( /OL. XV JANUARY 1961 NO. 1 PACIFIC SCIENCE A QUARTERLY DEVOTED TO THE BIOLOGICAL AND PHYSICAL SCIENCES OF THE PACIFIC REGION SISTER MARGARET JAMES ROE The Indigenous Hawaiian Species of Hibiscus PATRICIA R. BERGQUIST A Collection of Porifera from Northern New Zealand RITA D. SCHAFER Effects of Pollution on Free Amino Acid Content of Two Marine Invertebrates MAXWELL S. DOTY and ISABELLA A. ABBOTT Studies in Helminthocladiaceae, Part I. Helminthocladia G. BROWNLIE Studies on Pacific Ferns, Part III. Lindsaeoid Ferns ANGELES ALVARINO Two New Chaetognaths from the Pacific KEITH H. WOO DWICK Polydora rickettsi, a New Species of Spionid Polychaete SHOICHI F. SAKAGAMI An Ecological Perspective of Marcus Island PAUL N. SUND New Species of Chaetognatha from the Waters off Peru ALISON KAY A New Opisthobranch Mollusc from Hawaii L. J. DUMBLETON The Aleyrodidae of New Caledonia DONALD P. ABBOTT Ascidians of Point Barrow, Part I R. A. M. BERGMAN Anatomy of Coluber radiatus and Coluber melanurus DIPTIMAN CHAKRAVARTI and RONALD EISLER Strontium-90 and Gross Beta Activity in Rongelap Coconut Crabs UNIVERSITY OF HAWAII PRESS BOARD OF EDITORS O. A. BUSHNELL, Editor-in-Chief Department of Microbiology, University of Hawaii Robert Sparks, Assistant to the Editors Office of Publications and Information, University of Hawaii Agatin T. Abbott Department of Geology and Geophysics University of Hawaii Thomas S. Austin Bureau of Commercial Fisheries, Hawaii Area (U. S. Fish and Wildlife Service) Honolulu, Hawaii Brian M. Bary Institute of Oceanography University of British Columbia Vancouver, Canada E. H. Bryan, Jr. Bernice P. Bishop Museum Honolulu, Hawaii Koji Hidaka Geophysical Institute Tokyo University Tokyo, Japan Colin S. Ramage Department of Geology and Geophysics University of Hawaii Paul J. Scheuer Department of Chemistry University of Hawaii Albert J. Bernatowicz Department of Botany University of Hawaii Albert L. Tester Department of Zoology and Entomology University of Hawaii Howard Boroughs Instituto Interamericano de Ciencias Agricolas Turrialba, Costa Rica Miklos F. Udvardy Department of Zoology University of British Columbia Vancouver, Canada Thomas Nickerson, Managing Editor Office of Publications and Information, University of Hawaii SUGGESTIONS TO AUTHORS Contributions to Pacific biological and physical science will be welcomed from authors in all parts of the world. (The fields of anthropology, agriculture, engineering, and medicine are not included.) Manu- scripts may be addressed to the Editor-in-Chief, PACIFIC SCIENCE, University of Hawaii, Honolulu 14, Hawaii, or to individual members of the Board of Editors. Use of air mail is recommended for all communications. Manuscripts will be acknowledged when received and will be read promptly by members of the Board of Editors or other competent critics. Authors will be notified as soon as possible of the decision reached. Manuscripts of any length may be submitted, but it is suggested that authors inquire concerning possi- bilities of publication of papers of over 30 printed pages before sending their manuscripts. Authors should not overlook the need for good brief papers, presenting results of studies, notes and queries, com- munications to the editor, or other commentary. PREPARATION OF MANUSCRIPT It is requested that authors follow the style of Pacific Science described herein and exemplified in the journal. Manuscript Form. Manuscripts should be typed on one side of standard-size, white bond paper and double-spaced throughout. Pages should be consecu- tively numbered in upper right-hand corner. Sheets ( Continued on inside back cover ) PACIFIC SCIENCE A QUARTERLY DEVOTED TO THE BIOLOGICAL AND PHYSICAL SCIENCES OF THE PACIFIC REGION VOL. XV JANUARY 1961 NO. 1 Previous issue published October 12, I960 CONTENTS PAGE A Taxonomic Study of the Indigenous Hawaiian Species of the Genus Hibiscus (Malvaceae) . Sister Margaret James Roe .... 3 A Collection of Porifera from Northern New Zealand, with Descriptions of Seventeen New Species. Patricia R. Bergquist...... 33 Effects of Pollution on the Free Amino Acid Content of Two Marine Inverte- brates. Rita D. Schafer. 49' Studies in the H elminth o cladiaceae (Rhodcphyta): Helminthocladia. Maxwell S. Doty and Isabella A. Abbott. 56 Studies on Pacific Ferns, Part III. The Lindsaeoid Ferns. G. Brownlie.. 64 Two New Chaetognaths from the Pacific. Angeles Alvarino 67 Polydora rickettsi, a New Species of Spionid Polychaete from Lower California. Keith H. Woodwick 78 An Ecological Perspective of Marcus Island, with Special Reference to Land Animals. Shoichi F. Sakagami...... 82 ' Two New Species of Chaetognatha from the Waters off Peru. Paul N. Sund.. 105 A New Opisthobranch Mollusc from Hawaii. Alison Kay 112 The Aleyrodidae ( H emiptera—Homoptera ) of New Caledonia. L. J. Dumbleton 114 The Ascidians of Point Barrow, Alaska, Part I. Suborder Phlebobranchia (Enterogona). Donald P. Abbott. 137 The Anatomy of Coluber radiatus and Coluber melanurus. R. A. M. Bergman 144 Strontium-90 and Gross Beta Activity in the Fat and Nonfat Fractions of the Liver of the Coconut Crab (Birgus latro) Collected at Rongelap Atoll during March 1958. Diptiman Chakravarti and R onald Eisler 155 NOTE 160 Pacific Science is published quarterly by the University of Hawaii Press, in January, April, July, and October. Subscription price is $4.00 a year; single copy, $1.25. Check or money order payable to University of Hawaii should be sent to University of Hawaii Press, Honolulu 14, Hawaii, U. S. A. Printed by Star-Bulletin Printing Company, Inc., 420 Ward Avenue, Honolulu 14, Hawaii. SMITHSONIAN INSTITUTION MAR Z 3 1981 A Taxonomic Study of the Indigenous Hawaiian Species of the Genus Hibiscus (Malvaceae) Sister MARGARET JAMES ROE 1 The GENUS Hibiscus is a member of the Malva- ceae, a family of 80 or more genera distributed throughout the world except in frigid regions. Many abound in the tropics and subtropics, where they are important as ornamental and commercial plants. Since Hibiscus hybridization was first suc- cessfully accomplished in Hawaii in 1872 by Governor Archibald Cleghorn of Oahu, a grow- ing interest in the genus has prevailed not only in Hawaii, where the legislature has designated one of the species as the official flower, but also throughout global tropical and subtropical re- gions. Horticulturists striving to secure desir- able varieties have introduced to Hawaii many new species, subspecies, and forms (Naka- sone, 1953). The popularity of the genus has stimulated numerous amateur hybridizers to produce thousands of complex hybrids whose parentage is now impossible to trace. The at- tractiveness of these exquisite cultivars has en- couraged and nourished the organization of several societies to foster such hybridization. Little attention, however, has been given to the indigenous members of the genus, several of which have been useful in the production of these hybrids. Heretofore, no single compre- hensive work has been done on the native mem- bers of the genus. Individual descriptions have appeared in print as far back in 1819. Hille- brand (1888) and Rock (1913) both included 1 Former address : Maryknoll Sisters, Honolulu, Ha- waii; present address: Apartado postal 1121, Guate- mala City, Guatemala, C. A. Manuscript received October 1, 1959. several Hibiscus species in their Floras. It is the purpose of this work to record as complete as possible an investigation of the native species, varieties, and forms, resolving the groups of the indigenous Hibiscus populations into taxonom - ically recognizable groups. This study could neither have been under- taken nor accomplished without the help of friends who gave time, knowledge, and en- couragement. I am primarily indebted to Dr. Harold St. John, former Professor of Botany at the University of Hawaii, for introducing me to the problem. I am indebted to Mr. Irwin Lane for supervising the taxonomic and nomen- clatural conclusions. Dr. Joseph F. Rock also contributed many hours in discussing the prob- lem and collecting specimens. I am indebted to Miss Marie Neal, Dr. Otto Degener, Mr. and Mrs. Colin Potter, and Messers. Paul Weissich, Donald Anderson, Albert Duvel, and Henry Wiebke for their contributions to this work. Dr. Reed C. Rollins and Dr. Robert C. Foster of the Gray Herbarium checked and reported on material there. Dr. Richard S. Cowan investi- gated the Hibiscus specimens at the Smithsonian Institution (U. S. National Museum) and sent lengthy descriptions and explanations. On the island of Kauai, Mr. George Cliff acted as guide during my stay at Waimea Canyon. Mrs. Hector Moir of Poipu, Mr. John Santos of Eleele, Mr. Stephen Au of Lihue, Mr. Ronald Harker of Hale Manu, Mr. Adam Jacinto of Kalaheo, and Mr. Solomon Malina of Kipu were instrumental in helping me with collections on Kauai. The illustrations were made by Sister Chris- 3 4 PACIFIC SCIENCE, Vol. XV, January 1961 rina Francis of Maryknoll High School. All herbarium specimens mentioned in this paper are deposited at the Bernice P. Bishop Museum in Honolulu, unless otherwise stated. HISTORY OF THE GENUS Hibiscus IN HAWAII The history of the indigenous Hawaiian Hi- biscus dates back to Gaudichaud’s Hibiscus Youngianus, which was collected on the Frey- cinet Expedition, 1819. He mentioned it ( 1826) but failed to add any details; hence it was a nomen nudum as published, until Hooker and Arnott furnished the description (1832). Gray (1854) described both H. Brackenridgei and H. Arnottianus. In 1888 Hillebrand added H. Kokio to the three published species. This was followed closely by Heller’s description (1897) of the Kauai white H. waimeae. Hochreutiner (1900) included all members in the genus known in all parts of the world. The Hawaiian populations received treatment by him. Forbes (1912) found and named an endemic Kauai .species, H. kahilii, which is still considered rare. From 1912 to 1930 little was done in the way of publication until Caum presented Rock’s H. Brackenridgei var. molokaiana; a variety of his own, H. Brackenridgei var. kauaiana; and a variety to H. Kokio var. pukoonis. Skottsberg, on his Hawaiian Bog Survey assignment, found and described a variety and a form of H. Ar- nottianus. From that time, 1944 to 1957, no new groups were established. Degener (1957) recently raised Skottsberg’s var. punaluuensis of H. Arnottianus to the level of a species, H. punaluuensis. At present we recognize 15 populations of Hibiscus as indigenous to the Hawaiian Islands — 9 species, 5 varieties, and 1 form. DESCRIPTION OF GENERIC CHARACTERS The genus Hibiscus is a large taxon whose species, hundreds in number, are restricted to the warmer regions of the world. It is polymor- phic, including species that are herbaceous, shrubby, or arborescent, some of the latter at- taining heights of 30 ft. or more. Linnaeus (1754) listed Hibiscus under genus no. 75 6 and gave the fundamental description of this taxon. His original characters included double perianth; bracts 8 or more in number; calyx cup-shaped; 5 heart-shaped petals united at base; numerous filaments joined to column; anthers reniform; 5 -celled ovary; divided stigma; reniform seeds. Since this original description of Linnaeus, we have broadened our knowledge of the genus. It can be summarized in the following manner. Involucre consists of several free bracts. Calyx 5-lobed or toothed, persists with fruit. Flowers borne singly in axils, composed of 5 petals which are frequently wider toward their apices, nar- rowing towards base, where they usually unite with column. Flowers commonly large, showy, and of conspicuous colors. Staminal column com- posed of filaments which are united into hollow sheath for some distance from base; 5 -toothed at apex. Filaments extend out from column at various positions, usually close to stigma. Style runs through column and terminates in 5 branches, each of which has a terminal discoid stigma. Five carpels, each containing several ovules, united into a 5 -celled ovary. Capsule opens loculicidally. Seeds vary in shape from reniform to globose, with surfaces from glabrous to tomentose. Leaves variously lobed or entire, alternate, stipulate, and scarcely to distinctly palmately veined. The genus Hibiscus seems closely related to the genera Paritium and Abutilon. Paritium is allied to Hibiscus inasmuch as both genera have involucrate calyces which are 5-lobed; also 5- style branches with capitate stigmas. The in- volucral bracts of Paritium , however, are united, whereas those of Hibiscus are free. Abutilon and Hibiscus both have carpels with 2 or more ovules. Abutilon species do not have involucral bracts but have staminal columns ending in filaments. In distinguishing the various taxa under con- sideration, the following characters were most valuable: bract number (5-10), calyx length ( 1. 0-3.0 cm.), and venation of calyx and leaves. Staminal column lengths (2.0-19.0 cm.), ves- ture, petal shapes, position of filament exten- sion from column, and seeds are also important differentiating characters. Hawaiian Hibiscus — -Roe 5 KEY TO KNOWN HAWAIIAN MEMBERS OF THE GENUS Hibiscus A. Corolla yellow B B. Calyx 10-veined; petals 3 .0-4.0 cm. long C C. Leaves deeply lobed; bracts 8-10, glabrous, terete, nonglandular; calyx glandular; petals 3.0-3. 5 cm. long, ca. 2.5 cm. wide; staminal column 4.0-4.5 cm. long.. ..2. H. Brackenridgei C. Leaves moderately lobed; bracts 7-8, hirsute, terete, nonglandular; calyx nonglandular; petals 3. 0-4.0 cm. long, ca. 5.0 cm. wide; staminal column 7. 0-8.0 cm. long 5. H. Brackenridgei var. mokuleiana B. Calyx 5 veined; petals more than 4.0 cm. long D D. Leaves deeply lobed; bracts 8-9, hirsute, terete, glandular; calyx nonglandular; petals ca. 5.0 cm. long, 4.3 s — 4.5 cm. wide; staminal column 4.0-5. 0 cm. long 3. H. Brackenridgei var. molokaiana D. Leaves entire to shallow lobed; bracts 5, hirsute, nonterete, nonglandular; calyx gland- ular; petals 5.7-6.0 cm. long, 4.8-5.0 cm. wide; staminal column 2. 3-2. 5 cm. long 4. H. Brackenridgei var. kauaiana A. Corolla not yellow E E. Corolla pink; plant densely hirsute; bracts 10-12, bifid, 1.0-1. 3 cm. long, ca. 1.0 mm. wide; calyx glandular, lobes 2. 3-2. 5 cm. long, cleft 1.0 cm. from apex 1. H. Y oungianus E. Corolla not pink; entire plant not densely hirsute; calyx nonglandular F F. Corolla red G G. Calyx glabrous; petioles 2. 0-4.0 cm. long; staminal column 5.5-57 cm. long; petals 5.5 cm. long, 3-5 cm. wide H H. Calyx 2. 8-3.0 cm. long; 1.2 cm. wide; lateral calyx nerves fused immediately below cleft 6. H. Kokio H. Calyx 2.4-2. 5 cm. long; 1.0 cm. wide; lateral calyx nerves fused 5-10 mm. below cleft.... 7. H. Kokio var. pukoonis G. Calyx nonglabrous; petioles less than 2.0 cm. long; petals more than 5.5 cm. long, less than 3.5 cm. wide I I. Staminal column more than 5 cm. long; peduncles more than 1.0 cm. long J J. Calyx cleft 1.2-1. 3 cm., pilose; petioles 0.5-1. 8 cm. long; petals 6.5-67 cm. long, 2. 5-3.0 cm. wide; staminal column ca. 6.0 cm. long 8. H. Saintjohnianus J. Calyx cleft 2. 0-3.0 cm., pubescent; petioles 0. 3-0.8 cm. long; petals 6.0-6. 5 cm. long, 1.0-1. 2 cm. wide 9. H. kahilii I. Staminal column less than 5 cm. long; peduncles less than 1.0 cm. long 10. H. Newhousei F. Corolla white K K. Staminal column white, leaf margin denticulate, apex rounded 11. H. immaculatus K. Staminal column red, leaf margin entire or serrate, apex almost acute L L. Leaves velvety to the touch; calyx cleft 1.0-1. 5 cm.; bracts 1.5-2. 5 cm. long - 12. H. waimeae L. Leaves lack velvety touch; calyx cleft only 0.5 cm.; bracts 0. 5-1.0 cm. long M M. Leaves 4-10 cm. long, 4—7 cm. wide, ovate to elliptical-parabolical, glabrous; petioles 0. 5-2.0 cm. long, glabrous.... N N. Leaves 8-10 cm. long, 5-7 cm. wide; calyx 2-3 cm. long.. ..13. H. Arnottianus N. Leaves 4-6 cm. long, 4-5 cm. wide; calyx 1-2 cm. long 14. H. Arnottianus f. parviflora M. Leaves 10-25 cm. long, 8-20 cm. wide, ovate-subcordate, pubescent; petioles 3-12 cm. long, densely puberulent 15. H. Arnottianus var. punaluuensis 6 PACIFIC SCIENCE, Vol. XV, January 1961 l. Hibiscus Y oungiawus Gaud, ex H. & A. Figs. 1-3 Hibiscus Y oungianus Gaud, ex H. & A. Bot. Beechey Voy., p. 79. 1832. H. furcellatus Lam. var. Y oungianus (Gaud.) Flochr., Ann. Conserv. Jard. Geneve, p. 132. 1900. DESCRIPTION. Stiffly erect with a paucity of pithy branches, to 3 m. high (average ca. 2.5 m. ) , diameter at base ca. 8 cm., hirsute and pilose throughout, lateral branches very strongly ascending, nearly parallel to the main stem. Leaves more abundant above, "Scheele green” in color (cf. Ridgway); stipules 5-6 mm. long, 0.5 mm. wide, pubescent; petioles 9—1 3 cm. long, scabrous; blade 10-12 cm. long, 11-12 cm. wide, wide ovate, varying from nearly entire to mod- erately lobed, base cordate, upper portions acut- ish, margins serrate-dentate, both surfaces scab- rous-hispid, this more abundant on the veins; veins 5-9, palmately arranged throughout the blade. Peduncle axillary, 1.5 cm. long, 3 mm. in diameter, puberulent. Involucral bracts 10-12, 1-1.3 cm. long, ca. 1 mm. diam., terete, bifid at apex. Calyx "Scheele green” (cf. Ridgway), 2.2- 2.5 cm. long, 1.5-1. 8 cm. diameter at throat, hirsute, cleft 1 cm., lobes lanceolate, veins 10, prominent, gland on midvein of each lobe at the throat level. Petals "mallow purple” (cf. Ridgway), 5.8— 6.5 cm. long, 4-4.5 cm. wide, tubular at base, prominently nerved, microscopic hairs on both surfaces. Staminal column dark maroon-purple, 2. 0-3.0 cm. long; free filament tips extend 1-2 mm. from column, arranged in rings approximately 5 mm. apart; anthers deep purple; style inconspicuously exerted; style branches 1—2 mm. long, each terminating in a 1 mm. discoid stigma. Ovary ca. 1.0 cm. long, 8 mm. in diameter; seeds numerous, 2.8 mm. long, 2.0 mm. wide, dark brown, glabrous. This species was first brought to our atten- tion by Gaudichaud (1826): ". . . un hibiscus nouveau de la section 5 (furcaria) de M. De- candolle ( h. youngiana ) ...” ( nomen nudum ) . The first description was made in 1832 by Hooker and Arnott, who credited the species to Gaudichaud, ex Hooker and Arnott. Hille- brand (1888) included this species, as did Hel- ler (1897). Both authors regarded it as a true indigenous species. However, in 1900, Hoch- reutiner monographed the genus and reduced H. Y oungianus to a variety of H. furcellatus Lam., and cited specimens of the latter from Central and South America as well as from Ha- waii. I have grown the species in question for several years and I know its morphology and growth habits well. I have also obtained and observed H. furcellatus Lam. and find the two populations very similar. It is easy to see how Hochreutiner arrived at his conclusion. How- ever, after examining them minutely I feel that their differences warrant the classification of them as distinct species. Table 1 shows several important distinguish- ing characteristics. Skottsberg (1926) reviewed Hochreutiner ’s revision and concluded that the latter’s proposed close proximity of H. Y oungianus and H. fur- cellatus was an improbable one. A contradiction regarding locality might also be thought to exist when comparing a state- ment of Hillebrand’s found in small print fol- lowing his description of H. Y oungianus ("In marshes and abandoned taro patches here and there on all the islands. . . .”) with a statement of Mrs. Sinclair’s (1885) that the species . . . was once a common flower in nearly all valleys and sheltered places; seeming to flourish equally well on both the leeward and windward sides of the islands. Now cattle and cultivation have almost exterminated the plant on the dry lee-side, but it is still frequently met with on the windward side; where, owing to the more luxuriant vegetation, many plants, which have disappeared from the leeward side, are still found. The Hauhele {sic'] was once so plentiful in many parts that the aho’ (thatching sticks) of the houses were made of the stems, and any- one who knows what a great quantity of 'aho’ a single, old-fashioned house required, will readily see how abundant the plant must have been. Fig. 1 . Hibiscus Y oungianus Gaud, ex H. & A. Kawainui Swamp, Kailua, Oahu. Roe no. 205. Half scale. a, Cross section of ovary, XI \b, staminal column, X L Hawaiian Hibiscus — Roe 7 8 PACIFIC SCIENCE, Vol. XV, January 1961 TABLE 1 PART H. Y oungianus H. furcellatus Stipules length 5-6 mm. length 8-10 mm. vesture hirsute vesture short, stellate Leaf color "Scheele green” (yellowish green) "grass green” (bluish green) Leaf shape apex acute acuminate blade roundly lobed blade acutely cleft Stem puberulent, hirsute pilose Bracts length 1—1.3 cm. length 1. 5-2.1 cm. vesture densely hispid vesture short, stellate hairs 1-1.5 mm. in length hairs microscopic Calyx of fresh flower sepal length 2.2-2. 5 cm. sepal length 1.7-1. 8 calyx diameter at throat 1.5-1. 8 cm. calyx diameter at throat 1.1-1. 3 cm. densely hispid sparsely hispid cleft 1 cm. from sepal apex cleft 1.8 cm. from sepal apex veins raised veins extremely raised Petals length 5. 8-6. 5 cm. length 4-5 cm. width 4-4.5 cm. width 2.5-3 cm. color, "mallow purple” (deep pink) color, light "phlox purple” (pale pink) Hillebrand is correct in stating that H. Y oung- ianus is a marshland plant, for I have collected it growing in several inches of water in the cen- tral area of Kawainui Swamp, Oahu, and sim- ilarly in water at the foot of Wailua Falls, Kauai. But likewise, Mrs. Sinclair could be correct in her statement of locality, for I have recently observed it in upper Manoa on a rather dry slope, but in a rainy belt. The species, however, grows more luxuriantly and flowers continually in a wet environment. I have grown it in areas FIG. 2. Hibiscus Y oungianus Gaud, ex H. & A. of little rainfall. It continues to grow, but much less vigorously than in swamp areas. Heller (1897) is incorrect in stating, "It is found only on Oahu.” It has been collected on all of the major islands except Molokai, Lanai, and Niihau. (See Fig. 3.) SPECIMENS EXAMINED: Hawaii: Rainbow Falls, May 16, 1915, C. N. Forbes 525. H. Lava flow near Hilo, May 17, 1915, C. N. Forbes 545.H. Kauai: Hanalei, August 4, 1909, C. N. Forbes 128.K. Wailua Falls, October 5, 1916, C. N. Forbes 498.K. Wailua Falls, July 26, 1957, on the bank of the pool at the bottom of the falls, Roe 251. Weoweopilau Stream, on banks of stream near cane fields, altitude 500 feet, July 28, 1957, Roe 256 . Maui: Haleakala, October 10, 1922, Skotts- berg 793. Oahu: Woodlawn, Manoa, October 2, 1933, Neal. Waihole Valley, November 4, 1938, Neal. Waihole Ditch, May 1957, Pearsall. Kawainui Swamp, Kailua, May 4, 1930, St. John. Manoa about 500 yards off Woodlawn Drive, about mile west of Old Ti Slide, altitude 550 ft., Octo- ber 31, 1956, Roe 203. Kawainui Swamp, Kai- lua; plant growing in water, December 31, 1956, Roe 205. Hawaiian Islands: H. Mann & W . T. Brig- ham 598. Onekahakale Beach, August 15, 1936. V. O. Fosberg 64. Hawaiian Hibiscus— Roe 9 FlG. 3- Distribution of Hibiscus Y oungianus. Solid circles indicate localities for which exact data are avail- able; open circles represent localities where species have been reported but plants are unavailable. 2. Hibiscus Brackenridgei Asa Gray Figs. 4, 5 Hibiscus Brackenridgei Asa Gray. Bot. U, S. Expl. Exped, p. 175. 1838. DESCRIPTION: Shrubby plant, glabrous, leafy. Leaves rounded in outline and subcordate, diam- eter 5-10 cm., 5-7 lobes separated by acute and narrow sinuses, coarsely toothed, terminal lobe prolonged. Stipules setaceous, caducous. Flowers axillary; peduncle 5 mm. long, puberulent and sparingly hispid; bracts 8, rigid, setaceous- subulate, glabrous, nonglanduliferous. Calyx his- pid, cleft to below the middle, lanceolate lobes each bearing dorsal gland on midrib near the base. Corolla yellow, green when dry. Petals 4- 5 cm. long, externally stellate, pubescent. Staminal column antheriferous throughout, apex 5- toothed. Style branches hirsute; stigmas de- pressed-capitate. Ovary densely villous-hispid. Capsule 2 cm. long, ovoid, closely invested by calyx, silky-hispid. Seeds angled, minutely to- mentose. hoi, on: PE: From West Division of Maui, in Gray Herbarium. (In the Bishop Museum Her- barium are preserved a leaf and fruit of mate- rial used by Gray.) SPECIMENS EXAMINED: Lanai : Maunalei, March 1918, G, C. Munro 638 . Awalua, January 1919, Munro . Kanepuu, June 15, 1927, Munro . Kaena, April 1921, Munro. Puhiehelu, May, 1921, Munro . Maui : Pohakea Gulch, West Maui, on a very arid slope, June 11, 1927, Degener and Wiebke 3634 , 3633 . Brown Hill, Kula, Hillebrand and Lydgate. H. Brackenridgei , in at least one of its forms, has been reported and collected on the islands of Oahu, Maui, Lanai, Molokai, Kauai, and Ka- hoolawe. After gathering together specimens, descriptions, and illustrations of the group, and after studying the original type specimen, de- scription, and locality, the following conclusions have been reached. There are at least four dis- tinct groups in the population. As the original plant described by Gray (1854) was located "on a mountain in the west division of Maui,” this Maui plant, a portion of which type speci- men is in the Bishop Museum Herbarium, is rightly the true H. Brackenridgei Gray. In 1930 Caum published two varieties: H. Brackenridgei Gray var. molokaiana Rock and var. kauaiana Caum. The holotypes for these two Hawaiian Hibiscus — Roe 11 Fig. 5. Distribution of Hibiscus Brackenridgei populations: Solid circles, H. Brackenridgei (exact data) Open circles, H. Brackenridgei (plants unavailable) Solid squares, var. molokaiana (exact data) Open squares, var. molokaiana (plants unavailable) Solid triangles, var. kauaiana (exact data) Open triangles, var. kauaiana (plants unavailable) X, var. mokuleiana (exact data) varieties are at the Bishop Museum. I have studied various specimens of the Lanai plants and cannot find sufficient difference from the Maui population to maintain it as an en- demic variety. We have records that Jules Remy made col- lections of this species on the island of Kahoo- lawe between 1851-1855 (Remy no. 559). Remy s specimens are not available and the plants can no longer be found on Kahoolawe, so a statement cannot be made regarding their taxonomy. 3 .Hibiscus Brackenridgei var. molokaiana Rock Figs. 5, 6 Hibiscus Brackenridgei Gray var. molokaiana Rock. New Hawaiian Plants, Bishop Mus. Occ. Pap. 9(5): 4, pis. 2, 3. 1930. DESCRIPTION: Straggling shrub; leaves 6-8 cm. long, deeply lobed; involucral bracts 8-9, linear, subulate, terete, glanduliferous on inner side, 2.5 cm. long, 2.5 mm. wide, hirsute with long yellowish hairs, adnate to base of calyx. Calyx "carmine red,” 1. 5-2.0 cm. long, lobes prominently midribbed and nonglandular. Co- rolla "canary yellow,” tube "deep purple.” Petals 5 cm. long, 4.5 cm. wide, 10-12 veined, stellate puberulence on outside. Ovary densely villous- pubescent with whitish hairs. Staminal column 4 cm. long, irregularly antheriferous, naked at the base for 1 cm., puberulent. Free filament tips 2.5 mm. long. Style branches 2 mm. long, hirsute. Stigmas depressed, capitate. HOLOTYPE: Rock, west end of Molokai, Feb- ruary 1920. This variety was first found by Rock on west- Fig. 4. Hibiscus Brackenridgei Gray. From West Maui. (This illustration, presented by Otto Degener, was made from his specimens collected on Maui.) 12 PACIFIC SCIENCE, VoL XV, January 1961 ern Molokai, back of Kalaeokalaau in 1910, and collected there by him again in 1920. Cuttings were made and several plants were cultivated at the "Rockery” on the University of Hawaii campus. I have since tried to find traces of this variety again on Molokai but with little success. Henry Wiebke, principal of Holomua School, Hoolehua, told me that he collected it in the late 1920s but cannot find it now. Mrs. Cooke, Molokai resident, is anxious to preserve the variety also but has been unable to locate it. (She did have several plants growing along her driveway until a few years ago.) 4. Hibiscus Brackenridgei var. kauaiana Caum Figs. 5, 7 Hibiscus Brackenridgei Gray var. kauaiana Caum. New Hawaiian Plants, Bishop Mus. Occ. Pap. 9(5): 5-6, pi 4. 1930. description: Erect to spreading shrub to 1 m. Leaves dentate, cordate, 8 cm. long, 8 cm. wide, shallowly 3 dobed, densely pubescent on both surfaces. Petioles 4-5 cm. long, pubescent. Stipules setaceous. Peduncles 1.5 cm. long; 5 bracts, 3 cm. long, 8 mm. wide, halberd shaped, not terete, not glanduliferous, hirsute with short hair. Calyx 2.5 cm. long, triangular lobes each with an oblong gland at the base of a thick mid- FlG. 6. Hibiscus Brackenridgei var. molokaiana Rock. (Picture taken by Joseph F. Rock.) rib, hirsute with colorless hairs, glabrous on in- ner surface. Corolla 12 cm. in diameter, "lemon yellow” to "canary yellow,” dark "reddish-pur- ple” throat 3 cm. deep. Petals spreading, slightly reflexed at tips, 6 cm. long, 5 cm. wide, 14-16 nerved. Ovary pubescent with colorless hairs. Staminal column 3 cm. long, irregularly an- ther if erous for entire length, glabrous; lower filaments 5 mm. long, upper ones shorter. Style branches 2 mm. long, hirsute. Stigmas capitate, depressed, deep "reddish-purple.” Capsule ovoid, acute at tip, 2.5 cm. high, 17 mm. maximum diameter. Seeds broad-reniform, 3 mm. long, hirsute with short gray hair. HOLOTYPE: E, L. Caum. From plants in cul- tivation at 1420 Piikoi Street. Honolulu. SPECIMENS EXAMINED: Kauai : Waiawa, April 1919, Rock 16038; also Rock 17141. "V. Knudsen’s home,” 1919, Rock. Waimea, near Hale Manu, altitude 3300 feet, July 28, 1957, Roe 233, 234, 231. Oahu : 1420 Piikoi Street, cuttings from orig- inal type plant, Caum. 1508 Alexander Street, cuttings from Hale Manu plant, Roe 307. H. Brackenridgei var. kauaiana is quite un- common. I located it at Hale Manu near Kokee, but there were only three plants of it at this location. In tracing its origin to the Waimea area I found cuttings had been brought up there from a wild plant at a lower elevation. The original plant has not been found. Several collectors, including Rock, have men- tioned two distinct varieties of H. Bracken- ridgei on Kauai, one more arborescent than var. kauaiana Caum. In 1957 I made cuttings of a plant not in flower on Kauai that I felt had different growth habits. This plant is in cul- tivation now, with one also thought to be of this arborescent Kauai variety, given to me in July, 1958, by Rock. They have not flowered, so at present taxonomic judgment cannot be passed on their distinctness. 5. Hibiscus Brackenridgei var. mokuleiana nov. var. Figs. 5, 8 DESCRIPTION: Arbor 10-12 m. alta basi 20- 25 cm. diametro. Stipulis 1-2 cm. longis pilosis. Petiolo 6-9 cm. longo piloso, lamina 9-11 cm. Hawaiian Hibiscus — Roe 13 longa 8-10 cm. lata, 5-lobatis sinibus angustis, basi cordate, margine serrata, nervls palmatis. Pedunculo L5 cm. iongo, piloso echinatoque. Bracteis 7-8, 2-3 cm. longis subulatis teretibus, nonglanduliferibus, hispidis. Calyce campanu- lata 3 -4 cm. longa, 1.5— 1.8 cm. lata .lobis eglanduliferis inter lobes 1.5 cm. partitis exras hispida deci-nervata. Corolla citrea, tubo atro purpurea. Petalis 8-1 0 : cm. longis 5-8 cm. latis extus pubescentis, 12-16 nervatis. Ovario 0.8- 1.0 cm. Iongo, 5 cm. kto, dense albo-villoso. Columna staminea 7-8 cm. longa puberula ex- tremitates filamentomm liberae verticillatae per totam columnam longitudinem exsertae, ramuli styli 2 mm. long! tomentosi, stigmis capitatis compressis. Capsula 2. 0-2. 5 cm. longa lignea pilosa. Seminibus angulosis hirsutis. Tree 10-12 m. tall, 20-25 cm. in diameter at base. Stipules 1-2 cm. long, pilose. Petiole 6-9 cm. long, pilose. Leaf 9-11 cm. long, 8-10 cm. wide, 5 -lobed with narrow sinuses, base cordate, apex acute, margin serrate, palmately veined. Peduncle 1.5 cm. long, pilose and bristly; bracts 7-8, 2-3 cm. long, subulate, terete, nonglanduliferous, hirsute hairs 2-3 mm. long; calyx campanulate, 3-4 cm. long, 1.5— 1.8 cm. wide, lobes nonglanduliferous, deft 1.5 cm. from apex, hirsute with colorless hair on outer sur- face, inner surface glabrous, 1 0 veined. Corolla "lemon yellow,” tube "deep purple.” Petals 8- 10 cm. long, 5-8 cm. wide, pubescent on outer surface, 12-16-veined. Ovary 0.8-1. 0 cm. long, .5 cm. in diameter, densely pubescent with white villous hairs. Staminal column 7-8 cm. long, puberulent, free filament tips extend in rings the entire length of column. Style branches 2 mm. long, tomentose. Stigmas capitate, depressed. Capsule 2. 0-2. 5 cm. long, woody, pilose. Seeds angulate, hirsute. holgtype: Roe 210. Second gulch east of Mt. Kaala, Oahu. Altitude 600 feet; about 40 ft. up a stream bank. specimens examined: Oahu: Gulch north of middle ridge between Puu Kamaokanui and Puu Pane, Jan. 10, 1932, O. Degener 8961 , also 8962, 8963. First small gulch northwest of Puu Pane Peak, April 2, 1950, 0. Degener 20943. Kapalama Heights, Kamehameha Girls FIG. 7. Hibiscus Brackemidgei var. kauaiana Caum. School (in cultivation) March 21, 1932, A. F. Judd . Waianae Mountains, near Mokuleia Forest Reserve, near valley on slope of Puu Iki, altitude 500 ft. May 13, 1941, Karl Korte. Waianae Mountains, Mokaleha Cliffs, May 1917, Rock 12986. Waianae Mountains, Mokuleia area, Feb- ruary 16, 1957, Roe 219. In cultivation on Alex- ander Street, Honolulu, April 9, 1958, Roe 269. Waianae Mountains, Mokuleia, second gulch east of Puu Kaupakuhale, northeast of Puu Kaala, May 14, 1933, St. John 13184 . There exists on Oahu in the Mokuleia sec- tion a population of the species that shows dis- tinct differences from the original Maui plants. This Mokuleia variety has been collected by Rock, A. F. Judd, Potter, St. John, Degener, K. H. Korte, and others. Degener made a notation on his plants nos. 8961, 8962, 8963 from this area, stating that this was a variety of H. Brack - enridgei Gray. Rock recently gave me a photo of a tree of this taxon and on the back of the picture he had written " Hibiscus Brackemidgei var., 1917. J. F. Rock, Makaieha Cliffs, Mo. 12, 986.” This variety takes the form of an erect tree (reaching 30 ft.). It differs from the species in its leaf shape, size, and pubescence, and in bract number and surface lengths of petiole, stipules, calyx, petals, and column. 14 PACIFIC SCIENCE, Vol. XV, January 1961 SUMMARY OF DISTINGUISHING CHARACTERS OF H. Brackenridgei POPULATIONS Upon observation of the H. Brackenridgei populations, it is evident that four groups are distinct. Var. molokaiana stands apart for its long petioles and small, deeply lobed leaves; var. kauaiana has larger leaves, entire to shallow lobed; and var. mokuleiana is definitely arbores- cent, bearing larger flowers. Table 2 compares the differentiating characters. 6. Hibiscus Kokio Hbd. Figs. 9-11 Hibiscus Kokio Hillebrand. Flora of the Ha- waiian Islands, p. 173. 1873. H. Boryanus Hook & Arn. Bot. Beechey Voy., p. 79. 1841. H. Arnottianus A. Gray in Bot. U. S. Expl. Exped. 1: 176 (pro parte). 1854. H. Arnottianus var. Kokio Hochr. Ann. Conserv. Jard. Bot. Geneve 4: 133. 1900. H. Arnottianus forma Sinclair. Indig. FI. Haw. Isl, p. 9. 1885. DESCRIPTION: Tall shrub with straggly branches from the base, growing to 20 ft.; few lateral branches. Stipules 8 mm. long, glabrous. Petioles, 2-4 cm. long. Leaves 9-12 cm. long, 4-6 cm. wide, elliptical-ovate, long-acuminate apex, margin sinuately crenate, smooth surface, scarcely palmately veined. Peduncle 3.0 cm., articulate 5 mm. from involucre base. Involucre 6-7 lanceolate bracts, 1.3 cm. long, 1.5 mm. wide, slightly puberulent. Calyx 2. 5-2. 8 cm. long, 1.3 cm. wide, 10-veined, cleft 1 cm. from tips, 5 lobes, lanceolate, no glands, "lettuce green” in color, glabrate. Flowers axillary. Staminal column 5.5 cm. long, "scarlet red.” Free filament tips extend 4 mm. from column, occupy apical 1.5 cm. of column. Style extension 4 mm.; style branches (horizontal) 8 mm. long. Petals 5.5 cm. long, 3.5 cm. wide, "scarlet red.” Ovary truncate-conical, 6 mm. long, 4 mm. wide, pubescent (short, colorless hairs). Capsule gla- brous, 2.5 cm. long. Seed reniform, 4 mm. long, brown, coarsely pubescent. The nomenclature of this taxon is closely correlated with that of H. Arnottianus Gray. A discussion of their confusion can be found in the section of this paper dealing with the latter species. Gray’s original description of H. Ar- nottianus included a fragment of a red-flowered hibiscus as well as the white. TABLE 2 Distinguishing Characters of H. Brackenridgei Population CHARACTERS H. Brackenridgei VAR. molokaiana VAR. kauaiana VAR. mokuleiana Growth habits shrub rambling shrub rambling shrub tree to 30 ft. Leaf lobing deeply cleft (5-7) deeply cleft (5) shallow-lobed (3) moderately lobed Leaf length 5 cm. 3-5 cm. 7-9 cm. 9-1 1 cm. width 5 cm. 5-6 cm. 7-9 cm. 8-10 cm. Stipules 0.5-1 cm. long 1 cm. long 1 cm. long 1—2 cm. long Petiole nearly glabrous puberulent densely pubescent puberulent Bract number 8-10 8-9 5 7-8 Bract shape setaceous, terete subulate, terete hastate, not terete subulate, terete Bract glands none glands on inner side none none Bract surface almost glabrous hirsute hirsute hirsute Bract length 1 cm. 2.5 cm. 3.0 cm. 2-3 cm. Calyx venation 1 0-veined 5 -veined 5 -veined 10-veined Calyx length 1.5 cm. 1.5 cm. 2.5 cm. 3—4 cm. width 1.5 cm. 1.0 cm. 1-1.5 cm. 1.5 cm. Calyx glands present none present none Petal length 3.5 cm. 5.0 cm. 6.0 cm. 8-10 cm. width 2.5-3 cm. 4.5 cm. 5.0 cm. 5-8 cm. Column length 4.0 cm. 4-5 cm. 2.5 cm. 7-8 cm. naked at base for 1 naked at base for 1 completely antheri- completely cm. cm. ferous antheriferous Hawaiian Hibiscus — Roe 15 Fig. 8. Hibiscus Brackenridgei var. mokuleiana Roe. (Picture was presented by Joseph F. Rock.) 16 PACIFIC SCIENCE, Vol. XV, January 1961 Fig. 9- Hibiscus Kokio Hbd. Plant collected on Nov. 23, 1956 at Kawaiiki Ditch Dam. Roe no. 204. (Thin straggly branch reaching 30 ft. in length.) Altitude 1,080 ft. Hawaiian Hibiscus — Roe 17 This red flower, as insignificant as it is in comparison with other luxuriant Hawaiian members of the genus, was designated in 1923 as the official flower of the Hawaiian Islands. This specimen has been collected by many bot- anists at the end of the Kawaiiki Ditch Trail overhanging the dam (Kawailoa, Koolau Range, elevation 1,080 ft.). From this location, many cuttings have been made and brought to cultivation. As H. Kokio is known to hybridize freely, taxonomists hesitate to classify new groups. However, there seems to be a native red- flowered plant on eastern Kauai still undescribed. Sufficient material is not available to permit satisfactory classification; however, it appears that this group is a new species, closely allied to H. Kokio Gray and H. kahilii Forbes. The re- maining Kauai native reds fall into either of the latter two taxa or that of the newly estab- lished species, H. Saint johnianus. At Puu Ka Pele in Waimea the Kokee rangers recently reported a red-flowered hibis- cus. This has been collected by several botanists who have labelled it Hibiscus sp. After bring- ing it to cultivation and watching it for a year, the author has concluded that it is an introduced species bearing a large, reddish-purple flower. HOLOTYPE: Fragment of the original type now in the Bishop Museum Herbarium reads "Ex Museo botanico Berolinensi.” (From cul- tivated species in Hillebrand’s garden.) SPECIMENS EXAMINED: Kauai: In cultivation, 1913, Dewitt Alexander. Nonau Mountains, Oct. 16, 1916, C. N. Forbes 599.K. West side of Nualolu Pali, Waimea Drainage Basin, July 3, 1917, C. N. Forbes 963.K. Oahu : Kanaikupai, Waianae Range, February 14, 1913, C. N. Forbes 1814.0. Kalihi Valley, October 1928, A. F. Judd 37. In cultivation at Queen Liliuokalani’s Residence, Honolulu, Spring, 1916, Rock. Kawaiiki Ditch Trail, Ka- wailoa, Koolau Range, altitude 1080 feet, No- vember 23, 1956, Roe 204 . In cultivation at Foster Gardens, Honolulu, July 7, 1957, Roe 226. In cultivation at the Hibiscus Gardens at Waikiki, July 11, 1958, Roe. Ditch Trail (near intake), Koolau Range, 1922, Skottsberg 203. In cultivation, Honolulu, 1913, Gerrit P. Wilder. FIG. 10. Hibiscus Kokio Hbd. Hawaiian Islands: H. Mann and W . T. Brig- ham, 218. 7. Hibiscus Kokio var. pukoonis Caum Fig. 11 Hibiscus Kokio var. pukoonis Caum. New Hawaiian Plants, Bishop Mus. Occ. Pap. 9(5): 7.1930. description: A shrub 2-3 meters tall. Stip- ules 0.5-0. 8 cm. long, setaceous. Petioles 2-4 cm. long. Leaves 8-12 cm. long, 4-7 cm. wide, ovate to elliptico-oblong, acuminate, sinuately crenate, scarcely palmately veined; soft-chartace- ous. Peduncles 2. 5-4.5 cm. long, solitary in axils near end of branches. Bracts 6-8, linear-lance- olate, 10-15 mm. long, 1.0- 1.5 mm. wide, not adnate to clayx, glabrate. Calyx 2.5 cm. long, 1.0-1. 3 cm. wide, cleft for 1.0 cm.; lobes acute, glabrate, tripli-nerved, the lateral nerves fuse 0.5-1.0 cm. below cleft, the calyx being 10- nerved at the base, nonglanduliferous. Petals 3- 3.5 cm. wide, 6-nerved, slightly ciliate at mar- gin, obovate, "scarlet red.” Staminal column 5.0- 6.0 cm. long, slender, light red, glabrous, acutely 5-lobed; free filament tips extend 0.3-0.4 cm. 18 PACIFIC SCIENCE, Vol. XV, January 1961 from column, crowded on distal fifth of column. Style extension 2 mm. above column, branches 0.7 cm. long, red, ciliate, spreading slightly up- ward from the horizontal. Stigmas capitate, 0.1 cm. in diameter, red. Ovary truncate-conical, 0.5 cm. long, 0.3 cm. wide, scarcely puberulent with short appressed colorless hairs. Capsule glabrous, 1. 5-2.0 cm. long. Seeds reniform, 0.4- 0.5 cm. long, brown, coarsely pubescent. In 1930, Caum published a variety of H. Kokio which he had collected just inside the native forest at the bottom of Pukoo Valley on Molokai (E, L. Caum, no. 155, Feb., 1921). Fie named this group of plants var. pukoonis. These plants do not flower freely and very rarely set fruit. They differ from the species in the tex- ture and other characters of the leaf, petal shape and color, venation of the calyx, and other characters. HOLOTYPE: E. L. Caum no. 155. Planted at Caum’s Honolulu residence (1420 Piikoi St.), taken from bottom of Pukoo Valley, Molokai, just inside the native forest, February, 1921. (Specimen in the Herbarium of Bishop Mu- seum.) SPECIMENS EXAMINED: Molokai: Wailau Trail, 1912, C. N. Forbes 327. Mo. Halawa (the ridge south of the valley), August 1912, C. N. Forbes 472.Mo. Moist woods of Wailau Valley, elevation 500 feet, July 3, 1933, St. John 13230. Oahu: In cultivation 1420 Piikoi St., Hono- lulu, October 8, 1930, E. L. Caum 153 (dupli- cate). In cultivation in the University of Ha- waii Arboretum, October 25, 1956, Roe 202; also April 14, 1957, Roe 222. In cultivation on Alexander Street, Honolulu, May 30, 1958, Roe 270. In cultivation, 2365 Oahu Avenue (brought from Molokai by W. T. Pope), 1931, St. John 11170. 8. Hibiscus Saint johnianus sp. nov. Figs. 12-14 DESCRIPTION: Arbor 6 m. alta basi 5-8 cm. in diametro. Stipulis 7-12 mm. longis. Petiolo 0. 5-2.0 cm. longo piloso; lamina 6-12 cm. longa 3-5 cm. lata elliptica vel oblonga, margine basi versus integro, nervo medio conspicuo, nervis haud palmatis, viride spinacii colorata (cf. Ridgway). Pedunculo 1-1.5 cm. longo 3 FIG. 11. Distribution of Hibiscus Kokio populations: Solid circles, Hibiscus Kokio Open circles, Hibiscus Kokio var. pukoonis Hawaiian Hibiscus — Roe 19 FIG. 12. Hibiscus Saintjohnianus Roe. Headland west of Hanakaiai, Napali Coast, Kauai. Altitude 700 ft. St. John 25989. Dec. 22, 1956. 20 PACIFIC SCIENCE, VoL XV, January 1961 mm. infra involucrum articulate; bracteis 7-8 lanceolatis 5-8 mm. longis. Calyce 3 cm. longo 1.5- 1. 8 cm. lato tubulate partite 1-1.3 cm. piloso 10-nervato lobis acutis viride lactucae coloratis (cf. Ridgway); petalis 6.5 cm. longis 2.5- 3 .0 latis rubro-coralaceis coloratis extus pub- escentis; columna staminarum 6 cm. longa gra- cilis glabra in apice 5-lobata acuti, apicibus fil- amentium liberis 4-5 mm. longis in quarta superiore columnae collacatis. Ramuli styli 5, 1.0 cm. longi modice puberulenti in 1.5 mm. stigma globosa terminanti. Ovario 5-loculare 1.0 cm. longis 6-8 mm. latis. Capsula 2.0 cm. longa 2.5 cm. lata superne. Seminibus 4 mm. longis 3 mm. latis globoso-reniformibus, fusco- sericeis. Tree 6 m. tall, diameter at base 5-8 cm. Stip- ules 7-12 mm. long. Petiole 0.5-2 cm., slightly pilose. Leaves 6-12 cm. long, 3-5 cm. wide, elliptic-oblong, upper three-fourths of margin serrate, lower fourth entire, midrib prominent, veins scarcely palmate, "spinach green.” Ped- uncle 1-1.5 cm. long, articulate 3 mm. below the involucre. Lanceolate bracts 7-8, 5-8 mm. long. Calyx tubular, 3 cm. long, 1. 5-1.8 cm. wide, cleft for 1-1.3 cm., lobes acute, pilose, 10 -nerved, "lettuce green.” Petals laterally sub- falcate, 6.5 cm. long, 2. 5-3.0 cm. wide, "coral red,” puberulent on outer surface. Staminal column 6 cm. long, slender, glabrous, apex acutely 5-lobed. Free filament tips extend 4-5 mm. from column, located on the upper fourth of column. The 5 -style branches slightly puberu- lent, 1 cm. long, each terminating in a 1.5 mm. discoid stigma. Ovary 5 -parted, 1 cm. long, 6- 8 mm. wide. Capsule 2.0 cm. long, 2.5 cm. wide at upper section. Seeds 4 mm. long, 3 mm. wide, globose-reniform, covered with a seriaceous pu- bescence, dark brown. HOLOTYPE: Headland west of Hanakapiai, Napali Coast, Kauai. Altitude — 700 feet; on partly precipitous slope. December 22, 1956. H. St. John 25,989. SPECIMENS EXAMINED: Kauai: Hanakoa, al- titude of 500 ft., Jan. 12, 1956, H. F. Clay. Kala- lau Trail, Hanakapiai, September 1913, C. N. Forbes 462.K. Hanakapiai, between Hanakapiai Stream and Hoolulu Stream, 450 m. alt., Decem- ber 24, 1956, 1. E. Lane 56-561. In cultivation at Fig. 13. Hibiscus Saintjohnianus Roe. (Picture taken by Irwin Lane on the Napali Coast, Kauai.) Eleele, brought there from Awaawa Puhi Trail, Na Pali Kona Reserve, Waimea, July 29, 1957, Roe 258. In cultivation at Alexander Street, Honolulu; brought from Na Pali Cliff Trail, June 1, 1958, Roe 271. This species was brought to my attention by H. St. John, who has made several collections of it from Hanakapiai, Napali coast, Kauai. It has been confused with H. Kokio Hbd. and H. kahilii Forbes. To one who is familiar with these two native species, this plant is differen- tiated by the distinct leaf shape and by the vermilion-orange corolla. It has closest affinity to H. Kokio, from which it differs by its' longer stipules, shorter bracts, larger calyx, longer yet narrower petals, longer staminal column, larger ovary, longer style branches, and shorter pe- duncle. There are records at the Bishop Museum stating that J. M. Lydgate brought it to cultiva- Hawaiian Hibiscus— Roe 21 tion at his home on Kauai. I investigated the grounds of his estate in 1957, desiring to locate this and other native Hibiscus that have been recorded as planted there by him, but none of them remain. This species is in cultivation on Kauai, as I recognized it on several occasions, collecting it in Eleele. The Eleele plants were recorded as having been brought from Awaawa Puhi Trail, Na Pali Kona Reserve (Waimea) . Irwin Lane, University of Hawaii, located the species in 1957 at Hanakapiai. I have the plant in cultivation on Alexander Street, Honolulu. It is also growing at Foster Botanical Gardens in Honolulu. Both plants are from cuttings made by St. John on the Napali coast, Kauai. 9. Hibiscus kahilii C. N. Forbes Fig. 14 Hibiscus kahilii C. N. Forbes. New Hawai- ian Plants, Bishop Mus. Occ. Pap. 5(1): 4. 1912. description: Tree growing to 8 m. Petiole 3-8 mm. long, scabrous on both sides of blade and petiole. Leaves 5-7 cm. long, 3-5 cm. wide, ovate-elliptic, margin serrate on the upper half, lower portion entire. Peduncles 1.5 cm. long, flowers axillary. Calyx 2. 5-3.0 cm. long, 1-1.5 cm. width at throat, cleft 2-3 mm. from top, pubescent. Petals 6.0-6. 5 cm. long, 1.0 cm. wide, bright red, pubescent on outer side only, oblong- spatulate. Staminal column 5 cm. long. Styles 8-10 mm. long. Capsule unknown. HOLOTYPE: Near Wahiawa Swamp, foot of Mt. Kahili, Wahiawa Mountains, Aug., 1909- C. N. Forbes 239. K. SPECIMENS EXAMINED: Kauai: Wahiawa Mountains, Lydgate. May 1909, Rock 40 ( 2701 ). Northwest facing slope, ridge 34 mile north of Laakahi, Koloa, 900 ft. altitude, moist gulch, December 24, 1947, St. John 23023. Oahu: In cultivation on Palolo Street, brought there from Kauai, Oct. 25, 1956, Roe 201. This species was described by Charles N. Forbes in 1911. It is closely related to H. Kokio Hillebrand, from which it differs by its pubes- cent calyx, smaller leaf, shorter petioles, shorter peduncle, longer yet narrower petals, shallower cleft, and longer style extension. This species is poorly known. Forbes claimed it was first collected by Lydgate several years before he made his type collection in August, 1909. Rock ( 1913^) cited Forbes’ species but did not give it lengthy treatment. Rock recently tried to find H. kahilii again on Kauai but was unsuccessful. In a letter dated August 15, 1958, from Albert Duvel, Kauai forester, we find H. kahilii still difficult to obtain. Duvel wrote, "I am not able to locate a place or specimen of H. kahilii The last record of it in its native environment was in 1947, when St. John (no. 23,023) collected the species on a ridge .75 mi. north of Laaukahi, Koloa, Kauai. It was in a moist gulch at 900 ft. altitude. This species has been found in cultivation in Honolulu. In the herbarium at the Bishop Museum there are several specimens of a plant closely allied to Hibiscus kahilii. These were collected by Forbes (643-K) on October 22, 1916, in the Hii Moun- tains, Kauai. This population is undoubtedly a variety of H. kahilii, as it resembles it in most characters. The stem surface of these specimens is distinct, and the leaves are larger and by far more pubescent. I am unable to completely describe this variety at present, as sufficient material is not at hand. It has not been reported since 1916. Fig. 14. Distribution of H. kahilii and H. Saint- johnianus: Solid circles, H. kahilii Solid triangles, H. Saintjohnianus 22 PACIFIC SCIENCE, Vol. XV, January 1961 10. Hibiscus Newhousei sp. nov. Fig. 15 DESCRIPTION: Arbor 6 meters aita. Stipulis 4-6 mm. longis. Petiolo 0.3-1. 8 cm. longo piloso. Lamina 4. 5-9-0 cm. longa 2. 5-4.0 cm. lata. Pedunculo 0.7-0.9 cm. longo articulato 2 mm. infra involucrum; bracteis lanceolatis 7- 8 mm. longis. Calyce 1. 8-2.0 cm. longo 1.0- 1.4 cm. lato, furcato 0. 3-0.7 cm. cum lobis acutis, 10-nervoso piloso. Petalis 5. 0-5. 2 cm. longis 1.3-1. 6 cm. latis, rubro colorato. Columna stam- inarum 3-0-3 .8 cm. apicibus filamentium liberis ca. 4 mm. Tree 6 m. tall. Stipules 4.0-6.0 mm. long. Petiole 0.3-1. 8 cm. long, pilose. Leaves 4.5-9.0 cm. long, 2. 5-4.0 cm. wide; margin irregular, apex distinctly acuminate. Peduncle 0.7-0.9 cm. long, articulate 2 mm. below the involucre. Lanceolate bracts 6-8, 7-8 mm. long. Calyx tubular 1. 8-2.0 cm. long, 1.0- 1.4 cm. wide, cleft for .3-7 cm., lobes acute, pilose, 10- nerved. Petals 5. 0-5. 2 cm. long, 1.3-1. 6 cm. wide, dark red, puberulent on outer surface. Staminal column 3. 0-3. 8 cm. long, slender, gla- brous, apex 5-lobed. Free filament tips extend ca. 4 mm. from column, located on the upper half of the column. HOLOTYPE: Moloaa Forest Reserve, Kauai; ca. 500 feet altitude. November 10, 1958. I. E. Lane 44 . Fig. 15. Solid circles, Hibiscus ivaimeae; solid tri- angles, Hibiscus Newhousei. This group is the most recent of the indig- enous hibiscus to be discovered. It had been re- ported several years ago and rediscovered in 1957. Mr. and Mrs. Jan Newhouse found it growing along the Moloaa Stream in the Mo- loaa Forest Reserve, northeastern section of Kauai. They brought it to cultivation at that time. A year later Irwin Lane made a second collection in its original habitat. Lane sent cut- tings to the Foster Gardens in Honolulu, where this species can now be found. One of its notable characters is its distinctly acuminate leaves. H. Newhousei shows closest proximity to the native H. Kokio , from which it differs by leaf shape and size, shorter pedun- cles, shorter bracts, smaller calyx, shallower clefts, shorter column, and other characters. Flowers are dark red and leaves are deep green on both surfaces. 11. Hibiscus immaculatus sp. nov. Figs. 16-18 DESCRIPTION: Arbor 3 meters alta basi 6 cm. in diametro. Petiolo 1.0-1. 5. Lamina 5-7 cm. longa 4-6.5 cm. lata ovati-obovata margine den- tate, nervis baud palmatis viride lactucae nervo medio subter pubescente stellato. Pedunculo 2- 3 cm. longo 2-3 mm. lato articulato 1 cm. infra involucrum; bracteis lanceolatis 5-8 mm. longis. Calyce 2. 5-3.0 cm. longo, 0.8-1. 0 cm. lato, fur- cato 4 mm., cum lobis acutis, 10-nervoso piloso flavi-virente. Petalis 8-11 cm. longis 2. 5-3. 5 cm. latis glabris albis. Columna staminarum 10- 14 cm. paulo papillosa apicibus filamentium liberis 10-20 mm. longis. Ramuli stylorum 1.5- 2.0 mm. longi erecti papillosi. Ovario 1.0 cm. longo 7-8 mm. lato. Tree 3 m. tall, diameter at base 6 cm. Petiole 1.0-1. 5 cm. Leaves 5-7 cm. long, 4-6.5 cm. wide, ovate-obovate, margin denticulate, veins scarcely palmate, "lettuce green,” midrib bears slight stellate puberulence on underside. Pedun- cle 2-3 cm. long, 2-3 mm. wide, articulate 1 cm. below involucre. Lanceolate bracts 6, 5-8 mm. long. Calyx 2. 5-3.0 cm. long, 0. 8-1.0 cm. wide, cleft for 4 mm., lobes acute, 10-nerved, pilose, "dull green yellow.” Petals 8-11 cm. long, 2. 5-3. 5 cm. wide, glabrous, white. Staminal col- umn 10-14 cm. slightly papillate. Free filament Hawaiian Hibiscus — Roe 23 Fig. 16. Hibiscus immaculatus Roe. Pali of Olokui above Waiehu, Wailau Valley, Molokai. Sept. 1912. C. N. Forbes 551. Mo. 24 PACIFIC SCIENCE, Vol. XV, January 1961 FIG. 17. Hibiscus immaculatus Roe. tips 10-20 mm., located on the upper third of column. Style branches 1. 5-2.0 mm., erect, pap- illate. Ovary 1.0 cm. long, 7-8 mm. wide. HOLOTYPE: Pali of Olokui above Waiehu, Wailau Valley, Molokai, September 1912. C. N. Forbes 5 51. Mo. specimens examined: Wailau Valley, Mo- lokai, C. N. Forbes 550.Mo. Hibiscus Gardens in Honolulu, introduced from Molokai, July 10, 1958. Roe 301. In September, 1912, Forbes made several col- lections of a distinct white hibiscus from the Pali of Olokui above Waiehu, Wailau Valley, Molokai. He noted that the staminal column was entirely white, which is not the case in our other Hawaiian white-corolled species. Forbes at first considered this a form of H. Arnottianus; in fact an early label read: "H. Arnottianus Gray Form nov.” Forbes, following subsequent study of his specimens of the plant, concluded it was a distinct species and noted this on his herbarium sheets. However, he did not leave any description of the new species. In 1913 Rock related that this "pure white flowered one occurs on the beach of Wailau Valley on Mo- lokai.” A letter written on July 16, 1958, by Henry Wiebke, Hoolehua, Molokai, confirmed the fact that the species continues to exist in the original locality. Mr. Wiebke wrote: "There is a native white hibiscus still growing in Wailau Valley in and around Kalae. Records show that these have been reported a number of times. This one is completely white.” The plant has also been found in cultivation in Honolulu, and in each case its origin has been traced to Molokai. It is attractive, but smaller and less showy than H. Arnottianus. Its rounded leaf of "lettuce green” is quite distinc- tive and this, together with the pure white corolla and column of the flower, makes it a desirable plant for cultivation. 12. Hibiscus waimeae Heller Figs. 15, 19, 20 Hibiscus waimeae Heller. Observation of the Ferns and Flowering Plants of the Hawai- ian Islands, Minn. Pub. of Botany (1897) 851, pi. 53. H. Arnottianus Gray forma. Mrs. Sinclair, Indigenous Flowers of Hawaiian Islands (1885), pi. 8. H. waimeae var. Helleri. Hochreutiner, Ann. Conserv. Jard. Bot. Geneve 4: 132, 1900. DESCRIPTION: Tree 7-8 m. in height; DBH, 16 cm. Stipules 5-6 cm. long, lanceolate. Petioles 3-4 cm. long, 2-3 mm. wide, puberulent. Blade 6-9 cm. long, 5-7 cm. wide, obovate-orbicular, surfaces entirely velvety pubescent, margin ser- rate, "forest green.” Peduncles 2-3 cm. long, articulate 6 mm. below involucre. Bracts 7, lan- ceolate, 1.5-2. 5 cm. long, pubescent. Calyx 3.5- 4.0 cm. long, 1.0-1. 5 cm. wide, widest at base of teeth, cleft 1-1.5 cm. from apex, teeth ovate- lanceolate, velutinous. Corolla white, pubescent on outer surface. Petals ca. 14 cm. long, 4-5 cm. wide (at greatest width), prominently veined. Staminal column stout, long exerted, FIG. 18. Distribution of Hibiscus immaculatus: Solid circles, represent localities of collections Open circles, represent areas where plants have been reported Hawaiian Hibiscus— Roe 25 FIG. 19= Hibiscus Waimeae Heller. Kokee region of the Na Pali Kona Reserve, altitude 3,200 ft., on bank of canyon. JuL 29, 1957. Roe 261. 26 PACIFIC SCIENCE, Vol. XV, January 1961 red. Filament tips extend approximately 2 cm. from column. holotype: Kaholuamanoa above Waimea, 900 m. September 2-9, 1895. A. A. Heller 2785 . ( Isotype examined. ) SPECIMENS EXAMINED: Kauai: Waimea Can- yon, January 12, 1956, H. F. Clay. Mountains back of Waimea, September 1909, C. N. Forbes 422. K. Waimea Drainage Basin, West side (hu- mid forest), July 3-August 18, 1917, C. N. Forbes 993. K. Halemanu February 20, 1909, Rock 39 (1559). Below Kaholuamano in gulches at 1800 feet, September 1909, Rock 5629. Octo- ber 1916, Rock 17093. Kokee region of Na Pali Kona Reserve, altitude of 3200 feet on bank of canyon, July 29, 1957, Roe 261. In cul- tivation at Kalaheo, altitude of 750 feet, July 27, 1957, Roe 252. In cultivation at the Hibis- cus Gardens of Honolulu, July 2, 1958, Roe 290. This species, although closely related to H. Arnottianus and unfortunately confused with it, is very different in numerous distinct characters, such as petals, column, stamens, calyx, leaves, tree shape, pubescence, and texture. Hillebrand considered all Hawaiian white-flowered hibiscus to be H. Arnottianus, Heller studied the situa- tion and felt that the white-flowered plant from Kauai was distinct and unnamed. He published its description in T897. In 1900, Hochreutiner reopened the case of H. Arnottianus. (A thor- ough discussion of this situation will be found in my treatment of H. Arnottianus.) To the large Oahu white, Hochreutiner attached the name of H. waimeae (Heller) var. Hookeri Hochr., and to the real H. waimeae of Heller he gave the name H. waimeae var. Helleri Hochr. Hochreutiner gave the name H. Arnott- ianus to what is known today as H. Kokio Hbd., thus establishing two varieties. T. A. Sprague in 1914 also added to the confusion. He sug- gested the name H. Arnottianus be dropped com- pletely and replaced by H. waimeae. Skottsberg in 1926 concluded that the treatment given by Hochreutiner and Sprague is ". . . inadmissible and has resulted in a more and more hopeless confusion....” He continued, "...to replace 'Arnottianus’ with 'Waimeae’ Heller, to dis- tinguish two varieties of this and to call Heller’s type var. Helleri is wrong for the simple reason FIG. 20. Hibiscus waimeae Heller. that 'Arnottianus’ Hillebrand and 'waimeae’ are different species.” The major differences between H. Arnottianus and H. waimeae are these: H. Arnottianus is nearly glabrous throughout, while H. waimeae has a velvety pubescence. The petioles of the Kauai species are almost twice as long as those of the Oahu plant; their calyces are also longer and wider. The calyx cleft of H. Arnottianus is only 5 mm. deep and that of H. waimeae is 1-1.5 cm. deep. The staminal column of H. ivaimeae is stouter than that of the Oahu species. 13. Hibiscus Arnottianus Gray Figs. 21-23 Hibiscus Arnottianus Gray. Bot. U. S. Expl. Exped. 1: 176. 1854. H. Boryanus H. & A. Bot. Beechey Voy., p. 79, partim, non DC. 1832. H. Waimeae var. Hookeri Hochreutiner. Ann. Conserv. Jard. Bot. Geneve, 4: 132. 1900 . H. Fauriei Leveil. Fedde Repert. 10: 120. 1911. description: Tree growing to 35 ft., well branched, nearly glabrous. Stipules subulate, ca- ducous. Petioles 0. 5-2.0 cm. long, glabrous. Blades 8-10 cm. long, 5-7 cm. wide, ovate to elliptical-parabolical, apex acute-acuminate, chartaceous, margin sinuately crenate to entire, Hawaiian Hibiscus — Roe 27 Fig. 21. Hibiscus Arnottianus Gray. North end at junction of Manoa Cliff and Pauoa trails; 1,300 ft., Tantalus, Manoa. (Tree 8 m. X 2 dm.), Apr. 17, 1957. St. John 26053. 28 PACIFIC SCIENCE, Vol. XV, January 1961 scarcely palmately veined with 3 strong nerves and 2 less prominent nerves branching from base of blade. Flowers solitary in axils of leaves. Involucre with 5—7 linear-lanceolate bracts 0.7— 1.0 cm. long. Calyx tubular, 2-3 cm. long, 0.8- 1.0 cm. wide, 3 acute lobes, cleft 5 mm. from apex, glabrous to slightly pilose. Corolla white often with a slight pink tinge. Petals 7-12 cm. long, 2-3 cm. wide. Staminal column long exerted, 10-15 cm. long, red, free filament tips 1.0— 1.8 cm. long, from upper half of column. Style branches erect, 5-8 mm. long. Capsule chartaceous, 2-3 cm. long. Seeds reniform, 4- 5 mm. tomentose, dark brown. SPECIMENS EXAMINED: Oahu: North slope of Mt. Tantalus (tree 6 m. tall; 45 cm. DBH), June 15, 1930, Christopherson 1373. Beside stream in open forest, Wailupe, Kului, June 23, 1937, F. E. Egler No. 37-68. Konahuanui, Jan. 6, 1909, C. N. Forbes 1000.0. East side of Nuu- anu Valley, Oct. 1910, C. N. Forbes 1601.0. Waialae Valley, October 15, 1914, C. N. Forbes 1948.0; also May 4, 1914, Forbes 2496.0; and January 1919, Forbes 2522.0. Manoa Cliff Trail, April 11, 1920, Gerber D. Wesley 372. "Hau- hele” Hillebrand. Niu, Hillebrand and Lydgate. Tantalus Ridge, September 5, 1909, H. L . Lyon. "Oahu” H. Mann and W. T . Brigham 530 (2 sheets). Mt. Tantalus, July 8, 1922, C. Skotts- berg 84. North end at junction of Manoa Cliff and Pauoa Trails, 1300 feet, Tantalus, Manoa (tree 8 m. x 2 dm.), April 17, 1957, St. John 26053. There has been so much confusion and con- troversy regarding Gray’s H. Arnottianus that a presentation of its history is necessary. His original description (1838) of H. Arnottianus reads: H. fructicosus, glaberrimus; foliis ovatis sen ovalibus subcoriaceis integerrimis (nunc sub- dentates) basi tri-nervatis; floribus solitariis pedunculatis; involucelli phyllis 5-7 parvis de- ciduis; petalis (rubis) oblongis basi attenuatis atque in tubum gracilem calyce cylindrico long- iorem coalitis; columna staminea proelonga spit- hamoea; capsula polysperma. Hibiscus Arnottianus, Gray in herb. Hook anno 1837 H. Boryanus, Hook & Arn Bot. Beech. Voy. non DC Hab. Sandwich Islands; on the Kaala Moun- tains behind Honolulu, Oahu; where it was gathered by Macrae, Lay & Collie, Diell, Barclay, etc. (Byron’s Bay, Hawaii; Macrae, Diell.) This description includes a red-flowered hi- biscus having the long staminal column char- acteristic of the white-flowered Oahu plant. There are at least two distinct species included in Gray’s description of H. Arnottianus. Hillebrand described as new the small red- flowered hibiscus, calling it H. Kokio Hbd. He stated his belief that the red-flowered hibiscus from Byron’s Bay, referred to by Hooker and Arnott under H. Boryanus and considered by Gray as the H. Arnottianus from "Byron’s Bay, Hawaii, Macrae,” belonged to the taxon H. Kokio. Hillebrand applied the name H. Arnottianus to the plant collected during Beechey’s voyage and distinguished it from his newly described H. Kokio. It is presumed that Hillebrand studied the types. Heller in 1897 studied both plants in ques- tion and concluded that Gray’s type for H. Ar- nottianus was probably collected by Diell at Byron’s Bay and that it had red flowers. Heller at that time was in a position to designate the type but did not do so. Gray mentioned first "the Kaala mountains behind Honolulu” and secondly, in brackets, "Byron’s Bay, Hawaii, Macrae, Diell.” Heller felt that the white- FlG. 22. Hibiscus Arnottianus Gray. Hawaiian Hibiscus—- Roe 29 flowered Kauai hibiscus was distinct from both H. Arnottianus and H. Kokio, and he described it as H. waimeae. In 1900 Hochreutiner published his revision of the genus Hibiscus, including the Hawaiian species. He presented a new interpretation of H. Arnottianus. While Hillebrand and Heller had applied Gray’s description to the plant col- lected during Beechey’s voyage, Hochreutiner felt that the name H. Arnottianus belonged to the plant collected by Diell, called by Hillebrand H. Kokio. Hochreutiner classified the plants in question as follows: H. Waimeae Heller var. Hookeri Hochr. var. Heller i Hochr. (white-flowered species) (Beechey plant) (Kauai white) H, Arnottianus Gray (red-flowered species) var. Kokio Hochr. (Hillebrand’s H. Kokio ) var. genuinus Hochr. (Byron’s Bay- red- Macrae, Diell) Rock (1913^) makes no mention of Hoch- reutiner ’s revision, which leads one to believe that he was not familiar with it. He was aware, however, of the original problem regarding H. Arnottianus and, to verify his classification, he sent a specimen of the white-flowered H. Ar- nottianus to the Gray Herbarium, where B. L. Robinson made comparisons and replied, "There can be no question that the white-flowered species (No. 8831) from Oahu is precisely the real H. Arnottianus Gray.” As far as Rock was concerned the case of H. Arnottianus was set- tled. However, in 1914, T. A. Sprague began again to dispute its nomenclature. He wrote in the Kew Bulletin, "So much confusion has arisen in the past in connection with the name H. Arnottianus that it is perhaps desirable to abandon the use of it altogether.” Sprague claimed that Gray had a red-flowered species in mind when describing H. Arnottianus, for he sent a specimen of this species collected by Diell to Sir William Hooker under the name H. Ar- nottii Gray (later altered to Arnottianus) . How- ever, Sprague disregarded the name H. Arnot- tianus entirely and retained H. Kokio Hbd. for the red-flowered species. Skottsberg (1926), in reviewing the nomen- clatural history of H. Arnottianus, exclaimed, "The confusion is very complete.” His conclu- sion, however, is very sound and offers the only logical solution to the problem. He rebuked Hochreutiner and Sprague for their "inadmis- sible” treatment given to the species in ques- tion, and joined Rock in his view of the situa- tion. It is concluded that Gray had two species on hand when he described H. Arnottianus — the white-flowered Oahu hibiscus bearing the long staminal column, and the small red-flowered plant. The red plant was the material sent to Sir William Hooker and the white-flowered species was retained in the Gray and the U. S. National Museum herbaria. This white species compares to our Oahu white according to the judgment of Dr. Robinson. In a recent letter, Reed C. Rollins, director of the Gray Herbarium, also testified that H. Arnottianus from the Wilkes Expedition is "white-flowered, with a very long, apparently red, staminal column.” Likewise, Richard S. Cowan of the U. S. National Museum, in a letter dated June 28, 1958, confirmed this belief regarding the mate- rial deposited there. Cowan wrote: We have one sheet bearing one flowering branch collected by the U. S. Exploring Ex- pedition; the label is the standard label for this Expedition but the locality is in long-hand and reads: "Oahu, S. Islands.” The only other nota- tion is the name "Hibiscus Arnottianus Gray.” I have compared both the quoted items with a scrawl of Gray’s in our holograph collection and I feel certain that they were made by the same person. There is no question in my mind that the specimen is of the white-flowered element, for red-flowered H. Kokio clearly shows red flowers even in the dried condition. The length of the staminal column in one of the two flowers on our specimen is 10.5 cm. In most respects, this specimen appears very similar to the following collections in our herbarium as- signed to H. Arnottianus: Mann & Brigham 530 and Christopherson 1373. 30 PACIFIC SCIENCE, Vol. XV, January 1961 Skottsberg agreed that Hillebrand could re- tain either of the two species as H. Arnottianus and then describe the other as new (for Gray did not designate a type ) . Hillebrand considered the white-flowered Oahu plant as H. Arnottianus and described the red as H. Kokio. According to the international rules of nomenclature, the solution worked out by Hillebrand must be ac- cepted and is correct. We have the testimony of Rock and Skottsberg to this. The full syn- onomy given by Skottsberg is correct: H. Arnottianus ( A. Gray p.p. ) Hbd. s. str. ( ex- cluding Kauai specimens H. waimeae ) ; Rock, non Hochreutiner. Syn: H. waimeae (Heller) var. hookeri Hochr.; T.A.S.; H. fauriei Levi. H. waimeae Heller; Rock. Syn: H. Arnottianus Hbd. quoad spec. kauaiens H. Arnottianus forma, Mrs. Sinclair, pi. 8. H. waimeae (Heller) var. helleri Hochr. H. Kokio Hbd; Rock, T.A.S. Syn: H. Arnottianus Gray; Hochreutiner; non Hbd., Rock. H. Arnottianus forma, Mrs. Sinclair, pi. 9. FIG. 23. Distribution of H. Arnottianus popula- tions: Solid circles, H. Arnottianus Open circles, f. parviflora Solid triangles, var. punaluuensis H. Kokio Hbd; Rock, T.A.S. Syn: H. Arnottianus Gray pro parti; Hoch- reutiner; non Hbd., Rock. H. Arnottianus forma, Mrs. Sinclair, pi. 9. We conclude from this discussion that we have three distinct species: the two whites, H. Arnottianus and H. waimeae; and one red, H. Kokio. Hillebrand effectively restricted appli- cation of H. Arnottianus to a white by naming the red. 14. Hibiscus Arnottianus f. parviflora Skotts- berg Fig. 23 Hibiscus Arnottianus f. parviflora Skottsberg. Vascular Plants from the Hawaiian Islands, IV. Acta Hort. Gotoburg. 15: 396. 1944. DESCRIPTION: Tree to 30 ft.; well branched. Stipules subulate, caducous. Petioles 0.5—1. 5 cm. long. Leaves 4-6 cm. long, 4-5 cm. wide; cren- ulate to entire, glabrous to slightly stellate-pilose on the underside of leaf. Bracts 5-7, linear-lan- ceolate, 5-8 mm. long. Calyx tubular, 1-2 cm. long, 0.5-1. 1 cm. wide, slightly pilose. Petals 6-6.5 cm. long; 2. 5-3-0 cm. wide. Staminal col- umn long exerted, 8-12 cm. long, red, free fil- ament tips extend 1.0-1. 5 cm. from column, style branches erect, 5-8 mm. long. HOLOTYPE: One mile south of Kolekole Pass, Waianae Mountains. (Hawaiian Bog Survey) September 3, 1938, O. Selling 3364 . SPECIMENS EXAMINED: Oahu: Palehua, Wai- anae Range, C. M. Cooke. Kupehau Gulch, ele- vation 1800 feet, Waianae Range, October 13, 1946, Cowan 163. Kukuiula, Mokuleia, Waianae Range (Forest Trail N. 20a) elevation 1200 feet, tree 30 feet, William H. Hathaway 163 . Puu Hapapa, Waianae Mountains, elevation 2400 feet, October 19, 1930, E. Y. H osaka 323. Ekahanui Gulch, Waianae Mountains, elevation 1200 feet, September 28, 1947, Yoshio Kondo . Keekee Gulch, in unfenced forest reserve, eleva- tion 1500 feet, October 13, 1941, Korte. Ma- kaha Valley, Waianae Mountains, January 31, 1927, MacDaniels 369. Waikane Mauka Trail on the Waianaeuka-Wahiawa Divide, elevation 1250 feet, March 15, 1958, Roe 266. Kealia, Hawaiian Hibiscus — Roe 31 elevation 1500 feet, edge of woods; tree 8 m. X 25 cm., November 8, 1936, St. John 17,655* Mokuleia Trail, Waianae Range, elevation 2200 feet, tree 6 m., January 11, 1948, Webster l L. radyJISUr- ^ X Grat ~~T his population was brought to our attention in 1944 when Skottsberg published it as a form of H. Arnottianus Gray. Skottsberg felt that the variation was slight and showed principally in the smaller flower. Resides the flower difference, the leaves, calyx, and bracts of this form appear smaller than those of the species. 15. Hibiscus Arnottianus Gray var. punaluuen- sis Skottsberg Fig. 23 Hibiscus Arnottianus Gray var. punaluuensis Skottsberg. Vascular Plants from the Ha- waiian Islands, IV. Acta Hort. Gotoburg. 15: 396. 1944. H. punaluuensis (Skottsb. ) Deg. & Deg. Flora Hawaiiensis 5: Family 221. 1957. DESCRIPTION: Tree to 35 ft. Stipules subulate, caducous. Petioles densely puberulent, 3-12 cm. long, scarcely palmately veined ( 5 radiating ribs), veins strongly pubescent. Blade 10-25 cm. long, 8-20 cm. wide with puberulent-pilose sur- face, ovate, base subcordate, apex acuminate. Bracts 5-7, linear-lanceolate, 1. 0-2.0 cm. long, pilose. Calyx scabrous, 1.8-2. 5 cm. long, 1.0—1. 5 cm. wide, cleft 5 mm. Petals white, 8-12 cm. long, 2.5-4 cm. wide, stellate pubescent on outer surface. Staminal column red, 12—19 cm. long, free filament tips 1.5-2. 5 cm. long, red. Style branches 1.0 cm. long. HOLOTYPE: Punaluu, Koolau Mountains, Oahu (Hawaiian Bog Survey) Sept. 27, 1938. 0. Selling 3638. (Isotype examined; type not seen. ) SPECIMENS EXAMINED: Oahu: Edge of Ka- luanui Stream, Castle Trail, December 22, 1940, E. H. Bryan, Jr. 1501. Between Punaluu and Kaipapau, Koolau Mountains, November 14-21, 1908, C. N. Forbes. Near Head Gate, Wahiawa, August 17-20, 1915, C. N. Forbes 2198.0. Puna- luu, elevation 2000 feet, September 28, 1930, E. Y. H osaka 304. Kaluanui River Bank, Koo- lau Mountains, elevation 2800 feet, September 28, 1930, Inafuku. Upper Kaluanui Valley, Castle Trail, October 21, 1945, R. Kuykendall 117. Waiahole Ditch Trail, June 1932, A. Mee- hold. Near Mt. Stream, Koolau Mts., Nov. 14-21, 1908, Rock 37 (309) (duplicate). In cultivation in Wilder’s Garden, J. F. Rock. Along Puunahia streams near Castle Camp, tree 30 feet, August 1911, Rock 8831. Kaluani, Castle Trail, Koolau Range, November 14, 1955, Roe 18. Kaluanui, by stream 1950 feet altitude, rain forest, tree 7 m. X 15 cm., November 25, 1956, St. John 25977. Skottsberg in 1944 established this new vari- ety, punaluuensis, and called it after the locality where the type was collected. It differs from the species by a characteristic venation: between the base with the five radiating ribs and the next strong pair (which forms an angle of 45 degrees with the midrib) is a wide stretch with only faint veins issuing at an angle of about 90 degrees. The leaves are longer and wider than those of the species, the petiole being approxi- mately 5 cm. longer. The veins are strongly pubescent. The free filament tips and style branches are longer in the variety. The calyx is conspicuously hairy. In 1957 Otto Degener raised this population to the status of a species. These plants differ from H. Arnottianus but show such similarity that I hesitate to agree with Degener. The dif- ferences are not so great as they are between the other taxa which have been accorded specific rank. Skottsberg’s treatment seems more satis- factory. REFERENCES Gaum, Edward L. 1930. New Hawaiian plants. Bishop Mus. Occ. Pap. 9(5). Degener, Otto. 1957. Flora Hawaiiensis, Book V. Honolulu. Forbes, Charles N. 1912. New Hawaiian plants, III. Bishop Mus. Occ. Pap. 5(1). Gaudichaud, Charles. 1826. Botanique du Voyage de 1’Uranie. Paris. Gray, Asa. 1854. Botany Phanerogomia, Vol. I. U. S. Expl. Exped. During the Years 1838. ...George P. Putnam & Co., New York. Heller, A. A. 1897. Observations of the ferns and flowering plants of the Hawaiian Islands. Minn. Pub. Bot. 32 Hillebrand, William F. 1888. Flora of the Hawaiian Islands. Heidelberg. Hochreutiner, B. P. G. 1900. Revision du Genre Hibiscus. Ann. Conserv. Jard. 4. Leveille, H. 1912. Plantae novae sandwicenses. Repert. Spec. Nov. Regn. Veg. 12. Berlin. Linnaeus. 1754. Genera Plantarum. 5th ed. Stockholm. Mann, Horace. 1867. Enumeration of Ha- waiian Plants. Proc. Amer. Acad. Arts Sci. 12 . Nakasone, Henry Y. 1953. Ornamental Hi- biscus. Hawaii Agr. Exp. Sta. 37. PACIFIC SCIENCE, Vol. XV, January 1961 Rock, Joseph F. 1913*. The Indigenous Trees of the Hawaiian Islands. Honolulu. 1913 b. New species of Hawaiian plants. Coll. Hawaii Bull. 2: 48-49. St. John, Harold. 1954. Review of Mrs. Sin- clair’s 'indigenous flowers of the Hawaiian Islands.’ Hawaiian Plant Studies 23. Pacif. Sci. Sinclair, Mrs. Francis, Jr. 1885. Indigenous Flowers of the Hawaiian Islands. London. Skottsberg, C. 1926. Vascular plants from the Hawaiian Islands, I. Acta Hort. Gotoburg. 1944. Vascular plants from the Hawai- ian Islands, IV. Acta Hort. Gotoburg. Sprague, T. A. 1914. Curtis’ Bot. Mag. 140. A Collection, of Porifera from Northern New Zealand, with Descriptions of Seventeen New Species Patricia R. Rergquist 1 The material described herein has been collected by the author during the course of in- vestigations into the intertidal and sublittoral ecology of the sponges of Northland. Burton (1932), in reporting the "Discovery” Antarctic sponges, remarked on the relatively advanced state of knowledge of that fauna and added only 35 new species from this collection. In strong contrast stands the lack of knowledge of the littoral sponge faunas of New Zealand, particularly of the siliceous groups. The present collection contains 34 species, all relatively com- mon members of the northern shore faunas. Of this number, 17 are new species, and 11 are recorded for the first time from New Zealand. No Calcarea are included in the present study. They are, in general, better known than the Demospongiae, as most of the published work by Kirk dealt with this group. Ail type material is to be deposited in the Dominion Museum, Wellington. SYSTEMATIC DISCUSSION The scheme of classification followed is that of de Laubenfels ( 1936) . class DEMOSPONGIAE (Sollas) ORDER KERATOSA (Grant) family SPONGIIDAE (Gray) GENUS Ircinia ( Nardo ) Ircmia novae zealandiae , sp. nov. Fig. la OCCURRENCE: Noises Islands, Hauraki Gulf. 12/10/56. 2 description: The sponge Is irregularly palmo-digitate in shape, the surface irregularly conulose. Oscules are small, 1-2 mm. in diam- 1 Department of Zoology, University of Auckland, New Zealand. Manuscript received August 7, 1959° 2 In this paper, dates are given as day/ month/ year. eter, few in number, and indiscriminately scat- tered. The texture when dried is hard and the colour blackish-brown. The texture when fresh Is elastic and the colour blackish. The skeleton is a reticulation of strongly fasciculated fibres showing obvious differentia- tion into ascending and connective fibres. These fibres vary from .02 to 1.0 mm. in diameter. Foreign inclusions are occasionally present, sometimes filling the whole of a fibre. Filaments are not common and are .004 mm. in diameter. GENUS Spongia (Linnaeus) Spongia reticulata (Lendenfeld) Eu spongia reticulata (Lendenfeld, 1386, p. 541). Hippospongia reticulata (Lendenfeld, 1889, p. 300, pi. 13, fig. 3). OCCURRENCE: Rangitoto sublittoral fringe. DISTRIBUTION: Australia. FAMILY DYSIDEIDAE (Gray) GENUS Dysidea (Johnston) Dysidea cristagalli, sp. nov. Fig. lb OCCURRENCE: Noises Islands, 2/5/37 (coll L. B. Moore). Rangitoto, 7/6/57. In rock pools in caves. DESCRIPTION : The sponge is erect, tubular in shape, with several tubes coalescing to give a tubula-flabellate condition. The surface is uneven and the oscules apical, giving access to deep cloacae. The texture is firm and friable, the colour ash-grey. The skeleton is an irregular, closely knit reticulation of fibres varying in diameter from .02 to .2 mm. and having no obvious distinction between ascending and con- nective fibres. The fibres are filled with broken sponge spicules. There is no special dermal skel- eton differentiated. 33 PACIFIC SCIENCE, VoL XV, January 1961 34 FlG. L a, Ircinia novae zealandiae, sp. nov. b, Dy- sidea cristagalli, sp. oov. order HAPLOSCLERINA (Topsent) family HALICLONIDAE (de Laubenfels) GENUS Haliclona (Grant) Halicloma isodictyale, sp. nov. Fig. 2a, b OCCURRENCE: Waitawa Bay, Clevedon, 31/ 10/35 (coll L. B. Moore) . Waitawa Bay, Cleve- don, 2/10/58. Point Chevalier Reef, 6/7/57. DESCRIPTION: This sponge is an encrusting form with a maximum thickness of 3 mm. Its surface is minutely shaggy and hispid. The oscules are apical on tubular processes, as is characteristic of many haliclonids. The texture is soft and friable. In life the colour is pale cream, in spirit pale brownish-white. The skele- ton is a subregularly-isodictyal reticulation, mainly unispicular. Spicules . Megascleres: oxea .13 mm. X .007 mm. Microscleres: absent. Haliclona tenacior , sp. nov. Fig. 3 a, b OCCURRENCE: Waitawa Bay, Clevedon, on rocks or seaweed. Rangitoto, on rocks. DESCRIPTION: An encrusting to irregularly massive sponge infested with commensal poly • chaete worms. Its surface is uneven, minutely and irregularly subpapillose. Oscula are few and inconspicuous. Texture is firm and friable. The colour in life is dirty-muddy cream, in spirit Fig. 2. a, Haliclona isodictyale, sp. nov. b, Hali- clona isodictyale, typical oxea (X 350). New Zealand Porifera— Bergquist 35 FIG. 3= a, Haliclona tenacior , sp. nov. b, Haliclona tenacior, typical oxea (X 350). light greenish-brown. The skeleton is a closely knit reticulation of systems of fibres ranging from unt- to multispicular. Spicules . Megascleres: oxea .14 mm. X .007 mm. Microscleres: absent. Haliclona glabra f sp. nov. Fig. 4 OCCURRENCE: Stanley Bay, under sides of boulders at low tidal levels. DESCRIPTION : A thinly encrusting sponge with even, minutely hispid surface. The oscules are few, minute, and scattered. Texture is firm and friable. Colour in life is dull cream, in spirit pale brownish-white. The skeleton is an iso- dictyal reticulation, the ascending fibres of which are triangular. Connectives are unispicular. Spicules. Megascleres: oxea .152 mm. X -007 mm. Microscleres: absent. Haliclona heterofbrosa (Lundbeck) Renter a heterofibrosa Lundbeck, 1902, p 47, pis. 2, 11, figs. 8, 14; Br0ndsted, 1923, p. 121; Hentschel, 1929, p. 983. OCCURRENCE: Rangitoto Island. Point Che- valier Reef. Under sides of stones up to half tide. REMARKS: Quite typical specimens, soft tex- ture, ranging in colour from creamy yellow to faintly purple. DISTRIBUTION: Arctic; Campbell Islands Sub- antarctic. Haliclona clathrata (Dendy) Reniera clathrata Dendy, 1895, p. 237; Brpndsted, 1923, p. 125; 1924*, p. 453); ? R. spec. 4, Hentschel, 1912, p. 410. OCCURRENCE: Karaka Bay (St. Hellers). REMARKS: Single specimen: collet. ted agrees so closely in external form with the holotype of H . clathrata (Dendy) that there seems every reason to refer it to this species. Structure of the skeleton is closely similar. AUSTRA- LIA N. Z. CAMPBELL IS. N. Z. (Br.) Oxea .083 .157 O 00 \j\ J ITS o 2 © X X X X .005 mm. .007 mm. .005 mm. .004 to .006 mm. This species appears to be a typical haliclonid insofar as the spicule sizes vary from specimen Fig. 4. Haliclona glabra, sp. nov. 36 PACIFIC SCIENCE, Vol. XV, January 1961 to specimen. Small variations in this respect cannot be considered grounds for differentiating the present specimen from H. clathrata. DISTRIBUTION: Australia (south coast); New Zealand; Campbell Islands. Haliclona petrosioides (Burton) Haliclona petrosioides Burton, 1932, p. 269, fig. 7. OCCURRENCE: Rangitoto sublittoral fringe. remarks: Specimens are thinly encrusting, up to 3 mm. thick, deep cream in colour. Sur- face conspicuously porose with conspicuous channels radiating in a stellate manner from small ill-defined oscules. The spiculation is iden- tical with that of the holotype. DISTRIBUTION: Tristan da Cunha (South At- lantic). family CALLYSPONGIDAE (de Laubenfels) GENUS Callyspongia (Duchassaing & Michelotti) Callyspongia ramosa (Gray) Synonymy: See Burton, 1934, p. 603. OCCURRENCE: Noises Islands; Kawau Island; Narrow Neck. REMARKS: An erect branching sponge, purple in life. Commonly washed up on Auckland beaches. DISTRIBUTION: New Zealand; Australia; Ant- arctica. order POECILOSCLERINA (Topsent) group PHORBASIFORMES (de Laubenfels) family PHORBASIDAE (de Laubenfels) GENUS Phorbas (Duchassaing & Michelotti) Phorbas intermedia, sp. nov. Fig. 5a, b OCCURRENCE: Rangitoto; Karaka Bay (St. Helier’s ) . Fig. 5. a, Phorbas intermedia, sp. nov. b, Phorbas intermedia, acanthotornota, acanthostyles, isochelae, sigmata, raphides (X 350). DESCRIPTION: An irregularly massive sponge with an uneven to minutely papillate surface. Oscula are not apparent. The texture is firm. Colour in life is a rich yellow, in spirit a pale yellow. The skeleton, a confused system of as- cending fibres branching and anastomosing, is composed of bundles of acanthostyli of two sizes and tornota. There is a dermal palisade of brushes of tornota. Spicules . Megascleres: (a) acanthostyles .17 to .22 mm. X .005 to .007 mm.; (b) acantho- styles .09 X .004 mm.; (c) tornota with sub- tylote microspined ends — .15 X .004 to .006 mm. There are numerous intermediates between acan- thostyli and tornota. Microscleres : (a) arcuate isochelae .04 X .045 mm. chord; (b) sigmata .017 X .021 mm. chord; (c) raphides .04 X 07 mm. long. New Zealand Porifera— BERGQUIST 37 family ADOCIIDAE (de Laubenfels ) GENUS Adocia (Gray) Adocia parietalioides , sp. nov. Fig. 6a, b OCCURRENCE: Rangitoto Island; littoral. DESCRIPTION: An encrusting sponge with even, porose surface. Oscules small, 1-2 mm. in diameter, few in number, and level with the surface. Texture firm and friable. Colour in life Fig. 6. a, Adocia parietalioides , sp. nov. b, Adocia parietalioides , typical oxea (X 350). faintly pink, in spirit pale brownish white. Main and dermal skeletons unispicular. Spicules. Megascleres: oxea .15 X .01 mm. Microscleres: absent. REMARKS: This species is most closely related to A. parietalis (Topsent) from the Mediter- ranean, from which species it differs in: (a) the absence of subdermal canals radiating from the oscules; (b) the character of the undersurface of the free edges (which in A. parietalis "releve . . . et y developpe a sa face inferieure un feutrage plus serre des spicules ...”); (c) the size of the oxea, which are .01 mm. in thickness, as opposed to .005 mm. in A. parietalis. Adocia venustina, sp. nov. Fig. -la, h OCCURRENCE: Noises Islands, in mouth of cave; Rangitoto, in mouth of cave. DESCRIPTION: A sponge ranging in habit from encrusting to massive and depressed. Sur- face is uneven and glabrous. Oscules are numer- ous, subpapillate, 1-3 mm. in diameter. This sponge is of firm, friable texture. Colour in life yellow, in spirit dull yellowish brown. The main skeleton is isodictyal or sub-isodictyal, chiefly unispicular, but having occasional bispicular as- cending fibres. The dermal skeleton is unispic- ular with mainly triangular mesh. Spicules. Megascleres: oxea .1 X .004 mm. Microscleres: absent. GENUS Toxadocia (de Laubenfels) Toxadocia toxophorus (Hentschel) Gellius toxophorus Hentschel, 1912, p. 392, pi. 21, fig. 46. Gellius toxotes Hentschel, 1912, p. 392, pi. 21, % 47. OCCURRENCE: Rangitoto sublittoral fringe. REMARKS: The species is represented by cushion-shaped masses up to 1 cm. thick, pale cream in colour, with firm uneven surfaces. The oscules are scattered, 1-2 mm. in diameter. Skeleton is a loose subhalichondroid reticula- tion, mainly unispicular, of oxea tending to be 38 strongylote, .24 X .014 mm., and with toxa .035 to .052 mm. long, for microscleres. The old genus Gellius has been split by de Laubenfels (1936) into several genera and the present species is referred to Toxadocia. distribution: Aru Islands (Malay Archi- pelago), 4-6 fathoms. GROUP MICROCIONIFORMES (de Laubenfels) family MICROCIONIDAE (Hentschel) GENUS Microciona (Bowerbank) Micro ciona coccinea, sp. nov. Fig. 8 a, b OCCURRENCE: Stanley Bay, under sides of stones at low tide. Onetangi, Waiheke Island, on roof of cave at low tide. DESCRIPTION: A thinly encrusting sponge with uneven surface over which oscula are not apparent. Texture is soft. Colour when alive is scarlet, in spirit rusty brown. The skeleton is low plumose columns of basally spined styli and of acanthostyli. Scattered loose between the columns and forming a loose tangential layer in the dermis are auxiliary subtylostyli. Spicules. Megascleres: (a) basally spined styli .24 to .4 X .012 to .016 mm.; (b) acanthostyli .08 to .21 X .007 to .014 mm.; (c) subtylostyli .14 to .28 X .003 to .004 mm. Microscleres: toxas .05 to .07 mm. chord. Microciona rubens, sp. nov. Fig. 9 a, b OCCURRENCE: Waitawa Bay, Clevedon, en- crusting rocks at 2 fathoms. DESCRIPTION: An encrusting sponge with an uneven, minutely and irregularly mammilate surface on which oscules are not apparent. Tex- ture firm. Colour in life vermillion red, in spirit pale brownish yellow to dark brown. The skele- ton is made up of plumose columns of basally spined styli (rarely completely smooth) and of acanthostyli. Auxiliary subtylostyli are asso- ciated with these columns and form an irregular tangential layer in the dermis. PACIFIC SCIENCE, Vol. XV, January 1961 Fig. 7. a, Adocia venustina, sp. nov. b, Adocia venustina, typical oxea (X 350). Spicules. Megascleres: (a) basally spined styli .14 to .53 X .021 mm.; (b) acanthostyli .09 X .011 mm.; (c) subtylostyli .11 to .28 X .003 to .005 mm. Microscleres: (a) palmate isochelae .008 mm. chord; (b) toxa .04 to .07 mm. chord. REMARKS: This species is most closely related to Clathria mortens enii ( Brpndsted ) , which has been transferred to Microciona by de Lauben- fels (1936). Spicule dimensions in the present specimens differ widely from those described by Brpndsted. New Zealand Porifera — BERGQUIST 39 Microciona heterospiculata (Br0ndsted) M. heterospiculata Br0ndsted, 1924^, p. 465, fig. 20. OCCURRENCE: Point Chevalier Reef; Karaka Bay; Stanley Bay. Up to half tide encrusting rocks. DISTRIBUTION: New Zealand. Fig. 8. a, Microciona coccinea , sp. nov. b, Micro- ciona coccinea, acanthostyle, basally spined style. FIG. 9- a, Microciona rubens, sp. nov. b, Micro- ciona rubens, spined style, acanthostyles, subtylostyle, isochelae (X 350). family OPHLITASPONGIIDAE (de Laubenfels) GENUS My cate (Gray) Mycale rara (Dendy) Esperella rara Dendy, 1896, p. 18. OCCURRENCE: Ahipara Bay. REMARKS: The one specimen, yellow and soft in life, agrees closely with Dendy’s original speci- men, except that trichodragmata appear to be absent. Occasional isochelae .017 mm. chord occur. distribution: Australia. 40 PACIFIC SCIENCE, Vol. XV, January 1961 family AMPHILECTIDAE (de Laubenfels) GENUS Biemna (Gray) Biemna rhabderemioides , sp. nov. Fig. 10^, b OCCURRENCE: Rangitoto, under stones at about half tide. DESCRIPTION: An encrusting to massive and low-lying sponge, the surface of which is mi- nutely reticulate, conulose, and hispid. The oscules are not apparent. Texture is firm. Colour in life bright yellow, in spirit pale yellowish- brown. The skeleton is an irregular reticulation of styli and subtylostyli bent in the basal third and crooked at the proximal end. FIG. 10. a , Biemna rhabderemioides, sp. nov. b, Biemna rhabderemioides, subtylostyli, raphides (X 350). Fig. 11. a, Halichondria moorei, sp. nov. b, Hali- chondria moorei, oxea (X 80 ). Spicules. Megascleres: (a) styli .42 to .47 X .01 to .016 mm.; (b) subtylostyli .42 to .48 X .01 to .016 mm. Microscleres: (a) sigmata (two sizes) .042 to .045 and .012 to .014 mm. chord; (b) raphides .09 mm. long; (c) microxea .05 X .003 mm. order HALICHONDRINA (Vosmaer) family HALICHONDRIIDAE (Gray) GENUS Halichondria (Fleming) Halichondria moorei, sp. nov. Fig. 1 la, b OCCURRENCE: Point Chevalier Reef; Karaka Bay ( St. Helier s ) ; Parnell Reef. DESCRIPTION : A massive, somewhat depressed sponge with surface wrinkled and irregularly folded. Oscula not apparent. Colour alive pinkish- New Zealand Porifera — BERGQUIST 41 orange (Munsell, rY-R 7 /s), in spirit nearly white. Both main and dermal skeletons typical for the genus. Spicules. Megascleres: oxea .3 to .8 X .005 to .017 mm. Microscleres: absent. REMARKS: The tissues of this sponge are so heavily filled with opaque pigment cells that it is possible to study the arrangement of the skele- ton only with the utmost difficulty. Halichondria panicea (Pallas) Spongia panicea Pallas, 1766, p. 388. Halichondria panicea Johnston, 1842, p. 114. OCCURRENCE: Point Chevalier Reef. distribution: Arctic; Atlantic coasts of Europe and North America; Mediterranean; Azores; South Africa; Antarctic; New Zealand; extreme southern shores of South America; Japan; Sea of Japan. family HYMENIACIDONIDAE ( de Laubenfels ) GENUS Hymeniacidon (Bowerbank) Hymeniacidon perlevis (Montague) Spongia perlevis (Montague, 1818, p. 86) = H. sanguinea (Grant) and H. caruncula ( Bowerbank ) . OCCURRENCE: Waitawa Bay, Clevedon; Ana- whata; Stanley Bay. Lower midlittoral. REMARKS: Bright orange -yellow, tending to- ward orange when in shadow. The earlier name, H. perlevis (Montague), almost entirely over- looked in the literature, must take precedence over the more familiar names. DISTRIBUTION: Arctic; Atlantic coast of Eu- rope; Mediterranean; West Africa; South Africa; Australia; New Zealand; Japan. family AXINELLIDAE (Ridley and Dendy) SUBFAMILY AXINELLINAE (de Laubenfels) GENUS Axiamon (Hallman) Axiamon erecta (Brpndsted) Fig. 12 Hymeniacidon erecta Brpndsted, 1924*, p. 479, fig. 32. OCCURRENCE: Kawau Island. Commonly washed up on Auckland beaches. remarks : Two of Brpndsted’s species of Hy- menacidon, novae zealandiae and erecta, were relegated to Axiamon by de Laubenfels (1936: 130). This species has never before been figured, and therefore a photograph is appended. DISTRIBUTION: New Zealand. order HADROMERINA (Topsent) family CHOANITIDAE (de Laubenfels) SUBFAMILY CHOANITINAE (de Laubenfels) GENUS Rhahderemia (Topsent) Rhabderemia stellata, sp. nov. Fig. 13*, b, c OCCURRENCE: Rangitoto Island, under stones at low tide. Fig. 12. Axiamon erecta (Br0ndsted). PACIFIC SCIENCE, Vol. XV, January 1961 description : An encrusting sponge with porose surface copiously marked by subdermal grooves. Oscules are small, sometimes the grooves radiate in stellate manner from them. Colour in life yellow, in spirit pale brown. The skeleton is a reticulation with subplumose ascending fibres and connectives, mainly unispicular, of rhabdostyli. Spicules. Megascleres: rhabdostyles .21 to .32 X .015 to .021 mm. Microscleres: (a) acantho- styli .039 to .045 X .003 to .004 mm.; (b) much- contorted sigmas .011 to .017 mm. chord; (c) quadriradiate spicules (calthrops?) with rays .014 to .024 mm. (These are found occasionally in a subdermal position and appear to be foreign inclusions. ) FAMILY SUBERITIDAE (Schmidt) GENUS Suberites (Nardo) Suberites cupuloides, sp. nov. Fig. 14 a, b OCCURRENCE: Rangitoto, littoral in swiftly flowing water. Onetangi Beach in similar habitat. description: A massive sponge with large rounded lobes. The surface is even and minutely hispid. Oscules are not apparent. The texture is firm and fleshy. Colour in life is yellowish-scarlet, in spirit yellowish-brown. The skeleton is of loose, subplumose ascending fibres, ending at the surface in paniculate brushes. Megascleres are of two distinct sizes, the smaller occurring mainly in the dermal brushes. Spicules. Megascleres: tylostyli .17 to 8 X .011 to .018 mm. Microscleres: absent. GENUS Isociella (Halimann) Isociella incrustans, sp. nov. Fig. 15 a, b OCCURRENCE: Ahipara Bay, on ledge under rocks at low tide. DESCRIPTION: This sponge ranges in form from encrusting to massive and spreading. Its surface is uneven and minutely mammillate. Os- cules are small, scattered or irregularly grouped. Often with radiating subdermal channels. The texture is firm and resilient. Colour in life scar- FlG. 13. a, Rhabderemia stellata, sp. nov. b, Rhab- deremia stellata, rhabdostyles, acanthostyli (X 350). c, Rhabderemia stellata, contorted sigmas (X 350). New Zealand Porifera — BERGQUIST 43 let, in spirit whitish. The skeleton is an irregular reticulation of fibres cored and quasi-echinated by main styli (often subtylostylote ) of two sizes. Auxiliary subtylostyli of two sizes present. Dermal skeleton is a tangential layer of larger subtylostyli echinated by a palisade of smaller subtylostyli. Spicules. Megascleres: (a) main styli (2 sizes) : .32 to .35 X .013 to .014 mm.; .2 to .24 X .007 to .011 mm.; (b) auxiliary subtylostyli (2 sizes) : .28 to .34 X .007 to .011 mm.; .14 to Fig. 14. a, Suberites cupulotdes , sp. nov. b, Suber- ites cupulotdes . tylostyli (X 350 ). Fig. 15. a, Isociella incrustans, sp. nov. b, Isociella in crust arts, styli, subtylostyli, isochelae (X 350). .23 X .003 to .006 mm. Microscleres: palmate isochelae .011 to .014 mm. chord. GENUS Polymastia (Bowerbank) Polymastia fusca, sp. nov. Fig. 16^, b OCCURRENCE: Burgess Bay, Kawau Island, under rock ledge. Ahipara Bay, under rock ledge. Spirits Bay, under rock ledge. DESCRIPTION: A massive, spreading sponge, with numerous low wartlike papillae. Surface is even and minutely hispid. The oscules are small, apical on the papillae. The texture is firm and 44 PACIFIC SCIENCE, VoL XV, January 1961 fleshy. Colour in life greenish to chocolate brown (Munsell, rY-R 8/4), in spirit pale brown. The skeleton is of stout radiating bundles of large subtylostyli, with medium-size and small subtylostyli scattered thickly in the choanosome and forming a dense subdermal layer. The der- mal skeleton is a palisade of mainly small sub- tylostyli. Fig. 16. a, Polymastia fusca, sp. nov. b, Polymastia fusca, subtylostyli (X 350). Spicules. Megascleres: subtylostyli of three sizes: (a) .64 X .011 mm.; (b) .4 to .48 X .008 mm.; (c) .14 X .004 mm. Polymastia granulosa ( Brpndsted ) Polymastia granulosa Brpndsted, 1923, p. 162, fig. 36. OCCURRENCE: Anawhata, Piha; lower mid- littoral. REMARKS : Colour bright yellow. DISTRIBUTION: Auckland Islands; Subant- arctic. GENUS Aaptos (Gray) Aaptos aaptos (Schmidt) Synonymy: See Dendy and Frederick, 1924, p. 508. OCCURRENCE: Stanley Bay; lower midlittoral. REMARKS: Specimens all typically reddish- purple externally, brownish-yellow internally. DISTRIBUTION: Mediterranean; West Indies; Indian Ocean; Malaya; Australia. family CLIONIDAE (Gray) GENUS Cliona (Grant) Cliona celata ( Grant ) Cliona celata Grant, 1826, p. 79. OCCURRENCE: Point Chevalier; Kawau Island; Piha. Midlittoral. distribution: Arctic; Atlantic coasts of Eu- rope and North America; West Indies; Indian Ocean; Malaya; Australia. Cliona muscoides (Hancock) Cliona muscoides Hancock, 1849, p. 335, pi. 15, fig. 11. OCCURRENCE: Spirits Bay at 4 fathoms. REMARKS: The habit of the present specimen, especially the appearance of its perforations at the surface of the shell, conforms closely with New Zealand Porifera— -Bergquist 45 Hancocks description. The spicules are identical in appearance, save that the oxea are faintly microspined; the measurements differ, however. CHILE NEW ZEALAND Tylostyli... . Oxea .18 mm. long .07 mm. long .1 to .12 X *003 mm. .052 to .07 X *003 mm. A difference between the present specimen and the holotype is the presence in the New Zealand sponge of small spiny microrhabds, .007 X .002 mm. No mention is made of these in the holotype. Their distribution is, however, sparse and irregular; they themselves are inconspicuous and could easily have been overlooked. The holo- type had infested a shell of Monoceras fusoides , a species recorded for Chile only. Other species of Mollusca are common to Chile and New Zealand, and there is no reason to suppose the Cliona should not share this wide distribution. DISTRIBUTION: Chile. order CARNOSA (Carter) FAMILY HALINIDAE (de Laubenfels) subfamily CORTICIINAE (Vosmaer) GENUS Corticella (Sollas) Corticella novae-zealandiae, sp. nov. Fig. 17 a, b, c GENOTYPE: Corticum stelligemm Schmidt, 1868, p. 25, pi 3, %. 6. Corticella stelligera Sollas, 1888, p. 281. OCCURRENCE: Rangitoto Island; sublittoral fringe. DESCRIPTION: An encrusting to massive and depressed sponge, its surface smooth but uneven. Oscules never apparent. The texture is firm and granular. Colour in life white, in spirit pale brown. Spicules , Megascleres: calthrops, rays .12 to .21 mm. X .026 to .038 mm. The number of rays may sometimes be multiplied to 5 and 6 FlG. 17. a, Corticella novae-zealandiae, sp. nov. b, Corticella novae-zealandiae, typical calthrops (X 80). c, Corticella novae-zealandiae, strongylasters (X 350). and some can show a slight dicho modification. Microscleres: (a) strongylasters .011 mm. in diameter; (b) oxyasters, 5-7 rays .05 to .07 mm. in diameter. REMARKS: This specimen differs from the only other known species from the Mediter- ranean in the dimensions of the spicules. 46 PACIFIC SCIENCE, Vol XV, January 1961 FIG. 18. Location map of Hauraki Gulf (New Zealand inset). Legend: 1 , Burgess Bay, Kawau; 2 , Noises Islands; 3 , Rangitoto; 4 , Stanley Bay; 5 , Parnell Reef; 6 , Point Chevalier Reef; 7, Karaka Bay, St. Helier’s; 8 , Onetangi, Waiheke; 9 , Waitawa Bay, Clevedon. New Zealand Porifera — Bergquist 47 GENUS Plakina (Schulze) Plakina monolopha ( Schulze ) Synonymy: See Burton, 1929, p. 4 14. OCCURRENCE: Rangitoto; midlittoral. REMARKS: A small yellow incrustation. DISTRIBUTION: Mediterranean; Atlantic coast of France; West Indies; Antarctic; Japan. Plakina trilopha (Schulze) Synonymy: See Burton, 1929, p. 4 14. OCCURRENCE: Rangitoto; midlittoral. REMARKS: A small incrustation, cream to deep purple. DISTRIBUTION: Mediterranean; Antarctic. REFERENCES Bowerbank, J. S. 1864. A Monograph of the British Spongiadae, Vol. I. Roy. Soc. Lond. Pp. 1-290, 37 pis. 1866. A Monograph of the British Spongiadae, Vol. II. Roy. Soc. Lond. Pp. 1-388. Br0ndsted, H. V. 1923. Sponges from the Auckland and Campbell Islands. Vidensk. Medd. Kbh. 75: 117-167. — 1924^. Sponges from New Zealand, Part I. Vidensk. Medd. Kbh. 77: 435-483. — 1924 A Sponges from New Zealand, Part II. Vidensk. Medd. Kbh. 81: 295-331. — - — 1927. Antarctic and Subantarctic sponges. Ark. Zool. Stockh. 19A(6) : 1-6. Burton, M. 1926. Observations on some British spp. of sponges belonging to the genus Reniera. Ann. Mag. Nat. Hist. ser. 9, 17: 415-424. — 1929. Porifera, Part II. Antarctic sponges. Brit. Antarct. Terra Nova Exped. 6 (4): 393-458, 5 pis. 1932. Sponges. Discovery Rep. 6: 237— 392, 24 pis. — 1934. Sponges. Gr. Barrier Reef Exped. Sci. Rep. 14: 513-614, 2 pis. 1935. The family Plocamiidae with description of four new genera of sponges. Ann. Mag. Nat. Hist. 15(90): 651-653. Carter, H. J. 1885. Descriptions of sponges from the neighbourhood of Port Phillip Heads, South Australia. Ann. Mag. Nat. Hist. ser. 5, 15: 107-117, 196-222, 301-321, 6 pis.; ser. 5, 16: 277-294, 347-368. 1886. Descriptions of sponges from the neighbourhood of Port Phillip Heads, South Australia. Ann. Mag. Nat. Hist. ser. 5, 17: 40-53, 112-127, 431-441, 502-516. Dendy, A. H. 1894-1895. Catalogue of non- calcareous sponges collected by J. Bracegirdle Wilson, Esq., M.A., in the neighbourhood of Port Phillip Heads, Part I. Proc. Roy. Soc. Viet. ser. 2, 7: 232-260. — 1885, 1896. Catalogue of non-calcareous sponges collected by J. Bracegirdle Wilson, Esq., M.A., in the neighbourhood of Port Phillip Heads, Part II. Proc. Roy. Soc. Viet, ser. 2, 8: 14-51. 1897- Catalogue of non-calcareous sponges collected by J. Bracegirdle Wilson, Esq., M.A., in the neighbourhood of Port Phillip Heads, Part III. Proc. Roy. Soc. Viet, ser. 2,9: 230-259. 1924. Porifera, Part I. Non- Antarctic- sponges. Brit. Antarct. Terra Nova Exped. 6(3): 269-392, 15 pis. Dickinson, M. G. 1945. Sponges of the Gulf of California. Allan Hancock Pacif. Exped. 11 ( 1 ). Haeckel, E. 1889- Report on the deep sea Kera- tosa collected by H.M.S. "Challenger” during years 1873-1876. Rep. Sci. Res. Challenger Zool. 32, part 82: 1-92. Kirk, H. B. 1911. Sponges collected at the Ker- madec Islands by Mr. W. R. B. Oliver. Trans. N.Z. Inst. 43: 574-581, 1 pi. Laubenfels, M. W. de. 1930. The sponges of California. Stanf. Univ. Bull. ser. 5, 5(98): 24-29. 1932. The marine and fresh water sponges of California. Proc. U.S. Nat. Mus. no. 2927, 81: 1-140. — 1934. New sponges from the Puerto Rican deep. Smithson. Misc. Coll. 91(17): 1-28. 48 PACIFIC SCIENCE, Vol. XV, January 1961 1936. Sponge fauna of the Dry Tor- tugas. Pap. Tortugas Lab. 30: 225 pp., 22 pis. 1948. The order Keratosa of the phy- lum Porifera — A Monographic Study. Allan Hancock Fdn. Occ. Pap. 3: 217 pp., 30 pis., 31 figs. 1949. Sponges of the western Bahamas. Am. Mus. Wor. 1431: 1-25. 1950. Porifera of the Bermuda Archi- pelago. Trans. Zool. Soc. Lond. 27(1): 1-154, 2 pis., 65 figs. Lendenfeld, R. von. 1884. A monograph of the Australian sponges, Parts I and II. Proc. Linn. Soc. N.S.W. 9: 121-154. 1885. A monograph of the Australian sponges, Part III. Proc. Linn. Soc. N.S.W. 9: 1083-1150. 1888. Descriptive catalogue of the sponges in the Australian Museum, Sydney. Pub. Aust. Mus. Lond. 16: 260 pp., 12 pis. 1907. Die Tetraxonia. Wiss. Ergebn. "Valdivia” 11: 59-374, 37 pis. Lundbeck, W. 1902. Porifera, Part I. Homor- rhapidae and Heteroraphidae. Dan. Ingolf- Exped. 6: 1-108, 19 pis. Montague, G. 1818. An essay on sponges with descriptions of all species discovered on the coast of Great Britain. Mem. Werner Soc. 2: 67-122, 13 pis. Ridley, S. O. 1887. Report on the Monaxonida collected by H.M.S. "Challenger” during the years 1873-1876. Rep. Sci. Res. Challenger Zool. 20(59, 67): 275 pp., 51 pis. Schmidt, O. 1868. Die Spongien der Kiiste von Algier, mit Nachtragen zu den Spongien des adriatischen Meeres. (Drittes Suppl.) Englemann, Leipzig. Vol. IV, 44 pp., 5 pis. Sole as, W. J. 1888. Report on the Tetractinel- lida collected by H.M.S. "Challenger” during the years 1873-1876. Rep. Sci. Res. Chal- lenger Zool. 25: i-clxvi, 1-458, 44 pis. ToPSENT, E. 1900. Etude monographique des Spongiaires de France, III. Monaxonida (Ha- dromerina). Arch. Zool. Exp. Gen. 8: 1-331. Effects of Pollution on the Free Amino Acid Content of Two Marine Invertebrates Rita D. Schafer 1 Recent reports have been received from some inhabitants of the southern California coastal area of a change in the texture and taste of cer- tain edible marine invertebrates. Investigation showed that the specimens so designated had been taken from areas known to be polluted. The term "polluted” as used here has been given the same meaning as that used in the ecological study by Reish (1956) on the San Gabriel River area: this meaning is the dictionary definition, "the act of making or rendering unclean.” It has been shown for some invertebrates that a change in environmental factors will produce a change in either histological or mor- phological composition. Kinne (1958) showed that a change in ectodermal cell shape can be induced in Cordylophora craspia by varying the environment from marine to brackish. Wilson and Armstrong (1958), after experimentation, concluded that Echinus eggs and larvae are af- fected structurally by the properties of sea water. It has also been demonstrated (Lane and Scha- fer, in progress) that a difference in diet may change the amino acid composition of muscle tissue in some invertebrates. Since this is known to be true under controlled conditions, the pos- sibility exists that a variation in tissue com- position might occur as a result of a polluted and consequently altered environment. This study was thus undertaken to determine (1) if the change in appearance and taste was accompanied by a change in amino acid com- position, and (2) the nature of the change, if one had occurred. Acknowledgments . The author is indebted to Dr. Norman Mattox of the Department of Bi- ology, University of Southern California, for his direction and valuable suggestions throughout the course of this investigation, and to the 1 Allan Hancock Foundation, University of South- ern California, Los Angeles. Manuscript received July 8 , 1959 . administration of the Allan Hancock Founda- tion for the use of laboratory facilities. She also wishes to express her thanks to Mr. Fred Zie- senhenne of the Allan Hancock Foundation and to Mr. John Fitch of the California Depart- ment of Fish and Game, for the collections of specimens from the islands; and to Dr. Donald Reish for the use of his data of measured oxygen from the Los Angeles Harbor area. This study was supported by a research grant from the United States Public Health Service of the Na- tional Institute of Health, no. RG-4911. MATERIALS AND METHODS The animals used for investigation were the abalone, Haliotis cracherodii Leach, and the crab, Pachygrapsus crassipes Randall. The abalone was selected for study, as it is the form in which the greatest difference in tissue texture and taste has been observed. P. crassipes was selected because of its wide distribution and marked tolerance of polluted conditions. These two forms also offer two extremes as to length of time in which the animals are directly subjected to the polluted water. Haliotis is exposed to the air only during periods of lower low tides and therefore is almost continuously within the in- fluence of the conditions prevailing in the pol- luted water. Pachygrapsus, on the contrary, lives in the high tide zone along a rocky shore or on floats rather than in the water in a wharf region and consequently is submerged only during pe- riods of high tide, and is directly subjected to the pollutants for comparatively less time than is the abalone. Collections were made from pol- luted waters and, for a basis of comparison, from waters known to be free of pollution. Specimens of Haliotis were collected along the shore at White’s Point in the Palos Verde region of the southern California coast, from San Clemente Island, Anacapa Island, and Santa 49 50 PACIFIC SCIENCE, Vol. XV, January 1961 Fig. 1. Areas of collection. Catalina Island. Pachygrapsus eras sipes were collected from a point in the west channel of the Los Angeles Harbor area designated as L.A. 7, the mouth of the San Gabriel River at Ala- mitos Bay, White’s Point, Santa Catalina Is- land, Anacapa Island, and Morro Bay (Fig. 1). White’s Point, L.A. 7, and the mouth of the San Gabriel River at Alamitos Bay can be con- sidered polluted areas; while Morro Bay, Santa Catalina Island, San Clemente Island, and Ana- capa Island are surrounded by nonpolluted wa- ter. In the polluted areas both the nature and the degree of pollution vary. White’s Point is a sewer outfall area in which the pipes convey the treated sewage to a site 6,716 ft. from the shore line. According to Stephenson and Grady (1956), oxygen defi- ciency at 6,000-8,000 ft. from the outfall varied from 10-30 per cent, at one time of measure- ment, to 0 per cent (or normal) at another time. It was reported, further, that ammonia content was not normal in any area within 11.000 ft. of the outfall. Measurements taken at 3.000 ft. from the outfall showed an increase of silicates and phosphates. Although dilution un- doubtedly occurs in the remaining 3,700 ft., it would not be safe to assume that the water in the intertidal area where the specimens were taken is of normal composition. The White’s Point area is divided at low tide into a northern cove and a southern cove by a narrow expanse of rock which extends out from shore for approximately 200 ft. A survey of the fauna of these two sections shows them to be quite different. On the southern side of the rocky projections were found scattered speci- mens of Haliotis cracherodii, an occasional Pach- ygrapsus crassipes, masses of the tubed worm, Phragmatopoma calif ornica, and specimens of the limpets Fissurella volcano and Acmaea lima- tula. Empty Olivella shells were also present in relatively large numbers. In sharp contrast to this paucity of forms and individuals was the fauna of the northern section. Here were noted specimens of Pis as ter ochraceus, Strongylocen- trotus purpuratus. S. franciscanus, Ophiothrix spiculata, Bulla gouldiana, Aplysia calif ornica. Octopus bimaculatus, Conus calif ornicus, Pagu- rus samuelis, in addition to those animals found on the southern side. This difference can be ex- plained by the fact that the sewer outfall opens offshore in a line with the rocky projection which separates the two sides. The current flows primarily from north to south carrying the polluted water in the direction of the southern section. The point designated as L.A. 7 is located in the west channel of Los Angeles Harbor. The term "L.A. 7’’ was given to this particular point in a pollution survey conducted in 1952 by the Los Angeles Regional Water Pollution Control Board, and has been retained in this study so that this site may be recognized and related to data taken from that point. Specimens of Pachy- grapsus crassipes tested were taken from the piling and floats at one of the small yacht har- bors in this area. The nature of the pollution at this point has been designated as primarily raw sewage from approximately 500 persons. Oxygen content of the water, measured at monthly intervals over a period of 3 years (1956 through 1958), showed fluctuations from a low of 2.0 to a high of 8.2 parts per million. An ecological survey of the area shows the custom- ary wharf fauna of the southern California wa- ters. Anemones, the wharf mussel ( Mytilus edulis ) , tunicates both solitary and colonial, hydroids, barnacles, and colonial serpulid worms are attached in great abundance to the under- side of the wooden floats, while numerous sped- Free Amino Acid Content — Schafer 51 mens of Pachygrapsus crassipes live on the floats and piling just above the water line. Since the area is maintained as a small yacht harbor, the floats are scraped occasionally and invariably are soon resettled by young forms of the species mentioned. The conditions at the mouth of the San Gabriel River at Alamitos Bay were thoroughly investigated and reported by Reish in 1956. He reports the primary source of waste dis- charge into that region to be the Dow Chemical Company, the Santa Fe Springs Waste Disposal Company, two domestic sewage disposal plants, and the Los Angeles Bureau of Water and Power Steam Plant. Since specimens of the same species living in the same conditions are known to have a con- sistent pattern of free amino acids in their mus- cle tissues, an investigation of these amino acids should indicate whether or not a change from a normal to an abnormal metabolism has taken place. An analysis of free amino acids was thus undertaken by means of two-dimensional paper chromatography. Muscle tissue only was used for the deter- minations. All samples studied were taken from individual animals. Pooled samples were not used. Specimens were quick-frozen; muscle tis- sue from the foot of the abalone and from the legs of the crab was excised and subsequently lyophilized without thawing. This precaution was taken to inhibit the activity of autoenzymes and bacterial enzymes. After lyophilization the tissue was extracted with cold 70 per cent ethanol. Two 50 ml. aliquot portions were used, and extraction was permitted for at least 12 Fig. 2. Chromatogram of free amino acids of Haliotis cracherodii taken from Anacapa Island, a nonpolluted 52 PACIFIC SCIENCE, Vol. XV, January 1961 Fig. 3. Chromatogram of free amino acids of Haliotis cracherodii taken from White’s Point, a polluted hours for each portion. Extracts were then con- centrated and stored under refrigeration as a 4.4 ml. solution of 10 per cent isopropanol. Chromatograms were run on 1814- by 2214- inch Whatman no. 1 filter papers. A mixture of butanol, acetic acid, and water (4:1:5) was used as the first phase; water-saturated phenol was used as the second. Development was car- ried out by dipping in a 0.2 per cent solution of ninhydrin in acetone. RESULTS The abalone, Haliotis cracherodii, from both polluted and nonpolluted areas contained the amino acids alanine, arginine, aspartic acid, cystine, glutamic acid, glycine, histidine, leucine- isoleucine, tyrosine, and valine. Specimens taken from the nonpolluted areas (Santa Catalina Island, San Clemente Island, and Anacapa Is- land) contained asparagine in addition to the above amino acids. The specimens taken from White’s Point showed no asparagine, but did show a definite spot identified as aspartic acid. This acid was either absent or only very faintly discernible in the specimens from the non- polluted areas. Phenylalanine and three uniden- tified spots were present in the specimens ob- tained from the White’s Point area (Figs. 2, 3; Table 1). All specimens of the crab, Pachygrapsus crassipes, contained the free amino acids alanine, arginine, aspartic acid, cystine, glutamic acid, glycine, histidine, leucine isoleucine, lysine, me- thionine, proline, serine, threonine, tyrosine, and Free Amino Acid Content— Schafer 53 valine. Specimens taken from Santa Catalina Island, Anacapa Island, and Mot.ro Bay con- tained asparagine as well. Those from Santa Catalina and Anacapa islands contained an un- identified amino add located to the right of arginine on the chromatographic pattern (Figs. 4,5; Table 2). DISCUSSION Both Haliotis cracker odii , which occurs in one markedly polluted area, and Pachygrapsus eras - sipes, which was collected from a variety of polluted areas, have a free amino acid composi- tion different from that of the same animals collected from nonpolluted areas. Haliotis from clean water gives a more consistent pattern than does Pachygrapsus, Both forms taken from pol- luted waters are marked by the absence of as- paragine. This indicates that in some manner the metabolism dealing with this amino acid has been altered. In the case of Haliotis from White’s Point, phenylalanine and three uniden- tified spots appeared, indicating a more exten- sive change in metabolic pattern. The change in the abalone is greater than that in the crab. This may be due either to a difference in re- sponse to polluted conditions because of the difference in the animals; or it may in some way be associated with the difference in length of time during which the animals are submerged and are subjected to the polluted conditions. The factors of pollution common to the areas considered were a depletion of available oxygen and an increase in nutrients resulting Fig. 4. Chromatogram of free amino acids of Pachygrapsus crassipes from Anacapa Island, a nonpolluted 54 PACIFIC SCIENCE, Vol. XV, January 1961 TABLE 1 Free Amino Acids Present in Specimens of the Abalone, Halt otis cracherodii. from a Polluted and Three Nonpolluted Areas 3NINVTV ARGININE ASPARTIC ACID ASPARAGINE CYSTINE | GLUTAMIC ACID GLYCINE 1 HISTIDINE C/3 z 0 D u; LYSINE METHIONINE PHENYLALANINE PROLINE SERINE THREONINE TYROSINE VALINE 5 z o z u z 3 UNKNOWN #2 UNKNOWN #3 || Santa Catalina Is X X X* X x. X X ><^ >< X X X X X X Anacapa Is. \x_ x 1 X b X x^ X_' X X X X X X San Clemente Is X x'b 1 X X X X x^ X X X X X X X X White’s Point X X x X VI X X X X X X X X X X X X X X * Very faint. Fig. 5. Chromatogram of free amino acids of Pachygrapsus eras sipes from the mouth of the San Gabriel River at Alamitos Bay, a polluted area. Free Amino Acid Content — Schafer 55 TABLE 2 Free Amino Acids Present in Specimens of the Crab, Pachygrapsus crassipes, FROM FOUR NONPOLLUTED and Three POLLUTED AREAS ALANINE ARGININE ASPARTIC ACID ASPARAGINE | CYSTINE GLUTAMIC ACID GLYCINE HISTIDINE LEUCINES LYSINE | METHIONINE [ PHENYLALANINE W z 0 PS SERINE THREONINE TYROSINE VALINE UNKNOWN #1 Santa Catalina Is X X X X X_ X X X 2< X X 22 X X 2< Anacapa Is X X X X X X X X X x_ 2< X X X X 1 2< San Clemente Is X X X xl X X X X 21 2< 21 X X xl Morro Bay X xl x i X X X X X X X X X X X Xl 2< White’s Point X x xl xj X X X X 2! 2< X X X xl L. A. Harbor X x 1 X 1 X X X X X n X X X X xl Alamitos Bay X X| X| X X X X X X X X X X X | from sewage. Since the role of the free amino acids in the animal body is as yet not completely understood, it is not possible to state what physiological mechanisms have been altered by these environmental changes or whether one or both of the factors have operated to bring about the altered free amino acid metabolism. That a biochemical change as well as a morphological change has occurred is evident. SUMMARY 1. Specimens of the abalone, Haliotis cra- cherodii, and the crab, Pachygrapsus crassipes, were collected from clean and polluted waters. 2. Chromatographic analysis of the free amino acids of these forms were made by two-dimen- sional paper chromatography. 3. Specimens from polluted areas were found to be lacking in asparagine. This amino acid was present in specimens from clean waters. 4. Phenylalanine and three unidentified amino acids, not found in Haliotis cracherodii from clean water, were present in this species taken from polluted water. 5. Polluted waters differed from nonpolluted waters in that they showed a depletion of avail- able oxygen and an increase in nutrients. REFERENCES Kinne, Otto. 1958. Uber die Reaktion erbglei- chen Coelenteraten-gewebes auf verschiedene Salzgehalts und Temperaturbedigen, II. Mit- teilung fiber den Einfluss des Salzgehaltes auf Wachstum und Entwicklung mariner, brack - ischer und limnischer Organismen. Zool. Jahrb. Abt. Zool. Physiol. Tiere 67(4): 407- 488. Lane, C. E., and R. D. Schafer. 1959. Work in progress. Reish, D. J. 1956. An ecological study of lower San Gabriel River, California, with special reference to pollution. Calif. Fish and Game 42(1): 51-61. — — — 1959- Unpublished data. Personal com- munication. Stephenson, R. E., and John R. Grady. 1956. Plankton and associated nutrients in the wa- ter surrounding three sewer outfalls in south- ern California. Report to Hyperion Engineers, Inc. Univ. of Southern Calif. Press. 48 pp. Wilson, D. P., and F. A. J. Armstrong. 1958. Biological differences between sea waters: Experiments in 1954 and 1955. Jour. Marine Biol. Assoc. United Kingdom 37(2): 331- 348. Studies in the Helminthocladiaceae (Rhodophyta) : Helminthocladia 1 Maxwell S. Doty 2 and Isabella A. Abbott 3 During recent work in the field with the algae of Hawaii several interesting red algae have been found. Two of these interesting algae which seem to be members of the Helmintho- cladiaceae are reported here, in the hope that more work with such algae will be encouraged. The Helminthocladiaceae is accepted as be- ing a family, the limits of which would include eight genera that are rather well known and perhaps three genera that are not well known, Ardissonea, Dorella, and Endosira. Ardissonea was described by J. Agardh (1899: 99) and is treated by Kylin (1956: 127), under the name of Neoardis sonia Kylin, as a member of the Naccariaceae. Dorella , which may be a member of this family, has terminal cortical cells which are not enlarged. According to a personal com- munication from Dr. H. B. S. Womersley, the type of Ardissonea is a very finely branched alga and Endosira appears to be a juvenile of a different order. Kylin (1956: 557) suggests that Endosira may be related to Nemastoma. Consequently, we shall consider these genera no further in connection with the algae being described below. Of the eight easily recognizable genera, only Helminthocladia possesses lateral carpogonial branches and zygotes (post-fertilization carpo- gonia) which divide transversely, longitudinally, or obliquely and give rise to a dense gonimo- blast from both division products. In addition, 1 The field and laboratory expenses for this work were provided in part by contract G-3833 between the University of Hawaii and the National Science Foun- dation, and by Graduate Research Funds of the Uni- versity of Hawaii. The authors appreciate the use of facilities at the Hopkins Marine Station of Stanford University for part of this work. Contribution No. 143 from the Hawaii Marine Lab- oratory. Manuscript received September 8, 1959- 2 Department of Botany, University of Hawaii, Ho- nolulu 14, Hawaii. 3 Hopkins Marine Station, Stanford University, Pa- cific Grove, California. in Helminthocladia the terminal vegetative cells in the cortex are strongly enlarged. Kylin (1956: 108) uses this latter as a key characteristic to separate this genus from Helminthora. There are other differences between the two genera: In Helminthora, for example, only the upper cell of a transversely dividing zygote gives rise to gonimoblast filaments. Trichogloea differs from Helminthocladia and from other well-known genera in having straight terminal, rather than curved lateral, carpogonial branches and in its calcification. Dermonema has long been a rela- tively unknown genus but is distinct in form, being erect cushions formed of noncalcified closely dichotomous branches, sometimes like Chnoospora minima in looks and habitat. Both Dermonema and Cumagloia (Gardner, 1917: 401) are distinct in having a diffuse gonimo- blast ramifying among the cortical filaments near the zygote from which it originated as a few protuberances with no previous division of the zygote. The genus Liagoropsis of Yamada (1944) is like Nemalion (Desikachary, 1957 a), having straight carpogonial branches, but differs in being calcified. On the basis of various characteristics the algae to be described are judged to be distinct, new species of Helminthocladia. They repre- sent the only records of this genus of the Rhod- ophyta for the Central Pacific Ocean. Helminthocladia simplex sp. nov. Figs. 1-18 DESCRIPTIO TYPI: Thalli irregulariter cylin- drici, usque ad 9.5 cm. alt., acibus ramisque subsimplicibus saepissime 1 ad 1.5 mm. diam. Rami pauci, irregulariter dispositi. Thalli saepe simplices, qui saepe latiores quam thalli ramosi, raro, autem, plus quam 2 ad 4 mm. diam. Thalli simpliciores forma magis irregulares, clore magis obscuri, statura breviores saepe carpo- goniales sunt. Thalli antheridiales ubique vel 56 Helminth ocladia — Doty and Abbott 57 Fig. 1. The type of Helminth ocladia simplex, a preparation on a herbarium sheet. plerumque tenuiores, clore nitidiores, altiores, magis ramosi. Thalli textura ubique lubrici mol- lesque. Frondes mnltiaxiales, filamentis corti- calibus nullo modo inter se adhaerentibus. The holotype is a preparation of six thalli on one herbarium sheet deposited in the Bernice P. Bishop Museum in Honolulu, Hawaii. These specimens, along with a small Liagora, formed a turf on an almost horizontal rock surface just above extreme low tide line. The type material was collected by Jan Newhouse and Henry Ke- koanui (M. Doty no. 12691) at Kahanahaiki, Waianae, Oahu, Hawaiian Islands (21° 32' N., 158° 14' W.), Jan. 2, 1954. Isotypes are be- ing sent to the herbaria of the University of California, University of Michigan, Hopkins Marine Station of Stanford University, Cryp- togamic Laboratory of the Paris Museum, Uni- versity of Adelaide, South Australia, and Hok- kaido University at Sapporo, Japan. During some of the years since the original collection was made by Newhouse and Keko- anui, the type locality has been revisited. The sand shifts a great deal throughout the year at this site; sometimes the place is completely covered with sand. Until recently the alga had not been refound, though other members of the Helminthocladiaceae were often present in abundance. However, on Apr. 10, I960, while Doty was accompanied by Newhouse and Er- nani Menez, a dense stand of H. simplex (M. Doty no. 19135A, Menez no. 201) was found. The thalli were essentially of the same mor- phological form and were collected in the same place under the same conditions as the type. The living material was yellowish brown, with no taste or odor, and hard like a Gracilaria rather than soft like a Trichogloea. Perhaps a half liter of the species was obtained at this time. In time the algal population dwindled and the sand as well, until by May 8 there was very little of either on the site. The form of the Helminth ocladia simplex present at this time was still the same as the other collections of this species, or perhaps a bit more eroded. Thalli (Fig. 1) of irregularly cylindrical branches, up to 9.5 cm. tall, with most of the subsimple axes and branches 1-1.5 mm. in diam- eter. Branches irregular in arrangement and few. Thalli often simple, and these simple ones, while often of larger diameter than branched thalli, are rarely more than 2-4 mm. in diameter. Often the simpler thalli of more irregular form that are duller in color and shorter in stature are carpogonial. The antheridial thalli are gen- erally more slender, more brightly colored, taller, and more branched. Texture rubbery and pliant throughout. The fronds are multiaxial and the cortical filaments do not adhere to each other in any way. Our material is dioecious. No evidence of what might be a tetrasporangial generation was seen. The male thalli produce spermatangia on terminal cortical cells among the vegetative filaments of the surface. The spermatangia are produced on cells (Fig. 2s) that are smaller than the adjacent vegetative cells and terminate in dichotomous rows of small cells. These small masses of cells do spread out under the cover glass on a microscope slide as do the terminal fans described for Helminth ocladia by Martin (1939), but one suspects them of covering the surface area of the vegetative terminal cells they replace; i.e., occurring in round brushlike clusters rather than in two-dimensional fans. More than one spermatium may adhere to a trichogyne; those seen stuck to trichogynes were colorless. The female apparatus develops laterally from the fourth or fifth cell (Figs. 3, 4, 5) below the enlarged terminal superficial cortical cell. 180 JJ Helminthocladia — Doty and ABBOTT 59 The terminal cell of the developing carpogonial branch is large at first (Fig. 3). Only three- celled carpogonial branches were seen (Fig. 15 notwithstanding) . It often appeared (Figs. 5, 6, 7) that the trichogyne cytoplasm became separated from the zygote cytoplasm; this we accepted as evidence that fertilization had taken place. Few cases were studied where we were certain that only the first division of the zygote had taken place. A number of cases were seen where two (Figs. 6, 7, 8), three (Figs. 9, 10, 11), four, or five divisions (Fig. 12) had taken place. From these it was clear that division of the zygote is usually longitudinal or oblique, as in the case of H. papenfussii as illustrated by Martin (1939, figs. 17, 18). Conspicuous post-fertilization changes were not apparent in the carpogonial branch cells other than in those derived from the car- pogonium itself. No placental cell formation was seen. Pit connections within the carpogonial branch (Figs. 13, 14) and to the supporting cell and to the supra-supporting cell were not enlarged, or those between the carpogonial branch cells were only slightly enlarged. The hypogynous carpogonial branch cells in older stages (Figs. 13, 14) were "lighter staining” than during earlier stages. In the six or eight cases in older gonimoblasts studied in this re- gard (e.g., Fig. 15), the central complex of densely staining cells presumably derived from the carpogonial branch had only, at most, "broad- ened” pit connections. The contents of the sup- porting and supra-supporting cell were in some cases darkened and shrunken in diameter in this formalin-fixed material. As in H. papenfussii (Martin, 1939), Hel- minth ora lindaueri ( Desikachary, 1955: 131), and in Helminthocladia australis (Desikachary, 1957&), encircling sterile rhizoid-like filaments (r in Figs. 6, 7, 9, 11, 14, 15) after fertiliza- tion grow especially around the hypogynous cells of the carpogonial branch. These were not seen to invade or surround older gonimoblasts of H. simplex when these were producing sur- faceward- growing filaments. In fact they seemed largely to have disappeared or become lost in our preparations of older stages. Gonimoblast initials appear from both pri- mary division products of the zygote (Figs. 9-13). The cells of the young gonimoblast (Figs. 10-12) are at first in a dense regular mass which becomes lobed (Figs. 13, 14, 15) in time. In this respect our organism is like other Helminthocladia species and unlike other genera in the family. From the dense indefinitely lobed central gonimoblast mass, sparsely branched rather par- allel filaments (Fig. 16) grow towards the sur- face of the thallus. The gonimoblast filaments are usually unbranched for the terminal three cells. It may well be, though not followed out closely, that the terminal two cells tend to be- come carposporangia and the bottom (third one) gives rise (Fig. lib) to a two-celled branch that in turn grows to look like the terminal two cells of the parent filament be- fore they began enlarging into carposporangia. This same third cell may produce another branch. It is interesting to note this sympodial manner of growth here. In some cases the third cell in the row be- comes a carposporangium. In this respect the organism at hand approaches that condition illustrated by Kylin (1930, fig- 2D) for H. calvadosii. That is to say, there is nothing like the branching which Papenfuss ( 1946, fig. 25) illustrates for the homologous structures in Trichogloea . DISCUSSION: The type, MD 12691, is dis- tinguishable from the classical Helminthocladia hudsoni and H. calvadosii (accepted as the type species of the genus as circumscribed by Hamel, FIGS. 2-18. Reproductory and anatomical features of Helminthocladia simplex. 2, Cellular details of a cortical filament system bearing spermatangia (s) apically and a rhizoid (r) from the assimilatory region. 3-5, Young carpogonial branches (stippled). 6—15, Stages in the development of the gonimoblast with its enveloping rhizoids. Individual cells are indicated by dark stippling; groups of cells are indicated by light stip- pling; enveloping rhizoids are indicated, and an assimilatory apical cortical cell. 16—17, Branching of the carpospore-producing filaments (b) . 18, Cellular details of a cortical vegetative filament system showing the strongly dichotomous branching and, from the inner cortical regions, rhizoidal filaments (r). 60 PACIFIC SCIENCE, Voi. XV, January 1961 1930), and from all other species known to the authors, in the reproductory structures described for these taxa and on the basis of their being more branched and larger in size. Speci- mens of H. calvadosii from France (University of California Herbarium no. 407401, identified by Kylin) measured 26-36 cm. tall, and in the parts examined lacked any trace of the sterile rhizoid-like filaments characteristic of our species and of H. papenfussii. Another specimen (University of California Herbarium no. 218320, labeled by Rosenvinge H. purpurea) was up to 60 cm. tall and likewise lacked the peculiar rhizoidal filaments around the hypo- gynous cells. This latter specimen was the most nearly simple in branching of any Helmin- th ocladia examined by us aside from H. simplex . In regard to the enveloping rhizoidal filaments our organism is unlike H. calvadosii (Kylin, 1930), H. hudsoni (Feldmann, 1939) which have no such filaments, and H. papenfussii as described by Martin (1939) which has many such filaments. The most striking of these sterile filaments (Figs. 8, 9, 12, 14) arise from the cell above the supporting cell in the vegeta- tive branch, but they are more complex than those Balakrishnan (1955) illustrates for Lia- gora erecta. Martin (1939) ascribes both a fusion cell to Helminth ocladia papenfussii, derived from the carpogonial branch, and a sterile envelope; these are illustrated in her figures 20 and 21. By the time a gonimoblast is this far developed in this Hawaiian species, there is no indication of either such a fusion cell or such an enveloping basket of sterile branches. The sterile rhizoidal branches develop in H. simplex as in some other Hel- minth ocladia species where such may be found, primarily from the cell above the supporting cell in the vegetative branch, as Kylin ( 1938, fig. 1C) illustrates H. papenfussii. The first to appear tend strongly to encircle the young gonimoblast but they were not seen in older stages. Desikachary (1956, figs. 25, 29) illus- trates a similar situation in Helminthora lind- aueri from New Zealand. The material reported and figured as Hel- minth ocladia australis by Okamura (1916: 21) and by Segawa (1957: 58, fig. 254) seems to be similar to ours in habit, except for the larger size and greater degree of branching. However, Narita’s ( 1918) figure of H. yendoana, which in that author’s opinion includes H. australis of Okamura, does not resemble our alga at all. Furthermore, our examination of certain speci- mens (University of California Herbarium nos. 335335, a female thallus apparently identified by S. Narita; 279932, a female thallus identified by Y. Yamada; and 418162 ) shows the Japanese material to be different in other details as well. From the materials illustrated and discussed as H. australis by Desikachary (1957&), our mate- rial differs in being far simpler and smaller. We refer here only briefly to the rhizoids (Fig. 18 r) which develop from the lower cells of the cortical filament systems. Only rarely was there any indication of such a rhizoidal filament (see r in Figs. 2) in the outer part of the cortex that even recalls slightly the rhizoids peculiar to the next species. Figure 18 f i -?*4 perhaps illustrates the ontogeny of this rhizoidal type. Note that the cortical cells (Fig. 18) are, in general, characteristic for Helmin- th ocladia. Helminth ocladia rhizoidea sp. nov. Figs. 19-24 DESCRIPTO TYPI: Thallus 9 cm. alt., valde mucosoideus, in partibus inferioribus radiatim ramosus; filamenta corticea usque ad 350 /x long., irregulariter dichotome tritomeve ramosa; cel- lulae apicales amplificatae, pyriformes, 13-26.5 fx lat., 45 fx long.; rhizoidea multa 4.8-7.2 g lat., e filamentis corticeis exterioribus producta; gonimoblastus involucro filamentorum e cel- lulis minoribus quam cellulae corticis vegeta- tivi consistantium, aliter, autem, filamentis as- similativis corticeis propinquis similium, cir- cumdatus; filamenta involucri e cellulis vege- tativis infra superque cellulam sustinentem pro- ducta. The holotype is a preparation bearing the collection number MDoty 12860. The specimen was collected by Mr. Tetsuo Matsui at Lahaina (156° 41' W., 20° 53' N.), on the island of Maui, Hawaii, and it is deposited in the Bernice P. Bishop Museum, Honolulu, Hawaii. Helminth ocladia - — Doty and Abbott 61 Thallus (Fig. 19) of cylindrical branches, 9 cm. high, branching radial in the lower por- tions, the branches 4 mm. in diameter at their bases, gradually tapering to 1-2 mm. in diam- eter at the tips, strongly mucosoid. The main branches give rise laterally to shorter ones of irregular length between 1 and 3 cm. long. Basal disc 5 mm. in diameter. Cortical filaments (Fig. 20) up to 350 g in length, irregularly dichotomous or trichotomous, the lower cells ovoid to cylindrical, the terminal cells inflated and pyriform, 13-26.5 g wide X 45 fx long. Short unbranched filaments, usually without the terminal pyriform cells, are com- monly borne at the tops of the cortical filaments (Figs. 21/, 24/). Numerous, sometimes branched rhizoids 7.2-12 ji wide are produced from the medullary filaments. Rhizoids are also produced (Fig. 21 r) by upper cells of the cortical fila- ments at first as protuberances 4.8-7. 2 /x wide on the lower edge of the cells (Fig. 22 f), then by elongation cutting off segments (Fig- 23 r). They are linear, unbranched, and seem to con- Fig. 19- The type of Helminthocladia rhizoidea, a single specimen preserved on a herbarium sheet. nect neighboring assimilatory branches to each other, whereas those rhizoids produced nearer the axial strand add thickness to the axis. Oc- casional hairs (Fig. 21 h) may be formed on the terminal cortical cells. Carpogonial branches 3 -celled, curved, mostly lateral as specially formed branches (stippled in Figs. 20, 23) at or near a dichotomy, but occasionally taking the place of a vegetative branch at a dichotomy. If unfertilized, they may develop into cortical filaments (Fig. 20) that are of smaller diameter than the ordinary cortical filaments. The first division of the gonimoblast is longitudinal. Gonimoblast dense, regular in shape (Fig. 22), soon becoming an irregularly shaped dense mass of filaments with only the end cells producing carpospores. A few sterile filaments are produced from the vegetative cells above the supporting cell (Fig. 24) or from those subtending the supporting cell (Fig. 22). They loosely bracket and overtop the gonimoblast (Fig. 22/), together with the vegetative filaments deflected by the growth of the cystocarp. No fusion cell is formed. DISCUSSION: In its vegetative appearance ( Fig. 19), H. rhizoidea is similar to certain other species of Helminthocladia, such as some forms of H. australis ( Desikachary, 1957 h, pi. 16, fig. 3). It is a strikingly different alga from H. simplex, described above (Fig. 1 ) , which it does not resemble in either external or internal struc- ture. However it is generally similar in external appearance to other well-described species of Helminthocladia; i.e., H. calvadosii (Kylin, 1930), H. papenfussii (Martin, 1939), and H. australis (Levring, 1953; Desikachary, 1957£). H. rhizoidea differs from these because of the production of decumbent rhizoids (r in Figs. 21-23) from the basal ends of the vegetative cells, which constitute the assimilatory filaments. While this fact in itself may not be of first importance, it does clearly separate this species from other species of Helminthocladia. Rhizoi- dal structures do appear nearer the medulla in H. simplex ( Fig. 2 ) and in the Helmintho- cladia studied by Desikachary (1957 h: 442, fig. 5), but these seem to have a different origin (see Fig. 18). It would seem that vegetative 62 PACIFIC SCIENCE, Vol. XV, January 1961 Figs. 20—24. Cellular reproductive and vegetative peculiarities of Helminthocladia rhizoidea. 20, A cortical heterofilamentous system, wherein some branches are of slender long cells (v) and some terminated by "normal” obpyriform cells, normal (e.g., cp) and seemingly abnormal carpogonial branches. 21, Cellular details of a cortical vegetative system illustrating a hair base (h), slender cortical filaments (/), and two of the rhizoids (f) which characteristically issue from the assimilatory region. 22, Origin of a rhizoid (r) from an assimila- tory cortical filament, a well-developed gonimoblast (lightly stippled) with three hypogynous cells (darkly stippled) and several small-celled involucral filaments (/). 23, One of the rhizoids (r) peculiar to this species and well-formed carpogonial branch (cp). 24, A young gonimoblast with two one-celled encircling rhizoids (r) developed from supra-supporting cells, and a slender cortical filament (/). characters of this kind are necessary aids to dis- tinguishing the ever growing number of species in this genus. The sterile filaments surrounding the cysto- carp of H. rhizoidea (Fig. 22^') appear to re- semble closely those in H. papenfussii as illus- trated by Martin (1939), although the deriva- tion of the sterile filaments may not be the same in both species. Martin states that the sterile filaments arise from the vegetative cell above the supporting cell in H. papenfussii. This is true also in H. australis ( Desikachary, 1957&) where, however, they may also arise from the cell below the supporting cell. The derivation in H. rhizoi - dea also may be from above or below the sup- porting cell. The possession of a loose basket of sterile filaments around the gonimoblast in H. rhizoi- dea seems to furnish a further characteristic for distinguishing this species from H. simplex. Only the initial few cells of the involucre are illustrated in Figures 22 and 24 for H. rhizoidea , while perhaps the ultimate in development of rhizoids is given in Figures 6, 7, 9, 11, 14, 15 for H. simplex. Most species of Helminthocladia appear to have fairly regularly dichotomous assimilatory filaments; see the illustrations of H. calvadosii (Kylin, 1930), hi. hudsoni (Feldmann, 1939), H. australis (Desikachary, 1957&), and those of H. simplex , especially Fig. 18 in this paper. In this respect H. papenfussii and the present Helminthocladia — Doty and Abbott 63 species are similar to each other in that the branches near the tops of the filaments are often trichotomous. The ultimate cells of the cortical filaments are more crowded, therefore, than those of most other species. Often in this genus where the cortical filaments are not dichotomous the production of carpogonial branches or rhizoidal branches (of the type illustrated in Fig. 18) seems to have been involved. Either normal carpogonial branches may have appeared (Figs. 20 cp, 23 cp) , or abortive carpogonial branches may have become reorganized (Fig. 20v), pos- sibly into vegetative branches. REFERENCES Agardh, J. 1899. Analecta Algologica . . . con- tinuatio V:' 159 pp. Lunds Univ. Arsskr. 35, Afdeln. 2, n. 4. BalakrishmaN, M. S. 1955. On Liagora ere eta Zeh. J. Indian Bot. Soc. 34(3): 207-212. Desikachary, T. V. 1955. On a Flelminthora from New Zealand. Amer. J. Bot. 42: 1 26— 131. — — — 1956. Observations on two species of Liagora ( Rhodophyta ) . Pacif. Sci. TO (4): 423-430. — 1957a. On a "Liagoropsis” from Anda- mans. Rev. Algol, n.s. 3(2): 51-56. 1957 b. Helminthocladia from India and New Zealand. J. Indian Bot. Soc. 36(4): 441-456. Feldmann, Jean. 1939. Une nemalionale a carpotetraspores : Helminthocladia Hudsoni (C. A g.) J. Ag. Bull. Soc. Hist. Nat. Afr. N. 30: 87-94. Gardner, Nathaniel Lyon. 1917. New Pa- cific Coast marine algae, I. Univ. Calif. Publ. Bot. 6: 377-416. Hamel, Gontran. 1930. Floridees de France, VL Rev. Algol. 5: 61-109. Kylin, Harald. 1930. Ueber die Entwicklungs- geschichte der Florideen. Lunds Univ. Arsskr. n. f. Avd. 2, 26(6), 104 pp. — 1938. Verzeichnis einiger Rhodophy- ceen von Siidafrika. Lunds Univ. Arsskr. n. f. Avd. 2, 34(8), 26 pp. 1956. Die Gattungen der Rhodophy- ceen. C. W. K. Gleerups Press, Lund, Sweden, xv + 673 pp. [G. F. Papenfuss tells us that through correspondence with Mrs. Kylin he determined the date of publication to be Oct. 30, 1956.} Levring, T. 1953. The marine algae of Aus- tralia, I. Ark. Bot. ser. 2, 2(6): 457-530. Martin, Margaret T. 1939. Some South Afri- can Rhodophyceae, I. Helminthocladia Papen- fussii Kylin. J. Bot. 77: 234-244. Narita, Seiichi. 1918. Enumeratio Specierum Nemalionis et Helminthocladiae Japonicae. Bot. Mag. Tokyo 32(381): 189-193. Okamura, Kintaro. 1916. leones of Japanese algae 4: 2 If. Papenfuss, G. F. 1946. Structure and repro- duction of Trichogloea Requienii , with a comparison of the genera of Helminthocladia- ceae. Torrey Bot. Club Bull. 73(5): 419-437. Segawa, Sokichi. 1957. [Coloured illustrations of the seaweeds of Japan.} Hoikusha Press, Osaka, Japan. 175 pp. (In Japanese.) Yamada, Yukio. 1944. Notes on some Jap- anese algae, X. Sci. Pap. Inst. Algol. Res. Hokkaido Univ. 3(1): 19-22. sir: Studies on Pacific Ferns, Part III The Lindsaeoid Ferns G. Brownlie 1 The present study arose from the fact that the genus Lindsaea has proved of particular interest from a cyto-taxonomic viewpoint, and, as the author hopes to deal with some of these ferns in the Pacific islands from that aspect, he took the opportunity of a period spent at the Herbarium of the Royal Botanic Gardens, Kew, to examine the collections located there. It quickly became evident that the species of lindsaeoid ferns, particularly those of New Caledonia, were poorly understood. There was not sufficient time and, in some cases, not suf- ficient material to completely clear up all doubt- ful aspects, but it is hoped that the present paper will go some distance towards clarifying the situation. To accomplish the task fully for the Pacific and neighbouring regions would require a work almost as large as the excellent revision of Lindsaea in the New World by Kramer (Acta Bot. Neerlandica 6, 97-290, 1957). Taxonomically, the lindsaeoid ferns have proved a difficult group almost everywhere, and this is especially apparent in New Caledonia where both specific and generic distinctions are extremely doubtful. It is possible that they are a very old group whose close relationships are more apparent than real. This possibility has already been indicated by the limited cytological work published to date. Acknowledgments . I wish to thank Mr. F. Ballard of the Royal Botanic Gardens, Kew, for his assistance, and I record my appreciation of the help given by the late Mr. A. H. G. Alston in connection with photographs of type material in Berlin herbaria. I also acknowledge the leave granted me by the Council of the University of Canterbury, during which this research was un- dertaken. 1 Department of Botany, University of Canterbury, Christchurch, New Zealand. Manuscript received June 22, 1959. Lindsaea dimorpha Bailey. Lindsaea dimorpha Bailey. Handbook, Queens- land Ferns 19, 1874. L. anogrammoides C. Chr. Viert. Nat. Ges. Zurich 70, 223, 1925. Christensen himself suggested that these two were closely related, and comparison of his type with Australian material shows that there is nothing to distinguish the New Caledonian specimen. It should consequently be recognized as L. dimorpha. SPECIMENS EXAMINED: Franc 1308 in Herb. C. Chr. (Christensen’s type; B. M.), Bailey un- numbered (Kew). Lindsaea lapeyrousii ( Hk. ) Bak. Davallia lapeyrousii Hk. 2nd. Cent. f. 56, 1861. Lindsaea lapeyrousii (Hk.) Bak. Syn. Fil. 106, 1874. L. kajewskii Copel. Journ. Arnold Arb. 12, 48,1931. In describing L. kajewskii , Copeland suggested that it was possibly related to L. lapeyrousii, but he apparently did not see Hooker’s type speci- men. Almost all recent collections of L. lapeyrou- sii have come from Fiji, where it is not uncom- mon, whereas Kajewski’s specimen was obtained from Vanikoro Island in the Santa Cruz group. Hooker’s type has, however, the locality given as "Vanicolla Island,” which appears to be nothing more than an older rendering of Vani- koro, so that both specimens originated from the same island. The lack of subsequent material can be due only to the infrequency of collecting there. Copeland’s paper gives the collection number as Kajewski 523, whereas a Kajewski specimen from Vanikoro at Kew inscribed "Lindsaea kajewskii n. sp.” is numbered 533. The director of the Arnold Arboretum has 64 Lindsaeoid Ferns— -BROWNLIE 65 kindly informed me that the specimen there, used by Copeland, is also 533, so that the num- ber quoted in the published description is ob- viously an error. SPECIMENS EXAMINED: From Vanikoro — Herb. Hk. without collector (Kew), Kajewski 533 (Kew). Lindsaea nervosa Mett. Lindsaea nervosa Mett. Ann. Sc. Nat. IV, 15, 62 , 1861 . A comparison of New Caledonian material collected by Vieillard showed a discrepancy for a particular collection number. Mettenius gave the type specimen of his L. nervosa as Vieillard 1540, locality Mons Dore, which is presumably Mont Dore, whereas sheets at both Kew and the British Museum (Natural History) bearing that number were specimens of Schizoloma prolongata, collected at Wagap. The Mettenius type obtained from Paris proved to be exactly what other collectors had understood as Lindsaea nervosa, and did indeed have the number and locality given in the published description. It appears that two widely separated collections have received the same number. Accidents such as this have contributed in some measure to the confusion surrounding many species from New Caledonia. Schizoloma prolongata (Fourn.) Brownlie, comb. nov. Lindsaea prolongata Fourn. Ann. Sc. Nat. V, 18, 334, 1873. L. cheiroides Fourn. Ann. Sc. Nat. V, 18, 334, 1873. Comparison of the collections at Kew and the British Museum with one of Fournier’s type specimens of L. cheiroides (Balansa, 1602) leads to the conclusion that these are all better re- garded as examples of one variable species. The name L. cheiroides has been applied to a form with a markedly elongated apical portion of the frond and with the lobing of the sterile pinnules accentuated to serrations. L. prolongata as un- derstood by Fournier and others has pinnules with rounded lobes, and the terminal division of the frond is not elongated. However, varia- tion from rounding to serration of the pinnules may be observed even within a single specimen, and since the other characters appear to be identical there are no grounds for retaining them as separate species. It appears possible that these may also grade into Schizoloma het- erophyllum var. majus (Christensen, Viert. Nat. Ges. Zurich 77, 8, 1932). On the basis of Holttum’s understanding of the lindsaeoid gen- era (Flora of Malaya 2, 320, 1954), it should be placed in the genus Schizoloma. SPECIMENS EXAMINED: MacGillivray 27 (B.M.), unnumbered (Kew), Herb. Macleay unnumbered (Kew), Vieillard 1540 (Kew and B.M.), Balansa 1602 (Paris), le Rat 943 and 965 (Kew), Franc 11 and 33 (B.M.) and 365 and 679 (Kew), Pancher unnumbered ( B.M. ) , Compton 541 (B.M.), Buccholz 1260 (Kew). Tapeinidium flavicans (Mett. ex Fourn.) Hier. Hedwigia 62, 13, 1920. Lindsaea flavicans Mett. Mem. Soc. Sc. Nat. Cherb. 10, 317, 1867 (nomen); Fourn. Ann. Sc. Nat. V, 18, 334, 1873. L. tenuifolia Mett. Ann. Sc. Nat. IV, 15, 64, 1861 ( non Bl. ) . L. balansae Fourn. Ann. Sc. Nat. V, 18, 335, 1873. L. campylophylla Fourn. Ann. Sc. Nat. V, 18, 335, 1873. Davallia lenormandi Bak. Syn. Fil. 471, 1874. Odonto soria lenormandi ( Bak. ) C. Chr. Ind. Fil. 465, 1906. This is another example of several graded forms having been described under different names. Baker’s Davallia lenormandi is repre- sented at Kew by Vieillard 1548 and 1549 and by Pancher unnumbered. These specimens show considerable variation in degree of fineness and dissection of the fronds, but all are recognized as the one species. A single sheet, MacGillivray 14, which is given by Fournier as an example of Lindsaea flavicans, is identical with the most dissected form of Baker’s species. Balansa 1652, the type of L. balansae, is another very dissected form with somewhat more open habit but is otherwise indistinguishable. This possibility had already been suggested by Christensen (Viert. 66 PACIFIC SCIENCE, Vol. XV, January 1961 Nat. Ges. Zurich 74, 60, 1929). Similarly Balansa 854, the type specimen of L. campy lo- phylla, differs only in the slightly longer ter- tiary and quaternary segments. All should be united under the one species Tapeinidium flavi- cans. It seems possible that the related species T. moorei (Hk.) Hier. may even represent a decidedly coarse form of the same species, but enough comparative material was not available to decide this with certainty. Christensen identified Lindsaea tenuifolia of Mettenius with Sphenomeris chusana (L.) Copel., which does occur in New Caledonia, but Vieillard 1548 and 1549, which Mettenius gives as his species, represent Baker’s Davallia len- ormandi. Since that has been reduced to Tapein- idium flavicans, Lindsaea tenuifolia Mett. must also be a synonym of that species. specimens examined: MacGillivray 14 (B.M.), unnumbered (Kew), Vieillard 1548 and 1 549 ( Kew ) , Pancher unnumbered ( Kew ) , Balansa 854 and 1652 (Paris), Schlechter 15046 ( Kew and B.M. ) . Sphenomeris chusana (L.) Copel. Lindsaea retusa Mett. Ann. Sc. Nat. IV, 15, 63, 1861. This appears to have been an incorrect identi- fication, because the plant is not the same as L. retusa of Indonesia. Mettenius gives his speci- men as Vieillard 1547, but an examination of the sheet used by him shows this to be a large example of the widespread Sphenomeris chusana which has been collected frequently in New Caledonia. SPECIMEN EXAMINED: Vieillard 1547 (Paris). SUMMARY Five species of the genus Lindsaea from New Caledonia and one from Vanikoro Island are reduced to synonyms either of other species of Lindsaea or of species of related lindsaeoid ferns. One species has also been transferred from Lindsaea to Schizoloma. A number of confusing aspects of collections of Lindsaea from New Caledonia have been clarified. Further collections from New Caledonia ap- pear to be necessary to elucidate the relation- ships among the three species L. macgillivrayi Carr., L. nervosa Mett., and L. neocaledonica Compton, and to define the specific limits within the genus Schizoloma. Two New Chaetognaths from the Pacific Angeles Alvarino 1 The present paper deals with two species of chaetognaths observed when working on the distribution of this group in California waters, and on their relation to the hydrographic condi- tions of the area. The material was collected by the California Cooperative Oceanic Fisheries Investigations (CalCOFI) off the coast of Cali- fornia, a project sponsored by the Marine Re- search Committee of the State of California. Acknowledgments. I wish to thank Dr. M. W. Johnson, who has given me a great deal of valuable advice and many helpful suggestions during the preparation of this paper, and who has also corrected the typescript and provided 1 Scripps Institution of Oceanography, University of California, La Jolla, California. Manuscript received July 22, 1959. space for me in the Marine Invertebrates De- partment to carry out my work on zooplankton. I am also greatly indebted to Dr. R. Revelle and Professor J. D. Issacs for their interest in pro- viding an opportunity for me to work and con- tinue my research at this Institution. I also offer my sincere thanks to Gail Holden Theilacker for doing the drawings contained in this paper. Sagitta bierii n. sp. The body is rigid and slender, retaining its cylindrical shape. It is transparent, so that the ovaries are easily seen and the gut appears as a dark straight line along the body. The small head is attached to a long narrow neck which gradually broadens out into the body, which is widest at the level of the ovaries. The shape TABLE 1 Measurements of Sagitta bierii n. sp. NUMBER OF SPECIMENS TOTAL LENGTH, MM. TAIL LENGTH, MM. OVARY LENGTH, MM. HOOKS ANTERIOR TEETH POSTERIOR TEETH Median Range Median Range Mode | Range Mode | Range Mode Range 2 17.0 3.5 3.5 (3. 0-4.0) ( 5 ( 5 (14-16) 6 16.0 3.2 (3.0-3.5) 2.5 (1.3 — 4.5 ) ( 5 6 (6-8) 14 (12-16) 18 15.0 3.08 (3.0-3.5) 2.4 (0.5-4. 5) 6 1 (5-6) 6 (6-7) 14 (12-18) 7 14.5 3.01 (3.0-3. 5) 2.6 (1. 0-4.0) 6 (5-6) 6 (6-7) 14 (12-16) 30 14.0 3.0 (3.0-3.2) 2.37 (1. 0-4.0) 6 (6-7) 6 (6-8) 14 (12-18) 33 13.5 2.9 (2.8-3. 5) 1.9 (1. 0-4.0) 6 (6-7) 6 (6-8) 14 (12-18) 42 13.0 2.8 (2. 5-3. 5) 2.0 (1. 0-4.0) 6 ! (6-7) 6 (6-8) 14 (12-18) 29 12.5 2.7 (2. 5-3.0) 1.3 (0.8-3.0) 6 [ (6-7) 6 (6-7) 14 (12-18) 40 12.0 2.78 ( 2. 2-3.0) 1.7 (0. 5-2.4) 6 (6-7) 6 (5-8) 12 (12-18) 26 11.5 2.6 (2. 5-2.9) 0.9 (0. 2-2.0) 6 (6-7) 6 (5-6) 12 (12-14) 29 11.0 2.6 (2. 2-2.8) 0.7 (0. 0-2.0) 6 (6-7) 6 (4-7) 12 (10-16) 23 10.5 2.3 (2. 2-2. 6) 0.3 ( 0 . 0 - 1 . 0) 6 (6-7) 6 (4-6) 12 (10-14) 20 10.0 2.29 (2. 0-2. 5 ) 0.14 (0.0-1. 2) 6 (6-7) 4 (4-6) 10 (10-12) 14 9.5 2.1 (2. 0-2. 5 ) 0.16 (0.0-0. 5) 6 (6-7) 4 (3-6) 10 (10-13) 18 9.0 2.1 (2. 0-2.4) 6 (6-7) 4 (3-6) 10 ( 8-12) 10 8.5 2.08 (2.0-2. 2) 6 (6-7) 4 (3-5) 10 ( 8-12) 11 8.0 2.0 (1. 9-2.2) 6 (6-7) 4 (3-4) 10 (10-14) 7 7.5 1.77 ( 1.7-2. 0) 6 (6-7) 4 (3-5) 10 ( 8-10) 6 7.0 1.75 (1. 6-2.0) 6 (6-7) 3 (3-4) 10 ( 8-10) 5 6.5 1.72 (1.6-1. 8) 6 (6- ) 3 (3-4) 8 ( 8-10) 3 6.0 1.6 (1.5-1. 7) 1 6 (6-7) 3 (3-4) 8 ( 2-10) 1 5.0 1.8 1 7 3 6 67 68 PACIFIC SCIENCE, Vol. XV, January 1961 Fig. 1. Sagitta bierii. a, Ventral view; b, dorsal view; c, hook; d, left eye. of the body resembles Sagitta decipiens Fowler. That species, however, is more transparent and flaccid and differs in the position of the fins and in having hooks without serrations. The anterior fins begin at the level of the posterior end of the ventral ganglion. They broaden at their posterior end and are without a rayless zone. The posterior fins begin at a short distance from the posterior end of the anterior fins and end at the level of the seminal vesicles. They are somewhat roundly triangular in shape and about the same or slightly longer than the an- terior fins. They are broadest behind the tail septum, but their greatest length lies in the trunk region. They do not have a rayless zone (Fig. la). The head is small, slightly elongated in the longitudinal axis, and thus differs from 5*. decip- iens Fowler, which has a larger head (Fig. lb). The hooks have strong serrations as in other species of the " serratodentata group.” The ser- rations are very conspicuous with low magni- fications (Fig. 1 c). The eye pigment forms a kidney-shaped body which in the long axis is one-half the width of the eye, with the longitudinal diameter longer than the transverse (Fig. Id ) . The collarette is small. The seminal vesicles are triangular to pear shape and not very conspicuous. The species is not as protandric as most chaetognaths (Figs. 2a, b, 3 a). In about 60,000 specimens studied, less than a dozen appeared with the seminal vesicles well developed or undamaged. The sem- inal vesicles do not touch the tail fin. The ovaries are long, fine tubes; when mature they reach to the posterior end or the middle of the anterior fins. The right ovary is usually shorter. Viewed laterally, the alternated posi- tion of the ovulae is clearly seen (Fig. 3 b) . The number of ovulae in mature specimens is from 30 to 48. LENGTH: In a well-mixed population the most abundant size is 12-14 mm. In April, 1958, from Monterey northwards, the prevailing size was from 14 mm. to 17 mm. The northern specimens all had stronger bodies and those from 13 to 15 mm. in length were mostly im- mature (Fig. 4). Two New Chaetognaths— Alvarxno 69 Fig. 2. Sagitta bierii. a, Seminal vesicles, dorsal view; b, ventral view. Immature specimens 16 mm. long with al- most no trace of ovaries were found. Others were observed with ovaries 1.0-1. 5 mm. in length. A summary of measurements is given in Table 1. DISTRIBUTION: The species was recorded by R. Bieri (1957) in Peruvian waters, from col- lections of the Transpacific Expedition, and from Cape Blanco to Punta Eugenia (Baja Cali- fornia) (Bieri, 1959). It was found in the CalCOFI collections of April, 1958, from Cape Mendocino, the northern limit of the sampling, down to the southern part of Baja California, as well as in the plankton samples from Peru- vian waters taken by the Consejo de Investi- gaciones Hidrograficas del Peru in February and April, 1958. S. bierii is found mainly from 100 to 580 mi. offshore, and hence it is not coastal in hab- itat. RELATIONSHIPS: The shape of the seminal vesicles in the various forms of the ''serrato- dentata group” is discussed by Ritter-Zahony, 191 1^; Tokioka, 1939, 1940; Thomson, 1947; and Furnestin, 1953, 1957. In the present re- port, based on material from the Pacific and the Atlantic, the differences observed are given in Figures 5, 6, and 7 to compare with Figures 2 and 3. S. serratodentata tasmanica reported by To- kioka (1959) apparently is S. bierii n. sp. for the following reasons: He says that S. selkirki Fagetti, from Chilean waters "resembles most closely tasmanica in the appearance of the seminal vesicles.” This is true and the species are certainly synonymous. The characteristic of the seminal vesicles as well as the other specific characters are identical for both. The anterior part of the seminal vesicles in both 5'. serratodentata tasmanica Thomson 70 PACIFIC SCIENCE, Vol. XV, January 1961 and S. selkirki Fagetti have numerous short, soft protuberances (Fig. 6). This being true, S. serratodentata tasmanica cannot possibly be the species Tokioka describes (1959). The seminal vesicles which Tokioka (1959, fig. 7, p. 368) illustrated and described as belonging to S. serratodentata tasmanica Thomson, differ from the descriptions and figures of the seminal vesicles for this species as given by Thomson (1947) and Furnestin (1957), and for S. sel- kirki by Fagetti ( 1958) . S. bierii n. sp., recorded Fig. 3. Sagitta bierii. a, Seminal vesicles, lateral view; b, ovary, lateral view. Fig. 4. Sagitta bierii. Distribution of size in relation to the sexual stages, April 1958. Stage I, ; Stage II, Stage III, ..... from the same area where Tokioka (1959) re- ports S. serratodentata tasmanica, has seminal vesicles with one prominence at the anterior end (Fig. 2a, b) as figured by Tokioka for S. serratodentata tasmanica. Tokioka’s (1959, fig. 12, p. 372) figure of the left seminal vesicle for ? S. serratodentata atlantica does not agree with his own descrip- tion of the seminal vesicles for this species (To- kioka, 1940, p. 373 and fig. 7d, p. 374) nor with that given by Furnestin (1957, fig. 53, p. 150). The seminal vesicle which Tokioka (1959) describes as belonging to ? S. serrato- dentata atlantica agrees with his drawing for S. serratodentata tasmanica in the same paper. In revising the chaetognaths of the " serrato- dentata group” from both the Atlantic and Pa- cific, it was easy to separate specimens belong- ing to the different species. Since the char- acteristics for each species appear constant with no characteristics in common among the differ- ent species except for the serrations on the hooks, each one can be considered to rank as a species. The fact that each one of these species typifies different masses of water offers addi- tional evidence for considering each one a good species. A revision of the name of each species appears at the top of each column in Table 2. Named after Dr. Robert Bieri, who recorded it in his work as Sagitta sp. in the " serratoden- tata group.” HOLOTYPE: USNM no. 29919. PARATYPES: USNM no. 29920. Two New Chaetognaths — Alvarino 71 TABLE 2 Principal Differential Characteristics of the "Serratodentata Group” S. serratodentata Krohn 1853 S. serratodentata atlantica Toki- oka 1940, Thomson 1947, Furnestin 1953, 1957 S. pseudoserrato- dentata Tokioka 1939 S. pacijica Tokioka S. serratodentata pacijica, Toki- oka 1940, Thomson 1947, Furnestin 1953, 1957 S. t as manic a Thomson S. serratodentata tasmanica Thomson 1947, Furnestin 1953, 1957 S. selkirki Fagetti 1958 S. bierii n. sp. Total length, mature, mm. 10-13 7-10 12-14 15-20 14-17 % tail 23-26 26-29 23-26 22-26 20.5-23.3 Body shape Same width from head to tail septum. Same. Same. Same. Neck region long, thin. Body widest at ovary region. Collarette Well apparent. Smaller than in S. serratodentata. As in S. pseudo- serratodentata. As in S. serrato- dentata. Very small, if present. Posterior fins Long, rounded. About same length on tail as on trunk. At level of tail, septum begins ray- less zone which ex- tends over anterior part of fin. Long, rounded, more on tail than trunk. Very close to posterior fins. Internal portion of fins on trunk is ray less. As in S. pseudoser- ratodentata. Very close to anterior fins. No rayless zone. Continuous with anterior fins by ex- tremely narrow connexion, not visible with low power 10X and in badly preserved specimens. More on tail than trunk. Rayless zone be- gins at level of first third of tail ex- tending over an- terior part of fin as in S. serratodentata. T riangular, rounded, more on trunk than tail. No rayless zone. Seminal vesicles Separated from tail fin, close to posterior fins. Two prominences at an- terior lateral cor- ner. Thickening of collarette tissue in front of anterior end. Markedly pro- tandric. Close to posterior end of posterior fins and at longer distance from tail fin. One promi- nence at lateral an- terior corner and small teeth at an- terior end. Pro- tandric. Oval with thicken- ing frothy tissue. 5 — 10 chitinous teeth at anterior end. Protandric. Close to posterior fins, separated from tail fin. Numerous short protuber- ances at anterior end. Notable thick- ening in front and behind vesicle. Markedly protan- dric. Pear shape. One prominence and no soft protuber- ances as in S. tasmanica. Ovaries Long tubes extend- ing to anterior end of anterior fins. Ovulae small and in one row. Fine tubes extend- ing to anterior end of ventral ganglion. Ovulae large for size of specimens, arranged in one row. Very fine tubes, ex- tending up to neck. Ovulae in one row. Long tubes extend- ing to ventral gan- glion. Ovulae in two rows. Fine tubes ex- tending to pos- terior end of anterior fins. Ovulae in two rows. 72 PACIFIC SCIENCE, VoL XV, January 1961 Fig. 5. Sagitta pseudoserratodentata Tokioka, seminal vesicles, two phases observed. Sagitta euneritica n. sp. The body is very translucent, almost trans- parent as in Sagitta setosa Muller; however, the larger specimens which mature at a size over 13 mm. in length appear less transparent. The body is uniform in width, narrowing at the head and at the tail ends. The rather poorly marked neck region is filled by a collarette. Viewed dorsoventrally on a dark background, the gut appears as a narrow white line extending from the head down to the body as in S. setosa. The anterior fins do not reach the ventral ganglion. They are wider at the posterior end and have no rayless zone. The posterior fins lie more on the trunk than on the tail. They are longer than the anterior fins, broadest behind the level of the tail septum, and they end touching the seminal vesicles. They are without a rayless zone (Fig. %a) . The head is somewhat square and regular in size. The eye pigment is gathered into a rather perfect square with several clear spots which differ in position and shape from those of 5". setosa and S. euxina Moltschanoff ( Furnestin, 1958) (Fig. 8c, d). The collarette is well developed with the characteristic structure of round cells in a net- work. The sensorial cells are spread all over the body as they are in S. setosa and S. friderici Ritter-Zahony. The seminal vesicles are in contact with the tail fin and with the posterior fins. They are not very prominent, being small and similar to Two New Chaetognaths — Alvarino 73 TABLE 3 Measurements of Sagitta euneritica n. sp. NUMBER OF SPECIMENS TOTAL LENGTH, MM. TAIL LENGTH, MM. OVARY LENGTH, MM. HOOKS ANTERIOR TEETH POSTERIOR TEETH Median Range Median Range Mode { Range Mode | Range Mode Range 2 15.5 3.2 (3.0-3.4) 2.0 (2. 0-2.0) 7- -9 6-8 14-16 7 15.0 3.2 ( 3-0— 3.4 ) 1.6 (1. 0-2.2) 8 (8-9) 6 (5-7) 14 (10-16) 7 14.5 3.1 (3.0-3.2) 1.3 ( 1.0-2. 5) 8 (7-9) 6 (5-7) 12 (12-16) 30 14.0 2.9 (2. 5-3. 2) 1.59 (1. 0-2.3) 8 (7-8) 6 (6-7) 14 (10-16) 34 13.5 2.8 (2. 5-3.0) 1.52 (1. 0-2.0) 8 (7-9) 6 (4-7) 12 (10-16) 54 13-0 2.58 (2. 5-3.2) 1.5 ( 0.4-3. 0) 8 (7-9) 6 (5-7) 12 (10-16) 16 12.5 2.6 (2. 5-3.0) 1.2 (0.6-2. 6) 8 (7-8) 6 (5-6) 12 ( 10—12) 44 12.0 2.7 (2.3-3.0) 1.2 (0.0-2. 5) 8 (7-9) 6 (4-7) 12 (10-16) 16 11.5 2.6 (2. 3-2. 6) 0.8 (0. 3-2.0) 8 (7-8) 6 (4-7) 12 (10-12) 15 11.0 2.2 (2.0-2.6) 0.66 (0.3-1. 3) 8 (7-8) 6 (4-6) 10 (10-14) 10 10.5 2.05 (2.0-2. 3) 0.56 (0. 3-1.0) 8 (6-8) 6 (4-6) 10 (10-12) 9 10.0 2.0 (2.0-2.2) 0.54 ( 0.3-1. 1) 8 (7-8) 6 (4-6) 10 (10-12) 7 9.5 1.9 (1. 8-2.0) 0.41 (0.2-0. 7) 8 (7-8) 6 (3-6) 10 ( 8-12) 6 9.0 1.8 (1. 6-2.0) 0.25 (0. 1-0.4) 8 (7-8) 4 (2-6) 8 ( 6-12) 7 8.5 1.8 ( 1.5-2. 0) 0.2 (0.1-0. 3 ) 8 (7-8) 4 (3-6) 8 ( 8-10) 6 8.0 1.7 ( 1.5-2. 0) 0.18 (0.1-0. 3) 8 (7-8) 3 (3-6) 8 ( 8-10) 6 7.5 1.7 ( 1. 5-2.0) 0.1 (0.1-0. 2) 8 (7-8) 3 (3-5) 8 ( 8-10) 4 7.0 1.6 (1.4-1. 9) 0.1 (0.1-0. 2) 8 (7-8) 3 (3-5) 8 ( 8-10) 5 6.5 1.4 ( 1.3-1. 5) 0.1 (0. 1-0.2) 8 (7-8) 3 (2-4) 8 ( 6- 8) 5 6.0 1.4 ( 1.3-1. 5) 0.1 8 (7-8) 3 (2-4) 7 ( 7- 8) 2 5.0 1.2 (1.2-1. 3) 7 (2-3) ( 6- 8) 1 4.5 1.2 8 2 6 1 4.0 0.9 7 2 6 TABLE 4 Principal Differential Characteristics of Forms Closely Related to Sagitta euneritica n. sp. S. setosa Miiller S. friderici Ritter-Zahony S. euneritica n. sp. Total length, mm., mature 10-14 10-15 10.5-15.5 % tail 16-21.7 22.6-27 19-22 Anterior fins Begin somewhat behind ven- tral ganglion. Begin at posterior end of ven- tral ganglion. Begin a small distance behind posterior end of ventral gan- glion. Posterior fins More on trunk than tail, and end at seminal vesicles. More on tail than trunk, and extend to seminal vesicles. More on trunk than tail, and extend to seminal vesicles. Collarette Very small. Noticeable. Well developed. Seminal vesicles Far from tail fin and close to posterior end of posterior fins. Touching both tail fin and pos- terior fins. Touching both tail fin and posterior fins. Ovaries Short, extend to anterior end of posterior fins. Ovulae large and in small numbers. Extend forward of anterior end of posterior fins. Ovulae regu- lar in size and numerous. Short, extend to anterior end of posterior fins. Ovulae small, filling ovaries. 74 Fig. 6. Sagitta tasmanica Thomson, seminal vesicles, ventral view. PACIFIC SCIENCE, Vol. XV, January 1961 those of S. neglecta Aida (Fig. 9), and they belong to the "bedoti- type” (Tokioka, 1939). To compare the seminal vesicles of S. euneritica n. sp. with those of S. friderici Ritter-Zahony, see Tokioka, 1955, and Furnestin, 1953 and 1957; and to compare with S. setosa Muller see Furnestin, 1958. The ovaries almost fill the width of the body and extend up to the anterior end of the poste- rior fins. The ovulae are not as large as those in S. setosa and they completely fill the ovaries. The ecology of this species is very similar to S. setosa and S. friderici which are also found in coastal waters. In the California region S. euneritica n. sp. appears in large numbers (thousands per sample) in the neritic areas. In some places they are carried offshore by local superficial currents or eddies. In these cases the total number per haul gradually de- creases as the distance from shore increases. Table 3 gives the meristic characters and armature formulae. Fig. 7. Sagitta pacijica Tokioka. a, Seminal vesicles, dorsal view; b, ventral view. Two New Chaetognaths — AlvARINO 75 It appears that S. euneritica n. sp. is closely related to S. setosa Muller and S. friderici Ritter - Zahony, but there are several points of difference as shown in Table 4 and Figure 10. DISTRIBUTION: In the collections studied, CalCOFI cruises 5405 and 5804, it occurred close to shore from Cape Mendocino to Punta Eugenia in Baja California. Dr. Fager of the Scripps Institution collected some specimens using a hand-operated dredge net, when diving off the Tijuana River mouth (June, 1958) at a 50 -ft. depth and also south of Scripps Pier (August, 1958) at 20- and 26-ft. depths. Bieri (1957 and 1959) recorded this species as S. friderici ? in the same area along the coasts of North America extending south to the waters of Peru and Chile. holotype: USNM no. 29917. PARATYPES: USNM no. 29918. Fig. 7 c. Sagitta serratodentata Krohn, seminal ves- icles, dorsal view. Fig. 8. Sagitta euneritica. a , Dorsal view; b, lateral view; c, left eye, mature specimen; d. left eye, im- mature specimen. 76 Fig. 9. Sagitta euneritica, seminal vesicles, dorsal view. REFERENCES Bieri, R. 1957. The Chaetognath fauna off Peru in 1941. Pacif. Sci. 11(3). — 1959. Distribution of the planktonic Chaetognatha and their relationship to the Pacific water masses. Limnol. Oceanogr. 4(1). Fagetti Guaita, Elba. 1958. Quetognato nuevo procedente del Archipielago de Juan Fernandez. Rev. Biol. Marina, Univ. Chile 8(1, 2, 3). Eaure, M. L. 1953. Contribution a F etude mor- phologique et biologique de deux Chaetog- nathes des eaux Atlantiques du Maroc: S. friderici Ritter-Zahony et S. bipunctata Quoy et Gaimard. Vie et Milieu 3(1). PACIFIC SCIENCE, Vol. XV, January 1961 Fraser, J. H. 1952. The Chaetognatha and other zooplankton of Scottish area and their value as biological indicators of hydrographical conditions. Scottish Home Dept. Mar. Res. ( 2 ). Furnestin, M. L. 1953. Contribution a letude morphologique, biologique et systematique de S. serratodentata Krohn des eaux Atlan- tiques du Maroc. Bull. Inst. Oceanogr. Mo- naco ( 1025). — 1957. Chaetognathes et zooplancton du secteur Atlantique Marocain. Revue des Trav- aux de 1’Institut des Peches Maritimes 21(1, 2 ). — - 1958. Les variationes morphologiques de S. setosa Muller et ses rapports avec deux especes voisines. Rev. Trav. Inst. Peches Marit. 22 ( 2 ). Ritter-Zahony, R. 191 T*. Revision des Cha- tognathen. Dtsch. SiidpolExped. 1901-1903. 12 Zool. 5(1). — 1911 b. Die Chatognathen der Plankton Expedition. Plankton Exped. Humbolt Stif- tung, 2 H.e. Rose, M., and M. Hamon. 1953. Nouvelle note complementaire sur les Chetognaths de la Baie d’Alger. Bull. Soc. Hist. Nat. Afr. N. 44 (5-6). 5 6 7 8 9 10 II 12 13 14 15 16 17 TOTAL LENGTH mm. Fig. 10. Relation of total length to tail length. S. euneritica , — — •; S. friderici, S. setosa, ..... Ac- cording to records of Ritter-Zahony, 19114, Faure, 1953, Tokioka, 1955, and Furnestin, 1958, for Sagitta friderici Ritter-Zahony. Furnestin, 1958, and Ritter- Zahony, 1911 b, for Sagitta setosa Miiller. Two New Chaetognaths— Alvakino 77 Thomson, J. M. 1947. The Chaetognatha of South-Eastern Australia. Court. SdL Industr. Res. (14). Tokioka, T. 1939. Chaetognatha collected chiefly from the bays of Sagami and Suruga with some notes on the shape and the struc- ture of seminal vesicles. Rec Oceanogr. Wks. Jap. 10(2). — 1940. A small collection of Chaetog- naths from the coasts of New South Wales. Aust. Mus. Rec. 20(6). — - 1955. Notes on some Chaetognaths from the Gulf of Mexico. Bull. Mar. Sci. Gulf Caribbean 5(1). — — 1959'. Observations on the taxonomy and distribution of Chaetognaths of the North Pacific. Seto Mar. Biol. Lab. 7(3). Polydora rickettsi, a New Species of Spionid Polychaete from Lower California Keith H. Woodwick 1 One complete polychaete specimen and two fragments collected by E. F. Ricketts in 1940 from Cape San Lucas, Lower California, are described here as members of a new species. The individuals concerned were taken from the tubes of Spirobranchus incrassatus Morch, a ser- pulid polychaete. The new species resembles other polydorids in some characteristics, but is significantly different from all known species; the differences are discussed below. Polydora rickettsi, n. sp. The body is flattened in shape in the anterior segments, becoming more rounded posteriorly. The posterior end is not sharply tapered but is only about half the width of the first few seg- ments. The modified 5 th segment is greatly enlarged. The complete specimen measured 10.0 mm. in length and included 104 segments. The anterior and posterior regions have a brown surface pigmentation, anteriorly along the lateral edges of the prostomium and posteriorly on all surfaces of the five prepygidial segments. The rounded prostomium (Fig. 1) with its lateral lines of pigment produces an anterior end closely resembling that of the common California spionid, Boccardia proboscidea Hart- man (1940). The greatly pronounced caruncle extends to the anterior margin of the modified 5 th segment. There is a slight fold in the car- uncle in the region bordered by the posterior half of the palpal bases but there is no nuchal tentacle. The eyes were not visible and may have been faded by the preservative, although the 1 Department of Biology, Fresno State College, Fresno, California. Manuscript received January 15, 1960. body pigmentation was not greatly affected. The thick palpi are short, being equal in length to the first seven segments of the worm; they are tapered at the distal end. In dorsal view the peristomium is barely visible lateral to the palpi. The 1st segment lacks notosetae but the notopodial lobe is present. The lobe is very small and is crowded in at the lateral edge of the palpal base. The neuropodial lobe is shifted dorsally but remains ventral to a line created by the notopodial lobes of the 2nd, 3rd, and 4th segments. These segments have notopodial and neuropodial lobes and setae. Their septal lines are erased dorsally by the presence of heavy longitudinal muscles paralleling the car- uncle. The enlarged 5 th segment has a heavy mus- culature which overlaps most of segment 6 and part of segment 7 dorsally ( Fig. 1 ) . Anteriorly it crowds segment 4 and has a small anterior roll or cuff in that region. The heavy muscle bands associated with the specialized setae orient from the anterior lateral portion of the seg- ment to the posterior median line. The an- terior dorsal capillary setae precede the row of specialized setae. The latter have a main fal- cate tooth and a sharply curved pointed ac- cessory tooth. There also is a flange located laterally on the main tooth at the level of the accessory tooth (Figs. 2, 3, 4). The companion setae are smaller plumose setae (Fig. 6). There is a very poorly developed bundle of neuropo- dial capillary setae. These are difficult to see and ordinarily are not taken when the special- ized setae are removed for further study. The 6th segment is about half the size of the 8th and the setal lobes are poorly developed. FIGS. 1-7. Polydora rickettsi, n. sp. 1, Anterior end, dorsal view (X 144); 2, 3, 4, specialized setae of modi- fied 5th segment, new, worn, greatly worn, each at a different angle (X 555); 5, ventral hooded hooks from the 7th segment (X 900); 6, companion setae from modified 5th segment (X 1800); 7, pygidium, in poste- rior dorsal view (X 120). 78 Polydora rickettsi — -W OODWICK 79 7 80 PACIFIC SCIENCE, Vol. XV. January 1961 TABLE 1 Comparative Characteristics SPECIES PROSTOMIUM LENGTH OF CARUNCLE BRANCHIAE FIRST SEGMENT NOTOSETAE POSTERIOR SEGMENTS, NEUROPODIAL SPINES HABITAT Polydora rickettsi n. sp. rounded (lateral pigmentation) to the 5th segment begin 7th, large to 30th absent absent Spirobranchus tube Polydora hoplura bifid to the 3rd begin 7 th, to 10th-20th last absent present (boathook) shell of Balanus, sponges; shore, bottom of ship Polydora giardi bifid anterior end or middle of the 4th begin 10th, to 25th present absent shale coralline algae Polydora anoculata bifid to the 4th or 5 th begin 11th or 12 th present absent broken shells, Amaroucium Polydora ciiiata weakly bifid to the 3rd or mid-2nd begin 7th, to 10th last absent absent burrow in shells of many forms The dorsal septal line is broken by the oblique muscular bands of the 5 th segment. The mus- cles also reach into the 7th segment, but the latter is nearly full-sized and holds the 1st gill (not fully developed), and also the first rep- resentation of the distally bidentate neuropodial hooded hooks (Fig. 5). In these neuropodial setae the main tooth forms an oblique angle, with the shaft following the angle measure- ment system of Soderstrom (1920). There are seven hooded hooks mixed with capillaries in the 7th segment; the capillaries do not persist. The branchiae are full-size from the 8th seg- ment to the 30th; they decrease in size to the 66th segment, where they are continued pos- teriorly as small papillae. There are no posterior notopodial hooks or spines. The disclike pygidium has a dorsal notch and, although not broadly flared, is greater in diam- eter than the prepygidial segments (Fig. 7). DISCUSSION Other polydorids reported from Mexican wa- ters are discussed in Rioja ( 1943 ) and include Polydora armata, P, ciiiata, P. cirrosa, P. corn- mens alts, P. flava, P. giardi, P. heterochaeta, P. ligni, P. socialis, and P. tricusp a. Of these species P. giardi and P. ciiiata most closely resemble P. rickettsi in characteristics and habitat. Some of the characteristics of these three forms, and of the morphologically similar P. hoplura and P. anoculata from other waters, are compared in Table 1. In addition to the contrasting char- acters shown in the table the five species have characteristics in common, including the fol- lowing: specialized setae with a main falcate tooth and a subterminal accessory tooth in the 5 th segment, hooded hooks beginning in the 7th segment, disclike pygidia, and the absence of nuchal tentacles. TYPE MATERIAL: The holotype and additional material have been deposited in the polychaete collections of the Allan Hancock Foundation, University of Southern California. TYPE LOCALITY: P. rickettsi, known only from the tube of Spirobranchus incrassatus Morch, was collected March 18, 1940, at Cape San Lucas, Lower California, Mexico. BIOLOGY: The specialized setae of the 5th segment, which are like those present in other boring polydorid species, and the habitat of Polydora rickettsi — Woodwick 81 this worm suggest the species to be a boring form. Steinbeck and Ricketts (1941: 368) re- port the following for Spiro branchus incras- satus, the associated serpulid species, "An im- portant feature of the low intertidal landscape at Cape San Lucas, where the anastomosing calcareous tubes of this large and spectacular worm encrust the rocks.” REFERENCES Hartman, Olga. 1940. Boccardia proboscidea, a new species of spionid worm from Cali- fornia. Jour. Wash. Acad. Sci. 30: 382-387, i %. Rioja, Enrique. 1943. Estudios anelidologicos, VIII. Datos acerca de las especies del genero Polydora Bose, de las costas mexicanas del Pacifico. Anal. Inst. Biol. Mex. 14: 229-241, 25 figs. Soderstrom, Adolf. 1920. Studien liber die Polychaeten Familie Spionidae. Dissertation, Uppsala. 286 pp., 1 pi., 174 figs. Steinbeck, John, and Edwin Ricketts. 1941. Sea of Cortez. Viking Press, New York, viii + 598 pp., 41 pis. An Ecological Perspective of Marcus Island, with Special Reference to Land Animals Shoichx F. Sakagami 1 Marcus IS A small, remote reef island in the vast western Pacific. It is located at N. 24° 20', E. 154° (Bryan, 1903), being 1,000 km. ENE. of Farallon de Pajaros (the northernmost of the Mariana Islands), 1,300 km. E. of Iwo Jima, and a little farther WNW. of Wake (Gressitt, 1954). 2 Prior to World War II the island was a Japanese dependency. Now it is a part of the Trust Territory of the United States, but there is no active establishment upon it except for a weather station belonging to the Central Meteorological Observatory of To- kyo. Through the courtesy of the Observatory, I had an opportunity to visit the island, together with Dr. N. Kuroda of the Yamashina Orni- thological Institute (birds) and Mr. M. Ya- mada of our Institute (marine invertebrates), during April 30 to May 6, 1952, and to observe its land biota. Although our observations were not extensive because of lack of sufficient time, I believe that the results are worth publishing because of our scanty knowledge of the ecology of the smaller Pacific islands and the lack of comprehensive biological research on this is- land since Bryan’s visit in 1903. TOPOGRAPHY AND SOIL TEXTURE Marcus Island is a raised atoll formed on an elevation of submarine mountains in northern Micronesia. As seen in Figure 1, it is triangular, with south and north shores of about 2 km., and the northwest shore a little longer. The 1 Contribution No. 486 from the Zoological Insti- tute, Faculty of Science, Hokkaido University, Sap- poro, Japan. Manuscript received March 19, 1959- 2 Location of the island differs slightly from one record to another: N. 24° 17' 30", E. 153° 58', ac- cording to the notification by the Tokyo Prefectural Office (1898); and N. 24° 17' 35", E. 154° 4' 30", and N. 24° 17' 02", E. 154° 1', respectively, accord- ing to observations by two Japanese cruisers, the Kasagi and the Takachiho ( Yoshida, 1902) . lagoon between the island and the fringing reef is about 200 m. on the NW. shore but is much narrower on the S. and E. shores ( Fig. 2 ) . All of the shores are lined by sandy beaches, ex- cept at the northernmost parts of the NW. coast, where the old, already mineralized reef occurs along the beach (Figs. 1, 3). The reef is con- nected with the outer ocean by means of two indentations in the E. and S. shores, respectively. Only the southern indentation is used, however, as the harbor for landing by boats (Fig. 1 c) , as large ships cannot approach the harbor be- cause of the dangerous underwater reef. The island is very flat. Formerly, the highest altitude was reported as 22 m. by Bryan (1903), but now, because of the leveling undertaken during the war, it is only 7 m. near the northern cape. Also, the trace of an old lagoon discovered by Bryan was filled up with earth by the wartime activities (Matsubara, private communication to the writer). A runway of about 1,700 m. running across the island parallel with the NW. shore and a broad road near the southern shore now divide the island into three areas, the NW. zone, the S. zone, and the E. triangle ( Fig. 1 ) . As previously mentioned, the weather station and accompanying facilities are the only estab- lishments now active on the island. But re- mains of ruined buildings constructed by both Japanese and American military forces during or after World War II are scattered everywhere. The earth consists exclusively of coral sand and pebbles. The latter vary in dimensions from mere large sand grains to pieces of gravel more than 5 cm. in length (Fig. 10). Accumulation of humus was observed only in the E. triangle, where the vegetation was relatively well de- veloped. In summary, Marcus is extremely poor in land area, soil texture, and topographical diversities. How such a poverty reflects on the land biota will be described subsequently. It must be men- 82 Ecology of Marcus Island — Sakagamx 83 tioned also that the fringing reef may serve to a certain degree as a physical barrier against the immigration of various terrestrial organisms. CLIMATE Thanks to the occurrence of a weather sta- tion, which initiated its postwar activities in April, 1951, we possess a rather precise picture of this mere heap of sand and pebbles in the vast ocean. Means of maximum, mean, and minimum daily temperatures during my stay were 25.9°, 22.7°, and 21.0° C., respectively; the average annual trends of various climatic factors are shown in Table 1. From these data, Fig. 1. Marcus Island. Drawing based upon a map used in the Observatory, show- ing Messerschmidia and Pisonia (dots), papaya (triangles), coconut palms (crosses), buildings (including ruined ones). Minute dots denote the density of Ipomoea. a , Office of weather station; h, lodging house; c, harbor; d, ruined barracks. 84 PACIFIC SCIENCE, Vol. XV, January 1961 TABLE 1 Climate of Marcus Island (The data are the averages obtained during 1952—4. Maximum and minimum values are the extreme ones noted during the four years. ) TEMPERATURE (°c.) MEAN RELATIVE HUMIDITY (%) WIND VELOCITY (m/sec) RAINFALL (mm.) Max. Mean Min. Max. Mean Total Max/hr January 29.7 22.4 17.2 72 18.3 7.6 66.9 28.5 February 28.5 22.4 16.3 75 18.1 7.5 53.5 25.7 March 29.9 23.0 18.0 76 18.4 7.8 39.5 10.9 April 31.9 24.8 18.4 78 17.6 8.1 37.8 8.7 May 333 26.3 19.8 78 13.2 5.7 48.7 15.9 June 33.8 28.1 23.1 75 12.4 4.5 43.0 23.0 July 35.3 273 22.8 78 164 6.0 252.8 59.2 August 33.7 27.3 21.8 79 15.6 6.4 189-1 23.9 September 35.3 27.9 22.8 76 16.5 73 82.4 31.0 October 33.5 26.8 21.9 78 18.9 7.1 117.8 28.6 November 34.2 26.0 22.0 77 18.5 7.1 45.7 12.3 December 31.6 23.5 18.5 73 21.6 8.6 66.1 16.5 Warmth Index, W — 245-8° C. Humidity Index, K = 5.4. it is suggested that Marcus has a relatively dry climate in spite of its oceanic position. Actually, it occupies an intermediate position between Aw and Bs of Koppen’s climate formula, al- though the differentiation of seasons is relatively less conspicuous. According to the climate clas- sification by Kira (1953), who established an excellent climate system based upon two very simple indices, warmth and humidity, 3 the is- land lies at the cool-arid corner of his type B (! ( tropical semiarid climate ) . From the climo- graphs and hithergraphs shown in Figure 1 1 , 4 together with those of Chichijima (Bonin Is.), Yap, and Honolulu, the annual cycle can be roughly divided into two seasons, namely, Octo- ber to April, which is dry, cool, and windy; and May to September, which shows the opposite 3 Warmth Index: W = T (t-5), where t = mean temperature of each month; i = number of months when t>5. Humidity Index: K = 2P/(W+ 140), where P = annual rainfall, W = Warmth Index. 1 In the hithergraph, high rainfall in July is mainly caused by an abnormally rainy weather in 1953 (500.2 mm.). In other years, 76.8 (1951), 177.1 (1952), 144.7 (1953), and 1893 0954), respec- tively. trends. Bryan also reported the danger of land- ing during October to April, because in that season the waves beat violently upon the reefs and shores. This was also confirmed in my trip by the staff of the weather station. As the is- land is located in the western part of the north- east trade-winds belt, the prevailing winds are from the east, but certain northern trends mingle during October to April Furthermore, the in- fluence of typhoons, which frequently visit in September and October, must not be overlooked. For instance, the island was completely washed by violent waves from the south to the north- west and eastern shores, when typhoon Sara passed over the island in October, 1951. Max- imum wind speed was 40.5 m/s; maximal in- stantaneous speed, 50.9 m/s; rainfall, 154.9 mm. (For the effects of typhoons, see also the Appendix. ) The climatic features mentioned above may be well explained by the location and topog- raphy of the island. Gressitt (1954) mentioned that there was occasionally found a dry local climate within the generally wet, oceanic cli- mate of Micronesia, especially in low islands Ecology of Marcus Island-— SAKAGAMI 85 and atolls. Its minute size and the poor con- servation of water by coral sand may be the main causes of the dry climate of Marcus, as in Wake Island with a similar topography and climate. Consequently, the climate of Marcus is, in spite of its subtropical position, inadequate to support a luxuriant flourishing of organic and ecological diversities. (Rain is the only source of fresh water in the island.) FLORA AND VEGETATION The flora of Marcus has been reported by Yabe (1902), Bryan (1903), and Tuyama ( 1938). The plants collected by me were kindly determined by Dr. Tuyama. They are listed in Table 2, together with those reported by the earlier publications. Comparison of the present flora with those of previous studies will be dis- cussed later. Here the discussion is limited to the plants collected by myself. Judging from the size and topography of the island, I believe that the collection of the plants which were growing there during my stay is almost com- plete. It is obvious from Table 2 that the flora is extremely poor both in number of species and in endemism. Most of the species are either cosmopolitan or tropicopolitan, or are those which behave as dominants in many com- munities because of their great vigor. In other words, we find here no more than a typical example of the poor flora of oceanic atolls. The structure of the vegetation, too, is very simple. The arboreal stratum was composed of M esserschmidia and Pisonia mixed in an ap- proximate ratio of 7:3, although the latter was relatively scarce outside the E. triangle (Figs. 1, 4). The density and resulting coverage was highest in the E. triangle and next highest along an abandoned road in the northern section of the NW. zone. In addition to these two dom- inants, about a dozen papayas were observed TABLE 2 Synoptic Table of Plants Recorded from Marcus Island SAKAGAMI YABE (1902) BRYAN (1903) TUYAMA (1938) (Identified by Dr. Tuyama) Species recorded at least in two occasions Tounefortia argentea Tounefortia serviced Messerschmidia argentea Messerschmidia argentea Cocos nucifera Cocos nucifera Cocos nucifera Cocos nucifera Morinda citrifolia Rubiaceae gen. sp. Morinda citrifolia Carica Papaya Carica Papaya Pisonia grandis Pisonia grandis Portulacea oleracea Portulacea lutea Portulacea oleracea Portulacea oleracea Tobacco tobacco Nicotiana Tabacum Boerbaavia repens Boerbaavia repens Ipomoea pes-caprae Ipomoea pes-caprae Eleusine indica Eleusine indica Species recorded only once Graminae gen. sp. Euxolus sp. Dactyloctenium aegypticum Bryophyllum pinnatum Panicum pruriens Setaria lutescens Pennisetum setosum Rottboellia sp. Syntherisma sangunalis Cenchrus ecbinatus a low trailing herb Scaevola frutescens Erigeron sumatrensis Malvastrum tricupidatum Euphorbia hirta an unknown herb Lepturus repens E. prostrata Sonchus oleraceus Boerbaavia diffusa 86 PACIFIC SCIENCE, Vol. XV, January 1961 Fig. 2 Fig. 5 Fig. 3 Fig. 4 FIGS. 2-10. Some topographical and biological as- pects of Marcus. Explanation in text. Fig. 7 Ecology of Marcus Island— Sakagamx 87 Fig. 9 along two paths penetrating the E. triangle (Figs. 1, 8). The coconut palm, which formerly had been the leading member of the arboreal stratum, was represented by only three under- nourished saplings, as is indicated by the crosses in Figure 1. The simplicity of the herbaceous layer was much more surprising. It was practically no more than an overwhelming dominance of Ipomoea pes-caprae. The density was also highest in the E. triangle, except its NW. section, but the stout runner extended its domain through- out the island except on the outermost margins of the sandy beaches. In the center of the E. triangle, this creeper constituted a pure com- munity of about 1 sq. km., excluding all other herbs; there one could walk hundreds of meters on a thick bed of intermingled vines, both Fig. 10 living and withered, without touching the soil surface (Fig. 4). Consequently, other herbs and grasses, al- though most of them were very vigorous weeds, grew only in limited areas, apparently where the pressure of Ipomoea was not conspicuous, namely, in wooded edges, roadsides, and the MW. section of the E. triangle. In such zones, Portulaca and Cencbms were dominant mem- bers of the lower stratum, and Pennisetum, Ni- cotiana, Eleusine , and S one bus of the higher one. The area richest in species was the con- fluent point of the runway and the other broad road, where most species of herbs and grasses were collected. On the other hand, no plants other than Ipomoea were discovered on the beaches ( Fig. 9 ) . From this description, it may be recognized that the island is extremely simple in both floristic and vegetational aspects. FAUNA The birds and mammals collected or observed in our survey were described by Kuroda ( 1954) . 88 PACIFIC SCIENCE, Vol. XV, January 1961 All other land animals collected or observed by me, or those later sent me from the weather station, are listed in Table 3, accompanied by notes on distribution and abundance. The fol- lowing remarks will explain the data presented in the table. 1. The number of species given in parentheses after the names of the major taxonomic groups does not always coincide with the number listed under each group, because the familial char- acters were not determined for some specimens which were not caught or were lost before or during the preparation of our data. 2. Under the column showing range, the dis- tribution of identified species in other districts is mentioned. In the majority, however, only the pattern of geographic distribution is given, using the following abbreviations: E, endemic at present; C, cosmopolitan; T, tropicopolitan, including Indo-Pacificopolitan; P, Pacificopoli- tan; and Pa, Palaearctic. These patterns are naturally very conventional, for the distinction among C, T, and P is often subjective. 3. Under the column showing abundance, the relative abundance of each species is shown with marks: + + , very abundant; ±, abundant; +, common; — , rare. The last observation may express not an actual rareness, but only a cryp- tic life-mode. 4. The distribution and relative abundance of each species in the various habitats (see the definition of A, B, etc., in the next paragraph) are indicated by O (for occurrence) or A (for abundance). Where holometabolic insects are concerned, the distribution is considered only with respect to adults, but in the sphingid and noctuid moths, only with respect to their cater- pillars, inasmuch as the adults were collected only at lights. 5. The species observed but not collected are marked with an asterisk, and those which were only indirectly confirmed are marked with a dagger. DISTRIBUTION OF ANIMALS IN VARIOUS HABITATS In order to obtain a closer perspective with regard to the ecological distribution of animals listed above, the island was divided into the following habitat zones, based upon topography and vegetation (Fig. 12). A: Areas with both arboreal and herbaceous strata (Fig. 4 and Fig. 5, back) : Ai : Floor stratum, including earth surface and sites beneath gravel and stones. A 2 : Herbaceous stratum, consisting of Ipo- moea foliage alone. A 3 : Arboreal stratum, consisting of Mes- serschmidia and Pisonia as dominants. B: Areas without arboreal stratum, with rel- atively tall grass and herbs, and with poor development of Ipomoea (Fig. 5, left) : Bi: Floor stratum corresponding to Ai. B 2 : Stratum of short grass and herbs. B 3 : Stratum of tali grass and herbs. C: Areas with short grass and herbs alone; Ipomoea cover is more developed than in B (Fig. 5, middle): Ci: Floor stratum corresponding to Bi. Co: Stratum of grass and herbs. D: Areas largely exposed, with patchy de- velopment of grass and herbs; Ipomoea cover less developed than in C: C' (Ci and C 2 ): Littoral zones corre- D' sponding to C and D in habitat structure. However, C'i con- sisted of scattered establishments of Ipo- moea frontiers alone, and D' is almost aphy- tic. H: Areas disarranged by human activities. The relative size of these habitats was ap- proximately A greater than or equal to C ap- proximately equal to D / > D > B approximately equal to C. The richness of each habitat in num- ber of species and in ecological endemicity may be roughly estimated by comparing the total species number with the number of species found exclusively in each habitat (see Table 4). Conclusions derived from these data are: 1. With respect to vertical distribution, the floor strata are far richer both in species number and in ecological endemicity than are the upper strata. Apparently, this is caused by the poor development of vegetation in the latter. 2. Horizontally, A is the richest section in Ecology of Marcus Island — Sakagami 89 Fig. 11 . Climograph (above) and hithergraph of Marcus, Chichijima (Bonins), Yap, and Honolulu. 90 PACIFIC SCIENCE, Vol. XV, January 1961 TABLE 3 Terrestrial Macroscopic Animals on Marcus Island, Excluding Mammals and Birds ABUN- | DISTRIBUTION AND RELATIVE ABUNDANCE IN VARIOUS HABITATS FAMILY SPECIES RANGE DANCE ( Ai Ao As | Bi Bo B 3 | Ci Co | D 1 Ci C 2 ' D' | H Mollusca-Orthurethra ( 1 ) .| gen. sp. I — fl O I I Annelida (1) Megasco- lecidae.. . ? Allolobophora sp | O Crustacea (6) Grapsidae Geograpsus grayi 1 (Millne-Edwards) T ++ A O O O A o o d A Coenobi- tidae f Coenohita sp (±> (O) (O) Porcellidae.... Armadillo sp. O Porcellio sp. 1 — • ' O Porcellio sp. 2 — o gen. sp — O Myriapoda (2) Mecisto- cephalidae.. Henicopidae. Mecistocephalus marcusensis Miyoshi.. Lamyctes sp E O O o Arachnoidea-Araneae (6) Pholcidae Pholcus crypticoleus Bosenberg & Strand... Pa — o Salticidae Plexippus paykulli Aud.. C + o o o o o Heteropodi- dae Heterepoda venatoria (L:) T + o o Argiopidae... Neoscona theisi (Walckenaer) T ++ A Arachnoidea- -Acari (3) Oribatidae.-.. | gen. sp I ii l i I o 1 o 1 1 Arachnoidea-Cheriferidea ( 3 ) Dithidae Ditha (Par adit ha) mar- cusensis (Morikawa).. E — O Cnthoniidae.. Lechytia sakagamii Morikawa E — O Garypidae Geogarypus ( Geogary- pus ) micronesiensis Morikawa E — o Apterygota (5) Entomo hyridae. Lepismatidae Machilidae... Drepanocyrtus terrestris Folsom Sira jacohsoni Borner. Lepidocyrtus sp. Ctenolepisma villosa Escherich *gen. sp. Ha- waii P Pa + O O o Ecology of Marcus Island-— SAKAGAMI 91 TABLE 3 ( Continued ) ABUN- DISTRIBUTION AND RELATIVE ABUNDANCE IN VARIOUS HABITATS FAMILY SPECIES RANGE DANCE Ai As A 3 | Bi Bo B 3 | Ci Co | D | C/ C 2 ' D' | H Odonata ( 1 ) Libellulidae.. * Diplacodes bipunctata Brauer P ( + ?) (O) (O) Orthopteroidea (13) Biattidae Periplaneta americana CL.) T -H- P. australasiae (L.) T ++ IBlatta sp. — O tBlaffella sp. Leucophaea surinamen- 9 • I O sis (L.) T o Anisolabi- idae Anisolabis martima (Borelli) Euborella annulipes C + o O O o (Lucas) C + o O O O Labiduridae.. Labidura sp O Gryllidae Landreva clara Walker... T ++ A O A O Ornebius sp. + o Locustidae Aiolopus famulus (Fabricius) T + O A O A Locus fa migratoria ssp.... ++ i A A O O O O O Embioptera ( 1 ) Oligo- 1 tomidae Oligotoma saundersi Westwood j Ori- | ental ++ Psocoptera ( 1 ) 1 gen. sp 1 ! + II I | 1 O 1 Hemiptera (9) Coccidae Coccus hesperidum L T + | O O o o Aphidae Aphis gossypii Glover. . . . C + o o Coreidae Liorhyssus hyalinus (Fabricius) c o Miridae Cyrtopeltis (Nesidio- coris) tenuis (Reuter) T o Nabidae Nabis capsiformis Germer C + o o o Lygaeidae Nysius pulchellus (Stal). + o o o Pachybrachius nigriceps (Dallas) P o o o Antho- coridae Gardiastethus fulvescens (Walker) T Cydnidae Geotomus pygmaeus (Dallas) T o o o o Lepidoptera (5) Sphingidae... Herse convolvuli L C + 1 o O Noctuidae Prodenia litura Fabricius T ++ | A O Achaena melicerta Drury T Arctidae JJtethesia pulchella ssp... P -H- A ? a micro-moth Hh O 92 PACIFIC SCIENCE, Vol. XV, January 1961 TABLE 3 ( Continued ) 1 1 DISTRIBUTION AND RELATIVE ABUNDANCE IN ABUN- VARIOUS HABITATS FAMILY 1 • SPECIES RANGE DANCE Ai A 2 As | Bi Bs Bs | Ci C 2 | D | CT C 2 ' D' | H Coleoptera (7) Curculio- nidae Oxyderha fusiforme Wollaston P O Cylas formicarius Fabricius T -+- o o o o o o Tenebrio- nidae Calandra oryzae L. Triboiium castaneum C o Oedemeridae. Coccinell- (Herbst) Eobia chinensis Hope C Pa -H- A o A idaf 1 Scymnus sp. o Elateridae ?Harmivius sp o Hymenoptera (6) Sphecidae Sceliphron cementarium (Drudy) Nearc- tic + O - o Vespidae fEumeninae gen. sp. Formicidae... Solenopsis geminata Fabricius Papin oma melano- T — o O O cephalum Fabricius.... T -H- A O O A O O A O o o o o Lasius niger ssp T etramorium caespitum Pa ++ A O O A O O A o o o o o L Pa — O Diptera (11) Syrphidae. ... Drosophil- idae Anthomyidae Muscidae Sarcopha- gidae Phoridae Ephydridae... Agromyzidae Sphaeroceri- dae Dolicho- podidae Ischiodon scutellaris Fabricius Drosophila melanogaster Metgen-. Atherigona excisa Thompson Musca domestica L Lucilia sericata L Parasarcophaga ( Liosar- cophaga) mis era (Walker) Aneurina sp gen. sp gen. sp. gen. sp — - gen. sp..._ T + o o o c ± o o o o o p ++ o A c + o c + o o o c + o o o + o — o o — o o — o o o o Reptilia Scincidae Cryptoblepharus bouto- nii nigropunctatus (Hallowell) Bonins ++ A O O A O A O A O O A Geckonidae. Gehyra variegata ogasa- warasimae Okada Bon ins O O Ecology of Marcus Island— Sakagami 93 both species number and ecological endemicity. This is natural because this habitat occupies more than half of the island and is biologically the most productive and stable zone. It must be mentioned, however, that A has a relatively poor fauna, depending on its very simple vegetation, as in C' 2 . 3. D' is obviously the poorest habitat because of its aphytic conditions; this conclusion, of course, pertains only to our observations upon the macroscopic animals. Bio-economically, this habitat really is the front of the marine littoral ecosystem extending into the land. On the other hand, the relatively rich number of species found in B and C, in spite of their small size, is ap- parently due to their ecotonal character. 4. The poor differentiation of Q and Bi (compare the two serial orders in Table 4) may be understood if these strata are considered as a mere extension of an ecological gradient, of which the peak lies in Ai. The structure of the floor fauna varies, therefore, at first when the plant cover almost disappears in D. C'i has also a few characteristic species corresponding to its littoral nature. DESCRIPTION OF EACH HABITAT The several habitats distinguished above must not be considered to be like cages or walled areas which confine various inhabitants within them. They are merely devices of a coordinated system for the clear understanding of the ecolog- ical make-up of the island. Eventually, certain species pass freely from one habitat, or from one stratum, to another. Before describing each habitat and its inhabitants, brief notes will be given concerning these mobile species. The rat, Rattus rattus ssp., is the only mam- mal inhabiting the island. Formerly, the staff of the weather station kept cats which con- trolled a considerable number of rats. In the absence of any intensive controls, the rats are now fairly abundant and their activities were traced everywhere in the island. The skink, Crypto blepharus, and the land crab, Geograpsusp were also seen everywhere, except ° As most recorded genera are represented by a sin- gle species, only generic names will be given in the following descriptions. Ba and D' in the case of the former species, and except B 2 , Ba, C 2 , C' 2 , and D' in the latter one. Both can climb up Messerschmidia and Pisonia to fairly high twigs. They even appear in the upper stories of buildings: crabs were often observed when they were crawling up vertical walls nearly to the ceiling. It is certain that these animals, one as a predator and the other as a scavenger, play important roles in the bio-economy of the island. Two ants, Lasius and Tapinoma, may be added to the list of widely roaming species. They were observed utilizing the runners of Ipomoea to invade even into area C' 2 » where other animals were scarcely seen. Although it is a relatively sedentary creature, a cricket, Landreva, was col- lected in almost every floor stratum except Q and D'. Its songs could be heard in the daytime, but they were more impressive at night, dom- inating this tiny bit of land in the midst of the immense ocean. Setting these mobile species aside, some char- acteristic features of each habitat will be out- lined. Zone A is the largest, richest, and most stable habitat in the island. This is also the only area where the formation of humus is relatively con- spicuous. Consequently, because of the lodging it affords various cryptic animals ( roaches, land- isopods, myriapods, etc., under stones, Oxydema in decayed wood), Ai has the richest fauna in the island. A 2 consisted of Ipomoea foliage alone. Sphinx, Prodenia, and Coccus were the major pests of the vigorous creeper. Prodenia, especially, was locally very abundant, and con- siderable damage was observed, as is shown in Figures 6 and 7. Locusta and Utetheisa, both feeding on Mes- serschmidia, surprised us by their spectacular abundance. The adults of JJtetheisa are active irrespective of diurnal rhythm. In daytime, they were seen everywhere in the A zone, feebly fluttering from one tree to another. At night they swarmed abundantly around lamps. The first instar larvae live concealed within the young sprouts (Fig. 14); older ones feed on exposed leaf surfaces, and pupae are seen near the tips of leaves, in a thin hammock spun by themselves (Fig. 13). 94 PACIFIC SCIENCE, Vol. XV, January 1961 Fig. 12. Distinction of various habitats based upon vegetation and topography. Adults and nymphs of all stages of Locusta were collected on M esserschmidia. From their extreme abundance, high activity, and great vo- racity, I have the impression that this popula- tion might change from phasis solitaria to phasis transiens. Inside still younger buds of Messer- schmidia, a small cricket, Ornehius, was often discovered. They always directed the head and antennae upward (Fig. 15), and when dis- turbed rolled down very quickly into the earth. Coccus were also found in Pisonia and, espe- cially in papaya, were eagerly visited by two milkers, Lasius and Tapinoma. Moreover, vari- ous flies and their predators, Heteropoda and Neoscona, were abundant throughout the arbo- real foliage. Considerable numbers of the latter species were found in nests of an introduced American wasp, Sceliphron. Corresponding to their ecotonal nature, B and C were relatively rich in number of species but possessed only two characteristic bugs: Cyrtopeltis on tobacco and Liorhyssus on Son- chus. The activities of skinks and land crabs decrease in B due to a relatively thick growth of herbs and grass but increase again in C. The most characteristic species in these transient zones is Aiolopus, which, in contrast to its cousin, Locusta , does not invade zone A. Aphis and its predator, Ischiodon, were found in this zone on Portulaca, the dominant plant in C and D, although the former species was found in zone A as milk cows inside a nest of Tetramorium. Solenopsis was also found only in this zone. With the further decrease of plant cover, animals adapted to bare surfaces appeared in D. The characteristic species was Oligotoma, which was extremely abundant in runways and adjacent exposed areas, dwelling in a characteristic nest spun by themselves (Fig. 16). If they were driven away from the nest, they were hunted by Lasius as soon as they were discovered by this ant. At night, winged adults were collected around the lamps situated near the runway. Zones C and D' are reproductions of C and D in the littoral zone. A characteristic animal assemblage was collected under the stones and large gravels in CT : it consisted of Geogarypus , an oribatid mite, two collembola, myriapods, etc. On exposed surfaces, however, there were very few animals, except for Lasius and Tapin- oma walking on the runners of Ipomoea. D\ especially, was macroscopically a complete abio- tic zone. The only animals collected were An - isolahis, found under the decayed matter. Al- though it did not belong to the land biota, an endemic marine collembola, Polyacanthella oce - anica Uchida, was discovered at the northern rocky reef of the NW. shore, together with some polychaetes, crabs, etc. According to a Ecology of Marcus Island — SAKAGAMI 95 staff member of the weather station, a marine strider seems to occur in the lagoon. The area receiving direct human influences possesses no more than a well-known assem- blage of domestic species. The number of species is far less than that found in similar environ- ments on continents, but, reflecting the diversity of environmental conditions, it is fairly large in comparison with other habitats, in spite of the small space. A rat, two roaches, some domes- tic flies as omnivorous scavengers, two granary beetles, and domestic silver fish, Ctenolepisma, were the chief members in or around the weather station and accompanying buildings. Skinks, land crabs, and the two ants invaded all build- ings. Gecko and Heteropoda lived there as residential predators, although they were found in the A zone, too. Earthen nests of Sceliphron were abundant on ceilings, walls, and other parts of buildings. In a ruined cottage standing near the northern point of the island, a fairly large compound nest containing 62 cells was observed attached to a broken chimney (Fig. 17). In other nests, the number of cells counted was as follows: 1 cell alone (1 instance), 2 cells (3 instances), 3 (4) , 6 ( 3 ) , 7 ( 1 ) , 8 ( 1 ) , 12 (3), 14 (1), 30 (1), 57 (1). Numerous dead insects were observed in win- dow screens of the dining room, etc., due to treatment with DDT. Examination of these ac- cumulations showed an overwhelming abun- dance of Atherigona, although the main species found within the dining room during our stay were Musca, Lucilia, and Sarcophaga. INDIVIDUAL NUMBER (Sexes not separately SPECIES counted) Atherigona excisa 1,505 Drosophila melanogaster 43 Lucilia sericata 22 Aneurina sp 22 Dolichopodidae gen. sp 18 Parasarcophaga misera 16 Agromyzidae gen. sp. 13 Sphaeroceridae gen. sp 10 Musca domestica 4 Tribolium castaneum 3 Ephydridae gen. sp 2 Gardiastethus fulvescens 1 Lasius niger ssp 1 A micro-moth 1 All of the moths listed in Table 3, as well as Oligotoma and Eobia, were attracted to lamps. Because of its oedemogenic secretion, Eobia is the only insect species injurious to human be- ings. No fleas, mosquitoes, or blackflies occur on the island. After this brief sketch of the different hab- itats, a mystifying fact must be mentioned: a dragonfly, Diplacodes bipunctata, occurred on the island, even in the absence of fresh water. The adults of this species appeared a consider- able time after our visit. I observed only a single specimen, at a passway penetrating the E. triangle, but a staff member of the weather station repeatedly confirmed the appearance of numerous dragonflies, and later he kindly sent me the specimen which was identified. If this species multiplies on the island, then not only must fresh water be available somewhere but also a number of aquatic organisms to be preyed upon by its nymphs. In the absence of any evidence of fresh water, the only other ex- planation must be the seasonal migration of this relatively delicate species across thousands of kilometers of ocean — although this is an ex- planation that I myself find hard to believe. FURTHER ECOLOGICAL NOTES Ecological interactions among various organ- isms in a given area, however few there may be, are always difficult to demonstrate clearly. But, the uncomplicated environment and simple biota of Marcus permit schematizing it as in Figure 18. Even if the schema is still far from complete in many points, the principal courses of biotic energy-flow in the island are obvious and may be classified into two major groups with respect to the energy sources: those start- ing from green plants, and those from the products of human activities. The two groups are relatively independent of each other, al- though, as discussed later, many elements con- stituting the former group were brought to the island by various human activities. As a glance at the figure will show, the extreme disharmony between the food chains and the occurrence of numerous unoccupied niches is impressive. The extraordinary abundance of a few dominant species depends, without doubt, on this too 96 PACIFIC SCIENCE, VoL XV, January 1961 simple bio-economic structure. Such dishar- mony, a common feature of remote oceanic is- lands, is also recognized by comparing the num- ber of genera and species occurring upon the island. Except for birds, the total number of families, genera, and species of land animals is 54, 70, and 72, respectively. Only seven fam- ilies contain more than two species: the Blatti- dae (5 spp. ) ; Formicidae (4 spp. ) ; Porcellidae, Entomophyridae, Lygaeidae, Noctuidae, and Muscidae (2). There are only two genera con- taining two species: Periplaneta and Porcellio. In connection with this disharmony, it may be interesting to consider here the association of closely related species, for it has often at- tracted the attention of ecologists on account of competition or isolation. However, most species belonging to families represented by more than two species show obvious habitat- segregation. Species found in one and the same habitat were Leucophanea and Blatta, Anisol- abis and Euborella in Ai and D; two species of Periplaneta, Lucilia, and Muse a in H; two Porcellio in A T ; Sira and Lepidocyrtus in C'i; Lasius and Tapinoma almost everywhere. But, most of them differ from each other either in habit, as in the two ants mentioned above, or in their relative abundance in various habitats. Those species possessing similar habits, eating similar food, and collected from the same hab- itat were only two pairs of cosmo- or tropico- politan species, Musca and Lucilia, and two Peri- planeta. In former times, the island offered a favor- able breeding site for various sea birds. Sub- sequent reckless catching resulted in a rapid decrease of both the species and numbers of individuals. During our visit, two species, noddy terns and sooty terns, still bred on the island. Moreover, about 20 golden plovers and five American wandering tattlers were seen. Of these birds, only the plovers may have an intimate relation to the land biota. They were seen usually on the surface of the runway or on other roads through the E. triangle. According to Ku- roda ( in litt. ) , some vegetable matter was found in their crops. On the other hand, he did not find any food other than cuttlefish in the crops of the terns. Therefore, terns and tattlers are connected intimately to the marine ecosys- tem but possess little relation to land biota. Finally, some phenological trends are cited here, based upon the experience of the weather station staff (especially of Messrs. Y. Nakada and K. Fujisawa), as follows: Fructification of papaya, September to October; flowering of Ipomoea, April, and September to October; flowering of Messerschmidia, March to August; nymphs of Locus ta, seen throughout the year, but abundant during July and August; Dipla- codes adults, June to July, and September to October; larvae of Herse, throughout year, but TABLE 4 Number of Species Found in Each Habitat (Those in parentheses were found exclusively in that habitat.) VERTICAL DISTRIBU- TION HORIZONTAL DISTRIBUTION TOTAL A B c D c' D' H 1 2 3 27(15) 14(0) 20(3) 13(0) 7(0) 9(2) 19(1) 10(0) 20(4) 10(3) 3(0) 1(0) 45(35 j-29(9) Total 47(19) 21(2) 23(1) 20(4) 10(3) 1 (0) Order of richness in number of species: A, D = Art r> Ci > Au > B, > C/ = C a ^ B. t > B a > o>' > d' Order of ecological endemicity: A, D > A:; = C? B« > Ci > A« = B, = C a = B a = c»' = d' Ecology of Marcus Island — Sakagami 97 Figs. 13-17. Some aspects of insect life on Marcus. Explanation in text. abundant in June; assemblage of Eobia in lamps, May to June, and August to October. These data are still insufficient but indicate the monotonous and inconspicuous phenological trends on this island. This may be also recog- nized from the occurrence of all developmental stages of Locusta, Landreva, and Utetheisa dur- ing our short stay. BIOGEOGRAPHICAL REMARKS Marcus is of little interest from the point of view of regional biogeography. According to Tuyama, who not only identified all plants col- lected by me but also kindly informed me of their distribution and ecological characteristics, all the plants are species of wide distribution and high vigor. After comparing the very sim- ple flora of Marcus Island with that of the Bonin Islands — where 46 per cent of a total of 321 species are endemic, and where five endemic genera are found (Nakai, 1930) — it is prob- ably futile to discuss the phytogeographical position of Marcus. The same conclusion can be applied to land animals. According to Gressitt (1956), the is- land belongs by its location to the Oriental Zoogeographical Region, Polynesian Subregion, Division Polynesia Proper, and Subdivision Mi- cronesia. But the order of frequency of the vari- ous distributional patterns is: Tropicopolitans (including Indo-Pacificopolitans ) (18 spp.); Cosmopolitans (13); Pacificopolitans (6); Pan- Palaearctic (5); Endemic (4); Species with a 98 PACIFIC SCIENCE, VoL XV, January 1961 limited range (4). Distinction of these patterns is rather arbitrary but may be sufficient to con- clude that most of the species belong to types which can hardly be said to be the regional, although in general the Oriental elements are predominant. It is remarkable that four endemic terrestrial species, one centipede and three pseudo-scor- pions, were discovered upon this tiny island. One of the latter group, Lechytia sakagamii Mo- rikawa, is very interesting because it belongs to a genus which, up to the present time, has been recorded only from Nearctic, Neotropical, and Ethiopean regions (Morikawa, 1952). FORMATION OF LAND BIOTA The land biota described above has been com- pared to the earlier results published by Yabe (1902), Yoshida (1902), Bryan (1903), and Tuyama (1938). The plant species reported by those writers and by me are given synoptically in Table 2. From this table and from informa- tion kindly given me by Mr. Matsubara, the commander of Japanese Marcus Garrison dur- ing World War II, we can trace the floristic change of the island during the last 50 years. With respect to trees and shrubs, only Cocos and M esserschmidia have continued to exist throughout half a century. This combination, one of the commonest edaphic climaxes on sandy beaches of the Pacific islands, in all prob- ability had been already well established when the island was discovered. Later, but before 1938, the island received Pisonia as a new member of its flora, and it is now a chief member of the vegetation. On the other hand, Morinda dis- appeared between 1938 and 1940, because this was reported by Tuyama but not by Matsubara. Although still surviving at the present time, the coconut palms received remarkable damage from human interference (cf. Appendix). When Bryan visited the island in 1903, palms grew densely in the central area of about 3 acres. According to Matsubara, there were only 30 trees, about 4.5 m. high, when he arrived upon the island in 1941. Half of them were cut down at the end of that year. Moreover, as seen from the Appendix, all trees on the island were com- pletely damaged by repeated bombing during the war. The present arboreal stratum is, there- fore, the outcome of postwar regeneration. The origin of papaya now existing in the island is obscure. Bryan gave seeds of various plants, including papayas, to the Japanese in- habitants when he left the island. Later Tuyama reported this plant from the island. But no papaya trees were growing in 1941 according to Matsubara. He planted a few seeds in 1945, and some seedlings grew to the height of a child before being damaged by bombing. The plants now growing in the island seem to have been brought in by the U. S. Navy after the war. Of the herbs and grasses, tobacco and Por- tulacea are the only species reported by all writers, including myself. Judging from the small size and simple topography of the island, which permit one to walk around it within 2 hours, it is hard to believe that any abundant plant species escaped the eyes of other col- lectors. 6 Therefore, the lack of accord among four collections suggests the unstable character of the herbaceous strata, with new inhabitants appearing and being replaced in their turn by other ones, under the influence of human activ- ities during the last 50 years. Ipomoea was first reported in 1938, but Matsubara wrote me that in 1943 it was found only in scattered patches on the island. The overwhelming dominance of this species throughout the island at the present time is, therefore, a postwar event. Previous information concerning land animals is scanty. The most important change may be the extinction of numerous sea birds which bred on the island. A catastrophic decrease may well be recognized if the report of Bryan ( 1903) is compared with that of Kuroda (1953). With respect to other land species, Yoshida (1902) briefly described a skink, gecko, "flies,” "red moths,” and "small flies.” Bryan also re- ported a skink ( Ablepharus boutonii ) and a gecko {Perochirus articulatus) . Therefore, both have been constant inhabitants during 50 years, although their scientific names have been changed since Yoshida s visit. Among three land crabs mentioned by him — Grapsus grap- 6 Actually, except for Bryophyllum, all of the plant species collected by me were discovered on the first day of our survey. Ecology of Marcus Island-— SAKAGAMI 99 H B C ' -8 & y Z® n o m, or 7* f£^rru£a,di//m\ \ [Lechytra., l~«.myctes\ P/SOM ft Xfeastoc,^m,lus J f '^^adreim, 1 JS£eLUA,lenco^a.H€A PAPAYA \^uborel/cL, JlruSofaSis bem&mas M "ffpo/ym hoNCHosp /^TOBACCO 4- • — Landresm, ’'■■■'• 3ttfctev.-T 3 CoTvsumet^j ^T^W«.TCt.s" [STORED G&MWlfr——Ca.tanctri£L, Jriho/itun JMtdiajstetAiis 1 "'‘\ ! FRESH FOOD P^^Misca. L allied , f. ! ) Imumam waste *\fertplAn€tcL,Anisdc£iy ip&P^Jipkfs • 4 OTHER PRO Pf<-QW^W \2k .Urfe/W« /ftdpGekini 1 . /7j | yy UfetAtsio, ( larvae)] J J^Eumenh ''Ornehius Umd \sopo^f£^* s W /o 'A^0C 1 ^*'' ftachilcd ^Qxydema Pl&ZtppUSj , HVkoSCOna. I . f .occus ?<— 1 Various flies XAfefg^oA' “/Hicroi€pidfipteTOLCla.rira^ j\. ., PobsA i ~^Artts/ 7etra.monu.fi 'Herse PmdefiiCL (.larvae) ^/as"~ - - — ----/ PngyMus ^^ Anlc/l } us \TapmomJ •-LMreutV^ RAM,NA 6 ^— i— ^ 1 1 ■ — ga^re//aN QTHeR '1/POMEA AanJmk XPORTULA CB4 €y/as -sfyt/lisP \f scktodon* Ants(‘ <"i \ Lajsius \ TapinjomOLj Gcotofnus j | OTHER WEEDSl^— 7^ — PSOCOPTERA J —Olipdtonm, » —Dt^o-nocyrtus^ POffEA (jHuAorell jj/OTHER WEEDS / d aMdureLPpP' ________ «&7*a , LepifoeWfuix -Oribdtid^' / — -fbretllio 1 r S^o^arypttes) fTfecist&eep/m/us) -Anisolo-Pis j Ciypthi I^koms - — —M£tus Geoyrapsus- -Plover- Fig. 18. Food-nexuses in Marcus Island. sus (abundant), Geograpsus crisipes (less abun- dant), and G. grayi ( abundant ) —only the last species still remains on the island. According to a personal communication from a staff mem- ber of the weather station, he is sure that at least one species of land hermit crab still exists on the island. But it is uncertain whether this species is either Coenobita olivieri or C. com- pressa reported by Bryan, for no specimens were collected by myself. It is very regrettable that Bryans collection of insects, which he made by various methods (including lantern collecting, barking, attract- ing with decaying flesh, etc.) was damaged by ants and other pests during his return voyage. His miscellaneous notes based upon memory are so interesting, however, that I will cite them here: A small red ant was quite common as well as troublesome, especially about the settlements. I fancy it had been imported since the colony was established. Two species of flies were very abun- dant, one a blowfly ( Callipbora ? ) which per- sisted in laying its eggs on the birds both before and after they were skinned; the other species, a small vinegar fly of a genus unfamiliar to me, 100 PACIFIC SCIENCE, Vol. XV, January 1961 . . . were to be seen in moist, shady places all over the island. A small miller was common dur- ing the night, and I am of the opinion that the skinks and geckos feed on it as well as on the small flies just mentioned. . . . The only spider that had established itself was the widely distributed web-spinning species, Epeira nautica. . . . Trees and grass showed little or no signs of insect pests. In fact, I found only one species of plant that had been molested by biting insects. Since these depredations were to be seen only in a very limited area, and as I was unable to secure the miscreant either by day or night, I concluded the species must have been a recent Japanese introduction that had not had time to thoroughly establish itself. No species of Coleoptera were secured [pp. 117-8}. No land shells were noted, and I believe there were none [p. 120}. Believing that a collection of any earthworms that might occur on the island would be of interest I requested Mr. Sedgwick and his as- sistants to keep a close lookout for them. Al- though they made a large number of excavations in various places while prosecuting their inves- tigations, they were unable to discover a single specimen. ... I am persuaded that worms of this class have not as yet found their way thither [p- 122}. Compare these citations with results obtained by me, and remember that both surveys were made approximately in the same season and during the same interval (cf. Appendix, 1902). It may be assumed that, in all probability, many species now inhabiting the island were estab- lished there after 1902. Only Neoscona theisi (Walckenaer) (= Ep- eira nautica Bryan nec Koch) and, seemingly, some domestic dipterans are the inhabitants col- lected in both surveys. It is not certain whether Bryan’s "red, small ant” corresponds to any of four ant species collected by me or not. But it surely differs from Lasius niger, the com- monest ant in 1952. Consequently, judging from their conspicu- ousness and present abundance, the following species may safely be regarded as immigrants since 1902: Plexippus paykulli Prodenia litura Heteropoda venatoria Herse convolvuli Periplaneta americana Sceliphron caementarium P. australasiae Landreva clara Locusta migratoria ? Allolobophora sp. A land snail Lasius niger Calandra oryzae Cylas formicarium Tribolium castaneum Eobia chinensis Although with less certainty, the following species are also assumed to be relatively recent immigrants: Armadillo sp. Ctenolepisma villosa Euborella annulipes Anisolabis martima Aiolopus tamulus Oligotoma saundersi It is uncertain whether or not a "red moth” mentioned by Yoshida corresponds to Utetheisa. But, from Bryan’s notes, it is highly probable that the number of individuals was very small, even if this species was present in 1902. Matsubara wrote me only about the skink, gecko, land crab, flies, and cricket as being the impressive animals during his wartime service. From these accounts, we can assume that Land- reva was established before 1943. On the other hand, Locusta, Eiobia, and Periplaneta must have arrived after World War II, for these animals, if they occurred, certainly would have attracted the attention even of persons not biologically observant, either by their conspicuousness (as in Locusta) or by their sanitary importance. Needless to repeat, the species now most abun- dant are, in general, the relatively recent im- migrants. Thus, most members of the land biota of Marcus are immigrants since 1902. Consider- ing the extremely isolated location of the is- land, it must be obvious that most of these species gained their chances to arrive on the island and to establish themselves there only through direct and indirect human interference at the island. My conclusion, therefore, is that the present land biota is, in its origin, largely an outcome of human activities directed upon the island. CONCLUDING REMARKS AND GENERAL CONSIDERATIONS In connection with the zoogeography of Pa- cific islands, Gressitt (1956) gave an appro- priate summary on the nature of land fauna in Ecology of Marcus Island — SAKAGAMI 101 low coral islands: "Atolls and other low coral islands have a small fauna — similar in widely separated groups of islands— -which is limited by the lack of ecological diversity, the limited haplophytic strand flora, the presence of brack- ish ground-water, the scarcity of soil, and ex- posure to salt-spray.” The land biota of Marcus, with its extreme poverty in both taxonomic and ecological com- ponents, offers nothing other than a very typical example of Gressitt’s generalization. He also wrote: "The extent of speciation is directly re- lated to the island’s age, size, isolation and diversity of environment.” This proposition can be applied to biocoenology if the word "specia- tion” is replaced by the phrase "differentiation of ecological components.” On Marcus Island the isolation is fairly great, but its size and its diversity of environment are incomparably small to be able to promote any ecological differentia- tion. Moreover, this isolation may modify a given biota only when human interference is absent or at least negligible, because this factor acts, however locally, with an incomparably more rapid tempo and more violent means than do other natural agents. It would be rather surprising if Marcus Island had maintained any ecological peculiarities — even if such had existed in this most simple environment — despite the accumulation of various human interference dur- ing 50 years, including intensive skinning, co- conut collecting, public works which modified the appearance of the island, a high human pop- ulation during wartime (when 4,000 persons were living on this mere heap of coral sand and pebbles), and, finally, violent bombing. However, although Marcus Island may be little more than a disappointment to biologists who approach the island to study its flora or its biogeography, investigation of such an undif- ferentiated biota does reveal some important problems, as follows: 1. Our knowledge of the ecology of Pacific islands, as mentioned by Gressitt ( 1954) , is still very far from complete. In this account, the study of a relatively simple biota as that of Marcus may serve as a useful guide either to study more complex biotas or to find general principles underlying their diversities. 2. Considering the fact that any given eco- logical assemblages, either simple or complex, consist of interactions among numerous parts and processes, it is obvious that the analysis of such entities is far easier to do in simpler biota than in more complicated ones. It should be remembered that, while we may be interested in discovering any specificities and comparing them with each other, we must always seek gen- eral rules governing such specificities. 3. Because of their extreme isolation, eco- logical simplicity, and lack of industrial im- portance, the remote low islands such as Marcus may serve as the best laboratories in field eco- logy for the study of the intra- and inter- specific ecology of given species, both native and purposely introduced, as living isotopes. The clarification of land biota should be a pre- requisite for such experimental studies. ACKNOWLEDGMENTS The present survey was made in close col- laboration with Dr. N. Kuroda of Yamashina Ornithological Institute, and Mr. M. Yamada of the Zoological Institute, Hokkaido University, under a plan prepared by Professor Tohru Uchida of Hokkaido University. The survey was made with the permission of Dr. S. Wadachi, Head of Central Meteorological Observatory in Tokyo, aided by numerous staff members of the Observatory, especially Mr. N. Yamada, Chief Secretary, Mr. Y. Nakada, Head of the Remote Islands Section, Mr. T. Doi, Head of the Supply Section, Mr. H. Hasegawa of the Entomological Laboratory, the National Institute of Agricultural Sciences, Tokyo, and Mr. H. Okuyama, Botanical Laboratory of the National Science Museum, Tokyo. The specimens collected by me were identi- fied by the following gentlemen, who also gave me valuable information on the distribution and habits of the species collected: Mr. T. Aoto ( Reptilia ) , Dr. S. Asahina (Odonata), Mr. S. Ehara (Acari), the late Dr. T. Esaki (Embiop- tera ) , Mr. H. Hasegawa ( Heteroptera ) , Mr. K. Hori (Muscidae and Sarcophagidae ) , Dr. A. Kawada ( moths ) , Mr. S. Kato ( Anthomyidae ) , Mr. M. Konishi (Cossoninae) , Mr. K. Kosugi ( Coleoptera ) , Mr. Y. Miyoshi (Myriapoda), 102 Mr. K. Morikawa ( Cherif eridea ) , Mr. M. Mo- ritsu ( Aphidae), Dr. T. Nakane ( Coleoptera ) , Mr. N. Nozawa ( Orthopteroidea ) , Mr. H. Ni- shijima (Diptera), Dr. S. Saito (Araneae), Dr. T. Sakai (land crabs), Dr. T. Shiraki (Orthop- teroidea), Dr. R. Takahashi (Coccidae), Dr. K. Tsuneki (ants), Dr. H. Uchida (Aptery- gota), Mr. T. Yaginuma (Araneae), Dr. E. Yamaguchi ( Oligochaeta ) , and Dr. T. Tuyama (plants) . Some names have been added or changed on the basis of studies done upon the insects of Micronesia by J. C. M. Carvalho, H. G. Barber, and H. de Souza Lopes not cited in my ref- erences. Valuable information on the animals and plants during wartime was obtained from Mr. M. Matsubara, the commander of the Japanese garrison on Marcus Island during World War II. Messrs. Y. Nakada and K. Fujisawa of the Remote Islands Section of the Observatory gave me suggestions on the land biota. Dr. T. Kira of Osaka City University kindly answered my inquiries on the climatic and vegetational fea- tures of the island. Suggestions for improving the manuscript were given by Dr. J. L. Gressitt of the Bernice P. Bishop Museum, Honolulu, based upon his wide biological knowledge on the Pacific is- lands. I should like to express my sincere thanks to all of these gentlemen, whose help was in- dispensable in preparing the present paper. SUMMARY Based upon information obtained directly dur- ing the period from April 30 to May 6, 1952, and from previous works and personal com- munications, a general perspective of the land biota of Marcus Island in the western Pacific is outlined. As might be expected from the small size and lack of environmental diversity, the land biota shows the typical poor structure com- mon to low reef islands of the Pacific. Most constituents of the biota seem to have been in- troduced during relatively recent years, prob- ably aided by direct and indirect human activ- ities upon the island. PACIFIC SCIENCE, Vol. XV, January 1961 APPENDIX AN ANNOTATED HISTORY OF MARCUS ISLAND As mentioned by Bryan, the discovery, nam- ing, and early history of the island cannot be thoroughly traced in the obscurity of the chron- icles from the last century. The following table was prepared from the accounts of Yoshida ( 1902 ) , Bryan ( 1903), and Shiga ( 1903 ) , and from personal communications from Mr. Ma- tsubara and staff members of the weather sta- tion. Before I860: Some reports of Pacific whalers give some information on the island, but with much confusion about its name and location ( Bryan ) . 1868: Captain Kilton, aboard the "David Hoad- ley,” visited in May and described the place as a low sandy island covered with trees and bushes (Bryan) . Discovered in this year by an American, and thereafter visited occasionally by French and British ships (Shiga). 1874: U. S. survey ship "Tuscarora” (Com- mander Belknap ) visited. The Hawaiian Mission ship "Morning Star” ( Captain Gelett) visited and reported a dense cover of trees and shrubbery, with a white sandy beach ( Bryan ) . Tsunetaro Shinzaki visited as a passenger in a British ship (Yoshida) . This was the first visit by a Japanese (Shiga). 1889: Captain Rosehill landed in June while engaged in trading in the Pacific. He rec- ognized the island’s value as a source of coconuts. Believing himself to be the dis- coverer, he claimed it for the United States (Bryan, Yoshida) . 1896: A stone lantern (Ishi-doro) , with an inscription of February 12, 1896, written in Japanese, existed on the island until its destruction by U. S. bombers during World War II ( Matsubara ) . Shinroku Mizutani, Chief of the South Sea Section, Tokyo Animal Company (Tokyo Kinju Gaisha) , while he was a sailor aboard the "Tenyu-maru,” was cast ashore in a storm (Yoshida, Matsubara). Ecology of Marcus Island— SAKAGAMI 103 1898: In July, the Tokyo Prefectural Office claimed the island as a Japanese depend- ency, named it Minami-Torishima (South Bird Island), and incorporated it into the Ogasawara Section (the Bonins) of To- kyo Prefecture (Yoshida). In September, tenanting the island from the Tokyo Pre- fectural Office, Shinroku Mizutani began the skinning of sea birds, aided by the in- vestment of Shichigoro Kamitaki, a trader in Yokohama (Yoshida) . Haruzo Ogawa, a lieutenant in the second reserve of the Japanese Navy, called the inhabitants of Hachijozima and of the Bonins to Marcus Island for help in skinning the sea birds ( Matsubara ) . 1899-1902: According to grave posts (now missing), three Japanese died in the island during these years (Matsubara). 1901: In October a violent typhoon attacked the island for 10 days, sending the sea as far as 22 ft. above the normal level (Bryan). 1902: Hearing of Captain Rosehill’s expedition (see below), the Japanese Government sent the cruiser "Kasagi” to the island. Aki- yuki Toyoguchi, a sub-lieutenant, landed with 15 men (July 27). Captain Rosehill arrived at the island on July 30, accom- panied by Dr. Bryan and Mr. Sedgwick, in order to claim it as a U. S. territory, but left on August 5 because of its occupation by the Japanese Navy. Bryan and Sedgwick made a scientific survey of the island dur- ing the 5 days. August 28, the Japanese Government sent another cruiser, the "Ta- kachiho.” S. Kamitaki (a trader mentioned above), S. Shiga, M. P., and O. Yoshida, a geologist, landed. Two Japanese shrines, Kotohira and Ohtori, were built there ("Tengaisei”). September 2, a typhoon passed over the island. All inhabitants sought safety at the highest point. Until December 25, no food other than birds and fish was available. Sixteen died during this period (Nakada). In September, the Japanese Department of Foreign Affairs again claimed the island for Japan. The following publications appeared: Plants of Marcus (Yabe), Miscellaneous notes on the geology and topography (Yoshida) , Chron- icle of a journey to the island ("Tengaisei”). 1903: Shiga published an essay describing the discovery of this island. He asserted its im- portance from the national standpoint. Bryan’s comprehensive monograph was published. Han-emon Tamaoki, a Japanese, went to the island to collect coconuts but left without success (Matsubara). 1906-16: Many Japanese were landed for phos- phate mining. Nineteen died during these years (Matsubara). 1930: In November, all 32 inhabitants, who had been engaged in coconut collecting and fishing, left the island (Matsubara). 1931: The island was purchased by the Japanese Navy ( Matsubara ) . 1935: The Hydrographical Department of the Japanese Navy began meteorological ob- servations ( Matsubara ) . 1937: Establishment of the Japanese Navy air- port commenced (Matsubara). 1938: Tuyama published his Flora of Marcus Island. 1939: February 22, a large flock of terns visited the island. March 15-16, a large flock of "swallows” passed through. Terns and swallows appeared also in autumn (Ma- tsubara ) . 1941: The island was armed with six 15 cm. cannons and six 8 cm. aeroguns (Matsu- bara). Japan declared war upon the United States. 1942: March 4, the island was bombed by 40 U. S. carrier-based planes (Matsubara). 1943: A garrison consisting of 1,100 navy, 2,250 army, and 650 civilian personnel was in- stalled with M. Matsubara as commander. 1944: May 20-21, bombed by 132 U. S. car- rier-based planes. October 9, bombarded by a U. S. naval squadron consisting of one battleship (Pennsylvania type), two heavy cruisers (Pensacola type), and five large destroyers (Matsubara). 1945: Received 171 attacks by a total of 759 bombers from September, 1944, to the armistice on August 15, 1945. October 7, the Japanese garrison left the island (Ma- 104 PACIFIC SCIENCE, Vol. XV, January 1961 tsubara). November, occupied by U. S. Navy (Weather Station). 1946: U. S. Navy left the island because of the great damage to the establishment by typhoon Martha (Weather Station). 1950: The Central Meteorological Observatory in Tokyo made a survey in order to re- establish the runway and weather station on the island (Weather Station). 1951: Meteorological observations began again on the island (Weather Station). REFERENCES Bryan, W. A. 1903. A monograph of Marcus Island. Bernice P. Bishop Museum Occas. Pap. 2(1): 77-126. Esaki, T., E. H. Bryan, Jr., and J. L. Gressitt. 1955. Insects of Micronesia, II. Bibliography. 68 pp. Gressitt, J. L. 1954. Insects of Micronesia, I. Introduction. 257 pp. 1956. Some distribution patterns of Pa- cific island faunae. System. Zool. 5: 11—32. Ima, [?]. 1904. Minami -Torishima ( Marcus Island). Chigaku Zasshi 16: 730. (In Jap- anese. ) Kira, T. 1953. In: Handbook of New Geog- raphy (Tokyo) 4: 256-267. (In Japanese.) Konishi, M. 1955. Cossoninae of Marcus Is- land (Col., Curculionidae) . Ins. Mats. 19: 64. Kuroda, N. 1953. On the subspecific name of the sooty tern of Marcus Island. Misc. Bull. Yamashina Inst. Ornith. Zool. 2: 17-21. (In Japanese, with English resume.) 1954. Report on a trip to Marcus Is- land, with notes on the birds. Pacific Sci. 8: 84-93. Miyoshi, Y. 1953. Beitrage zur Kenntnis jap- anischer Myriapoden, 9. IJeber eine neue Art von Leptodesmidae (Diplopoda). Anhang- sel; Eine neue Art von Mecistocephalidae aus Marcus-Insel. Zool. Mag. Tokyo 62: 186- 188. (In Japanese, with German resume.) Morikawa, K. 1952. Three new species of false scorpions from the Island of Marcus in the west Pacific Ocean. Mem. Ehime Univ. Sec. II (Sci.) 1: 241-248. Nakai, T. 1930. The flora of Bonin Islands. Bull. Biogeogr. Soc. Japan 1: 249-278. Sakagami, S. F. 1953. A trip to Marcus Island. Shin-Konchu 6(5): 23-29. (In Japanese.) Shiga, S. 1903. Minamitorishima and the North Pacific problem. Chigaku Zasshi 15: 42. (In Japanese.) "Tengaisei .” 1902. Chronicle of a trip to Mi- nami-Torishima. Chigaku Zasshi 14: 683. (In Japanese.) Tuyama, T. 1938. Plants of Marcus Island. J. Jap. Bot. 14: 425-426, 554. (In Japanese.) Yabe, Y. 1902. Plants of Marcus Island. Bot. Mag. Tokyo 16: 258. (In Japanese.) Yoshida, O. 1902. An inspection to Marcus Island. Chigaku Zasshi 14: 674-678. (In Japanese.) Two New Species of Chaetognatha from the Waters off Peru Paul N. Sund 1 Since February, 1958, the Consejo de Investi- gaciones Hidrobiologicas of Peru has been mak- ing regular cruises to investigate the biological and hydrographic conditions off the coast of Peru. Through the kindness of Dr. Z. Popovici, I have been able to examine the Chaetognatha from plankton tows taken in the course of these investigations. The tows were made at the sur- 1 Inter-American Tropical Tuna Commission, Scripps Institution of Oceanography, La Jolla, Califor- nia. Manuscript received April 29, I960. Sagitta peruviana n. sp. Fig. 2 A-L Holotype, 1 specimen, USNM no. 29921 Paratypes, 10 specimens, USNM no. 29922 Body rigid, opaque, hispid. Tail segment 23-29 per cent of total length. Anterior fins com- pletely rayed, start at posterior end of ventral ganglion. Posterior fins separate from anterior fins, completely rayed, reach posteriorly to seminal vesicles; greatest portion on tail segment, widest behind tail septum, about same length as anterior fins, posterolateral margin somewhat concave. Caudal fin triangular with rounded lateral apices; in contact with seminal vesicles. Anterior teeth 4-9; posterior teeth 10-21; hooks 7-9. Seminal vesicles, when mature, with expanded anterolateral corner. Ovaries may extend to neck; ovaries with ova not filling body cavity, ova in two dorsoven- trally placed rows. Collarette conspicuous, extends posteriorly to approximately Vi distance between head and ventral ganglion. Corona ciliata extends from between eyes to about 2 A distance from head to ventral ganglion. Intestinal diverticula absent. Eye pigment dense, apparent shape quadrangular, with elongate extension arising medially from center of median side of pigmented area. BODY LENGTH IN MM. TAIL SEGMENT Vo TOTAL HOOKS ANTERIOR TEETH POSTERIOR TEETH MATURITY STAGE LENGTH OF OVARY IN MM. 13.2 24.0 8 9 21 IV 7.1 12.8 24.2 8 9 19-21 II-III 2.1 11.9 24.4 7 9 15-18 K?) 1.1 11.4 24.5 7 — 17-18 Ill 1.3 11.3 25.7 8 — — III-IV 4.1 10.9 25.7 8 8 20 hi 4.1 10.5 25.7 7 — — ii-iii 1.2 10.4 24.0-25.0 7-8- 7-8 16 III-IV 2. 5-2. 7 10.1 25.7 8 6 19 in 2.9 9.8 23.0 8 6 12 III-IV 3.0 9.1 24.2 7 5 13 III 2.7 8.4 25.0 8 8 15 Juv. — 7.2 26.4 8 5 13 Juv. — 6.8 25.0 8 5 12 Juv. — 5.8 29.3 9 4-6 10-11 Juv. — 5.3 26.4 8 4 10 Juv. — face, using Vi -meter nets. The first specimens of the two new species reported were noted in a plankton sample taken on March 5, 1958, near the entrance to the Port of Talara, Peru. Inspection of other samples taken during the expedition (cruise 5802) re- vealed numerous individuals of one of the new species, Sagitta peruviana ; but none of the sec- ond species, Sagitta popovicii, was noted (Fig. 1 ) . The descriptions of the new species are based on specimens preserved in formalin. 105 106 PACIFIC SCIENCE, Vol. XV, January 1961 Fig. 1. Locality records of S. peruviana and S. popovicii. Figures indicate number of specimens present in sample. Chaetognaths from Peru — S und 107 Fig. 2. Sagitta peruviana n. sp. A, Entire animal, ventral view; B, ventral view of head; C, anterior teeth; D, posterior teeth; E, typical hooks; F, detail of tip of hook; G, eyes, showing pigment distribution; H, dorsal view of anterior end of animal; l-L, stages of development of seminal vesicle. 108 PACIFIC SCIENCE, VoL XV, January 1961 FORMULA: The body form and general shape of the lateral fins is reminiscent of S. bipunctata Quoy et Gaimard 1827 and S. hispida Conant 1895. But the fact that the posterior and caudal fins of S. peruviana both are in such close proximity to the seminal vesicles distinguishes it from either of these two species. The seminal vesicles in the later stages of development resemble those of S. neglecta Aida 1897 and S. friderici Ritter-Zahony 1911. Features that separate S. peruviana from S. neglecta and S. friderici are the robust body form, length of ovary, size of ova, shape of the lateral fins, and shape of the eye pigment. The fact that the new species has no intestinal diverticula separates it from both S. neglecta and S. hispida. Table 1 compares the features of the species discussed above. Sagitta popovicii n. sp. Fig. 3 A-F Holotype, 1 specimen, USNM no. 29923 Paratype, 1 specimen, USNM no. 29924 Body short, translucent, rigid; constriction at tail septum inconspicuous. Anterior fins start at posterior end of ventral ganglion, completely rayed, slightly tapered, widest near posterior end. Posterior fins separated from anterior fins by short but distinct interval, wider than anterior fins, widest posterior to septum, completely rayed, reach seminal vesicles. Caudal fin with rounded pos- terior margin, reaching seminal vesicle. Anterior teeth 3-4; posterior teeth 6-10; hooks 6-7. Sem- inal vesicle large, opaque, very conspicuous. Ovaries reach posterior region of anterior fins. Ova large, filling body cavity; up to 9 or 10 ova in a single row. Collarette present, small. Intestinal diverticula absent. Corona ciliata not observed. Pigmented area of eyes with an elongate extension arising from median side. The above description is from four specimens, three of which were in such condition that the armature formulae and measurements could be determined. All four specimens were fully mature. The one specimen for which the measurements and formula were not made was found preserved in a distorted position. FORMULA: BODY TAIL LENGTH LENGTH SEGMENT ANTERIOR POSTERIOR MATURITY OF OVARY IN MM. °/o TOTAL HOOKS TEETH TEETH STAGE IN MM. T8 32^8 7 3 6^7 IV LI 5.9 (2 indiv.) 26.3-28.6 6-7 3-4 9-10 IV 1.1 REMARKS: The present new species is similar in size and certain characteristics to S. bedfordii Doncaster 1903 (redescribed by Tokioka, 1942, and considered by him as synonymous with S. pseudoregularis, Tokioka, 1942 and 1952), S. sp. Tokioka 1954, and S. pseudoregularis Oye 1918. The differences and similarities are readily compared in tabular form. Inspection of Table 2 clearly illustrates that S. popovicii differs significantly enough from the other species to justify its designation as a new species. The name, Sagitta popovicii , is given in honor of Dr. Zacarias Popovici. REFERENCES Tokioka, T. 1942. Systematic studies of the plankton organisms occurring in Iwayama Bay, Palao, III. Chaetognaths from the Bay and adjacent waters. Palao Trop. Biol. Sta. 2(3): 527-548. 1952. Chaetognaths of the Indo-Pacific. Annot. Zool. Jap. 25(1-2): 307-316. Fig. 3. Sagitta popovicii n. sp. A, Entire animal, ventral view; B, eyes, showing pigment distribution; C, anterior teeth; D, posterior teeth; E-F, stages of development of seminal vesicle. TABLE 110 PACIFIC SCIENCE, Vol. XV, January 1961 H3 ’E, g 6 .-a G ^ XT 3 Oh S ° o 3 -g ’ 5 ) & 3 -y o -o ^ ol q j w) -G • -rt C C G O « It g» ^ a co § 8 2 u § G 'Eh .2 > .9 ^ G LJ rd 2 c 1 1 § oi » O G V, a si 8 bO A! *o — 1 G - 5 6 8 cS 5 § s 3 *" £ 3 - >3 2 ^ <3 > ID 3 13 a .2 § 6 O qj U U3 O CD > X °3 3 u ^ w O G G 8 c* -8 O G 8 X X .8 W £ o . -o g (N 5 G M -Q 0 3 .„ a w >* 1 2 CD G 1 > 2 o JD co o 2 a 'G 8 ^ CD . s jy ^ G r* 5 G o (?s bO rt 3 3b c 3 3 q> >, «s - G 4h qj O X qj O ^ -s 8 G > jj 8 G G ^ qj 8 O O .-, Oh bJO «_ G U X! . -2 G O G H qlJ3qC bJTG G S o> qj h-1 «« M £V Pd o Z. s nd qj «-4 bo qj c G 'S G qj a x o « qj -A 8 .2 8 v> 0 -G a q> nd ^ § > o E? M 1 S G ^ O ^ u • - w G •- 8 ~o ^ 'a — i 1 *§ r- > g-a o ,G O qj — < G qj U ^ 2 "6 II ^ a | 3 a ^ qj 8 — — ' Jd O <3 «'SU 8 G M o qj O > _g a 8 c s g s to - Ji n3 ^3 G § o 8 O u, ^ s •£ G qj ur> "d ^ os "d 8 § CO >* I_r 8 1 < 2 g 'G 8 a qj 3 co . W 3 8 ^ a ^ o . 'I I J a | Tb 3 G G qj ci C u G 3 G » o qj M >, ’gl S a -8 8 6 » o •la ^ •- qj G cr . a 6 o a _r cn .9 •- qj bO ^ S r V Q & . H_. — H 2 g g >, 8 S 6 2 S 3 CO > Oh qj O bO G . — i i-i O -d 'C G G l-l CS qj "d G LT ^ *8 C 3 £ O ^ G ■G •- w S nd > ■g > 8 ^ s l un nd bo -s G 8 Gd G x bO G qj G CO CG TABLE 2 Chaetognaths from Peru — SuND 111 -d la * a tj o -g 3 C/D u 5 6 1 6 d m 3 u Z2 -g 2 «3 6 c cS AS a c o .2 3 "3b O c3 u *3 ^ G ti d i > bo __r G <3J C3 X ^ U CO o3 -d ‘i 03 O 3^ o ... g a 0 3 d bQ bO a -a 2 &J. £ C T3 • G £ • 2 § o jG aJ bX) S5 ^ o I i a u ftT3 co d G > O O G 6 .3 d G d .. -d o ^ — r ^ ~3 6 >> w •a g il d 5 !> a co _Q O g a O O H "be G G s* o3 bQ bo ”3 "3 Wl M G G d d > > 3 S O _ M _ G •2 | 6 d a " d bQ ^ G AS S -a 3 a « ^ R2 03 o U, ■ti c j -d o d -d G a c G O $ a, '2 aj C o s 2 -g c3 o m_ JQ ° _ I 3 . bJD 1 •P G i d -d y o g psj 2 f o s c ■M — 1 u ■ «-* Uh a d o3 3 bJO Wl G a .2 03 a QJ CO o d d d G o3 c X G a . or s a g cr G O d * U W jQ ^ bC O G ^ -2 G 3 d3 -d 173 G u -ri 2 G "3 g ' d > ^ n3 aj G< 3 6 S 8 ex.% w Tb d c (SS .2 60 "So — . aj c3 PQ 2 • G — ' G « a g a <-p «-< G O O •d « .2 G X aj -Q G a (U 3 ^ CO . G bfi w l § 3 g si M3 g”S - bC-M U .c ° - ^2 J3 ' . s_ bo^: c 2 OJ c3 co — i d *5 •d '$■ 6 4-4 0 1 6 or 6 03 ^ a . o s ^■S m b0 a c O . a d ^ o ° ib — , G b0 d g^g > ^ -d -*-i i_ d ° .° ^ O 8 2 d ^ c*2 £ < m=i ^ a .& § 0J G O - .2 'Z S S « i a 3 C/D a o3 G JS ^2 lb '§ s I « .2 "3 'd 2 G m— c d O > d >: 3 co a d co jg a u S 5 o 6 »5 M-< O G 1-4 X J* Si ‘C cn d ^ G O a C4P 8 d d bo X d a > 2 *d bJO ^ G X d 6 2 So 2 d ^ A New Opisthobranch Mollusc from Hawaii Alison Kay 1 The opisthobranch genus Arthressa was pro- posed by Evans (1950) on the basis of a revi- sion of the genus Volvatella Pease I860. Pease’s genus was known only from four species, three described by Pease (1860, 1868), based on sin- gle specimens from the Pacific, and a fourth described by G. and H. Nevill (1869) from Ceylon. Evans (1950) pointed out several dis- crepancies in Pease’s species descriptions, and, while retaining V. fragilis Pease, the type species, he proposed the genus Arthressa to include A. cincta (G. and H. Nevill) from Ceylon and A. elioti Evans which was newly described from Zanzibar. Four specimens of a species congeneric but distinct from A. cincta and A. elioti have been collected in the Hawaiian Islands, and are here described as a new species. GENUS Arthressa Evans 1950 SHELL: Thin, calcareous, covered by a thicker periostracum extending beyond the calcareous margin. Body whorl ovate, broadest in the mid- dle, and contracting posteriorly to a spout. An- terior aperture as broad as shell in its anterior half. Right lip overlapping reflected left lip be- hind aperture; left lip depressed behind margin and passing at the junction of right and left lips under recurved left margin into a deep umbilical cleft; margin continuous with the spout. On left side of spout columella of a sunken spire of approximately three turns visible through per- iostracum. ANIMAL: Tentaculate area of head rimless, frontal, and oriented ventrally. Dorsal tentacles short and smooth. Lateroventral tentacles deeply grooved laterally, slightly bilobed at tip. Eyes immediately posterior to grooves. Anal opening on dorsal surface of visceral mass beneath man- 1 Department of General Science, University of Ha- waii. Contribution No. 144 from the Hawaii Marine Laboratory. Manuscript received July 29, 1959- tie. Foot small and triangular with blunt cor- ners. TYPE SPECIES: A . cincta (G. and H. Nevill). Ceylon. Arthressa evansi, new species Fig. 1 HOLOTYPE: Bishop Museum, no. 8901. 11 mm. in length; 6 mm. in breadth; collected in a tidepool, Diamond Head Beach Park, Oahu, Hawaii, November, 1956. Paratype: Bishop Mu- seum, no. 8902. 11 mm. in length; 5 mm. in breadth; collected with holotype. Length of anterior aperture more than half that of shell. Body whorl rounded, not flattened on the left, and with the spiral curve of the shell continued into it. Length of spout less than half shell breadth. Opening of spout tri- angular with a ventral slit continuous with aperture. Mantle edge smooth in spout. Animal orange, freckled with a darker shade, and with a band of freckling crossing body whorl. Foot similar in coloring to that of body but freckling of a lighter shade. Foot bluntly rounded, not ex- tending anterior to head, and with a groove separating anterior two-thirds from posterior third. Animals extruded a viscid white sub- stance from both posterior spout and anterior aperture. Specimens collected vary from 8-11 mm. in length and from 4-6 mm. in breadth. The species has been named for the late J. T. Evans. The specimens were collected in a tidepool on the reef flat of Diamond Head Beach Park, Oahu. They appeared to be associated with the algae Padina and Gracilaria. The mode of pro- gression on a hard substrate is such that the anterior end of the foot extends forward and attaches to the substrate while the posterior por- tions move up. The animals readily suspend themselves upside down on the surface of the water, progressing as waves of muscular con- traction pass across the foot. 112 Arthressa evansi , n. sp. — K ay 113 FIG. 1. Arthressa evansi. Drawn from the living animal. A. Dorsal view; B, ventral view. DISCUSSION Of the three species now included in Ar- thressa, A. evansi appears to be intermediate in characters between A. cincta and A. elioti. A. ■evansi resembles A. cincta in coloring and in the smooth mantle edge; it resembles A. elioti in the length of the anterior aperture and in the rounded body whorl. Although Evans (1950) suggested that Vol - vatella fragilis Pease should be retained as it was sufficiently described, no collections of this species have been recorded in Hawaii since the original description. Pease (I860) merely de- scribed the type specimen as from the "Sand- wich Islands,” and the animal was illustrated as being white. Evans (1950) distinguished Vol- vatella from Arthressa on the basis of Peases (I860) description of the anal opening into the posterior spout in Volvatella; the anus opens on the dorsal surface of the visceral mass in Arthressa . If specimens agreeing with Pease’s (I860) description of Volvatella, including a posterior anus, are collected in the future, Evans’ distinction between the genera will merit rec- ognition; however, it is possible that Pease’s description was misleading, and the pore which he described as the anus may have been the pore of the gland which secretes the viscid fluid which these animals emit. If this is the case, Evans could have gone further and entirely suppressed the genus Volvatella of Pease. Evans (1950) described Arthressa as a sac- coglossan genus, and it should therefore be re- moved from the position assigned Volvatella in the Akeratidae of Thiele (1931). Arthressa is of interest in the scheme of opisthobranch relationships in that it is a suctorial form feed- ing on various algae. The occurrence of a shell and a generalized mantle cavity and alimentary canal suggest that Arthressa holds a position in the Saccoglossa analogous to that of Actaeon (Fretter and Graham, 1954), which also pos- sesses many prosobranchiate characters. REFERENCES Evans, T. J. 1950. A review of Pease’s genus Volvatella, together with a preliminary re- port on a new saccoglossan genus. Proc. Malac. Soc. London 28: 102-106. Fretter, V., and A. Graham. 1954. Observa- tions on the opisthobranch mollusc Actaeon. Journ. Mar. Biol. Assoc. U. K. 33(3): 565- 585. Nevill, G., and H. Nevill. 1869. On some new marine Gastropoda from the Southern Province of Ceylon. Journ. Asiatic Soc. Ben- gal 38, pt.2: 65-69. Pease, W. H. 1860. Descriptions of new species of Mollusca from the Sandwich Islands. Zool. Soc. London Proc. 28: 18-37. 1868. Descriptions of marine Gastrop- oda inhabiting Polynesia. Am. Journ. Conch. 4: 71-80. Thiele, J. 1931. Handbuch der systematischen W eichtierkunde. Verlag von Gustav Fisher, Jena. Bd. I, 37 6 pp., 470 figs. The Aleyrodidae (Hemiptera-Homoptera) of New Caledonia L. J. Dumbleton 1 A previous paper ( Proc. R. Ent. Soc. Lond. (B) 25, pts. 7-8, pp. 129-141, 1956) included descriptions of one species of Dialeurodicus Ckll. and six species of Orchamoplatus Russell. This paper completes the study of a collection of Aleyrodidae from New Caledonia. The material was collected largely by myself during three years’ residence in New Caledonia, and as col- lection was concentrated specifically on this fam- ily it is believed that the fauna described here is representative and probably includes a ma- jority of the species. The following table compares the size and endemicity of the New Caledonian Aleyrodid fauna with those of adjoining areas, in so far as these are known. The fauna of the New Guinea-Solomons- New Hebrides chain, from which direction an ities between the faunas of New Caledonia and New Zealand. The faunas of New Zealand and the South Pacific are smaller, but while the majority of the New Zealand species are un- doubtedly endemic the South Pacific species are predominantly introduced forms occurring on introduced economic crop plants or weeds or on native food plants whose movements about the area have been considerable. The majority of the species from these four areas belong to the subfamily Aleyrodinae. The subfamily Udamoselinae is poorly represented, there being one species of Aleurodicus Douglas in Australia and another in Fiji, one species of Dialeurodicus in New Caledonia, and one species of the endemic genus Synaleurodicus Solomon in Australia. The subfamily Uraleyrodinae is not represented. AREA NUMBER OF GENERA NUMBER OF SPECIES Endemic Nonendemic Total Endemic * Nonendemic Total New Caledonia 3 7 10 25 — 25 Australia 1 12 13 19 3 22 New Zealand...... 1 3 4 9 1 10 South Pacific — 7 7 5 4 9 * In each area the number of species presumed to be endemic includes two which are possibly nonendemic. important element of the New Caledonian fauna was probably derived and with which it could be expected to show strong affinities, is un- fortunately completely unknown. The Australian fauna has been little studied and is undoubtedly much larger than the above figures suggest. Those of New Zealand and the South Pacific are reasonably well known. The New Caledonian fauna has affinities with the Australian fauna (for example, in the presence of endemic species of Alem acanthus, Bemisia . Dialeurodes, and Tetraleurodes) , but there is a relatively stronger element of endemic genera in New Caledonia. There are no obvious affin- 1 Entomology Division, D. S. I. R., Christchurch, New Zealand. Manuscript received June 1, 1959- KEY TO THE GENERA OF THE ALEYRODIDAE OF NEW CALEDONIA ( based on pupal cases ) 1. Thoracic tracheal pores, clefts, and combs absent 2 Thoracic tracheal pores, or clefts, or combs present 6 2. Large elongate brown species, vasiform orifice large, operculum short transverse; lingula long, more than half exposed, occupying whole orifice Dialeurodicus Cockerell Not as above 3 3. Submarginal line distinct, complete T etraleurodes Q. & B. Not as above 4 114 Aleyrodidae of New Caledonia— Dumbleton 115 4. Vasiform orifice triangular; operculum oc- cupies only half orifice; lingula long, pointed, exposed; distinct caudal furrow Bemisia Q. & B. Not as above 5 5. Marginal teeth well developed, submar- ginal series of papilla-like pores absent, but simple pale small spots or pores may be present Aleurocanthus Q. & B. Marginal teeth absent or ill-defined, sub- marginal linear series of papilla-like pores present, broken by 9 processes on each side Leucopogonella n.g. 6. Abdomen with only 7 visible segments, top-shaped pit-making species Gomenella n.g. Not as above ™ 7 7. Vasiform orifice long, triangular; oper- culum occupies half orifice; lingula long, pointed, exposed Parabemisia Takahashi Not as above 8 8. Thoracic tracheal comb absent Dialeurodes Cockerell Thoracic tracheal comb present 9 9. Thoracic tracheal comb with 2 teeth, ab- dominal tracheal comb absent, margin sinuate; submargin with 10 prominent setae on each side; without submarginal papillae Dothioia n.g. Thoracic tracheal comb with 2 -many teeth, abdominal comb present, margin not sinuate, submargin without prominent setae, specialized submarginal papillae present Orchamoplatus Russell subfamily UDAMOSELINAE Enderlein GENUS Dialeurodicus Cockerell Dialeurodicus elongatus Dumbleton. Proc. R. Ent. Soc. Lond. (B), pts. 7-8, pp. 129— 131, 1956. SUBFAMILY ALEYRODINAE Enderlein GENUS Aleurocanthus Quaintance & Baker KEY TO NEW CALEDONIAN SPECIES (based on pupal cases) 1. Long spines absent 2 Long spines present 3 2. Shape regular ovate or elliptical, without short spinous processes; lateral abdom- inal rays present nudus sp. n. Shape somewhat pointed anteriorly, shal- lowly emarginate posteriorly, short spi- nous processes present, lateral abdominal rays absent brevispinosus sp. n. 3. Long spines on thorax only, shape some- what pointed anteriorly, marginal teeth rounded, lateral abdominal rays present spinitkorax sp. n. Many long spines on both thorax and abdomen, shape broadly rounded ante- riorly, bluntly pointed posteriorly, mar- ginal teeth pointed, lateral abdominal rays absent multispinosus sp. n. Aleurocanthus brevispinosus sp. n. Fig. 1 a-d LARVA: Second instar (Fig. la ) : Length 0.39 mm., width 0.24 mm. Colour pale. One pair spines anterior to and 2 pairs laterad of cepha- lic hairlike setae. One pair short and 1 pair long spines posteriorly on thorax. One pair hair- like setae on 1st abdominal segment, those on 8th minute, caudal setae longer, 6 pairs of spines on abdomen. Margin obscurely toothed. Third instar (Fig. lb ) : Length 0.60 mm., width 0.41 mm. Sometimes pigmented brown medially. Two pairs of spines anterior to ce- phalic setae and 2 pairs caudad. One pair short spines on thorax. First abdominal segment with 1 pair paramedian processes, 8th abdominal and caudal setae hairlike, 6 pairs spines on abdomen. Marginal teeth similar to those of pupal case. PUPAL CASE (Figs. Ic, d) : Length 0.97 mm., width 0.74 mm. Colour black, eye spots paler. Slightly pointed anteriorly, sides a little concave from one-quarter to one-half length, broadest at mid-length, caudal margin more broadly rounded than anterior but with slight concavity. Not con- stricted across thoracic tracheal folds. A white wax fringe of columnar elements as wide as teeth and up to half as long as body width. Margin prominently toothed; teeth longer than wide, rounded apically. Submargin not defined by submarginal line, an inner row of teeth pres- ent. A continuous row of minute pale spots or 116 PACIFIC SCIENCE, VoL XV, January 1961 pores, 1 mesad of each outer tooth. A row of minute pores in the position of the submarginal line. Thoracic and abdominal tracheal folds, pores, clefts, and combs absent. Cephalic area posteriorly with 2 paramedian lobes mesad of the eye spots, each bearing 4 short spines. Pores and spines on thorax as in figure. Abdomen, first segment with paramedian processes, a pair of nearly contiguous pores on second segment, a median abdominal ridge laterad of which are 5 pairs of short blunt spinous processes, 1 pair more acute laterad of orifice. Eighth abdominal seta bases laterad of orifice about mid-length. Caudal seta bases slightly closer together than width of orifice and in same line as submarginal line of pores. Vasiform orifice (Fig. Id) sub- circular, slightly transverse. Operculum sub- cordate, filling orifice. Lingula not visible. holotype: Pupal case on slide mount. De- posited in Institut Frangais d’Oceanie, Noumea, New Caledonia. PARATYPES: In author’s collection. TYPE LOCALITY: New Caledonia, PDothio, coll. F. Cohic. FOOD PLANT: Undetermined. OTHER MATERIAL: From undetermined hosts, Carenage 17/12/54 and Montagne des Sources 8/4/55, coll. L.J.D., also in association with A. spinithorax at Dothio. This species shows some affinities with Aleu- rotrachelus because the outer boundary of the clear pores mesad of the marginal teeth is raised and simulates an inner row of teeth. In this the species resembles A. dissimilis Q. & B. and A. serratus Q. & B. Aleurocanthus multispinosus sp. n. Fig. 2 a-e LARVA. Second instar (Fig. 2a ) : Length 0.34 mm., width 0.2 mm. Colour pale. Cephalic re- gion with 1 pair of paramedian hairlike setae and anterior to these 2 pairs of longer stout spines. Posteriorly on the thorax is 1 pair of short conical spines and a pair of very long FlG. 1. Aleurocanthus brevispinosus sp. n. a, Sec- ond instar larva; b, third instar larva; c, pupal case; d, pupal case, vasiform orifice and caudal margin. (Scale in mm.) 0.3 Aleyrodidae of New Caledonia— Dumbleton 117 Fig. 2. Aleurocanthus multispinosus sp. n. a, Sec- ond instar larva; b, third instar larva; c, pupal case; d, pupal case, vasiform orifice and caudal margin; e, I pupal case, margin. (Scale in mm.) spines. Abdomen with 3 pairs of short conical spines and 2 pairs of long spines. Eighth ab- dominal setae and caudal setae long, hairlike. Third instar (Fig. 2b): Length 0.54 mm., width 0.31 mm. Colour pale. Cephalic area with 2 pairs of long spines anterior to cephalic setae, lateral pair shorter. Thorax with 2 pairs long spines anteriorly, lateral pair shorter; 2 pairs of long spines posteriorly. Abdomen: 1 pair of paramedian processes on 1st segment, followed by 1 pair of long spines, 3 pairs of pores, 2 pairs of long spines. Eighth abdominal and caudal setae hairlike. PUPAL CASE (Figs. 2 c-e): Length 1.0 mm., width 0.71 mm. Colour black. Shape elliptical, wider behind mid-length, posterolateral mar- gins slightly concave and caudal region bluntly conical. No constriction across thoracic tracheal folds and no caudal invagination. Margin (Fig. 2d) regularly toothed, teeth not broad and rounded but acute and pointed with the slope of the sides concave. Submargin not defined by submarginal line, finely ridged near margin. Thoracic and abdominal tracheal folds, clefts, pores, and combs absent. Cephalic area with 5 pairs of long, black, somewhat frayed spines anterior to 1 pair of short finer paramedian cephalic setae. Eye spots not evident. Thoracic area with 5 pairs of long spines and 1 short thornlike spine midway between median line and margin metathorax. Abdomen with ill-de- fined median ridge, segmentation ill-defined, rhachis absent. Segment 1 with a pair of para- median processes. Segments 2-6 each with 1 pair of short thornlike paramedian spines. Seg- ment 7 with long spines. Eighth segment with 1 pair of long stout spines, setae absent. Caudal setae short, hairlike, with bases about same width as vasiform orifice but nearer the margin than to the orifice. Five pairs of long lateral spines in same line as 8th abdominal spines and 3 pairs of long discal setae midway between paramedian and lateral spines. Vasiform orifice (Fig. 2e) raised, not clearly discernible, hem- ispherical. Operculum filling orifice. Lingula not discernible. HOLOTYPE: Pupal case on slide mount. De- posited in Institut Francais d’Oceanie, Noumea, New Caledonia. 118 Fig. 3- Aleurocanthus nudus sp. n. a, Pupal case; b, pupal case, margin; c, pupal case, vasiform orifice. (Scale in mm.) PARATYPES: In author’s collection. TYPE LOCALITY; New Caledonia, Montagne des Sources, 3,000 ft., coll. L.J.D. FOOD PLANT: Hibbertia sp. Similar to A. calophylli Kotinsky from Fiji but differing in the form of the marginal teeth. Aleurocanthus nudus sp. n. Fig. 3 a-c pupal CASE (Fig. 3 a-c): Length 1.0 mm., width 0.72 mm. (holotype) but reaching 1.4 mm. long. Colour black. Shape broadly elliptical, rounded anteriorly and posteriorly. Rather flat PACIFIC SCIENCE, Vol. XV, January 1961 but with median ridge and 2 thoracic lateral ridges. Wax fringe white about one-third body width. There is a submarginal line of wax, an inverted U-shaped wax line more or less con- touring the adult head and thorax, and 3 or 6 transverse lines on the abdomen. Inside the mar- ginal line of wax there are narrow radial striae of wax and the disc is powdered with wax. Mar- gin (Fig. 3 b) strongly toothed, teeth rounded, 7 teeth in 0.1 mm. Submargin not defined from dorsal disc, radially striate, striae same width as teeth; with a number of minute simple pores, 3 cephalic, 2 thoracic, 2 mid-length on abdomen. Thoracic and abdominal tracheal folds, pores, combs, and clefts absent. Eye spots present, sub- triangular. Cephalic and thoracic areas without setae, pores as in figure. Abdomen, segment 1 with 1 pair paramedian pores, segments 2-6 limited laterally and with paramedian pores. Be- tween segments 3 and 6, 4 lateral rays on each side. Seta bases on 8th abdominal segment on orifice rim about mid-length of orifice. Caudal seta bases close together, nearer margin than orifice. Vasiform orifice (Fig. 3 c) subcordate, 0.06 mm. wide, 0.065 mm. long, internally toothed. Operculum subcordate, filling orifice 0.05 mm. long. Lingula parallel-sided, apex rounded. HOLOTYPE: Pupal case on slide mount. De- posited in Institut Francis d’Oteanie, Noumea, New Caledonia. PARATYPES: In authors collection. TYPE LOCALITY: New Caledonia, Tinip, coll. L.J.D., 4/11/54. FOOD PLANT: Leucopogon sp. On upper sur- j face of leaf. This species comes nearest the genus Aleuro- canthus and is similar to dissimilis Q. & B. in the absence of long spines and to T - signatus Mask., serratus Q. & B., and spinosus Kuwana in the presence of lateral rays on the abdomen. Aleurocanthus spinithorax sp. n. Fig. Aa-d LARVA. Second instar (Fig. 4a): Length 0.4 mm., width 0.27 mm. Colour pale laterally, brown medially. Margin with short broad trun- cate teeth. Cephalic area with 2 pairs of short conical spines anterior to cephalic hairlike setae Aleyrodidae of New Caledonia-— Dumbleton 119 cl Fig. 4. Aleurocanthus spinithorax sp. n. a, Second instar larva; b, third instar lava; c, pupal case; d, pupal case, vasiform orifice and caudal margin. (Scale in mm.) and 1 pair posterior. Thoracic area with 3 pairs of short conical spines. Abdomen: First seg- ment with 1 pair hairlike setae, following seg- ments with 4 pairs of short conical spines, 8th abdominal setae small or absent, 1 pair of short conical setae laterad of orifice, caudal setae hair- like. Third instar (Fig. Ab ) : Length 0.68 mm., width 0.52 mm. Colour brown. Marginal teeth as in pupal case with pore mesad of each. Cephalic area with 2 pairs of short conical spines anterior to and 2 pairs posterior to cephalic hairlike setae. Thoracic area with 3 pairs of short conical setae. Abdomen: First segment with 1 pair of paramedian processes followed by 6 pairs of short conical spines. Eighth ab- dominal setae not discernible, caudal setae hair- like. pupal case (Fig. Ac, d ) : Length 1.1 mm., width 0.82 mm. Colour dark brown to black. Shape elliptical but slightly pointed anteriorly, angulated opposite end of transverse moulting suture and again at 2 points midway to anterior median line. Very slightly emarginate caudally. Not constricted across thoracic folds. Margin toothed, 9 teeth in 0.1 mm., teeth longer than wide, apex rounded. Thoracic tracheal folds in- dicated; pores, clefts, and combs absent. Ab- dominal tracheal folds, pores, clefts, and combs absent. Submargin with a continuous row of minute simple circular pores one behind each marginal tooth, not defined mesally by sub- marginal line, without setae but with 4 minute pores on each side of thorax and 2 on each side of abdomen mesad of the line of pores behind the teeth. Cephalic area delimited by line sub- parallel to margin commencing anterior to thor- acic fold and meeting at median suture. On each side a group of 3 long spines and 2 short spines. Thoracic area with the median part limited by lateral lines midway between median line and margin. Near the anterior end of these lines, on each side, a group of 2 long and 2 short spines. In the posterior lateral angle of the meta- thorax, and nearer the margin than the median line, is a single long spine. Abdomen broadly ridged in the median area with the segments limited laterally and with 3 rays extending half- way to margin from intersegmental sutures 4-5, 120 PACIFIC SCIENCE, Vol. XV, January 1961 5-6, and 6-7. Segment 1 subrectangular with 2 paramedian processes close together. Segments 2- 5 each with 2 paramedian nearly contiguous pores. Laterad of the boundary of the median ridge are 4 short spines opposite sutures 2-3, 3- 4, 4-5, and 5-6, a similar one at mid-length on 7, and 1 laterad of orifice on 8. The 8th ab- dominal seta is not discernible. Caudal seta bases as wide apart as spines on 8th segment. Vasiform orifice subcordate (Fig. Ad) . Operculum fills orifice. Lingula not discernible. HOLOTYPE: Pupal case on slide mount. De- posited in Institut Frangais d’Oceanie, Noumea, New Caledonia. PARATYPES: In author’s collection. TYPE locality: New Caledonia, Dothio, coll. F. Cohic. FOOD PLANT: An undetermined climbing plant. This species is very similar in facies to A. brevispinosus and shows affinities with Aleuro- trachelus Q. & B. The moulted skins of 2 larval instars are usually adhering to the dorsum of the pupal case. GENUS Bemisia Quaintance & Baker Bemisia cordylinidis sp. n. Fig. 5 ctr-c LARVA. Third instar (Fig. 5 a): Length 0.7 mm., width 0.42 mm. Resembles pupal case but anal furrow less developed and caudal setae long and stout. PUPAL CASE (Fig. 5 b, c): Length 1.15 mm., width 0.80 mm. Colour transparent or white. Shape elliptical, flat. No constriction across thoracic pores, little or no caudal invagination. Margin finely crenulate. Submargin not sepa- rated from dorsal disc. Anterior and posterior marginal setae present. Area between submargin and tubercles with dense hemispherical pustules in radial arrangement, each row tending to con- sist of tubercules 2 deep. Cephalic area with 2 pairs of minute paramedian setae. Thoracic area with tracheal folds, pores, combs, and clefts ab- sent; 3 pairs of tuberculate warts nearly midway between median line and margin. Abdomen, first segment with 1 pair of minute, widely spaced setae. Segments 1-5 with a median tuberculate wart, less developed on segments 1, 4, and 5. A Fig. 5. Bemisia cordylinidis sp. n. a, Third instar larva, vasiform orifice and caudal margin; b , pupal case; c, pupal case, vasiform orifice and caudal margin. (Scale in mm.) Aleyrodidae of New Caledonia — Dumbleton 121 row of 7 or 8 tubercles on each side, the pos- terior members tending to become continuous with the ridges bounding the anal furrow. These abdominal tubercles and those on the thorax outline the adult body. Seventh segment nar- rower than 6 and 8. Eighth segment with a pair of minute setae level with anterior margin of vasiform orifice, bases as wide apart as orifice width. Vasiform orifice (Fig. 5 c) subtriangular, long, acutely pointed, with 2 or 3 oblique ridges on each side of apical half. Operculum semi- circular, not more than half the length of orifice. Lingula long, exposed, spatulate, pointed. Ridges of anal furrow contouring the orifice where the tubercles are somewhat pectinated, smooth tubercles between orifice and cauda. Caudal setae minute, on ends of furrow ridges near margin. HOLOTYPE: Pupal case on slide mount. De- posited in Institut Frangais d’Oceanie, Noumea, New Caledonia. PARATYPES: In author’s collection. TYPE LOCALITY: New Caledonia, Montagne des Sources, coll. L.J.D. FOOD PLANT: Cordyline sp. The tuberculation suggests Aster obemisia Tre- han but the thoracic and abdominal tracheal folds and combs are absent. The species re- sembles Bemisia shinanoensis Kuwana but lacks setae on the abdomen. GENUS D idem odes Cockerell KEY TO NEW CALEDONIAN SPECIES (based on pupal cases) 1. With 5 large simple pores on each side dothioensis sp. n. No large pores 2 2. Pale species, adult body outlined by tuber- culate line; transverse moulting suture not continued anteriorly; anterior and posterior marginal, 8th abdominal and caudal setae present psychotrim sp. n. 3. Black species; transverse suture continued anteriorly and meeting in median line and enclosing eye spots, lateral abdominal line passing from transverse suture to level with orifice; without setae ...dumbeaensis sp. n. Dialeurodes dothioensis sp. n. Fig. 6a, b PUPAL CASE (Fig. 6a, b): Length TO mm., width 0.73 mm. Colour transparent or white. Shape elliptical, widest at mid-length, with slight concavities in the thoracic and abdominal tra- cheal pore areas. Wax secretion in the form of pale blue filaments. Margin somewhat irregularly toothed, anterior and posterior marginal setae present. Submargin not defined from dorsal disc, with 12 short spearhead-like marginal setae on each side, 5 anterior to thoracic fold and 7 pos- terior to it. Cephalic area with 1 pair of para- median cephalic setae and anterior to these and nearer to the margin 1 pair of very large simple Fig. 6. Dialeurodes dothioensis sp. n. a, Pupal case; b. pupal case, vasiform orifice and caudal margin. (Scale in mm.) 122 Fig. 7. Dialeurodes dumbeaensis sp. n. a, Pupal case; b, pupal case, thoracic tracheal pore; c, pupal case, vasiform orifice and caudal margin. (Scale in mm.) PACIFIC SCIENCE, Vol. XV, January 1961 pores with thick walls of external diameter 0.045 mm. Thoracic area with 1 pair of similar pores immediately anterior to thoracic folds which are not sculptured and end in a comb of about 5 minute teeth. Abdomen with three pairs of large simple pores near margin, one opposite the transverse moulting suture, one slightly anterior to the orifice, and one caudad of the orifice. Ab- dominal tracheal fold and comb not evident. First abdominal segment with one pair of short setae. Eighth abdominal setae present longer than orifice, bases level with anterior margin of orifice. One short seta cephalad of each large caudal pore. Caudal setae caudad of these pores. Vasiform orifice (Fig. 6b) subcordate, 0.06 mm. wide, 0.045 mm. long, no teeth on margin. Operculum nearly filling orifice, 0.04 mm. wide, 0.035 mm. long. Lingula included, constricted near base, 0.02 mm. long, 0.015 mm. wide, sub- truncate apically. HOLOTYPE: Pupal case on slide mount. De- posited in Institut Frangais d’Oceanie, Noumea, New Caledonia. PARATYPES: In author’s collection. TYPE LOCALITY: New Caledonia, Dothio River Bridge, coll. F. Cohic, 17/6/53. FOOD PLANT: Undetermined. This species is closely related to D. decem- punctata Q. & B. but differs in that the thoracic fold is not between the first and second pores, in the spearhead shape of the marginal setae and in having only one seta anterior to each caudal pore. It differs from D. viburni Takahashi in having only one seta anterior to each caudal pore and in the form of the marginal setae. It belongs to the subgenus Dialeuropora Q. & B. Dialeurodes dumbeaensis sp. n. Fig. la-c PUPAL CASE (Fig. 7 a-c): Length 1.25 mm., width 1.2 mm. Colour black. Shape subcircular, faintly constricted across thoracic folds but in- vaginated caudally at abdominal tracheal pore. The marginal fringe, of white wax rods half as wide as body, may split in stellate fashion into 10 rays. Margin with rounded teeth, 10 teeth in 0.1 mm., anterior and posterior marginal setae present. The submarginal line is higher in elevation than the margin and the dorsal Aleyrodidae of New Caledonia — Dumbleton Fig. 8. Dialeurodes psych otriae sp. n. a, Third in- star larva; b, pupal case; c, pupal case, thoracic tra- cheal pore; d, pupal case, vasiform orifice and caudal margin. (Scale in mm.) 123 disc and the transverse and median moulting sutures are also elevated. Submargin radially striate 0.25 mm. wide, on outer half with radial rows of 4—5 minute circular poriferous papillae mesad of alternate teeth, intervening areas darker, inner half with radial rows of less well-defined subcircular nonporiferous papillae. Submarginal line on cephalothorax continuous with transverse moulting suture. Cephalic area with eye spots oval, long axis 0.03 mm. long, oblique, enclosed by submarginal line, without setae or evident pores. Thoracic area, without setae or evident pores. Tracheal folds not vis- ible dorsally, cleft present but closed, pore (Fig. Ih) minute, subcircular, minutely toothed, inset from margin. Cephalothorax with scattered irregular minute pores. Abdomen. Submarginal line of abdomen ending level with orifice, the two halves joined by a semicircular raised line anterior to orifice, from the junction on each side a short raised line runs mesocaudally so that the orifice is enclosed except caudally. Ab- domen, segments 1-6 subequal in length, with- out setae but with transverse row of minute pores, seventh segment half as long. Vasiform orifice (Fig. 7c) semicircular, 0.03 mm. long, 0.035 mm. wide. Operculum semicircular, 0.02 mm. long, 0.03 mm. wide, filling orifice. Lingula not visible. Eighth abdominal setae represented by pores, level with anterior margin of orifice and distant from it. HOLOTYPE: Pupal case on slide mount. De- posited in Institut Frangais d’Oceanie, Noumea, New Caledonia. PARATYPES; In author’s collection. TYPE LOCALITY; New Caledonia, Durnbea, coll. L.J.D., 9/6/52. FOOD PLANT: Homalium sp.? This species would run to Dialeurodes sub- genus Gigaleyrodes Q. & B. in Quaintance & Baker’s key, and is reminiscent of D.( G.) cerifera Q. & B. Dialeurodes psychotriae sp. n. Fig. 8 a-d LARVA. Second instar: Length 0.30 mm., width 0.28 mm. Elliptical. Colourless. Similar to 3rd instar but tubercles absent and marginal teeth more distinct. 124 PACIFIC SCIENCE, Vol. XV, January 1961 Third instar (Fig. 8 a): Length 0.75 mm., width 0.56 mm. Elliptical. Colourless. Without tuberculate pattern but with scattered irregular small tubercles. Caudal setae situated on margin, strong. Tracheal pores with few teeth on margin, no clefts. Tracheal folds present, dotted. Mar- gin as in pupal case. PUPAL CASE (Fig. 8 b-d): Length 1.44 mm., width 1.15 mm. Colourless except for a circular pigmented middorsal area on meso- and meta- thorax and anterior half of first abdominal seg- ment, and another more elongate area on 7 th and 8th abdominal segments including the ori- fice and extending caudad of it. Shape elliptical, flat, widest at mid-length. Margin slightly cren- ulated nearly entire with numerous marginal ridges and grooves extending a short distance mesally, anterior and posterior marginal setae present. Submargin not defined from dorsum. Body of adult outlined by a row of tubercles, single on cephalothorax but tending to be double on abdomen and ending caudad of orifice. Be- tween this and the margin on the abdomen is a second row of slightly smaller tubercles in single series. Thoracic tracheal folds present, marked by dots, extending from pore to line of tubercles where they are dilated. Pore (Fig. 8c) simple, without teeth, distant 0.05 mm. from margin, cleft present. Abdomen with scat- tered small circular pores especially between margin and outer row of tubercles. Setae not evident on segment 1. Posterior half of 6th abdominal segment between paramedian pores is finely spinulose. Eighth abdominal setae 0.03 mm. long level with anterior margin of orifice, distant from orifice. Caudal setae 0.01 mm. long, one on each side of tracheal pore at mid-length. Vasiform orifice (Fig. 8 d) subcordate, not toothed internally on lateral or posterior mar- gins, 0.04 mm. long, 0.04 mm. wide. Operculum filling orifice, 0.03 mm. long, 0.03 mm. wide. Lingula not exposed, parallel sided, apex rounded, as long as operculum. Abdominal tra- cheal fold present, marked by dots. Pore sim- ple, distant from margin by closed cleft. holotype: Pupal case on slide mount. De- posited in Institut Fran^ais d’Oceanie, Noumea, New Caledonia. PARATYPES: In author’s collection. TYPE LOCALITY: New Caledonia, Plum, coll. L.J.D. FOOD PLANT: Psychotria deplanchei. This species would run to the subgenus Gigaleyrodes in Quaintance & Baker’s key al- though the tracheal pore differs. A similar tuber- culation of the dorsal disc occurs in Dialeurodes ( Dialeuronomada ) dissimilis Q. & B. in which however the orifice is toothed. GENUS Dothioia nov. PUPAL CASE: Apparent margin not toothed, margin reflexed with small conical teeth. Tho- racic and abdominal tracheal folds not evident. Thoracic tracheal pore represented by two pro- jecting processes. Margin in the position of the abdominal tracheal pore with four or five ill- defined teeth. Apparent body margin sinuate, i.e., concave between the submarginal setae of which there are five on each side of the cephalothorax and four on the abdomen in addition to the caudal setae. TYPE SPECIES: D. hidentatus sp. n. Dothioia hidentatus sp. n. Fig. 9 a-h larva. Third instar (Figs. 9 a, h): Length 0.45 mm., width 0.31 mm. Shape elliptical. Colour pale. Margin toothed. Submargin with 10 fine short setae on each side. Cephalic area with one pair of fine setae and four pairs of spines. One pair of posterolateral thoracic spines. Abdomen, one pair of paramedian processes on first segment, seven pairs of spines laterad, eighth abdominal setae hairlike, level with ori- fice, caudal setae long. pupal CASE (Fig. 9c— f ) : Length 0.75 mm., width 0.56 mm. Colour black. Wax pattern of narrow linear raised elements; a median line along suture on anterior half of cephalothorax, a short transverse line on first abdominal seg- ment, and a longer oblique line in posterolateral angle of thorax. Apparent margin not toothed, Fig. 9- Dothioia hidentatus sp. n. a, Third instar larva; h, third instar larva, vasiform orifice and caudal margin; c, pupal case; d, pupal case, margin; e, pupal case, thoracic tracheal pore; f, pupal case, vasiform orifice and caudal margin; g, female, vasiform orifice and lingula; h, male, claspers and aedeagus. (Scale in mm.) Aleyrodidae of New Caledonia — Dumbleton 125 0.03 126 PACIFIC SCIENCE, Vol. XV, January 1961 sinuate, with 7 rounded prominences on each side. True margin may be visible between prom- inences near cauda and there show small conical teeth. Margin (Fig. 9 d) reflexed ventrally, marked by a line of small conical teeth mesad of which is a clear membranous line separating the venter which is densely covered by small circular pigmented tubercles. Apparent margin with 9 submarginal setae 0.05 mm. long, 4 anterior to thoracic pore, 1 on thorax caudad of pore, and 4 on the abdomen in addition to the caudal seta on each side. Submargin in bleached specimens with elongated pigmented rectangular thickenings internally. Cephalic area with a pair of cephalic setae and anterolaterad of these about 4 small tubercles in the position of the eye spots. Tho- racic tracheal folds not evident, pores (Fig. 9e) represented by 2 projecting rounded proc- esses 0.01 mm. long. Abdomen with median ab- dominal ridge. Segments clearly defined laterally by raised tuberculate line, the lateral margins of the seventh segment produced caudad. Eighth abdominal setae as long as orifice, distant from it, and level with its anterior margin. Vasiform orifice (Fig. 9/) subcircular, 0.04 mm. long, internally ribbed. Operculum filling orifice. Lin- gula not visible. Abdominal tracheal fold not evident, no pore or comb, caudal margin with 4 or 5 indistinct teeth between caudal setae. ADULT: (Fig. 9 g—h): Pale yellow. Wings immaculate. Forewings with Rg and Cu present. Hind wings with single vein. Antennae with third segment longer than 4-7 combined, 4 and 7 subequal and shorter than 5 and 6, which are subequal. About 18-19 setae in comb on hind tibia. Lingula (Fig. 9 g) with sides subparallel, apex conical and reduced in thickness. Male with claspers slender, pointed; aedeagus (Fig. 9 h) tapering slightly falcate apically. holotype: Pupal case on slide mount. De- posited in Institut Frangais d’Oceanie, Noumea, New Caledonia. paratypes: In author’s collection. TYPE LOCALITY: New Caledonia, Dothio River, coll. L.J.D., 9/5/53. FOOD PLANT: Undetermined. genus Gomenella nov. PUPAL CASE: Circular. Submargin defined by submarginal line. Submarginal setae present or absent. Dorsum densely covered with simple pores which may be large and contiguous. Thoracic tracheal pores flasklike, folds absent. Abdominal tracheal fold and pore present or absent. Anal furrow present. Vasiform orifice subcircular, operculum filling it. Lingula finger- like. Only 7 visible abdominal segments. Top- shaped, pit-making species. type SPECIES: G. multipora sp. n. One of the species described below resem- bles Aleurochiton Tullgren in the presence of dense dorsal pores but differs as above and also in the absence of a vertical wax palisade. The other species conforms to Dialeurodes (s.g. Gigaleyrodes ) except for the reflexed margins which suggest Tetralicia Harrison but it is otherwise unlike that genus. In assigning these 2 species to a new genus, stress is placed on the apparent presence of only 7 visible abdominal segments and the top-shaped form correlated with the pit-making habit in spite of the dif- ferences noted in the following key. KEY TO THE NEW CALEDONIAN SPECIES ( based on pupal cases ) Submargin reflexed, without setae; dorsum without dense large contiguous pores; ab- dominal tracheal pore present. . . .reflexa sp. n. Not as above multipora sp. n. Gomenella multipora sp. n. Fig. 10 a-f LARVA. First instar (Fig. 10*): Length 0.13 mm., width 0.10 mm. Submarginal setae as in other stages but with very long caudal setae functional legs and long antennae, without large pores on dorsum, vasiform orifice more elongate. Second instar (Fig. 10 b ) : Distinguished from other stages by sparse regularly arranged dorsal pores on thorax and a line of four subdorsal pores on each side of abdomen. Third instar (Fig. 10c): More numerous pores on the thorax, 3 rows of 3 pores on each side of abdomen not extending across median line as in pupal case. Two pores on each side of orifice. PUPAL CASE (Fig. 10 d-f): Length 0.5 mm., width 0.5 mm. Pupae yellow. Pupal case colour- Aleyrodidae of New Caledonia-— Dumbleton Fig. 10. Gomenella multipora sp. n. a, First instar larva; b, second instar larva; c, third instar larva; d, pupal case; e, pupal case, thoracic tracheal pore; f, pupal case, vasiform orifice and caudal margin. (Scale in mm.) 127 less. Shape circular, fiat dorsally but convex ventrally, i.e., top-shaped where it fits into a pit on the leaf. Bearing a thick dorsal plate of wax which covers the body except for the sub- margin, orifice, and caudal area. Margin crenu- late, anterior and posterior marginal setae pres- ent. Submarginal groove present, contouring the margin except caudally. Fourteen submarginal setae on each side, 5 anterior to thoracic tra- cheal pore and 9 posterior. Thoracic tracheal fold absent. Thoracic pore (Fig. lOe) flask- shaped, reaching mesally nearly to submarginal groove; walls thicksided with numerous thick- enings, 4 or 5 small teeth on lip of pore which is slightly concave. Cephalic, thoracic, and ab- dominal areas of dorsal disc covered with dense circular simple pores. Lateral arms of transverse moulting suture straight meeting in very obtuse angle medially. Abdominal area densely covered with circular simple pores 0.01 mm. diameter. Segments each with a single transverse row of pores. Only 7 segments apparent. Eighth ab- dominal setae longer than orifice, bases distant from rim, and level with anterior margin of orifice. Vasiform orifice (Fig. 10/) subcircular, 0.04 mm. long, 0.037 mm. wide. Operculum fill- ing orifice. Lingula subparallel-sided, apex rounded. Anal furrow present between posterior end of vasiform orifice and caudal setae, the ridges conturing the posterior part of the orifice. Caudal setae subequal in length to submarginal setae, bases nearer to margin than to orifice and narrower than orifice. Abdominal tracheal fold and pore absent. HOLOTYPE: Pupal case on slide mount. De- posited in Institut Frangais d’Oceanie, Noumea, New Caledonia. PARATYPES: In author’s collection. TYPE LOCALITY: New Caledonia, Tinip, coll. L.J.D., 4/11/54. FOOD PLANT: Undetermined. Occurring in pits on the tomentose under- surface of the leaf. Gomenella reflexa sp. n. Fig. 11 ctr-g LARVA. ? Instar (Fig. lla-c) : Length 0.26 mm., width 0.21 mm. Colour pale except diamond-shaped area surrounding orifice and 8th 128 PACIFIC SCIENCE, Vol. XV, January 1961 segment. Subcircular dorsal surface, body top- shaped. Margin barely crenulated, submarginal pores broad. Thoracic tracheal fold visible on ventral surface as asperities. Thoracic tracheal combs (Fig. 11 b) present as small marginal teeth with 2 or 3 obscure teeth mesally. Thoracic and abdominal segmentation indistinct. Legs tapering, then widened distally. Vasiform orifice (Fig. 11c) subcordate with a small cell apically. Eighth abdominal setae short, fine, level with an- terior margin of orifice. Operculum subcordate, filling orifice. Lingula included, parallel-sided, apex rounded. Caudal setae short, fine, bases not as far apart as bases of 8th abdominal setae. These larvae occurred in pits on the upper surface of the leaves of Grevillea gillivrayi on Mont Dore (coll. L.J.D.). The disc of the dor- sum is covered with a shieldlike raised white wax plate, except for the orifice and the caudal area (Fig. lid). They are considered to be those of reflexa because of the brown pigmenta- tion and the pit-making habit on the upper surface of the leaf. PUPAL CASE (Fig. 11 e—g ) : Length 0.60 mm., width 0.58 mm. Colour dark brown. Shape cir- cular, flat dorsally but convex ventrally where it fits into a pit on the leaf. Apparent margin straight. Margin reflexed ventrally, distant about 0.1 mm. from apparent margin except posteriorly where it arises from near the anterior end of the abdominal tracheal pore, crenulated or with 11-12 short rounded teeth in 0.1 mm. A sub- marginal line between the apparent margin and the margin, nearer to the latter. Area between margin and apparent margin radially striate. Setae absent, sutures and segmentation as in Figure 11^. Minute scattered simple pores on body. Thoracic tracheal folds (Fig. 11/) pres- ent, beset with minute dots. Thoracic pore flask-shaped, invaginated, and with thick sides, the inner walls covered with short transverse projections, outer margin toothed. Abdomen, first abdominal segment short, 7th apparently FIG. 11. Gomenella reflexa sp. n. a, Third instar larva; b, third instar larva, thoracic tracheal pore; c, third instar larva, vasiform orifice and caudal mar- gin; d, third instar larva, wax plate and pit; e, pupal case; f, pupal case, thoracic tracheal pore; g, pupal case, vasiform orifice and caudal margin. (Scale in mm.) Aleyrodidae of New Caledonia — Dumbleton 129 absent. Abdominal tracheal fold present, with- out dots, abdominal tracheal pore similar to thoracic pore. Yasiform orifice (Fig. llg) cor- date, 0.05 mm. long, 0.05 mm. wide. Eighth abdominal setae minute, at half the length of orifice. Operculum 0.04 mm. long, 0.04 mm. wide, subcordate, completely filling orifice. Lingula 0.025 mm. long, 0.012 wide, finger- like, slightly constricted before base, apex rounded, completely included. HOLOTYPE: Pupal case on slide mount. De- posited in Institut Francais d’Oceanie, Noumea, New Caledonia. PARATYPES: In author’s collection. TYPE LOCALITY: New Caledonia, Riviere des Pirogues, coll. L.J.D., 31/7/54. FOOD PLANT: Maxwellia sp. The species is unusual in that it occupies pits on the dorsal surface of the leaves; the ventral surface of the leaves bears dense brown tomentum. GENUS Leucopogonella nov. PUPAL CASE: Margin entire or teeth only faintly indicated. A submarginal linear row of pores or papillae, giving the effect of an inner row of teeth. Submargin not separated from dorsal disc except in pale species. The row of pores is broken by 9 processes on each side, producing either 2 minute teeth on the margin or a marginal concavity between these points. Thoracic tracheal folds, pores, and combs ab- sent. Without rhachis. Without contiguous pores on first abdominal segment. Vasiform orifice small subcordate. Eighth abdominal setae slightly anterior to orifice. Caudal setae close together, bases not as wide apart as width of orifice. TYPE SPECIES: L. sinuata sp. n. This genus perhaps comes nearest Aleuro - trachelus but differs in the absence of an inner row of marginal teeth and in the absence of the nearly contiguous paramedian processes on the 1st abdominal segment. KEY TO THE NEW CALEDONIAN SPECIES ( based on pupal cases ) 1. Pale species . pallida sp. n. Black species 2 2. Without abdominal tracheal comb, margin not sinuate apectenata sp. n. Abdominal tracheal comb present, margin sinuate 3 3. Wide abdominal ridge, no dark-bordered pores laterad on dorsal disc, caudal setae near margin sinuata sp. n. Narrow abdominal ridge, 2 dark-bordered pores laterad on abdomen and on thorax, caudal setae nearer apex of orifice than to margin simila sp. n. Leucopogonella apectenata sp. n. Fig. Ylct^d PUPAL CASE (Fig. 12 a-d) : Length 0.68 mm., width 0.41 mm. Colour black. Shape elliptical, not constricted across thoracic folds or invag- inated caudally. Pupal case sunk in raised vi- treous wax over which lie 8 ribbons of white £2 , d FIG. 12. Leucopogonella apectenata sp. n. a, Pupal case; h, wax fringe; c, pupal case, margin; d, pupal case, vasiform orifice and caudal margin. (Scale in mm.) 130 wax (Fig. 12 b) on each side and single an- terior and posterior median ribbons. Margin (Fig. 12c) reflexed, obscurely toothed. Apparent margin smooth except for 9 minute paired teeth on each side more or less equidistant, the margin between them straight or nearly so; 4 on the cephalothorax, 5 on the abdomen with the anterior one opposite the end of the trans- verse moulting suture. Caudal margin with 2 prominences. Submargin not separated from dorsal disc and pigmented to margin. On the submargin is a continuous linear series of contiguous subquadrate pores, about 10 between each pair of marginal processes. The pores tend to be longer than wide. A minute pore mesad of alternate large pores. The line of pores be- tween the processes is slightly concave. The processes are flanked on each side by a smaller pore. Cephalic area with pores as in figure. Meso- and metathorax each with 2 pairs of double pores, one larger and one smaller. Tho- racic tracheal folds, pores, and combs absent. Lateral arms of transverse moulting suture nearly straight, meeting in very obtuse median angle. Abdomen, pores on first abdominal seg- ment wide apart in same line as 2 pores on each side at mid-length, paramedian pores on second segment, 8th segment seta bases separated from orifice and level with its anterior margin. Vasi- form orifice (Fig. 12*7) semicircular or broadly cordate, wider than long. Operculum fills orifice. Lingula short included, parallel-sided apex rounded. Distance between caudal seta bases about half width of orifice, bases nearer apex of orifice than to margin. Abdominal tracheal folds, pore, and comb absent. HOLOTYPE: Pupal case on slide mount. De- posited in Institut Fran^ais d’Oceanie, Noumea, New Caledonia. PARATYPES: In author’s collection. TYPE LOCALITY: New Caledonia, Tontouta River, coll. L.J.D., 27/4/53. food plant: Leucopogon sp. Leucopo gonella pallida sp. n. Fig. 13 a-c PUPAL case (Fig. 13 a-c ) : Length 0.78 mm., width 0.42 mm. Colour white or colourless. Shape elongate-ovate with an anterior median PACIFIC SCIENCE, Vol. XV, January 1961 prominence (Fig. 13 b) , without indentations in thoracic pore area. Margin crenate, scarcely toothed. Submargin defined by submarginal line. On the submargin a linear series of contiguous pores broader than long except in cephalic area, broken by 9 processes or openings, 5 on the cephalothorax and 4 on the abdomen. Pores as in figure. Most prominent are 2 submarginal pores on posterior half of thorax. Thoracic tra- cheal fold, pore, and comb absent. Thorax with a median ridge obscurely tuberculate and dim- inishing in width anteriorly. Abdomen with slight median ridge faintly outlined by minute tubercles. First segment without evident setae or pores. Segment 7 longer than 6. Eighth ab- dominal segment setae minute, slightly anterior Fig. 13. Leucopogonella pallida sp. n. a, Pupal case; b, pupal case, anterior margin; c, pupal case, vasiform orifice and caudal margin. (Scale in mm.) Aleyrodidae of New Caledonia — Dumbleton 131 to orifice, and distant from it. Vasiform orifice (Fig. 13c) subcordate, not notched posteriorly but contoured posteriorly by a semicircular line. Operculum subcordate, filling orifice. Lingula short, rounded apically. Caudal setae minute, bases as wide apart as those of the 8th abdominal setae. Two lateral discal pores on each side op- posite segments 5 and 7. Dorsal disc, especially between abdominal ridge and submargin, beset with subcircular pustules. Abdominal tracheal fold faintly indicated, finely stippled from ori- fice half way to margin. Abdominal tracheal comb present with 4 to 6 teeth which are longer than wide. HOLOTYPE: Pupal case on slide mount. De- posited in I ns ti tut Frangais d’Oceanie, Noumea, New Caledonia. PARATYPES : In authors collection. TYPE LOCALITY: New Caledonia, Plum, coll. L.J.D. FOOD PLANT: Leucopogon sp. This species shows some affinities with Aleuro - paradoxus Q. & B. and Pentaleyrodes Takahashi. Leucopogonella simila sp. n. Fig. lia^c PUPAL CASE (Fig. 1 4a-c) : Length 0.89 mm., width 0.57 mm. Colour dark but paler than sinuata. Shape elliptical, margin less sinuate than sinuata. Margin (Fig. 1 4b) with 9 proc- esses on each side less prominent than in sinuata. Between the processes on the submargin are 14 or 15 pores in a linear series. Pores tending to be longer than wide. Disc pores as in figure. Laterad on or near the submargin is a row of minute paired pores. Mesad of these on the ab- domen at mid-length is a pair of larger dark- bordered pores and a similar pore laterad on the mesothorax. The abdominal ridge is nar- row. Bases of the 8th abdominal setae slightly anterior to orifice and wider apart than the orifice width. Vasiform orifice (Fig. 14 c) sub- cordate, similar to that of sinuata. The bases of the caudal setae are slightly nearer to the apex of the orifice than the margin and about the same distance apart as the orifice width. Ab- dominal tracheal fold absent, comb of 8 to 9 teeth whose rounded apices are level with mar- gin. HOLOTYPE: Pupal case on slide mount. De- posited in Institut Frangais d’Oceanie, Noumea, New Caledonia. PARATYPES: In authors collection. TYPE LOCALITY: New Caledonia, Plum, coll. L.J.D. Fig. 14. Leucopogonella simila sp. n. a, Pupal case; b, pupal case, margin; c, pupal case, vasiform orifice and caudal margin. (Scale in mm.) 132 PACIFIC SCIENCE, Vol. XV, January 1961 food plant: Leucopogon sp. This species occurred on the same food plant with L. sinuata. LeucopogoneUa sinuata sp. n. Fig. 13 a, b PUPAL CASE (Fig. 15 a, b ) : Length 1.0 mm. Width 0.66 mm. Colour black. Shape elliptical. Margin entire or faintly crenulate, sinuate. Sub- margin distinctly paler than disc but not sep- arated by submarginal line. Margin with 9 proc- cesses on each side, concave between these proc- esses. Submargin with continuous linear series of pores. These are pale subcircular or sub- elliptical areas partly separated by blunt conical black processes of the dorsal disc. About 12-13 pores between processes. A somewhat irregular row of minute pores mesad of the larger linear series. Thoracic area without tracheal folds, pores, or combs. Segments bounded laterally about midway between median line and mar- gin. Abdomen with segments laterally bounded. Eighth abdominal segment with setae bases wider apart than orifice and distant from and slightly anterior to it. Vasiform orifice (Fig. 15 b) transverse, subcordate. Operculum filling orifice. Lingula. Abdominal tracheal fold absent, tracheal comb present with about 10 teeth, each longer than wide, rounded apically, and some- what fluted. Caudal seta bases nearer to margin than to apex of orifice, as wide apart as width of orifice. holotype: Pupal case on slide mount. De- posited in Institut Frangais d’Oceanie, Noumea, New Caledonia. PARATYPES: In author’s collection. TYPE LOCALITY: New Caledonia, Plum, coll. L.J.D. FOOD PLANT: Leucopogon sp. Occurred with simila at Plum on the same food plant, and also at Yate, Montagne des Sources, and Mt. Mou. GENUS Orchamoplatus Russell ( Proc. Hawaiian Ent. Soc. 16 (3): 389- 410. New name for Orchamus Q. & B. pre- occupied in Orthoptera.) The following species of this genus were described and keyed in a previous paper (Proc. R. Ent. Soc. Lond. (B) FIG. 15. LeucopogoneUa sinuata sp. n. a, Pupal case; b, pupal case, vasiform orifice and caudal margin. (Scale in mm.) Aleyrodidae of New Caledonia — DUMBLETON 25, pts. 7-8, pp. 131-141, 1956): caledonicus Dumbleton, dentatus Dumbleton, incognitus Dumbleton, montanus Dumbleton, plumensis Dumbleton, and porosus Dumbleton. Russell described another species, O. noumeae, from citrus in New Caledonia and keyed the existing species of the genus, which include 0. calophylli Russell from Tonga. Orchamoplatus per dentatus sp. n. Fig. I6a-c pupal CASE (Fig. 1 6a-c ) : Length 0.95 mm., width 0.79 mm. Colour black. Widest slightly behind mid-length, flat subcircular, deeply con- stricted across thoracic combs. Margin with teeth prominent, rounded, as long as wide. Sub- marginal line distant from margin, meeting and continuous with transverse moulting suture on each side and together completely enclosing the cephalothoracic area. Submargin with a single row of papillae, about 18 anterior to the tho- racic tracheal comb and 37 posterior to it on each side. Eye spots triangular, distant from submarginal line. Thoracic tracheal fold not in- dicated, comb (Fig. 1 6b) with about 6 teeth which are more rounded apically than in denta- tus. First abdominal segment with a pair of setae and the usual paramedian pores on seg- ments 2 to 5. Segment 6 with 1 pair of spe- cialised papillae identical with those on the sub- margin. Eighth abdominal seta on tubercles about mid-length on rim of orifice. Vasiform orifice (Fig. 1 6c) subcircular, 0.07 mm. long, slightly transverse. Operculum filling orifice. Lingula not visible. Caudal setae 0.05 mm. long. Abdominal tracheal fold faintly indicated, comb with 6 or 7 teeth, separated, rounded apically. HOLOTYPE: Pupal case on slide mount. De- posited in Institut Fran^ais d’Oceanie, Noumea, New Caledonia. TYPE LOCALITY: New Caledonia, coll. F. Cohic. food PLANT: Undetermined. The unique specimen is distinct. O. citri Ta- kahashi, the only other species with papillae on the disc of the dorsum, has several papillae laterad on the abdomen and is pale in colour. The species is separable from O. dentatus Dum- bleton by the following characters: more cir- 133 h Fig. 16. Orchamoplatus perdentatus sp. n. a, Pupal case; h, pupal case, thoracic tracheal pore and comb; c, pupal case, vasiform orifice and caudal margin. (Scale in mm.) 134 PACIFIC SCIENCE, Vol. XV, January 1961 cular shape, more broadly rounded anteriorly, more numerous submarginal papillae especially on abdomen, cephalothoracic area completely enclosed by submarginal line and transverse moulting suture, 2 papillae on the 6th abdominal segment, eye spots distant from submarginal line, suture between 7th and 8th abdominal segments and pockets very close to anterior margin of vasiform orifice. In this latter char- acter the species resembles plumensis Dumble- ton. GENUS Parabemisia Takahashi Parabemisia reticulata sp. n. Fig. lla-c pupal CASE (Fig. lla-c ) : Length 1.0 mm., width 0.78 mm. Colour transparent or white, j Shape elliptical or subcircular, flat, constricted | between thoracic clefts, caudal cleft invaginated. Margin with teeth absent but with short nar- row ridges immediately mesad of the margin. Anterior and posterior marginal setae present. Submargin not defined. Segmentation as in fig- ure. Thoracic tracheal combs absent, clefts pres- | ent. Pore (Fig. lib) inset from margin, opens into tracheal fold which is hexagonally reticulate for half its length. The reticulated area is some- what variable, sometimes longer and expanded mesally. Abdomen without setae on 1st segment, 7th segment shorter than 6th and 8th. Eighth 1 abdominal setae minute, close to anterior lateral : margin of orifice. Vasiform orifice (Fig. 17c) longer than wide, subtriangular but rounded apically, floor with 8-9 subparallel transverse ridges. Operculum sub-semicircular, wider than long, less than half as long as orifice, more than half length exposed with lateral knob at base of exposed part on each side. Caudal setae small, midway between apex of orifice and margin. Abdominal tracheal cleft and pore similar to thoracic, fold reticulate for only a short distance cephalad. HOLOTYPE: Pupal case on slide mount. De- posited in Institut Frangais d’Oceanie, Noumea, New Caledonia. FIG. 17. Parabemisia reticulata sp. n. a, Pupal case; b, pupal case, thoracic tracheal pore; c, pupal case, ! vasiform orifice and caudal margin. (Scale in mm.) Fig. 18. Tetraleurodes submar ginata sp. n. a, First instar larva; b, third instar larva; c, second instar larva; d, pupal case; e, pupal case, abdominal sculpturing; f, pupal case, vasiform orifice and caudal margin. (Scale in mm.) P ARATYPES : In author’s collection. type LOCALITY: New Caledonia, Montagne des Sources, coll. L.J.D., 17/12/54. FOOD PLANT: Undetermined. The facies of this species is that of Bemisia but the lateral knobs on the lingula suggest Parabemisia and it is placed in that genus, though in the absence of marginal setae and tracheal combs and the presence of tracheal clefts like those of Dialeurodes it is not typical. GENUS Tetraleurodes Cockerell Tetraleurodes submar ginata sp. n. Fig. 18a- f larva. First instar (Fig. 18 a): Length 0.24 mm., width 0.13 mm. Colour pale. Shape elon- gate-elliptical. Submargin with 6 short setae on each side. Cephalic area with a raised median prominence and 2 short spines anterior to it; cephalic spines long, reaching to first abdominal 136 PACIFIC SCIENCE, Vol. XV, January 1961 segment. Spines on first abdominal segment long, reaching to orifice. Caudal spines as long as anterior cephalic spines. Second instar (Fig. 18c): Length 0.39 mm., width 0.21 mm. Colour pale. Shape elongate- elliptical, slightly tapering posteriorly. Margin crenulate. Inner row of teeth faint or absent. Submarginal line present. Caudal setae and first abdominal setae present. Third instar (Fig. 18 b) : Length 0.70 mm., width 0.52 mm. Colour light brown, submargin darker. Shape elongate-elliptical, distinctly ta- pering posteriorly. Margin with 2 rows of teeth. Transverse moulting suture not evident. Proc- esses on first abdominal segment thick and long. No sculpturing evident on abdominal segments. PUPAL CASE (Fig. 18 d-f): Length 1.1 mm., width 0.72 mm. Colour black. Shape elliptical, convex. Margin toothed with 2 rows of teeth. Outer row longer than wide, rounded apically, dark brown or black. Submargin wide, sharply defined by groove which contours the margin completely. On submargin mesad of inner teeth are small circular dots behind alternate teeth. Anterior and posterior marginal setae present, short. Submargin in cephalothoracic region with 4 larger pores on each side abdomen with 2 such pores on each side posteriorly at mid-length. Cephalic region with subcircular eye spots and 1 pair short paramedian setae. Pores on ce- phalothorax as in figure, many paired, especially laterad. Thoracic tracheal folds, pores, and combs absent. Abdomen, first segment with 1 pair of short paramedian setae and with toothed sculp- turing (Fig. I8e) anterior to base of each seta. Segments 2-7 with similar sculpturing and with paramedian pores. Paired pores laterad on disc and adjoining submarginal line. Eighth segment with seta bases distant from and anterior to orifice. Vasiform orifice (Fig. 18/) subcordate, not toothed, slightly raised. Operculum sub- cordate, filling orifice. Lingula not discernible. Caudal setae on submargin, long, projecting be- yond margin, bases as wide as orifice. Abdominal tracheal fold, pore, comb, and anal furrow absent. HOLOTYPE: Pupal case on slide mount. De- posited in Institut Frangais d’Oceanie, Noumea, New Caledonia. PARATYPES: In author’s collection. TYPE LOCALITY: New Caledonia, Dothio, coll. F. Cohic. FOOD PLANT: Undetermined (? Eugenia). This species would run to Hempelia Sampson & Drews in Sampson’s key but is unlike that genus. In facies it is a T etraleurodes with 2 rows of marginal teeth, or an Aleurotrachelus with a submarginal groove. The Ascidians of Point Barrow, Alaska, Part 1 Suborder Phlebobranchia (Enterogona) Donald P. Abbott 1 During the years 1948 to 1951 sizeable col- lections of ascidians were made by workers at the Arctic Research Laboratory, Point Barrow, Alaska. Most of the specimens were dredged during the summers of 1948 to 1950 under the supervision of Professor G. E. MacGinitie of the California Institute of Technology, in the course of an extensive survey of the distribu- tion and ecology of the marine invertebrates of Point Barrow. 2 This material was forwarded to me for study through the kindness of Professor MacGinitie and of Drs. Paul Illg and Fenner A. Chace, Jr., of the Smithsonian Institution, U. S. National Museum. Additional small col- lections of ascidians were made in 1950-1951 by Dr. Ira L. Wiggins and Mr. James Bohlke of Stanford University; these have been made available to me by Miss Margaret Storey of the Natural History Museum, Stanford University. It is a pleasure to express my thanks to those named above. A general account of the environment and marine invertebrate fauna of Point Barrow is now available (MacGinitie, 1955). This gives complete locations and descriptions of dredging stations 1-62 (MacGinitie, 1955: 60-85, fig. 1), and includes some information on the dis- tribution and ecology of the ascidians (pp. ISO- 183). The ascidian names used by MacGinitie are based on determinations made by myself. McGinitie’s locality records are more compre- hensive than those included here, for he has listed some records resulting from field identi- fications (using determined comparison speci- mens) of material which was not preserved for later study. In the present systematic account I 1 Hopkins Marine Station of Stanford University, Pacific Grove, California. Manuscript received October 12,1959. 2 Supported by funds from the Office of Naval Re- search, Contract and Task Order no. N6-onr 24316, Project no. NR 162 911. have cited only specimens which I have exam- ined personally. The collections contained 27 ascidians be- longing to the suborder Phlebobranchia, rep- resenting 3 species: Ascidia callosa , Chelyosoma macleayanum, and C. inae quale. The first 2 species are familiar arctic forms; while well known, much remains to be learned of their variability. C. inaequale is previously known only from 10 specimens, plus three very dubious records from deep water off southern California and Panama (Van Name, 1945: 209). Collect- ing data on the material examined are listed in Table 1. In the following section the references listed are limited to a few earlier works which have made important contributions or which con- tain good descriptions or extensive bibliog- raphies on the species concerned. The descrip- tions given are intended to supplement, not duplicate, the existing accounts cited for each species. order ENTEROGONA suborder PHLEBOBRANCHIA family ASCIDIIDAE Ascidia callosa Stimpson, 1852 Ascidia adhaerens Ritter, 1901, pp. 227-230, pi. 27, figs. 1-5. A. ( Ascidiopsis ) columbiana Arnbiick, 1934, pp. 53-55, text figs. 13-14. A. callosa Hartmeyer, 1924, pp. 41-53; Huus, 1930, pp. 1-11, figs. 1-6; Arnback, 1934, pp. 49-52, pi. 4, 6, figs. 19-26, 43-44, text fig. 12; Van Name, 1945, pp. 178-180, pi. 4, fig. 5, text figs. 92-94. Ascidiopsis columbiana Huntsman, 1912, pp. 110-113, pk 10, fig. 5, pi. 14, figs. 5, 7, 8. DESCRIPTION: The following notes supple- ment the description given in Van Name 137 138 PACIFIC SCIENCE, Vol. XV, January 1961 TABLE 1 Collecting Station Data on Phleboranchiate Ascidians Examined Stations 20-61 were occupied by MacGinitie, stations B51-31 to B51-33 by Bohlke ! SPECIMENS EXAMINED TYPE OF BOTTOM AND Chely- osoma Chely- STATION DEPTH REMARKS Ascidia macleay- osoma NUMBER DATE (ft.) (after MacGinitie, 1955) callosa anum iu ae quale 20 9 Sep 48 125 stones (sea urchins, Psolus, sea 3 anemones) 23 15 Sep 48 130 stones (sea anemones, Psolus, sea 3 urchins ) 26 9 Aug 49 130 stones, gravel 1 27 9 Aug 49 420 stones, gravel 1 1 32 17 Aug 49 741 mud (worm tubes) 1 36 6 Sep 49 477 few rocks (worm tubes) 1 37 6 Sep 49 227 stones, large perforated rocks 4 42 6 Oct 49 216 rocks, stones ( Psolus , sea urchins) 2 44 1 1 Oct 49 453 rocks, stones, small amount of 1 2 1 gravel ( Psolus ) 45 1 1 Oct 49 341 stones, gravel, few rocks (sea urchins) 1 61 5 Aug 50 204 mud, stones, gravel 1 B51-31 29 Jul 51 ! 151 1 B51-32 29 Jul 51 164 2 B51-33 29 Jul 51 | 144 1 (1943). Tunic highly variable with size, age, and environment; in smaller specimens less than 1 mm. thick, membranous, colorless, trans- parent, smooth or slightly tuberculate, some- times papillate about the apertures; in larger individuals up to 9 mm. thick, gelatinous to cartilaginous, dirty tan to olive brown, often wrinkled and encrusted with debris. Siphons ranging from very short to nearly one-third of the total body length. Oral aperture with 6-8 lobes, atrial with 6 lobes; some individuals with orange pigment spots between adjacent lobes on both siphons. Tentacles of 3 or 4 sizes, variable in number (Table 2), and regular to irregular in arrangement. Dorsal tubercle usu- ally with a U-shaped slit, the opening some- times canted toward the left; horns of the slit conspicuously outrolled in the largest specimen examined (sta. 44). Internal longitudinal ves- sels variable in number (Table 2), bearing both regular and intermediate papillae (the lat- ter often irregularly distributed). Transverse vessels of 5 orders present in larger specimen from sta. 42, the smallest order being parastig- matic, irregularly distributed, incompletely de- veloped, and corresponding in position to the intermediate papillae of the internal longitu- dinal vessels. Stigmata 2-20 per mesh. Renal vesicles forming a conspicuous layer of small bladders, particularly on the gut and adjacent mantle on the left side. Gonads well developed only in the 3 specimens from sta. 20 ( 1 speci- men, only 11 mm. long, bore a single row of large eggs in its oviduct), all of these with de- veloping larvae in the atrial chambers. DISCUSSION: This species has often been con- fused with Ascidia obliqua and A. prunum, which also occur in arctic waters. It differs from A. obliqua in bearing intermediate papillae on the internal longitudinal vessels and in its pos- session of abundant renal vesicles. It is dis- tinguished from A. prunum most conspicuously by its much smaller number of internal longitu- dinal vessels and by its habit of brooding Alaskan Ascidians — A bbott 139 FIG. 1. Chelyosoma macleayanum. Dorsal view of disc; growth lines indicated only on intersiphonal plate and two siphonal plates, a, Individual with 7 marginal plates; b, individual with 8 marginal plates. embryos and immature larvae within the atrial cavity. The tentacle counts for the 4 largest speci- mens (Table 2) are higher than those usually obtained in A. callosa, although Arnback ( 1934: 53) records similar numbers. The difference here is probably not significant. Tentacle counts vary a good deal depending on the degree of development of a fourth cycle of very small tentacles; in the present study these tentacles were counted when present, whereas earlier workers have often ignored or overlooked them in making counts. The specimen from 74 1 ft. (123 fathoms) represents an unusual depth record for the species; Van Name (1945: 180) indicates a maximum depth for A. callosa of about 80 fathoms. DISTRIBUTION: Norway, N. Ireland (?), the Faeroes, Iceland, Greenland, eastern N. America south to Cape Cod, arctic N. America, western N. America south to Puget Sound, Bering Sea, and Korea. family RHODOSOMATIDAE Chelyosoma macleayanum Broderip & Sowerby, 1830 Fig. la, b Chelyosoma macleayanum Huntsman, 1921, pp. 27-33, pi. 1, fig. 1, text fig. 1; Hart- meyer, 1924, pp. 2-9, text fig. 36; Arn- back, 1934, pp. 76-84, pi. 5, figs. 27-33; Van Name, 1945, pp. 205-207, text fig. 118. DESCRIPTION : The following notes supple- ment the description of Van Name (1945). Dorsal disc of test divided into plates arranged as follows: 6 triangular plates forming a ring about each aperture; remainder of disc bearing 1 central intersiphonal plate and 7 or 8 mar- ginal plates, the 8th marginal plate, when pres- ent, occupying an asymmetric position to the right of the center of the intersiphonal plate ( Fig. 1 ) ; lateroventral plates below rim of disc TABLE 2 Ascidia callosa STATION NUMBER DIMENSIONS ( mm.) ORAL LOBES ATRIAL LOBES TENTA- CLES INTERI Right 'sfAL LONGITl VESSELS Left JDINAL Total 20 11X 6X 1* 8 32 19-20 16 35-36 27 I4x 9X 4 8 6 24 20 26 19 14 33 42 14X10X 4 8 6 61 16X16X 5 7 6 7 Of 20 16 36 32 33X24X12 7 6 ca. 50 18 16 34 42 41X28X13 6 6 ca. 65 17-18 19 36-37 44 75X53X35 ca. 55 26 23 49 * Measured on animal removed from tunic; all other size measurements in table made on intact animal in tunic, t Fourth order tentacles unusually numerous. 140 PACIFIC SCIENCE, Vol. XV, January 1961 TABLE 3 Chelyosoma macleayanum STATION NUMBER LENGTH OF DISC (mm.) NUMBER OF MARGINAL PLATES ON DISC NUMBER OF GROWTH LINES ON PLATES OF DISC 23 — 7 — 23 — 7 — 36 3 8 0 23 7 8 — 44 8 7 2 37 8 8 2 44 8.5 7 2 B51-32 10 8 2 37 11 8 2 45 12 8 2 B51-33 12 8 2 B51-32 13 7 2 37 14 8 1? 37 17 8 3 B51-31 30 8 3 indistinct or absent except in largest specimen (disc 30 mm. long); 0-3 growth lines present on plates of disc. Quantitative information is presented in Table 3. DISCUSSION: All of the specimens of this well-known arctic species obtained at Point Bar- row are assigned to the form typica (Arnback, 1934), which is characterized by possession of a single central intersiphonal plate and 5-9 marginal plates. Van Name (1945) considers the individuals with 7 more or less symmetrically placed marginal plates (Fig. la) to represent the normal condition, and this appears to be the case in northern Atlantic waters, but Arn- back has described a population from Pitlekaj on the Siberian arctic coast (67° 7 ' N., 173° 24' W.) in which the modal number of mar- ginal plates is 8, the 8th plate being located on the right as in the Point Barrow specimens. The sparse comparative data available ( Table 4 ) sug- gest a zoogeographic cline with respect to num- ber of marginal plates, and a re-examination of earlier collections of C. macleayanum with this possibility in mind should prove worthwhile. Growth lines on the test plates show a gen- eral increase in number with the size of the animal (Table 3), and suggest that the largest individuals taken were in their 4th year (see Huntsman, 1921). distribution: Circumpolar arctic: Russia, Norway, Iceland, Greenland, eastern N. Amer- ica south to Cape Ann, arctic N. America, Bering Strait, Sea of Okhotsk, Siberian arctic. Chelyosoma inaequale Redikorzev, 1913 Figs. 2, 3 a-e C. inaequale Redikorzev, 1913, pp. 206-208, fig. 2; Arnback, 1934, pp. 80-81; Van Name, 1945, p. 209. DESCRIPTION: Body elongate-oval, the dorsal surface flattened to form an oval disc bearing the siphons and surrounded hy a marginal ridge whose prominence varies with degree of body contraction. Both apertures 6-lobed. Tunic tough, flexible, translucent, grey-white to yel- lowish, locally thickened to form triangular plates on all siphon lobes (Fig. 3 b) and a con- spicuous array of warts, ridges, and small plates on the disc and marginal ridge, giving the dor- sal surface a tesselated appearance (Figs. 2, 3 a). Internal surface of tunic bearing a ridge below Fig. 2. Chelyosoma inaequale. Dorsal view of disc. Alaskan Ascidians- — Abbott 141 TABLE 4 Chelyosoma macleayanum , PER CENT OF POPULATION Bearing SPECIFIC Numbers OF Marginal Plates REFERENCE AND LOCATION NUMBER OF SPECIMENS EXAMINED NUMBER OF MARGINAL PLATES 5 6 7 8 9 Uncertain % % % % % % Hartmeyer, 1924; material mostly from Greenland 53 2 9 79 2 0 8 Present collections; Point Barrow, Alaska 15 0 0 33 67 0 0 Arnback, 1934; material from Siberian arctic. mostly Pitlekaj 185 0 0 11 87 2 0 each siphonal plate and numerous papillae dor- sally and dorsolaterally to which mantle muscles attach. Mantle thin, with muscles strongly de- veloped about the siphons and below the mar- ginal ridge of the disc but nearly absent else- where. Tentacles simple, of 2-4 orders (Table 5). Dorsal tubercle with a crescentic slit (Fig. 3 c); 20-40 dorsal languets of 1 or 2 sizes (Fig. 3 c; Table 5). Pharynx relatively complex and thick-walled (Fig. 3 d)\ 18-31 internal longitudinal vessels on each side, often inter- rupted, where incomplete their position being marked by papillae; transverse vessels irregularly branched, in places forming a perforated sheet medial to the stigmata; stigmata scattered, vary- ing from oval slits to spiral infundibula, ex- hibiting complex anastomoses with other phar- yngeal structures; external pharyngeal wall in places provided with branching bundles of mus- cle fibers. Anal margin lobulate (Fig. 3e) . Gonads situated in the intestinal loop and rami- fied over most of the inner and outer surfaces of the intestine. DISCUSSION: This remarkable species is known with fair certainty only from Redikor- zev’s original material (7 specimens, Sea of Ok- hotsk) and from 3 small individuals from Teller and Point Hope, Alaska, reported by Van Name (1945). It differs from all other mem- bers of the genus in lacking a clear series of dorsal plates arranged with fair geometrical reg- ularity and precision, the test instead being relatively smooth or dotted with scattered thick- enings and protuberances. The latter are espe- cially well developed in the present specimens, and, while irregular, do sometimes suggest a crude pattern consisting of a central intersi- phonal plate (divided in the median sagittal TABLE 5 Chelyosoma inaequale STATION DIMENSIONS OF TEST (mm.) NUMBER OF DORSAL LAN- INTERNAL LONGI- TUDINAL VESSELS NUMBER L W Ht TENTACLES GUETS Right Left 26 20 10 19 ca. 36 of 1st 2 orders; ca. 39 of 3rd and 4th orders 22, of 2 sizes 19 18 44 25 16 20 ca. 58 of 1st 2 orders; ca. 55 of 3rd order; 4th order very small, numerous, and ir- regularly placed 20 19 19 27 45 35 18 ca. 58 of 1st 2 orders; few scattered smaller tentacles 40 28-31 20-22 142 PACIFIC SCIENCE, VoL XV, January 1961 FIG. 3. Chelyosoma inaequale. a, Dorsal view of disc; b, oral siphon from anterior left side; c, dorsal tubercle and anterior dorsal languets; d, small area of pharynx, viewed from interior; e , anal margin. plane in the largest specimen) flanked by 2 or 3 rows of peripheral plates. Internally the speci- mens agree favorably with Redikorzev’s ac- count, particularly in the unusual complexity of the pharyngeal wall. The only previous figure of the species is Redikorzev’s sketch of the en- tire animal, which shows few details. distribution: Sea of Okhotsk, west coast of Alaska off Teller and Point Hope (but see Van Name, 1945 ) . REFERENCES Arnback-Christie-Linde, A. 1934. Tunica ta, 4. Cionidae, Ascidiidae, Agnesiidae, Rhodo- somatidae. In: Northern and arctic inverte- brates in the collection of the Swedish State Museum. K. Svensk. Vetens. Akad. Handl. (3rd ser.) 13(3): 1-91, 6 pis., 22 figs. Hartmeyer, R. 1924. Ascidiacea, Pt. 2. Dan. Ingolf-Exped. 2(7): 1-278, 45 figs., 1 map. Alaskan Ascidians — Abbott 143 Huntsman, A. G. 1912. Hoiosomatous ascid- ians from the coast of western Canada. Contrib. Canad. Biol. 1906-1910, pp. 103- 185, pis. 10-21. 1921. Age-determination, growth and symmetry in the test of the ascidian Chelyo- soma. Trans. Roy. Canad. Inst. 13(1): 27- 28, 1 pi., 3 figs. Huus, J. 1930. Ascidiologische Notizen, I. Zur Verbreitung, Morphologic und Biologie von Ascidia callosa Stps. Bergens Mus. Arbok Nat. Rekke 5: 1-11, 6 figs. MacGinitie, G. E. 1955. Distribution and ecology of the marine invertebrates of Point Barrow, Alaska. Smithson. Misc. Coll. 128(9): 1-201, 8 pis., 3 figs. Redikorzev, V. 1913. Neue Ascidien. Zool. Anz. 43: 204-213, 6 figs. Ritter, W. E. 1901. Papers from the Harriman Alaska Expedition, XXIII. The Ascidians. Proc. Wash. Acad. Sci. 3: 225-266, pis. 27-30. Van Name, W. G. 1945. The North and South American Ascidians. Bull. Amer. Mus. Nat. Hist., N. Y. 84, viii -f 476 pp., 327 figs., 31 pis. The Anatomy of Coluber radiatus and Coluber melanurus R. A. M. Bergman 1 Coluber radiatus is a well-known snake in Java. However, we found only 43 of these in a total of nearly 6,000 snakes collected over a num- ber of years before and after World War II in the north coast of Java. Looking at the small number of Coluber radiatus in our catches, one may get the impression that it may be much less common than was generally thought. The name radiatus is explained by Schlegel (1837: 135) as referring to the very typical black stripes on the head and on the sides of the body: L’occiput est marque dune raie transversale noire et un peu en croissant, a laquelle se joig- nent de chaque cote une raie plus etroite pro- venant de loeil: deux autres, naissant egale- ment sur les bords de cet organe descendent sur les levres. Le dos est orne de deux larges raies noires longitud inales; une autre plus etroite regne le long des flancs The popular name in Java is "ular tikus,” meaning the snake (that hunts the) rats, and it is a very apt name, as this snake is indeed a predator on rats and mice and as such is a very useful animal. It is incredibly swift in its move- ments: we had a C. radiatus in a rather small cage, about 100 cm. long and 80 cm. wide and deep, and a rat was put into it. For the first minutes nothing happened, but all of a sudden the snake moved and, in a matter of a few seconds, it caught the rat in a coil, crushing it to death. The whole attack happened so quickly that it was difficult to follow the series of move- ments involved. This snake is not poisonous and is in no way harmful to man, but it is aggressive, it strikes and bites easily. Sometimes it takes a remarkable attitude more of defense than of at- tack, as is described by van Heurn ( 1929 ) , who gives the following details: the first third of the body is bent in a U that points laterally, the 1 Royal Tropical Institute, Amsterdam. Manuscript received June 8, 1959- head is directed forward towards the enemy, the mouth is wide open. The neck is compressed laterally, contrary to the manner of cobras. The author illustrates his article with a clear draw- ing. MATERIALS AND METHODS: Our material con- sists of 43 snakes, 28 from Surabaia (8 females, 20 males), and 15 from Djakarta (5 females, 9 males, one new-born animal that has not been sexed). Kopstein (1941) measured the length of the body and that of the tail in 15 females and 16 males. In our series we took each animal that was offered, without attempting a selection. The snakes were killed by occipital puncture, weighed, perfused through the aorta with saline followed by Bouins liquid for hardening, and then the distances from the snout to the top and to the end of each organ were measured. For statistical analysis the work of Simpson and Roe (1939) has been followed. PATHOLOGY: Infestation with round worms was fairly common. Two males (body length r 1249 and 1252 mm.) were wounded, the first one at about one-third of its length behind the j head, the other one close behind the head. BLOOD: The blood of a female (body length 1156 mm.) was analysed in the laboratory of Prof. Radsma. The following results were ob- tained: Na 509, K 18.8, Ca 17.2, where the figures represent the number of mgs. per 100 ml. of plasma. SIZE: The maximum lengths observed by various authors are summarized, together with our data, in Table 1. In our material combined with the data gath- ered by Kopstein, the female group includes six very young animals, hardly more than new-born ones, as against a single very small one in the male group. To make both groups more com- parable, the five shortest females are not in- cluded in the data used for Figure 1. For our reasoning it does not make much difference whether we include them or leave 144 Coluber radiatus and C. melanurus — Bergman 145 Body I ength COLUBER RADIATUS + + oO ooqoq< p<^ 0 ^ 00 -f 0Gp0PC 000090°" + ) , + + FlG. 1 . Body lengths, male and female specimens. them out, because, in any event, the point of intersection between the first steep part of the curve and the plateau is to be found at a length of about 950 mm. The curve does not suggest a sexual dimorphism in body length. SEX RATIO: It is always a difficult question whether the findings on a small group can be used in a more general way. In the case of the sex ratio in C. radiatus, there are three inde- pendent observations, each one on a small group, but all pointing in the same direction. In Table 2 the figures of Kopstein (1938) and those of our material from Surabaia and Djakarta are shown. These figures are very suggestive of a sex ratio of two males to one female. However, in 1941 Kopstein published a list of measurements on 16 males and 15 females, those of 1938 in- cluded. This may signify that a few years more of collecting can change the first impression. On the other hand Kopstein, when collecting, has not included in his method the factor of TABLE 1 Coluber radiatus, LENGTH OF BODY AND OF TAIL author MALES FEMALES SEX UNKNOWN Body Tail Body I Tail Body Tail de Rooy Smith 1520 1267 1380 370 (293)* (311) 1445 1218 1352 350 (308) 308 1280 330 Kopstein Bergman * The figures for the length of the tail in parentheses do not refer to the animal whose body length is given. TABLE 2 Coluber radiatus, Sex RATIO AUTHOR MALES FE- MALES TOTAL Kopstein 1938 9 6 15 Bergman Surabaia 9 5 14 Bergman Djakarta . 20 8 28 Totals 38 19 57 not-selecting as a conditio sine qua non, and it seems probable that his first figure is more the result of a random sample than the second one. But even when we take Kopstein’s later figures, there is a greater number of males, still in the order of two males to one female. MATURITY: In our small series, the shortest female carrying growing eggs (which are 6 mm. in length) has a body length of 1006 mm. As we saw in the curve of successive body lengths ( Fig. 1 ) , the intersection of the ascent and the plateau is at about 950 mm. body length. We will not be very far from the real value when we take this figure as the lower limit of the adult size for both sexes. BODY LENGTH: When all animals are taken together, the average figure for the length of the body is 1096 mm. in males, 898 mm. in females. However, we have already noted that there is an excess of five new-born animals in the female group, which, in a total of 28 ani- mals, makes the average figure too low. If these five females are removed from the series, the average length for the females would be 1018 mm. Taking the adults alone (in this case the animals with a body length longer than 950 mm.), the average length for 38 males is close to 1170 mm., and for 16 females is 1178 mm., which is practically the same figure. It can be TABLE 3 Coluber radiatus, MUTILATION OF THE Tail TAIL Whole Mutilated Totals Males 35 10 45 Females 24 4 28 Total 59 14 73 146 PACIFIC SCIENCE, VoL XV, January 1961 TABLE 3^ TABLE 5 Coluber radiatus, MUTILATION OF THE Tail WHOLE MUTI- LATED TOTALS Young 17 2 19 Adult 42 12 54 Total 59 14 73 concluded, therefore, that there is no suggestion of sexual dimorphism in body length. TAIL: For a fighting species the number of broken tails does not seem excessively high: 14 in a total of 73. The figures for both sexes are given in Table 3- One might ask whether the males show more mutilated tails than do the females, or whether it is the other way round, or perhaps whether there is no difference at all. For the data shown here, x 2 ls very small, and the deviation from the hypothesis of in- dependency is very probably due to chance. In other words, there is no association between frequency of mutilation and sex. If we should take the young animals of both sexes together and, similarly, the older ones of both sexes, and divide these into groups of mutilated and of nonmutilated individuals, then Table 3 a will show these new figures, x 2 1 . 21 , and it is not. possible to see here an as- sociation between the length of life and the number of casualties. The length of the tail and the body for non- mutilated animals is given in Table 4, and the illustration in Figure 2. In both sexes the relative length of the tail TABLE 4 Coluber radiatus, LENGTH OF Body AND OF TAIL AVERAGE LENGTH OF THE: N Body Tail Tail length in °/ 00 of body length Males, juv 6 658 165 250 adult 28 1155 288 252 Female, juv. 6 640 153 239 adult 13 1170 281 240 Coluber radiatus, THE LENGTH OF THE TAIL in %cc of the Body Length N R Md= 950 mm.), TOPOGRAPHY N R M±trm o' ± o a V± .FEMALES r ,_ ( — - X : Lrzri Heart Liver ^.''Spleens Gonad Ga II b I adder Cloaca K i d ne Pacreas Fig. 3. Topographical pattern of the internal organs in males and females of Coluber radiatus. in the females. For the other organs the range of V is from 10 to 17. For the intervals the difference seems to be somewhat more marked, the values of V being greater in the female series. However, the small number of females influences the value of o-. Furthermore, the quotient D/ 950 mm.), TOPOGRAPHY N R M±crm a±aa V± 950 mm. body 38 957-1380 1170.0±15.2 94. 0± 10.8 8.0±0.9 tail 28 237-351 287.9^=4.7 24.6±3.3 8.5 ±1.1 heart 27 30-45 36.0±0.8 4.2±0.6 11.7 ±1.6 liver 27 185-291 236.8±3.9 20.2±2.7 8.5 ± 1.2 gallbladder 27 22-42 29.4±1.0 5.2±0.7 17.6±2.4 pancreas 27 12—27 18.9±0.6 3.3 ±0.4 17.5±2.4 spleen 27 4-12 6.9±0.4 2.3±0.3 33.0±4.5 sex R 27 41-61 47.7±1.1 6.1 ±0.8 12.7 ±1.7 sex L 27 30-59 4l.2± 1.4 6. 6 ±0.9 16.0±2.2 sex both 27 76—1 10 88.8±2.3 1 1.0± 1.5 12.4± 1.7 kidney R 27 76-134 105.7±2.9 14.9±2.0 14.1 ±1.9 kidney L 27 64-112 89-8±2.0 10.4 ±1.4 1 1.7 ±1.6 both kidneys 27 150-232 197.0 ±1.2 20.2±2.7 10.2± 1.4 Females > 959 mm. body 16 989-1352 1178.5±29.0 116.0±20.5 8.4±1.5 tail 13 247-308 280.4±5.0 1 7.9 ± 3 -2 6.4 ±1.3 heart 9 25-41 34.5 =±=1.7 5.2 ± 1 .2 1 5.1 ±3.6 liver 9 184-287 230.7 =±=11.3 33.1 ±7.8 14.3 ±3.4 gallbladder 9 19-33 26.5 ±177 5.0 ±1.2 19.5 ±4.6 pancreas 9 12-22 16.2 ±1.1 3.2±0.8 19.8±4.7 spleen 9 4-9 6.5±0.6 1.7 ±0.4 26.2±6.2 ovar. R 9 61-174 106.9±13.2 39-6±9.3 37.0±8.7 ovar. L 9 47-117 75.0±8.2 24.6±5.8 32.8±7.7 ovar. both 9 108-291 182.0±23.4 70.0±16.5 38.6±9.1 kidney R 9 71-126 95.7±4.9 16.3 ±3-9 17.0±4.1 kidney L 9 70-96 83.2±3.0 9.1 ±2.1 10.9±2.6 both kidneys 9 141-216 178.9±7.9 23.8±5.6 13-3 ±3-1 TABLE 11 Coluber radiatus, Male Adults, INTERVALS BETWEEN THE Organs N R M±cm a±aa V±o-v A1 snout— heart 27 199-264 241.2±2.9 14.8±2.0 6.1 ±0.8 A2 heart-liver 27 43-87 66.3±2.1 1 1.0± 1.5 16.6±2.5 A3 liver-gallbladder 27 95-179 1 27.3 ± 3 .8 19.8±2.7 1 5.5 ± 2.6 A 27 362-506 435.— ±7.5 39-— ±5-3 9— ±1.2 B1 pancreas— gonad 27 68-141 1 0 1 .5 ± 2.7 14.— ± 1.9 1 3.9± 1.8 B2 gonad R-kidney R 26 38-125 85.6±4.2 20.7±2.8 24.2 ±3.4 B3 kidney R-cloaca 26 77-120 1Q2.7±2.2 11.— ±1.5 1 0.7 ±1.5 B 26 215-333 289-2±4.9 25.0±3.5 8.6±1.2 Cl pancreas-gonad L 27 121-208 182.0±3.5 18 .-±2.4 9 .9 ±1.4 C2 gonad L-kidney L 26 48-113 82.6±3.4 17.3±2.4 21.2±2.8 C3 kidney L-cloaca 26 39-59 47 .6 ± 1 . 1 5.4±0.8 1 1 .4 ± 1 .6 C 26 218-366 31 1-5 ±5.8 30 .-±4.2 9.6± 1.3 A+B 26 577-839 724. 3± 10.8 55.— ± 7.6 7 . 6 ± 1 .— A+C 26 580-872 742.8 ±12.1 60.9±8.5 8.2 ±1.1 Dr pancreas-kidney R 26 162-284 232.9±4.7 24.7±33 10.6± 1.5 Dl pancreas-kidney L 26 213-376 305.5 ±5.5 30.6±4.3 10.-±1.4 Coluber radiatus and C. melanurus — Bergman 151 TABLE 12 Coluber radiatus, Female Adults, Intervals between the Organs N R M±crm a ± phalange, apical view, X 34; d, carpel apices and stigmas, an inner one at lower right, and three outer ones, oblique view, X 3; e, leaf base, lower side, X 34; f, leaf middle, lower side, X 34; g, leaf tip, lower side, X 34; b, leaf tip, lateral view, X 34; h leaf tip, lateral view, X 3. FIG. 7. Pandanus cochleatus, from holotype. a, Phalange, lateral view, X 34; b, phalange, longitudinal median section, X 34 \ c, phalange, apical view, X 34 \ d, carpel apices and stigmas, an inner one at left, and two outer ones, oblique view, X 3; e, leaf base, lower side, X 34; f, leaf middle, lower side, X 34; g, leaf margin at middle, X 3 \ h, leaf apex, lower side, X 34- Fig. 8. Pandanus Zea, holotype. a, Younger syncarp, X 3/14; b, mature syncarp, X 3/14; c, lateral drupes from above, X 34; d,e, lateral drupes, from above, X 3; f, lateral drupe from side, X 3; g, lateral drupe from side, X 3 4; ^ lateral drupe from side, X 3; i, lateral drupe, longitudinal median section, X 3; j, apical drupes, from above, ^>< 34; k,l, apical drupes, from above, X 3; m, apical drupe, oblique lateral view, X 3; n, apical drupe, lateral view, X 34; o, apical drupe, lateral view, X 3; p, apical drupe, longitudinal median section, X 3; q, leaf base, lower side, X 34; f, leaf middle, lower side, X 34; s, leaf apex, lower side, X 34* Fungi from Raroia in the Tuamotu Archipelago William Bridge Cooke 1 Between July 21 and September 6, 1952, M. S. Doty collected a number of specimens of fungi on Raroia (ca. 16°05' S., 142°25' W.) in the Tuamotu Archipelago. These were sent to the writer for identification. In turn, certain speci- mens were forwarded to specialists for more accurate identification than the writer could furnish. The geographic distribution of the fungi of the islands in the Pacific Ocean is poorly known, largely because usually only the more obvious things, such as Pycnoporus sanguineus, are col- lected. It is probable that the fungi of the Tua- motus have never before been collected as thor- oughly as they were on Dr. Doty’s visit. There are few habitats on an atoll for fungi. These are largely the dead wood of native or introduced trees, and driftwood. In addition to the seven species reported be- low, based on material in 19 collections, other species were found. However, some specimens were unidentified because they were immature at the time of collection, had passed maturity and were no longer producing spores, or were poorly dried because of inadequate drying facilities. The writer wishes to acknowledge the help of B. Lowy, J. H. Miller, L. E. Wehmeyer, John Eriksson, S. J. Hughes, J. A. Stevenson, and J. L. Lowe, in identifying this material. Duplicate specimens of all species listed have been for- warded to the Bishop Museum, Honolulu; and the National Fungus Collections, Beltsville, Maryland. Duplicates of several species were sent to the New York Botanical Gardens. ASCOMYCETES Anthostomella sp. On fallen rachis of leaf of Cocos nucifera. 11274 , 11277. July 21, 1952. 11277 is older than 1 Robert A. Taft Sanitary Engineering Center, Bu- reau of State Services, Public Health Service, U. S. Department of Health, Education, and Welfare, Cin- cinnati, Ohio. Manuscript received January 18, I960. 11274 and has more crowded perithecia. Dr. L. E. Wehmeyer stated that these specimens have the spores of A. melanosticta but that this species shows no surface blackening. Additional species which are similar include A. contaminans (Dur. & Mont.) Sacc, and A. lugubris (Rob. & Desm.) Sacc. The description of A. cocoes- capitaiae Cab. was not available at the time of examination. Hypoxylon chrysoconium Berk. & Br. Six collections were made from decaying wood of Guettarda speciosa or M esserschmidia argentea. 11279, 11280, 11374, 11389, 11390, 11393. July 21, 28, 29. Dr. J. H. Miller stated that "this is intermediate with Rosellinia but the perithecia are often united into an effused stroma. It occurs from India through the Ha- waiian Islands.” The mycelium on the surface of the wood is golden in color, making the species readily identifiable in the field. Two specimens were without perithecia and so were identified by analogy. Hypoxylon investiens (Schw.) Berk. Three specimens were obtained from decaying wood of Guettarda speciosa or M esserschmidia argentea. 11267, 11273, 11606. July 21, August 7. Dr. Miller indicated that the specimens were rather old, showed few perithecia, and that blackening of the wood was characteristic of the species. This species forms thin, brown, effused patches on decaying twigs, branches, and wood. Phaeopeltosphaeria irregularis Wehmeyer Two specimens of this species were found on old wood, probably drift wood. 1 1 164, 11394. July 9, 28. Dr. Wehmeyer indicated that this species has now been found on three atolls of the Marshall Islands and on the Galapagos Islands. He said that "it is apparently common in the South Pacific on driftwood.” 186 Fungi from Raroia-— Cooke 187 BASIDIOMYCETES Auricularia mesenterica Pers. Three collections of this widely distributed tropical species were taken from wood of Mes- ser schmidia argentea. 11262, 11386, 11661. July 21, 28, August 8. Schizophyllum commune Fr. Although the shape of the specimens obtained by Doty is typical of more northern material usually assigned to this species, this specimen could as easily have been placed in S. radiatum (Swartz) Fries 2 because of its tropical distri- bution. Throughout the Pacific Basin only one species is recognized in this genus at present. 11425. July 26. Thirteen collections of thelephoraceous fungi were sent to John Eriksson for identification. Unfortunately, Dr. Eriksson was unfamiliar with the species these collections represent. In addi- tion, the methods of preservation of these speci- mens make their identification quite difficult. Representatives of Peniophora sect. Coloratae , and sect. Hyphales were present. No spores and no fertile fruiting structures were present in the collections, although in some cases holobasidia were observable. It was suggested that through habitat, knowledge of sterile hymenial structures, and other features of a collection, a person fam- iliar with tropical species in this group could make tentative identifications. The same difficulties were found in the several collections of pore fungi which were made. A large trametoid specimen, a medium-size black Phellinus-like species represented by two normal specimens and two overgrown with a Penio- phora-\i\s.e species, and nine specimens of re- supinate polypores were among the materials obtained. Because of his wide familiarity with tropical polypores these specimens were sub- mitted to John A. Stevenson, National Fungus Collections, for study. The writer had found no satisfactory basidial or spore characters and hoped that sight identification might be possible. Neither Mr. Stevenson, nor Josiah L. Lowe who 2 Now considered to be synonymous with S. com- mune. also examined them, could come to any conclu- sion concerning species names for these speci- mens. The collection and adequate preparation of specimens of this type in tropical regions is dif- ficult. It has been suggested that if quick drying equipment is not readily available, the hymenia should be inverted to prevent loss of spores dur- ing drying. Care should also be taken to prevent the specimens from getting too wet during peri- ods of prolonged rain or high humidity. This condition tends to reactivate the hymenial tissues and exhaust the spore-producing capacity of the basidia, which then become unrecognizable structures. FUNGI IMPERFECT! Helicomyces roseus Link Two specimens composed of pieces of wood partially covered with a white effused growth were assigned to this species. Spores from this white growth were very similar to those described by Linder (Ann. Mo. Bot. Gard. 16: 271, 1929) for this species. The material was assigned to this species on that basis as well as on the fact that it resembles closely material described under this name by Rogers (Pacif. Sci. 1: 106, 1947), from the Marshall Islands. Dr. S. J. Hughes did not agree with this identi- fication but offered no substitute. 11283, 11602. July 21, September 6. Helicosporium lumbricoides Sacc. em. Matr. Two collections were made from material on old spathes of Cocos nucifera. 11269, 11270. July 21. Identification by S. J. Hughes. Oidium curtisii (Berk.) Linder Found twice on rotten wood. 11385, 11400. July 28, 29. Identified by S. J. Hughes. Additional collections of moldlike fungi were found to be sterile, or when spores were present it was difficult or impossible to determine the method of their production, so that they were unidentifiable. On the basis of the material seen, both that listed above and that which proved to be un- workable, it would seem that a considerable 188 number of fungus species is at work reducing the dead organic matter of the Pacific islands to humus and mineral material. In addition to fungi of these types there are also plant path- ogens, soil molds, and probably aquatic fungi on land, and marine fungi parasitizing algae and growing on the ocean floor, especially where there is decaying wood. Recent work in the North Atlantic indicates that there is a popula- tion of fungi in marine sediments to consider- able depth. According to the unidentifiable material in PACIFIC SCIENCE, Vol. XV, April 1961 Doty’s collections, there are several species of polypores, thelephores, and agarics in the my- cobiota of Raroia. In addition there are other types of molds associated with wood. In collect- ing such fungi great care must be exercised in the drying process to insure against the second- ary invasion of mold fungi in the tissues and sporiferous areas of the sporophore. It is hoped that additional collections will be made in the Pacific Basin in order that the fungal popula- tions and their activities may become better known. Notes on Indo-Pacific Scleractinian Corals, Part 3 A New Reef Coral from New Caledonia John W. Wells 1 In 1957 Dr. R. L. A., Catala of the Station de Biologic Marine, Noumea, New Caledonia, dis- covered the brilliant fluorescence in ultraviolet light of the polyps of reef corals living in deeper waters (see, Nature 1(83): 949, 1959; Life 47(3): 64-65, 1959; 2 6 franc postage stamp of New Caledonia issued March 21, 1958 ) . Dr. Catala kindly sent the writer a collection of these corals, together with photographs of many of them living in the Aquarium de Noumea. The specimens came from a depth of 35-40 m. on Banc Gail, in the lagoon of Noumea about 10 mi. from the Aquarium, and were collected by Dr. Yves Merlet, for whom the new species described below is named. The scleractinian fauna includes these species: Montip ora caliculata (Dana) M. verrucosa (Lam.) Goniopora lobata Milne Edwards & Haime Pavia speciosa ( Dana ) Favites ahdita ( Ellis & Solander ) Platygyra lamellina (Ehrenberg) Ecbinopora hirsutissima M. E. & LI. Trachyphyllia geoffroyi ( Audouin) Galaxea fascicularis ( Linn. ) Bantamia merleti sp. nov. Cynarina lacrymalis (M. E. & H.) 2 Protolobophyllia japonica Yabe & Sugiya- ma 2 Lobophyllia corymbosa (Forskaal) L. hemprichi (Ehrenb.) Symp by Ilia recta (Dana) Mycedium elephantotus (Pallas) Eupbyllia picteti Bedot 3 Piero gyra sinuosa (Dana) 1 Department of Geology, Cornell University, Ithaca, New York. Manuscript received February 1, I960. 2 These two species will be the subject of a subse- quent note. 3 Photographs by Dr. Catala of the living coralla of this species show an exact similarity, including color, between its polyps and those of Pectinia jardinei Saville-Kent (1893: 39; 1958: pi. 25, fig. 3, and chromo pi. 4, fig. 7) from the northern part of the Great Barrier Reef, and this form probably should be termed Euphyllia jardinei (Saville-Kent). E. picteti was originally described from Amboina by Bedot and has been reported by the writer (1955: 26) from Port Newry, Queensland. family OCULINIDAE SUBFAMILY GALAXEINAE genus Bantamia Yabe & Eguchi 1943 Bantamia merleti sp. nov. Figs. 1-4 Corallum fasciculate, 10 cm. in height, 10 cm. broad (holotype), formed by cylindrical corallites, 5-7 mm. in diameter, 4-7 mm. apart, arising by extratentacular budding from a very narrow edge -zone near the calices, at first sub- horizontal, then becoming erect and subparallel and losing organic connection with parents. Ex- terior of corallites costate only near calices, epithecate below, completely lacking any exo- thecal structures. Calices circular or slightly com- pressed, 5-7 mm. in diameter, shallow. Septa exsert 1-1.5 mm. near the wall, dropping to low inner lobes near the columella. Septal margins thickened where exsert, nondentate but finely granulated, the granulations extending down septal sides where they are evenly distributed. Septa of first two cycles ( 12 ) equal and extend- ing to columella; a few very thin, short septa of the third cycle developed in some systems. Costae weakly developed by narrow edge-zone near calices; in some corallites the edge-zone is not developed and the wall appears epithecal. Columella formed by interlaced loose trabecular processes from inner margins of septa, with one to three granulate papillae arising in bottom of the calice and commonly having a sublamellar aspect. Endotheca consisting of widely spaced, deeply concave single vesicles, the latest ones about 5 mm. below bottom of calice. Polyps pale brown with pale yellow-green peristomes. Peristomes fluorescing a pale emer- ald green, the remainder a pale orange-brown. The reference of this species to Bantamia is fairly certain on the basis of Yabe and Eguchi ’s careful description of the unique specimen of B. gerthi from the Miocene deposits of Java. B. merleti , also known from a unique specimen, lacks the feeble development of vesicular exo- 189 Indo-Pacifk Corals, 3 — Wells 191 theca found in the angles of the branches, has fewer septa in the adult corallites, and has con- cave rather than tabulate endothecal dissepi- ments. The range of Bantamia thus appears to be extended from the Miocene time to the Re- cent epoch. B. merleti is apparently not a surface reef type but is confined to the deeper lagoon waters. Yabe and Eguchi considered Bantamia to be closely related to Galaxea , differing only in the almost complete reduction in Bantamia of the extensive vesicular endotheca of Galaxea (or the converse?). This skeletal difference implies an almost complete lack of edge-zone in the polyps of Bantamia , a distinction confirmed by B. merleti. The striking similarity of the calice, septa, and columella of the two genera is em- phasized by comparison of the calices of B. mer- leti (Fig. 3) and G. fascicularis (Fig. 5). On the other hand, a color photograph by Catala of the living holotype colony of B. merleti in- dicates a considerable difference in the polyps of the two genera. This is not clearly shown in the enlargement (Fig. 4), for the polyps are evidently not fully expanded. According to Fowler (1890: 409) and Matthai (1914: 59), the tentacles of Galaxea are ento- and exocoelic, arranged in two rings of 12 each around the prominent (when expanded) oral cone. Saville- Kent (1893: chromo pi. 4, figs. 8-12) and Yonge (1930: 24, fig. 10) show them in a sin- gle ring near the edge of the oral disc, the 6 over the first cycle septa standing erect around the oral cone, the 6 of the second cycle and the 12 of the third cycle smaller and normally point- ing outwards. The polyps of Galaxea are usually partly expanded even in daylight. In B. merleti the tentacles total 96, arranged in five alternating rings over the outer half of the oral disc, in the center of which is a small protuberant oral cone. It would appear that the inner three rings, total- ing 24 tentacles, are entocoelic, overlying the entocoeles of the mesenterial couples containing the first two cycles and developed parts of the third cycle of septa, in decreasing order of size. The next 24 tentacles, near the periphery, are also entocoelic without corresponding fourth cycle septa. The outer marginal ring of 48 small tentacles is probably exocoelic and extends like a fringe around the edge of the oral disc. The external morphology of the living reef corals is not yet well enough known to evaluate the taxonomic worth of such differences in ten- tacular number and arrangement. Holotype to be deposited in the U. S. Na- tional Museum. Locality: 35-40 m. depth, Banc Gail, Nou- mea lagoon, New Caledonia. REFERENCES Bedot, M. 1907. Madreporaires dAmboine. Rev. Suisse Zool. 15: 143-292, 46 pis. Fowler, G. H. 1890. The anatomy of the Ma- dreporaria, Part V. Quart. J. Micr. Sci. 30: 405-422, pi. 28. Matthai, G. 1914. A revision of the Recent colonial Astraeidae possessing distinct coral- lites. Trans. Linn. Soc. Lond. Ser. 2 (Zool.) 17: 1-140, pis. 1-38. Saville-Kent, W. 1893. The Great Barrier Reef of Australia. London. Wells, J. W. 1955. A survey of the distribu- tion of reef coral genera in the Great Barrier Reef region. Rep. Gt. Barrier Reef Comm. 6: 21-29, chart. Yabe, H., and M. Eguchi. 1943. Note on the two Hexacoralla, Goniocorella dumosa (Al- cock) and Bantamia gerthi , gen. et sp. nov. Proc. Imp. Acad. Tokyo 19: 494-500, figs. 1-5. Yonge, C. M. 1930- Studies on the physiology of corals, I. Feeding mechanisms and food. Gt. Barrier Reef Exped. 1928-29 Sci. Rep. 1 : 13-57, figs. 1-34, pis. 1-2. FIGS. 1-4. Bantamia merleti n. sp. 1, 2, Lateral and calicular aspects of holotype corallum, XL 3, calices, X4; 4, partially expanded polyps of holotype, X4 (from color photo by R. Catala). Fig. 5. Calice of Galaxea fascicularis (Linn.), Amboina, X8 (Bedot, 1907: pi. 11, fig. 44). Porifera of Friday Harbor and Vicinity M. W. de Laubenfels 1 The Porifera of the northwest coast of North America were described rather thoroughly in three papers by L. M. Lambe (1892, 1893, 1894). Many sponges which he treated were from Van- couver Island, near the location herein discussed. Otherwise the Puget Sound Porifera have been little studied. The present paper is based on a study made by the author during the summer of 1931. In- tertidal species were studied living, in the field. Dredged specimens were examined practically immediately upon removal from the water. T. G. Thomson, director of the Oceanographic Lab- oratories at that time, and other members of the staff, rendered valuable cooperation for which gratitude is hereby expressed. While other invertebrates were found in con- siderable variety, amazingly few species of sponges were found to occur in the San Juan Archipelago. The number of individual speci- mens was large, so that one may conclude that conditions were suitable for the survival and growth of sponges. Notwithstanding careful search of the shores and almost daily dredging for several weeks, the collection consists of rep- resentatives of 16 genera, 17 species in all, 2 of which, Stylissa stipitata and Syringella amphi- spicula , are new to science. Two species of the class Calcispongiae were found. No dredging was done deep enough to bring up sponges of the class Hyalospongiae. The 15 species of the class Demospongiae represent 14 genera. CALCISPONGIAE Leucosolenia nautilia de Laubenfels, 1930 The sponge thus identified was collected July 4, 1931, near the town of Lopez in Lopez Island. It was dredged from a depth of about 60 m. It consists of a mass of tubes, each somewhat less than 1 mm. in diameter, occasionally an- astomosing, but much more frequently branch- 1 A posthumous paper, edited by Frank J. Little, Jr., Department of Zoology, University of Texas, Austin, Texas. Manuscript received February 11, I960. ing. The mass of tubes arises from a basal rhi- zome or reticulation of tubes affixed to the sub- stratum. The maximum distance from this which is reached by the ascending or descending tubes is barely 1 cm. The color in life was white; the consistency softly fragile. The occasional oscules are located at the distal extremities of the ascon tubes and are less than 1 mm. in diameter. The surface is optically smooth, and microscopically somewhat roughened by spicules which pro- trude, but not at right angles, they lie tangen- tially in the ectosome, their points directed to- ward the oscule. Around the oscule in each is found a coronal fringe of special oxeas about 6 X 300/a. The obliquely placed dermal oxeas are somewhat larger, say about 8 X 420/a. The prin- cipal spicules in among the flagellate cells of the ascon layer are triaxons with rays about 8 X 140/a. There are a few hypogastral quadri- radiates, of approximately the same size as the triradiates. This is the second record of this species, the first being by de Laubenfels (1930: 25) from Monterey Bay, California. The Puget Sound specimen has somewhat smaller oxeas echinant- ing the surface than do the California speci- mens, but otherwise it agrees very closely. The other species of Leucosolenia which is probably most closely to be compared here, and which therefore indicates the zoogeographical relation- ships of the form under discussion, is Leucoso- lenia echinata Kirk (1894: 177) from New Zealand. All its spicules are noticeably larger than those of nautilia and it is remarkable that in it the quadriradiates considerably exceed the triradiates in abundance. Furthermore, it is not recorded by Kirk that there were special coronal oxeas, although this may have been the case and they were overlooked by the author. Scypha raphanus (Schmidt, 1862) de Lauben- fels, 193 6a The species thus tentatively identified was found growing on the pilings at Friday Harbor. 192 Porifera of Friday Harbor — DE LAUBENFELS It is in the form of great pendants tapering at each end, and cylindrical in the central region. Common sizes range from 3 X 30 mm. to 8 X 50 mm. The color in life was dirty white but was considerably obscured by accumulations of foreign material on the surface. The consistency is somewhat elastic but in general very fragile. The surface is minutely hispid, minutely cavern- ous to the naked eye. The critical difference be- tween this genus (frequently known as Sycon) and Grantia is in the presence or absence of a special dermis, which is present in Grantia, The structure of the chambers is typically sycon, with elongate thimble shaped chambers frequently as large as 180' X 700/a, but in some portions of the surface shallow and cuplike in shape, in this case being about 130 /a in diameter and only 100/a in depth. The oscules are about 2 or 3 mm. in diameter and to the naked eye have smooth rims without conspicuous coronal fringes. Microscopical study shows that coronal spicules are present; they are about 6 X 800/a. The spicules of the chamber layer include smaller oxeas about 5 X 400/a and abundant triaxons with rays 5 X 75/a to 6 X 105/a. Among them occur a very few tetraxons with cladomes of about the same size as the triaxons, but with rhabds considerably shorter. There are raphide- like spicules perhaps to be classified as micro- scleres. They are only 0.75/a in diameter and so frequently broken because of their delicacy that maximum lengths cannot satisfactorily be as- signed. Lambe (1893: 38) described a Sycon (that is to say, a Scypba ) with specific name com - pactum , from this vicinity. Its dermal oxeas are very much smaller than those of the specimen here described, and the triacts of the chamber layer have shorter rays of the same thickness, but much more sagittal in shape. It possesses a conspicuous subgastral category of triradiates, and in its gastral quadriradiates the rhabds, in- stead of being shorter than the clads, are longer. Its shape, moreover, is thicker in proportion to the length than is true to the Friday Harbor sponge. There are two species of Sycon , that is to say Scypba , described from California. S. coronatum was first placed in the genus Spongia by Ellis and Solander (1786: 190) . It does not have any 193 microxeas. S. coactum was described by Urban (1905: 55) as of the genus Sycandra. It does have microxeas, but does not have typical tufts to the flagellate chambers, that is to say, spicules projecting from the surface, as is true of the Puget Sound sponge. Furthermore, all the spic- ules of coactum are very much larger than those of raphana. S. raphana was first described by Schmidt ( 1862: 14) from the Adriatic. There are many minute points of difference between it and the American sponge, but at least pro tern this identification may be made. DEMOSPCNGIAE Haliclona per mollis ( Bower bank, 1866) de Laubenfels, 1936* The sponge thus identified was found grow- ing abundantly in a channel at a beach near Argyle on San Juan Island. This was so placed that a strong current rushed past it each time the tide changed. The shape is encrusting with very conspicuous oscular projections. The color in life was a beautiful purple. The consistency is somewhat elastic but rather fragile. The sur- face, aside from the evident pores, is optically smooth. The oscular projections referred to are about 7 to 9 mm. high, each terminating in a round oscule about 2 mm. in diameter. The total projection, however, is much larger than this, frequently reaching a thickness of 4 or 5 mm. It is very unusual to find such conspicuous oscu- lar tubes on a sponge placed in a strong current, although such are very common in sponges which grow in calm water. The pores are about 100/a in diameter and 200/a or 300/a apart, and as mentioned above, they show very plainly. The internal structure is an isodictyal or "renierid” reticulation of spicules united chiefly at their apices. These are of one sort only, oxeas approx- imately 7 X 100/a. This species is clearly conspecific with that which Lambe (1893: 2 6) recorded from British Columbia and identified as Reniera cinerea , a name which unfortunately can no longer be em- ployed, as was shown by de Laubenfels ( 1936a: 39) . A lavender species of Haliclona of this gen- eral type is found in many places in the world. They are separated from each other only by such 194 PACIFIC SCIENCE, Vol. XV, April 1961 items as minute differences in average length of spicule, and it may well be that they are really conspecific, although this is far from certain. From the majority of them the Puget Sound species differs in having considerably shorter spicules. Those of Europe and California, for example, are 140/x long, instead of only 100 ft. Xestospongia vanilla (de Laubenfels, 1930) 1932 The specimen thus identified was found grow- ing intertidally on Brown Island, July 1, 1931. It is a thinly encrusting specimen, at most 2 or 3 mm. thick, and spreading indefinitely later- ally. The color is a pale drab, and the consistency is almost stony hard. The surface is smooth and even except for the minute pores and an occa- sional oscule well under 1 mm. in diameter. The endosomal structure is very dense, the spicules being crowded close together, but permeated by canals which are arranged so frequently at right angles to each other that their pattern is that of a reticulation. The spicules are of one sort only, hastately pointed oxeas about 11 X 137/x. A few which are much smaller are possibly to be regarded as being not yet fully developed. This species was originally described by de Laubenfels (193$: 28) as Haliclona vanilla and was transferred by him (1932: 116) to the new genus Xestospongia with a fairly complete dis- cussion of the species and genus. The Puget Sound specimen is entirely typical, differing in no important respects from the common Califor- nia species. Sigmadocia edaphus (de Laubenfels, 1930) 1936^ Fig. 2 a The sponge so identified was collected on July 3, 1931, being dredged from a depth of 15 m. in Peavine Pass. Large masses aggregating sev- eral handfuls, amorphous in shape, were taken at this time. The color in life was pale drab, almost white, and the consistency is stony hard. The surface is comparatively smooth; micro- scopically it is seen to be abundantly provided with pores only a little more than 100/x in diam- eter. The oscules average about 10 mm. apart and are nearly 1 mm. in diameter. The interior is rather dense, but is somewhat breadlike in structure, without any conspicuous reticulation of canals. They are united in a somewhat iso- dictyal fashion, that is to say, connected to each other at their apices in such a way as to make triangular or polygonal meshes. Those imme- diately at the surface placed horizontally make a network of even finer mesh than that of the endosome. The megascleres of this sponge are of one type only, oxeas approximately 22 X 285 ix. There is also one type of microsclere pres- ent, a sigma varying from 40/x to 66ft in length. The Puget Sound specimen is in very com- plete agreement with the specimen described as Gellius edaphus by de Laubenfels (1930- 28) and discussed by him in more detail and com- pared to related species (1932: 111). The spe- cies was transferred to Sigmadocia by de Lau- benfels ( 1936^: 69). Lissodendoryx frma (Lambe, 1894) new transfer Fig. 1 ,A1,A2, A3 The sponge thus identified was collected on July 3, 1931. It was dredged from a depth of 75 m. near Turn Island. It is a compactly massive sponge, only 4 or 5 cm. in greatest diameter and in life was a rich orange color. The consistency is firm, somewhat elastic. The surface is slightly tuberculate, but in general might be described as smooth, a very evident special dermal reticulation being fine grained. The pores are minute and the oscules are represented only dubiously by a few open- ings much less than 1 mm. in diameter which possibly were mere accidental ruptures in the surface. The interior is dense, provided with only a small amount of open space in the form of canals or gross chambers. The special dermal tornotes are hastately pointed tylostrogyles 7 X 343ft. The principal skeleton is made up of a more or less confused dense mass of smooth styles, exceptionally large for this genus. They are 36 X 440 ft. The microscleres include arcuate isochelas of one sort, length 4 3 ft, and sigmas of one sort, length 30 to 3 2 ft. This species was first described as Myxilla firrna by Lambe (1894: 122). It may appropri- ately be compared to Lissodendoryx kyma de Laubenfels, 1930 (p. 27). This California sponge lacks the sigmas, and all of its spicules are notably smaller than those of frma. It should be commented, however, that kyma and frma are remarkable in the genus for the exceptionally Porifera of Friday Harbor — DE Laubenfels large size of the spicules; they are doubtless re- lated forms. Lissodendoryx noxiosa de Laubenfels, 1930 Fig. 1, Bl The specimen thus identified was dredged on July 3, 1931, at a depth of 15 m., in Peavine Pass; masses aggregating several handfuls came up in the dredge. The shape is amorphous and the color in life a dull yellowish drab. The characteristic un- pleasant odor was very much in evidence in life, and seemed to be identical with the odor of the California specimens to the best of the recol- lection of the author. The consistency is softly spongy, fragile. The surface is irregularly tu- berculate, but otherwise is optically smooth be- cause the dermal skeleton is of very fine mesh. Rather numerous oscules are scattered about, averaging somewhat under 1 cm. apart, and 0.5 to 1.5 mm. in diameter. The endosome is very much like the struc- ture of a crumb of bread, the spicules being ar- ranged primarily in bunches, but otherwise in a somewhat isodictyal reticulation in the masses surrounding the numerous gross chambers. The special dermal spicules may be described in gen- eral as being tylotes, but very frequently there is a rounded promulgation at each end, as though they were strongyles with a tylote enlargement near but not at each end. Others of them have one end larger than the other and the smaller end somewhat spined. The variety of shapes is quite remarkable. The endosomal spicules are smooth styles, with here and there what may be interpreted as an incipient spine. Only arcuate chelas 28/x were in evidence as microscleres. This species was described by de Laubenfels (1930: 27) from California, where it is very abundant, and yet it is doubtful if any Califor- nia specimen has attained the very large size of the Puget Sound one. Lambe (1894: 121) had a Lissodendoryx which he identified as Myxilla barentsi Vosmaer, 1885 (p. 27). This was al- most certainly the form at present under dis- cussion and not the arctic species of Vosmaer. Lambe s specimen had as its principal spicules styles with small spines on them. Whereas this is not common, it is really the case in the Pea- vine Pass specimen and various California speci- 195 mens also show this tendency. Lambe’s speci- men had sigmas, as do the California speci- mens. The latter variety has megascleres about thirty percent shorter but not thinner than those of the Puget Sound sponges. Taken by itself alone, this is a trivial difference. The gen- eral agreement, and especially the very distinc- tive odor render it possible to make the iden- tification with noxiosa very confidently. Burtonanchora lacunosa (Lambe, 1892) de Lau- benfels, 1936^ The sponge thus identified was dredged on July 3, 1931, in Peavine Pass, at a depth of 15 m. It is a small subovate mass about 2X3 cm. The color in life was fleshy pink, and the con- sistency is softly spongy to fragile. The surface, otherwise smooth, is rendered irregular by the pore areas and oscules mentioned below. There is definite evidence of the presence of a special dermal layer. The round oscules are nearly 1 mm. in diameter and are scattered here and there more than 1 cm. apart in most places, but occasionally as little as 2 mm. apart. Except within regions about 2 or 3 mm. from the os- cules, the entire surface of the sponge is dotted with pore areas each nearly 1 mm. in diameter and about 2 mm. apart, center to center. Each of these circular areas is a sieve with the pores approximately lOOg in diameter, and crowded closely together. The endosomal structure is between subiso- dictyal and confused. The special dermal spicules are inequi-ended hastately pointed tornotes, about 7 X 220g. The endosomal spicules are somewhat spiny styles 12 X 220g. The micro- scleres include sigmas 23g long, and isochelas slightly longer. The latter are here interpreted as being anchorate, but they are far from being typical anchorate chelas, verging strongly to- wards the arcuate. Lambe (1892: 70) described a sponge from the vicinity of Vancouver as Myxilla lacunosa , which is rather clearly the form under discussion but which has one difference, i. e., that the principal spicules according to Lambe were not spiny. At the same time (p. 71) Lambe records another Myxilla which he identifies as being rosacea of Lieberkiihn (1859: 521). This second specimen has the principal spicules like the Pea- 196 PACIFIC SCIENCE, Vol. XV, April 1961 as o Fig. 1. A, Lissodendoryx firma (Lambe) : 1, chelas, X 880; 2, style, X 212; 3, der- mal tornote, X 212. B, Lissodendoryx noxiosa de Laubenfels; chelas, X 880. C, lophon chelifer var. calif orniana de Laubenfels: 1, anisochela, X 212; 2, bipocilli, X 212; 3, acanthostyle, X 212. D, Mycale adhaerens (Lambe); anisochela, X 880. E, Ophlitaspongia pennata (Lambe); toxa, X 212. F, Stylissa stipitata new species; style, X 212. G, Choanites suberea var. lata (Lambe): l, microstrongyle, X 880; 2, tylostyle, X 212. (All from camera lucida drawings.) vine Pass specimen now being discussed. Its isochelas and sigmas were both, however, about double the length of those of his lacunosa and those of the Puget Sound sponge. It is very fre- quently the case in the Myxillinae that there are two categories of chelas, a larger and a smaller, and two categories of sigmas, a larger and a smaller. One or the other of these may be com- mon while the other is rare. It may be that there exists in the vicinity of Puget Sound but a sin- gle species, properly to be termed lacunosa , hav- ing a full complement that includes larger and smaller chelas, and larger and smaller sigmas. It may be that in a specimen which Lambe described as lacunosa he found only the smaller microscleres, and in the one he identified as rosacea he found only the larger. It is here pro- posed that the specimen identified by Lambe as being rosacea should be dropped in synonymy to his lacunosa. Neither of these two specimens described by the latter author possesses the pe- culiar arrangement of the surface found in the Porifera of Friday Harbor- — DE LAUBENFELS sponge from Peavine Pass, but Lambe (1894: 121) described from the vicinity of Alaska a sponge which he named Myxilla behringensis. The dermal spicules and microscleres are very like those of his lacunosa , and significantly the endosomal spicules vary from entirely smooth as in his lacunosa to spiny as in the specimen he identified as rosacea. This behringensis did have the peculiar arrangement of the pores and oscules of the specimens herein described. It seems probable that only one species is involved in all of these descriptions, and it is therefore proposed that Myxilla behringensis Lambe be dropped in synonymy to his lacunosa , which was transferred to the genus Burtonanchora by de Laubenfels (1936^: 94). Ectyodoryx parasitica (Lambe, 1893) de Lau- benfels, 1936* This species is exceedingly abundant in the vicinity of Friday Harbor, occurring apparently always on the shells of bivalve mollusks belong- ing to the genus Pecten. In fact, practically every specimen of Pecten collected in this vicinity seems to have been utilized by the sponges. Usu- ally the sponges are of this species, but occa- sionally My cole adhaerens , to be described below, is also found on the shells. Two hundred and two such sponge-covered pectens were taken for study, and of them 183, or well over 90 per cent, proved to have Ectyodoryx . Nineteen had the Mycale. It was further noticed that in each case when both the valves of the pelecypod were abundantly covered with sponge, that the mol- lusk was always a male, though not all males were so distinguished. The crust has a relatively smooth surface, al- though showing many openings such as pores and oscules. Thickness may attain to 12 or 15 mm., and lateral extent is of course limited by the size of the shell. The color in life is a dull grayish drab and the consistency is weakly spongy or fragile. The surface is abundantly covered with rather coarse pores and oscules about 1 mm. in diameter, irregularly scattered. The endosome is "crumb-of -bread” in appear- ance and consistency. The special dermal spic- ules are hastately pointed tornotes, verging slightly upon the tylote shape. They are from 4 X 130 /x to 6 X 145/a. The endosomal or princi- 197 pal spicules are styles, usually acanthose, about 11/x in diameter, but ranging in length from 190 to 270/x. Occasionally the spines are almost or completely wanting, which is very interesting in comparison to the preceding species, as is also the fact that there are two categories of isochelas present. These are anchorate, the larger being 54/a in length, while the smaller are only 14/a in length. There are also two sizes or cat- egories of sigmas, each respectively about the size of the corresponding isochelas. This species was originally described by Lambe (1893- 31) from Vancouver Island, as Myxilla parasitica . It was reviewed by de Lau- benfels ( 1936a: 84) and its correct location in the genus Ectyodoryx was shown by him. Tedania fragilis Lambe, 1894 The specimen thus identified was dredged northeast of Blakeley at a depth of between 50 and 66 m., on July 6, 1931. It is an encrusta- tion less than 1 cm. thick, and about 2 X 4 cm. in area. In life it was flesh colored, and the consistency is very soft, fragile and compres- sible. The surface is comparatively smooth, al- most glabrous. Pores and oscules can not be made out. The endosome is rather dense to the unaided eye and microscopically is seen to be packed with plumose tracts, which is an axinellid char- acteristic usually not so pronounced in the genus Tedania . The special dermal tornotes are hast- ately pointed, 4 X 200/a. As may be expected the principal skeleton consists of styles, 3 X 150/a to 7 X 400/a, but it is most astonishing to note that some of these are more or less cov- ered with small spines. The microscleres are onychaetes as is absolutely typical for the genus Tedania] these are long thin rhaphides 1 X 150/a to 3 X 200/a which are covered by char- acteristic roughenings, rather than pronounced spines. A specimen which must surely have been conspecific with the one under discussion was described as Tedania fragilis by Lambe (1894: 116). It was a more typical Tedania , not pos- sessing the unusual shape of the styles. Identi- fication of the Puget Sound sponge with Tedania is rather confidently made, however, because of the very characteristic microscleres. 198 PACIFIC SCIENCE, Vol. XV, April 1961 lophon chelifer calif orniana de Laubenfels, 1932 Fig. l,Cl,C2,C3 The sponge so identified was dredged north- east of Blakeley on July 6, 1931, at a depth of between 50 and 66 m. It is an amorphous mass about 2X3X5 cm. In life it was a dull drab color, gradually turning black in alcohol. This is characteristic of the genus lophon , and is ex- hibited by few, if any, other genera in quite the same way, i. e., some turn black suddenly upon exposure to air, etc. The consistency is softly fragile, crumbling. The surface is irregularly tu- berculate, with a smooth dermal membrane much in evidence. This is broken off in many places, leaving numerous internal cavities ex- posed. It is difficult to say whether any oscules show or not; there are what appear to be oscules, but these may be merely places where the der- mal membrane has been broken off, exposing the underlying cavities. The interior is "crumb-of-bread” with an iso- dictyal reticulation of masses of spicules placed around minute cavities. The special dermal spic- ules are tylotes 4 to 7/x in diameter and 280/x long. The principal spicules are acanthostyles, reaching a maximum size of 16 X 320 /x. The microscleres include palmate isochelas of the peculiar shape characteristic of the genus, and also bipocilli 13 /x long, which resemble triden- tate unguiferate anisochelas. Ridley and Dendy (1886: 349) described lophon chelifer and Lambe (1893: 30) re- corded a sponge from the vicinity of Vancouver as of this species. There are differences from the typical race which were pointed out by de Lau- benfels (1932: 82) in connection with speci- mens which he had from California, for which he established the subspecies calif orniana. The Puget Sound specimens correspond very closely with those from California, and one need have little hesitation in regarding them as of the same variety. It is very probable that the speci- mens described by de Laubenfels (1928: 361), as Bartonella melanokhemia are also conspecific with this variety of chelifer. Ophlitaspongia pennata (Lambe, 1894) de Lau- benfels, 1927 Fig l, El This species was found growing on Brown Island, just across from the laboratory of the University of Washington. It is an exceedingly thin crust as found at Puget Sound, only 1 or 2 mm. thick, and spreading indefinitely laterally. The color at the time of collection was dull brown. The consistency is somewhat compress- ible, rather like that of soft wood. The surface is irregular with notable grooves radiating around minute oscules, only 200/x or 300/x in diameter. The structure consists of innumerable little plumose columns extending up perpendicularly from the base and little if any connected to each other except by protoplasmic structures. The bulk of the spicules in these tracts are sub- tylostyles 16 to 20 /x in diameter and about 330 /x long. Near the surface many are found which are only 3 X 200/x; these may or may not con- stitute a separate category. Toxas about 40/x long are occasionally found among the smaller spicules near the surface. This species was described as Desmacella pen- nata by Lambe (1894: 129). It is an exceed- ingly abundant sponge along the Pacific Coast of the United States, extending far down in California, almost to the Mexican boundary. It is noteworthy for its ability to survive intertidal conditions, often growing nearer the high tide mark than is true of other species of sponge. The species was transferred to the genus Gph- litaspongia by de Laubenfels (1927: 265). For a further discussion of the species, reference may be made to his paper on the sponges of California (1932: 103). My dale adhaerens (Lambe, 1893) de Laubenfels, 1936^ Fig. 1, Dl, 2 B This species occurs abundantly in the vicinity of Friday Harbor upon bivalve mollusks of the genus Pecten. As noted above in connection with Ectyodoryx parasitica, about 10 percent (or slightly less) of the Pectens in this vicinity have Mycale as the sponge which covers the shell. The color in life is a dull grayish drab, and the consistency is softly spongy to fragile. The surface is nearly smooth, and what openings are found are susceptible to interpretation as being accidental ruptures in the dermis rather than structural oscules. Porifera of Friday Harbor — DE Laubenfels 199 Fig. 2. A, Sigmadocia edaphus de Laubenfels. cies. E, Choanites suberea var. lata (Lambe). B, Mycale adhaerens (Lambe). C, Stylissa stipitata new spe- The structure of the interior is very fibrous, the fibers themselves being plumose as seen under the miscroscope. They probably contain a small amount of spongin, but are composed chiefly of abundant rows of spicules, which are smooth styles, reaching a maximum size of about 10 X 320 /*. However many are as small as 8 X 300 /* and the smaller spicules occa- sionally show a tendency to be tylostylote. There are very numerous stout palmate anisochelas of three size ranges, 14 to 15/*, 27 to 28/* and 56 to 60/*. This largest category frequently has the spicules associated together in symmetrical groupings known as rosettes. There are also two size ranges of sigmas, the smaller having a chord length of about 20/*, and the larger of 35/*. This sponge may be readily distinguished in the field from the other Pecten -covering sponge, Ectyodoryx parasitica , by the very fibrous struc- ture of the interior of the Mycale , and the greater ease with which its ectosome may be detached from the endosome. This species was originally described as Es- perella adhaerens by Lambe (1893: 27) and the genus was later shown to be synonymous with Mycale by de Laubenfels ( 1936a: 122). Stylissa stipitata new species Fig. 1, FI, 2 C The holotype, USNM no. 22687, was col- lected near False Bay of San Juan Island. It was dredged at a depth of 20 m., only 100 m. off 200 shore. Other specimens were taken later in the same month, July, 1931, at a depth of 45 m. at Griffin Bay. This species is typically funnel-shaped with an elongate stem. The maximum diameter of the cone is at the distal end and at that place is 20 mm. across. The thickness of the wall is only 2 or 3 mm., and at the thinnest the stem is also only 2 or 3 mm. in diameter. The total length, or height, of the sponge is approxi- mately 10 cm., of which 5 or 6 cm. may be described as stem, the remainder being the hol- low funnel, shaped like an inverted cone. The color in life was light drab. The con- sistency is spongy, very flexible, tough, and not easily torn. The surface is moderately smooth to the naked eye, especially on the exterior of the sponge, where there is an obvious ectosomal specialization of very fine meshed reticulation. This is frequently cracked, but other than this, no oscules are evident. It is probable that the efferent openings are the abundant minute ones on the interior of the hollow funnel. The spicules on the interior are arranged in a rather confused manner, but in general with the points towards the surface. The adjective "plumose” might be applied to the appearance in many places. There can be little doubt that this is axinellid structure. The megascleres show typical axinellid variation in size. Only smooth styles have been found; these are 2 X 180^ to 4 X 3 66/x. The genus Stylissa was established by Hall- man (1914: 349) to have as genotype the sponge described as Stylotella flabelliformis by Hentschel (1912: 355). This East Indian spe- cies is very much like the Puget Sound one here described, except that its spicules are some- what larger, and it is cup-shaped without any stem. Another similar sponge is that for which Gray (1867: 513) established the genus Tra- gosia , a species originally described as Spongia infundibuliformis by Linne ( 1858: 1348). It is very like the one under consideration except that it is shorter of stem and is usually broader and more cup-shaped. It is tremendously more hispid as to the surface, and its spicules include oxeas as well as styles. It is typically an Euro- pean species. PACIFIC SCIENCE, Vol. XV, April 1961 Syringella amphispicula new species Fig. 2 D The holotype, USNM no. 22707, was dredged July 24, 1931 at a depth of 45 m. in Griffin Bay. It is a ramose sponge, with a very few anastomoses between the gnarled and misshapen branches. The latter are 3 to 10 mm. thick and about 6 cm. high, with obtuse, club-shaped ends. The color in life was ochre yellow, the consistency very tough and flexible. The surface is even but undulating, nearly lipostomous. There is a dense axial specialization of long- itudinally arranged megascleres, among which spongin may have been present, but not con- spicuous. This axis often has a diameter slightly more than half that of the entire branch. In the flesh around this axial core the spicules are ar- ranged perpendicular to the surface and at right angles to the axis. The spicules are exclusively tylostyles, many about 18 X 670g, but also much smaller. Numerous very small ones (say 2 X 100/x) may be immature forms, or may constitute a separate category, but are connected to the larger type by numerous intermediates. The different sizes are mixed among each other, not localized. This new species is unique in the genus Syringella for its abundance of the smaller spic- ules, that in it nearly make up a second size range. Many (but by no means all) of the species of this genus have a central hollow to the branches. Many have more spongin than does amphispicula. Cliona celata Grant, 1826 The sponge thus identified was dredged from a depth of about 5 or 10 m. on July 3, 1931, south of Turn Island. It was found growing in a mass of broken barnacle shells, making sub- circular tunnels 0.5-1. 5 mm. in diameter. From them minute papilles protrude here and there a distance of about 1 mm. in life. The color in life was lemon yellow. The consistency is soft and the spicules are arranged in a rather con- fused fashion within the tunnels. The spicules are of one sort only; tylostyles about 10 X 285/*. This sponge was originally described by Grant ( 1826: 79) from Europe, but it is a well- Porifera of Friday Harbor— DE LAUBENFELS known cosmopolitan sponge. Its wide distribu- tion may be due to the fact that it bores in the shells of mollusks which may transport it freely from place to place. It frequently disintegrates the shells completely in the course of time, and may do serious damage to- commercial oyster beds. As an older sponge it grows up and out from its earlier boring habitat to make large yellow masses, frequently the size of a man’s head. Such were not found in Puget Sound, how- ever; only a few small boring specimens being recorded. Choanites suberea var. lata (Lambe, 1892) de Laubenfels, 1936$ Fig. 1, Gl-2, 2 E The sponge thus identified is moderately com- mon in the vicinity of Friday Harbor, fre- quently occurring on hermit crab shells, which may in the course of time be completely re- placed by the sponge so that no trace of the calcareous material remains, except that the sponge preserved the shell-shape as a mold might. The total mass is frequently more than 3 cm. in diameter. The color in life was gray white and the consistency was stiff, cartilaginous or cork-like. The surface is optically very smooth; pores are not in evidence, but a few oscules 2 to 4 mm. in diameter are found, usually from only one to three per sponge. Within the sponge the spicules are densely packed, although there are distinctive gross canals 0.5 to 1.5 mm. in diameter. An even denser ectosomal region which might almost be described as a cortex extends from 0.3 to 0.7 mm. below the surface. In it practically all the spicules are placed with their points toward the surface, whereas this is frequently but not al- ways the case in the interior of the sponge. These ectosomal spicules are perhaps slightly smaller than those of the endosome. The meg- ascleres of this species are of one sort only, tylostyles 5 X 280 to 6 X' 180 /i. The micro- scleres are centrotylote microstrongyles 24/x long. This species was described from European specimens as Spongia suberea by Montagu (1818: 100). The Puget Sound species in ques- 201 tion was first described by Lambe (1892: 71) as Suberites latus. This was recorded from Cali- fornia by de Laubenfels (1932: 52) where it is reduced to a subspecies of Viculina suberea. The genus Viculina being a complete synonym of Choanites , however (de Laubenfels 1936$: 144), the genus must be designated according to the prior appellation. The ■principal or Eu- ropean variety has slightly smaller spicules, and is frequently, though not always, rosy red in color. REFERENCES Ellis, J. 1786. The Natural History of Many Curious and Uncommon Zoophytes Collected from the Various Parts of the Globe. System- atically Arranged and Described by the Late Daniel Solander. London. 206 pp., 63 pi. Grant, R. E. 1826. Notice of a new zoophyte ( Cliona celata Gr. ) from the Firth of Forth. Edinb. New Phil. J. 1: 78-81. Gray, J. E. 1867. Notes on the arrangement of sponges with the description of some new genera. Proc. Zool. Soc. Lond.: 492—558, pi 27-28. Hallman, E. F. 1914. A revision of monaxid species described as new in Lendenfeld’s "Cat- alogue of the Sponges in the Australian Mu- seum.” Proc. Linn. Soc. N.S.W. 39: part I, 263-315, pi. 15-24; part II, 327-376; part III, 398-446. Hentschel, E. 1912. Kiesel- und Horn- schwamme der Am- und Kei-Inseln. Abh. Senckenb. Ges. 34: 295-448, pi. 13-21. Kirk, H. B. 1894. Contribution to a knowledge of the New Zealand Sponges. By.... In: Trans. N. Z. Inst. 26: 175-179, pi. 22. Lambe, L. M. 1892. On some sponges from the Pacific Coast of Canada and Behring Sea. Proc. Trans. Roy. Soc. Can. 10, sec. 4: 67- 68, pi. 3-6. 1893. Sponges from the Pacific Coast of Canada. Proc. Trans. Roy. Soc. Can. 11, sec. 4: 25-43, pi. 2-4. 1894. Sponges from the western coast of North America. Proc. Trans. Roy. Soc. Can. 12, sec. 4: 113-138, pi. 2-4. 202 Laubenfels, M. W., de. 1927. The red sponges of Monterey Peninsula, California. Ann. Mag. Nat. Hist. ser. 9 , 19: 258-266. 1928. A new genus and species of sponge from Puget Sound. Publ. Puget Sd. Biol. Sta. 5: 361-363. 1930- The sponges of California. Stanf. Univ. Bull. ser. 5, 5(98): 24-29. 1932. The marine and freshwater sponges of California. Proc. U. S. Nat. Mus. No. 2927, 81: 1-140. 1936^. A discussion of the sponge fauna of the Dry Tortugas in particular and the West Indies in general, with material for a revision of the families and orders of the Porifera. Carneg. Instn. Publ. 467: 1-225. 193 6b. A comparison of the shallow water sponges near the Pacific end of the Panama Canal with those at the Caribbean end. Proc. U. S. Nat. Mus. No. 2993, 83: 441-466. PACIFIC SCIENCE, Vol. XV, April 1961 Linne, C. von. 1758. Systema Naturae. Tom. I: Animalia. Editio decima, reformata. Hob miae. Montagu, G. 1818. An essay on sponges, with descriptions of all the species that have been discovered on the coast of Great Britain. Mem. Werner Soc. 2: 67-122, pi. 3-16. Ridley, S. O., and A. Dendy. 1886. Prelimi- nary report on the Monaxonida collected by H.M.S. Challenger, Parts I, II. Ann. Mag. Nat. Hist. ser. 5, 18: 325-351, 470-493- Schmidt, O. 1862. Die Spongien des Adriati- schen Meeres. Leipzig, vii + 88 pp., 7 pi. Urban, F. 1905. Kalifornische Kalkschwamme ...Von.... In: Arch. Naturgesch. 72: 33- 76, pi 6-9. Vosmaer, G. C. J. 1885. The sponges of the Willem Barents Expedition 1880 and 1881. Bijdr. Dierk. 12: 1-47, pi. 1-5. Studies of the Biology of Polychoerus carmelensis (Turbellaria: Acoela) Kenneth B. Armitage 1 Little is known about the biology of the Acoela. Hyman has summarized both earlier (1951) and more recent studies (1959: 731 and f.). The Acoela are of particular interest because typically they lack a gut and lack protonephridia, and frequently lack eyes. Thus they carry on a number of biological activities without having the structural modifications associated with these activities. The biology of the Acoela is of fur- ther interest because of the hypothesis of Hadzi that the Acoela are the stem group of the Eume- tazoa and were derived from ciliates (de Beer, 1954; Hanson, 1958). Polychoerus carmelensis is found in the pools of the mid-tide horizon in the vicinity of Mon- terey, California (Ricketts and Calvin, 1952: 49 ) . Costello and Costello have described copu- lation ( 1938^) and egg laying (1939) in this species. These studies were conducted at the Hopkins Marine Station of Stanford University, Pacific Grove, California, where the author was study- ing marine biology as a National Science Foun- dation science-faculty fellow. I wish to express my appreciation to Dr. L. R. Blinks, director of the Hopkins Marine Station, for providing facilities, and to Dr. Donald P. Abbott for in- troducing me to P. carmelensis. REACTIONS TO SALINITY Animals were collected from the tide pools during low tide at Point Pinos and Carmel Point. They were returned to the laboratory and placed in a flat, rectangular glass dish through which sea water (s.w.) flowed. Solutions of 25, 50, 75, 100, 125, and 150 per cent s.w. were made. The dilute solutions were made by mix- ing sea water from the laboratory pipes with the appropriate amount of distilled water. The concentrated solutions were made by evaporat- ing sea water to form a 200 per cent solution 1 Assistant Professor of Zoology, University of Kan- sas, Lawrence. Manuscript received February 25, I960. and then diluting this with appropriate amounts of distilled water. The first experiment was to determine the range of tolerance to salinity. Twenty animals were placed in each of the dilutions of s.w. They were observed for activity every hour during the first 12 hr. and then were checked every 12 hr. for 5 days. Any animals surviving after 5 days were periodically checked for another 2 weeks after which time the experiments were discon- tinued. Two tests were made for activity. The dishes were shaken gently; healthy animals re- acted to this agitation by showing some move- ment in place or by locomotion. A bright light caused normal animals to locomote. Animals that did not respond to either stimulus were considered inactive until disintegration of epi- dermal cells was evident, at which time the animals were considered dead. 23 per cent s.w. All animals curled into a U-shaped position immediately upon being placed in the dish. After 1 hr. all animals were dead. The individuals showed a marked swell- ing and disintegration of the epidermal cells. 30 per cent s.w. Most animals curled into the U-shaped position within 10 min. After 1 hr., only abnormal body movements, characterized by twisting and contractions, were evident. The worms would not attach to the dish, and several extruded material, including copepods that had been engulfed, from the mouth. After 5 hr. 10 animals were transferred to normal sea water. Two hr. later half the worms were active. After 3 hr., 7 of them were active and the other 3 were uncurled. By the end of 16 hr. all animals displayed normal activity. Seven of this group were active 2 weeks later when the experiment ceased. At 54 hr. 5 of the remaining 10 worms in 50 per cent s.w. were disintegrating and the other 5 showed some slight muscular movement when stimulated. The 5 active worms were trans- ferred to normal s.w., but none of them sur- vived. 203 204 PACIFIC SCIENCE, Vol. XV, April 1961 74, 100, 125 per cent s.w. All animals in these concentrations showed normal activity for 2 weeks. After 16 days, 3 animals in 125 per cent s.w. died. 150 per cent s.w. Most animals curled up shortly after being placed in the dish. One hr. later, 3 responded to the light stimulus and 7 responded upon being touched with a glass rod. The low level of activity was maintained with- out detectable change for the next 14 hr. By the 15 th through the 24th hr. movement was limited to 4 worms who could expand and con- tract their anterior ends, but could not move their posterior ends. Most animals were in the U-shaped position, but some curled into an oval position. At 42 hr. a few showed feeble move- ments, 4 were disintegrating, and the remainder were in the U-shaped position. Twenty-four hr. later all of the animals except 3 were disinte- grating and only 1 was capable of any move- ment. The second experiment was set up to deter- mine if the worms are capable of acclimating to levels of salinity. The experiment was not designed to determine the range of acclimation to salinity, but to demonstrate that acclimation was possible. Fifty animals were placed in 75 per cent s.w. 96 hr. prior to testing. Survival times in 30, 40, 50, 60, and 70 per cent s.w. were determined for worms acclimated to 7 5 per cent s.w. and 100 per cent s.w. Ten animals were placed in each solution. 30 per cent s.w. All worms were immobile. In 2 hr., 7 of the group acclimated to 100 per cent s.w. and 3 acclimated to 75 per cent s.w. were sloughing epidermal cells. At 8 hr., all of both groups were disintegrating. 40 per cent s.w. All animals of both groups were immobile and failed to respond to normal stimuli. However, at 8 hr. 7 of the group ac- climated to 75 per cent s.w. responded to a light stimulus from a no. 2 photoflood. Six of these were still active at 12 hr. and 5 showed feeble movements at 24 hr. None of the group accli- mated to 100 per cent s.w. showed any response at any time. 50, 60, 70 per cent s.w. The number of an- imals surviving in each of these three concen- trations is plotted against time in Figure 1. The curves for the animals tested in 50 per cent s.w. FIG. 1. Number of active animals plotted against time for P. carmelensis acclimated to 75 per cent s.w. and 100 per cent s.w. and tested in 50, 60, and 70 per cent s.w. indicate that acclimation has taken place and has acted to increase survival time in the accli- mated animals. The curve also indicates that the difference in survival time between the two acclimated groups increases with increasing time in the test environment. The curves also illus- trate the variability in tolerance to salinity among individuals. This is particularly evident in the curves for the animals tested in 60 per cent s.w. and in 70 per cent s.w. After the least tolerant individuals of the worms acclimated to 100 per cent s.w. died, all the groups reached an equilibrium with essentially all the remain- ing animals surviving. DIURNAL RHYTHM The only known case of a persistent rhythm in the phylum Platyhelminthes has been reported for the acoel Convoluta roscoffensis (Harker, 1958). The worm comes to the surface of the sand at low tide and disappears into the sand when the tide returns. The rhythmic behavior is maintained in the laboratory for about 1 week in vessels of still water and is independent of day or night. P . carmelensis is found on the up- per surfaces of algae or gravel during low tide and disappears into the gravel as the tide re- turns. The similarity between the two species makes desirable a study to determine if a diurnal or tidal rhythm occurs in P. carmelensis. About 150 worms were collected near Pt. Pinos and brought into the laboratory. Ten an- imals along with a few pieces of rock were Polychoerus carmelensis — Armitage 205 placed in each of 10 plastic bowls. Two addi- tional bowls were filled with gravel and 10 worms were placed in each. All the bowls were placed in an aquarium with the sides and top covered to keep out all light. In addition, the room was also darkened. Sea water from the laboratory system was circulated around the plastic bowls as a temperature control. The num- ber of animals visible in each bowl was deter- mined approximately every 2 hr. from 0800 until 2200 hr. for 2 days and spot-checked for 2 more days. The counts were made under a dim light. During the first day there was a gradual in- crease in the number of worms that were visible. This leveled off by the morning of the 2nd day. The number of animals visible at the time of each count for the 2nd day is given in Table 1. There is no relationship between the number of animals visible and the time of day or the con- dition of the tide. Nor did the subsequent spot checks indicate any relationship. From this it is concluded that P. carmelensis does not show any diurnal or tidal rhythm of activity. Observations in the field indicated that P. carmelensis came to the surface during low tide when the water was quiescent in the tide pools and when the light was relatively dim. If one waded through a pool in which the animals were located, one could observe the animals begin a downward movement. A similar downward movement was observed when the first waves of the incoming tide reached the pool. These ob- servations indicated that water agitation was the stimulus for movement downward. The following laboratory experiment tested the field observation. The data in Table 1 show that there was not much variation in the num- ber of animals visible in the plastic bowls in the darkened aquarium. The number observed was particularly stable in the plastic bowls with gravel. Two bowls with rocks and two bowls with gravel were lighted sufficiently so that the animals could be counted. There was no reaction to the light in any of the dishes. Each of the bowls with gravel had 10 animals visible. One bowl was shaken gently; all the animals imme- diately became active and began crawling down into the gravel. After 3 min., only 2 worms were visible and both of them were crawling. All 10 animals remained visible in the control dish which was not shaken and only 1 animal was active. There were 5 worms on top of a rock in a third bowl. When this bowl was agitated, all the worms became active; 4 of them crawled under the rock within 2 min. The other animals in the bowl also became active, but could not move down as they were on the bottom of the dish. There was no activity in the fourth bowl, which also had a rock and served as an unshaken control to compare with the third bowl. From these experiments and from the field observa- tions, it was concluded that P. carmelensis is negatively geotaxic in quiet water and positively geotaxic in agitated water. TABLE 1 Total Number of Animals Visible in Plastic Bowls with Rock and with Gravel (Ten animals were placed in each bowl) TIME OF NO. VISIBLE IN NO. VISIBLE IN TIME OF TIDE OBSERVATION 8 BOWLS WITH ROCK 2 BOWLS WITH GRAVEL 0800 56 19 0810 low 1000 64 20 1200 62 20 1400 48 19 1444 high 1500 54 19 1600 62 20 2000 51 20 2105 low 2200 53 17 206 PACIFIC SCIENCE, Vol. XV, April 1961 During low tides when bright sunlight was present, Polychoerus was found under rocks and gravel, indicating that the genus might be posi- tively geotaxic under bright light. Twenty-one animals were distributed in six 25 X 75 mm. plastic vials so that there were at least 3 worms in each vial. The worms were allowed to come to rest after being added to the vials. The number of worms on the side of the vials was determined. Then the worms were stimulated for 3 min. with various intensities of light. All activity was recorded. The worms were kept in total darkness for 1 hr. between subsequent tests. The same worms were used throughout. The results are summarized in Table 2. Low light intensities had essentially no effect on the animals. Higher light intensities resulted in an increase in over-all activity with all an- imals becoming active at the highest intensity of light. Animals never crawled upward. There was an increasing percentage of animals on the sides of the vials that crawled downward with the increase in light. This experiment indicates that P. carmelensis becomes more active with increasing light intensity, and where directed movement is possible the animal crawls away from the source of light. Thus the absence of animals on the upper surface of rock and gravel during low tide and bright sunlight can be at- tributed to the reaction to light demonstrated above. REACTIONS TO LIGHT In the previous section some of the reactions of P. carmelensis to light were described. That experiment was designed to explain part of the upward and downward movement of the worms in the intertidal substrate in the absence of a diurnal rhythm. The reactions to light led to some further exploration of the behavior of the worms in relation to light. The increase in the number of animals show- ing locomotion on the bottom of the vials at higher intensities of light (Table 2) indicated a photokinetic response. Photokinesis is usually defined as a change in the rate of undirected locomotion resulting from a change in the in- tensity of light. The photokinetic response in P. carmelensis was measured in two experiments. The first experiment was designed to meas- ure the rate of crawling under varied intensities of light. Two narrow strips of plastic were fas- tened 6 mm. apart in the bottom of a petri dish with a diameter of 14 cm. The plastic strips and the bottom of the dish were covered with black friction tape. A 5 cm. course was marked off between the plastic strips. The course was illuminated from one end. A worm was dropped into the dish at one end of the course with the light turned on, and the time that elapsed until the worm reached the end of the course was determined. The behavior of the worms was highly erratic. Some of them spent considerable time in turning the head from side to side, others ceased crawling before reaching the end of the course, and some crawled directly down the course. All of the worms were photonegative at the intensities used. Because of the variability in behavior, the experiment was discontinued after a small series of determinations were made. The rates of crawling for worms that crawled directly down the course illuminated by means of 5 ft. c. and 37 ft. c. were analysed by means of an analysis of variance and the between groups variance was statistically significant. The mean rate of travel was 0.86 mm/sec at 5 ft. c. and 1.34 mm/sec at 37 ft. c. The rate of crawling increased about 55 per cent when the light in- tensity was increased about 640 per cent. Sim- ilar slight increases in the rate of crawling with large increases in the intensity of light were found for Dugesia gonocephala and Plagiosto- mum sp. (Carthy, 1958: 37). A second experiment attempted to measure photokinesis by determining the amount of ac- tivity initiated in a population of quiescent worms illuminated at various intensities of light. Five worms were placed in each of five petri dishes filled with sea water. After 1 hr., the worms were illuminated dorsally at various light intensities for 3 min. The time in seconds for a worm to respond was determined, as well as the nature of the response. Responses were of three types, ( 1 ) head raising; ( 2 ) body move- ment in which the animal might swing the an- terior end side to side several times or show other changes in body form, but remaining es- sentially in the same location in the dish; (3) locomotion, in which the animal actively crawled about the dish. An activity index was determined Polychoems carmelensis — Arm IT AGE 207 TABLE 2 Reactions of P. carmelensis to Light of Various Intensities (A total of 21 animals in 6 vials were tested for 3 min. at each intensity) LIGHT INTENSITY ft. C. NO. ON SIDES ! NO. MOVING DOWN PER CENT MOVING DOWN HEAD MOVEMENT AT SURFACE LOCOMOTION ON BOTTOM 1.7 15 1 10.0 9 1 37.0 8 3 2 130.0 11 4 36 3 5 225 8 7 87 1 12 400 5 5 100 16 by assigning an arbitrary value of 5, 10, and 20, respectively, to each of the above responses and dividing the sum of the values for individual worms by the sum of the number of seconds that elapsed until the animal responded or until 180 sec. elapsed. Animals that did not respond were included. This may be illustrated in the following equation: sum of arbitrary values of ... . , response for all worms Activity index = £ sum of number of sec. un- til response or until elapse of 180 sec. for all worms The activity index will be higher if the number of animals responding to the stimulus increases or if the nature of the response is of a higher level or if both of these occur. The results are presented in Figure 2. Activity appears to be linearly related to light intensity over the ranges of light intensity studied. It is difficult to know if this relationship is a real one, for it might be an artifact of the method used to determine ac- tivity. However, the raw data indicate such a linear relationship, so that this relationship seems reasonably accurate. It probably does not exist at higher intensities, for once an animal has responded fully it can no longer respond to an increased stimulus. On contrasting backgrounds, planarians come to rest on the darker ground (Ullyott, 1936). This reaction was tested in P. carmelensis in the following manner. Half the bottom, the sides, and the upper edge of the sides of a 41/2 -in. petri dish were painted black and the other half of the bottom was painted white. The dish was half-filled with sea water and placed in a metal trough through which sea water circulated to maintain a relatively constant temperature and illuminated from above by 15 ft. c. of light. Ten (in some experiments, 20) animals were added to the petri dish as close to the center as possible. The animals were allowed 1 hr. to come to rest, after which the animals on each back- ground were counted. The dish was rotated 90° between trials so that if the animals were re- acting to light being reflected from the wall of the room or from the sides of the dish, such a directional orientation could be detected. The position of rest of each animal for each trial was plotted on a drawing of a petri dish. There was no evidence of directional orientation in any of the experiments. Forty different individ- uals were used in the first set of experiments. All animals were kept in the dark between trials. Any animal that crawled up on the side of the dish was considered to be on a black background. The experiments extended over 5 days. Of 40 animals tested, 33 came to rest on the white background and 7 came to rest on the black background. The marked orientation to the white background (chi square = 1 6.8, p < 0.001) was unexpected in view of the results with planarians. The testing was repeated four times using the same animals. The results were similar in all cases. Since all of the animals came from an area of broken shell, most of which was white, it was postulated that the animals were acclimated to the white background. If this hypothesis were correct, then it should be pos- sible to acclimate the worms to a black back- 208 PACIFIC SCIENCE, Vol. XV, April 1961 FIG. 2. The activity index of a population of P. carmelensis plotted against light intensity. Curve fitted by eye. ground. Twenty animals were kept under con- stant light in a black plastic bowl for 96 hr. The animals were tested at 24, 72, and 96 hr. At 24 hr., 12 animals oriented to white and 8 to black, showing no significant difference (chi square = .8, p < 0.5), 72 hours, 8 oriented to white and 12 to black, again showing no significant difference. At 96 hr., 5 oriented to white and 1 5 to black, now showing a significant difference (chi square = 5.0, p < 0.05). The animals were tested again 2 days later. During the two days, they were kept on the black background but received light only from 0800 to 1800 hr. Several tests were run, with the animals being kept in darkness between tests. In the first test, 5 animals stopped on the white background and 15 on the black background. The marked orien- tation to the black background was significant (chi square = 5.0, p < 0.05). However, when the test was repeated using the same animals, two of the tests gave essentially the same re- sults, but the third showed no difference in background selection. A second series of tests for background choice was made with newly collected worms from Carmel Point. Forty worms were placed in black dishes and 40 were placed in white dishes. The worms were kept under constant light and tested after 96 hr. The testing extended over 2 days and the animals were kept in darkness during the period of testing. The animals were tested under 35 ft. c. of light. The animals kept on a white background went to the white side of the dish 27 times and to the black side 1 1 times. The selection of the white background is highly sig- nificant (chi square r=. 6.7, p = c. 0.01). The animals kept on the black background went to the white side of the dish 9 times and to the black side 29 times. The selection of the black background is highly significant (chi square = 10.5, p < 0.01). Each of these tests was re- peated once using the same animals. Results were essentially identical. The animals were kept in their respective dishes and given 48 hr. of constant light. They were then placed in the dark and subsequently tested under 70 ft. c. of light. The animals kept on a white background went to the white side of the dish 2 1 times and to the black side 16 times. There is no indica- tion that background selection occurred (chi square = 0.66, p < 0.5 ) . However, the animals kept on a black background went to the black side of the dish 33 times and to the white side 7 times. Selection of the black background was significant (chi square r= 16.0, p < 0.001). Each of these tests was repeated twice using the same animals. The worms acclimated to the black background showed about the same pat- tern, but with lowered chi-square values. The animals acclimated to the white background showed almost complete randomness in selection of background (chi square = 0.01). These is no ready explanation for the shift in background selection by the worms acclimated to the white background. The experiments con- cerning photokinesis demonstrated that the worms had a differential sensitivity to light, some reacting to a weak stimulus, others to a strong stimulus. Because of the variation in sen- sitivity to light, it seems reasonable to postulate that under the increased light intensity, photo- kinesis was stimulated more in the light-sensi- tive animals. These sensitive animals then ori- ented to the black background to reduce the amount of stimulation. The less sensitive ani- mals continued to orient to the white back- ground to which they were acclimated. That there is a threshold of sensitivity to light whereby the reaction to light stimulation is reversed is indirectly indicated. Costello and Costello (1938 b) reported that P. carmelensis may ”... be positively phototropic to moderate light intensities”; the positive phototropism was Polychoerus carmelensis — Armitage 209 evidenced by the gathering of the worms on the lighted side of the aquarium in which they were kept. Taxic reactions were not studied as such in this series of experiments, but the experi- ments on diurnal rhythm demonstrated that the animals were negatively phototaxic at high light intensities and showed no response at low in- tensities. Thus the possibility of a differential response to low and high light intensities exists, as has been found for other animals (Clarke, 1930; Baylor and Smith, 1957). Clarke (1932) found that a change of illumination must rise above a certain threshold to be effective in caus- ing a reversal of phototropic signs in Daphnia. It was mentioned previously that worms ac- climated to a black background showed a lesser degree of choice of the black background when the tests were repeated. Since the animals were kept in darkness except while being tested, it seemed possible that some of the worms were losing their acclimated condition and perhaps were moving in a more random manner. There- fore, both the animals acclimated to a white background and the animals acclimated to a black background were illuminated with 70 ft. c. for 12 hr., placed in darkness for 12 hr., and then tested. Thirty of the animals acclimated to the black background came to rest on the black background while 8 selected the white back- ground. The orientation to the black background was highly significant (chi square =: 12.6, p < 0.001). Nineteen of the animals acclimated to the white background came to rest on the white background and 19 selected the black background. Thus it was not possible to condi- tion the animals to select the white background under 70 ft. c. of light under the conditions of the experiment. However, the animals accli- mated to the black background responded al- most to the same degree as in the original test. This experiment suggests that failure to main- tain an orientation to a white background at 70 ft. c. is a result of animals more sensitive to light changing their orientation from the white background to the black background. FEEDING BEHAVIOR Five P. carmelensis were placed in a Syracuse watch glass with a dozen copepods, Tigriopus calif ornicus. One of the copepods came to rest near the left anterior end of a Polychoerus which had stopped crawling. The anterior end of the worm was raised and with a sudden whiplike movement it was brought down over the Ti- griopus. The worm assumed a cup-shaped posi- tion over the copepod. The copepod was quickly engulfed and its movements inside the body of the worm could be observed. These movements continued for 10 min. The capture of the crus- tacean by Polychoerus was similar to the man- ner of prey capture by the acoel Convoluta par- adoxa (Jennings, 1957). Dead Tigriopus were not ingested. CONCLUSIONS 1. P. carmelensis can tolerate salinity condi- tions ranging from 75 to 125 per cent s.w. in- definitely. Worms were quickly inactivated at concentrations of sea water above and below these values. Animals kept in 50 per cent s.w. for 5 hr. and transferred to 100 per cent s.w. recovered normal activity by 16 hr. after trans- fer. 2. P. carmelensis was acclimated to 75 per cent s.w. and survival time was increased at the range of salinities tested over controls acclimated to 100 per cent s.w. 3. There was no evidence of a diurnal or tidal rhythm of activity. Worms tended to be ne- gatively geotaxic in quiet water at low light intensities and positively geotaxic in agitated water or at high light intensities. 4. Photokinesis, measured as the amount of activity in a population of worms, was linearly related to light intensity over the range of light intensities used. Only slight differences were found in the rate of crawling of worms over a measured course under highly different inten- sities of light. 5. At 15 ft. c. light intensity, worms collected from tide pools with white shell and rock chose the white background when placed in a petri dish with half the bottom painted white and the other half black. Worms were acclimated for 96 hr. in dishes painted black and in dishes painted white. The black-acclimated worms chose the black background and the white-accli- mated worms chose the white background when tested in the petri dish with contrasting back- grounds of black and white. The reaction was 210 PACIFIC SCIENCE, Vol. XV, April 1961 highly significant at 15 ft. c. and 30 ft. c. How- ever, at 70 ft. c, the worms acclimated to a white background showed no preference when tested on contrasting backgrounds. The worms acclimated to the black background continued to orient to the black background when tested on contrasting backgrounds. It was postulated that the change in response at 70 ft. c. of animals acclimated to the white background was a re- sult of crossing a threshold of light sensitivity so that the more sensitive animals tended to orient to the black background while the less sensitive animals tended to orient to the white background. 6. The capture of a copepod prey is described. REFERENCES Baylor, E. R., and F. E. Smith. 1957. Diurnal migration of plankton crustaceans. In: Re- cent Advances in Invertebrate Physiology, B. T. Scheer, ed., Univ. of Oregon Publications. 304 pp. Carthy, J. D. 1958. An Introduction to the Behavior of Invertebrates. Macmillan Co., New York. 380 pp. Clarke, G. J. 1930. Change of photo tropic and geotropic signs in Daphnia induced by changes in light intensity. J. Exp. Biol. 7: 109-131. 1932. Quantitative aspects of the change of phototropic sign in Daphnia. J. Exp. Biol. 9: 180-211. Costello, H. M., and D. P. Costello. 1938^. Copulation in the acoelous turbellarian Poly- choerus carmelensis. Biol. Bull. 75: 85-98. and 1938 A. A new species of Polychoerus from the Pacific Coast. Ann. Mag. Nat. Hist. ser. 11, 1: 148-155. and 1939. Egg laying in the acoelous turbellarian Polychoerus carmelensis. Biol. Bull. 76: 80-89. DE Beer, G. R. 1954. The Evolution of the Metazoa. In: Evolution as a Process, Huxley, Hardy, and Ford, ed., George Allen and Un- win, London. 367 pp. Hanson, E. D. 1958. On the origin of the Eumetazoa. Systematic Zool. 7: 16-47. Hyman, L. H. 1951. The Invertebrates: Platy- helminthes and Rhynchoeoela. McGraw-Hill Book Co., New York. 550 pp. 1959. The Invertebrates: Smaller Coe- lomate Groups. McGraw-Hill Book Co., New York. 783 pp. Jennings, J. B. 1957. Studies on feeding, di- gestion and food storage in free-living flat- worms ( Platyhelminthes : Turbellaria ) . Biol. Bull. 112: 63-80. Ricketts, E. F., and J. Calvin. 1952. Between Pacific Tides. 3rd ed., rev. by J. W. Hedgpeth. Stanford University Press, California. 502 pp. Ullyott, P. 1936. The behaviour of Dendro- coelum lacteum , II. Responses in non-direc- tional gradients. J. Exp. Biol. 13: 265-278. Preliminary Tests of the Toxin Extracted from California Sea Hares of the Genus Aplysia 1 Lindsay R. Winkler 2 During A STUDY of the biology of California sea hares (Winkler, 1957), the large amounts of absorbed but unused substances from the sea hares diet which are to be found in the digestive gland were noted. On the assumption that if there are toxic sea weeds in the diet of the sea hare these toxins might possibly be present in the digestive gland, samples of the exudate from frozen digestive glands on hand were removed with a small pipette and were placed in test tubes. The test tubes were then placed in a boil- ing water bath for 10 min. After being centri- fuged, milliliter aliquots were injected intra- peritoneally into 20-gm. C57 black mice. The digestive glands of Helix aspersa O. F. Muller were similarly treated as controls. The animals injected with the experimental extract showed almost immediate respiratory symptoms fol- lowed by an excitement stage and death in 4 to 6 min. Since neither the controls nor ashed sam- ples reconstituted and injected produced these effects, further study seemed desirable. Certain general aspects of the resulting study are re- ported here. It may or may not be significant that Latin and Medieval writers, beginning with Pliny ( ca . A.D. 60 ) considered the sea hare to be very poi- sonous and claimed its use in poisonings during the days of Imperial Rome. An excellent sum- mary of the beliefs and superstitions pertaining to the sea hare is given by Johnston (1850). MATERIALS AND METHODS Collections in March, April, and May were made at Doheney Beach near Dana Point, Cali- fornia. Later collections were made from May 1 This investigation was supported in part by a re- search grant, RG-6445-A, from the National Institutes of Health, U. S. Public Health Service. Manuscript received February 6, I960. 2 Department of Pharmacology, College of Medical Evangelists, Loma Linda, California. through August at Lunada Bay, Palos Verdes, Los Angeles County, California. The former were small specimens between 4 and 6 in., the latter were large breeding specimens measuring to 1 ft. in length. Animals were collected at low tides, packed in wet Pelvetia fastigiata , and transported back to the laboratory where they were either immediately dissected or were re- frigerated until the next day. In either case, the animals were alive when dissected and no dif- ference in the toxic effect was noted between those dissected immediately and those stored overnight. The digestive glands were dissected out by making a longitudinal, midpedal cut with scis- sors in the ventral surface of the foot from the tail to the lip area. The animal was then turned inside out. The digestive gland was removed, including the part of the intestine embedded in the gland and the ovotestis, which is an in- tegral part of the digestive gland complex. The percentage weight of the digestive gland com- plex to the weight of the intact animals was determined in a certain number of cases. Sam- ples of the crop and/ or intestinal contents were preserved in alcohol and studied where it seemed desirable in a rough attempt to ascertain if the diet was responsible for the toxicity. Specimens of Aplysia vaccaria Winkler were also collected and dissected for extraction. The digestive glands were stored in glass containers in a deep-freeze and were not thawed until ready for use. It early became apparent that a more refined method of extraction than the water methods used initially was necessary because of the large amounts of pigments, salts, and apolar materials present in the gland. After much trial and error the following method proved the most satis- factory for large-scale crude extract preparation. Thawed digestive gland (130-150 gm.) is placed in a Waring Blendor. When the gland is thoroughly liquified, 400 ml. of acetone is 211 212 slowly added, and after a few minutes the entire mixture is quickly filtered. The residue is then rinsed with an additional 100 ml. of acetone. The acetone extract, containing the tissue "wa- ter” from the digestive gland is then placed in the flask of a rotary evaporator, partial vacuum being supplied by an aspirator. As the percent- age of acetone is reduced by evaporation, the apolar fraction is thrown out of solution and deposited on the flask wall. When the remain- ing "water” has been evaporated to about 50 cc., it is placed in a refrigerator overnight, after which it is refiltered and becomes what is re- ferred to as the crude extract. The routine bio-assay procedure for indicat- ing relative strength of extracts in the toxic principle consists of injecting 1 ml. of an ex- tract intraperitoneally into mice of similar weight and observing the death time to the last heartbeat audible with a stethoscope. The death time serves as a rough indication of the relative toxicity. Initial LD 50 values were obtained by con- ventional methods using three groups of mice to establish three points on a graph plotting percentage mortality against dose given. The LD 50 values were then read from the graph. Subsequent LD 50 approximations and those for the chick were determined using a minimum number of animals by the "up and down” method proposed by Dixon (1959)- Since the LD 1-99 range is very narrow, this latter method gave adequate results for the present purposes. EXPERIMENTAL RESULTS The average percentage weight of the diges- tive gland/ovotestis complex of A. calif ornica based on 15 specimens was 10.8 per cent, and the range was from 8.4 to 14.0 per cent. Vary- ing amounts of sand were found in the intes- tines of this species, which influenced the ac- curacy of the weights. The accuracy was higher for A. vac curia, however, since no sand was found in the intestines. The average percentage based on 5 specimens of the latter species was 19.4 per cent, ranging from 18.0 to 20.3 per cent. The diet varied between the two collecting sites. At Doheney Beach the diet followed the predominant flora consisting of several coralline PACIFIC SCIENCE, Vol. XV, April 1961 algae and Hypnea calif ornica. A considerably wider variety of seaweeds was noted in the Lunada Bay collections. The Lunada Bay sea hares appeared to be more toxic but this may have been the result of more maturity or merely a reflection of variation in the extraction ef- ficiency. No correlation is possible at the pres- ent state of the research. Aplysia vaccaria , a sea hare rather distantly related within the genus, possesses a similar toxin in its digestive gland. The LD 50 was determined for 23-gm. mice and is expressed in grams of digestive gland tissue. Two different batches of raw material were used to obtain two somewhat removed values. Each batch represents the material ob- tained from 5 to 12 sea hares, depending on the animals’ size. One batch collected May 4, 1959, had an LD 50 value of 0.65 gm. tissue for a 23- gm. mouse (0.028 gm/ gm body weight). An- other collected at the same location July 28, 1959, had a value of 0.8 gm. per 23-gm. mouse (0.036 gm/ gm weight). However, the differ- ence in the two values may represent only dif- ferences in extraction efficiency rather than true variation in toxin concentration. The LD 50 for 3 -day-old baby chicks was found to be only slightly less than 25 per cent more than that for mice. When mice are injected intraperitoneally with the crude toxin of somewhat more than the LD 50 dosage, there is an almost instantane- ous hyperventilation. Ears are drooped and the mouse usually sticks his nose in a corner of the cage and salivates profusely. After a varying time in which hyperventilation is evident, the mouse starts scurrying about the cage, usually leaving a trail of urine. Perhaps it then returns to its corner or begins to demonstrate occa- sional muscular twitching which may turn into uncontrolled attempts at movement suggesting a convulsion. This uncontrolled movement may develop in waves and once begun is always terminal. Ataxia and inability to right itself usually develop before or during these uncon- trolled movements. The animal passes into a completely relaxed state. The heart continues for some time at a reduced pace, gradually be- coming weaker until it can no longer be heard with a stethoscope. The toxin also killed mice Toxin of Sea Hares- — Winkler 213 when given by stomach tube at approximately 12 times the intraperitoneal LD 50 dose. Helix aspersa O. F. Muller withstood doses of the toxin sufficient to kill several mice with only temporary effects. However, the purple shore crab quickly went into a relaxed state when small quantities were injected at the po- dial interstices. Since they were presumed dead, they were not observed further. When frogs are injected with the LD 50 dose/ gm weight for mice, the dorsal appendages be- come weakened and paralyzed in approximately 5 min. This is followed quickly by relaxation ( Fig. 1 ) , complete except for the anterior limbs, which become spastic. The rectus abdominis is also tensed. The muscles controlling the eyes are the last to become paralyzed. A complete deathlike stupor follows, lasting about 15 hr. The first reaction to reappear is the movement of the nictitating membrane and retraction of the eyes. Soon thereafter, when teased in the eye region, an isolated leg twitch may occur, usually in the thigh region. As time progresses, Fig. 1. The reaction of the frog to the toxin of A. californica. Note the tensed abdominal muscles and forelegs. FIG. 2. The reaction of the chick to the toxin of A. californica. The toxin was injected subcutaneously over the breast. teasing produces an initial kick followed by complete immotility for a time. In 3 to 5 hr. more the frogs recover completely. Three-day-old baby chicks were injected with lethal doses to observe symptoms. Immediate hyperventilation occurred, followed in order by ataxia, relaxation of the wings, and a stretching of the legs (Fig. 2), which were relaxed only terminally. Hypersalivation and difficulty in swallowing were apparent. Respiratory arrest preceded a final convulsive effort before com- plete cessation of movement. Rats and guinea pigs show symptoms quite similar to those of the mouse. Kittens, on the other hand, passed through a short but violent wretching and vomiting stage. Hyperventila- tion, relaxation of the vocal cords and nic- titating membrane, and dilation of the pupils followed. Relaxation of forepaws and neck muscles preceded respiratory distress, violent tail wagging, terminal respiratory arrest, and relaxation of the bladder sphincter muscles. DISCUSSION AND CONCLUSIONS The common denominator of the lethal symp- toms observed seems to be respiratory paralysis with no other noticeable lethal effects. Frogs, though able to survive doses paralyzing their lung respiration for 15 hr. or more, succumbed sporadically to a wide range of much larger doses. This may indicate other less dominant 214 lethal effects masked by the respiratory arrest. However, respiratory arrest seems to be the limiting factor in birds and mammals, though a suggestion of other contributing effects is noted in the chick. Studies on isolated prepara- tions are now in progress and will be reported later. At the present stage of the study it is im- possible to postulate the function, if any, or the ultimate source of the toxin. The absence of any method for the animal to inject the toxin along with the high dosages required to be effective orally would seem to preclude any de- fensive use. However, both concentration of the toxin from the seaweed diet or an endocrine function may be considered as possibilities. SUMMARY 1. The digestive gland of Aplysia calif ornica and A. vaccaria contain a water- and acetone- soluble toxin. 2. Crude extracts produced muscular weak- ness and death by respiratory arrest when in- jected intraperitoneally into various laboratory animals or given orally at about 12 times the IP dose. PACIFIC SCIENCE, Vol. XV, April 1961 3. Frogs survived a respiratory arrest and complete paralysis for 13 hr. When extracts are given in much larger doses, death ensues from causes not yet determined. 4. From observation of these symptoms, it is suggested that the primary lethal effect in mam- mals and birds is respiratory arrest, though other less dominant lethal effects seem to op- erate in the frog. REFERENCES Dixon, W. J. 1959. {A later, as yet unpub- lished, simplification of his method as pub- lished by A. W. Kimball, W. T. Burnett, Jr., and David G. Doherty, 1957.} Chemical pro- tection against ionizing radiation. Rad. Res. 7: 1-12. Johnston, George. 1850. Introduction to Con- chology; or Elements of the Natural History of Molluscous Animals. J. Van Voorst, Lon- don. Pliny, Secundus. ca. a.d. 60. Historia Natu- ralis, Lib. 9, 32. Winkler, Lindsay R. 1956. The biology of California sea hares of the genus Aplysia. Dissertation, Univ. So. Calif., Los Angeles. A Contribution to the Biology of the Convict Surgeonfish of the Hawaiian Islands, Acanthurus triostegus sandvicensis 1 John E. Randall 2 The surgeonfishes (family Acanthuridae ) , which are distinctive chiefly in their possession of a knifelike spine or spines at the base of the tail, are one of the dominant groups of tropical inshore marine fishes. Over much of their vast range, including Hawaii, where they are espe- cially prominent on the reefs, they are important components of subsistence fisheries. Little is known of their biology, however. There are only scattered references to the herbivorous food hab- its of the group and general remarks on the habitat of certain species. An analysis of the generic classification of the family and taxonomic revisions of some of the genera have been published (Randall, 1955 a, c, d\ 195 6b). The largest genus, Acanthurus , contains 32 species, 4 of which occur in the Atlantic, and the rest in the Indo-Pacific. The present paper constitutes a report of a study made during 1952-55 upon the life history and the ecology of one Hawaiian subspecies of this genus, A. triostegus sandvicensis, the convict tang or convict surgeonfish. In Hawaii and else- where in Polynesia this surgeonfish is known as the manini, and hereafter it will usually be referred to by that name. The manini is the most abundant species of surgeonfish in the Hawaiian Islands and com- mercially the most important. Judging from its prevalence in museum collections, it is also com- mon elsewhere in the Indo-Pacific region. Jordan and Seale (1906: 354) wrote, "This species is the most abundant of the genus about Samoa, swarming everywhere on the reefs.” The young reside in tidepools, and are therefore more ac- cessible for observation and experimentation 1 Contribution No. 149 of the Hawaii Marine Lab- oratory and No. 301 of the Marine Laboratory, Uni- versity of Miami, Florida. Manuscript received April 4, I960. 2 The Marine Laboratory, University of Miami, Miami, Florida. than the young of other acanthurids in Hawaii which are usually found in deeper water. The species has the widest distribution of all of the surgeonfishes — East Africa to the Gulf of Cali- fornia (a single record from West Africa by Fowler, 1936, should be confirmed). Thus in- terest in its biology may be greater than that of a localized species. A. triostegus (Fig. 1) has been described un- der 10 different scientific names and placed in six different genera. Nomenclatural considera- tions and description of the species and variants, with special reference to fin-ray counts and col- oration, have been dealt with previously (Ran- dall, 1956* ). ACKNOWLEDGMENTS This research was done in partial fulfillment of the doctoral thesis requirement at the Uni- versity of Hawaii. The author wishes to express his sincere thanks to the members of his doctoral committee, headed by William A. Gosline of the Department of Zoology, for their guidance. Personnel of the Hawaii Division of Fish and Game, then under the direction of Vernon E. Brock, helped materially in the tagging pro- gram of the manini. The tagging was financed by the Federal Aid in Fish Restoration Program. The larval fish collection of the Pacific Oceanic Fishery Investigations (POFI) of the United States Fish and Wildlife Service was made freely available. H. W. Manter and Leland S. Olsen of the University of Nebraska, Paul Illg of the Uni- versity of Washington, Cadet Hand of the Uni- versity of California, J. Percy Moore of the Uni- versity of Pennsylvania, and Alan Lewis of the University of Hawaii, assisted in the identifica- tion of parasites. Most of the photographs were taken by Charles E. Cutress of the U. S. National Mu- seum. 215 216 Fig. 1. Adult manini, Acanthurus triostegus sand- vicensis, from Hawaii. Standard length, 154 mm. HABITAT In the Hawaiian Islands the manini occupies a diversity of habitats. It occurs in bays, harbors, and exposed reef areas. It abounds in tide pools and shallow water, yet is known at depths of at least 100 ft. Areas of very turbid and brackish water are usually avoided, but a few manini have been collected even in these regions. Like most reef fishes, the manini avoids ex- tensive stretches of sand or mud bottom, pre- ferring instead the nearness of holes or crevices in the reef for shelter and solid bottom for the growth of its algal food. The manini appears to be more restricted in habitat in the Gilbert and Marshall islands, pos- sibly because of the sharper definition of the ecological zones of these atolls. It occurs in these atolls both in the lagoon and outer reef areas; however it seems to be confined to relatively shallow water. On the outer reef the fish were seen only on the reef flat, the young in great abundance in the shallower sections and the adults on the seaward portions (for data from a reef-flat transect in the Gilbert Islands, see Randall, 1955 h: 181). The species was only rarely observed in the turbulent surge channel zone and was never seen on the coralliferous terrace (benched area of living coral offshore from the reef flat). As in the Gilberts and Marshalls, the presence of heavy coralline areas in the Hawaiian Islands seems to limit the manini. This is apparent from the underwater transect work of Brock ( 1954). PACIFIC SCIENCE, Vol. XV, April 1961 Areas like the northern part of Kealakekua Bay, Hawaii, where the coral growth is extremely rich for the Hawaiian region, are almost devoid of manini. Since this species is herbivorous, its dim- inution in coral areas may be associated with the small amount of substratum available for the growth of algae. Surprisingly, other surgeon- fishes such as Acanthurus leucopareius , A. achil- les , Zebrasoma flavescens , and Ctenochaetus strigosus , usually less abundant than manini, are the most common fishes in this part of the bay. With the exception of C. strigosus which is a detritus feeder (Randall, 1955^) , these species are more inclined to crop close to the substratum than the manini, judging from the greater amount of inorganic debris in their stomach contents, and they may therefore be superior competitors under such conditions. The eggs and larvae of the manini are pelagic. This has been ascertained by a study of develop- ment following artificial fertilization of the eggs, by the capture of larvae in plankton nets, and by the taking of the late postlarval or acronurus stage at night lights offshore. A total of 11 acanthurid larvae (at least 3 of which are manini), 4.2-87 mm. in total length, were found in the plankton collections from Hawaiian waters of Cruises 4 (May 1950) and 6 (August 1950) of the "Hugh M. Smith,” a POFI research vessel. These fish were taken at eight stations ranging from about 10 to 140 mi. from the nearest island (for more exact locations, see King and Hida, 1954: fig. 1, sta- tions 1A, 10, 13, 15, 17, and 23 of Cruise 4 and stations 7 and 14 of Cruise 6). They were captured in horizontal tows made with fine- meshed plankton nets (1 m.) which sampled three different levels simultaneously. For both cruises combined, a total of 58 hauls were made at the surface, 24 at a depth of 50 m., and 112 at depths from 100 to 300 m. Six of the larvae were taken at the surface, 5 at 50 m., and none at greater depths. The failure to find more larvae in these collections is probably due to the dis- tance from land of most of the stations. Only 6 of the 29 stations of each cruise were within 25 mi. of any of the Hawaiian Islands and only 1 within 10 mi. The absence of larval Acanthurus longer than 8.7 mm. in the collections is prob- ably due to the ability of larger larvae to elude Acanthurus triostegus sandvicensis — Randall 217 the plankton net. The lack of specimens shorter than 4.2 mm. may be the result of a paucity of small larvae so far offshore. That acanthurid larvae, in general, are more abundant close to land has been shown by Reintjes and King ( 1953). These authors found a total of 1,067 acanthurid larvae in the stomach of 184 of 1,097 yellowfin tuna ( Thunnus ma- cropterus) sampled from the region of the Line and Phoenix islands. They stated that the acan- thurids were common in the stomachs of near- shore yellowfin, along with balistids and caran- gids. Representatives of the Bramidae, Exocoe- tidae, and Gempylidae predominated in the stomachs of offshore yellowfin. King kindly allowed me to examine the stom- ach content work sheets in order to compare the occurrence of larval acanthurids in the stomachs of tuna caught at the surface by pole-and-line fishing and by trolling with those caught at depths of about 30 to 160 mi. by long-lining. The comparison revealed more acanthurid larvae in the stomachs of surface-caught fish. The acronurus larval form of the manini, av- eraging about 26 mm. in standard length in the Hawaiian Islands, leaves the pelagic realm and enters very shallow water, often tide pools, to transform to the juvenile stage. Although ju- veniles tend to remain in shallow water, there is a progressive movement with increasing size to somewhat deeper water. Adults are not en- tirely absent from tide pools, however. Several were seen in pools 1-3 ft. deep cut off from the open sea at low tide on Moku Manu and Ma- nana (Rabbit) islands. Also, at night on Oahu, manini as large as 70 mm. were occasionally found in high tide pools. TOLERANCE TO TEMPERATURE AND SALINITY In order to determine if the manini is cap- able of surviving the extremes of temperatures and salinity which it encounters on Oahu, a sur- vey of these factors in tide pools and brackish areas known to be penetrated by the species was undertaken and compared with the tolerance of the species to these factors as determined by laboratory experiments on both juveniles and adults. Environmental Extremes The survey of environmental temperature and salinity was not made on any regular schedule. Only when conditions occurred which suggested that high and low values of temperature and salinity might be found, were measurements of these factors made. Temperatures were taken at mid-depth of the tide pools tested. Water sam- ples from which chlorinity was determined were obtained after stirring the pools. The Mohr method was used to determine chlorinity, and the results are expressed as grams of chlorine per kilogram of sea water (%o). The maximum temperature, 35.1° C., was recorded in a tide pool with basalt bottom (al- most black) at Makapuu Point on August 31, 1953, at 2:00 P.M. A reverse stratification of temperature existed in the pool at that time. A +0.2 low tide had occurred at 1:50 P.M. The maximum air temperature at Makapuu Point for August 31 was 79° F. (26.2° C.). Other fishes observed in the tide pool along with manini included aholehole ( Kuhlia sandvicen- sis ) and kupipi ( A budefduf sordidus ) . The minimum temperature, 16.2° C., was recorded in a small tide pool at Diamond Head on March 5, 1954, at 11:15 P.M. A -0.2 low tide had occurred at Honolulu at 10:42 P.M. The minimum air temperature during the night was 60° F. ( 15.5° C.) . A strong northerly wind was blowing. The wind velocity for the hour prior to the temperature reading was as high as 33 knots. The extremes of environmental temperature recorded by Tester and Takata (1953: 48) for the aholehole on Oahu are 20.1° C. and 32.1° C. Like the manini, young aholehole are tide- pool residents. Although found in the highest pools of the intertidal zone, they are less in- clined than the manini to enter small pools. A more notable difference of these two species is the ability of the aholehole to live in streams with little or no salt content. The lowest chlorinity recorded from pools where manini were observed was 2.65 %c. This reading was obtained from a high tide pool at Diamond Head during a heavy rain at 9:00 A.M. on March 1, 1954. A +0.1 low tide oc- curred at 9:02 A.M. A rivulet of rain water was observed entering the pool. 218 PACIFIC SCIENCE, Vol. XV, April 1961 Another low chlorinity (3.62 %©) was ob- tained in a sample taken from the shore of the Ala Wai Canal at the Ala Moana Bridge on February 23, 1954, following a heavy rain. Al- though manini could not be seen at the time because of the turbidity of the water, they were observed in the area prior to the rain and after the water had cleared. The highest chlorinity was 21.30 %c. The water sample was taken from the pool at Ma- kapuu Point in which the 35.1° C. temperature was recorded. Experiments on Temperature Tolerance The experiments on temperature tolerance were performed on manini which were main- tained in aquaria at a near-constant temperature of 24° C. for at least 24 hr. The fish were placed singly in a gallon jar of sea water at the tempera- ture of the aquarium from which they were taken. The jar was then lowered into a water- filled copper compartment containing either a heating unit or a refrigerating unit and brought to the test temperature in 1 hr. ± 10 min. The fish were kept at this temperature for 1 hr. If death ensued before the end of this hour, the time was recorded. The duration of the test period was chosen as a rough approximation of the temporal conditions of exposure to tempera- ture extremes experienced by manini in high tide pools. Throughout the tests the jar was strongly aerated. At about 5° C. before the maximum or mini- mum temperatures were reached, the fish began to increase their rate of swimming and darted around the jar. As the lethal temperature was approached more closely, the equilibrium of the fish was affected. They swam on one side, upside down, or in small circles. The last major activ- ity was usually a rapid, spasmodic, swimming movement, often in a short spiral. The results, except those of tests of both ju- veniles and adults run at temperatures of 36.5° C. or less and 13° C. or greater (which caused no deaths) , are shown in Table 1. Although the data are not adequate to deter- mine individual variation in the region of the temperature extremes, it seems evident that a range of about 13° to 36° C. is withstood by TABLE 1 Temperature Tolerance of Acanthurus triostegus sandvicensis STANDARD LENGTH (mm.) 1 TEST TEMP. (° c.) MINUTES TEST TEMP. ENDURED 96 37.0 45 97 37.0 60 137 ' 37.5 25 96 37.6 10 86 38.0 4 97 38.0 15 101 38.2 6 103 38.3 20 26 36.8 40 27 37.2 60 28 37.5 60 34 37.6 15 29 37.8 60 28 38.0 60 29 38.0 18 28 38.1 2 103 13.3 60 95 12.2 60 140 10.1 10 122 9.0 2 25 13.6 2 27 12.8 10 27 12.5 60 28 11.6 5 27 11.0 60 32 10.8 8 this species in the Hawaiian Islands, disregard- ing the possible extension of this range by ac- climatization. Although the range of temperature which ma- nini can withstand appears to be extralimital to the extremes normally encountered by the spe- cies in the Hawaiian Islands, information sup- plied by D. W. Strasburg in a letter suggests that juvenile manini in the southern Marshall Islands are, upon occasions, killed in tide pools by heat. On August 17, 1950, a high tide pool, about 30 sq. ft. in surface area and 1 ft. deep, on the ocean side of Arno Atoll was observed at low tide to contain two small Acanthurus triostegus triostegus and one small Istihlennius edentulus. Several hours later the two manini j were dead. The blenny was still living. The temperature of the pool at mid-depth was 41° C. Acanthurus triostegus sandvicensis — Randall Experiments on Salinity Tolerance Salinity tolerance experiments on manini were carried out in a 30 gal. aquarium. All fish were retained in aquaria at least 24 hr. before use in experiments. The fish were not exposed to the test salinity suddenly. Instead the salinity was gradually changed over a period of 2 hr. either by the removal of aquarium water and replace- ment with tap water (previously allowed to stand for at least 12 hr.) or the addition of salt from evaporated sea water. Fish were held at the test salinity for a maximum of 24 hr. The long period of exposure to salinity extremes was chosen because manini entering brackish areas may be subjected to water of low salt content for at least this length of time. The results are given in Table 2. Omitted are trials run at chlorinity values greater than 1.4%o, none of which caused any deaths. No trials were run at chlorinities greater than 38.25 foo. Although more data are needed to determine with accuracy the minimum salinity which ma- nini can withstand, it seems evident that manini in tide pools can tolerate a greater range in salinity than they normally experience. Since they cannot live in water of extremely low sa- linity, they probably do not enter fresh-water habitats. PREDATORS Predation on the manini is probably most acute during the early stages of the life history, but I have no information on the identity of 219 the many pelagic animals that must feed on the eggs and small larvae. The large number of post- larval stages of acanthurids found in the stom- achs of adult yellowfin tuna has been discussed. Only a few instances of predation on juvenile manini have been encountered incidentally and are presented here briefly. Juvenile manini have been found in the stomachs of moray eels (Mu- raenidae). A small lizard fish (Synodidae) was observed to catch a juvenile manini in its jaws in shallow water in Kaneohe Bay, Oahu. The ma- nini was too large to be swallowed, and it even- tually escaped. A 50 mm. specimen of Anten- narius was placed in an aquarium with six ju- venile manini. A half hour later it had eaten one of them ( length, 3 1 mm. ) . In the Society Islands a juvenile Caranx mel- ampygus about 90 mm. long was observed from shore to capture a small Acanthurus triostegus triostegus 25-30 mm. in length. This carangid occurs in the Hawaiian Islands where it prob- ably feeds in part on manini, as may other spe- cies of Caranx. Also in the Society Islands, the young Acan- thurus triostegus triostegus have been found in the stomachs of the groupers Epinephelus merra and Cephalopholis argus and the snapper Lut- janus vaigiensis (Randall and Brock, I960). It is believed that predation (man excluded) on the juvenile stage of the manini in the Ha- waiian Islands is much more pronounced than on the adult stage. In addition to the fishes men- tioned above, holocentrids, scorpaenids, cirrhi- tids, and sphyraenids, and possibly also certain TABLE 2 Salinity Tolerance of Acanthurus triostegus sandvicensis NO. OF FISH STANDARD LENGTH (mm.) CHLORINITY (0/00) NO. SURVIVING 24 HR. maximum hr. SURVIVED 3 26-29 1.4 3 4 26-29 0.7 0 5 3 27-29 0.1 0 less than 5 1 92 1.4 1 1 89 0.7 0 8 2 86-90 0.4 0 between 3 and 15 3 27-29 34.8 2 4 26-28 35.7 2 3 28-29 37.4 2 4 26-29 38.25 0 5 1 83 38.25 0 17.5 Fig. 2. Parasites of Acanthurus triostegus sandvicensis. A, Colony of Hydrichthys sp. on the dorsal fin of a manini in its first day of transformation from the acronurus to the juvenile state. B, Hydrichthys on the post- orbital part of the head of a manini in its second day of transformation. C, Benedenia sp. from the epidermis of an adult manini; length of trematode, 1.9 mm. D, Ancyrocephalus sp. from the gills of manini; length of trematode, 0.63 mm. E, Hapladena varia Linton from the duodenum of an adult manini; length of trematode, 2.9 mm. F, Adult female Spirocamallanus monotaxis Olsen from the intestine of adult manini; insert shows anterior end of the nematode (after Olsen, 1952); red in life, the females attain a length of about 45 mm. and the males about half this size. G, Unidentified leech (possibly Johanssonia sp.) from the gills of a manini; length of leech, 2.4 mm. Ac ant hums triostegus sandvicensis — RANDALL of the larger nocturnal crabs, may feed on small manini. Once a manini reaches adult size, it probably enjoys freedom from predation by most of these fishes in Hawaii, for the inshore marine fauna is peculiar in the paucity of moderately large carnivorous fishes. Shallow-water serranids and lutjanids, many of which are well adapted to prey on reef fishes, were almost totally lack- ing in the Hawaiian Islands until the recent ef- fort to introduce some of these fishes. The larger native predaceous fishes, such as adults of Ca- ranx , Sphyraena , and sharks, are not numerous at the present time, probably because of fishing pressure. PARASITES Seventeen species of parasites were detected on and within the manini. Seven specimens of transforming manini were found to be parasitized by Hydrichthys (Fig. 2 A), probably H. mirus Fewkes. This unusual hydroid occurred on the head, body, and fins and apparently does not have a preferred attach- ment site. It was never seen on juveniles or adults. Probably it flourishes only on postlarval manini, for it appears to be abraded and re- gressing on most of the transforming fish ( Fig. 2 B). Three specimens of a monogenetic fluke ( Benedenia sp.) were discovered on the epider- mis of two adult manini. The flukes are small (less than 2 mm. in length), perfectly trans- parent in life, and were hidden in the slime of the fish. They are difficult to find and are prob- ably more common than the few specimens would indicate. Figure 2C is a photograph of a stained specimen. Jahn and Kuhn ( 1932 ) have worked out the life cycle of Benedenia melleni. The most common fluke of the manini is an unidentified species of the gyrodactylid genus Ancyrocephalus. Figure 2D is a. drawing of a living specimen, 631 micra in length, which was obtained from the gills of a 132 mm. manini. This very small species of trematode was found on the gills of most of the manini which were examined. In some of the manini the infesta- tions were heavy, an average of about one fluke per gill filament being present. It is extremely active and highly extensible and was observed to change its position on a gill filament by at- taching with the anterior end and looping the 221 posterior end over to another location where the hooks were dug into place. Siwak (1931) re- ported on the life cycle of a species of Ancyro- cephalus. Two digenetic trematodes were found in the digestive tract of the manini, Hapladena varia Linton and Haplosplanchnus ohtusus (Linton), previously known only from the surgeonfishes Acanthurus hepatus ( = chirm gus) and Acan- thurus coeruleus at Tortugas, Florida. Manter (1953: 76) included a discussion of this dis- tribution in a recent paper reviewing the zoo- geography of trematodes of marine fishes. Hapladena varia (Fig. 2 E) was only occa- sionally found in the intestine of the manini. A total of 22 specimens, usually less than 3 mm. in length, were taken from 200 manini whose digestive tracts were examined for parasites. The manini harboring the flukes ranged from 43 to 136 mm. in standard length. Haplosplanchnus ohtusus was found only in juvenile manini, although it may have been over- looked in adults. It is smaller than Hapladena varia , usually less than 1.3 mm., and not as slender. Some of the juvenile manini contained a large number of these flukes. Seventy-one were counted in one 49 mm. specimen. Most of the flukes were in the duodenum, a few extended as far back as the halfway point of the intestine, and a few were found in the pyloric caeca. Four different nematodes were found in the manini. Two could not be identified to genus. One of these was a very attenuate immature male from suprapharyngeal tissue of the fish. The other was represented only by larval speci- mens from the pyloric caeca. A third worm, which was occasionally found encysted in the mesenteries, was considered to be the immature female of an unidentified species of Contracae- cum. The fourth nematode, Spiro camallanus mon- otaxis (Fig. 2 F), was described by Olsen (1952) from the sparid fish Monotaxis grandoculis , re- portedly a mollusk-feeder. In all probability this roundworm will be found in other Hawaiian fishes. Of 200 manini (mostly adults) from Oahu examined throughout the year for internal par- asites, 76 were found which harbored S. mon- otaxis in the intestine or pyloric caeca. The av- 222 PACIFIC SCIENCE, Vol. XV, April 1961 era ge number of worms found in these 7 6 fishes was 5. The largest number of worms in a single fish was 35. No obvious variation of the degree of infestation of this nematode in the manini was observed during the year. Two different leeches were found on the ma- nini. Both are species in the family of Pisicol- idae (possibly genus Johanssonia) . One is rep- resented by a single 2.4 mm. immature specimen (stained and photographed herein as Fig. 2 G) which was taken from the gills of an adult ma- nini. Six specimens of the other leech were col- lected from the body surface of three juvenile manini, 26-31 mm. in standard length, taken in the Ala Wai Yacht Basin on April 17, 1955. These leeches varied from 2.9 to 7.5 mm. in length, were dark brown in life, and easily de- tected on the host. Six species of parasitic copepods have been taken from the body surface and pharyngeal cavity of the manini. Lepeophtheirus dissimula- tus Wilson was the species most commonly en- countered: 112 of 164 adult manini examined for parasites throughout the year were infested with from 1 to 20 of these copepods. The other parasitic copepods found were two species of Caligus , a Dentigryps sp., Peniculus minuticau- dae Shiino, and Nessipus costatus Wilson, which encysts in the fins. A comprehensive report on these and other parasitic copepods from Hawai- ian acanthurid fishes is being prepared by A. Lewis. A mutualistic association was observed be- tween the manini (and other reef fishes) and a small colorful wrasse, Labroides phthirophagus Randall. In the clear water off Manana Island two adult manini were seen being pecked over the head and body by this wrasse. The manini remained motionless in the water except for slight undulation of the fins. Since the stomach contents of several of the wrasses consisted pri- marily of calagoid copepods, it is believed that the fish were removing crustacean ectoparasites from the manini. On another occasion a manini was observed to swim directly to the "domain” of two of these labrids. As the manini was ap- proached by one of them the surgeonfish was speared and brought ashore. Two adults of Le- peophtheirus dissimulatus were visible on its body. DISEASES The only data on disease were obtained from a series of unsuccessful attempts to maintain manini in an aquarium. Four adult fish were placed in a 23 gal. stainless steel aquarium which had previously been used to keep adult and ju- venile manini for periods of several weeks. The water was aerated and filtered. Two days later two of the fish died, and on the following day the remaining fish were dead. Several hours be- fore their death the manini swam in a dis- oriented manner. The fins were badly eroded. The head and body were excessively slimy, and large scattered sores were present. The fish were examined for parasites and found to have no more than the average complement. The tank was cleaned (but not sterilized) and two more adult fish added. They died in a similar fashion in 48 hr. The tank was again cleaned and a 57 mm. juvenile manini placed in it. This fish died just 48 hr. later. In spite of the filtration, the tank became cloudy on the second day after cleaning. A milliliter of the water was taken, successive di- lutions made, and the higher dilutions cultured with a sea water and agar mixture (sterile aged sea water, agar, and a small amount of peptone and ferric phosphate). The number of resulting bacterial colonies were counted; the bacterial count of the original milliliter of aquarium wa- ter was computed at 212,000,000. Before the fish were placed in the aquarium the number of bacteria were probably less than 1,000,000 per ml. Several smears were taken from eroded areas of dying manini. The dominant organism was a short, plump, gram-negative rod. It is not known whether this is the etiological agent of the disease or merely a secondary bacterium. FOOD HABITS Food of Larval Manini Only three specimens of larval manini (6.6- 7 mm. in total length) were available for an analysis of gut contents. Two of these fish were taken in a plankton net towed between 5:46 and 6:48 A.M. One was empty and the stomach of the other contained two appendicularian tunicates and a larval polychaete. The third fish Acanthurus triostegus sandvicensis — Randall 223 was captured between 3:20 and 4:25 A.M. and was empty. The digestive tracts of 57 manini in the acronurus stage (see section on transformation), which were obtained at night light stations in the Hawaiian Islands, Line Islands, and Phoenix Islands, were examined in the same way. Forty- five were completely empty. Four contained one to seven tiny fish scales; one of these also con- tained a shrimp chela. A few crustacean ap- pendages (mostly from shrimp) were found in three other fish, one of which had also eaten a barnacle metanauplius. The intestines of two fish had a small amount of unidentified chitinous remains. The crustacean appendages, metanaup- lius and fish scales were not free in the gut lumen but were caught between longitudinal folds of the gut. It seems evident that the acro- nuri do not feed at night. Four acronuri of Acanthurus were obtained from the stomachs of skipjack ( Katsuivonus pelamis) caught during the day in Hawaiian waters. Because of their being partially digested, I am not able to identify any of these acronuri to species; two, however, are too large to be manini. The stomachs and intestines of all four of these fish were filled with zooplankton. Shrimps and shrimp larvae were the principal food items. Also found were copepods, poly- chaetes, and the remains of larval fish. Food of Transforming Manini In an aquarium, manini in their first day of transformation from the acronurus to the juve- nile stage were not observed to feed. The di- gestive tracts of 30 specimens, collected in tide pools during the morning of their first day of transformation and preserved immediately, were devoid of food material. Of 24 similar speci- mens collected in the afternoon, 19 were empty, but 5 had eaten small amounts of fine filamen- tous algae and leptopel (for a discussion of leptopel see Fox, Isaacs, and Corcoran, 1951). Second-day transforming manini were seen to feed on algae in an aquarium, though not fre- quently, and preserved specimens of the same transformation age all contained small amounts of algae and leptopel. During the remaining 2 or 3 days of transformation, feeding is progres- sively heavier. Food of Juvenile and Adult Manini The gut contents of juvenile and adult ma- nini consist almost entirely of relatively fine filamentous algae. The teeth of the manini are close-set and denticulate on the margins (Ran- dall, 195 6 A fig. 2a) like other species of Acan- thurus. They are therefore well adapted for feed- ing on filaments of algae. Experiments on feed- ing in aquaria showed that the absence of coarser algae in the gut contents is due to limitation in the size and strength of the jaws and teeth. Only the smaller branches of coarser algae, like species of Hypnea, were eaten. Small juvenile manini only ate the fine end branches whereas larger fish ate progressively larger branches. Inorganic sediment was rarely found in the stomach or intestines of manini. When branches of a delicate species of the red alga Polysiphonia were mixed with fine sand and offered to aquar- ium manini, the fish were most adept at pick- ing out the filaments which projected above the sand. The stomach of the manini is thin-walled, similar to that figured and described by Breder and Clark (1947: 295, fig. 1) for Acanthurus coeruleus. It contrasts sharply with the thick- walled, gizzard-like stomachs of some species of Acanthurus. Unlike the manini, the latter species normally ingest large amounts of inorganic sedi- ment with the algae upon which they feed. In addition to avoiding the ingestion of in- organic debris, juvenile and adult manini do not seem to feed on animal material (although some species of Acanthurus , such as A. xan- thopterus , will accept animal food readily) . Even when hungry, manini declined to eat any of several kinds of animal food which were of- fered. In an aquarium the fish were very skillful in avoiding the intake of small crustaceans and small masses of tunicates or sponges which were in close proximity to the algae on which they were feeding. Occasional small animals are found •in the gut contents, however. These are prob- ably taken in accidentally. In shallow water at Coconut Island, Oahu, an adult manini was ob- served feeding on algae close to an egg mass of the damselfish, Abudefduf abdominalis (the guarding fish had been frightened away ) . When the manini reached the edge of the egg mass, it ceased to feed, swam over the mass, and began feeding on algae on the other side. Other fishes, 224 PACIFIC SCIENCE, Vol. XV, April 1961 such as labrids and chaetodonts, feed greedily on the eggs when afforded the opportunity to do so by the absence of the guarding parent. A study of the kinds of filamentous algae eaten by manini was undertaken. It was soon apparent that a great many different species of algae are eaten, and it is believed that at least a few filaments of virtually every filamentous alga in the Hawaiian area can ultimately be found in the gut of the manini if enough speci- mens are examined. Dawson, Aleem, and Halstead (1955: 21), reporting on the gastrointestinal contents of 42 Acanthurus triostegus triostegus (size of speci- mens not given) from Palmyra, Line Islands, found 40 species of benthic algae and a number of species of diatoms of several genera. The number of algal species exceeded that previously known from Palmyra. Five species were most common in the stomachs: Pterocladia sp., Spha- celaria furcigera , Lyngbya majuscula ( and/ or L. aestuarii ) , Bryopsis pennata , and Lophosiphonia sp. The authors assumed that these were either dominant algae in the grazing grounds of the fish or were especially selected as food. It is the opinion of this author that these are the dom- inant algae, for (as is discussed below) the blue-green Lyngbya majuscula and another Lyng- bya are not selected by this surgeonfish (at least not the Hawaiian subspecies ) . A simple piece of apparatus was devised to test the preference by manini in Hawaii for various kinds of common filamentous algae. Ten tongue depressors were attached 2 cm. apart to a piece of wood. The free end of each depressor was notched and wrapped with a short piece of pliable galvanized steel wire. Different kinds of algae were fastened to the depressors by wrap- ping the wire ends around them. All 10 de- pressors with the assortment of algae attached were lowered simultaneously into the center of an aquarium in which juvenile manini, 26-30 mm. in standard length, were kept. When ex- periments were repeated, the position of the algae was always changed. If an alga was not eaten during the course of the experiment, it was left in the aquarium as long as it appeared edible to see if it would ultimately be devoured. The results are given in Table 3. Many of the above algae were offered in a comparable manner to two adult manini, 123 and 128 mm. in standard length, with similar results. Some species of algae not listed in the table were offered but not eaten; these, however, were probably too coarse for the juveniles. Included were two species of Laurencia , a species of Gelidium , a Gracilaria , and a species of Dictyota. A thallus of one of the species of Laurencia was placed in the aquarium with the adult manini. The small branches were eaten, leaving the ma- jor branches denuded. Of the seven kinds of algae fine enough to be eaten but ignored in the preference experiments, one is a red (Asparagopsis taxiformis , well known for its high iodine content), one is brown (Ecto carpus breviarticulatusj , and the rest are blue-greens. These were the only blue- greens tested. Blue-green algae, some of which are the same species as those in the preference experiments, have been found in the stomachs and intestines of manini by the author ( and, as noted, by Daw- son, Aleem, and Halstead, 1955). This seems in- consistent with the results of the experiments. It was noted, however, that the blue-greens were never the exclusive food material, but were al- ways mixed with large amounts of other algae, usually reds or greens. In May, 1952, in water 10-20 ft. deep in Hanauma Bay, Oahu, the bottom was examined for the kinds of fine algae which reach a height of about 5 cm. or more. Three species this size were common: Lyngbya majuscula , Asparagop- sis taxiformis , and Plocamium sandvicense (a red alga). Since surgeonfishes are abundant in the bay and at least two of the three algae are distasteful to one surgeonfish, the success of these algae might be associated with distasteful qualities. Although there was little question from the observations that the manini requires no animal food, an experiment was designed to demon- strate that this species can subsist and grow on algae alone. A 2 3 -gallon aquarium was divided into two compartments with a piece of plexi- glass. Six manini, 26-31 mm. in standard length, were placed on each side and provided with cover in the form of several rocks grouped in the center of each compartment. The fish on one Accmthurus triostegus sandvicensis — RANDALL 225 TABLE 3 Preference Feeding Experiments on Juveniles of Acanthurus triostegus sandvicensis PREFERRED COMPLETELY EATEN SPARINGLY EATEN NEVER EATEN Polysiphonia sp. Hypnea sp. Ectocarpus indicus Lyngbya majuscula Enteromorpha sp. Hypnea sp. Sphacelaria sp. Lynghya sp. Lophosiphonia sp. Liagora sp. Hormothamnion Ceramium sp. Jania sp. enteromorphoides Centroceras sp. Rosenvingia sp. Hydrocoleum Gracilaria sp. Trichogloia sp. cantharidosmum Rhizoclonium sp. fixed diatoms Calothrix confervicola Dasya sp. Enteromorpha sp. Cladophora sp. Grateloupia sp. Herposiphonia sp. Microdictyon setchellianum ( many species ) Ectocarpus breviarticulatus Asparagopsis taxiformis side were fed with a Polysiphonia and those on the other with an Enteromorpha (these were the algae upon which manini fed most vigor- ously in the preference experiments ) . These algae are among the first macroscopic organisms to appear on the bottom of boats in harbor areas on Oahu. They are fast growing, especially the Enteromorpha which increased its length an average of 27 mm. per day on the eastern side of a boat during a period of 5 sunny days in April. If collected when the thalli first develop, they are almost devoid of animal life. A fresh mass of the Enteromorpha weigh- ing 1.75 g. contained one ciliate, one nematode, one copepod, and a few epiphytic pennate dia- toms and blue-green algal cells. Nevertheless the algae were thoroughly washed and picked over for animals before being fed to the fish. Fresh algae were added to the aquarium every day, and the previous day’s algae removed. At the end of 2 weeks the fish were measured. Those fed on Polysiphonia grew an average of 2.9 mm.; those fed on Enteromorpha grew an average of 3.1 mm. During the 2 -week interval the tem- perature in the aquarium was cool, varying from 23.0° to 23.8° C. The preference experiments indicated that neither of these algae was eaten more readily than the other. Manini often alternated between feeding on the two. After 2 weeks of feeding on one of these, both kinds were added at the same time. It was immediately apparent that the fish had been conditioned to the alga on which they had been feeding. The "new” alga was un- touched initially and only occasionally sampled an hour later. The preference was no longer obvious by the end of the day, however. Manini feed almost constantly during the day, both in an aquarium and their natural habitat. They do not feed at night (see section on be- havior ) . The volume of algae consumed is large. Four juvenile manini, 28.5-45 mm. in standard length and weighing a total of 8.3 g., were fed a known mass of Enteromorpha in an aquarium early in the morning. The alga was weighed after firm squeezing followed by blotting on paper towels. At the end of the day the remaining fresh algal material was weighed in the same manner. Fresh Enteromorpha is grassy green and can easily be distinguished from fecal alga which is brownish or blackish green and tends to remain in pellets. The manini ate 10.8 g. of this alga one day and 8.8 g. the next. The same procedure was utilized for two adult fish, 123 and 128 mm. in stand- ard length and 83.2 and 94.5 g. in weight, re- spectively. These two fish ate 27.4 g. of En- teromorpha in one day. The second day 16.3 g. of Polysiphonia was consumed. After being un- fed during the morning of the third day, they ate 25.2 g. of Enteromorpha during the remain- ing 6 hr. of the day. Although these two adult manini in the above experiment were well adapted to aquarium life (they were reared to this size as captive fish in a pond of the Hawaii Marine Laboratory at Coconut Island and were 226 PACIFIC SCIENCE, Vol. XV, April 1961 maintained in the aquarium 12 days before the experiments ) , it was observed that they did not feed as frequently as adult manini in the natural environment (whereas the juveniles did). The feeding by the adults was readily interrupted by the approach of an observer. Usually only one fish fed at a time, and since the larger one dom- inated the smaller and held it in a corner of the aquarium for much of the day, the former con- sumed the major part of the algae. Yet both fish disdained to feed at all when placed in separate aquaria. Stimulus to Feeding The following simple experiments and ob- servations were conducted in order to ascertain what sensory mechanism or mechanisms are utilized by the herbivorous manini in finding food. The juice from several grams of Enteromor- pha was squeezed into an aquarium in which two adult manini had been starved for a day. This was unfiltered and colored the water green when first dropped into the aquarium. No re- sponse was observed, even when some of the green color was seen to diffuse in the immediate vicinity of the nose and mouth of the fish. A similar experiment was performed on several juvenile manini with Poly sip honia, again with no visible response. By contrast, when an ex- tract of Polysiphonia was dropped into an aquar- ium containing a half -grown Ahudefduf abdom- inalis (omnivorous in food habits), this fish swam to the surface where the extract clouded the water red and snapped its jaws erratically in this region. Some Enteromorpha and Polysiphonia was boiled to the extent that the algae began to dis- integrate and lost most of their color. This was placed in the aquarium with fresh algae of the same kinds. Although juvenile manini fed ini- tially on the fresh algae, the boiled algae was ultimately completely eaten. Some maroon rayon fiber similar in texture and color to Polysiphonia was placed in an aquarium with juvenile manini. It was taken into the mouth temporarily by several of the fish. When the rayon was soaked in extract of the Polysiphonia , it was not taken in with greater frequency. When recently voided fecal Enteromorpha was placed at the surface of the aquarium in the same manner that fresh algae were offered, most of the manini approached it, hesitated, and swam away. One juvenile drew a small amount into its mouth but hastily ejected it. When de- fecated filaments were added together with fresh filaments of Enteromorpha, only the fresh alga was eaten. In the absence of fresh algae, fecal algae which have remained in the aquarium for several hours may be ingested. The results of offering the blue-green algae listed in Table 3 were similar to that just described for recently voided fecal algae. The blue-greens were ap- proached, but rarely taken into the mouth. The nasal organs of a 26 mm. manini were cauterized with a hot wire. Five min. after this, the fish was offered some Polysiphonia . It fed on this with the same "enthusiasm” as previ- ously. Another juvenile fish was blinded by wiping a crystal of potassium hydroxide over the eye ( other fish were blinded with silver nitrate crys- tals but these invariably died within 36 hr.). After contact with the potassium hydroxide, the surface of the eye became opaque white; this fish survived more than 3 days. It moved slowly about the bottom, occasionally swimming by chance into a mass of Polysiphonia. It never swam directly to the algae. When the algae touched the mouth it was often eaten. By the third day when the aimless movements carried the fish into a mass of algae, it fed voraciously. It is concluded from the above that vision alone is needed to locate algal food, and the olfactory sense functions in feeding only to avoid unsavory material. Chemoreceptor organs in the mouth and possibly tactile organs as well also seem to be associated with the acceptance of algae and the rejection of material which is not utilized as food. The tendency to feed more on one alga than another develops after the algae have been sampled. DIGESTION In the work on nutrition of marine animals much effort has been expended on the analysis of mode of feeding and stomach contents, whereas little has been applied to the important Acanthurus triostegus sandvicensis — Randall 227 aspect of the digestion of these animals. Yonge ( 1931 ) pointed out that information as to what an animal may collect and pass into its alimen- tary system may or may not indicate the true food of the animal. Thus, a study of digestion should logically accompany an investigation of food habits. The author is especially indebted to P. B. van Weel of the University of Hawaii for his counsel in the following research on di- gestion. Morphology of the Digestive Tract The similarity of the stomach of the manini to that of Acanthurus coeruleus as described by Breder and Clark (1947) has been mentioned. The manini stomach is elongate, and divisible into cardiac and pyloric portions. The cardiac part has prominent, longitudinal, irregularly scalloped folds on the inner surface; the pyloric part is smooth. A cross-section of the pyloric part showed the muscle tunic to be about the same width as the mucosa. There are few mul- ticellular glands. The stomach wall is less than 0.1 mm. thick except posteriorly near the pylorus where it is slightly thicker than 1 mm. Just posterior to the pylorus are five pyloric caeca. Several authors (including Yonge, 1931) have claimed that the pyloric caeca has taken over the role of the pancreas in certain teleosts, the latter organ supposedly being absent in these fishes. Dawes (1929) found that the pyloric caeca of the plaice ( Pleuronectes plates sa) had the same structure as the intestine with which it was in free communication, and Rahimullah ( 1945 ) came to the same conclusion after exam- ination of 119 species in 50 different families of fishes. The structure of the pyloric caeca of the manini is consistent with the findings of Dawes and Rahimullah. The bile duct opens into the base of one of the pyloric caeca. The pancreas of the manini was difficult to find. It was finally located in the form of two to four ( usually three ) small, round glands buff in color lying in the mesentery alongside the bile duct near its junction with the liver. The intestine of the manini is long, in keep- ing with the well-founded biological principle that herbivorous animals have lengthy intestines. The length of the alimentary tract of large adult manini is nearly six times the standard length of the fish. Most of this length is attributable to the intestine which is complexly folded within the body cavity. The length of the alimentary tract of small juvenile manini is only slightly greater than three times the standard length of the fish. The increase of the alimentary tract relative to standard length is shown in the graph of Figure 3. With increasing size the volume of a body requiring nutriment increases faster than the intestinal surface if the rate of growth of the two is equal. A disproportionate elonga- tion of the intestine is necessary to keep the area of absorptive surface adequate to the needs of the body. Hiatt (1947: 254, 257) noted a striking increase in the relative growth of the intestine of the herbivorous milkfish ( Chanos chanos) between 90 and 115 mm. standard length. The ratio of intestinal length to standard 1000 900 -g. 800 £ | 700 >. o 600 e> • | o . *5 500 • ' . • £ O' c a> - 400 • •• "5 • . 300 200 100 b£v • Ail-. 30 40 50 60 70 80 90 100 110 120 130 140 150 160 Standard length fnm) FIG. 3. Change in length of the alimentary tract of Acanthurus triostegus sandvicensis with increase in standard length. The group of points at A represents the length of the tract of 10 specimens (24-27.5 mm.) in the acronurus stage which were taken at a night light offshore. The points at B represent 10 tide-pool specimens (24—28 mm.) which have just completed transformation from the acronurus to the juvenile stage. 228 length increased from 3.5: 1 to 7.2: 1. He demon- strated a correlation between the increase in length of intestine and the substantial rise in the amount of larger algal types in the stomach con- tents and the reduced number of smaller food items like diatoms and blue-green algae. No sud- den change-over in the size of algal food which is eaten by the manini occurs at any size range once transformation has taken place, and the curve of alimentary tract length plotted against standard length is without sharp inflection. The great increase of the gut length which occurs when manini transform from the acronurus to the juvenile state is discussed in the section on transformation. The inner surface of the intestine is covered with small villi. The wall of the intestine is thin, its average thickness, not including any villi, is about 0.05 mm. The mucosa is about three to four times as broad as the muscle tunic. Time for Algae to Pass through Gut In order to find out the time which is required for algae to pass through the entire alimentary canal of the manini, two adults, 123 and 128 mm. in standard length, were fed only the red alga, Polysiphonia , for a period of 2 weeks. The tank was cleaned and the diet switched to the green alga, Enteromorpha. Two hr. and 25 min. later the fish were observed defecating the green alga. They had already deposited some on the bottom of the aquarium, so the following day the experiment was repeated by switching back to the red alga. In 2 hr. and 4 min. the first of the red alga was voided. In view of the great length of the digestive tract, this short interval of time seems extraordinary. However, when considered in the light of the nearly constant diurnal feeding and the volume of algae con- sumed, it becomes more understandable. The same procedure was repeated for small juvenile manini, ranging from 26 to 30 mm. in standard length. The Polysiphonia passed through their alimentary tracts in 1 hr. and 45 min. At night, when feeding ceases and the fish enter a state of torpor, the gut is not completely emptied in 2 hours or so. One 39 mm. manini, for example, which was caught at 1:15 A.M. PACIFIC SCIENCE, Vol. XV, April 1961 still had a small amount of algae in the stom- ach and in the last 26 mm. of the intestine. The total length of the alimentary tract of this speci- men was 177 mm. A 123 mm. manini was ob- served to defecate Enteromorpha 14 hours after it last fed on this alga. Digestion of Algae Lefevre (1940) has shown that different kinds of algae resist digestion by gastrointestinal juices of fishes in various degrees. Although sup- porting growth of manini as well or better than the Polysiphonia , the Enteromorpha appeared less digested. Filaments of this alga were found to be only slightly altered after their rapid pas- sage through the gut. The cellulose cell walls seemed unaffected, and the cell contents still in place. The only discernible change in most cells was the breaking up of the parietal grassy green chloroplast to brownish-green granular clumps. A visual comparison was made of the quantity of starch granules in the Enteromor- pha cells following staining of fresh and fecal algae, and no obvious differences were apparent. Clearly, more work is needed to elucidate the problem of algal nutrition of the manini. The possibility that the pectic sheath material of algae is digested should be investigated. Digestive Enzymes No reference was found in the literature to any assay of the digestive enzymes of a strictly herbivorous fish. From work on herbivorous an- imals other than fishes the generalization can be made that proteinase is less active in herbivorous forms than in carnivores and amylase more ac- tive. There are indications that this is true for fishes, at least for amylase. Kenyon ( 1925 ) com- pared starch digestion in the carp ( Cyprinus carpio) which, though omnivorous, eats large amounts of algae, and the carnivorous pike ( Esox lucius ) . He wrote, . . the carp, which is largely a vegetarian, possesses amylase in tre- mendous amounts in the hepatopancreas and to a less extent throughout the intestinal mucosa. The pickerel, on the contrary, . . . possesses only a negligible quantity of amylase, having little in the pancreas, esophagus, intestine, and prac- tically none in the stomach.” Vonk (1927) Accmthurus triostegus sandvicensis — Randall 229 wrote that the carp has more than 1,000 times the amount of amylase in the pancreas than that of the carnivorous pike or shark. Schlottke ( 1939) found amylase in large quantities in the carp, whereas the predaceous rainbow trout (Trutta iridea ) and perch ( Perea fluviatilis) evidently produced almost no amylase. Data comparing the activity of proteinase and lipase of omnivorous and carnivorous fishes are meager and conflicting. In view of the importance of the hydrogen ion concentration to enzyme activity, the pH of the contents of various parts of the digestive system of the manini was determined. The meas- urements of pH were made with a Beckman pH meter on six adult fish which averaged 120 mm. in standard length. The results, expressed in the ranges of pH found, are given in Table 4. The variation of pH within any one organ appears to be correlated with the degree of full- ness of the organ. The low pH values were found in the organs when they were filled with algae. Hydrochloric acid secretion in a stomach in which food is present is a probable explana- tion for the greater acidity at this time. Babkin and Bowie (1928) found a variation in pH of the duodenum of the killifish ( Fundulus heter- oclitus) similar to that shown above for the manini. These authors also noted that low values of pH were obtained when the duodenum con- tained food. They attributed this to the discharge of bile to the organ when filled with food. Extracts for the enzyme study were consist- ently prepared from the stomach (both cardiac and pyloric portions combined), pancreas, py- loric caeca, duodenum, and intestine of adult manini which were killed immediately before the removal of these organs. Because of the ex- cessive thinness of the gut wall, it was very dif- ficult to separate the mucosa from the muscle layers; therefore extracts were made of entire organs or linear parts of organs. All portions of the digestive tube to be extracted were first washed with sea water to remove food material. Tissues were ground in mortar and pestle with calcareous sand. This sand had previously been cleaned by repeated washings with water, boiling with 3 per cent potassium hydroxide and then for a short while with 2 per cent hydrochloric acid. In view of MacKay’s (1929) report that TABLE 4 pH of Organs of the Digestive System of Acant hunts triostegus sandvicensis ORGAN RANGE OF pH Stomach 6. 3-7.7 Duodenum 7. 7-9.1 Intestine 8.0-9. 1 Gall bladder 6.2-64 30 per cent alcohol yielded the most active amy- lase from the eel pout ( Zoarces anguillaris ) , this agent was also used to extract amylase in the present study. Lipase extracts were made in 40 per cent glycerol and protease extracts in 50 per cent glycerol. Extraction was carried out in a refrigerator for a period of 24 hr. Digestion by amylase and lipase proved to be rapid at room temperature (26°-27° C); thus no incubation was necessary in experiments with these enzymes. Digest tubes with protease were incubated at 36° C. Buffer solutions used in the digestion experi- ments were based on the mixtures of Clark and Lubs (Hawk and Bergeim, 1942: 24). Bacterial action was prevented by the addition of several drops of toluol to the extract and digest test tubes. The substrate for amylase experiments was 1 per cent starch solution. To each test tube con- taining 1 ml. of extract of the digestive organs 5 ml. of starch solution and 1 ml. of buffer of pH 6.8 were added. For each tissue there was a control tube identical with the experimental di- gest tube except, for the previous boiling of the extract to inactivate all enzymes. The progress of digestion was followed by re- moving small amounts of fluid from the digest tubes and testing with Lugol’s solution. The changes in the solution from deep blue-black through purple, red, yellow, and finally colorless indicated a breakdown of the starch at least to achroodextrine. Ultimately all of the tubes were colorless, thus disclosing starch digestion by the pancreas, pyloric caeca, duodenum, intestine, and stomach. The experiment was repeated three times with sections of the digestive tract vig- orously washed to minimize the possibility of enzyme from another source being adsorbed on the epithelial surface of the organ being tested. 230 PACIFIC SCIENCE, Vol. XV, April 1961 Again, there was a definite amylase reaction from each organ. The positive results seem unusual in view of the fact that most vertebrates ( except for mam- mals which secrete ptyalin in their saliva ) break down starch initially with pancreatic amylase and complete the process with intestinal maltase. These results on the manini seem less dubious, however, in the light of the finding by Kenyon of amylase throughout the whole gut of the carp ( although it was considered to lack a true stom- ach) and by the detection of stomach and duo- denal amylase in Zoarces by MacKay. Also Bab- kin and Bowie found amylase in the intestine of the killifish. They were certain it was not adsorbed pancreatic amylase, for they were un- able to observe any proteolytic action in the same extract. Pancreatic amylase of the manini is nearly 20 times more powerful per unit of tissue than the amylase from other organs, while that from the stomach was weakest (attempts were made to obtain extracts from equal amounts of glandular tissue of the organs under comparison). The pH optimum of the amylase, as deter- mined by color change with Lugol’s solution and the micro method of Linderstr0m~Lang (Linder - str0m-Lang and Hoi ter, 1933), is 6.7. The Schoorl method was utilized to test for the presence of maltase in the stomach, pyloric caeca, pancreas, duodenum, and intestine of the manini. One per cent maltose solution served as the substrate. Trials for all organs were run at pH 7.0 and 7.2 and incubated at 30° and 35° C for periods up to 12 hr., but results were consistently negative. The method of Michaelis and Rona (see van Weel, 1937: 245) was used in lipase experi- ments. Tri-n-butyrin solution was used for the substrate. Digestion occurred rapidly in all the organs tested. It was evident that the pancreas produced the most lipase and the stomach the least, although the difference was not as marked as with amylase. The pyloric caeca showed the greatest lipase activity of the remaining organs. Difference between the duodenum and the rest of the intestine was not discernible. The pH optimum determined for pyloric caeca lipase of the manini is 7.2. Detection of protein digestion was based on the formaldehyde titration of Sorensen (Jordan, 1927). The substrate was a 3 per cent colloidal solution of gelatin. In initial experiments protease was found in the pancreas, pyloric caeca, duodenum, and in- testine, but not in the stomach. In none of the organs was the proteolytic activity strong. The pH optimum of pancreatic protease is 8.4. Because of the acidic reaction in the stomach of the manini and the knowledge that protease in this organ can vary widely from individual to individual depending on the state of hunger of the animal (Schlottke, 1939), further effort was expended to localize this enzyme in the stomach. Extract of high concentration (pre- pared from trituration in 5 ml. of 50 per cent glycerol of three adult manini stomachs, two of which contained considerable algae) finally gave positive results. One ml. of this concentrated extract (thus containing the extractable enzyme from three-fifths of a stomach ) at pH 6.0 yielded acid equivalent to 0.2 ml. of 0.015 normal so- dium hydroxide after 4 hr. of incubation. A piece of the very thin covering (one cell layer thick) from one of the internal rings of an onion was peeled off and placed in a glass stender and covered with the fluid from the in- testine of an adult manini. The onion skin was examined after 24 and 48 hr. periods, but no digestion of the cellulose cell walls occurred. The experiment was repeated with fluid from the intestine of another adult specimen, again with negative results. Thus there appears to be no cellulase-secreting micro-organisms in the intestine of the manini. It is concluded that the results of the enzyme study of the manini are consistent with the gen- eralization previously made concerning the di- gestive enzymes of herbivorous animals except for the absence of cellulase. REPRODUCTION Sex Ratio No sexual dimorphism in external morphol- ogy was noted; therefore gonad examination was necessary for sex determination. The gonads lie in the ventroposterior part of the body cavity. No difficulty was experienced in distinguishing an ovary from a testis macroscopically except with immature fish. The ovaries are pinkish Acanthurus triostegus sandvicensis — Randall cream in color, smooth, slightly compressed lat- erally, and closely applied to one another in the mid-line. The testes are white, strongly com- pressed, irregularly lobular, and in contact with each other only ventrally. It was apparent from sampling the catch of trap fishermen throughout the year that approxi- mately twice as many male manini as females were taken. In the month of May, 1953, 291 trap -caught adult fish were sexed; 68 per cent of these were males. The traps in which the fish are caught are unbaited. A typical trap consists of a rectangular framework ( about 2X5X6 ft.) of steel rod covered with chicken wire. A cone of chicken wire with a narrow slitlike opening extends inward from one end to about the center. Some trap fishermen prefer to leave one or two butterfly fish or other fishes of little economic value in the trap in the belief that other fishes will be more prone to enter. If male manini are more gregarious or less cautious or more migratory than females, they would be caught in traps with greater frequency than females; therefore sex ratios ascertained from trapped fish should be viewed with caution. The sex was determined for 221 adult manini from islands in the tropical Pacific in the col- lections of the U. S. National Museum, the Ber- nice P. Bishop Museum, and the University of Hawaii. Most of these specimens were taken with rotenone, seine, and throw net. These col- lective techniques are less selective than traps and the sample would seem to be a better indi- cation of true sex ratio. Of the 221 fish, 134 (60.7 per cent) are females. A chi-square value of 10 results from testing whether 134 females could be obtained from 221 fish randomly sam- pled from a population in which the true sex ratio is 50-50. This high chi-square has a prob- ability greater than 0.01 that a sample more extreme than the above could be drawn by chance from a half male, half female popula- tion; thus the hypothesis of a 50-50 sex ratio is rejected. It is nevertheless possible that the apparent predominance of female manini is spurious. Most of the 221 specimens were taken in inshore areas. The traps from which 198 males of 291 manini were caught were set in from 30 to 90 ft. of water. Perhaps females are more abundant 231 in shallower water and the males predominate over deeper parts of the reef. More collections are needed to clarify this problem. The museum collections of manini offer some slight evidence for differential schooling by sexes. One sample of 7 manini from the outer reef at Eniwetok Atoll, Marshall Islands, is en- tirely male; another of 12 fish from the lagoon of Kwajalein Atoll in the Marshalls is wholly female. Spawning Cycle The manini in the Hawaiian Islands has a distinct spawning season. This has been ascer- tained by the examination of the gonads of adult fish and collections of young throughout the year. In more equatorial areas, on the contrary, the spawning of this surgeonfish appears to be year-round. There is also evidence both in Ha- waii and the Gilbert Islands that the pattern of the incoming young correlates with the lunar cycle. This is attributed to a lunar effect on spawning by adults. From October 31, 1952, to October 27, 1953, a total of 137 female and 156 male manini 100 mm. or more in standard length were purchased in 41 samples from a fisherman for the purpose of examination of the gonads. The fish were caught in traps set in depths of 30-90 ft. from Kewalo Basin to Koko Head, Oahu. The length of the gonad on each side of the body cavity of each specimen was measured and the average length recorded. This average measurement was divided by the standard length and the quotient multiplied by 100. The range and mean of the resulting percentages are presented in Table 5. Although the monthly samples are small, it is nevertheless apparent that the gonads of both sexes undergo a significant change in relative length during the year. They are largest in late winter and spring and smallest in late summer and early fall. The change which takes place in the ovary in late July and August is more com- plex than would be indicated from a mere short- ening. The ovary (presumably following the last spawning) is red, flattened, and has a large lu- men. Residual eggs are gradually absorbed ( eggs being absorbed appear as amorphous masses of variable size ) , and the entire organ regresses to small size and a translucent gray color. There is 232 PACIFIC SCIENCE, Vol. XV, April 1961 TABLE 5 Gonad Length as a Percentage of Standard Length by Month, 1952-53 MONTH NO. OF FEMALES GONAD LENGTH x 100 1 NO. OF MALES GONAD LENGTH \/ 1 nn STANDARD LENGTH X ioU STANDARD LENGTH Range Mean Range Mean Oct. 15 7.1-18.2 9.4 14 5.6-22.8 9.4 Nov. 2 7.7-14.4 11.1 12 8.3-22.8 15.5 Dec. 4 12.3-25.0 14.3 11 10.0-22.1 15.9 Jan. 9 8.6-25.6 13.5 10 8.4-25.0 17.9 Feb. 15 14.1-25.6 18.2 31 20.4-34.5 28.6 Mar. 19 13.5-32.2 19.3 25 12.8-35.7 28.6 Apr. 17 10.4-26.4 17.3 7 22.8-38.5 29.4 May 6 9.8-23.3 16.2 5 18.2-31.2 25.6 Jun. 9 7.0-34.5 13.5 5 15.0-25.1 20.0 Jul. 10 7.5-17.6 9.6 9 8.1-26.3 14.9 Aug. 9 5.5-11.9 7.2 8 6.5-23.8 13.2 Sept. 22 5.0-14.5 8.4 19 7.6-17.3 9.0 variation in the time when this occurs in indi- vidual fish, hence the ultimate small size is not apparent when the values of a number of fish are averaged by month. The ovary of a 145 mm. female taken on August 7, for example, was only 8 mm. in length. The gonad measurement data are more sig- nificant in indicating a definite spawning season of the manini when coupled with the results of sampling the young throughout the year. Trans- forming and small juvenile specimens were ab- sent from inshore areas during the fall and early winter months of 1952-53. Extensive observa- tion and collecting were undertaken during the winter and early fall to ascertain the time of the first and last arrivals of acronuri from the pelagic realm. The first transforming young were sighted on February 14; the last of the season straggled in on October 6. In 1954 the first young were discovered on February 9. During my absence from the state, my wife and Philip Helfrich made an effort to determine the date of the last influx of young in 1954. In October only a single small juvenile was collected; it was taken on the 9th of the month and measures 27.5 mm. in standard length. It is completely transformed; therefore it had been in the tide-pool zone for about 5-8 days (see sections on transformation and growth). The first transforming young of the 1955 season were observed on January 27. The time between the finding of the first ripe female of the season and the first incoming young and the time between the last ripe female and the last incoming young constitute estimates of the duration of larval life. The last ripe female of the 1952—53 season detected in the sampling program (see Table 6) was found on July 21. The first ripe female of the 1953-54 season was found on December 1. The last ripe females of this season were observed in the July 16 sample. The three estimates of the duration of larval life from the above data are 71, 78, and 83 days. Admittedly such estimates are subject to consid- erable error. Their average, 77 days, approxi- mates 2 l /z months. In order to obtain quantitative data on the recruitment of young to inshore areas of Oahu, a single large tide pool at Diamond Head ( Fig. 4) was chosen for repeated collections of ma- nini. This pool is nearly isolated at low tide. An overhanging ledge on the left (east) side of the pool provides excellent cover for the small fish. Pools as large as this but lacking such cover contain fewer manini. Each week at low tide the narrow outlet to the sea was occluded with a net, and all the manini in the pool were poi- soned with rotenone. A total of 157 manini less than 30 mm. in standard length were taken from the pool during the season (Fig. 5). Nineteen specimens 30 mm. or longer were collected; these are presumed to have been in the tide- pool zone for more than one week (see section on growth) and hence are probably migrants from adjacent pools. They are not included in Ac ant hums triostegus sandvicensis — Randall Fig. 4. Tide pool at low tide at Diamond Head, Oahu, from which the collections of Acanthurus trios- tegus sandvicensis of Figure 5 were made. the graph of Figure 5. The poison stations on August 13 and August 21 were interrupted by unexpected large waves, and no fish were col- lected. However, on these two dates manini were counted in the pool before rotenone was added and these numbers are indicated by the dotted 233 line on the graph. Probably there were at least twice as many small manini in the pool as are recorded, for most of the fish take cover rapidly when an observer approaches and are hidden before they can be counted. The sample of 36 fish taken on September 12, on the other hand, may be higher relative to the overall number of tide-pool young at this time than it should be. Observation of other pools at Diamond Head and along the shore of the Ala Wai Yacht Basin prior to, during, and after September 12 failed to disclose any striking influx of young. Never- theless, the number of incoming young in Sep- tember is high and this month should be in- cluded with the period May to August as in- dicating large tide-pool recruitment on Oahu. Although this sampling from a single pool is not sufficiently great to be correlated closely with the similarly inadequate samples of adult gonads, the May to September recruitment seems to cor- respond roughly to the greater degree of gonad development from February to June. In more equatorial waters the A. triostegus triostegus appears to spawn throughout the year. The 221 adult fish used for the sex ratio deter- mination were collected mostly from the Mari- Feb. March April May June July Aug. Sept. Oct. Fig. 5. Weekly collections of Axanthurus triostegus sandvicensis made in 1953 from a single tide pool at Diamond Head, Oahu. Only specimens 30 mm. or less are recorded. Times of new moon are indicated by black circles. 234 ana, Marshall, Gilbert, Line, and Phoenix is- lands — groups in which the sea surface tempera- ture exceeds 80° F. (26.7° C.) the year-round and where the annual variation in sea tempera- ture is slight (based on Hydrographic Office, 225). Examination of these fish revealed ripe gonads in every month of the year. Specimens of transforming or small juvenile manini col- lected every month of the year from these areas were found in museums. It is, therefore, puzzling that the seasonal spawning in the Hawaiian Islands seems to be associated with the colder part of the year. The mean monthly sea surface temperatures around Oahu (1945-55) vary from about 75° to 81° F. in an average year (Leipper and Anderson, 1950; Hydrographic Office, 280.) The tempera- ture of the warm part of the year is essentially the same as that of the lower latitudes where spawning is year-round. The initial enlargement of the gonads in Hawaii coincides with the time of decreasing sea surface temperature in the fall. The marked increase in tide pool recruitment in May appears to correspond to increased spawning in February-March, the time of cold- est water temperature. If temperature or some other factor such as change in length of day is the cause of inter- rupted spawning in Hawaii, then manini in a region of the South Pacific Ocean with com- parable latitude should spawn seasonally and 6 months out of phase from Hawaiian manini. Unfortunately no data are available from south- erly island groups such as the Australs or south- ernmost Tuamotus where sea temperatures closely approximate those in Hawaii. Small juvenile and transforming A. triostegus triostegus were taken or observed by the author in every month of the year (1956-57) in the Society Islands ( where temperatures range from about 77.5° to 81° F., or 25.2°-27.2° C). The manini is not as abundant in the Society Islands as in Hawaii, and the insignificant tide in the former island group made it difficult to find numerous individuals isolated in tide pools. Col- lections were insufficient to demonstrate any possible variation in reproductive activity dur- ing the year. A cyclic fluctuation in the abundance of in- coming young within the spawning season in PACIFIC SCIENCE, Vol. XV, April 1961 the Hawaiian Islands is apparent from Figure 5. The peaks of these fluctuations line up roughly with the time of new moon. This correlation is more evident from the extensive collections made from various tide pools plotted in Figure 6. The large mode of May 10 centered on a standard length of 26 mm. contains 54 trans- forming fish. The time of new moon was May 13. The large mode of April 17 is centered on a standard length of 27 mm. and contains 10 transforming manini. New moon in April oc- curred on the 13 th day. Tester and Takata (1953: 36, fig. 14) have demonstrated a similar lunar periodicity in the appearance of young aholehole ( Kuhlia sand - vicensis ) in tide pools on Oahu. If the periodicity in the influx of manini acronuri is a function of some lunar effect on the young and not on the spawning adults, then the peak fluctuations of incoming acronuri in areas of different temperature, and hence dif- ferent rates of development (see development section), should still correlate with the time of new moon. A sample of 108 small manini col- lected with rotenone by the author from tide pools at Onotoa Atoll, Gilbert Islands, on Au- gust 21, 1951 (Fig- 7), demonstrates that this is not the case. Onotoa is located in the central Pacific (1° 47' S., 175° 32' E,); the monthly mean sea surface temperatures vary only slightly from 82.5° F. (28.6° C.) throughout the year. Full moon occurred on August 16 (new moon on August 2) . The large mode of Figure 7 cen- tered on about 23 mm. standard length includes only 9 transforming specimens which average 23.5 mm. in standard length. Thus the peak in- flux of the fish comprising this mode is esti- mated at 4 or 5 days prior to August 21. This large Onotoa sample, therefore, is about 1 1 days out of phase with the lunar cycle of incoming young on Oahu (where the mean monthly sea surface temperatures vary from 75° to 76.5° F., or 23.9°-24.5° C., from January to April). In order to obtain direct evidence of lunar spawning by adult manini and to more sharply delimit the spawning season, the entire catch of two trap fishermen was examined for ripe fish on 62 days from June 19, 1953, to September 3, j 1954. The catch was usually brought to port alive. The fishermen would not permit the open- I Acanthums triostegus sandvicensis — -RANDALL 235 ing of the body cavity without purchasing the fish, but they allowed pressure to be applied to the abdomen of each manini. When running ripe males (the sperm of which were motile) were discovered in samples taken throughout the year, it was realized that the limits of the spawning season could not be defined by the development of the male gonad. As would be expected from Table 5, however, there were more ripe males during the spawning season than outside of it. Of 745 adult manini ( since trap caught, an estimated two-thirds were Do!* F»b. 14 A JO 40 Sample Size _____________ 3 50 Data April II I i.L. 40 30 Sample Size - 29 April 17 50 107 April 24 40 A, 50 47 .4 euuiLLL May 17 *J_L lji«. i. mli. ,.i Fig. 6. Tide-pool collections of Acanthums triostegus sandvicensis from Ala Wai Yacht Basin to Makapim Point, Oahu, 1953. Vertical distance between horizontal lines is proportional to number of days between samples. 236 Fig. 7. Collection of Acanthurus triostegus trios- tegus from tide pools of the outer reef flat, Onotoa Atoll, Gilbert Islands, August 21, 1951. Sample size, 108. Nine specimens 21—25 mm. in standard length are transforming from the acronurus to the juvenile stage; the rest are juveniles. Time of full moon, Aug. 16 . males ) examined from September to November, 11.8 per cent were running ripe males; 23 per cent of 2,419 adult manini (an estimated two- thirds of which were males) examined from February to June were running ripe males. No variation in degree of ripeness was apparent within the period of a single month. It should be emphasized, however, that no sharp dividing line exists between the condition of a running ripe male and one that is not quite running ripe, and these data tend to be subjective in this re- gard. Ripe females are more easily distinguished. Fertilizable eggs of female manini are about 0.7 mm. in diameter, prefectly transparent, and are released when the abdomen is only slightly squeezed. The data on ripe females in terms of the lunar month are summarized in Table 6. As may be seen in this table, the number of fish examined on the different days ranged from 9 to 252. This variability is entirely due to varia- tion in the catches. Also there is no regular sequence in time when the catches were exam- ined. Scheduled sampling was not possible largely because of the great dependence of trap fishing on hydrographic conditions. In addition to being impeded by rough seas, trap fishermen are restricted by turbid water (traps are not found by surface markers but by observing through a glass-bottom box). Although the fish- ermen ordinarily tend their traps twice a week, they may be held in port several weeks by dirty water. A successful sampling day was also con- tingent on being at Kewalo Basin when the catches were unloaded. Frequently the fishermen PACIFIC SCIENCE, Vol. XV, April 1961 came in unexpectedly early and their fish were promptly taken to market. Market fish were not sampled because catches of previous days may accumulate and the date when fish are caught is usually not available. A total of 43 running ripe females were found. Thirty-nine of these were taken from 12 days before to 2 days after the full moon and only 4 from the remaining half of the lunar month. A total of 2,552 adult manini were examined in the former period, however, and only 1,311 in the latter; therefore a correction is necessary for this discrepancy in sample size. A corrected figure of 7.8 ripe females instead of 4 is the proper number to compare with 39. Nevertheless, it still remains obvious that a sig- nificantly greater number of ripe females, and hence probably greater spawning, occurs before and shortly after full moon than before and shortly after new moon. Since at least a few acronuri of all sizes (see section on transforma- tion) enter tide pools at all times during the month and not totally in one portion thereof, some spawning might be expected throughout the month. One sample of adult manini was examined on July 29, 1955. There were 7 running ripe female fish among the 57 manini comprising the sam- ple. Full moon occurred on August 3. It is not known whether spawning by the same fish takes place every month. There is evi- dence, however, that individual fish spawn more than once a season. The long season in itself suggests repeated spawning. If large ripe females were found at the start and progressively smaller ones throughout the rest of the season, a long spawning season might be indicated with each individual spawning only once, but no obvious trend in size of the 43 ripe females was apparent. The first ripe female of the season, which was found December 1, measured 138 mm. in stand- ard length. The last (July 21) was also large, 140 mm. The shortest found were 106 mm. (March 13) and 101 mm. (July 14). More con- vincing is the failure to see any completely spent ovaries in fish during the spawning sea- son and the observation of spent and regressing ovaries in all of the females at the end of the season. The strongest evidence concerns the find- ing in ripe females of a second mode of egg size Accmtbums triostegus sandvicensis— Randall 237 TABLE 6 Results of Sampling for Ripe Female Acanthmus triostegus sandvicensis During the Spawning Season DATE days to neare Before ST FULL MOON After NO. OF FISH EXAMINED NO. OF RIPE FEMALES Jun. 19, 1953 8 40 1 21 6 9 1 Jul. 14 12 114 1 17 9 45 0 21 5 14 1 Dec. 1 11 141 1 8 12 126 1 22 2 139 0 - 29 9 96 0 Jan. 8, 1954 11 149 1 12 7 21 1 24 5 26 0 Feb. 5 12 30 0 9 8 30 0 12 5 41 1 Mar. 2 13 31 0 12 7 144 6 30 11 44 0 Apr. 2 13 87 0 6 12 198 0 9 9 121 0 13 5 111 0 16 2 193 7 20 2 132 6 23 5 76 0 27 9 151 0 May 1 13 88 0 6 11 83 1 11 6 226 3 14 3 149 2 18 1 18 0 21 4 204 1 25 8 159 0 28 11 103 2 Jun. 11 5 42 2 18 2 97 0 22 6 29 0 25 9 49 0 Jul. 2 14 27 0 9 7 28 0 16 0 0 252 4 about half the size of the ripe egg mode, coupled with the knowledge that this mode is not re- tained in the ovary through the summer months. This mode of intermediate-size eggs appears to be a persistent unit within the ovary of all mature female manini during and 1-2 months before the spawning season, suggesting that it constitutes a static mass of eggs from which modes of ripe eggs may develop periodically. It is not known whether this mass is built up constantly or at short intervals from the huge reserve of minute primordial eggs which is pres- ent in the ovary of every adult female regardless of the time of year (and is here not considered as a mode of egg size) . The procedure for the determination of egg diameters was as follows: Tiny sections were snipped from various parts of an ovary (which was previously preserved in 8 per cent formal- dehyde solution) and placed in water in a Syra- 238 PACIFIC SCIENCE, Vol, XV, April 1961 cuse watch glass; the eggs were teased from the ovarian tissue with dissecting needles; the egg diameters were then measured with an ocular micrometer at X 24. Many of the eggs were oblong, probably because of unequal pressure of adjacent eggs at the time of preservation. Instead of measuring the greatest or least diam- eter, the eggs were moved into position by mov- ing, but not turning, the watch glass on the micrometer image which remained fixed in a horizontal position. This method reduces the possibility of bias but has the disadvantage of increasing the spread of modal groups on graphs of egg diameter measurements. Graphs of the egg diameters of ovaries from 9 adult female manini are shown in Figure 8. These graphs do not show any definite sequence. Graph C, of a 15 mm. ovary from a female taken on January 2, for example, portrays the egg diameter pattern of an immature but matur- ing ovary. No graphs were made of egg diam- eters of ovaries from adult fish taken in August because the fish have only primordial eggs about 1 to 3 micrometer units (.035 to .105 mm.) in diameter except for those in which larger eggs are being resorbed. The relative size of the group of primordial eggs was estimated only for the ovary graphically illustrated in A. Except for the immature gonad of C, all of the ovaries display a mode of egg size centered on about 10 to 12 micrometer units in diameter. This prompted the examination of samples of eggs from 54 ovaries taken from female fish throughout the year (more, however, from mature females during the spawning season). Sufficient eggs from each ovary were measured to determine the presence of obvious modal groups. A definite mode at from 9 to 12 mi- crometer units in egg diameter occurred in 46 of the ovaries, 15 to 35 mm. in length, from fish caught from September 15 to July 21. The median of this mode of 38 of these fish appears to lie between 10 and 12 micrometer units. In the ovaries of 8 fish, the median lies between about 9 and 10 units. One of these fish is the only running ripe female of the 54 examined. The remaining eight fish lack an intermediate- size mode; all of these have small gonads. No specimens were found among the 54 females with a group of eggs which have an average diameter greater than 12 units but less than the 20 to 21 unit average size of ripe eggs. Also, when checking for ripe females among the 3,863 adult manini of Table 6, the eggs pressed from the body cavity of few, if any, of these were of a size between that of the usual mode and the ripe egg mode. It is expected that such a size would be extruded with moderate pressure on the abdomen of the fish, since eggs of about 10 micrometer units (0.35 mm.) are frequently forced out. Eggs as large or nearly as large as completely ripe eggs, but in which small opaque white regions were visible, were seen almost as often as the ripe, perfectly transparent eggs. These can not be fertilized, and females contain- ing them were not considered among the ripe fish previously discussed. The failure to find a mode of eggs intermediate in size between about 10 and 20 micrometer units suggests that eggs maturing from the former modal group (the eggs of which are opaque) do so rapidly. Per- haps the enlargement is essentially a process of hydration. In view of the finding of so few female ma- nini with fertilizable eggs during the season, it is further believed that the ripe eggs are not retained very long in the ovary. Admittedly a change in behavior of ripe females might in- fluence the frequency with which they enter traps ( a possible cessation of feeding, however, would not seem to affect the rate of entry, for the traps, as mentioned, are unbaited). Also, confinement in the traps might affect the mat- uration of the eggs. If, however, it is assumed that ripe eggs are held within a female no longer than 1 day on the average, and 43 ripe female fish is the num- ber that would be present in truly random sam- ples of the population equal in size to the ones taken, then the small number of ripe females actually constitutes an indication that every adult female spawns once each lunar month. Since the lunar month is 29.53 days, only about one- thirtieth of the females sampled from the pop- ulation during the spawning season would be ripe. As mentioned, 3,863 adult manini were examined during the 1953-54 season, an esti- mated one-third of which, or 1,288, were fe- males. One-thirtieth of 1,288 is 43, precisely the number of ripe females which were found. Acanthurus triostegus sandvicensis — Randall 239 2 4 6 8 10 12 14 16 18 20 22 24 26 Diameter of eggs 125 mm. specimen. Oct. 27, 1952. Ovary length 20 mm. 2 4 6 8 10 12 14 16 18 20 22 24 26 Diameter of eggs 142 mm. specimen. Dec. 5, 1952. Ovary length 18 mm. • Diameter of eggs 146 mm. specimen. Feb. 6, 1953. Ovary length 32 mm. 106 mm. specimen. March 13, 1953. Ovary length 20 mm. 138 mm. specimen. March 17, 1953. Ovary length 35 mm. 2 4 6 8 10 12 14 16 18 20 22 24 26 Diameter of eggs 123 mm. specimen. June 21, 1953. Ovary length 42 mm. FIG. 8. Egg diameters of Acanthurus triostegus sandvicensis plotted as units of an ocular micrometer. Scale: 1 unit = .035 mm. Except for A, where the number of eggs of 1 and 2 micrometer units were esti- mated at 1,575 and 627, respectively, no eggs less than 3 micrometer units were measured. Spawning The spawning of A. triostegus sandvicensis was never witnessed, in spite of considerable observation both at night and during the day and at various times of the month. On June 12, 1957 (full moon) A. triostegus triostegus was observed to spawn at the atoll of Tikahau in the Tuamotu Archipelago at dusk (5:30 P.M.). It is believed that the failure to see the Hawaiian manini spawn was due to the lack of effort at dusk. The following account of the reproduction of the Tuamotu manini is probably similar to that for the Hawaiian subspecies. Several hundred spawning A. triostegus trio- stegus were observed in the pass at a depth of about 25 ft. Many of the fish displayed a marked color change. The black bars on the side of the body were much broader (but the added width on either side of each bar was not as intense a black as the center ) ; the median fins were dark, almost black, except for the middle of the caudal which remained pale; the dark median band on the forehead was prominent. Five fish of this color pattern were speared (when pursued, an individual dark-colored manini resumed normal coloration within about 10 sec.) ; all 5 were ripe males. As many as 4 or 5 of these more melanis- tic fish were repeatedly observed chasing single fish which were unchanged in color pattern. Two of the normally colored manini were speared. One was a running ripe female and the other a female with opaque eggs which were squeezed from the body only with considerable pressure; thus not all females in the spawning aggrega- tion are capable of reproduction at the same time, which is in keeping with the finding of running ripe female manini in Hawaii along with fish with maturing ova on many different days of the lunar month (Table 6). Actual spawning took place among small groups of fish within the large school which became increas- ingly active and suddenly darted upward about 8 ft. above the rest. Eggs and sperm were re- leased at the apex of this movement. Swimming was more rapid as the apex was approached. The fish dispersed as they swam back down to mingle with the school. At times two or more small groups joined together in the first few feet of the upward movement, resulting in a dozen or more fish in a compact mass at the top. The attempts to spear individual fish broke up the one large aggregation into several lesser ones in which spawning took place infrequently. 240 PACIFIC SCIENCE, Vol. XV, April 1%1 Two other surgeonfishes, Ctenochaetus stri- atus (Quoy and Gaimard) and Zebrasoma sco- pas (Cuvier), were observed to spawn in the Society Islands. Like the manini, they exhibited the sudden upward rush prior to spawning. Pos- sibly the release of eggs and sperm is facilitated by the expansion of the airbladder from the de- creasing pressure caused by the upward swim- ming movement. Like the manini in the pass at Tikahau, both of these surgeonfishes were spawning in a region of strong current to the open sea. Size and Age at Maturity The smallest running ripe female seen by me was 101 mm. in standard length. The smallest running ripe male was 97 mm. in standard length. These are probably near the minimum lengths for mature manini around Oahu. Ten manini were reared in a large concrete tank of the Hawaii Marine Laboratory at Coco- nut Island from a size of 25 to 27 mm. in stand- ard length beginning on March 17, 1953 (see detailed discussion in section on growth). Two fish of this group, an 89.5 mm. male and a 99 mm. female were killed for gonad study on July 24, 1953. The female had an immature gonad, clear gray in color, and only 5 mm. in length. The ova were minute, only 0.015 to 0.03 mm. in diameter. The male was detected because it was possible to strip a small amount of milt from it, although considerable pressure on the abdomen was necessary. The sperm, however, were not motile. Its testis was slender and 6 mm. long. Subsequent examination of the gon- ads of the remaining 8 captive manini was not made because of arrested growth of these fish following inadvertent destruction of the algal food supply when seining the tank. There appears to be considerable variability in the size at maturity. This variability is ap- parent when the 101 mm. ripe female manini is contrasted with the 127 mm. immature speci- men whose egg diameters are graphically illus- trated in Figure 8, C. It is doubtful that any female manini spawn during the season in which they first arrived as acronuri in tide pools. Even if the W mm. female, which was killed on July 24 following its growth in captivity from a small juvenile on March 17, had been among the first few arrivals to shallow water of the season (middle Febru- ary), another month of growth and gonad de- velopment would hardly seem sufficient to bring a tiny immature ovary such as that seen in this specimen on July 24 to full maturity before the season ended. It also seems unlikely that males will spawn within their first season following recruitment to tide pools. It seems probable, from a knowledge of growth (see growth section) and the approxi- mate size at maturity, that some manini will spawn in the season following their first sojourn as juveniles in inshore waters. This would be very likely for the first young of the season; the last young of the season would net seem to have sufficient time to develop to sexual matu- rity before the following spawning season ended. Such a situation would probably increase the variation in the size of the fish at maturity. Fecundity Fecundity is defined as the total number of ripe eggs produced by a female in 1 year. It is impossible to ascertain this for fish like the manini which may spawn more than once a year, when the number of spawnings per year is unknown. Therefore only the number of eggs released at one spawning can be determined. This was attempted for only a single manini. The fish chosen for the egg count was the 123 mm. June 21 specimen, the egg diameters of which are plotted in Figure 8, 7. This fish had a very large ovary (42 mm. in length) which contained a high percentage of large eggs. These eggs were not completely transparent, and none appeared to have been expelled by the fish. The ovary was vigorously shaken in the vial in which it had been preserved with a minimum of fluid. The agitation released nearly all of the near-ripe eggs from ovarian tissue and placed them in uniform suspension. The contents of the vial were then placed in a graduated cylinder, al- lowed to settle, and the small amount of excess fluid poured off. A sample of the ovarian mass was drawn off. After its removal, the volume was computed as 26 per cent of the total. All of the large eggs in this sample were counted in successive lots placed in a Petri dish under a binocular dissecting microscope. The sample Accmthums triostegus sandvicensis— Randall contained 10,814 large eggs. The total number of large eggs in this ovary was therefore esti- mated at 40,000. DEVELOPMENT Fertilization of the eggs of the manini was effected artificially, using live fish obtained from trap fishermen. As previously discussed, ripe female fish are detected by the release of per- fectly transparent eggs following gentle pres- sure on the abdomen. The milt of running ripe male manini oozes even more readily from the fish than the eggs of ripe females; however, it was found that eggs could be fertilized with sperm from males in which definite squeezing was necessary to bring forth the milt. Eggs were fertilized both by stripping ripe males and females concurrently into sea water with genital apertures in close proximity and by stripping them separately into different containers and subsequently mixing the contents. The fertilized eggs are spherical, 0.66 to 0.70 mm. in diameter, and contain a single oil globule which is 0.165 mm. in diameter. If suspended in sea water, the eggs slowly rise to the sur- face. They exhibit no stickiness and do not ad- here to objects or one another, although they often lie in single-layered groups at the surface. Unfertilized (but fertile) eggs are also less dense than sea water. No difference in diameter of unfertilized and fertilized eggs could be detected. Developing eggs were placed in large finger bowls, an aerated battery jar, and an aerated 20 gal. aquarium. In all of these containers heavy mortality of the larvae ensued because of set- tling to the bottom (probably through contact with bacteria there and not to any physical in- jury). Efforts were made to keep the larvae in suspension by placing a fine-mesh false bottom in a battery jar and locating a bubbling aerator stone beneath a hole at one side and by installing a paddle which was slowly turned by an electric motor. Although these methods reduced the amount of settling, mortality was still high and only a few larvae survived to an age of 5 days or more, when oriented swimming began. In view of the difficulty of rearing larvae in rigid containers, a new approach was tried for the manini. A fine-mesh bag, 1.2 m. on a side, 241 with cork along the top was floated in the Ala Wai Yacht Basin. Fertilized eggs were placed within and allowed to develop. Approximately the same mortality resulted, however. Once the manini larvae were capable of ori- ented swimming, they were maintained as easily in unaerated finger bowls as any other way. At this time the critical factor became the supply of food. The principal stages in early development from 0 hr. to 5 days 22 hr. are shown in Figures 9-12. The drawings of eggs before hatching were made shortly after placing the eggs in 10 per cent acetic acid, a treatment which makes the blastodisc opaque white and the cleavage pattern more visible. The ages given are based on an average development temperature of 24° C. During development the temperature was 24° C. zb about 1°. Although temperature control was not suf- ficiently precise to permit an assay of the change of rate of development with changing tempera- ture, it was obvious from the rearing of one batch of eggs at 26° C. dz 1°, that temperature has a marked effect on the rate. The stage at- tained at 28 hr. 40 min. at 24° C. was reached in about 22 hr. 30 min. at 26° C. The 42 -hr. stage at 24° C. was reached in about 32 hr. at 26° C. No attempt was made to determine the tem- perature limits within which normal develop- ment of the manini is possible. Due to proximity of a light bulb, the temperature of one finger bowl in which 37-hr. larvae were developing was inadvertently raised in a short time to 29.3° C. None of the larvae died at this time. The blastodisc of the fertilized egg is 0.4 mm. long. One hour after fertilization the first cleav- age becomes apparent. At the end of the second hour 8 cells are present, and after 4 hr. about 55 cells can be counted in the blastodisc. At the age of 6 hr. gastrulation is under way. Epiboly is evident from the thin layer with a wavy edge which is extending downward over the yolk. The cells are very small and are not shown in the drawing of this and subsequent stages. At 14 hr. the primitive streak is apparent and the cephalic region of the embryo well devel- oped. By about 18 hr. epiboly is complete, and 242 PACIFIC SCIENCE, Vol. XV, April 1961 0 hours 0 hours 45 min. 12 hours 45 min. 14 hours I hour 15 min. 18 hours 20 min. 4 hours 6 hours 25 hours 27 hours Fig. 9- Developing eggs of Acanthurus triostegus sandvicensis. Egg diameter, 0.67 mm. the yolk plug is visible as a small clear area near the oil globule. By 15 hr. 30 min. the body of the embryo is beginning to take form and elongate over the yolk. At 18 hr. 20 min. 9 somites can be counted and the optic vesicles are prominent. At 21 hr. 45 min. 12 somites are present. At 25 hr. 19 or 20 somites are visible; the lens of the eye is just starting to form; the auditory vesicle is present and the statoliths are just forming; 24 tiny melanophores can be seen in lateral view on the body; twitching movements are common; the heart contains blood although it has not yet started to beat. Hatching occurs at the age of about 26 hr. At 27 hr. the larva is nearly 1.7 mm. in length and has 22 somites; the heart is still quiescent. At 31 hr. the heart was observed to beat; oc- casional short random swimming movements oc- cur by rapid vibration of the posterior half of the body. At the age of 42 hr. the amount of yolk is reduced to about half of what is present in the 31 -hr. larva. As the yolk is used up, the larvae show a progressive tendency to sink. Larvae of 39 hr. still float head down at the surface, but 42. 5 -hr. fish have begun to settle. The rate of settling of 1 larva at this age was measured at 1.2 cm. per min. The average rate of 3 44.5-hr. larvae was 1.67 cm. per min., and the average rate of 4 47-hr. larvae was 1.88 cm. per min. One 71-hr. larva sank at the rate of 8.2 cm. per min. At 42.5 hr. some larvae were already settling out at the bottom of the aquarium. Most, how- ever, maintained their level by swimming move- ments. 300 such movements were observed in an aquarium which contained several hundred larvae. Of these movements, 227 resulted in the larvae being in a higher position in the tank; the rest were sideways or downward. Movements which consisted only of a twitch were not Acanthurus triostegus sandvicensis — Randall counted. The ability to overcome settling is more efficient than 227 movements with an upward component out of 300 would indicate. Since the head is directed downward, swimming is initi- ated in this direction. In most cases the fish soon turn sharply and swim upward. If the in- terval of movement is short, there may be in- sufficient time to end up higher than the start- ing position, but usually the fish is not signif- icantly lower. The interval between movements is highly variable, but it averages about 1 min. The distance traveled by the larvae per move- ment is usually less than 40 mm. One, however, stopped 110 mm. above its starting position. In addition to geotaxic responses, swimming movements were made to avoid contact with an approaching object. This was first observed in I mm. Fig. 10. Early larval stages of Acanthurus triostegus sandvicensis. 243 42 -hr. larvae; however, it may be operative sev- eral hours earlier. The eyes do not seem suf- ficiently well developed to be functional at this early age, and a glass rod is detected as easily as an opaque one; thus vision would not seem to be the sense responsible for perception of an approaching object. Before such an object touches a larva, and sometimes when it is as far as 2 or 3 cm. away, a fleeing movement is ini- tiated. A second stimulus soon after the flight reaction will usually result in a second swim- ming movement. Contact with the bottom due to slow settling usually will not result in a swim- ming movement. Regular movements occur at about the same rate when the fish are on the bottom. They may enable a larva to become sus- pended again, but usually only for a short time. Areas of white pigment are beginning to form on the ventral part of the seventh and fifteenth somites of the 42 -hr. larva. These are large and conspicuous in the 54-hr. larva, and lesser white areas may be seen on the ventral part of the body anterior to the seventh somite and on the head. At this stage the intestine has just formed and the yolk mass is small. The melanophores over the yolk are large and dendritic. Figure 11 consists of microphotographs of living 7 5 -hr. and 4-day 1-hr. larvae with the light transmitted through the little fish. The dorsal fin fold is a prominent feature at 75 hr. (it was first noticed in the 66-hr. larva), and the pectoral fins are developing. Dark pigment around the eye is evident. The 4-day 1-hr. larva was observed to sink less rapidly than prior stages. One larva of this age sank 2.5 cm. per min. and another less than 1 cm. per min. The 4-day 12 -hr. larva remained suspended head down in the water, thus indicating that the air bladder was functional. The jaws were observed to move in the 4-day 5 -hr. larva (Fig. 12). Although still tending to float head down- ward, the 5 -day larva is capable of normal ori- ented swimming. The pectoral fins can be flut- tered rapidly, and the eyes move. The yolk is completely gone and the oil globule half re- sorbed. By 5 days 12 hr., the 3 individuals which survived to this age appeared to be feeding by short darting movements. Material from a cul- ture of marine ciliates was added to the finger bowl in which these 3 larvae were kept. One 244 PACIFIC SCIENCE, Vol. XV, April 1961 FIG. 11. Microphotographs of larval Acanthurus trios tegus sandvicensis. A, Side view of 75-hr. larva; length 2.3 mm. B, Four-day 1-hr. larva in top view; length, 2.5 mm. died after 5 days 17 hr. of development. Food was present in the intestine of this specimen. The second member of this trio was observed to be dying at the age of 5 days 22 hr. It was re- moved from the finger bowl for the lower draw- ing of Figure 12. As may be seen in this draw- ing, no trace of the oil globule remains, the in- testine is convoluted, the liver is prominent, and the gall bladder is visible. The two large white areas on the body have disappeared. The total length is nearly 2.7 mm. The last larva survived to the age of 6 days 12 hr. It grew very little in the last 24 hr., presumably because of insuf- ficient or inadequate food. Three of the 11 acanthurid lavae from the POFI plankton collections (see p. 216) were identified as manini by dorsal and anal fin-ray counts. They are 6.6-7. 0 mm. in total length. Figure 13 is a drawing of the 6.6 mm. specimen (standard length 5.3 mm.). The teeth are not visible without opening the jaws. They are simple canines. The total length of the anterior upper teeth is contained 3.7 times in the diam- eter of the pupil of the eye. Specimens smaller than this one could not be identified by ray counts, for the soft rays s~e not fully formed. A 4.2 mm. specimen could not be positively identified at this time to genus, for only the first three dorsal spines are suf- ficiently developed to distinguish them from soft rays; however this specimen is about interme- diate in structure to the 5 -day 2 2 -hr. larva and the 6.6 mm. larva (though is closer to the lat- ter). Its body depth is contained 1.8 times in the total length, and the relative length of the second dorsal, second anal, and pelvic spines is about two-thirds as great as that of the 6.6 mm. larva. The caudal fin is about half formed, the urostyle extending to the posterior part of the incipient fin. A 4.3 mm. specimen was the only acanthurid found in the POFI larval fish collections which were taken with a 6-ft. modified Isaacs-Kidd trawl, except for an 18 mm. Zebrasoma veli- ferum (Randall, 1955c: fig. 3). It was captured near the surface off Kahuku, Oahu. Although a little longer than the specimen just mentioned, it is in a slightly earlier stage of development and is probably a different species. If a large series of specimens were available instead of just a few it might be possible to identify the various stages of the manini to the size where fin-ray counts alone can provide definite iden- tification. No postlarval specimens of Acanthurus greater than 8.7 mm. in length but smaller than the acronurus form were found in any of the POFI collections or museum collections. As previously discussed, specimens larger than about 9 mm. FIG. 12. Larval stages of Acanthurus triostegus sandvicensis. Acanthurus triostegus sandvicensis—KANDKLL Fig. 13. Postlarval Acanthurus triostegus sandvi- censis taken in a 1 m. plankton net at a depth of 50 m. at 22° 38' N., 157° 11' W. in length probably elude a meter plankton net. The presence of the 18 mm. Zebras oma veli- ferum in the collections from the 6-ft. trawl suggests that this would be a better means of catching larger larval acanthurids. The failure to take more acanthurids in the trawl is prob- ably due to the great distance from land of most of the trawl tows. The only records of the early stages of Acan- thurus found in the literature are two postlarval A. coeruleus identified by Liitken (1880: pi. 5) and an unidentified 7 mm. postlarval Acanthurus in Weber (1913: fig. 70). Sparta (1928) re- ported on a long, ribbon-like, transparent, pela- gic egg mass of an unknown teleost fish from the Mediterranean. The early developmental stages which he succeeded in rearing from this egg mass are remarkably similar to those of the manini. Sparta concluded that the eggs were from a species as yet unrecorded from the Med- iterranean. TRANSFORMATION The late postlarval form of the genus Acan- thurus is characteristic in morphology, yet iden- tifiable to family Acanthuridae by the presence 245 of the caudal spine and a diagnostic number of fin rays. Originally this stage was placed in a genus by itself, Acronurus. The name acronurus has persisted as a common name for this stage, reminiscent of the leptocephalus of eels. Acronuri are commonly taken at night-light stations. Sixty-two manini acronuri and many of other species collected at night lights and by night-light traps offshore in the vicinity of the Hawaiian, Line, and Phoenix islands were exam- ined. No specimens in night-light collections were found which were smaller than the acron- urus stage. It is inferred, therefore, that a posi- tive phototaxis develops in the acronurus. The lack of food material in the gut of night-light specimens (see p. 223) indicates that the pres- ence of numerous small planktonic animals around a light at night is not the attractant. Once transformation to the juvenile stage has begun, the positive phototaxis disappears. Figure 14 is a photograph of a live specimen of the manini acronurus taken shortly after its arrival in a tide pool. The characteristic trans- parent scaleless body with silvery abdomen and head (except snout and nape), disk-like form, and vertical striae on the body are apparent. The vertical bars are just beginning to form. The acronurus comes into inshore regions Fig. 14. The acronurus of Acanthurus triostegus sandvicensis. Captured at dawn in a tide pool at Dia- mond Head, Oahu, and photographed alive. 246 only at night. New arrivals to inshore areas have been collected all through the night, some as early as 9 P.M. and others just before dawn. Most of the night collecting was done at low tide. It is not known whether the influx of acronuri is greater at one phase of the tide than another. On two occasions before midnight at low tide in ankle-deep water along the shore of the Ala Wai Yacht Basin near the entrance to the Ala Wai Canal, manini acronuri were observed just coming into the area from deeper water. They did not passively float into the shallow zone but swam in rapidly. One crossed several times through the beam of a head lamp before it was caught. If it is assumed that these fish were in deep water beyond the breaker zone before nightfall, then they must have actively swum into the harbor area and not been carried in by any tidal currents. Prior to low tide, tidal cur- rents would be flowing out of the yacht basin and not into it. This is contrary to the belief of Breder ( 1949^: 296) that acronuri of Acan- thurus hepatus (= A. chirm gus) are carried into shallow water by "vagaries of current.” Breder reported observing transforming speci- mens of A. chirm gus at sizes from 23 to 29 mm., and found juveniles from 10 to 20 mm. in length in tide pools at Bimini. He inferred that late postlarval surgeonfish reach a certain size in the plankton at which transformation is pos- sible but continue to grow. The size at trans- formation is dependent on the size of the acron- urus when it reaches shallow water. Breder s explanation of the cause of this variability in transformation size is plausible, although a range of at least 10 to 29 mm. for one species seems high. From April 11 to October 4, 1953, a total of 175 manini were found in shallow water on Oahu in their first day of transformation to the juvenile stage. Their standard lengths (meas- ured to nearest 0.5 mm.) ranged from 22 to 29.5 mm. Even this would seem, a priori, to be more variation in length than would be expected from mere growth variation alone in the pelagic habitat. A small amount of this variability in Hawaii is due to the change in temperature at which development occurs during the season (Fig. 16 and discussion below), but consider- able variation can be seen in the transformation PACIFIC SCIENCE, Vol. XV, April 1961 size of acronuri taken within the period of a single month, and therefore cannot be attributed to seasonal temperature differences. The fluctuation in abundance of incoming young manini has provided an opportunity to test Breder’s hypothesis. If the variation in size at transformation were entirely due to normal variation in growth, then no difference should be apparent in the size of manini which are transforming during the time of the month when a big influx is occurring and the period when the recruitment to inshore areas is minimal. If, however, more extremes in size were found when the recruitment is at a minimum, it could be assumed that growth of the manini occurs in the plankton after transformation is possible and that members of a modal group which ar- rive inshore and transform earlier or later than the majority of the group will be smaller and larger, respectively. In Figure 15 the lengths of the 116 trans- forming manini which were collected within the period 5 days before to 5 days after peak tide- pool recruitment (taken as 2 days before the time of new moon; see p. 234 and Fig. 6) are compared to the lengths of the remaining ma- nini caught outside this period. Clearly, a higher percentage of extremes in length at transforma- tion occur during the part of the month when the number of acronuri entering shallow water to transform is low. Fig. 15. Size of Acanthurus trio ste gus sandvicensis during the first day of transformation from the acron- urus to the juvenile stage. The solid line represents the percent of fish at the designated standard lengths from samples taken 5 days before to 5 days after peak tide pool recruitment (3 days before new moon). The dotted line represents the percent of fish from sam- ples taken during the rest of the lunar month. The solid line is based on a total of 116 fish and the dotted line on 59 fish. Samples include all of the first day transforming manini which were collected from Apr. 11 to Oct. 4, 1953. Acanthurus triostegus sandvicensis — RANDALL If spawning of the manini occurred only dur- ing a definite small segment of the month, the dotted curve of Figure 15 would probably be bimodal, one of the modes centered on a low and the other on a high standard length. Since, however, spawning appears to occur through- out the month and merely more prominently in one part than another, most of the young re- sulting from fertilization at a time of low spawn- ing intensity will come into shoal areas at a more nearly average size during the interval be- tween periods of high tide-pool recruitment and obscure the postulated bimodal effect. The extremes in size of this small group would enter the tide-pool zone at the time of peak recruit- ment; thus, if the samples were large enough, both curves would extend to the same limits. The maintenance of populations of manini around small islands such as Johnston Island, the shores of which are constantly swept by a strong current in one direction, would seem to demand some means on the part of the develop- ing pelagic larvae to remain close to land. The presence of small eddy systems around an is- land might afford the explanation, provided the larvae can stay within such systems. The ap- parent differentiation of the manini at Johnston Island (Randall, 195 6£: table 2) and also of Ctenochaetus strigosus (Randall, 195 5<7) would seem to rule out the possibility that the Johns- ton Island populations are derived primarily from fishes which spawn in the Hawaiian Is- lands proper. While examining museum specimens of trans- forming manini from many localities, a definite variation in size with locality was noticed. The standard lengths of the available early transform- ing museum specimens were measured. These lengths (except those of Hawaiian Islands speci- mens, already presented graphically) are given in Table 7 with localities and with collection dates when known. A correlation exists between the size of ma- nini at transformation and the temperature of the water of the locality. The warmest region listed in the table is the East Indies where the mean monthly sea surface temperature may reach 85° F. (29.4° C.) (Hydrographic Office, 225). Here we find the smallest size at transformation, 20-21 mm. Small standard lengths, averaging 247 less than 23.5 mm., are found in other warm areas such as the Palau, Marianas, Gilbert, and Marshall islands (all regions where the mean monthly sea surface temperature exceeds 81° F., or 27.3° C, during the entire year), and southern India in January, and Okinawa in September. Transforming specimens from re- gions where the temperature is 80° F. (26.6° C.) or less throughout the year or during the month of the date of collection are large, 24 mm. or greater in standard length. These areas include Durban (S. Africa), New South Wales, Ningpo (China), Hawaiian Islands, Mangareva, the Marquesas Islands ( in August ) , and Clarion Island. The large transformation size of the specimens from the Phoenix Islands, where the sea surface temperature exceeds 82° F. (27.8° C. ) all year, is an obvious exception to the above. Perhaps the manini population in these islands has differentiated in this respect, although the Phoenix group is not markedly isolated. In order to preclude the possibility that the different transformation sizes of the different areas are entirely due to genetic factors (re- flecting possible races of the manini in all these areas) a comparison was made of the size at transformation of specimens obtained at Oahu in an early cool and a late warm part of the season. Thirty-nine first-day transforming ma- nini were obtained in collections from April 1 1 to May 8, 1953. The standard lengths of these specimens are plotted as the solid line of Figure 16. The dotted line enclosing the stippled por- tion of the graph represents the lengths of 38 first-day transforming manini taken from July 7 to October 4, 1953. The average sea surface temperature at Oahu from February to May is 75.7° F. (24.2° C.) ; the average from July to October is 79.1° F. (26.2° C), based on av- erage temperatures from 1941 to 1947 presented by Leipper and Anderson (1950). The mean length of the April-May group is 26.256 mm.; the mean length of the July-October group is 25.447 mm. Although the difference in length appears highly significant, a t test was made by the group comparison method (Snedecor, 1948: 80) . The resulting t value of 3.11 gives a prob- ability of nearly 0.001 of obtaining a value greater than this; thus it is extremely unlikely that samples as diverse as the above could be 248 PACIFIC SCIENCE, Vol. XV, April 1961 TABLE 7 Size of Acanthurus triostegus at Transformation from the Postlarval to the Juvenile State LOCALITY DATE NO. SPECIMENS STANDARD LENGTH (mm.) Range Mean Durban, S. Africa May 8 2 23-25 24 S. India Jan. 19 1 22.5 22.5 East Indies Sumatra Dec. 19-25 1 21 21 Moluccas 10 19-21 20.2 New South Wales 3 24-25 24.3 Palau Is Aug. 3-6 7 21-23 22.1 Philippine Is Jun. 1 23.5 23.5 Okinawa Sep. 6-8 2 22 22 S. Japan (30.4° N.) 2 21-23 22 Ningpo, China (29-5° N.) 5 24-26 24.9 Mariana Is. Guam Jun. 28 3 22-24 22.7 Jul. 24 7 20.5-23 21.8 Nov. 25 2 21.5-23 21.25 Saipan Jul. 9-14 1 22 22 Rota Nov. 11-13 16 21.5-24 22.5 Marshall Is. Bikini Apr. 1 1 23.5 23.5 Aug. 17-18 2 22-22.5 22.25 Eniwetok May 20 2 23-23.5 23.25 Kwajalein Jun. 24 7 22-24 22.7 Sep. 1 3 22-24 22 Onotoa, Gilbert Is Aug. 21 9 21-25 23.2 Samoa Is. Tutuila Jun. 3 1 22 22 Rose Jun. 11-14 1 24 24 Swains May 3-9 1 25 25 Phoenix Is. Canton Apr. 15-28 3 25-26.5 25.8 May 13 3 25-27 25.8 Enderbury May 15-19 7 25-27 25.3 Hull Jul. 7-17 9 23-25.5 24.1 Howland I - Sep. 24 8 23.5-26 24.3 Line Is. Palmyra Jan. 18 1 23.5 23.5 Fanning.. Aug. 1 24 24 Wake I Jun. 9 2 23.5-24 23.75 Marcus I. --- Aug. 4 25-26.5 25.6 Marquesas Is. Feb. 3 1 26 26 Aug. 2 26-26.5 26.25 Makatea, Tuamotus Feb. 13 3 23.5-25 24.3 Mar. 15 8 23-25.5 24.4 Mangareva ----- - Feb. 3 7 24-25 24.7 Tahiti Feb. 1 26.5 26.5 Apr. 28 1 23 23 Jun. 16 11 22.5-26 24.2 Jul. 8 2 23.5-24 23.75 Aug. 12 4 22.5-24 23.4 Aug. 28 5 23-25.5 24.4 Oct. 4 3 25-27 25.7 Dec. 30 17 22.5-26.5 24.5 Clarion I., Mexico — 1 25 25 Acanthurus triostegus sandvicensis — RANDALL Fig. 16. Size of Acanthurus triostegus sandvicensis from Oahu during the first day of transformation from the acronurus to the juvenile stage. The solid line represents all of the specimens collected from Apr. 11 to May 8, 1953; the dotted line indicates those collected from Jul. 7 to Oct. 4, 1953 - drawn by chance from a population in which the transformation length was the same through- out the season. It is therefore concluded that a relationship as indicated above exists between the size of transformation of the manini and the temperature of the water. Breder (1949a) stated that the complete trans- formation of Acanthurus chirm gus occurred in a period of about 48 hr. He did not, how- ever, define the morphological state that con- stitutes complete transformation. In the manini, adult coloration is nearly acquired after about 48 hr., but other major changes are less than half completed after this length of time. Ad- mittedly it is difficult to define a precise stage at which larval characteristics have completely disappeared and miniature adult structure is assumed. Probably the most convenient criterion is the stage at which the scales have completely formed. Drawings of developing scales have been made by Poey (1875: pi. 3) for the West Indian species, Acronurus caeruleatus ( — Acan- thurus coeruleus ) and Acronurus nigriculus {— Acanthurus bahianus) . The first evidence of the formation of the scales is the appearance of scattered thornlike outgrowths which project upward as well as posteriorly from the narrow, vertical, fleshy ridges of the body (these spines do not occur on the anterior half of the body ) . 249 The spines are present on manini acronuri caught offshore at a night light, but they can be seen on these specimens only with the aid of a microscope. They enlarge and become visible (when wet) with the naked eye on the majority of first-day transforming individuals as early as 7:00 A.M. Each such spine on later stages was found to be a central stout ctenius of a single developing scale; the majority of scales, how- ever, do not possess these. As scale development proceeds, the large ctenii flatten and become overlaid with epidermis. When this occurs, scale development is essentially complete as are other major changes of the transformation process. Scale development normally requires 4-5 days. In an aquarium the whole process seems de- layed, and about 6 days are necessary. Figure 17 shows transforming manini on the first, second, third, and fourth to fifth days. The second-day and third-day stages were identified with assurance by comparison with specimens recovered in tide pools following fin clipping on the first day of transformation. No older marked specimens were recovered at the stage when scale structure is just complete, and the time this required had to be inferred from the percentage of specimens in the different stages collected in the weekly poison stations of the Diamond Head tide pool of Figure 4. Fig. 17. Transformation of Acanthurus triostegus sandvicensis from the acronurus to the juvenile stage. Upper left, first day; upper right, second day; lower left, third day; lower right, fourth or fifth day. 250 The vertical dark bars which are so charac- teristic of the manini are absent from the ma- jority of night-light specimens. They are faintly visible on a few such specimens, however. Ma- nini caught in tide pools during the same night when they came in usually possess traces of the dark bars, but some, presumably very recent arrivals, totally lack these markings. One of the two previously mentioned specimens which were observed just entering shallow water from deeper areas of the Ala Wai Yacht Basin had traces of the bars, and the other lacked them. The bars are not dark on any of the specimens collected at night. During daylight of the first day of transformation the bars become intensely dark brown. This suggests that sunlight might accelerate or perhaps even be necessary for the darkening. In a review on the biochemistry of the production of melanin, Lerner and Fitz- patrick (1950: 119) pointed out that ultra- violet irradiation appears to be concerned with melanin formation in at least four different ways, all of which tend to increase pigmentation. An experiment was designed to determine whether any changes occurred in the transforma- tion process when light is excluded. Two trans- parent specimens which displayed no evidence of dark vertical bars on the body were taken at midnight. These were placed in a battery jar containing IVz gal. of aerated fresh sea water. The jar was covered with a box which, in turn, was covered with a heavy black cloth. Three other transparent specimens taken at the same time were placed in a aquarium of about the same capacity which was not darkened. The battery jar was uncovered 12 hr. later. The transformation of the two specimens had barely started. The stage which was reached was about comparable to transforming fish taken shortly after dawn. The bars on the side were barely discernible and the ctenii of the developing scales on the posterior half of the body were just visible. One of the fish was on its side, dying. It was removed, and the jar was again darkened. Eight hours later the second specimen was found nearly dead. The bars were possibly a little darker; the transparency of the body was largely replaced by opaque white, an abnormal color alteration. The three aquarium specimens com- pleted their transformation. The results suggest PACIFIC SCIENCE, Vol. XV, April 1961 that light is essential to normal transformation. To be conclusive, however, the experiment should be repeated several times. If light were conclusively demonstrated to be essential to the metamorphosis of the manini, the mechanism by which it exerts its effect would still not be known. It might act directly on the tissues, or by way of the visual sense, or both. During the first day the transparency of the body is gradually lost. By the end of the day the dark brown bars on one side of the body cannot be seen through the body from the other side as they may be on specimens such as the one in Figure 14. The bright silver color on the abdomen and head is slightly dulled by the end of the day due to the development of melano- phores external to the silver layer. The iridocytes causing the silver sheen on the abdomen are present in the peritoneum. On the gill cover they occur in a layer on the underside of the opercular bones. They are found on the surface of the cleithrum at the edge of the gill opening. No silver color is present dorsal to the eye. In- stead there is a broad black band which covers the brain. This is found beneath the dorsal bones of the head. By the third day both the silver color and the black band on the head are nearly imperceptible. They are, like the rest of the body except the region of the dark bars, white with numerous, close-set, small flecks of, dark brown. The olivaceous cast of the body as seen on juvenile and adult manini is not conspicuous until about the fourth or fifth day of transforma- tion, probably because of slow development of yellow chromatophores. The second dorsal and second anal spines of the acronurus stage are longer and stouter than the other spines of these fins (Fig. 14). If stuck by either of these spines or by the pelvic spines, a stinging sensation almost comparable in in- tensity to a bee sting is felt. It is believed that these same spines, very elongate in the 6.6 mm. manini, are venomous in earlier stages. The poi- sonous nature of the spines appears to be a lar- val adaptation, for it is lost by the third day of transformation. A diminution in the relative length of the second dorsal and second anal spines may be seen in the series of transforming specimens of Figure 17. The relative change in length of these spines continues until, in late Acanthurus triostegus sandvicensis — RANDALL juveniles and adults, the third dorsal and third anal spines are much longer than the second dorsal and second anal spines, respectively. The latter, however, remain thicker. All of the acron- urus spines are T-shaped in cross-section; all are solid. The nature of the venom apparatus is un- known. The T-shape of the spines persists into the juvenile stage. One of the most striking changes during transformation is the alteration of the configura- tion of the head. This consists primarily of an increase in the length of the snout, resulting in the mouth being in a more ventral position. The rostral prolongation may be seen in Fig- ures 14 and 17. The diameter of the eye (which does not change during transformation) of the transparent specimen of Figure 14 is slightly longer than the length of the snout; in the just transformed manini the eye diameter is con- tained about 1.5 times in the length of the snout. The osteological changes which accom- pany the changes in the proportions of the head were not investigated. A slight increase in standard length occurs during the metamorphosis to juvenile form. After 5 days in an aquarium, three transforming specimens, 24.1-25.0 mm. in standard length, increased 1.2-1. 5 mm. in standard length. Most of the increase is attributable to the prolonga- tion of the snout. A remarkable lengthening of the digestive tract takes place during transformation (Fig. 3 ) . In a period of 4 to 5 days the total length of the alimentary tract of the manini increases about three-fold. This lengthening occurs at the time of a change in food habits from feeding on zooplankton to feeding on algae (see pp. 222-223). The teeth of the manini acronurus just be- fore transformation are very different from the simple conical teeth of the postlarval manini (as seen in 6.6 to 7.0 mm. specimens). They are close-set and flattened with denticulations on the edges similar to adult teeth. There are about 14 of these teeth, each with 6 or 7 den- ticulations, in the upper jaw with a broad gap between teeth at the symphysis of the jaw. There are 14 such teeth in the lower jaw with 5 den- ticulations each. During transformation 8 large adult-type teeth (with about 12 denticulations), 251 which are preformed within the premaxillary bone above the upper series of larval teeth, grow down over the larval teeth which are shed. The more medial of these large teeth are the largest and the first to emerge. In the lower jaw the same process occurs. Six large adult-type teeth with 7 denticulations grow over the larval teeth. Larval teeth lateral to these large ones may per- sist beyond the period of transformation. The sequence of tooth replacement is not without variation from specimen to specimen. At the stage when the scales are just completely formed on the body some specimens have fewer than 8 upper and 6 lower teeth in position. Some specimens taken at night lights in deep water appear to be getting their adult teeth prematurely. One specimen of Accmthurus tri- ostegus triostegus obtained at a night-light trap at anchorage at Christmas Island already had 6 large upper and 2 large lower teeth. GROWTH No information could be found in the litera- ture on growth of any species of surgeonfish. The closest family, phylogenetically, to the Acanthuridae in which growth data are avail- able for comparative purposes is the Teuthididae ( Siganidae of most authors ) . Like the acanthu- rids, the teuthidids are herbivorous (Suyehiro, 1942; Al-Hussaini, 1947). Whitehouse (1923: 68-70, 83-85) reported that juveniles of Teu- this java grew from a length of 25 mm. in April to 75 mm. in September in Silavatturai Lagoon, Tuticorin, India. Ommanney (1949: 48-49) stated that Siganus corallinus ( = Teuthis cor- allinus) reached a length of 180-220 mm. after 2 years’ growth, and about 280 mm. at the end of the third year, in Mauritius. Information on growth of the manini was obtained by the rearing of fish in a tank of the Hawaii Marine Laboratory, at Coconut Island, Oahu, analysis of the progression of modes among the juveniles of Figure 6, recovery of young fish marked by fin clipping, and the re- covery of tagged adults. The data are discussed separately under these headings below. The standard-length measurement for all of the work on growth of the manini is the length from the tip of the snout to the middle of the 252 PACIFIC SCIENCE, Vol. XV, April 1961 JOO 1 Time • Fish released after capture * Fish sacrificed for gonad study Aug. I Sept. I Oct. I Fig. 18. Growth of Acanthurus triostegus sandvicensts. Based on 10 juvenile specimens, 25 to 27 mm. in length, released in a concrete tank at Coconut Island, Oahu, Mar. 17, 1953. dorsal black bar posteriorly on the caudal pe- duncle. It was found that this measurement could be made with greater accuracy on living fish than the usual standard length to the base of the caudal fin (i.e., posterior edge of hypural plate). The original measurements are herein retained for all tables, graphs, and discussion (and designated simply "standard length”), rather than converting to true standard length. Should the latter measurement be desired, it may be obtained by increasing the peduncular bar length by 3.5 per cent. Growth of Captive Fish Ten manini, 25-27 mm. in standard length (mean length 26.4 mm.), were placed in a con- crete tank at Coconut Island on March 17, 1953. The tank measures 34.7 X 10.8 ft. and is an average 3.9 ft. in depth. A constant supply of sea water is pumped into the tank. When the fish were introduced, no other fishes were pres- ent and the walls were covered with an exceed- ingly luxuriant growth of many species of algae. The top of the caudal fin of the 10 juveniles was removed to permanently mark them (see below ) as a safeguard against the possible addi- tion of more manini to the tank by other per- sons. The growth of these fishes is recorded in Fig- ure 18. The 3 fish captured on May 13 (mean length 66.0 mm.) had grown an average of 20 mm. per month since March 17. On August 6 all of the manini were caught (2 had disap- peared and 3 were killed for gonad study). They averaged 90.9 mm. in standard length. By September 27 they had grown to an average standard length of 94.3 mm.; thus the average rate of growth from August 6 to September 27 had diminished to 1.8 mm. per month. The manini were caught by seining the tank after partially draining it. The efforts to catch them on September 27 failed several times, and much of the algae was detached from the walls of the tank before all of the fish were netted. Acanthurus triostegus sandvicensis — Randall Following this the amount of algae on the walls remained slight and apparently insufficient to support the growth of the fish. The 5 remaining manini were measured again on November 19. They were thin, and instead of growing they lost an average of 1.3 mm. in standard length. Manini this size shrink nearly 3 mm. in stand- ard length when preserved in 10 per cent form- aldehyde solution. Growth Measurement from Progression of Modes The collections of juvenile manini from Ke- walo Basin to Makapuu Point, Oahu, plotted in histogram form in Figure 6 permit an estimate of the early growth rate by analysis of the pro- gression in standard length of prominent modal groups from week to week. By this method the growth rate of juvenile manini from about 26 to 45 mm. in length is computed at about 12 mm. per month. Particular reference was made to the mode which first fully appears on the graph on April 17. Growth of Marked Juveniles Before undertaking the marking of juvenile manini in the field by fin clipping, experiments were performed in aquaria to test the possibility of fin regeneration. One pelvic fin and the pos- terior part of the soft portion of the dorsal fin were cut from 3 juvenile manini averaging 27.5 mm. in standard length. Although the fins were severed at the juncture with the body they regenerated rapidly and were nearly completely formed in a month. In a second experiment the top few rays of the caudal fin were cut away as well as the upper corner of the hypural plate to insure the removal of all fin elements. This cutting resulted in the permanent deformation of the caudal fin of the fish. Ten manini, 28.5 to 30 mm. in standard length (mean 29-2 5 ) were caught at the Ala Wai Canal near its entrance to the Ala Wai Yacht Basin on April 24, 1953. The top of the caudal fin of these fish was removed and they were released in the same area. Three of these marked fish were recovered, 1 on May 2 which measured 32.5 mm. in standard length and 2 on May 15 which were 34.8 and 37.0 mm. long 253 (Fig. 19) . The average growth rate was 9.6 mm. per month. Ninety-six juvenile manini, 25-28 mm. in standard length (mean 26.6 mm.), were caught, marked, and released in a Diamond Head tide pool on May 10, 1953. A total of 11 were re- covered from this pool on three different dates (Fig. 19). The average growth rate of these fish was 12.9 mm. per month. The 4 manini of this group which were recovered on June 23 ranged from 42.5 mm. to 48 mm. in standard length and averaged 45.1 mm. When first caught with the 92 others on May 10, these fish were a part of the prominent mode which may be seen centered on about 26.5 mm. standard length on the graph for this date in Figure 6. This same mode is still apparent in a sample of 147 fish which were collected at Diamond Head on June 23 in an effort to recover marked fish. At' this time the mode is composed of 27 fish, 40 to 49 mm. in standard length with a mean of 43.3. Since the 4 marked manini taken on this date are distributed well within the mode and of larger average size than its mean value, it is concluded that the fin clipping did not retard the growth. The difference in growth rate between the Diamond Head marked manini and the Ala Wai marked fish may be associated with the difference in algal flora of the two areas. Algal growth in the latter area is sparse; it consists primarily of diatoms. The extraordinary growth rate of the captive manini in the Coconut Is- land tank is probably also associated with the nature of the algal food supply; the stand of algae was much greater than normally observed in reef or bay areas around Oahu. Invertebrates, including herbivorous types such as echinoids, gastropods, and decapod crustaceans, were pres- ent in the tank. It is assumed that the large supply of algae was primarily due to the lack of herbivorous fishes. When introduced to this unusual environment, the 10 juvenile manini were given an unprecedented opportunity for rapid growth. In view of the more normal environment of the Diamond Head tide-pool region and the similarity in growth rate of marked fish from this region to the rate computed from the pro- gression of modes, 12-13 mm. per month is 254 PACIFIC SCIENCE, Vol. XV, April 1961 50 $ 96 fish, 27.5 to 28 mm, Diamond Head tidepool, May 10, 1953. 0 10 fish, 28.5 to 30 mm, Ala Wai Yacht Basin, April 24, 1953. *£ 45 E JZ 40 35 30 Apr. 20 Apr. 30 May 10 May 20 May 30 June 9 June 19 Time June 29 Fig. 19- Growth of juvenile Acanthurus triostegus sandvicensis on Oahu. Based on recovery of marked fish. considered as the probable rate of growth in spring of early juveniles of the bulk of the Oahu manini population. The growth of juveniles in the winter ceased. This was first noticed from shallow-water col- lections and observations made during the fall and winter months of 1952-53. Up to Novem- ber, manini less than 40 mm. could usually be found in the collections or seen in the water. In late November, December, and January no specimens less than 40 mm. were found, but the 40 mm. size persisted. This suggests that the last incoming juveniles of the season, which en- ter tide pools in early October, grow to a stand- ard length of about 40 mm. by November and then stop growing. During the next two winters the same phenomenon was noted. In order to demonstrate this cessation of growth and to ascertain the time when growth is resumed, 22 juvenile manini, 40 to 72 mm. long, were caught with a dip net, marked by fin clipping, and released on the nights of Jan- uary 5 and 6, 1954, at the entrance of the Ala Wai Canal to the Ala Wai Yacht Basin, Oahu. Fish were taken at both the northwest (Wai- kiki Yacht Club) and southeast shores of the canal entrance. On the two sides of the canal three basic methods of fin clipping were uti- lized: the top of the caudal fin was removed, the bottom of the fin, and both the top and bottom. Manini of variable size were chosen for each category, and another fin was cut on each specimen. Also any distinctive color marks ( usu- ally at the base of the pectoral fin) were noted. Subsequent efforts to recover the fish were limited by conditions of wind, tide, and tur- bidity of the water. Although more subject to the effects of turbidity than exposed shores, the Ala Wai area was preferred, for fish are more restricted to the shore area because of the mud bottom away from the bank of the canal. Fish of this size in a region such as Diamond Head move well out on the reef where the broader area, deeper water, and action of surf make them difficult to capture and recapture. Although, as mentioned, some of the larger manini come into tide pools at night, their numbers are not suf- ficient for such a marking program. Since fish were measured alive with a light at night, difficulty was experienced in obtaining Acanthurus triostegus sandvtmtisis — Randall 255 precise length determinations. Although the lengths were recorded to 0.5 mm., it should be added that variations as great as 1 mm. from the original measurement were noted in un- biased tests of accuracy. Eleven of the 22 marked fish were recovered; all from the area of release. Two of these were recaptured a second time. Of the 9 fish taken before February 1, few, if any, showed signif- icant growth (Table 8). The 4 recovered on February 21 and 27 displayed a definite increase in size, indicating a resumption of growth dur- ing this month in the area tested and for the year 1954. One other marked fish caught in February and 2 in early March had regenerated the fin other than the caudal fin which had been cut and could not be distinguished from at least 1 other fish of their respective caudal fin cate- gories. Although there is some indication of a rela- tionship between increasing water temperature and the onset of growth of these manini, the temperature data are inadequate to establish any definite correlation. Temperatures were taken in the canal at night between 9:30 and 12:00 P.M. about 3 in. beneath the surface (Table 9) . It should be emphazied that the temperature at the mouth of the Ala Wai Canal probably varies much more in a day or from day to day than the sea surface temperature off Oahu (mean water temperatures for Honolulu in 1955 are given in Hydrographic Office, 280, as follows: January 24.0; February 23.6; March 24.1). The TABLE 9 Temperature at the Mouth of the Ala Wai Canal DATE TEMP. (° C.) Jan. 5 22.8 Jan. 20 23.2 Feb. 1 23.1 Feb. 27 25.3 Mar. 2 24.7 Mar. 21 24.9 Ala Wai area is not large and the water is there- fore more subject to change due to variation in insolation and atmospheric temperature than the open sea. Probably more important are the changes produced due to variation in the run off of water from Palolo and Manoa valleys which drains to the canal and to the ever variable tides which affect this estuarine region. In order to provide adequate data for the determination of a correlation, temperatures should have been taken at regular intervals each day and an anal- ysis of growth made in terms of average daily water temperatures. Growth of Tagged Adults The first tag used on the manini was the At- kins type (Rounsefell and Kask, 1945: 330, fig. 1 ) . It consisted of a single rectangular piece of white cellulose acetate, 5 X 28 mm. in size, and was attached through the back of the fish beneath the posterior part of the dorsal fin with 4-lb. test monofilament nylon. Two small adult TABLE 8 Recovery Size of Juveniles of Acanthurus triostegus sandvicensis Marked on January 5-6, 1954 STANDARD LENGTH WHEN MARKED (mm.) STANDARD LENGTH WHEN RECAPTURED Jan. 20 Jan. 26 Feb. 1 Feb.. 21 Feb. 27 40.0 41.0 46.5 47.0 49.0 54.5 51.5 57.0 55.0 55.5 60.0 55.5 55.0 56.0 56.5 64.0 64.0 66.5 66.0 70.5 70.5 72.0 72.0 79.0 256 PACIFIC SCIENCE, VoL XV, April 1961 manini were tagged and held in a tank at Coco- nut Island. Within 3 weeks the tags were no longer present on the fish. They were recovered from the bottom of the tank with the loops of nylon intact, thus indicating that the nylon had pulled through the flesh. The same type of tag was then threaded through the upper part of the hypural plate of the 2 fish. As they swam, the tag of each oscil- lated slightly as it was drawn along in a position just above the caudal fin. Three months later both tags were in place, and the method looked auspicious. On March 3, 1953, while operating from the fishing vessel "Friendly Isle’’ owned by Samuel Kaolulo, 50 adult manini were tagged in this manner. Two fish were recovered 5 days later, both in the vicinity of the point of release. No other recoveries were reported. The two Coconut Island manini lost their tags be- fore 6 months elapsed, and further plans for tagging by this method were abandoned. Dis- appointing results with Atkins tags have been reported by other authors, such as Fraser (1955). Petersen disk tags were then tried. Since stainless steel seemed most promising among the Fig. 20. Petersen disk tag used in tagging Acan- thurus triostegus sandvicensis on Oahu. The ends of the monofilament nylon connecting the disks have been melted into knobs with a cigarette lighter. The heating caused a blackening of the one knob which is visible. metals commonly used to attach these tags ( Cal- houn, Fry, and Hughes, 1951), this material was chosen for initial trials. Considerable dif- ficulty was experienced, however, in bending the free ends of the wire on a struggling fish, a dis- advantage clearly pointed out by Calhoun et al. Attention was then shifted to nonmetallic sub- stances, of which nylon seemed the most propi- tious. It is superior to metal in its pliability, non- toxicity, and in not being subject to electrolytic action or fatigue. In order to keep the disks in place, the ends were melted into knobs with a flame (Fig. 20) . The position of the tag on the fish is shown in Figure 21. (For a detailed dis- cussion of this tagging method, see Randall, 195 6a.) On August 6, 1953, 3 captive adult manini were tagged with plastic disks by 60-lb. test monofilament nylon (diameter, 1 mm.) and 1 with stainless steel wire. Seven weeks after tagging, all 4 fish, along with 6 untagged fish which were measured at the time the 4 fish were tagged, were caught and remeasured. The 6 untagged fish, which had a mean standard length of 86.3 mm., had grown an average of 6.6 mm. The 3 tagged with nylon (mean standard length, 84.4 mm.) grew an av- erage of 7.7 mm. The 93 mm. fish whose disks were attached with stainless steel wire added the smallest increment of growth, 2.5 mm. Al- though there are insufficient data for definite conclusions, it seems apparent that manini tagged with Petersen disk tags fastened with monofilament nylon are capable of normal growth. One of the fish which survived 11 months until July when the tank was needed for other purposes still retained its disks fastened with nylon. It was noted that the tagging wound was small, and there was no evidence of deteriora- tion of the nylon. Since the experimental tagging demonstrated no obvious fault in this method, field tagging was begun. Forty-five manini were tagged with hand-made tags. This proved to be most time consuming, and assistance was requested of and granted by the Hawaii Division of Fish and Game. Manufactured tags of laminated cellulose acetate (Fig. 20) were purchased. In order to publicize the tagging, posters (Fig. 21) were Acanthurus triostegus sandvicensis — Randall REWARD ! IF A MANINI IS CAUGHT WHICH IS TAGGED AS SHOWN IN THE ABOVE DRAWING, PLEASE CONTACT FISH 8 GAME WARDEN OR BRING THE FISH TO THE DIVISION OF FISH 8 GAME (PHONE 92071). ONE DOLLAR ( 8 1.00) REWARD WILL BE GIVEN. Fig. 21. Poster used to publicize tagging of Acan- thurus triostegus sandvicensis on Oahu. distributed by game wardens. Carl M. Nemoto, then working for the Division of Fish and Game, tagged a total of 1,157 manini, 70 to 152 mm. in standard length, from July 6 to November 23 , 1954. Forty-five additional fish were tagged on May 4 and June 6, 1955. The tagging was carried out from Kaolulo’s vessel between Ke- walo Basin and Koko Head, Oahu. The manini were caught in traps, tagged, measured, and re- leased at the place of capture, which was re- corded. Due to the difficulty of using a flame when tagging from a vessel exposed to the strong winds of the Kaiwi Channel (Molokai Chan- nel), Nemoto decided to tag most of the fish by crimping brass leader sleeve on one end of the nylon (the other end was melted into a knob prior to the departure of the vessel ) . This method was devised by Kenji Ego of the Divi- sion of Fish and Game. Of the total of 1,247 manini tagged with disk tags, there have been 281 recoveries. Of the recovered fish, 173 have been returned alive, pro- 257 viding the opportunity for re-release. Twenty- four of these have been caught again, and 3 have been taken a third time (these recoveries are included in the 281 mentioned above). The following table shows the lapse of time by 2 -week periods between the release and re- covery of tagged manini. The fish with the longest recovery time was tagged on September 21, 1954, and recovered on July 29, 1955. The explanation of the large number of ma- nini recovered in the first few weeks following release and the sharp drop in returns thereafter probably involves a number of factors, such as mortality, loss of tags, and movement from the immediate area of release. It is known that at least some loss of tags has occurred, for manini with the tagging wound but without tags have been noticed by fisher- men. Since no fish have been seen with a tag in the process of moving up through the back and dorsal fin, it is assumed that most of the early losses, at least, occurred from a dropping off of a disk because of the knob being too small or the brass inadequately crimped. Analysis of the returned tags indicated that the knobs were too small in many cases. When the diameter of the hole in the disk is 1 mm., the knob should measure at least 2 mm. TABLE 10 Recovery Time of Tagged Adults of Acanthurus triostegus sandvicensis WEEKS FOLLOWING RELEASE OF FISH NO. OF FISH RECOVERED 2 96 4 57 6 42 8 25 10 17 12 9 14 6 16 5 18 4 20 4 22 3 24 5 26 2 28 2 30 2 34 1 40 1 258 PACIFIC SCIENCE, Vol. XV, April 1961 Movement from the trap area could account for a diminution in recoveries of tagged fish; however, evidence is strong against at least ex- tensive migration by this species (see section on migration ) . The marked drop in number of recoveries still seems too high to be accounted by any or all of the above reasons. One other possibility exists which is supported by the extraordinary high percentage of early returns. A tagged manini may show a greater predilection to re-enter a trap than untagged manini to enter a first time. This phenomenon has been conclusively demon- strated by Lawrence ( 1952 ) for bluegill sun- fish in Iowa farm ponds. If such a tendency exists, it might be expected to be greatest in the fish soon after release and gradually abate with time. One manini was caught, tagged, re- leased near the next trap, and caught again in this trap a few minutes later. The 173 tagged fish which were returned alive were measured alive a second time by Ne- moto and released. The remaining tagged ma- nini were measured after they had died, and usually not when fresh but following partial drying and/or freezing. Seventeen of the latter group were recovered within the first 2 weeks following tagging. These fish were all from 1 to 7 mm. shorter than when first measured. The average shrinkage was 3.53 mm. The average standard length of these specimens measured to the peduncular mark is 109.8 mm. The average shrinkage per millimeter of length is therefore 0.032 mm. A correction was made for each fish which was measured dead (unless freshly dead) by multiplying its recovery length by this fac- tor and adding the product to the length. A total of 70 tagged manini were measured alive a second time following recovery within 2 weeks of tagging. Surprisingly, 46 of these were from 1 to 5 mm. shorter than when first measured; 10 were the same size; 14 were from 1 to 3 mm. larger. The average "growth” was — 1.06 mm. A t test applied to these data to as- certain the probability of obtaining such a dis- tribution from a population in which no actual change in length occurred resulted in a / value of 5.1. With 69 degrees of freedom this gives a probability of less than 0.001 of getting a higher t value; thus the null hypothesis of no significant difference is rejected. Since the same person measured the fish both times, the reason for the significantly shorter length is not apt to be any bias in measurement. The shorter length is probably due to shrinkage from starvation in the traps. Manini which are caught in traps may have been held in the traps for a week or more. Although both measurements are made on fish caught in traps, the effects of confinement would be expected to be greater on fish spending an- other period in traps within 2 weeks of their first. In order to test the hypothesis that actual shrinkage occurs during starvation, 17 trap- caught adult manini (none showed evidence of long sojourn in traps) , 112 to 131 mm. in stand- ard length, were tagged, measured, placed in two large aquaria, and allowed to starve. Eleven days later the 5 fish which survived were re- measured. A definite shrinkage of from 1 to 5 mm. (mean 3.4 mm.) had taken place. The measurements were made by Nemo to. The large amount of variation ( — 5 to +3 mm.) in the series of differences between the measurements made on the 70 fish when tagged and recovered within 2 weeks was unexpected. When it is realized that five different sources contribute to this variability, it is more com- prehensible. There is the possible error in the first measurement of a living fish. The second measurement of this same fish constitutes an independent source of error. The shrinkage due to starvation is a third. Some manini which are caught have not been in a trap more than a few hours and will not have shortened at all; others held prisoner for a week or more may have shrunk several millimeters. Some manini initially caught after a very short stay in a trap, tagged with a minimum of injury, and recovered a full 2 weeks later after another brief period in a trap might have grown about a millimeter; thus growth itself probably forms a fourth source. An important fifth source of error in the meas- urement of the dead tagged fish is the variation in shrinkage after death. All of this variability has necessitated a statistical approach to the analysis of the data, and for this purpose the large number of recoveries has been most wel- come. Only those manini recovered after spending Acanthurus triostegus sandvicensis — Randall 259 Fig. 22. Growth of Acanthurus triostegus sandvicensis on Oahu during the period Jul. 6 to Nov. 19, 1954. The points represent growth increment of tagged fish recovered after at least 1 month in the water. Circles indicate the superposition of one point on another. The dotted line is the regression of growth on standard length. a month or more in the water following being tagged were utilized for growth study. In Fig- ure 22 the data from 68 such fish, tagged be- tween July 6, 1954, and September 28, 1954, are plotted. The dotted line represents the re- gression of millimeters of growth per month on standard length. The regression coefficient "b” (slope of the regression line) is — 0.0292. The t value testing whether this regression differs significantly from zero is 2.085. With 66 degrees of freedom, the probability of getting a larger value of t is 0.04; thus the null hypothesis of no significant regression is rejected. The comparison of the regression line with the upper part of the growth curve of Figure 18 is favorable. The average growth rate of the captive fish from August 6 to September 27 was given as 1.8 mm. per month. The average rate for this size range (90.0 to 94.3 mm.) from the regression line is 1.74 mm. per month. In order to further confirm the finding of ces- sation of growth in winter, 155 manini were tagged on November 19 and November 23, 1954, by Nemoto. Twenty-five of these fish were recovered from December 23, 1954, to Febru- ary 18, 1955 (see Fig. 23). Their average rate of growth is — 0.112 mm. It therefore seems likely that no growth occurred during this period. Tagged manini, recovered after a month or more in the sea, which were at liberty during winter and either fall or spring months have not yet been considered. Data from these fish are presented in Table 11. If it is assumed that these 22 fish did not grow during any part or all of the 3 -month period beginning November 19, and the growth rate of each is computed only from remaining months, an average growth rate of 1.8 mm. per month is obtained. This compares favorably with the rate of growth as indicated in Figure 22 (the average standard length in Table 11 is 110.6 mm.) . AGE In view of the lack of winter growth of the manini it was hoped that annuli might be de- tected on the scales; however, none could be seen. Even the circuli of the tiny scales of this species are not very distinct. Cross-sections of vertebrae and otoliths were also examined, but with negative results. It was necessary to grind otoliths to about 1 mm. thickness and view them with subdued light under a compound micro- scope before any concentric rings could be seen. These were numerous and variable, and none were more sharply defined or grouped closer together than others. 260 PACIFIC SCIENCE, Vol. XV, April 1961 TABLE 11 Growth of Acanthurus triostegus sandvicensis Recovered After 1 Month or More Following Tagging (Period of tag retention includes both winter and fall or spring months, 1954-55) DATE RELEASED DATE OF RECOVERY STANDARD LENGTH MEASURED DEAD GROWTH (mm.) Tul. 6 Dec. 14 105 X 3.4 23 Mar. 25 110 3.5 30 Dec. 23 115.5 5.5 Aug. 10 Feb. 6 113 X 7.8 10 Feb. 22 110 3.0 13 Jan. 28 98 X 3.1 27 May 17 97 8.0 Sep. 3 Feb. 6 101 X 7.2 3 Mar. 25 102 3.0 21 Jan. 24 98.5 4.0 21 Feb. 18 123 X 7.0 21 Feb. 21 105 X 7.4 21 Jul. 29 101 4.0 28 Dec. 23 121 X 1.8 28 Jan. 21 114 2.0 28 Mar. 25 122 X 1.8 28 Apr. 23 124 1.0 Nov. 19 Mar. 1 105 X 3.8 19 Apr. 5 152 1.5 23 Mar. 19 119 0.5 23 Apr. 27 105 3.0 23 Jul. 6 106.5 7.5 The next approach to obtaining the age of manini was the preparation of the length fre- quency curve of Figure 24. As the recruitment of the manini is seasonal, it seemed possible that year classes might stand out as modes on such a graph. At least for this curve, however, no definite modes are apparent. Probably the long period of recruitment (February to Octo- ber), coupled with individual variation in growth, precludes the formation of distinct modal groups in standard length. Age determinations of the manini would therefore seem possible only by rearing of fish, recovery of tagged fish (tagged at a size small enough to predict the age from growth rate with reasonable accuracy), or gross estimates from growth rate and knowledge of the maximum size. Although the age attained by the manini is unknown, the following information on the age of two surgeonfishes supplied by Spencer Tinker, director of the Waikiki Aquarium, indicates that at least some acanthurids are long-lived. Two unicorn fish ( Naso unicornis ) which were present in the aquarium when an employee ar- rived in 1935, lived until they succumbed as the result of an accident in 1955. In 1940 several specimens of Acanthurus xanthopterus approxi- mately 5-6 in. in length were placed in the aquarium. Two of these survived until 1951 in a small tank together; they had reached a size of about 20 in. At this time one slashed the _ 2 e 0 . : Li * o o» # • • # • _ -2 _£ . ... , — 2 100 104 108 112 116 120 124 128 132 136 140 144 Standard length (mmj Fig. 23= Growth of Acanthurus triostegus sandvi- censis on Oahu during the period Nov. 19, 1954, to Feb. 18, 1955, as based on the recovery of tagged fish which were in the water 1 month or more. The hori- zontal line was added at the zero point to facilitate comparison of the distribution of points with Figure 22 . Acanthurus triostegus sandvicensis — RANDALL 261 Fig. 24. Length frequency distribution of Acanthurus triostegus sandvicensis. Based on 426 specimens caught in traps from Kewalo Basin to Koko Head, Oahu, from Apr. 2 to 9, 1954. other with its caudal spine and the latter ulti- mately died. The remaining fish died in 1955 because of an accident. BEHAVIOR Migration Small juvenile manini tend to remain in the tide pool or shallow water area to which they first come from the pelagic state. Some evidence for this has already been presented. Of 176 ju- veniles taken in the weekly collections during a season from the tide pool of Figure 4, only 19 were 30 mm. or more in standard length and hence definite migrants to the pool from other areas. Since all of the fish in this pool were killed with rotenone each week, a much greater movement to such a large unoccupied area would be expected from the smaller adjacent pools if the species did not exhibit a tendency to remain in one location. When the adjacent pools were sampled for manini, they invariably contained many more specimens 30 mm. or larger, than less than 30 mm. It should be emphasized that these pools are isolated from one another for only about an hour or less during the infrequent low tides less than 0.0 ft. Even in areas like the shore of the Ala Wai Canal and the Ala Wai Yacht Basin where adjacent areas are freely available at all times and essentially identical in cover and food supply, appreciable movement does not take place. This is attested by the suc- cess in recovery of marked fish at the mouth of the Ala Wai Canal (Fig. 19, Table 8). Failure to take more of these fish was not due to their movement out of the area but to the difficulty in catching them. Less than one-fourth of the manini sighted were caught, and probably many others were hidden from view under large rocks, etc. None of the fish were taken farther than 20 ft. from the point of release, although the shore was searched as much as 100 ft. away. On March 7, 1953, a juvenile manini, esti- mated 36 mm. in standard length, was seen next to the sea wall of the Ala Wai Yacht Basin be- side a pier. It was probably among the first manini of the season to come into the area, and was the only individual this size visible for over 100 ft. of sea wall on either side of the pier. On the 9 different days up to March 28 when at- tempts were made to find this fish, it was seen. Its greatest movement from the place where first observed was only 12 ft. On March 28 it was caught (measured as 41 mm.), and the posterior part of the soft portion of the dorsal fin removed in order to identify it with greater certainty in further observations. It was found on 10 dif- ferent days up to May 2, but never more than 12 ft. to either side of the pier. On April 24 the dorsal fin had almost completely regenerated and by May 2 the fin appeared normal. As juvenile manini grow, they migrate sea- ward from the tide-pool zone (in the harbor area, as mentioned, the larger fish remain in shallow water because of the lack of cover and 262 PACIFIC SCIENCE, Vol. XV, April 1961 Fig. 25. Southeastern coast of Oahu, Hawaiian Islands (after U. S. Coast and Geodetic Survey Chart No. 4110). the lack of algae on the mud bottom away from shore). This movement is demonstrated in the graphs of Figure 6. The relative size of any one modal group is progressively smaller in subse- quent collections (predation would have to be extremely heavy to achieve the same diminu- tion). The one seeming exception is the large collection of May 29. This, however, was made from a very large pool at Diamond Head which averages about 4 ft. in depth. A higher per- centage of larger manini were expected from this collection than previous collections which were all made in pools less than 2 ft. deep. Information on movements of adult manini was obtained from the recovery of tagged fish. Unfortunately, exact locality data for the areas of release and recovery were usually not avail- able. Since Nemoto was operating from a com- mercial vessel solely by virtue of the kindness of Kaolulo, he could not delay the fishing to take bearings or soundings. Most locality data from fishermen who recovered tagged manini were also very nonspecific. As a result, little is known of local movements of adults, and only migrations from one major area to the next or the lack of such migrations could be ascertained. Kaolulo’s traps are arranged in groups, most of which are well separated. Nemoto designated these areas as offshore from the following: Yacht Basin, Royal Hawaiian Hotel (Waikiki), Natatorium (near Kapiolani Park), Diamond Head, Black Point, Kahala, KULA antenna, Wailupe, Aina Haina, Kuliouou, Niu Peninsula, Portlock, and Koko Head ( Fig. 25 ) . Of the 281 recoveries of tagged manini 237 were made by Kaolulo; 126 tagged fish, most of which were early recoveries by Kaolulo, were returned to the Division of Fish and Game with no record of where they were caught. This dif- ficulty arose because of his resorting to memory alone to recall the area where tagged fish were taken. When he caught several fish he usually admitted that he could not be sure where any of them were caught. He was finally persuaded to keep record of locality by tag number. One hundred and thirty-two tagged manini were recovered which exhibited no definite mi- gration. Ignoring fish released in Kewalo Basin for the moment, this leaves only 11 records which indicate that some migration might have occurred. Five of these were fish which were displaced from their home area by movement of Acanthmus triostegus sandvicensis — RANDALL the vessel to the next area and which apparently returned to the home area. One was caught at Wailupe, released at Black Point, and caught again at Wailupe. The same fish was not released the second time until the Kuliouou area was entered. It was caught a third time back at Wai- lupe. Another Wailupe manini was released op- posite the KULA antenna and was recovered at Wailupe. Two fish were caught at Black Point, released farther offshore at Kahala, and caught again at Black Point. The remaining 6 records are as follows: 1 fish supposedly migrated from Kuliouou to Wailupe; another went from the KULA antenna to Niu Valley; 2 apparently went from Kuliouou to Wailupe; the last 2 swam from Black Point to Kuliouou. The possibility of error must be kept foremost in mind for all of these apparent migrations. The 2 manini which moved from Black Point to Kuliouou were both caught by Kaolulo on the same day, which seems unusual. In view of the manner in which the locality data have been reported, it is surprising that more alleged migrations have not taken place. It is therefore concluded that extensive migra- tions of this surgeonfish probably do not nor- mally occur. Of the 172 tagged manini which were released a second time 125 were set free on Kewalo Basin. Twelve of these have been recovered. One was taken between piers 7 and 8 in Hono- lulu Harbor. The rest were all captured either at Kewalo or in the Diamond Head direction. One was taken from a tuna boat in the basin, and 2 off the Kewalo breakwater. One locality reads Ala Moana and another Ala Wai. One manini was caught by Kaolulo with no record of locality. Since his nearest trap is off the Yacht Basin, this fish must have migrated at least this far. One manini released at Kewalo was caught at a depth of 30 ft. directly off the Royal Hawaiian Hotel. A fish from Black Point was caught 6 months later at wading depth off Diamond Head not far from Black Point. The most interesting recoveries of Kewalo-released fish are 2 from Kuliouou which were originally captured at Kuliouou and 1 at Wailupe which was first caught off Wailupe. Although, again, the possibility of error must be kept in mind, it seems unlikely that the only 3 examples of ex- 263 tensive migration (about 8 mi.), all of which occurred among the 12 fish recovered following release at Kewalo Basin, could be mistakes. One of the Kuliouou manini was recorded from this area by tag number by Kaolulo. When ques- tioned of any chance of error, he insisted there could be no mistake. An interval of 3M> months elapsed between its release in Kewalo and its final capture. The other Kuliouou fish was ob- tained from the Young Market in Honolulu. The interval between the release and capture of this fish was 16 days. The Wailupe fish was re- covered by Kaolulo 1 month after release in Ke- walo Basin. The locality was reported verbally. The intriguing interpretation of these 3 records of long migration is that at least some manini removed from their home area will continue to wander on the reef until they return. Three records are too few on which to base such a con- clusion; therefore more data should be obtained by purposely displacing tagged manini consider- able distances from the original area of capture. Aggregations Adult manini are often observed in large feed- ing aggregations which slowly move over the bottom. These groups of fish may be dense and cover areas more than 50 ft. in diameter. What appears to be the same school was repeatedly seen in the same general location at the north- western end of Manana Island. This group of fish was observed in the area in fall as well as during the breeding season. The behavior of in- dividual manini of the school did not seem to differ from that of solitary adults. One school of about 300 adult Acanthmus triostegus triostegus was observed in the north- east side of the pass at Takaroa, Tuamotu Archi- pelago, over a period of a week. The school moved as a close-knit mass several feet off the bottom, stopping frequently for periods of sev- eral minutes to graze algae. One or two Aulos- tomus chinensis were observed to accompany the school, and the manini were almost constantly harassed by Acanthurus glaucopareius which chased individual fish. The school moved as much as 300 yd. in several hours. Juvenile manini are gregarious and are often seen swimming in groups. One group of 1 1 in- dividuals, all less than about 36 mm. in standard 264 length, was observed for a period of half an hour at the shore of the Ala Wai Yacht Basin. No fish became separated by more than 12 in. from the rest of the group. The schooling of A. triostegus triostegus dur- ing spawning has been discussed previously. Reaction to Danger Juvenile manini along the shore of a part of the Ala Wai Canal infrequently visited by man perceive the approach of a person walking along the bank from an average distance of about 20 ft. This distance is contingent on the position of the sun and the depth and turbidity of the water. The usual reaction is a swift darting to some immediate hiding place. If an observer remains immobile, the fish come out of hiding in from about 15 to 25 sec. and resume feeding shortly thereafter. Manini of the same size next to the sea wall of the Ala Wai Yacht Basin, where people pass almost continuously, do not interrupt their feeding at the approach of a per- son unless some unusual sharp movement is made, at which time they may hide. The manini in a large tide pool at Diamond Head were observed to have two hiding places, each beneath ledges at the side of the pool. One of these was preferred, and most of the fish swam to it when the pool was approached. When rotenone was spread in the region of this retreat, all of the fish swam rapidly in a direct line to the second ledge, in spite of the turbid- ity of the water (from the rotenone), even though they had to move in the direction of the observer and pass over a very shallow part of the pool. The part of the pool away from the observer was broader, deeper, free of rotenone at the time, but without good cover. It was evident that the fish were well aware of the lo- cation of the best places to seek refuge. An experiment was designed to determine the time required by juvenile manini to recog- nize and accept a new site of cover. A small opaque plastic bowl was inverted and placed with one edge elevated from the bottom of a 23 gal. aquarium in which 12 manini, 26 to 39 mm. in standard length, had been kept without a place of shelter for 3 days. Initially this bowl was avoided by all the fish. Fifteen min. elapsed PACIFIC SCIENCE, Vol. XV, April 1961 before any fish swam near it. One hr. after the bowl had been added to the tank, the manini were frightened by rapid arm movements all around the aquarium; however none sought ref- uge in the new shelter. Upon being similarly frightened 3 hr. later, 2 of the smallest fish went into the bowl. Three hr. after this, all of the manini swam to the bowl for cover. Adult manini are less inclined to seek shelter. Their reaction to an approaching swimmer usu- ally takes the form of retreat. When cover is sought, the hole into which the fish swim usu- ally has two or more entrances. As has been pointed out by Breder (1949^: 97) a reaction to attack which alternates between flight and hiding is common among fishes. Righting Juvenile manini may often be seen fighting among themselves in the natural environment. Also, several days after being placed in an aquar- ium, this behavior becomes noticeable. It in- creases with time to the extent that it interferes with feeding. After 2 weeks in an aquarium, fully half the diurnal time of 12 juveniles was spent at this activity. The fighting was not as- sociated with territoriality, but took place among all the fish and in all parts of the tank. Three of the larger fish were consistently the aggres- sors, and the smaller fish usually bore the brunt of their attack, but no definite peck order was established. When only 2 fish were present, however, the larger invariably dominated the smaller and forced it to remain in a corner of the aquarium for much of the day. A distinct color change, which results in an almost complete reversal of the usual color pat- tern, accompanies the fighting. The broad pale interspaces between the vertical black bars be- come very dark, and the upper seven-eighths of the bars change to yellowish white except for a fine dark line in the center of each. On the lower eighth of the body the black bars are faintly visible in their normal width. The more aggressive fish are darker than the less aggres- sive ones. The fighting usually involves little actual con- tact between fish. Most of the time a fish under Acanthurus triostegus sandvicensis — Randall attack hastily withdraws, a response which gen- erally satisfies the attacker for the moment. When there is no immediate retreat the two fish circle at close quarters and make pugnacious movements, mostly with the posterior part of the body. When a new manini was placed in the tank, it was attacked with ferocity by all of the other manini, but rarely more than one at a time. The hapless fish was pecked over the head and body, strongly bumped, and lashed at with the caudal spines of the others as it scurried from one part of the tank to the next. A manini twice the size of any in the tank was given the same treatment. After about half hour of inces- sant attack, this oppressed fish was observed to lie on its side on the bottom and submit com- pletely to the relentless abuse. It was caught and examined shortly thereafter. Only a few small scratches on the fins and two small cuts on the body could be seen. The manini has a much smaller caudal spine than other species of Acan- thurus and is not capable of inflicting much damage with it. Also the jaws do not seem strong enough to injure other fishes of equal size. Manini were never observed to fight with other species of fishes. They appear to display a dominant behavior pattern to at least some, how- ever. Six small pomacentrids ( Abudefduf ab- dominalis ) ranging from an estimated 18 to 38 mm. in standard length were observed to inter- mingle with 4 juvenile manini about 28 to 36 mm. in standard length at the shore of the Ala Wai Yacht Basin. Although the pomacentrids swam close to one another, they usually avoided coming within a body length of any manini. When one did and this was observed by a ma- nini, the latter responded by a slow sideward movement of the body toward the intruding fish. The posterior half of the manini’s body was curved toward the pomacentrid and vibrated slightly. The pomacentrid invariably retreated a short distance. Breder (1948: 293) observed similar behav- ior in an aquarium with A. chirm gus. He noted that a peck order was established among certain fishes in an aquarium. This involved 4 pomacen- trids and 1 gerrid. A small specimen of A. chirurgus was added which was about equal in size to the largest pomacentrid. Although the surgeonfish was not bothered much by the other 265 fish at first, it seemed to be low or lowest on the peck order. It was never noted to pursue the others, but was occasionally pecked at by the others. Fourteen days later the peck order was modified and the acanthurid was definitely num- ber one. There was little fighting but all kept clear of the surgeonfish which "showed” its peduncular spine to the more timorous. I have never observed fighting among adult manini. Larger individuals may display dom- inance over smaller ones, however. Behavior at Night At night, both in the natural habitat and in the aquarium, manini rest on the bottom, usually with the pelvic fins and anal fin touching the substratum, in what appears to be a state of torpor or sleep. In general the body color is darker than during the day. Large dark blotches which may be seen faintly during the day be- tween the vertical dark bars are more conspicu- ous at night. This resting phase is not without movement, for the pectoral fins are almost in constant slow motion and there are vertical un- dulations of the median fins. When a light was flashed on a darkened aquarium at night, short movements over the bottom were occasionally found to be in progress. These were directed backward more often than forward. Usually the manini in an aquarium or a tide pool at night are well dispersed. In the normal environment most of the fish choose a sheltered although not a confining location. Many, how- ever, may be seen in very exposed locations. Breder (1948: 294) observed that A. chirurgus rested on the bottom of an aquarium at night in the proximity of some shells, but in no case did it enter or hide in the shells. If light is turned on a sleeping manini at night, and no sharp movements are made, the fish can be lightly touched or even picked up without its making any effort to escape. If a sharp movement is made in the water of the aquarium, or if the fish is roughly handled, it will dart away for a considerable distance over the bottom. This behavior was put to good use for the collection of this species alive. Using a head lamp at night to locate a sleeping fish, a dip net was cautiously placed in front of the 266 quarry. A long stick was used to prod the pos- terior end of the fish which often responded by swimming into the net. The following observations on sleep were made on 10 juveniles, 27 to 40 mm. in standard length, which had been maintained in an aquar- ium for 12 days. With the onset of darkness the fish settled to the bottom of the tank, although they con- tinued to move about, and occasionally these movements would take them above the bottom. After total darkness ensued, 25 min. were re- quired before they entered the state of suspended animation completely. This state was ascertained by lightly stroking the fish with a glass rod. When they ceased to flee from such contact, it was assumed that they were asleep. Since the period of attaining sleep was prolonged by the interruptions, a series of observations on succes- sive nights was necessary to determine the nor- mal period. The stimulus to sleep was shown to be the onset of darkness and not merely the passage of a definite number of hours of diurnal exist- ence, for the manini remained active for many hours when a light was kept burning above the tank after darkness set in. In one experiment the fish were kept awake for a period of 24 hr. They remained active during the time; however they ceased to feed toward the end of this pe- riod and displayed "restless” swimming. When the fish were kept in total darkness for 24 hr., they remained relatively quiet on the bottom during the entire period. At the end of this period, however, it was noticed that they perceived the immersion of a hand into the tank regardless of how slowly it was inserted. After 3 hr. of uninterrupted sleep, beginning with onset of darkness, lights were turned on to see how long the state of torpor would per- sist. In 13 to 14 min. the first 3 fish awakened and began to swim around the aquarium. These were the largest of the 10 manini. The remain- ing fish awakened according to size; the 2 smallest required 25 and 28 min., respectively. All of the fish displayed very restless swimming, moving constantly back and forth in the aquar- ium. When the lights were turned off, they re- quired 39 min. to re-enter the state of torpor. PACIFIC SCIENCE, Vol. XV, April 1961 No distinction could be detected in the time needed for the different-size fish to go back to sleep. SUMMARY 1. In the Hawaiian Islands the common sur- geonfish, Acanthurus triostegus sandvicensis (known in Polynesia as the manini), is found in many different habitats, although never far from coral or rock for shelter and algae for food. The young are abundant in tide pools. Manini are not numerous in regions of heavy coral cover, although certain other surgeonfishes are surprisingly common there. 2. The eggs and larvae of the manini are pe- lagic. Eleven acanthurid larvae, 4.2 to 8.7 mm. in length, were taken in eight plankton tows 10-140 mi. from the nearest Hawaiian island. All were from the upper 50 m., although more tows were made in from 100 to 300 m. A higher percentage of acanthurid postlarvae have been found in surface-caught yellowfin tuna than those taken with long line at depths of about 30 to 160 m. Acanthurid larvae are more com- monly found in the stomachs of tuna caught inshore than offshore. 3. The limits of temperature tolerated for 1 hr. by manini (both young and adults) appear to be about 12° and 37° C. Chlorinity tolerance for 24 hr. approximates 1.4 %o to 37 %c. The most extreme readings of temperature found on Oahu in waters inhabited by manini were 16.2° and 35.1° C. and of chlorinity, 2.65 %c and 21.3 %o. 4. Apart from the postlarval acanthurids found in tunas nothing is known of predation on the pelagic stages of the manini. In Hawaii, juve- nile manini probably constitute prey to carangid, muraenid, synodid, scorpaenid, sphyraenid, cir- rhitid, holocentrid, and antennariid fishes, and possibly certain large crabs. In view of the pau- city of large piscivorous fishes in Hawaiian wa- ters adult manini are probably relatively free of predation ( man excluded ) . 5. The manini was found to be parasitized by one species of hydroid, four trematodes, four nematodes, two leeches, and six copepods. 6. Difficulty was experienced at one time maintaining manini in aquaria. They died in Acanthurus triostegus sandvicensis — Randall about 48 hr., displaying large sores, eroded fins, and excessive mucus on the body. The dominant bacterium was a short, gram-negative rod. 7. The stomachs of 3 postlarval manini 6.7 to 7 mm. in length were examined for food. Two were empty, and 1 contained two appendicular- ians and a larval polychaete. Of 57 manini in the acronurus (late postlarval) stage taken by night light offshore, 45 were empty, and the rest contained only occasional fish scales or crusta- cean fragments lodged in intestinal folds, thus suggesting that the acronuri do not feed at night. The guts of 4 unidentified acronuri taken from a skipjack stomach caught during the day were filled with zooplankton. 8. During the first day of transformation from the acronurus to the juvenile state, most manini do not feed at all. Feeding on algae becomes pro- gressively heavier as transformation continues. Juvenile and adult manini feed almost entirely on relatively fine filamentous algae of numerous species. Little inorganic sediment was seen in the gut contents. When 28 different kinds of algae plus diatoms were offered to manini in feeding experiments, a red alga ( Polysiphonia sp. ) and a green ( Enteromorpha sp. ) were pre- ferred. All of the blue-greens, the red Asparag- opsis , and one species of the brown alga Ecto- carpus were not ingested (although these may be found in stomach contents, mixed with other algae ) . 9. Juvenile manini were maintained in aquaria on algae alone. Fish 26 to 31 mm. in standard length grew at the rate of about 6 mm. per month (temperature 23.0° to 23.8° C.) when fed either Polysiphonia sp. or Enteromorpha sp. 10. Manini feed almost constantly during the day, and a large volume of algae is passed through the gut. Four juveniles weighing a total of 8.3 grams consumed 10.8 grams of Entero- morpha in 1 day and 8.8 the next. Two adults weighing 83.2 and 94.5 grams ate a total of 27.4 grams of Enteromorpha in 1 day. The adult fish did not feed as frequently in aquaria as in the natural environment. The larger fish dom- inated the smaller one and consumed the major part of the algae. 11. Manini showed no response to extracts of algae, and the visual sense alone is needed to locate algal food. The olfactory sense appears to 267 function in feeding only to avoid unsavory mate- rial. Chemoreceptor organs in the mouth and possibly tactile organs as well seem to be as- sociated with the acceptance of algae and the rejection of material not utilized as food. 12. The morphology of the digestive system is briefly discussed. The alimentary tract in- creases nearly three-fold during the 4 or 5 days of transformation from the acronurus to the ju- venile form, reflecting a change in food habits from feeding on zooplankton to feeding on al- gae. The length of the gut increases from about three times the standard length in small juve- niles to nearly six times in large adults. 13. Algae passed through the gut of juvenile manini in 1 hr. 45 min. and in about 2 hr. in adults. At night when feeding ceases, the gut is emptied slowly. 14. Although sustaining growth, the Entero- morpha appeared little affected after passing through the gut of manini. Additional work is needed to determine what is utilized by the ma- nini from its algal food. 15. The stomach of the manini varies from slightly acidic to slightly alkaline. The intestine is alkaline (pFl 7.7 to 9.1) and the gall bladder acidic (pH 6.2 to 6.4). Amylase was detected throughout the diges- tive tract, strongest in the pancreas and weakest in the stomach. The pH optimum of the amy- lase is about 6.7. Maltase was not found. Lipase, with a pH optimum of 7.2, was found in all digestive organs, strongest in the pancreas and pyloric caeca and weakest in the stomach. Proteolytic enzymes are weak, especially from the stomach; pH optimum of pancreatic pro- tease is about 8.4. 16. About two-thirds of the adult manini caught in unbaited traps off Oahu in from 30 to 90 ft. are males. Of 221 manini in museum col- lections from the tropical Pacific mostly taken from inshore waters, 134 (60.7 per cent) are females. 17. Although ripe males may be taken through- out the year ( fewer, however, in the fall ) , ripe female manini have been found in Hawaii only from early December to late July (and trans- forming young from middle February to early October ) . Average monthly sea surface tempera- 268 PACIFIC SCIENCE, Vol. XV, April 1961 tures in the Hawaiian Islands vary from about 75° to 81° F. Enlargement of the gonads ap- pears to coincide with decreasing water tem- perature in the fall. 18. An estimate of about 2 Vi months for the duration of larval life of the manini in Hawaii was made by noting the time between the find- ing of the first ripe females and first transform- ing young of the season and the last ripe fish and last incoming acronuri. 19. Ripe female manini ( Acanthurus trioste- gus triostegus) were found in collections made at islands of the tropical Pacific (where mean sea surface temperatures exceed 80° F. through- out the year) in all months of the year, as were transforming or early juvenile specimens, thus indicating year-round spawning in more equa- torial waters. 20. A lunar periodicity in the number of transforming manini was discovered from tide- pool collections on Oahu, with the peak in spring occurring a few days before new moon. Since a collection of young from the Gilbert Islands, where water temperatures average about 7° F. higher (and hence development is prob- ably more rapid) , indicates peak influx of acron- uri approximately at the time of full moon, the periodicity is not attributed to any lunar effect on the young but to greater spawning by adults at one time of the month than another. Thirty- nine ripe female fish were found among 2,552 adult manini examined from 12 days before to 2 days after full moon during the spawning season and 4 among 1,311 fish seen in the re- maining half of the lunar month. 21. Individual manini appear to spawn more than once during the season. Large and small ripe females were found near the start and end of the season. No completely spent ovaries were observed until July and August. In graphs of egg diameters of mature ovaries, a mode of egg size about intermediate to the diameter of trans- parent ripe eggs and primordial ova was in- variably present during the season. No specimens were found with modes of egg size lying between the intermediate-size mode and that of fertilizable eggs, thus suggesting that ripe eggs develop rapidly from the opaque ova of the intermediate group. 22. Although the spawning of manini was never witnessed in the Hawaiian Islands, A. triostegus triostegus was observed to spawn in the Tuamotu Archipelago. Several hundred fish were seen in a milling aggregation at dusk dur- ing the time of full moon. Many had broader bars on the body and darker fins than normal; 5 of these were speared and proved to be ripe males. Two normally colored manini were fe- males, 1 running ripe. Spawning occurred at the apex of sudden rapid upward movements by small groups of fish. A current carried the eggs to the open sea. 23. The smallest running ripe female manini which was found on Oahu was 101 mm. in standard length; the smallest male, 97 mm. Ma- nini probably do not spawn in their first year. 24. The number of ripe eggs in the ovary of one 123 mm. manini was estimated at 40,000. 25. Early development was studied following artificial fertilization of the eggs. The eggs aver- age 0.67 mm. in diameter, have a single oil globule, and float at the surface. Hatching occurs in 26 hr. at 24° C. As yolk is used up, the larvae show a progressive tendency to sink. They com- bat this by upward swimming movements. Feed- ing begins at the age of 51/2 days. No young were maintained in aquaria more than 6 V 2 days; however, 3 postlarval manini, 6.6 to 7 mm. in total length, could be identified from plankton collections by fin-ray counts. The most striking feature of their morphology is very elongate sec- ond dorsal, second anal, and pelvic spines. 2 6. Manini acronuri come into shoal water to transform only at night. They are discoid, scale- less, and transparent with a silvery abdomen. Their second dorsal, second anal, and pelvic spines are poisonous. During transformation, which requires 4-5 days, adult-type configura- tion and coloration is attained, scales form, and the venomous quality of the spines disappears. 27. Light appears to be essential to transfor- mation. Two acronuri taken at midnight were placed in a darkened 2 l /z gal. battery jar. They failed to transform, whereas 3 others in a com- parable but not darkened container completed their metamorphosis to juvenile form. 28. A variation in standard length from 22 to 29.5 mm. was noted in 175 manini acronuri from Oahu in their first day of transformation. One hundred and sixteen were collected in tide Acemthurus triostegus sandvicensis — RANDALL 269 pools from 5 days before to 5 days after peak tide-pool recruitment (2 days before new moon). When the lengths of these manini were com- pared with the lengths of the 59 acronuri from the rest of the lunar month, a smaller range of length (23.5-28.5 mm.) was apparent for the former group in spite of nearly twice as many specimens. Presumably individual acronuri are capable of transforming some days before the rest of their age group, and hence at a smaller size, when they come inshore earlier than the others. Similarly, those which reach shoal waters later than the majority will have attained larger than average size. 29. A variation in size at transformation was found with locality which appears related to water temperature. The smallest size occurs in the warmest regions (about 20 mm. in the East Indies) and the largest in cooler sectors (Phoe- nix Islands excepted ) . In the Hawaiian Islands the size at transfor- mation averaged 26.3 mm. for 39 specimens col- lected in April and May and 25.4 mm. for 38 specimens which were taken from July to Octo- ber (thus developing during warmer months). 30. Growth of juvenile manini, as determined by rearing of captive fish, recovery of fish marked by clipping rays from the top or bot- tom of the caudal fin, and progression of modes in successive graphs of tide-pool collections on Oahu is about 12 mm. per month. The recovery of 281 manini of 1,247 tagged with Petersen disk tags fastened with monofila- ment nylon provided information on the growth of larger fish. The growth of adults 100 to 120 mm. in standard length is slightly greater than 1 mm. per month; 120 to 14 1 mm. fish grow about 0.8 mm. per month. Growth of both juveniles and adults ceases during the winter months in the Hawaiian Is- lands. 31. No annular marks could be detected on the scales, otoliths or vertebrae of manini from which estimates of age could be made. Two larger acanthurids, Naso unicornis and Acanthu- rus xanthopterus, lived 20 and 15 years respec- tively in the Waikiki Aquarium in Honolulu where they died as the result of accidents. 32. Small juvenile manini tend to remain in the tide pool or shallow water area to which they first come as acronuri. With increasing size they move seaward. Locality data from the recovery of tagged adult manini was too general to determine local movements, but adequate to demonstrate that no extensive migrations are normally undertaken by this surgeonfish. If displaced from their home area, however, manini are capable of wandering distances as great as 8 mi. 33. Manini may be seen as solitary fish, in small groups, or in large feeding aggregations. One large group of fish, although observed to remain in one region in the Tuamotus, moved as much as 300 yd. in several hours. 34. Juvenile manini in a large tide pool on Oahu were observed to have two hiding places beneath ledges, one of which was preferred. In order to attain shelter they will swim toward an intruder, if necessary. From 4 to 7 hr. were required for aquarium fish to accept an inverted bowl tilted on edge as shelter. Adults are less inclined to seek shelter. Their reaction to an approaching swimmer usually takes the form of retreat. 35. Juvenile manini frequently fight among themselves, both in the natural environment and after being adapted to aquarium life. This does not concern territoriality, but is an expression of dominance by larger fish over smaller fish or old residents (in the case of aquarium fish) over new arrivals. A distinct color change, which involves an almost complete reversal of color pattern, is readily apparent, with the more ag- gressive fish becoming the darkest. Generally there is no contact between fish, and the fish under attack usually retreat. If not, there is a circling at close quarters with occasional overt movements, especially with the posterior part of the body. The manini’s caudal spine is small, and the juveniles were observed to inflict at most only superficial scratches. Manini were not observed to fight with other species but they may display a dominance in their behavior towards some. 36. With the onset of darkness juvenile and adult manini settle to the bottom, usually with pelvic and anal fins in contact, and enter a state of torpor or sleep. Occasional short movements are made, these more often directed backward than forward. Even when remaining in one place 270 PACIFIC SCIENCE, Vol. XV, April 1961 the pectoral fins are kept in slow motion and there are vertical undulations of the median fins. The color is darker than during the day, and large dark blotches are conspicuous between the vertical black bars on the body. A period of 25 min. after darkness ensued was found to be necessary for the fish to reach a state when they could be lightly touched without fleeing. When lights were kept on in an aquarium for 24 hr., manini remained active, however they ceased to feed toward the end of this period and swam restlessly. A light was turned on after 10 ma- nini, 27 to 40 mm. in standard length, were allowed to sleep for 3 hr. From 13 to 28 min. were required for the fish to awaken, the small- est taking the longest time. REFERENCES AL-HUSSAINI, A. H. 1947. The feeding habits and the morphology of the alimentary tract of some teleosts living in the neighbourhood of the Marine Biological Station, Ghardaqa, Red Sea. Mar. Biol. Sta. Ghardaqa (Red Sea) Publ. 5: 1-61, 12 figs. Allee, W. C, and P. Frank. 1948. Ingestion of colloidal material and water by goldfish. Physiol. Zook 21(4): 381-390. Babkin, B. P., and D. J. Bowie. 1928. The di- gestive system and its function in Fundulus heteroclitus . Biol. Bull. 54: 254—277. Breder, C. M. 1948. Observation on coloration in reference to behavior in tide-pool and other marine shore fishes. Amer. Mus. Nat. Hist. Bull. 92: 285-311. 1949'^. On the taxonomy and post larval stages of the surgeon fish, Acantlourus hepatus . Copeia 1949(4) : 296. 1949$. On the relationship of social behavior to pigmentation in tropical shore fishes. Amer. Mus. Nat. Hist. Bull. 94(2): 85-104, 8 pis., 2 figs. Breder, C. M., and E. Clark. 1947. A contri- bution to the visceral anatomy, development, and relationships of the Plectognathi. Amer. Mus. Nat. Hist. Bull. 88(5) : 287-320, 4 pis., 8 figs. Brock, V. E. 1954. A preliminary report on a method of estimating reef fish populations. J. Wild!. Mgmt. 18: 289-308, 1 fig. Calhoun, A. J., D. H. Fry, Jr., and E. P. Hughes. 1951. Plastic deterioration and metal corrosion in Petersen disk fish tags. Calif. Fish Game 37(3): 301-314, 3 figs. Dawes, B. 1929. The histology of the alimentary tract of the plaice {Fleur one ctes plates sa ) . Quart. J. Micr. Sci. 73: 243-274. Dawson, E. Y., A. A. Aleem, and B. W. Hal- stead. 1955. Marine algae from Palmyra Is- land with special reference to the feeding habits and toxicology of reef fishes. Allan Hancock Fdn. Occ. Pap. 17. 39 pp. Fowler, H. W. 1936. The marine fishes of West Africa. Amer. Mus. Nat. Hist. Bull. 70(2): 607-1493,291 figs. Fox, D. L., J. D. Isaacs, and E. F. Corcoran. 1952. Marine leptopel, its recovery, measure- ment and distribution. J. Mar. Res. 11(1): 29-46. Fraser, J. M. 1955. The smallmouth bass fishery of South Bay, Lake Huron. Fish. Res. Bd. Can. J. 12: 147-177. Hawk, P. B., and O. Bergeim. 1942. Practical Physiological Chemistry. Maple Press Co., York, Pa. xxii + 968 pp., 281 figs. HlATT, R. W. 1947. Food-chains and the food cycle in Hawaiian fish ponds, Part I. The food and feeding habits of mullet {Mugil cepha- lus) , milkfish {Chanos chanos) , and the ten- pounder ( Flops machnata) . Trans. Amer. Fish. Soe. 74: 250-261. Hydrographic Office (United States Navy). 1948. World Atlas of Sea Surface Tempera- tures. Publ. No-. 225, 2nd ed. 48 charts. JAHN, T. L., and L. Kuhn. 1932. The life his- tory of Epih della melleni Maccalium 1927, a monogenetic trematode parasitic on marine fishes. Biol. Bull. 62: 89-111, 2 pis., 2 figs. Jordan, D. S., and A Seale. 1906. The fishes of Samoa. U. S. Bur. Fish. Bull. 25: 175-455 + xxx pp., Ill figs., 5 pis., 16 col. pis. Acanthurus triostegus sandvicensis — Randall Jordan, H. J. 1927. Ubungen aus der Verglei- chenden Physiologic. Julius Springer, Berlin, vii + 272 pp. Kenyon, W. A. 1925. Digestive enzymes in poikilothermal vertebrates. An investigation of enzymes in fishes with comparative studies on those of amphibians,- reptiles, and mam- mals. U. S. Bur. Fish. Bull. 41: 181-199. King, J E , and T. S. Hida. 1954. Variations in zooplankton abundance in Hawaiian waters, 1950-52. U. S. Fish and Wildl. Serv. Spec. Sci. Rep. (Fish.) 118: 1-66, 16 figs. Lawrence, J. M. 1952. A trapping experiment to estimate the bluegill population in a farm pond. Iowa Acad. Sci. 59: 475-479. LEFEVRE, M. 1940. Sur la resistance de certaines algues d’eau douce a Taction des sues gastro- intestinaus des poissons. Acad. Sci. Paris C. R. 210: 347-349. Leipper, D. F., and E. R. Anderson. 1950. Sea temperatures, Hawaiian Island area. Pacif. Sci. 4(3): 228-248, 27 figs. Lerner, A. B, and T. B. Fitzpatrick. 1950. Biochemistry of melanin formation. Physiol. Rev. 30: 91-126, 4 figs. Linderstr0m-Lang, K., and H. Holter. 1933. Studies on enzymatic histochemistry. V. A micromethod for the estimation of sugars. Lab. Carlsberg C. R. 19: 1-12. LINTON, E. 1910. Helminth fauna of the dry Tortugas, III. Trematod.es . Tortugas Lab. Pap. 4: 1-98. Lutken, C. F. 1880. Spolia Atlantica. Bidrag til kundskab orn formforandringer hos fiske under deres vaext or udvikling, saerligt hos nogle af Atlanterhavets H0js0fiske. K. Dan- ske Vidensk. Selsk. 5(12): 409-613, 11 figs., 5 pis. MacKay, M. E. 1929. The digestive system of 'the eel pout (Zoarces anguillaris ) . Biol. Bull. 56: 24-27. MANTER, H. W. 1955. The zoogeography of trematodes of marine fishes. Exp. Parasit. 4: 62 - 86 . 271 Olsen, L. S, 1952. Some nematodes parasitic in marine fishes. Inst. Mar. Sci. Univ. Tex. Publ. 11(2): 173-215, 84 figs. Ommanney, F. D. 1949. Age investigations in Mauritius fishes. Roy. Soc. Arts Sci. Mauritius Trans. Ser. C 15: 38-59, 8 figs. Poey, F. 1875. Enumeratio piscium cubensium, Part I. Soc. Esp. Hist. Nat. An. 4: 75-161, 3 pis. Rahimullah, M. 1945. A comparative study of morphology, histology, and probable func- tion of pyloric caecae in Indian fishes. Indian Acad. Sci. Proc. 2 IB: 1-37. Randall, J. E. 1955^. An analysis of the genera of surgeon fishes ( family Acanthuridae) . Pacif. Sci. 9(3): 359^-367. — 195 5 A Fishes of the Gilbert Islands. Atoll Res. Bui. 47 : ix + 243 pp., 2 figs. 1955c. A revision of the surgeon fish genera Zebras oma and Par acanthurus. Pacif. Sci. 9(4): 396-412, 8 figs, 1 col. pi. 195 5 <7. A revision of the surgeon fish genus Cteno chaetus, family Acanthuridae, with descriptions of five new species. Zool- ogica 40(4) : 149-166, 3 figs, 2 pis. 195 6a. A new method of attaching Petersen disk tags with monofilament nylon. Calif. Fish Game 42(1): 63-67, 1 fig. 195 6 A A revision of the surgeon fish genus Acanthurus . Pacif. Sci. 10(2): 159- 235, 23 figs, 3 col. pis. — 1958. A review of the labrid fish genus Labroides , with descriptions of two new species and notes on ecology. Pacif. Sci. 12(4): 327-347, 6 figs, 1 col. pi. Randall, J. E, and V. E. Brock. I960. Ob- servations on the ecology of epinepheline and lutjanid fishes of the Society Islands, with emphasis on food habits. Trans. Amer. Fish. Soc. 89(1): 9-16. Reintjes, J. W, and J. E. King. 1953. Food of yellowfin tuna in the central Pacific. U. S. Fish Wildl. Serv. Fish. Bull. 81: 91-110, 10 figs. 272 PACIFIC SCIENCE, Vol. XV, April 1961 Rounsefell, G. A., and J. L. Kask. 1945. How to mark fish. Amer. Fish. Soc. Trans. 73: 320- 363, 4 figs. Schlottke, E. 1939. Untersuchungen fiber die Verdamingsfermente der Quappe ( Lota vul- garis L.) Z. Fisch. 37: 381-394. Siwak, J.- 1931. Ancyrocephalus vistulensis sp. n., un nuoveau trematode, parasite du Silure ( Silurus glanis L.) . Int. Acad. Pol. Sci. Lettr. Cl. Sci. Math. Nat. Ser. B. (Sci. Nat. II) Bull. 7(10): 669-679, 1 pi., 7 figs. Snedecor, G. W. 1948. Statistical Methods. Iowa State College Press, Ames, xvi + 485 pp. Sparta, D. A. 1928. Nastro galleggiante di vova di Teleostei. R. Com. Talasogr. Irak Mem. 145: 1-13, 1 pi. Suyehiro, Y. 1942. A study on the digestive system and feeding habits of fish. Jap. J. Zool. 10(1): 1-303, 15 pis, 190 figs. Tester, A. L, and M. Takata. 1953. Contri- bution to the biology of the aholehole, a po- tential baitfish. Industr. Res. Adv. Coun. Grant No. 29 Final Rep.: 1-54, 17 figs. U. S. Coast and Geodetic Survey. 1956. Surface water temperatures at tide stations Pacific coast North and South America and Pacific Ocean islands. Spec. Publ. U. S. Cst. Geod. Surv. 280: 74 pp, 1 map. VAN Weel, P. B. 1937. Die Ernahrungsbiologie von Amphioxus lanceolatus. Staz. Zool. Na- poli Pubbl. 16: 221-272, 25 figs. VONK, H. J, Jr. 1927. Die Verdauung bei den Fischen. Z. Vergl. Physiol. 5: 445-546. Weber, M. 1913. Die Fische der Siboga-Expedi- tion. E. J. Brill, Leiden, xii + 710 pp, 123 figs, 12 pis. Whitehouse, R. H. 1923. A statistical study of young fishes from Silavatturai Lagoon, Tuticorin. Madras Fish. Bull. 17: 49-103. Yonge, C. M. 1931. Digestive processes in marine invertebrates and fishes. J. Cons. 6(2): 175-212. A New Species of Micronereis (Annelida, Polychaeta) from the Marshall Islands 1 Donald J. Reish 2 During the course of surveying Eniwetok Atoll, Marshall Islands, for polychaetous an- nelids in the summers of 1957 and 1958, the author encountered from three localities four specimens of a new species of polychaete be- longing to the nereid genus Micronereis Clap- arede. Transects were made of both the ocean and lagoon sides of the islands shown in the map of Eniwetok Atoll (Fig. 1). One of the collecting methods employed was to preserve algae, coral, rocks, sand, etc., in formalin in the field and to sort out the polychaetes in the lab- oratory. The new species of Micronereis was collected in this manner. family NEREIDAE Micronereis Claparede Micronereis eniwetokensis n. sp. Figs. 2-6 Four complete specimens, each with 17 seti- gerous segments measuring 1.5 to 2.0 mm. in length, came from three localities in Eniwetok Atoll. The prostomium (Fig. 2) is broadest posteriorly and is clearly separated from the first setigerous segment. The prostomium is weakly indented at its anterior margin. Four pairs of tentacular cirri are broadest near the point of attachment and tapered distally. The short palpi are viewed from the ventral side. There are no prostomial tentacles or peristomium as is char- acteristic for the genus. Four large eyes are in trapezoidal arrange- ment, with the anterior pair being the larger, 1 The field work was made possible by the U. S. Atomic Energy Commission through its Eniwetok Marine Biological Laboratory. The author is indebted to Dr. Robert W. Hiatt for his assistance during the course of the field investigations. 2 Department of Biological Sciences, Long Beach State College, Long Beach, California. Manuscript re- ceived February 5, I960. farther apart, and provided with a clear lens. A variation was noted in the eyes of one speci- men ( Fig. 3 ) . The anterior pair was elongated and extended posteriorly and laterally to the second pair. Each anterior eye was provided with a large clear lens. The yellow jaws (Fig. 4) were seen by dis- section or by mounting the worm in glycerine. They were broadest at their base and terminated in one large tooth and three smaller teeth. The apical tooth was serrated along its inner margin. The first two setigerous segments are uniram- ous with homogomph spinigers. Each has a fili- form ventral cirrus. All parapodia (Fig. 5) are biramous from the third setigerous segment to the posterior end. The rami are widely separated and each has a single black aciculum. The inner margin between the notopodium and neuropodium is ciliated. Each ramus has a cirrus; the dorsal one is longer than the ventral one. All setae (Fig. 6) are homogomph spinigers with the appendages lacking spines along their margins. The setae number 12-20 in each lobe of the parapodium. The pygidium is characterized by having a bilobed ventral fleshy membrane and two blunt lateral lobes. DISCUSSION Three species are previously known for the genus Micronereis , M. variegata Claparede 1863, M. nanaimoensis Berkeley and Berkeley 1953, and M. halei Hartman 1954. Sexual dimorphism has been described for the former two species. Specialized setae in the third setigerous segment of the male, and a difference in the number of segments between the two sexes, occur in both M. variegata and M. nanaimoensis. The jaws vary according to sex in M. variegata . Additional dimorphic characters in M. nanaimoensis include the pygidium, coloration, and capillary setae in 273 274 PACIFIC SCIENCE, Vol. XV, April 1961 the male. Sexual dimorphism is unknown for M. halei and M. eniwetokensis. The large anterior eyes (Fig. 3) on one in- dividual of M. eniwetokensis are an unusual variation for the members of the genus. The appearance of secondary sex characteristics, such as enlargement of the eyes prior to sexual ma- turity, is widespread in the family Nereidae (Reish, 1957), but it is unknown for the genus Micronereis (Berkeley and Berkeley, 1953; Rullier, 1954). No other morphological differ- ences were noted in this specimen collected from 94 ft. of water in the lagoon. M. eniwetokensis comes closest to M. varie- Eniwetok Atoll 5 Statute Miles i J Fig. 1. Map of Eniwetok Atoll, Marshall Islands. Collections were made at the named islands. Micronereis eniwetokensis was collected from the areas indicated by the dots. Micronereis eniwetokensis — Reish 275 FIGS. 2-6: 2, Anterior end of the holotype of Micronereis eniwetokensis; 3, prostomium of M. eniwe- tokensis with enlarged eyes; specimen collected from a depth of 94 ft. in the lagoon of Eniwetok Atoll; 4, jaw of M. eniwetokensis ; 5, parapodium from the mid-region of M. eniwetokensis; setae omitted from figure; 6, homogomph spiniger from the notopodium of Figure 5. 276 PACIFIC SCIENCE, Vol. XV, April 1961 gata\ they differ in the nature of the homogomph spiniger, the pygidium, and the presence or ab- sence of cilia on the parapodia. There are some distinct characteristics for the male of M. varie- gata\ it is not known whether or not sexual dimorphism occurs in M. eniwetokensis. The diagnostic characters for the known species of the genus are summarized in Table 1. TYPE MATERIAL : The holotype, one paratype, and two additional specimens have been de- posited in the U. S. National Museum. TYPE LOCALITY: Intertidal region on the lagoon side of Aaraanbiru Island, Eniwetok Atoll, Marshall Islands ( Fig. 1 ) . ECOLOGY: As stated above, pieces of coral rock of loose substrate were preserved in the field at each station visited. M. eniwetokensis was collected in this manner. It was associated with the following species of polychaetes: Aaraanbiru Island, lagoon side, substrate coral rock and beach rock with algae attached. Polychaetes: M. eniwetokensis (2 specimens), Haplosyllis spongicola Grube ( 1 ) , Opistosyllis longicinata Monro (6), Ceratonereis mirabilis Kinberg ( 17 ) , Eunice afra Peters ( 1 ) , Lysidice collaris Grube (3 ), Palola siciliensis Grube ( 3 ) , Dodecaceria laddi Hartman ( 2 ) , and Poly- op hthalmus pictus (Dujardin) (19). Igurin Island, ocean side, substrate coral rock and fragments taken from under coral head near surge zone. Polychaetes: Micronereis eni- wetokensis (1 specimen), Eurythoe complanata TABLE 1 Comparisons of the Known Species of Micronereis CHARACTERISTIC M. variegata M. nanaimoensis M. halei M. eniwetokensis Length 2-4 mm. 13-15 mm. 7 mm. 1. 5-2.0 mm. Number of setigerous segments $ $ 17-18 21 23 26 20-25 (sex unknown) 17 ( sex unknown ) Palpi absent present present present Jaws dimorphic; $ 5 teeth 2 4 teeth 4-5 teeth 6 teeth 4 teeth Crotchets in $ , tnird segment 2 curved 5+ crested unknown unknown Spines on appendage of homogomph spinigers present present absent absent Cilia on parapodia absent present absent present Pygidium 2 anal cirri 2 anal cirri in $ and $ ; also 2 lateral lobes in $ 2 lateral lobes; ventral bilobed membrane Unique pigmentation none 2 red bands on tentacular cirri; see original account for body pigmentation brown band on tentacular cirri none Unique characters none capillary setae in 6 inferior notopodial and superior neuropodial lobes in parapodium none Geographical distribution Europe, Mediterranean Sea British Columbia, Canada South Australia Eniwetok Atoll, Marshall Islands Micronereis eniwetokensis — Rexsh 277 (Pallas) (2), Pseudeurythoe oculifera (Auge- ner) (1), Genetyllis gracilis (Kinberg) (1), Typosyllis variegata (Grube) (2), Platynereis pulchella Gravier ( 6 ) , Lysidice collaris ( 1 ) , Palola siciliensis (1), Dodecaceria laddi (2), and Terebella ehrenbergi Grube (1). Parry Island, lagoon side in 94 ft. of water collected by Mr. Mike Chamberlain, 450 ml. of light gray sand. Polychaetes: Micronereis eni- wetokensis ( 1 specimen) , Micronephthys sphae- rocirrata (Wesenberg-Lund ) (2), previously known only from the Sea of Iran (Wesenberg- Lund, 1949), and Scolelepis indica Fauvel (2). REFERENCES Berkeley, Edith, and Cyril Berkeley. 1953. Micronereis nanaimoensis n. sp.: with some notes on its life history. Canad. Fish. Res. Bd. J. 10 (2): 85-95. ClaparLde, Edouard. 1863. Beobachtungen iiber Anatomic und Entwicklungsgeschichte wirbelloser Thiere an der Kiiste von Nor- mandie angestellt. Leipzig. 120 pp. Fauvel, Pierre. 1923. Polychetes errantes. Faune de France 5 : 1-488. Hartman, Olga. 1954. Australian Nereidae including descriptions of three new species and one genus, together with summaries of previous records and keys to species. Trans. Roy. Soc. S. Aust. 77: 1-41. Rullier, Francois. 1954. Recherches sur la morphologic et la reproduction du Nereiden Micronereis variegata Claparede. Arch. Zook Exp. Gen. 91: 197-233. Wesenberg-Lund, Elise. 1949. Polychaetes of the Iranian Gulf. Dan. Sci. Invest. Iran, pt. 4: 247-400. Life Cycle of Mesostephanus appendiculatus (Ciurea, 1916) Lutz, 1935 (Trematoda: Cyathocotylidae) 1 W. E. Martin 2 For a number of years, the author has been studying trematode larvae which develop in the California horn-shell snail, Cerithidea calif ornica Haldeman. One of these, a pharyngeate, fur- cocercous cercaria, proved to be the larva of Mesostephanus appendiculatus. The adults of this species were first described from the small intestines of Rumanian cats and dogs by Ciurea (1916). Ciurea, however, placed this species in the genus Prohemistomum. Price (1928) found this parasite in the small intestine of a dog that had lived in the vicinity of Washing- ton, D.C. Lutz (1935) transferred this and some other species to his new genus, Mesoste- phanus. , naming M. fajardensis (Price, 1934) as type species. Dubois (1953) includes the fol- lowing species in the genus Mesostephanus: M. fajardensis (Price, 1934); M. appendicula- tus (Ciurra, 1916); M. appendiculatoides (Price, 1934); M. cubdensis Allegret, 1941; M. haliasturis Tubangui and Masilungan, 1941; and M. longisaccus Chandler, 1950. Caballero, Gro- cott, and Zerecero (1954) added M. micro- bur sa from the intestine of Pelecanus occiden- talis calif ornicus. Dubois (1953) believes that the members of the genus are natural parasites of certain fish-eating birds and accidental par- asites of dogs. MATERIAL AND METHODS Infected Cerithidea californica were isolated in finger bowls. Emerged cercariae were studied 1 These studies were aided by a contract between the Office of Naval Research, Department of the Navy, and the University of Southern California, NR 165-252. Manuscript received June 9, 1959. The opinions and assertions contained herein are the private ones of the author and are not to be con- strued as official or reflecting the views of the Navy Department or the naval service at large. 2 Biology Department and Hancock Foundation, University of Southern California, Los Angeles. alive and as fixed and stained whole mounts. Cercariae were fixed without pressure by forci- bly ejecting them into cold Bouin’s solution. Earlier larval stages and percentages of infec- tion were obtained by crushing snails. Unin- fected Fundulus parvipinnis parvipinnis (Gi- rard ) and Gillichthys mirabilis Cooper were col- lected in an isolated pool where there were no snails. These fish were exposed to cercariae and, following a lapse of 2-3 weeks, were fed to hatchery-raised chicks. The chicks were exam- ined 9 days after the experimental feeding. The adult worms obtained were fixed in Bouin’s solution under slight cover-glass pressure. Lar- val and adult stages were stained with para- carmine and mounted in Permount. All measurements are in millimeters. OBSERVATIONS The sporocysts and cercariae of Mesostepha- nus appendiculatus develop in the digestive gland of the brackish- water snail, Cerithidea californica. During a 12 -month survey (Mar- tin, 1955), in which at least 1,000 snails were examined each month, only 7 infections of this parasite were found in 12,995 snails. sporocyst ( Fig. 2 ) : Mother sporocysts were not observed. Daughter sporocysts are saccular and elongate. Measurements of 20 stained and mounted specimens are: length 1.368-3.355, av. 2.38; maximum width 0.173-302, av. 0.236. The wall of the sporocyst has transverse con- tractile bands. At intervals there are thicker bands which give a false appearance of seg- mentation. One end of the sporocyst has a thick wall which is traversed by a birth canal. CERCARIA (Figs. 3, 4): The cercariae are nonoculate and furcocercous. Though they lack eyespots, they show positive phototropism. The body surface is covered with minute, quincun- cially arranged spines and scattered papillae 278 Mesostephanus appendiculatus — Martin with bristles. Tubular glands are plentiful lat- erally and sparse dorsoventrally in the anterior half of the body. Ten to 12 glands have their cell bodies near the oral sucker and have ducts opening near the mouth. Measurements based on 20 stained and mounted specimens are: body length 0. 18-258, av. 0.192; maximum body width 0.078-.115, av. 0.094; oral sucker length 0.031-045, av. 0.038; oral sucker width 0.025- .037, av. 0.03; ventral sucker midventral, ru- dimentary, 0.009-012, av. 0.01 in diameter; lengths of prepharynx and esophagus approxi- mate that of pharynx; pharynx oval to spherical, 0.006-012, av. 0.011 long and 0.012-.016, av. 0.013 wide; intestinal caeca sinuous, termi- nating near excretory bladder; genital primord- ium median, immediately anterior to excretory bladder, 0.006-.022, av. 0.019 long and 0.019- .025, av. 0.021 wide; excretory bladder small, transversely elongate, with exit duct entering tail, dividing into two ducts which pass around the "Island of Cort” and rejoin, extending to the furcal region to divide into two ducts, each of which opens to the outside at the tip of a furca. Four collecting ducts empty into the blad- der, two laterally and two medial to the lateral ducts on the anterior wall of the bladder. The two medial ducts pass around the genital pri- mordium to unite and proceed as a single duct to a point near the bifurcation of the gut where it joins the lateral ducts which have proceeded anteriorly from the bladder. As the lateral ducts bend medially, each gives off a duct whose prox- imal portion contains a tuft of cilia. The latter duct extends posteriorly to about mid-body level where it divides into anterior and posterior branches each of which collects from three groups of three flame cells each. Three of the flame cells emptying into the posterior branch are located in the tail. The excretory formula is 2 [(3 + 3 + 3) + (3 + 3+ (3)] = 36. Short, moniliform concretions occur in the main col- lecting tubes. The tail is set in a dorsal socket near the posterior end of the body. The tail surface bears bristles and minute spines. The tail stem length is 0.3 58-407, av. 0.376, and maximum width near the junction with furcae 0.0 14-022, av. 0.021. The furcae are 0.18-.2, av. 0.19 long and 0.019-022, av. 0.021 in maxi- mum width near the junction with the tail 279 stem. Each furca bears a dorsoventral fin over the distal four-fifths of its length. METACERCARIA: Fundulus parvipinnis parvi- pinnis and Gillichthys mirabilis were exposed to cercariae which rapidly penetrated the skel- etal muscles and encysted. Penetration of large numbers of cercariae killed the fish. Death of the second intermediate host due to the pene- tration of large numbers of cercariae has been noted by Vernberg (1952) for a related par- asite. Metacercariae approximately 3 weeks old were dissected from the fish and were fed, along with some muscle tissue, to hatchery-raised chicks. ADULT ( Fig. 1 ) : Adult Mesostephanus ap- pendiculatus were obtained from the small in- testines of hatchery-raised chicks fed fish mus- cle and metacercariae. The chicks were exam- ined 9 days after the experimental feeding. The following description and measurements are based on nine specimens. Body surface covered with scale-like spines arranged quincuncially. Body length 0.547-763, av. 0.68; body width 0.346-518, av. 0.41; oral sucker length 0.04- .059, av. 0.049; oral sucker width 0.047-.078, av. 0.055; acetabulum 0.04-.068, av. 0.06 in diameter; tribocytic organ well developed, open- ing usually slitlike; prepharynx very short; pharynx 0.037-.058, av. 0.05 long and 0.031- .044, av. 0.037 wide; esophagus approximately one-half pharyngeal length, with transverse mus- cle fibers; intestinal caeca sinuous, with occa- sional short diverticula, reaching to near pos- terior end of body; testes oblique, in posterior half of body, 0.109-124, av. 0.116 long and 0.072-.087, av. 0.079 wide; cirrus sac and cirrus well developed; male genital opening communi- cates with common genital exit at posterior end of body; ovary intertesticular, 0.05-.08, av. 0.065 long and 0. 04-065, av. 0.05 wide; metraterm elongate, muscular, with sphincter at distal end where it empties into common genital exit; eggs yellow, operculate, 0.084-.137, av. 0.108 long and 0.058-.081, av. 0.07 wide; vitellaria com- posed of discrete follicles arranged in a circle in posterior half of body but not entering pos- terior conical body extension; excretory system more complex than in cercaria, anastomosing branches arise from main collecting ducts, some branches end blindly near body surface. 280 PACIFIC SCIENCE, Vol. XV, April 1961 Figs. 1-4: 1, Adult Mesostepkanus appendiculatus , ventral view; 2, sporocyst; 3, diagram to show most of excretory system; 4, cercaria, ventral view. Abbreviations : a, Acetabulum; b, birth pore; c, cirrus sac; e, egg ; g, genital primordium; i, Island of Cort; m, metraterm; o, ovary, or oral sucker; p, pharynx; t, testis; tr, tribo- cytic organ; v, vitellaria. All drawings made with the aid of a camera lucida unless otherwise stated. Mesostephcmus appendiculatus — -Martin DISCUSSION The body dimensions of the adult M. appen- diculatus described in this paper are smaller than those listed for the species by Dubois (1938). This may be due to the fact that the duration of the infection was only 9 days so that the worms probably had not attained their full size even though they were sexually mature. The range of egg size and the number of eggs ( 1-7 ) in the uterus were greater in my speci- mens that in those listed by Dubois (1938), which include measurements given by Ciurea (1916) and Prendel (1930). The anterior ex- tent of the cirrus was greater in some speci- mens than is shown in Figure 1. The extent of the cirrus in Figure 1 resembles that of M. microbursa Caballero, Grocott, and Zerecero (1954), recovered from the intestines of peli- cans, Pelecanus occidentalis calif ornicus, col- lected in Panama and in the Coronado Islands off Mexico. However, the sucker ratio, spina- tion, and extent of the posterior appendix are different in the two species. The present work extends the range of M. appendiculatus to the west coast of the United States. It has been found on the east coast of this country by Price (1928), in Rumania by Ciurea (1916), and in the Ukraine by Prendel (1930). Dubois (1953) states that pelicans are the natural hosts of this species. Since this parasite can develop also in dogs, cats, and chicks, the present author believes that other fish-eating birds probably serve as additional natural hosts. Maxon and Pequegnat (1949) examined Cerithidea calif ornica collected at Newport Bay, California, between October, 1947, and May, 1949. They found 21 per cent of the snails infected with furcocercous cercariae. They described one of the latter with 16 flame cells but did not describe the cercaria of Mesospepha- nus appendiculatus. SUMMARY The life cycle of Mesostephanus appendicu- latus (Ciurea, 1916) Lutz, 1935 has been dem- onstrated experimentally. Sporocysts and cer- cariae develop in the brackish-water snail, Ceri- thidea calif ornica Haldeman, collected at New- 281 port Bay, California. The cercaria is furcocer- cous and has a flame-cell pattern expressed by the formula 2 [(3 + 3 + 3) + (3 + 3+ (3)3 = 36. Second intermediate hosts are Fundulus parvipinnis parvipinnis (Girard) and Gillich- thys mirahilis Cooper. Experimentally infected fish were fed to hatchery-raised chicks. After a lapse of 9 days, egg-bearing worms were re- moved from the small intestines of the chicks. REFERENCES Caballero, E., R. G. Grocott, and Zere- cero y D., C. 1954. Helmintos de la Repub- lica de Panama, IX. Algunos Trematodos de Aves marinas del Oceano Pacifico de Norte. An. Inst. Biol. Mex. 24: 391-414. Ciurea, I. 1916. Prohemistomum appendicula- tum eine neue Holostomidien-Art aus Hun- den- und Katzen-darm, dessen Infektion- squelle in den Siisswasserfischen zu suchen ist. Z. Infektkr. 17: 309-328. Dubois, G. 1938. Monographie des Strigeida (Trematoda). Mem. Soc. Neuchatel. Sci. Nat. 6: 1-535. 1953. Systematique des Strigeida. Mem. Soc. Neuchatel. Sci. Nat. 8: 1-141. Lutz, A. 1935. Observances e consideranoes sobre Cyathocotylineas e Prohemistomineas. Mem. Inst. Osw. Cruz 30: 157-168. Maxon, M. G., and W. E, Pequegnat. 1949. Cercariae from Upper Newport Bay. J. Ent. Zool. 41: 30-55. Prendel, A. R. 1930. Ein Beitrag zum Studium der Helminthenfauna der Hunde in der U. d. S.S.R. (Siidliche Ukraine). Zool. Anz. 89: 323-326. Price, E. W. 1928. The occurrence of Prohem- istomum appendiculatum in the United States. J. Parasit. 15: 68. Vernberg, W. B. 1952. Studies on the trema- tode family Cyathocotylidae Poche, 1926, with description of a new species of Holo- stephanus from fish and the life history of Prohemistomum chandleri sp. nov. J. Parasit. 38: 327-340. Some Notes on the Hawaiian Monk Seal Judith E. King 1 and R. J. Harrison 2 Up to 1958 the collections of the British Mu- seum (Natural History) possessed neither skel- etal material nor skin of the Hawaiian or Laysan monk seal, Monachus schauinslandi Matschie 1905. Indeed the only remains of this animal in Europe until now have been those brought back by Dr. Schauinsland, amongst which was the skull of the seal later named after him by Mat- schie (Matschie, 1905). The stuffed skin of this animal is in the Bremen Museum and the type skull is in the Zoological Museum in Ber- lin, no. 32795 (Wahlert, 1956). So it was with great pleasure that we received, in the summer of 1958, through the kindness of Mr. Vernon E. Brock, then director of the Territorial Division of Fish and Game, Honolulu, Hawaii, a young male monk seal from Laysan Island that had been shipped complete and frozen to this coun- try. The animal was thawed, injected with col- oured gelatin in the vascular system, and dis- sected after fixation, but unfortunately the tis- sues were too poorly preserved for any accurate histology. DESCRIPTION The seal is young, has a nose-to-tail length of 163.5 cm. (5 ft. 6 in.) and weighs 74.4 kg. (164 lb.) complete. Kenyon and Rice (1959) give the estimated weights of recently weaned pups as from 95-160 lb. The present animal was caught June 4, 1958, and as it must have been very near to being weaned, its probable date of birth must have been about April 30. Although the weight is a little high for a recently weaned pup, the animal is in very good condition and is unlikely to have been a yearling, as yearlings are relatively thin and do not reach the weight and condition of pups that have been feeding from their mothers. Similarly the state of the epiphyses and the obvious youth of the skeleton make it unlikely that it is from a 2 -year-old animal. The age of the present animal is there- fore estimated to be about 5 weeks ( the thymus is large and weighs 32.5 g. ). The coat is dark 1 British Museum (Natural History). 2 London Hospital Medical College. Manuscript re- ceived January 15, I960. silvery grey dorsally, on the top of the head, on both sides of the fore and hind flippers, and on the dorsal surface of the tail. Laterally the grey shades to silvery white ventrally. The hind flippers are a little lighter grey on the inner side towards their insertion. There are also lighter patches round the eyes and surrounding the insertion of the supraorbital vibrissae, and along the upper lip. The lower jaw is light in colour. On the back the hairs are dark grey with a white tip. This white tip becomes longer towards the belly so that the ventral hairs are white with a short dark base. The moustachial whiskers are in five rows on each side of the nose, the upper row having four whiskers and the other rows approximately nine in each. The whiskers are dark brown at the base, shading to straw colour at the tip; they are oval in cross section and are not beaded as in Phoca (Fig. 1 ). The tongue has a notch in its anterior end. The length of the small intestine is approxi- mately 57 ft. The only food remains in the stomach are fish bones and skin and these have been identified as being most probably from the puffer fish, Arothron meleagris (Lac.), which is a poor swimmer usually found near coral formations. Nematodes, a small cephalopod beak, and an isopod are also present in the stom- ach. The nematodes have been identified as Contracaecum turgidum, a species previously described from this seal, and the isopod as Livoneca sp., usually found as an ectoparasite on fish and probably ingested attached to a fish. In the small intestine there are remains of a tapeworm. It is not in a sufficiently good con- dition to be identified exactly, but is probably Diphyllobothrium sp. The skin and skeleton of this seal are in the collections of the British Museum (Natural His- tory), the registered number of the skin being 58.521, and that of the skeleton 1958.11.26.1. OSTEOLOGY The skeleton of M. schauinslandi has not pre- viously been described and although it is at- tempted here to fill this gap the description and 282 Hawaiian Monk Seal — King and Harrison 283 Fig. 1. a, Anterior view of the face of M. schauinslandi. b, Lateral view of the face, showing dorsal posi- tion of nostrils. 284 PACIFIC SCIENCE, VoL XV, April 1961 Fig. 2. a . Dorsal view of right fore flipper, b, Dorsal view of right hind flipper. Hawaiian Monk Seal — King and Harrison comparison are considerably limited by the youth of the specimen. The form of this descrip- tion is based on that of the skeletons of M. monachus and M. tropicalis (King, 1956) and frequent reference will be made to the details in that paper. No skeleton of a young M. trop- icalis was available so the M. schauinslandi skele- ton was compared with that of an adult M. trop- icalis ( 1887.8.5.1.) and of a young M. monachus ( 1894.7.27.3), the skull of which showed by its condylobasal length (224 mm.) and suture age (13) that it was from an animal of approxi- mately the same size as the Laysan monk seal. Comparison of M. schauinslandi Skull with Those of M. monachus and M. tropicalis 1. The additional measurements now avail- able of the skull of M. schauinslandi confirm the previous statement (King, 1956) that skulls of both M. schauinslandi and M. tropicalis are slightly narrower in proportion to their width than those of M. monachus, though the accuracy of this conclusion is limited by the small num- ber of skulls available. M. M. tropi- M. schau- monachus calis inslandi % % % Zygomatic width... Snout width at 59-9-70.3 61.7-62.1 60.9-61.5 canines 20.9-26.0 20.6-20.9 20.3-20.5 Width at external auditory meatus Width at petrous 53.1-58.5 49.8-50.5 50.5-55.0 bones 1 60.3-64.9 56.3-59-2 59.2-64.1 2. The nasal bones of this young animal are as previously described from the type skull. The anterior ends are in the form of an inverted V with the point directed posteriorly; the posterior ends do not taper as much as in M. tropicalis. 3. The lower edge of the infraorbital foramen is, when seen from the front, narrower than the upper edge. This confirms Matschie ( 1905 ) and is similar to M. tropicalis. 4. The absence of a maxillary tubercle at the anterior edge of the orbit also confirms Mat- schie. 5. As previously noted, the shape of the zy- gomatic arch is more like that of M. monachus than M. tropicalis. 285 6. It was noted previously that the posterior edge of the palate of M. schauinslandi formed a wide V. It is now felt that it is better described as U in shape and rather more like that of M. monachus, though there is probably a certain amount of variation in this character. 7. The pterygoid bones curve outwards as in M. tropicalis and in this young skull of M. schauinslandi they are just visible when it is viewed dorsally. 8. The coronoid process of the lower jaw is narrow and like that of M. tropicalis. 9. Examination and comparison of the young skull of M. schauinslandi now available con- firms the previous report that it is more like M. tropicalis than M. monachus ( Fig. 3 ) . There are no supernumerary bones in the skull. Kenyon and Rice ( 1959) note that, in the few skulls of M. schauinslandi they examined, the shape of the palate and the shape of the zygomatic branch of the squamosal do not ap- pear to be constant and are thus not good dis- tinguishing characters. It is certainly agreed that the shape of the palate is variable and there is a need for the examination of as many adult skulls as possible to determine the amount of variation. Measurements of Skull of M. schauinslandi 1958.11.26.1 mm. % Condylobasal length 220 100 Condylobasilar length 211 95.9 Basal length 201 91.4 Basilar length 192 87.3 Snout width at canines 45 20.5 Width of skull at front end of last upper molars 59 26.8 Zygomatic width 134 60.9 Width at upper edge auditory meatus .. 121 55.0 Width at petrous bones (mastoid width) 141 64.1 Palatal length 101 45.9 Palatilar length 92 41.8 Width of occipital condyles.. 62 28.2 Length of nasal suture 50 22.7 Length of upper molar row 55 25.0 Suture age 13 286 PACIFIC SCIENCE, Vol. XV, April 1961 Fig. 3. ct. Dorsal, b, ventral, and, c, lateral views of the skull of M . schauinslandi. d , Lateral view of lower jaw. Hawaiian Monk Seal — King and Harrison 287 Dental formula ‘ C pC 2 1 5 The term "post canines” is used here instead of "molars” and "premolars,” as these latter cannot be used precisely when referring to pin- niped teeth. UPPER JAW TEETH: As previously mentioned, the upper incisors are set in a straight line across the anterior end of the premaxillae and in general the setting and shape of the teeth are like M. tropicalis. The incisors of this young animal do not have such a pronounced "waist” as those of the adult M. tropicalis, but it is more evident than in the young M . monachus. The canines are very small for a male animal. Those of the young Af. monachus, also a male, whose skull is only 4 mm. longer than that of the young M. schauinslandi, are much larger. M. mona- M. schau- chus inslandi Anteroposterior length at crown-root junction 15 mm. 12 mm. Anterior height of canine 23 mm. 14 mm. (in straight line from crown-root junction to tip) M. tropicalis also has small canines. The measurements given below of the upper canines of an adult male M. tropicalis (1889-11.5.1; condylobasal length 267 mm., suture age 25) are compared with those of an adult male M. mona- chus of approximately similar size (1863-4.1.1; condylobasal length 273 mm., suture age 26). M. mona- M. trop- chus icalis Anteroposterior length 16.5 mm. 11mm. Anterior height 27 mm. est. 17 mm. (v. worn) The height of the crown of the postcanine teeth in both M. tropicalis and M. schauinslandi is lower than in M. monachus and the main cusp is more rounded. In this respect M. schau- inslandi is more like M. tropicalis, but in the possession of a single anterior and posterior cusp it resembles M. monachus . The fourth post- canine has, however, two small anterior cusps and two small posterior cusps. The surface of the teeth is more rugose than in M. monachus and the anterior and posterior cusps are very much less distinct, as though it is in process of losing the second posterior cusps of M. trop- icalis. The postcanine teeth are not set obliquely. lower JAW TEETH: The lower incisors are similar to those of M. tropicalis, the canines are small, and the postcanines are similar to those in the upper jaw. No disease or irregularity in number is pres- ent in the teeth of either upper or lower jaws. Skeleton SCAPULA: The scapula of M. schauinslandi is very like that of M. tropicalis. It is similar in the way that the anterior edge is directed almost horizontally forwards from the neck before sweeping round to the dorsal surface. In M. monachus the anterior edge is directed forwards and upwards at an angle of approximately 45°. The spine in the young Laysan seal is repre- sented by a low ridge with a well-developed acromion process (Fig. 4b). HUMERUS, RADIUS, ULNA: No real compari- son can be made because of the youth of M. schauinslandi, though the humerus appears to be slightly more robust than that of M. mona- chus of similar size. MANUS: Except that the terminal phalanges are not so extensively grooved for claws as in the young M. monachus, although the claws themselves are of approximately similar size, the manus is not noticeably different in the three species (Fig. 4c). PELVIS: The pelvis is very similar in shape to that of the young M. monachtis although it is more mature as there is no sign of the sym- physis between ischium and pubis, and the acetabulum is deep and well formed, while in the young M. monachus the symphysis is about half fused and the acetabulum is shallow and more obviously immature. The shape of the ischium and pubis is like that of M. monachus except that the pubis is possibly slightly more slender. There is no sign of the stout pubis and very narrow ischium posterior to the ischiatic spine as in M. tropicalis. The ilium is slightly narrower than in M. monachus. There is a large foramen for the obturator nerve just posterior 288 PACIFIC SCIENCE, Vol. XV, April 1961 Fig. 4. a, Right ilium, b, Right scapula, c, Right manus. d, Right pes. Hawaiian Monk Seal — King and Harrison to the cotyloid notch that is not present in M. monachus or M. tropicalis (Fig. 4a) . FEMUR, TIBIA, FIBULA: These bones are not sufficiently well formed to be used for any comparison. PES: The previous description of the pes of M. monachus applies also to the pes of the present animal. There is practically no indica- tion of the insertion of the small claws on the terminal phalanges (Fig. 4 d) . VERTEBRAL COLUMN: Vertebral formula: cervical 7, thoracic 15, lumbar 5, sacral 3, caudal 13. Cervical vertebrae. These are more like M. tropicalis than M. monachus in that the trans- verse processes of vertebrae 4-6 are divided into two branches, though the division is less pro- nounced than in the adult M. tropicalis. There is a general similarity to M. monachus, though this is probably more because of the similarity in age. The neural arches have completely fused. Thoracic vertebrae. These are similar to those of the young M. monachus except that the neural arches are narrower and do not lean so far posteriorly. Lumbar and caudal vertebrae. These are as previously described (King, 1956). RIBS: The articulation of the ribs is similar to that in the other monk seals. In order to in- ject the vascular system and remove the soft parts the cartilaginous parts of the ribs were cut away. ABDOMINAL VEINS The abdominal veins are thin and easily dis- tensible. The posterior vena cava is duplicated as in Phoca but displays a complicated arrange- ment of large anastomotic channels (Fig. 5 ) not hitherto described in other Pinnipedia. The right limb of the posterior vena cava is the larger, is almost straight, and lies a little to the right of the midline. The smaller left limb ex- tends from the pelvis on the left, passing some- what obliquely cranialwards to the right to join the right limb near the upper pole of the left kidney. A large anastomotic channel passes from the right limb at the level of the lower pole of the right kidney obliquely across the midline to join the left limb at the level of the middle of the left kidney. A smaller channel arises from the right cranial end of the anastomosis just 289 described, passes dorsal to the right limb, and enters it on the right at the level of the upper pole of the right kidney. Each limb and the two anastomotic channels receive numerous tribu- taries draining the renal stellate plexus as well as many vessels from the lumbar and pelvic venous plexuses. The right limb of the posterior vena cava is 2.5 cm. in diameter where it is formed by union of the iliac, lumbar, and most caudal renal tributaries. It is 3.0 cm. in diameter where it is joined by the left limb. The common trunk is 3.0 cm. in diameter throughout the 8.0 cm. of its extent to the point where it is enclosed by hepatic tissue. The common trunk of the posterior vena cava is enlarged considerably where it enters the sub- stance of one lobe of the multilobed liver. It has the form of a dilated tube, 15 cm. in length and 6 cm. in diameter in its cranial portion, lying on the ventral surface of the liver and sur- rounded on three sides by a narrow strip of hepatic tissue. Several large orifices of hepatic veins are present on the lateral walls of this dilated part of the posterior vena cava. Cranially this part of the posterior vena cava enters a hepatic sinus, nearly spherical in shape and ap- proximately 10 cm. in diameter. The sinus is partially surrounded by hepatic tissue, but in regions only a thin translucent wall covered by peritoneum separates it from the diaphragm. The sinus is divided by two narrow septa arising from its right wall. Six large hepatic veins open into the sinus. The capacity of the sinus is estimated to be 450 cc. Figure 6 shows the ap- pearances of the dilated vena cava and the he- patic sinus. The intrathoracic part of the posterior vena cava is 5 cm. in length and 3.5 cm. in diameter. No pericardial plexuses of vessels, such as are found in Phoca, were present and no veins drained into this part of the vena cava. An in- complete sphincter of striated muscle encircles the vena cava just cranial to the diaphragm. The dorsal part of the sphincter was 3.5 cm. high and was closely adherent to the vena cava; it was separated from the diaphragmatic muscle by a narrow strip of connective tissue. The fibres of the sphincter only partially encircle the vena cava so that on its ventral aspect the sphincter is narrowed to a bundle of closely packed fibres 290 PACIFIC SCIENCE, Vol. XV, April 1961 FIG. 5. A drawing from the ventral aspect to show the duplicated posterior vena cava, its anastomotic branches, and the stellate renal plexus on the surface of the multilobular kidney. 0.5 cm. high. The sphincter is shaped like a signet ring. It is supplied by branches of the right phrenic nerve. KIDNEYS The right kidney lies 2 cm. more cranial than the left. The right is 18 cm. long, 8 cm. broad, and 3.5 cm. thick; the left is 16 cm. long, 7 cm. broad, and 3.5 cm. thick. Each kidney is com- posed of large numbers of renules, each about 2.0 cm. in diameter. The papilla of each renule projects into a single calyx with a ductule that unites with others to drain eventually into the pelvis of the ureter as in Phoca (Harrison and Tomlinson, 1956). Each kidney is surrounded by a stellate venous plexus, the communicating vessels of which lie in the sulci between the renules on the surface of the kidney. The plexus is more marked on the ventral aspect of the organ. There are numerous anastomoses with lumbar and lateral wall veins, with the intravertebrai vein, and with the azygos vein. The plexus is drained mainly by a series of tortuous tributaries that extend round the upper and lower poles and transversely across the ventral and dorsal surfaces of the body of the kidney. These tributaries unite at the medial border of each kidney to form three or four short trunks that enter the two main limbs of the posterior vena cava or its large anastomotic Hawaiian Monk Seal — King and Harrison 291 channels (Fig. 5). The major portion of venous blood is conveyed by interlobular veins reach- ing the surface between renules to enter the stellate plexus. Some interlobular veins, how- ever, pass towards the hilum of the kidney to drain into small channels that extend medially to the limbs of the posterior vena cava. HEART AND GREAT VESSELS The aorta is markedly constricted at a point immediately to the left of the origin of the left subclavian artery and below the ductus arteri- osus. This condition is known as coarctation ( coarctus — pressed together) and is the result Fig. 6. A drawing from the right side of a sagittal section through the dilated intrahepatic part of the posterior vena cava, the hepatic sinus, and the sphinc- ter about the intrathoracic part of the vena cava. The hepatic sinus is divided by thin, falciform septa; the main openings of the hepatic veins are shown as black circular areas. of partial obliteration of the dorsal aorta either between the 4th and 6th arch ( above the ductus arteriosus) or below the 6th arch and the dorsal aorta (below the ductus arteriosus). It occurs rarely in man: Wood (1956) found coarctation of the aorta in 9 out of 900 cases of congenital heart disease. It appears to be very rare indeed in mammals and has not been reported in any animals dying at the Zoological Gardens, Re- gent’s Park, London (R. W. Fiennes, personal communication). Cordy and Ribelin (1950) describe its occurrence in a bullock associated with dextraposition of the heart. It occurs, in man, more frequently in males (4.5: 1 ) , is most often found in young adults and 1 per cent of the cases have hereditary links (Wood, 1956). The transverse diameter of the monk seal aorta at the point of coarctation is 1.0 cm., that of the first part of the descending aorta is 1.4 cm. There does not appear, therefore, to be any post-stenotic dilatation of the aorta as is often found in man. The ascending aorta and its arch are dilated with marked thickening of the wall. The most dilated part is 4.5 cm. in diameter; the thickened wall is 3.0 mm. thick as opposed to the wall of the descending aorta, which is 1.0 mm. thick. At the point of coarctation the wall of the aorta is thickened by fibrous tissue to 4.0 mm. cranially and to 3.0 mm. caudally; the other parts of the aortic wall are less thick. Three aortic valves are present (only two are present in 23 per cent of human subjects with coarctation, Hamilton and Abbott, 1928). The aortic ring appears of normal size (aortic ste- nosis is present in 7.5 per cent of cases in man). The left ventricular musculature appears hy- pertrophied, but otherwise the heart is normal. There is no patent foramen ovale and the ductus arteriosus is closed (7 per cent of human cases show a patent ductus ) . The right and left atria appear to be of normal size and have walls that do not look hypertrophied. There is no evidence of enlargement of vessels that provide collateral circulations above and below the constriction (internal mammary arteries). No notching of the posterior margins of the ribs (which can be caused by raised blood pressure in the inter- costal vessels) is present. The lack of any such findings could well be due to the immaturity of the animal. 292 PACIFIC SCIENCE, Vol. XV, April 1961 OTHER VEINS A large azygos vein is present just to the right of the midline; it is 1.25 cm. in diameter and terminates in the anterior vena cava. No left- sided azygos vein is present. An extradural intravertebral vein is present; at the level of the 12 th thoracic vertebra it lies to the left of the cauda equina. At this level it measures 1.75 cm. by 1.25 cm. It is somewhat larger at the level of the 3rd thoracic vertebra. The poor preservation of the specimen pre- vented detailed examination of the connexions of the vein. It definitely has connexions with the azygos vein and with the lumbar veins related to the stellate renal veins at the lower poles of the kidneys as in Phoca. These are clearly shown after the venous system had been injected with coloured gelatin. LUNGS Superficial examination (no casts were made) suggests that the arrangement of the bronchial tree is symmetrical as has been described in other seals (Brown, 1958). Histological examination reveals the presence of numerous features described by Pizey ( 1954) in the lung of Phoca. The cartilages of the tertiary bronchi are con- tinued far to the periphery of the lung, and bronchial mucous glands are frequent. The lung is divided into numerous lobules by well-marked septa of loose connective tissue. The bronchioles possess a series of myo-elastic valves which are similar to those of Phoca but not as marked as those in Tur slops (Wislocki, 1929). TESTIS Each testis weighs 1.9 g. and measures 25 X 15 X 10 mm. in its greatest diameters. The sem- iniferous tubules are inactive, immature, devoid of lumina, and average 50/ /x in cross-sectioned diameter. The interstitial tissue is relatively abundant. Some of its cells are large, polyhedral, and heavily vacuolated, but the majority are small, fusiform, and have densely stained nuclei. The connective tissue between the seminiferous tubules is loosely arranged and oedematous. These appearances suggest that the gonad may well have been precociously enlarged at birth with hypertrophy of the interstitial tissue as has been described in Halichoerus, Phoca, and Mi- rounga (see Harrison, I960, for references). DISCUSSION It is difficult to come to any definite conclu- sions from the examination of this skeleton of Monachus schauinslandi , not only because of the lack of comparative material but also because of the extreme youth of the animal. It is obvious that it is more closely related to M. tropicalis than to M . monachus but this is to be expected because of its geographical position. Kenyon and Rice (1959) suggest that further study may indicate a closer relationship between the Lay- san and the West Indian monk seals and that M. schauinslandi may possibly be a race of M. tropicalis. The Laysan monk seal possesses certain vas- cular ( venous ) modifications seen in other Pin- nipedia. These are a duplicated posterior vena cava, a stellate renal plexus, a caval sphincter, a hepatic sinus, and an extradural intravertebral vein (Harrison and Tomlinson, 1956). There are, however, certain differences. The pattern of duplication of the posterior vena cava is more complicated than in any seal so far described. There is evidence of persistence of several anastomotic channels between the two limbs of the posterior vena cava. This could be inter- preted as persistence of an embryonic state in which primitive anastomoses have become en- larged rather than suppressed. The hepatic sinus is not as large as in Phoca or Mirounga , whereas the curious dilatation of the intrahepatic part of the vena cava is undescribed in seals. The sphinc- ter is not as large or as complete as in other Pinnipedia. It could be argued that the monk seal shows less vascular (venous) specialization than Phoca, Mirounga, Leptonychotes, Lobo- don, Halichoerus, and Hydrurga, but more than Zalophus. This could mean that monk seals are not able to dive for so long a period or as deep as these forms. No observations have been made on the diving abilities of the Laysan monk seal, but Kenyon and Rice (1959) note that these seals occur regularly only on islands having ex- tensive areas of shallow shoal water and that they appear to feed primarily on bottom-living fishes that they could obtain only in shallow water. They do travel over deep water, though not necessarily at any great depth. It must, how- ever, be emphasized that the specimen described here had coarctation of the aorta. We are not Hawaiian Monk Seal — King and Harrison certain that its venous pattern is that prevail- ing in all Laysan monk seals, and it is frustrating that the first whole specimen available for exam- ination should be congenitally abnormal. It is hoped that continuing interest in this rare and relatively unknown seal will, in due course, result in the acquisition of more speci- mens for study. In the meantime it has been a great pleasure to see and work on the present animal and we are much indebted to Mr. Vernon E. Brock and his fellow workers for their kind- ness. We are also grateful to Dr. J. D. W. Tom- linson for injecting the specimen and for making some initial observations. SUMMARY A young male monk seal Monachus schauins- landi of an estimated age of 5 weeks has been received by the British Museum (Natural His- tory). A brief description is given of the ex- ternal features, stomach contents, and parasites. The skull and skeleton are described and a gen- eral similarity to that of M. tropicalis is noted. Certain modifications in the venous system are described; coarctation of the aorta is present. REFERENCES Abbott, M. E. 1928. Coarctation of the adult type. Amer. Heart J. 3: 574-618. Brown, D. 1958. The bronchial tree in aquatic mammals. J. Anat. Lond. (Proc.) 92: 656. Cordy, D. R., and W. E. Ribelin. 1950. Six congenital cardiac anomalies in animals. Cor- nell Vet. 40: 249^-256. 293 Hamilton, W. F., and M. E. Abbott. 1928. Coarctation of adult type. Amer. Heart T. 3: 381-492. Harrison, R. J., and J. D. W. Tomlinson. 1956. Observations on the venous systems in certain Pinnipedia and Cetacea. Proc. Zool. Soc. Lond. 126: 205-233. Harrison, R. J. I960. Reproduction and re- productive organs in common seals ( Phoca vituina ) in the Wash, East Anglia. Mam- malia ( in press ) . Kenyon, K. W., and D. W. Rice. 1959. Life history of the Hawaiian monk seal. Pacif. Sci. 13(3): 215-252. King, J. E. 1956. The monk seals ( Genus Monachus ). Bull. Brit. Mus. (Nat. Hist.) Zool. 3(5): 203-256. Matschie, P. 1905. Eine Robbe von Laysan. S. B. Ges. Naturf. Fr. Berl. 1905: 254-262. PlZEY, N. C. D. 1954. The structure of the pin- niped lung. J. Anat. Lond. (Proc.) 88: 552. SviHLA, A. 1959. Notes on the Hawaiian monk seal. J. Mamm. 40: 226-229. Wahlert, G. von. 1956. Die Typen und Ty- poide Des Uberseemuseums Bremen, 4. Die Laysan-Robbe, Monachus schauinslandi Mat- schie 1905. [?}. Veroff Uberseemuseum Bremen, A.2. Heft 6: 365-366. WlSLOCKl, G. B. 1929. On the structure of the lungs of the porpoise ( Tursiops truncatus). Amer. J. Anat. 44: 47—77. WOOD, P. 1956. Diseases of the Heart and Cir- culation. Eyre and Spottiswoode, London. Hermaphroditic Skipjack 1 Richard N. Uchida 2 Two PAIRS of skipjack ( Katsuwonus pelamis ) gonads, each having male and female compo- nents, have come to my attention in the past 3 years. The first was received from Mr. Thomas S. Higa, proprietor of a Honolulu fish retail market, in April 1937. Mr. Higa recovered the gonads from a neighboring fish retailer who had discovered them in a freshly eviscerated 16-lb. skipjack. This fish was captured by the "Orion,” a commercial skipjack vessel, on April 11, 1957, off Makua, Oahu, from a school of 15- to 18-lb. skipjack. The ovo-testes ( Fig. 1 ) were typically paired and elongate, joined posteriorly, and weighed 121 gm. (fresh weight). The male and female components were easily discernible. The left gonad was divided into three segments; the an- terior one-sixth and posterior one-third were ovarian and the remainder testicular. The right gonad was divided into two segments; about two-thirds of the anterior portion was testicular and the remainder ovarian. Figure 2 shows the length of each of these segments. The cream-colored testicular sections were solid and somewhat flattened in cross-section, and did not seem to be atypical in any way. There was no running milt present, but their size suggested that they were in a rather ad- vanced stage of maturity. All ovarian sections were pinkish, but differed in firmness. The pos- terior segments were hollow and flabby, with a ribbed internal cavity extending their entire length. The single anterior ovarian section was nearly round in cross-section and rather turgid. Detailed examination of the ovo-testes was made after they had been preserved in 10 per cent formalin. The longitudinal duct of each 1 Published with permission of the Director, U. S. Bureau of Commercial Fisheries. Manuscript received July 29, I960. 2 Fishery Research Biologist, Honolulu Biological Laboratory, U. S. Bureau of Commercial Fisheries, Honolulu, Hawaii. testicular section could be followed posteriorly to the juncture of the testicular and ovarian segments, after which the ductus deferens be- came obscured in the network of blood vessels on the surface of the ovarian section. No open- ings of the ductus deferens were detected any- where in the area of the urogenital sinus. The two posterior ovarian sections were joined posteriorly by a single oviduct which opened into the urogenital orifice. No duct could be found in the single anterior ovarian segment. - Microscopic examination of the ova was made to determine their degree of maturity. The lu- men of the anterior ovarian section was filled Fig. 1. The ovo-testes of a hermaphroditic skipjack (ventral view). 294 Hermaphroditic Skipjack — UCHIDA 295 ANTERIOR POSTERIOR Fig. 2. Oudine of the ovo-testes, showing the loca- tion and size of the male and female components (ventral view). with a compact mass of resorbing ova, while the peripheral portion contained ova ranging from small, transparent stages to larger, opaque ova. The average diameter of 25 randomly se- lected larger, opaque ova was 0.504 mm. The posterior ovarian sections contained similar small and large ova, but no resorbing ova were present. Average diameters of randomly selected ova from these parts were 0.543 mm. for the left posterior member and 0.514 mm. for the right. The presence of residual ova in the anterior section suggests that the fish had been a func- tional female. Assuming that the ova in each of the ovarian sections ripened at the same time, it is possible that ova were extruded from the posterior sections at spawning time, whereas ova in the anterior section could not be extruded because of the lack of an adequate duct. This could account for the presence of the resorbing ova in the lumen of the anterior ovarian sec- tion. Since the testicular sections were well de- veloped and constituted a large part of the gon- ads, the fish may also have been a functional male. The discovery of the second pair of ovo-testes was reported to the Honolulu Biological Lab- oratory on March 10, I960, by Mr. Richard Na- kashima of Honolulu, who found them in a 10-lb. skipjack. The gonads, outlined in Figure 3 and labeled A and B for purpose of identification, were not joined posteriorly as a result of mutilation to both posterior ends during removal, and, there- fore, could not be identified as to position ( left or right ) . Both gonads were distinctly separated into three parts; an anterior ovarian section, a middle testicular portion, and a torn, fragmen- tary ovarian segment posteriorly. The testicular segments were similar to those found in the ovo- testes previously described, and contained no running milt. The ovarian sections were pre- dominantly yellow with a tinge of pink. The anterior segments were flabby and hollow. A starch suspension colored with powdered carmine was injected into the posterior ovarian section of gonad A and the flow of injected material indicated the presence of a duct con- necting the two ovarian parts. A similar injec- tion was attempted with gonad B, but failed MM. Fig. 3. Outline of the ovo-testes, showing the location and relative size of the male and female components. 296 PACIFIC SCIENCE, Vol. XV, April 1961 to reveal any ducts owing to excessive tissue damage. Ova diameter measurements revealed that the posterior ovarian section of gonad A and both ovarian segments of gonad B were identical in their degree of development. The ova ranged from small, primitive ova to larger, opaque ova. Average diameters of 25 randomly selected ova of the most advanced group were 0.479 mm. for the posterior ovarian section of gonad A and 0.476 mm. and 0.479 mm. for the anterior and posterior ovarian sections, respectively, of gonad B. The development of the anterior ovarian sec- tion of gonad A was dissimilar to any of the sections previously examined. Its lumen con- tained a loose mass of large, ripe, resorbing ova, together with a few completely degenerated ova. The peripheral portion of the section con- tained only primitive ova. Those intermediate to the primitive and the large, resorbing ova were absent. The presence of residual ova indicates that this fish also had been a functional female. Again, it is not clear whether the ova remnants were the unexpelled portions of a previous suc- cessful spawning or a group of ripe ova that were not expelled owing to the inadequacy of the duct. Damage to the posterior portions of both gonads made it impossible to determine whether the male ducts were connected to the exterior, but since the testicular segments were well developed, this fish may also have been a functional male. Hermaphroditism in skipjack is an extremely rare occurrence. So far as is known, only one other example has been recorded and described (Nakamura, H., 1935, Trans. Nat. Hist. Soc. Formosa 25(141): 197-198; in Japanese). Studies on Pacific Ferns, Part IV The Pteridophyte Flora of Pitcairn Island G. Brownlie 1 Although the ferns of southeastern Poly- nesia have been described in several papers, no separate list has previously been provided for Pitcairn Island. The present paper outlines the details of a collection made there by Mr. W. H. Lintott of the Botany Department of the Uni- versity of Canterbury, and includes mention of other collections where these were available. When the changes which have occurred in the indigenous vegetation are considered, it is re- markable that, as well as finding all the other species noted earlier, one species not previously recorded from Pitcairn is included in the most recent collection. Of the 20 species listed, 15 are widespread throughout Polynesia, and 2 others are prob- ably local derivatives of similarly widespread species. This large group represents the dom- inant Malay-Papuan element found in decreas- ing numbers of species from west to east in the tropical Pacific. Of the remaining 3, 1 species of Asplenium and 1 of Trichomanes are found in S.E. Polynesia and New Zealand, and are the only ones to which Copeland’s ( 1938) ideas of an Austral group can be applied. The single species of Cyathea is related to ferns found in Rapa and the Society Islands. There are only 2 species recognised as endemics, and each of these may in fact be better regarded as varieties. Viewed as a whole, the pteridophyte flora of this region, Pitcairn, Rapa, and the Australs, appears to be merely an extension of that of the Society Islands. Abbreviations: C, Department of Botany, University of Canterbury; W, Dominion Mu- seum, Wellington; S, Department of Agricul- ture, Suva, Fiji; K, The Herbarium, Royal Bo- tanic Gardens, Kew. 1 Department of Botany, University of Canterbury, Christchurch, New Zealand. Manuscript received April 25, I960. PSILOTACEAE Psilotum nudum ( L. ) Grisebach Pendant on trees and shady rocks; not com- mon. Widely distributed throughout Polynesia. Lintott 167a and 167b (C), Williams 3068 ( W ) , Twyford S.P. 129 (S). MARATTIACEAE Angiopteris chauliodonta Copeland On damp hillsides in dense shade. I doubt- fully retain this as a distinct species, and it may be only a local form in which the sterile apices of the pinnules are more toothed than is usual in A. eve eta. Lintott 150 (C), Twyford S.P. 130 (S). GLEICHENIACEAE Dicranopteris linearis (Burm.) Und. Gleichenia linearis (Burm.) Clark. Brown, B.P. Bishop Mus. Bull. 89, 97, 1931. Copeland, Occ. Pap. B.P. Bishop Mus. 14(5): 52,1938. Forming tangled masses on dry hillsides. Widely distributed in the tropics and subtropics of the Old World with many recognized vari- eties in S.E. Asia. It is probable that there are also several varieties in the Pacific area, but these have not been worked out. Lintott 163, 164 (C), Twyford S.P. 135 (S), Williams 3038, 3081 (W), Fosberg 11221 (K), Quayle No. X (K). HYMENOPHYLLACEAE Trichomanes endlicherianum Pr. Damp rocky faces in shady stream bed. This is the only record of a member of this family from Pitcairn, and as it appears to be fairly un- common, it has presumably been overlooked by previous collectors. The specimens are some- 297 298 PACIFIC SCIENCE, Vol. XV, April 1961 what smaller than those from the Austral Is- lands and New Zealand, but they have the char- acteristic single row of elongated marginal cells. The species is found in New Zealand, the Ker- madecs, Samoa, Fiji, Tahiti, the Australs, and Rapa. Lintott 155 (C). PTERIDACEAE Adiantum hispidulum Sw. In isolated clumps on dry slopes and in rocky crevices. All examples are considerably smaller than those seen from the larger Pacific islands. The species is widely distributed in the tropics and subtropics from Africa to Polynesia and south to New Zealand. Lintott 152, 152b (C), Twyford S.P. 51 (S), St.John 15031 (K). CYATHEACEAE Cyathea cumingii Bak. Tree fern up to 10 ft. high, occurring in sev- eral places on the island. Elsewhere known only from the Australs, but C. rapensis and C. so- cietarum are very similar, if not identical. Lintott 165 (C), Williams 3010, 3011 (W), Twyford S.P. 132 (S), Quayle No. 1 (K), Cuming 1393 (K) , Matthews No. 7 (K) , Fos- berg 11241 (K), St. John 14978 (K). DAVALLIACEAE Davallia solida (Forst.) Sw. Common everywhere, creeping on the ground and as an epiphyte on old trees. Four local forms have been recognized in S.E. Polynesia by Brown (1931), but the species is every- where somewhat variable in degree of division of the fronds and in shape of the indusia. It is distributed from Burma to Pitcairn. Lintott 162 (C), Williams 3013, 3052 (W), Twyford S.P. 133, 134 (S), Quayle No. X (K), St. John 15048 (K). Nephrolepis biserrata (Sw.) Schott Common throughout the island underneath Pandanus. Christensen ( 1943 ) regards the Poly- nesian-Asiatic forms as probably distinct from genuine African N. biserrata, but until a com- plete revision of the genus has been carried out it is preferable to retain the well-known name. The distribution of the species in its widest sense is pantropic. Lintott 158 (C), Williams 3074, 3079 (W), Twyford S.P. 59 (S), Quayle No. 1 (K). Nephrolepis hirsutula (Forst.) P'r. Fairly common, particularly around clearings. Very variable in all its characters, often ap- proaching close to N. biserrata, although it is usually smaller than that species. Distributed from tropical Asia to Pitcairn. Lintott 169 (C), Williams 3051, 3078 (W), Twyford S.P. 136 (S). ASPIDIACEAE Thelypteris uliginosa (Kze.) Ching Dry op ter is setigera (BL) Kze. Brown, B.P. Bishop Mus. Bull. 89, 30, 1931. On cleared banks along shady tracks. I have seen no fertile specimens of this fern from Pit- cairn, and it appears to be relatively uncommon. The species is widespread in the tropics and subtropics from Asia through the Pacific to Pitcairn. Lintott 168 (C), Twyford S.P. 131 (S) . Rumohra aristata (Forst.) Ching Polystichum aristatum (Forst.) Pr. Brown, B.P. Bishop Mus. Bull. 89, 38, 1931. Copeland, Occ. Pap. B.P. Bishop Mus. 14(5) : '57, 1938. Forming small patches under trees on hill- side ridge of Adamstown. Brown was in error in comparing specimens of this fern from Rapa with New Zealand forms, for, although found in the Kermadecs, it does not occur on the mainland of New Zealand. Distribution is throughout the tropics and subtropics of the Old World. Lintott 154 (C), Quayle No. X (K), St. John 14969 (K). Ferns of Pitcairn Island — Brownlie 299 Cyclosorus parasiticus (L.) Farwell Dryopteris parasitica (L.) Kze. Brown, B.P. Bishop Mus. Bull. 89, 22, 1931. Dryopteris dentata (Forst.) C. Chr. Cope- land, Occ. Pap. B.P. Bishop Mus. 14(5): 56, 1938. Common everywhere in shady positions. The habit of these specimens is almost intermediate between typical C. parasiticus and C. nymphalis, but since the leaves are not fascicled it is better to place it in the former species. I feel that Brown’s recognition of a local variety based chiefly on characters of size is, however, un- sound. Widely distributed from tropical Asia through the Pacific. Lintott 156 (C), Williams 3049, 3076 (W), Twyford S.P. 100 (S). Athyrium polyanthes (Sol. ex Bak.) Copel. Diplazium polyanthos (Sol.) C. Chr. Brown, B.P. Bishop Mus. Bull, 89, 55, 1931. Athyrium pitcairnense Copel., Occ. Pap. B.P. Bishop Mus. 14(5): 60, 1938. Diplazium harpeodes Moore. C. Chr., B.P. Bishop Mus. Bull. 177, 76, 1943. Not common; found in isolated clumps in shady valleys. Much of the confusion concern- ing the nomenclature of this species was cleared up by Christensen (1943), but I prefer to fol- low Copeland in uniting Diplazium and Athyr- ium. Copeland’s local species falls within the range of the species as outlined by Christensen. Distribution is throughout the Pacific islands from Fiji to Pitcairn. Lintott 166 (C), Cuming 1389 (K) , Mat- thews unnumbered (K). BLECHNACEAE Doodia media R. Br. Fairly common on open ground and under- neath Pandanus. Distributed from Australia and New Zealand to Pitcairn, but absent from Sa- moa and Tahiti. Lintott 159a and 159b (C), Williams 3044 (W), Twyford S.P. 133, 134 (S), St. John 14976 (K). ASPLENIACEAE Asplenium nidus L. Common in one or two valleys and cultivated in local gardens. Widely spread throughout the tropics of the Old World. Lintott 153 (C). Asplenium ohtusatum Forst. On rocks on the south coast, and in small caves. Always within or close to the spray zone. The specimens are similar to those found on Rapa, the Australs, and other Pacific islands, but are somewhat smaller than the typical state represented in New Zealand. Elsewhere the spe- cies is found in Juan Fernandez and southern Chile, with closely related, if not identical, spe- cies on Tristan da Cunha and Tasmania. Lintott 170 (C), Williams 3083 (W), Twy- ford S.P. 71, 73 (S), Fosberg 11343 (K). Loxoscaphe gibberosum (Forst.) Moore LoxosCaphe gibberosum var. pitcairnense Brown, B.P. Bishop Mus. Bull. 89, 67, 1931. Common everywhere in deep shade. This spe- cies shows so much variation in the degree to which the leaf segments extend beyond the sorus that I feel that Brown’s varieties are dif- ficult to maintain. It is found in the Pacific re- gion from Fiji to Pitcairn, but is absent from Samoa. Lintott 157 (C), Williams 3012, 3053 (W), Twyford S.P. 43 (S), Matthews unnumbered (K), Cuming 1373 (K), Fosberg 11297 (K). POLYPODIACEAE Pyrrosia angustata (Sw.) Ching Cyclophorus angustatus (Sw.) Desv. Brown, B. P. Bishop Mus. Bull. 89, 93, 1931. Common on rocks in inland situations, and as an epiphyte on large trees. The species is 300 widely distributed in tropical areas from Malaya to Polynesia. Lintott 161a, l6lb (C), Williams 2997, 3075, 3082 (W), Twyford S.P. 153 (S), Mat- thews unnumbered (K), Cuming 1394 (K) . Phymatodes pitcairnense (Copel.) Brownlie, comb. nov. Polypodium phymatodes L. Brown, B.P. Bi- shop Mus. Bull. 89, 87, 1931 (in part). Micro sorium pitcairnense Copel. Occ Pap. B.P. Bishop Mus. 14(5): 74, 1938. Common, creeping on banks along paths. This is a close relative of P. nigrescens, but with- out the distinct veins of that species. Unlike Copeland, I do not find that the sod of any of the specimens are relatively marginal, so that these may not be the same as his species. How- ever, until a full revision of the Pacific members of this genus is undertaken, I feel that it is unwise to introduce further new names for minor differences. Lintott 160 (C), Williams 3067, 3077, 3084 (W), Twyford S.P. 36, 65 (S), Matthews No. 12 (K). PACIFIC SCIENCE, Vol. XV, April 1961 VITTARIACEAE Vittaria elongata Sw. Local, in one area growing on moist rocks. The limitations of this species are somewhat uncertain in the Pacific area, and the Pitcairn example is doubtfully assigned to it. It may in fact be closer to V. rigida var. samoensis de- scribed by Christensen ( 1943 ) . In the widest sense the species extends from tropical Asia to Polynesia. Lintott 151 (C),Fosberg 11309 (K). In addition to the above-listed species there are two sheets at Kew (Cuming 1388) identi- fied as Polypodium sandwicense. The specimens belong to the genus Ctenitis and approach close to Dryopteris samoensis (C. Chr 1943), but the fact that it has not been found by any subse- quent collector indicates that it is probably extinct on Pitcairn or has been wrongly ac- credited to that island. I wish to thank the professors W. R. Philip- son and Harold St. John for making the Lintott collection available for study, and I wish to acknowledge the use of material from the Do- minion Museum, Wellington, the Department of Agriculture, Suva, and the Herbarium, Royal Botanic Gardens, Kew. The Principal Weedy Melastomaceae in Hawaii D. L. Plucknett and B. C. Stone 1 None of the numerous species of Melasto- maceae is native to the Hawaiian Islands, yet at present at least a dozen species in nine different genera are found as weeds or seminaturalized plants in various localities in the Islands. Some of these species are hardly known out of cul- tivation, others are frequent in certain small areas, while a few are well-known common plants on most of the islands in the chain. Cer- tain areas have become thickly populated with one or another of these species, and waste and pasture areas have been invaded, becoming in some cases nearly impenetrable thickets. Two or three of these species must be regarded as noxious weeds. In Hillebrand’s Flora of the Hawaiian Is- lands (1888) there is no mention of any melas- tomaceous species. Degener (1935: family 274) mentions Heterocentron, and (1930: 226) mentions Tibouchina semidecandra and Melas- toma malab athricum. Earlier, W. T. Pope (1929: 147) had mentioned Melastoma decem- fidum and stated that it had escaped from cul- tivation on Kauai, and had been introduced (presumably from a Florida nursery) in 1916. It seems from recent collections that this plant mentioned by Pope is, as Degener indicated, M. malab athricum; although both species are pres- ent in the Islands, only this one has been found on Kauai. The Tibouchina is said to have been introduced to Hawaii in 1910. Degener indicated the pestiferous nature of these plants, and his prediction that they would spread has come true. At present they have a very spotty distribution, but where they have spread they often form virtually pure stands which have blighted many areas of natural veg- etation and have become a foe of both the agri- culturist and conservationist. Hosaka ( 1945 ) referred to Melastoma malabathricum as "an ag- 1 Department of Agronomy and Soil Science, and Department of Botany, University of Hawaii, Hono- lulu. Manuscript received June 15, I960. gressive shrub that forms dense stands — crowds out other plants . . . has no forage value.” Recently the extent of bauxitic or aluminous soils in Hawaii has been publicized (Sherman, 1954), and plants which accumulate aluminum have been the subject of a paper by Moomaw, Nakamura, and Sherman (1959). Among the Hawaiian plants found to accumulate aluminum was M. malabathricum, a known accumulator of aluminum in other areas (Webb, 1954). None of the other melastomaceous plants in Hawaii was used in the study by Moomaw et al., but other species of the Melastomaceae are re- ported by Webb to accumulate aluminum. The purpose of this paper is to discuss the distribution and spread of the noxious Melas- tomaceae in Hawaii with special reference to the most common species and to provide a key and brief descriptions of each as a preliminary step for a possible study of bauxitic soils and aluminum-accumulating species of Melastoma- ceae. Three species are found on more than one island and may be considered common. These are Tibouchina semidecandra, Melastoma de- cemfdum, and M. malabathricum. Three other species are rather abundant in a few isolated locations on one or two islands. These are Clidemia hirta, Pterolepis glomerata, and Heterocentron subtriplinervium. The remaining species are restricted to a few small areas and do not threaten to become nox- ious weeds. These are Oxyspora paniculata, Ar- throstemma latifolium, Tetrazygia bicolor, and Medinilla magnifica. Some of these species are referred to by Neal (1948: 569). KEY TO COMMON SPECIES Petals and all anthers brilliant purple; leaves densely silvery-pubescent with soft hairs, especially on undersides; open shrubs up to 12-15 ft. tall. ...Tibouchina semidecandra (1) Petals pink, stamens pink or yellow; leaves 301 302 PACIFIC SCIENCE, Vol. XV, April 1961 pubescent but not densely silvery; with stiff hairs; compact shrubs usually 10 ft. tall or less. 1. Petioles and nodes set with sparse long stiff hairs up to 5 mm. long; calyces rather sparsely scaly; ribs be- low sparsely scaly; lateral veins more or less glabrous Melastoma decemfdum (2) 2. Petioles and stems set with numerous short spinules to 1 mm. long; nodes with a few long scales; calyces silky with numerous scales; ribs and small veinlets rather densely scaly-spiny be- low Melastoma malabathricum ( 3 ) 1 .Tibouchina semidecandra (Schrank and Mart.) ' Cogniaux distribution: Kauai: Kokee, open grass area by museum, on ditch bank by roadside, Dec. 23, 1959, D. L. Plucknett 81-85. Hawaii: N.W. Kilauea Crater, wet fern for- est around houses, 1150 m., Aug. 30, 1933, Fos- berg 10121. Volcano Road, near volcano, forest reserve along road, in mixed fern forest, tall shrub 5 m. high, Nov. 10, 1926, L. H. Mac Daniels 223. Volcano House, July 24, 1926, Aug. 24, 1926, C. S. Judd. 29 miles, Glenwood, Hawaii, July 23, 1926, Degener 8188; June 22, 1929, persistent in forested region, Degener 9641. Glenwood, el. 1000 m., in forest, Dec. 25, 1930, E, H. Bryan, Jr., 714. Kalanilehua, Aug., 1917, J. F. Rock 13029, 13030. Kurtis- town, Jan. 20, I960, along roadside, D. L. Pluck- nett 95- Hilo, May, 1932, A. Meebold. Oahu: Nuuanu Valley, upper part near Up- side Down Falls, well established in underbrush in thick woods, shrub 2 m. tall, flowers deep purple, Fosberg 27067, Sept. 15, 1946. Tantalus, grounds of E. F. Bishop, Oct. 3, 1930, H. E. Gregory. Observed, Manoa Valley, by Manoa stream, B. C. Stone, Feb. I960. Altitude range 1,500-4,500 ft.; requires moist relatively cool habitat. At lower elevations it is usually under shade, but at Kilauea Crater it may be found in the open (elev. 3,400 ft.). This species is a spectacular shrub and in cultivation is very attractive. It does not appear to be as aggressive as the two Melastoma spp. and tends to spread only in disturbed areas. It has not been found deeply penetrating the native forest, but is often very common along roadcuts, houselots, trails, and near buildings. 2. Melastoma malabathricum L. distribution : Kauai: Kilohana Crater, dom- inant shrub, 320 m., 7 Aug. 1928, E. H. Bryan 626. Reservoir near Hanahanapuni Crater, Mar. 24, I960, Plucknett 1 16, 117. Wailua bauxite project area, Mar. 24, I960, Plucknett 118-123. Hatvaii: Anauulu Rd. above Hilo, side of gully in canefield, flowers pink, Dec. 7, 1933, alt. 400 m., Fosberg 10490. Kaumana, alt. 200 ft., flowers pink, Sept. 1953, Amy Suehiro. Ke- aau Orchard- Volcano Highway Intersection, by roadside, Dec. 4, 1959, Plucknett 78. Hilo-Vol- cano Road, 4 miles mauka, Jan. 20, I960, road- side, Plucknett 93, 94. Hilo radio tower, Jan. 20, I960, Plucknett 89, 90, 92. Altitude range 0-1,000 ft. In contrast to Tib- ouchina, Melastoma malabathricum is tolerant of drier and warmer habitats and tends to spread in lowland areas rather than in disturbed areas of native forest. Usually found in open habitats such as pastures, waste areas, and fields with weedy vegetation. On Kilohana Crater on Kauai and the adjoining areas where it once was dom- inant, Rhodomyrtus tomentosa (Ait.) Hassk. appears now to be replacing it, but it still re- mains an important weedy species in this area. Pope’s reference (1929: 147) to Melastoma decemfidum on Kauai is probably in error since only M. malabathricum has been found on that island. The plant is known on Kauai as "Isen- berg bush” probably because of its association with Kilohana Crater and the Isenberg home there. Rhodomyrtus tomentosa has also been called "Isenberg bush” on Kauai but it more commonly is known as "Indian gooseberry.” On Hawaii Melastoma malabathricum may be found in great abundance in the Keaukaha area near the Hilo Airport and along the Volcano Road. M. malabathricum is vigorous in growth and spreads rapidly from numerous seeds, usually spread by birds. 3. Melastoma decemfidum Roxb. DISTRIBUTION : Hawaii: Hilo, sent to J. Kim, Honolulu, B. Ag. Forestry, 10 Dec. 1957. Hilo- Melastomaceae in Hawaii — PLUCKNETT and Stone 303 Volcano Road, 4 miles above Hilo, top of road cut, Dec. 5, 1959, Plucknett 79, 80. Altitude range 0-1,000 ft. Flourishes in open areas of high rainfall on Hawaii. Together with its weedy relative, M. malabathricum, M. de- cern fidurri has become dominant in the Keau- kaha area near the radio tower and also may be seen in thickets along the Volcano Road. This shrub has been observed as a small tree with a trunk 4-5 in. in diameter and up to 12 ft. high in the Keaukaha area. LESS COMMON MELASTOMACEOUS PLANTS IN HAWAII There are several species which seem to be spreading in certain areas but which at present cannot be classed as dangerous or even common weeds. It seems useful to indicate briefly their presently known distributions. Clidemia hwta (L.) D. Don. This rather small plant has been collected on Oahu on Mt. Tantalus, twice on the Poamoho Trail in the Koolau range, and twice near the Hawaiian Sugar Planters’ Association nursery in Wahiawa. There are no reports of Clidemia from the other islands. Pterolepis glomerata (Rottb.) Miquel. Speci- mens of this plant have been collected from Palikea in the Waianae range, and from Poa- moho and Pupukea in the Koolau range on Oahu. Heterocentron subtriplinervium (Link and Otto) Br. and Bouche. According to the labeled specimens in Bishop Museum there are four species of Heterocentron in Hawaii, but prob- ably only one or two of these species are actually represented. Neal (1948: 568) reported only H. roseum Br. and Bouche. Clarification of the spe- cies of Heterocentron will probably be pre- sented in the new edition of Neal’s In Gardens of Hawaii ( in preparation ) . There are five specimens determined as H. subtriplinervium in the Bishop Museum. These were collected from Mt. Tantalus, Oahu, and from Hawaii along the Hilo-Kona and Hilo- Kilauea roads. Arthrostemma latifolium D. Don. This plant has been found escaping at the Makiki Nursery, in Honolulu, and along south Opaeula ridge on Oahu. REFERENCES Degener, O. 1930. Plants of Hawaii National Park. Honolulu. 1935. Flora Hawaiiensis. Honolulu. Hosaka, E. Y. 1945. Noxious weeds of Hawaii. Bd. Comm. Agric. For. Honolulu. (Loose- leaf. ) Moomaw, J. C, M. T. Nakamura, and G. D. Sherman. 1959. Aluminum in some Hawai- ian plants. Pacif. Sci. 13(4): 335-343. Neal, M. C. 1948. In gardens of Hawaii. Bishop Mus. Spec. Plibl. Pope, W. T. 1929. Manual of Wayside Plants of Hawaii. Honolulu. Sherman, G. D. 1954. Some of the mineral resources of the Hawaiian Islands. Hawaii Agr. Exp. Sta. Spec. Publ. 1. Leaf and Air Temperature under Hawaii Conditions T. L. Noffsinger 1 Air temperature may give a poor indication of the actual temperature experienced by a plant leaf under various meteorological conditions. Leaf temperature is dependent upon the type and condition of the plant as well as on a num- ber of meteorological elements including wind, humidity, cloud cover, solar radiation, and air temperature. MATERIALS AND METHODS During the period from March 9 through April 15, I960, leaf and air temperatures were measured using a 12 -channel single-input ther- mister-type thermometer. Leaf temperatures of pineapple and papaya plants were measured using a 2 2 -gauge hypodermic probe. Air tem- perature was measured with the thermister air temperature probe and checked with a standard mercury-in-glass thermometer. The wet-bulb temperature was determined by a standard psy- chrometer and solar radiation was measured by a recording bimetallic actinograph (pyrheli- ometer) . Soil temperature was measured from a mercury-in-glass thermometer implanted 4 in. in the soil. In measuring leaf temperatures the needle was carefully inserted from the underside of the leaf, parallel to the leaf veins, to approximately Vi in.; i.e., at least V 2 in. of the probe was enclosed within the leaf tissue. All temperatures were read to the nearest 0.1° C. The average tempera- ture of the plant was taken as the mean value obtained from one leaf exposed to solar radia- tion and one in shade. RESULTS Data were collected at 0800, 1200, and 1600 hr. during the period of March 9 through April 15. In addition, a continuous hourly collection was made over a 24-hr. period from 0800, April 11, through 0700, April 12. 1 Climatologist, Land Study Bureau, University of Hawaii. Manuscript received May 19, I960. During the daylight hours pineapple leaf tem- perature was consistently higher than the air temperature measured in an instrument shelter at the same elevation as the plants. The values usually ranged from 1.5° to 3.5° C. above the air temperature but occasionally a leaf exposed to direct sunlight had a temperature as much as 7.6° C. higher than the air temperature. The corresponding average temperature of the papaya leaf was consistently lower than the air temperature. Temperature of leaves exposed to direct solar radiation and temperature of leaves on the same plant but shaded by higher leaves are given in Table 1 together with the average air temperature, average leaf tempera- ture, soil temperature, and solar radiation values. Table 2 gives air temperature, leaf tempera- ture, insolation values, soil temperature, cloud cover, and wind speed at 4-hr. intervals for a 24-hr. period April 11 through April 12. Maxi- mum air temperature (27.0° C. ) occurred at 1300; maximum soil temperature for the pine- apple (28.0° C. ) occurred at 1400; maximum soil temperature for the papaya (30.5° C.) oc- curred at 1500 and 1600. Maximum average leaf TABLE 1 Leaf and Air Temperature and Related Meteorological Data for March 9-April 15, I960 TIME 0800 1200 1600 Air temperature °C. 21.5 26.7 26.4 Leaf temperature °C. Pineapple average 23.2 29.1 26.6 in sun 24.7 31.0 28.0 in shade 21.7 27.2 25.2 Papaya average 21.3 25.4 24.5 in sun 21.7 25.9 24.9 in shade 20.9 24.9 24.1 Insolation Langleys /min.. 0.21 0.99 0.74 Soil temperature Pineapple 19.5 26.4 26.3 Papaya 19-5 27.1 29.8 304 Leaf and Air Temperature — Noffsinger 305 TABLE 2 Leaf and Air Temperature and Related Meteorological Data for 0800 April 11-0800 April 12, I960 TIME 0800 1200 1600 2000 2400 0400 SUNRISE Air temperature °C 22.0 25.7 26.7 22.3 22.2 21.9 22.0 Leaf temperature °C. Pineapple average 23.5 29.5 27.1 21.7 21.5 22.1 . 21.7 in sun 25.0 31.5 29-0 in shade 21.9 27.5 25.1 21.7 21.5 22.1 21.7 Papaya average 21.3 25.1 24.8 21.2 20.5 21.0 21.7 in sun 21.6 25.0 26.0 in shade 21.0 25.1 23.5 21.2 20.5 21.0 21.7 Insolation Langleys/min 0.27 1.43 0.98 Soil temperature Pineapple 20.2 26.5 27.0 23.7 21.4 21.4 20.5 Papaya 20.5 27.0 30.5 24.2 21.2 21.2 20.5 Cloud cover (1/8 S.) 5 6 1 1 1 5 1 Wind (knots) 8 8 10 5 4 0 0 temperature in the pineapple plant (29.6° C.) occurred at 1500 and in the papaya plant (26.8° C) at 1300. DISCUSSION Plants undergo irradiation from the sun, clouds, and sky during the daylight hours, and at night they radiate heat outward. Made (Geiger, 1950: 278-280) kept a continuous record of leaf temperature over a period of two days and found that during the middle of the day the leaf surface of Bilbergia nutaus (a hothouse plant of the pineapple family) was as much as 10° C. higher than the air temperature. The in- terior portion of the pineapple fruit exposed to the direct rays of the sun at latitude 2 1 ° N. may experience temperature from 5° to 8° C. above the free air temperature. In Formosa it has be- come a practice to shade the pineapple fruit in order to prevent heat damage from solar radia- tion (Ekern, personal communication). Wag- goner and Shaw ( 1952 ) have shown that energy losses from potato and tomato plants at night due to transpiration may be very small. Geiger (1950: 276) states that plant temperature is generally higher than that of air temperature when the ground surface is warmer than the air layer resting upon it, and that by night the plant is, for the most part, cooler than the air. An examination of air, soil, and leaf tempera- tures in Tables 1 and 2 shows that the gen- eralization made by Geiger holds for pineapple, but that the average leaf temperatures of the papaya were consistently lower than the air temperature even during periods of relatively high soil temperature. In one observation dur- ing the period (not shown in the table), the temperature of the papaya leaf exposed to the sun registered 2.6° C. higher than the air tem- perature. At that time the value for solar radia- tion was 1.42 calories per sq. cm. per min., the sky was five-eighths covered with cumulus and stratocumulus clouds and a light sprinkle of rain was falling from clouds immediately to the N.E. of the station. SUMMARY Air temperature in an instrument shelter at the level of the plants does not provide a good measure of plant temperature. Papaya, with a respiration and transpiration pattern character- istic of the mesophytes, shows a leaf temperature which remains relatively near the air tempera- ture but may be higher or lower than the air temperature, depending upon the condition of radiation, cloud cover, and wind. Pineapple, a xerophyte, has leaf temperature during the day- light hours which average 1.5° to 3.5° C. above 306 PACIFIC SCIENCE, Vol. XV, April 1961 the air temperature, and for short periods of high solar radiation a leaf temperature 7.6° C. above the air temperature was recorded. Gen- erally, leaf temperatures of both pineapple and papaya were below the air temperature during the night except for one period in which clouds moved into the area. At 0400, April 12, with five-eighths cloud cover and calm wind, the tem- perature of the pineapple leaf rose to 0.3° C. above the air temperature. REFERENCES Ekern, P. C. Personal communication. Geiger, R. 1950. The Climate Near the Ground. Harvard Univ. Press, Cambridge. P. 276. Waggoner, P. E., and R. H. Shaw. 1952. Tem- perature of potato and tomato leaves. Plant Physiol. 27: 710-724. NOTE The Species Commonality Index: A Method for Comparing Habitats Ecologists have used a variety of methods for comparing habitats in order to point up their similarity and to predict the efficacy of trans- plants. Some methods are based upon what are believed to be outstanding physical differences and similarities between the habitats, while others are based upon faunal and floral similari- ties and differences. The following is a method based upon faunistic similarity. For 3 years the writer has used a simple technique involving the number of species in common between two habitats in order to ar- rive at a single value for comparison. This num- ber, the Species Commonality Index (SCI), is arrived at by dividing the number of species common to both habitats by the total number of species present, as shown in the following example. HABITAT I HABITAT II Species Present Species Present A A B B C C L L M N O P R Q S u T V w Species in Common = 4 Total Number of Species =13 Species Commonality Index = 4 = .27 15 In order to make even semivalid compari- sons using this technique, the collections should be thorough and reasonably concurrent. The writer has found that a large group of interested students (for example a group of young biol- ogy majors), make very comprehensive collec- tions. The following data were compiled by 30 students in an elementary ecology class at West- ern Washington College of Education during the months of March and April, I960, and show to what purpose the SCI might be used. From inspection of the data shown in Table 1, it is apparent that Whatcom Rapids and Bad Pond are least similar. One would expect a rapids area to bear little faunistic similarity to a pond habitat, and thus it is seen that when- ever Whatcom Rapids is compared to a pond, the SCI is relatively quite low, except when compared with Mud Lake, and here is seen the third highest index. From this, it might be con- cluded that, ecologically, Whatcom Rapids is more similar to Mud Lake than to any other habitat shown, and thus, if no other informa- tion were available, it could be assumed that a transplant from the Rapids to Mud Lake might stand a fair chance of surviving. Further inspection of the data shows that Lake Fragrance and Good Pond are ecologically most similar. This is surprising in view of the fact that the two habitats are separated by more than 50 mi. and lie at elevations differing by more than 1,000 ft. Good Pond and Bad Pond, however, lie less than 100 ft. apart, but, accord- ing to the SCI, they seem to be relatively quite different. The writer had assumed that the two bodies were connected until subsequent investi- gation spurred by the low SCI revealed com- plete separation. Lacking other information, the writer would stake more on a transplant being successful between Lake Fragrance and Good Pond than between Good Pond and Bad Pond — even though the latter two are adjacent. This opinion is supported somewhat by knowledge 307 308 PACIFIC SCIENCE, Vol. XV, April 1961 TABLE 1 Data Comparing Various Fresh Water Habitats in the Bellingham, Washington, Area Using the Species Commonality Index (SCI) COMBINATION TOTAL NO. SPECIES NO. SPECIES IN COMMON SCI Whatcom Rapids Lake Fragrance 28 5 17 Good Pond Bad Pond 24 8 33 Bad Pond Peat Bog 18 9 50 Good Pond Peat Bog 19 6 32 Lake Fragrance Good Pond 22 12 55 Lake Fragrance Bad Pond 24 8 33 Lake Fragrance Peat Bog 23 ' 7 30 Whatcom Rapids Mud Lake 28 11 39 Whatcom Pools Good Pond 23 8 35 Whatcom Pools Bad Pond 21 8 38 Whatcom Pools Peat Bog 21 8 38 Whatcom Rapids Good Pond 26 5 19 Whatcom Rapids Bad Pond 24 4 16 Whatcom Rapids Peat Bog 22 4 18 Fragrance Lake Mud Lake 24 10 42 Mud Lake Good Pond 24 9 37 Mud Lake Bad Pond 25 7 ' 28 Mud Lake Peat Bog 24 7 29 Whatcom Pools Whatcom Rapids 24 9 . 38 Whatcom Pools Lake Fragrance 28 6 21 Whatcom Pools Mud Lake 27 6 22 common to fishermen in the locality, that trout stocking is much more successful in Good Pond than in Bad Pond — thus the names for the two ponds. It should be emphasized that the method is useful only if: 1. Thorough and complete collections are made. It is obvious that the greater the number of species in common, the greater the similarity of niches, and the more similar the physical attributes to account for the species similarity. Thus, the more thorough the collection, the more valid the results. 2. The collections are made as concurrently as possible. This prevents seasonal variation from giving too low an index. If an index is desired which covers a longer period, then of course concurrent collections should be made over sev- eral seasons. 3. Statements such as, "Mud Lake and Bad Pond are twice as dissimilar as are Lake Fra- grance and Good Pond,” are unwarranted. It is certainly possible to have a zero SCI but for the habitats to have many similarities. 4. The method is applied within limited areas where biotic communication between habitats is constantly possible and likely. For example, it would be unrealistic to compare a pond in New Zealand with one in Washington by this method. Even though the physical attributes of the habitats might be very similar, there is no chance of obtaining a high index. It is the writer’s intention in the future to make more thorough collections of these same areas and of others, and to attempt relating the various indexes to the physical characteristics of the habitats . — Charles J. Flora , Department of Biology, Western Washington College of Educa- tion, Bellingham, Washington. should not be fastened together in any \ray, and should be mailed flat. Inserts should be either typed on separate sheets or pasted on proper page, and point of insertion should be clearly indicated. Original copy and one carbon copy of manuscript should be submitted. The author should retain a car- bon copy. Although due care will be taken, the editors cannot be responsible for loss of manuscripts. Introduction and Summary. It is desirable to state the purpose and scope of the paper in an introductory paragraph and to give a summary of results at the end of the paper. Dictionary Style. It is recommended that authors fol- low capitalization, spelling, compounding, abbrevia- tions, etc., given in Webster’s New International Dic- tionary (unabridged), second edition; or, if desired, the Oxford Dictionary. Abbreviations of titles of pub- lications should, if possible, follow those given in World List of Scientific Periodicals. Examples of the usual abbreviations are to be found in Pacific Science. Footnotes. Footnotes should be used sparingly and never for citing references (see later). When used, footnotes should be consecutively numbered by supe- rior figures throughout the body of the paper. Foot- notes should be typed in the body of the manuscript on a line immediately below the citation, and sepa- rated from the text by lines running across the page. Citations of Printed Sources. All references cited should be listed alphabetically by author at the end of the paper, typed double-spaced. References to books and to papers in periodicals should conform to the following models : Batzo, Roderick L., and J. K. Ripkin. 1849. A Treatise on Pacific Gastropods. Rice and Shipley, Boston, vii + 326 pp., 8 figs., 1 map. Crawford, David L. 1920<*. New or interesting Psyllidae of the Pacific Coast (Homop.). Proc. Hawaii. Ent. Soc. 4(1): 12-14. 1920A Cerotrioza (Psyllidae, Homoptera). Proc. Hawaii. Ent. Soc. 4(2) : 374-375. Rock, Joseph F. 1916. The sandalwoods of Hawaii; a revison of the Hawaiian species of the genus Santalum. J. Aust. Inst. Agric. Sci. 12(3): 1—43, 13 pis. In the text, sources should be referred to by author, date, and page, as follows: "It was noted (Rock, 1916: 18) that . . .” or "Rock (1916: 21-24) says . . Quotations. Quoted matter of fewer than five printed lines (about 200 characters) should be given in the text in the usual form, using double quote marks. Longer quotations should be set flush with left mar- gin. The author is responsible for the accuracy of quoted material. Numbers. Decimals, measurements, money, percent- ages, time; enumerations in which any figure is 10 or over; and isolated enumerations of 10 and over should be given in Arabic figures, rather than spelled out, except when the number begins a sentence. ILLUSTRATIVE MATTER Only the minimum number of illustrations required to supplement the text will be accepted by the editors. Reproduction costs of illustrations in excess of the number allowed by the editors will be paid by the author. Illustrations should accompany manuscript, on sep- arate sheets. Often more valuable than a photograph is a good line drawing, which may be reproduced by the zinc etching process. Figures and Graphs. Copy for figures and graphs should always be drawn large enough to allow for at least one-third reduction by the engraver. Copy should consist of carefully prepared line drawings in one color only, drawn in India ink on plain white draw- ing paper or tracing cloth. Co-ordinate paper with lines in light blue (a color which will not show in a photograph) may be used; but co-ordinates which should appear in the finished graph must be drawn in India ink. If original figures may not be conven- iently submitted with manuscript, duplicate rough sketches or photographic prints may be furnished to aid the editors in their decisions. It is strongly urged that an indication of scale be incorporated as a part of all drawings in which mag- nification and size are critical considerations. Photographs. Photographs should be chosen for clarity in portraying essential information. They should be printed for contrast, on glossy paper, and should be sent unmounted. They should be identified with serial number written in soft pencil on the back to corre- spond with list of captions. Illustrations will be returned to the author. Tables. Tabular matter should be kept to a minimum. Each table, prepared to conform with Pacific Science style, should be typed on a separate page, and its posi- tion indicated on the manuscript. Captions. Readily identifiable captions for figures, graphs, photographs, and other illustrative matter should be supplied on a separate page. PROOF Proof should be corrected immediately and returned at once to Robert Sparks, assistant to the editors. Authors are reminded that the editors will allow only a minimum number of corrections on galley proof. Additions to the printed text and changes in style and content are not permitted. All queries on proof should be answered. If cor- rected proof is not received within four weeks after being sent to the author, author’s changes cannot be accepted. REPRINTS Reprints or separates should be ordered on the form provided and returned with author’s proof. All correspondence concerning separates must be directed to the printer, Star-Bulletin Printing Company, Inc., 420 Ward Avenue, Honolulu 14, Hawaii. JULY 1961 NO. 3 pin ' VOL. XV PACIFIC SCIENCE A QUARTERLY DEVOTED TO THE BIOLOGICAL AND PHYSICAL SCIENCES OF THE PACIFIC REGION | MARTIN W. JOHNSON Zooplankton of Arctic Coastal Lagoons HAROLD ST. JOHN Revision of the Genus Pandanus Part 3. A New Species from the Tuamotu Archipelago Part 4. Disposition of Some Later Homonyms Part 5. Pandanus of the Maidive and the Seychelles Islands V. J. CHAPMAN Ecology of the Kermadec Islands C. K. WENTWORTH , H. A. POWERS , and J. P. EATON Feasibility of a Lava-diverting Barrier at Hilo , Hawaii GORDON A. MACDONALD and TAKASHI KATSURA Lava of the 1 959 Eruption in Kilauea Iki E. YALE DAWSON Literature of Benthic Algae from the Eastern Pacific MARION GREY Fishes Killed by the 1 950 Eruption of Mauna Loa, Part V YOSHINORI KANEHIRO and LYNN D. WHITTIG Amorphous Mineral Colloids in Pacific Soils FRANKLIN G. ALVERSON Daylight Surface Occurrence of Myctophid Fishes I AUG-’ 7 U \\ UNIVERSITY OF HAWAII PRESS BOARD OF EDITORS O. A. Bushnell, Editor-in-Chief Department of Microbiology, University of Hawaii Robert Sparks, Assistant to the Editors Office of Publications and Information, University of Hawaii Agatin T. Abbott Colin S. Ravage Department of Geology and Geophysics Department of Geology and Geophysics University of Hawaii University of Hawaii Thomas S. Austin Bureau of Commercial Fisheries, Hawaii Area (U. S. Fish and Wildlife Service) Honolulu, Hawaii Albert J. Bernatowicz Department of Botany University of Hawaii L. H. Briggs Department of Chemistry University of Auckland Auckland, New Zealand Ai Kim Kiang Department of Chemistry University of Malaya, Singapore Paul J. Scheuer Department of Chemistry University of Hawaii Donald W. Strasburg Bureau of Commercial Fisheries, Hawaii Area (U. S. Fish and Wildlife Service) Honolulu, Hawaii Albert L. Tester Department of Zoology and Entomology University of Hawaii Miklos F. Udvardy Department of Zoology University of British Columbia Vancouver, Canada Thomas Nickerson, Managing Editor Office of Publications and Information, University of Hawaii SUGGESTIONS TO AUTHORS Contributions to Pacific biological and physical science will be welcomed from authors in all parts of the world. (The fields of anthropology, agriculture, engineering, and medicine are not included.) Manu- scripts may be addressed to the Editor-in-Chief, PACIFIC SCIENCE, University of Hawaii, Honolulu 14, Hawaii, or to individual members of the Board of Editors. Use of air mail is recommended for all communications. Manuscripts will be acknowledged when received and will be read promptly by members of the Board of Editors or other competent critics. Authors will be notified as soon as possible of the decision reached. Manuscripts of any length may be submitted, but it is suggested that authors inquire concerning possi- bilities of publication of papers of over 30 printed pages before sending their manuscripts. Authors should not overlook the need for good brief papers, presenting results of studies, notes and queries, com- munications to the editor, or other commentary. PREPARATION OF MANUSCRIPT It is requested that authors follow the style of Pacific Science described herein and exemplified in the journal. Manuscript Form. Manuscripts should be typed on one side of standard-size, white bond paper and double-spaced throughout. Pages should be consecu- tively numbered in upper right-hand corner. Sheets (Continued on inside back cover) PACIFIC SCIENCE A QUARTERLY DEVOTED TO THE BIOLOGICAL AND PHYSICAL SCIENCES OF THE PACIFIC REGION VOL. XV JULY 1961 NO. 3 Previous issue published April 17, 1961 CONTENTS PAGE On Zooplankton of Some Arctic Coastal Lagoons of Northwestern Alaska, with Description of a New Species of Eurytemora. Martin W . Johnson.... 311' Revision of the Genus Pandanus Stickman, Part 3. A New Species from Oeno Island, Tuamotu Archipelago. Harold St. John . 324 Revision of the Genus Pandanus Stickman, Part 4. Disposition of Some Later Homonyms. Harold St. John 327 Revision of the Genus Pandanus Stickman, Part 3. Pandanus of the Maidive Islands and the Seychelles Islands, Indian Ocean. Harold St. John 328 A Contribution to the Ecology of the Kermadec Islands. V. J. Chapman 347 Feasibility of a Lava-diverting Barrier at Hilo, Hawaii. C. K. Wentworth, H. A. Powers, and J. P. Eaton.... 332 Variations in the Lava of the 1939 Eruption in Kilauea Iki. Gordon A. Macdonald and Takas hi Katsura 338 A Guide to the Literature and Distributions of Pacific Benthic Algae from Alaska to the Galapagos Islands. E. Yale Dawson... 370 Fishes Killed by the 1930 Eruption of Mauna Loa, Part V . Gonostomatidae. Marion Grey... 462 Amorphous Mineral Colloids of Soils of the Pacific Region and Adjacent Areas. Y oshinori Kanehiro and Lynn D. Whittig. 477 note: Daylight Surface Occurrence of Myctophid Fishes Off the Coast of Central America. Franklin G. Alverson 483 NEWS NOTE ..; 484 Pacific Science is published quarterly by the University of Hawaii Press, in January, April, July, and October. Subscription price is $4.00 a year; single copy, $1.23. Check or money order payable to University of Hawaii should be sent to University of Hawaii Press, Honolulu 14, Hawaii, U. S, A. Printed by Star-Bulletin Printing Company, Inc., 420 Ward Avenue, Honolulu 14, Hawaii. > On Zooplankton of Some Arctic Coastal Lagoons of Northwestern Alaska, with Description of a New Species of Eurytemora Martin W. Johnson 1 A CHARACTERISTIC FEATURE of the low-lying Alaskan coast of the Chukchi Sea is the presence of a more or less broken beadlike series of "lagoons” extending from Cape Prince of Wales to Point Barrow. The general scientific interest in these lagoons is succinctly summarized by R. H. Fleming and staff in a preliminary report on the "Brown Bear” cruise no. 236 in 1959: The geological and oceanographic processes that have led to development and life history of these features are of major scientific interest. Because each of them may represent a variable but unique micro-environment, the biology of these lagoons is also of unusual interest because they represent a transitional series of marine to fresh- water environments. At one extreme these lagoons are, in effect, the complex estuaries of rivers that flow only during the summer. At the other extreme the older lagoons, now per- manently isolated from the sea and clogged with sediment and vegetation, are only distinguish- able from aerial photographs. Between these two extremes are bodies of water, varying greatly in size, that must from time to time be flooded with sea water and then are closed off again and slowly diluted by the accumulation of precipitation and runoff. During August 4-15, 1959, an opportunity was provided by the U. S. Atomic Energy Com- mission’s Committee for Environmental Studies of Project Chariot and the University of Wash- ington Department of Oceanography 2 to con- duct a survey of the plankton in a number of these more or less landlocked lagoons situated at the immediate coast in the region of Cape Thompson, Alaska. The samples collected, be- ing from various lagoons, are important in pro- viding a broad picture of the deviations or 1 Scripps Institution of Oceanography, University of California, La Jolla, and the University of Washing- ton, Department of Oceanography, Seattle. Contribu- tion from Scripps Institution of Oceanography, New Series. Manuscript received September 12, I960. 2 Contract AT-45-1-540 and Nonr-477 (10). similarities that characterize the populations of these bodies of water. The fauna encountered are of special interest in furthering our knowl- edge of the geographic distribution, biology, and taxonomic status of certain calanoid cope- pod species that are in varying degrees transi- tional between the purely marine and fresh- water forms. Prior to the initiation of this project, there had apparently been no study of the zooplank- ton of these remote and relatively inaccessible lagoons. Some studies that provide especially useful records for comparison have been made previously of the offshore plankton of the Alas- kan coast of the Chukchi and Beaufort seas (Willey, 1920; Johnson, 1953, 1956), and in ponds situated at Point Barrow (Comita and Edmondson, 1953; Comita, 1956; Johnson, 1958) . Coincident with the present restricted study, a general plankton survey was made in the offshore waters of portions of Chukchi and Bering seas by members of the University of Washington Oceanographic Department aboard the "Brown Bear” (R. H. Fleming and staff, 1959) . It is a pleasure to thank Dr. Richard H. Flem- ing for providing this opportunity to participate in the research expedition of the University of Washington oceanographic vessel "Brown Bear” to the Chukchi Sea area. The assistance of Dr. Norman J. Wilimovsky, Philip Buscemi, and Howard Smith in the field is gratefully acknowledged, as is also the cooperation of the administrators and contractors of the project site. The water samples were titrated for chlo- rinity by the Geological Survey laboratory at the project camp site. PROCEDURE AND RESULTS The plankton samples were collected with a 30 -cm. net, 1 m. long, constructed of no. 6 bolting cloth. An 311 inflated rubber boat was used Institution AUG i M 312 PACIFIC SCIENCE, Vol. XV, July 1961 Fig. 1 . Numbered lagoons sampled in the Cape Thompson, Alaska area. Based on U. S. Coast and Geodetic Survey Topographic Map T-9425 Alaska. while obtaining the deeper hauls and the water samples. Most surface hauls were made by wad- ing along shore with the net attached to the end of a pole. Lagoons The position, approximate size, and identi- fication of the lagoons sampled are shown in Figure 1. Cape Thompson divides them geo- graphically into a northern and a southern group, as numbered in Figure 1. The only lagoon for which a name occurs on U.S.C. and G.S. Topographic Map T-9'423, on which Figure 1 is based, is Tusikpok, which corresponds to no. 6 south. All of the lagoons are shallow; the depths at the positions of sampling (about one- half the distance across the lagoon) were from 1.3 to 3 m., and, as anticipated from wind con- ditions and shallowness, they showed only mod- erate or no thermal or haline stratification ex- cept in no. 2 south where the salinity at the bottom was 1.65 per mille higher than that at the surface (Table 1). It is clear that they are at times subject to invasion of salt water from the sea. This is evidenced by the brackish water of some, by high wave-washed channels on the gravel berms separating the lagoons from the sea, and, in several instances, by the presence of marine and brackish water animals found in the plankton of some lagoons. Lagoons no. 2 Arctic Zooplankton — Johnson 313 TABLE 1 Water Temperature and Salinity in Coastal Lagoons Immediately South and North of Cape Thompson, Alaska, Aug* 1959 (The lower sample was taken just above the bottom depth indicated) LAGOON TEMP. °c SALIN- ITY %o Lagoon no. 1 south, Aug. 12 surface. 11.0 0,83 bottom (1.5 m.). 11.0 0.83 Lagoon no. 2 south/Aug. 12 surface....... 11.2 14.31 bottom (1.3 m.) 11.1 15.96 Lagoon no. 3 south, Aug. 12 surface. 11.0 0.16 bottom (2.1 m.) 10.4 0.17 Lagoon no. 4 south, Aug. 13 surface........... 12.3 6.42 bottom (1.3 m.)... 12.1 7.16 Lagoon no. 5 south, Aug. 13 surface..... 13.6 0.83 bottom (1*3 m.) 13.6 0.83 Lagoon no. 6 south, Aug. 13 surface 12.6 0.73 bottom (2.4 m.). 12.4 0.73 Lagoon no. 7 south, Aug. 13 surface 12.6 3.58 bottom (2 m.) 12.1 3.58 Lagoon no. 1 north, Aug. 14 surface 13*5 0.18 bottom (3m.) 13.0 0.18 Lagoon no. 2 north, Aug. 1 5 surface 13.8 0.46 bottom (2.5 m.) 13.0 0.55 and 4 south each have a narrow above-sea-level outlet that probably floods with sea water dur- ing high storms and undoubtedly accounts for the higher salinities observed there* Judged by the composition of the contained zooplankton, the lagoons are strikingly dissim- ilar ecologically. How persistent this dissimilar- ity is cannot be said from this preliminary sur- vey. Basically, the differences probably arise from the geographic position of each lagoon with respect to (1) extent of influx of fresh water in relation to the amount and frequency of that which spills over from the sea during high winds, (2) the height of the lagoon with respect to sea level, and (3) the effectiveness of the berm in serving as a barrier (reinforced by permafrost?) to water percolating out from the lagoon through the gravel* However, the envi- ronmental conditions characteristic of each la- goon have been sufficiently unique in the imme- diate past to permit selection and development of one or two species to a position of over- whelming dominance numerically, as shown by the analysis of the plankton fauna. Zooplankton The percentage composition of the fauna is shown in Table 2. Specific attention was given mainly to the calanoid copepods, because of the interest that some of the species hold as tran- sitory forms or as permanent residents in the overlapping environmental conditions between the sea and fresh water* In all cases the popula- tions were composed dominantly of fresh-water or brackish-water forms. But in lagoon no. 2 south, there was a conspicuous element of marine plankton, which was probably only re- cently recruited from the sea (Table 2). Al- though Acartia bifilosa, a brackish-water species, was overwhelmingly dominant in this lagoon, several marine copepods were also present, but no larval stages of any species were found that might indicate local production. However, if only one generation is produced each year in the lagoon, as is common in some Arctic lakes, the season of nauplii-hatching was probably about past. Most of the marine species found are considered typical of strongly neritic waters, but Calanus finmarchicus , Pseudo calanus minu- tus, and Acartia longiremis do also occur well offshore and are very common in the Chukchi Sea, especially off the eastern coast and well north into the Arctic. Eurytemora pacifica is one of the more in- frequent neritic species encountered. Rather little is known regarding its preferred habitat. Hitherto, on the Alaskan coast it has been taken only at Grantley Harbor, and in the region of Point Hope, where it was reported by Willey (1920) as E. johanseni n. sp. Excepting lagoon no. 2 south, it was not found in any of the pres- ent lagoons, but it was common in a tow taken immediately offshore from lagoon no. 1 north. It is also common in Kivalina Lagoon, where it constituted 7 per cent of the calanoids in a sam- ple kindly provided by Mr. Robert W. Owen of Percentage Composition by Number of Plankton in Coastal Lagoons Immediately South and North of Cape Thompson, Alaska (c = common; + = present, but neither in sufficiently large numbers to constitute 1 per cent of population) 314 PACIFIC SCIENCE, Vol. XV, July 1961 LAGOONS NORTH CN i/"C PQ + 28 \T\ 57 + \r\ CO - r^- \r\ - 40 - » rs PQ - 86 U u xr CO u 99 u o LAGOONS SOUTH r-' rcc PQ + rO> CO o + u NT 80 CO m PQ u 06 + 85 GC CO co CN m ire GC + + ire rCC PQ u + 38 + 37 25 CO 1 CO u + u + CO Gc + u "St* cfC PQ + 72 CN + + CO CO 06 + + i r\ + - + + rO> CN PQ + pH - + 97 + CO I CO u + 66 + CN CN PQ o ON i i 00 + u + + U - + + m CO CN CO CO PQ CO 00 CN l CO + u U + - u + + + + 1 CN 00 CO - + \T\ \o rCC 28 CN CN u Date of sampling Surface or bottom Arcartia bijilosa | A. claus i A. longiremis Calanus finmarchicus Centropages abdominalis Cyclops spp Eurytemora canadensis | E. herdmani 1 E. pacifica 1 E. foveola, n. sp 1 Limnocalanus grimaldi L. johanseni Pseudocalanus minutus T ortanus discaudatus Harpacticoids Daphnia Podon Evadne Clam shrimp Fairy shrimp (Anostraca) Ostracods Neomysis (juvenile) Sagitta Fish larvae (total found) Rotifer Arctic Zooplankton — Johnson 315 the University of Washington from collections made near the outlet on Aug. 22, 1959. These are the only known locality records for the American coast. Elsewhere the species occurs on the Asian coast. The other neritic species encountered are quite common in small numbers on the Alaskan coast to Point Barrow and in diminishing num- bers eastward (Johnson, 1953, 1956). In lagoon no. 4 south, the plankton was sparse and Acartia bifilosa was the most nu- merous copepod in the surface water on Aug. 6, but an unidentified harpacticoid was the dom- inant form at the bottom on Aug. 13. It is probably a benthic species since the net con- tained much bottom debris. Intermingled in the plankton on both dates were the neritic cladoc- erans, Podon and Evadne , as the chief evidence of marine invasion. The plankton fauna of all of the remaining lagoons in the numbered series was characteris- tically fresh-water. However, there was still a strange marine affinity evidenced by the pres- ence of Limnocalanus grimaldi and L. johanseni. The former species was common in lagoon no. 1 north, and the latter in all other lagoons with the exception of no. 2 south. The species of this copepod genus have long been the subject of much speculation with respect to their geo- graphic distribution and affinities to the sea. One species, L. macrurus , commonly occurs in deep fresh-water lakes, and is generally believed to be a marine relict of glacial times. L. grimaldi is an Arctic marine and brackish-water form occurring along the Arctic coast of the U.S.S.R. and is also considered to be a relict of Arctic fauna when found in such widely different iso- lated localities as the Caspian Sea, the Gulf of Bothnia, and the Baltic Sea. Aside from its present occurrence in lagoon no. 1 north, it has been reported from the Alaskan coast on two previous occasions, once from a collection taken about 100 yd. from the shore at Collinson Point (Willey, 1920), and once at a series of nine offshore stations in the same region, and off the mouth of the Colville River (Johnson, 1956). The present discovery is therefore of special interest in extending the known range of the species on the Alaskan coast, and in recording its occurrence in virtually fresh water, together with Eurytemora foveola n. sp. that apparently thrives best at very low salinities. Less is known about the distribution of Lim- nocalanus johanseni. It was originally described by Marsh (1920) from a fresh-water pond at Collinson Point, just inland from, but not con- nected with, the shore where Willey recorded L. grimaldi. Subsequently, Comita and Edmond- son (1953) reported it from Imikpuk Lake, a fresh-water lake near Point Barrow. In the pres- ent survey it was a conspicuous element ranging from small to dominant numbers in all lagoons except no. 2 south, the one which most nearly approaches marine conditions. In lagoon no. 5 south, it was extremely abundant, and although a considerable number of early copepodid stages occurred mingled with adults no nauplii were observed. This agrees in general with the ob- servations of Comita ( 1956) in a more ex- tensive analysis which indicated that the eggs of this species in Imikpuk Lake hatched early in spring, and only copepodid stages are to be found in late July and August. His data show no nauplii after Jul. 31 and the first appearance of adults (copepodid stage VI) was on Aug. 10. In the present collections from lagoon no. 5, adults already outnumbered the other stages on Aug. 6, when the first sampling was done. Comita concluded that L. johanseni in Imikpuk Lake produces only one generation a year and that the species winters over in the egg stage. This agrees with the known life cycle of an- other fresh-water species, L. macrurus , discussed by Gurney (1931). Another similarity between the populations of Imikpuk Lake and lagoon no. 5 south was the presence of at least two size groups. This bimodality was not so pro- nounced in the males as in the females, but some tendency is shown (Fig. 2). The present analysis of the population applies only to the standing crop at the time of sampling, and hence further comparisons cannot be made other than to note that the large-size group was the most abundant. But it cannot be said whether they were produced before or after the small- size group. TAXONOMIC NOTES The following observations can be made re- garding a few of the species encountered. 316 PACIFIC SCIENCE, Vol. XV, July 1961 Fig. 2. Length frequency histogram for Limnocalanus johanseni adult (copepodid VI) population. Limnocalanus grimaldi (de Guerne) A total of 25 specimens was taken in lagoon no. 1 north, mostly near the bottom at a depth of only 3 m. Specimens were compared directly with L. grimaldi taken in 1950 in the Beaufort Sea at various stations offshore near Collinson Point and the Colville River. The fresh-water species L. macrurus Sars has not been seen for comparison. The Beaufort Sea specimens which were re-examined at this time agree well with the descriptions given in the literature for L. grimaldi . The head is not in the least vaulted as occurs in various degrees in L. macrurus , and there is only a slight cervical depression (Sars, 1897: pi. 4, figs. 2, 3, 17). The structure of the fifth feet also agrees. The first antennae of the female reach beyond the anterior margin of the anal segment. In the male there are five distinct segments in the distal end of the right antenna beyond the geniculation. The deviations from descriptions of grimaldi are mainly in the posterior margins of the fifth thoracic segment. These were smoothly rounded in some cases, but many specimens had the characteristic tri- angular or spine-like point, and this was some- times present only on one side. The caudal rami were about eight times as long as broad and were slightly longer in relation to the urosome than given by Sars. The specimens collected in lagoon no. 1 north are here also considered to be L. grimaldi , since they agree in nearly all respects with the Beaufort Sea material. The posterior margin of the fifth thoracic segment was acutely pointed in most specimens, but a few occurred with smoothly rounded margins (a characteristic of L. macrurus ) ; there were, however, only four distinct segments in the distal end of the male right antenna beyond the geniculation. This is also a characteristic of L. macrurus , although Marsh (1933) states that in exceptional cases there may be five segments. The lagoon speci- mens were slightly smaller than those from the Beaufort Sea, namely 2.6 to 2.7 mm. as opposed to 2.9 to 3.3 mm. for females. Since the condition of the fifth thoracic seg- ment is obviously an unreliable character, one is left to choose between the relative specific value of the outline of the head, and the seg- mentation of the distal end of the male first antennae. Both of these are variable for L. macrurus . The habitat presented by lagoon no. 1 north is probably an important consideration, for although it was virtually fresh at the time and place of sampling, it is shallow and so near the sea that it may be reached by very high storm waves. This environmental situation may constitute an interesting natural experiment in which a landlocked population of L. grimaldi Arctic Zooplankton — Johnson 317 has assumed some characteristics of its fresh- water counterpart, particularly in the fusion of two of the segments in the distal portion of the male right antenna. Limnocalanus johanseni Marsh A comparison of specimens of this species from the lagoons and from collections I made in Imikpuk Lake in 1957 shows only small variations in structure. Mention should be made, however, of some features that are not included in the original description. It was noted that in many specimens the fifth thoracic segment may be smoothly rounded or with only slightly angular outline. Marsh states that this segment is rounded on the sides, and each side is armed with a small spine which may be either sharp or blunt. Apparently inadvertently omitted from Marsh’s drawings is a long, heavy seta that oc- curs on the inner anterior distal angle of the second basis of the first feet. Centropages abdominalis Sato 1913 Centropages mcmurrichi Willey 1920 Willey (1920) was apparently unaware of Sato’s (1913) publication, and described this species as new under the name C. mcmurrichi n. sp. His description is without figures, but he considered the species to be identical with a copepod reported and figured by McMurrich (1916), with some reservations, as C. hamatus Lilljeborg. Examination of Alaskan material and reference to McMurrich’s figures, and to those of Sato ( photostatic copies of which are at hand through the courtesy of Dr. Takasi Tokioka and Isamu Yamazi of the Seto Marine Biological Laboratory), and of Mori (1937) leaves little doubt that these species are identical. Eurytemora pacifica Sato 1913 Eurytemora johanseni Willey 1920 This is a clearly defined species, principally on the basis of the fifth feet in the female (Figs. 7-10). In the female, the fifth thoracic segment is provided with broad triangular wings, but breeding females were also found with this segment smoothly rounded (Fig. 8). The hya- line wing apparently may be shed or fails to develop. In the present study both males and females were found in the same catches. Willey (1920) described the species under the name E. johanseni n. sp. Sato described only the adult male, but his figures agree especially with Willey’s figures 11 and 12 of the fifth feet. Smirnov (1931) first pointed out the probable identity of the species. Eurytemora foveola n. sp. TYPES: Holotype, male, U. S. National Mu- seum no. 105996; allotype, female, USNM no. 105997. Type locality, lagoon no. 6 south. In lagoon no. 6 south and nos. 1 and 2 north the dominant copepod was this hitherto un- known species. It was present in these lagoons in vast numbers and despite its small size con- stituted 90 per cent or more of the biomass in the samples from lagoon no. 6 south and no. 1 north where it was actively reproducing and many females were found carrying eggs and spermatophores. In lagoon no. 2 north it was greatly surpassed in mass by the larger clado- ceran Daphnia. It is closely allied to Eurytemora gracilis, both sexes of which were described by Sars (1898) from the lower Yana (Jana) River and, judging from Sars’ figures and description, the females may be nearly indistinguishable. The chief differences separating the present species from that of Sars’ are structural details in the fifth feet of the males. In view of these small but apparently persistent differences, as shown by examination of a great many speci- mens, it seems best to designate it a new species rather than a variety of E. gracilis. FEMALE (Figs. 3,4): Length 1.15-1.25 mm. The body is slender in both dorsal and lateral aspects with about the following linear propor- tions: metasome (anterior division) 67, genital segment 11, first abdominal segment 6, anal segment 11, caudal rami 17. The metasome is slightly widest in the region of the first pe- digerous segment. There is a slight cephalic de- pression, but no medial knob on the postero- dorsal margin of the cephalic segment. (In E. gracilis the greatest width appears to extend somewhat farther forward according to Sars’ pi. 8, fig. 8; and in his fig. 9, the lateral profile also differs in that his species is relatively thicker through the midbody and the cephalic 318 PACIFIC SCIENCE, Vol. XV, July 1961 FIGS. 3-6. Eurytemora foveola n. sp. 3, Female, dorsal; 4, female, lateral; 5, male, dorsal; 6, male, lateral. FIGS. 7-8. Eurytemora pacijica. 7, Female, lateral with eggs and spermatophore; 8, female, lateral outline of fifth thoracic segment occurring in some adults. Arctic Zooplankton — JOHNSON segment is slightly raised dorsally at the hind edge.) The fifth thoracic segment is extended into laterally projected "wings” that vary con- siderably in length and hyalinity. Many adult specimens were observed in which the left wing was only slightly developed, although the right wing was always well developed. The first an- tennae reach to, or slightly beyond, the fifth thoracic segment. The genital segment is nearly symmetrical and only slightly constricted lat- erally. It bears no lateral expansions but in lat- eral aspect has a rather conspicuous genital operculum. The anal segment is armed dorso- laterally with patches of fine spinules, best ob- served in aqueous media. (Sars states that in E. gracilis "the last caudal segment is perfectly smooth without any trace of the densely crowded spikes clothing the dorsal face of this segment in T. (E.) hirundoides .”) The caudal rami are provided with fine "hairs” on both their inner and outer margins. The longest caudal setae are only slightly longer than the rami. The terminal spine of the exopod of the second, third, and fourth swimming feet (Fig. 11) is cultriform as in E. gracilis. The fifth feet (Fig. 13) are symmetrical as in E. gracilis , and the first exopod segment bears a heavy, slightly setose, ungui- form process directed inward. There are two spines on the outer margin of this segment. The end segment is short and bears one long ter- minal spine and one short outer spine. The terminal spine of the exopod of the third and fourth feet is somewhat broadened in the mid- dle and is shorter than the end segment. In the male this spine is longer and more slender. The genital operculum is rounded (Fig. 16) in con- trast to the pointed operculum of the closely related species E. affinis as described and figured by Gurney (1931), and which he considers synonymous with E. hirundoides. MALE (Figs. 5-6): Length 1 .0-1.2 mm. The body is slender and nearly parallel-sided ante- riorly but tapers posteriorly. The body propor- tions are about as follows: metasome 57, uro- some (without caudal rami) 25, caudal rami 15. As in the female, the anal segment bears dorsal patches of tiny spinules. The caudal rami have fine "hairs” on the inner and outer margins. The right geniculated first antenna has 25 seg- ments with moderately heavy spines on seg- 319 ments 8, 9, and 12. The fifth feet (Figs. 14, 15) are asymmetrical, uniramous, and each exopod consists of two segments. The specific characters separating the species from closely related forms are seen mainly in the basal seg- ments and in the relative proportions of the segments of the rami. At the inner proximal margin of the second basipod of the right foot there is a well-defined rounded projection and associated smooth notchlike depression suggest- ing the specific name. The inner margin of the first basipod of the left foot projects down- ward, forming an irregular blunt process over- lapping about one-third the length of the sec- ond basipod. The second basipod is broadened with an inward expanse forming a dull flange along the long axis of the segment. Sars’ plate 8, figure 16, shows no trace of these three fea- tures for E. gracilis. Otherwise the exopod seg- ments agree with that species which, however, appears to have a shorter first exopod segment on the left foot. The uncertain species described by Kiefer (193d) also differs in the absence of these characters and in the shape of the second basal segments of each foot. The second an- tennae and the mouth appendages of E. foveola are similar in the two sexes and appear to have no specific character except perhaps for the shape of the mandibular blade (Fig. 17). Few illustrations of other species are available with which to make comparisons, but the gap be- tween the first and second tooth is considerably wider than that figured by Gurney (1931) f° r E. affinis, Sars ( 1903 ) for E. velox, and Wilson (1953) for E. yukonensis and E. composita. It should be mentioned that a single damaged female specimen of a Eurytemora was collected by the Canadian Arctic Expedition offshore near Cape Thompson. Willey (1920) tentatively re- ferred it to E. gracilis. It is not possible to con- clude what species he dealt with but the prox- imity of the catch to the lagoons so richly pop- ulated by E. foveola makes it highly likely that it had been washed out from a nearby lagoon. Although there may be an intermittent present- day connection between the Yana River popula- tion and that in the Cape Thompson area, it must be a very tenuous one for a fresh- or brack- ish-water species, especially in view of the pre- vailing oceanic circulation northward through 320 PACIFIC SCIENCE, Vol. XV, July 1961 FIGS. 9—10. Eurytemora pacifica. 9, Female, fifth feet; 10, male, fifth feet. FIGS. 11-15. Eurytemora foveola n. sp. 11, Female, fourth foot; 12, female, first foot; 13, female, fifth feet; 14, male, fifth feet; 15, male, left fifth foot. Arctic Zooplankton- — JOHNSON 321 the Bering Strait and eastern Chukchi Sea. How- ever, other studies indicate that plankton or- ganisms may be transported across rather well- defined currents by eddy diffusion, and the pre- vailing currents along the Siberian coast might bring expatriates into the western portion of the Chukchi Sea. Brodskii (1950 ) does not in- clude E. gracilis in his list of copepods of the Polar Basin. Gurney (1931) has pointed out the specific variabilities that may occur in several species of Eurytemora living in different habitats in Eu- rope. It remains to be shown to what extent E . foveola may also be morphologically variable in diverse environments. The great range of Alaskan lagoon and estuarian habitats and the seven or more species on the Alaskan coast should provide excellent material for such a study. SALINITY RELATIONS Although the data are too few to give more than a glimpse into the salinity tolerances ob- served for the various copepod species identi- fied, still it is worthwhile to record such analyses as can be made from this area. Figure 19 ap- pears to divide the copepods encountered into three groups with respect to salinity range and preference. ( 1 ) Acartia bifilosa and A. clausi occurred over nearly the whole range encoun- tered but not below 0.83 %o at which value only relatively few specimens were found, especially A. clausi. The former was the dominant copepod in the two lagoons having highest salinities 6.42 to 15.96 /co. (2) Cyclops spp., Eurytemora can- adensis , and Limnocalanus j o bans eni ranged from 0.1 6 to about 7 %c but appeared to thrive best at 0.73 to 0.83 %o. Eurytemora foveola n. sp. probably belongs to this group but stands somewhat alone in occurring only in ranges from 0.18 to 0.73 %o at which salinities it was dominant. Limnocalanus grimaldi, in this unique situation, occurred also at only a very low salinity, but elsewhere in the arctic it has a strong affinity for the sea. ( 3 ) The final group, into which Acartia bifilosa and A . clausi ex- tended, are the characteristically marine forms occurring in lagoon no. 2 south with salinities of 14.31 to 15.96 /co. Only in the more saline lagoons, nos. 2 and 4 south, was there any ap- preciable range in salinity from top to bottom. These two lagoons are also probably the ones most readily invaded by marine forms, but the influx of fresh water appears also to be quite high in these lagoons. Hence the communities are probably more or less transitory and fluc- tuate with repeated recolonizations. Whittaker and Fairbanks (1958) have studied the occurrence of various nonmarine copepods in inland lakes and ponds of different salt con- tent. The salinities and fauna they dealt with are of course not directly comparable with those of coastal lagoons, since the ratio of salts differ and their more saline species have different taxonomic relationships. But it is interesting to note that the transition between saline and fresh-water communities in their study was "somewhere between 425 and 875 ppm” for most bodies of water. If a comparison is justi- fied it must be with the fresh-water fauna of the lagoons, then lagoons nos. 1, 3, 5, and 6 south and 1 and 2 north could be expected to foster truly fresh-water species and the occur- rence of Eurytemora canadensis and Limnocal- FlGS. 16—18. Eurytemora foveola n. sp. 16, Female, genital field and operculum, ventral; 17, male, mandi- bular blade; 18, male, labrum. 322 PACIFIC SCIENCE, Vol. XV, July 1961 Fig. 19. Total salinity ranges over which certain copepod species occurred in nine coastal lagoons near Cape Thompson, Alaska, 1959- Cross bars denote salinity at stations of sampling. anus johanseni in most of these and in lagoons nos. 4 and 7 south would mark them as quite euryhaline but with preferences for the lower salinities. REFERENCES Brodskii, K. A. 1950. Copepoda, Calanoida, of the far-eastern waters of the USSR and the Polar Basin. Zool. Inst. Acad. Sci. USSR Mosc. : 441 pp. (In Russian.) COMITA, G. W. 1956. A study of a calanoid copepod population in an arctic lake. Ecology 37(3): 576-591. CoMiTA, G. W., and W. T. Edmondson. 1953. Some aspects of the limnology of an arctic lake. Stanf. Univ. Publ. Biol. Sci. 11: 7-13. Fleming, R. H., and Staff. 1959. Oceano- graphic survey of the eastern Chukchi Sea, 1 August to 2 September 1959. Progress re- port as of 1 December 1959, brown bear Cruise No. 236. Dept. Oceanogr. Univ. Wash. Ref. 59^47: 17 pp. Gurney, R. 1931. British fresh-water Cope- poda. Ray Soc. Publ. 1(118): 1-238. Johnson, M. W. 1953. Studies on plankton of the Bering and Chukchi seas and adjacent areas. Proc. Seventh Sci. Congr. 4: 480-500. 1956. The plankton of the Beaufort and Chukchi Sea areas of the arctic and its rela- tion to the hydrography. Arct. Inst. N. Amer. Tech. Pap. No. 1: 32 pp. 1958. Observations on inshore plankton collected during summer 1957 at Point Bar- row, Alaska. J. Mar. Res.: 17: 272-281. Arctic Zooplankton — -Johnson 323 Kiefer, F. 1938. Freilebende Siisswassercope- poden von den Nordkurilen. Bull. Biogeogr. Soc. Japan 7(4): 75-94. MARSH, C. D. 1920. Report of the Canadian Arctic Expedition, 1913-18. Southern Party, 1913-16. Vol. 7: Crustacea, pt. J: Freshwater Copepoda. 25 pp. — 1933. Synopsis of the calanoid crusta- ceans, exclusive of the Diaptomidae, found in fresh and brackish waters, chiefly of North America. Proc. U.S. Nat. Mus. 82, art. 1-8: 1-58. McMurrich, J. P. 1916. Notes on plankton of British Columbia Coast. Trans. Roy. Soc. Can. Ser. Ill, 10, Sec. IV: 75-89. Mori, T. 1937. The Pelagic Copepoda from the Neighbouring Waters of Japan. 150 pp. Sars, G. G. 1897. Pelagic Entomostraca of the Caspian Sea. Ann. Zool. Mus. Imp. Acad. Sci. St. Petersb.: 1-73. 1898. The Cladocera, Copepoda and Ostracoda of the Jana Expedition. Ann. Zool. Mus. Imp. Acad. Sci. St. Petersb. 3: 324-359. Sato, T. 1913. Pelagic copepods. Rep. Fish. Res. Hokkaido Fish. Exp. Sta. No. 1: 79 pp. (In Japanese.) Smirnov, S. 1931. Zur Kenntnis der Copepo- dengattung Eury femora Giesbr. Zool. Anz. 94(5/8): 194-201. Whittaker, R. H., and C. W. Fairbanks. 1958. A study of copepod communities in the Columbia Basin, Southeastern Washing- ton. Ecology 39(1): 46-65. WILLEY, A. A. 1920. Report of the Canadian Arctic Expedition, 1913-18. Vol. 7: Crustacea, pt. K: Marine Copepoda. 46 pp. Wilson, M. 1953. New Alaskan records of Eury femora ( Crustacea, Copepoda ) . Pacif. Sci. 7: 504-512. Revision of the Genus Pandanus Stickman, Part 3 A New Species from Oeno Island, Tuamotu Archipelago Harold St. John 1 Previous parts of this revision of Pandanus have appeared in Pacific Science. A single spe- cies is here published in order that its name may be available for use in another projected publication. PANDANACEAE (section Pandanus ) Pandanus feruliferus sp. nov. Fig. 11 diagnosis holotypi: Arbor 10 m. altus 30 cm. diam. corona rotundata, foliis longioribus 1 B. P. Bishop Museum, Honolulu 17, Hawaii, U.S.A. Manuscript received December 12, 1959- 92 X 5.4 cm. longe diminuentibus in basi paene latioribus subcoriaceis, nervo medialein parte tertia infera inerme in media cum denti- bus acicularibus luteis adpressis 1.5-2 mm. longis 14-38 mm. distantibus in parte tertia ultima dentibus nigris 1 mm. longis 1.5-4 mm. distantibus, marginibus in basi inermibus sed in parte quarta infero cum dentibus subulatis luteis adpressis 2-2.5 mm. longis 2-7 mm. dis- tantibus in parte media simulantibus sed nigris in parte quarta ultima ad eos marginis simu- lanties, syncarpio terminale, phalangibus 6-6.2 cm. longis 2.3-37 cm. lotis 2.1-3 cm. crassis oblanceo-ellipsoideis paene compressis epiderme FIG. 11. a, b, Phalange, lateral view, X 1; c, phalange, longitudinal median section, X 1; d, phalange, apical view, X 1; leaf base, lower side, X 1; f, leaf middle, lower side, X 1; g, leaf tip, lower side, X 1; h, marginal teeth of leaf near base, X 4. 324 Page 26: Revision of Pandanus , 3. Tuamotu— St. John 325 326 PACIFIC SCIENCE, Vol. XV, July 1961 in sicco brunneo lucido cum 5-8 angulis epro- minentibus lateribus laevibus lucidis subplanosis sinibus lateribus subcausis dimidia supera li- bera apice convexo et 7—24 mm. lato, apicibus liberis carpelorum anguste conicis eis centrali- bus erectis sed eis marginis extus curvatis sini- bus centralibus 3-5 mm. profundis, 40% pha- langiorum cum 1-3 carpelis lateralibus parvis abortivis et 40% minoribus, carpellis 5-11; stigmatibus apicalibus 2-3 mm. longis subor- bicularibus vel ovatis fissis centripetalibus in 45° obliquis, mesocarpio apicale fibroso et ca- vernoso, mesocarpio basale fibroso et carnoso, endocarpio mediale osseoso obscure mahogani- brunneo gracile 1-2 mm. crasso, seminibus 12- lb mm. longis 5-7 mm. latis oblique ellipsoideis. DIAGNOSIS OF holotype: Tree 10 m. tall, 30 cm. in diameter, round-topped; longer leaves 92 cm. long, 5.4 cm. wide, ligulate, long taper- ing to an apparently acute apex, at basal attach- ment only slightly broadened, subooriaceous, the midrib unarmed on the lower third, near the middle with ascending yellowish acicular ap- pressed teeth 1.5-2 mm. long, 14-38 mm. apart, on the upper third the teeth 1 mm. long, 1.5- 4 mm. apart, black; the margins at base un- armed, near the lower quarter the teeth 2-2.5 mm. long, 2-7 mm. apart, appressed subulate yellowish; those near the middle similar but black; those of the upper quarter like those on the adjacent margins; heads terminal, size and shape not recorded; phalanges 6-6.2 cm. long, 2. 3-3.7 cm. wide, 2.1-3 cm. thick, oblance- ellipsoid, a little compressed, the surface (when dried) brown, smooth, shining, with 5-8 slight angles, the sides shining, smooth, nearly plane, the sutures between the lateral carpels mostly invisible, only a few of them with distinct furrows about half as long as the phalange, upper half free and marked with numerous longitudinal light-colored cracks, the sides curv- ing upward like an ellipse, the apex convex and 7-24 mm. wide, the free carpel apices narrowly conic, the central ones erect, the marginal ones curved, diverging, the central sinuses 3-5 mm. deep; of the phalanges 2/5 show 1-3 small abortive lateral carpels about 2/3 as long as the fertile ones; carpels 5-11; stigmas apical, 2-3 mm. long, suborbicular to ovate, creased, cen- tripetal, oblique, at about 45° facing the cen- ter; apical mesocarp fibrous and cavernous; basal mesocarp fibrous and fleshy; endocarp median, bony, dark mahogany brown, forming a thin coating only 1-2 mm. thick; seed 12-16 mm. long, 5-7 mm. wide, oblique ellipsoid. HOLOTYPUS: Oeno Island, Tuamotu Archipel- ago, forest on coral sands, 3 m. alt., June 23, 1934, H. St. John & F. R. Posher g 15,197 (bish). In the Bishop Museum there is another col- lection from Oeno, 1922, E. H. Quayle 410. The leaf has larger, coarser, more widely spaced, mar- ginal teeth. It might be a leaf from a vigorous seedling of this species, but the collection lacks fruit, so it will be left undetermined. A similar plant is P. odoratissimus L. f. var. oahuensis Martelli from Hawaii, but this has phalanges 6.5 cm. long; and the carpel apices indexed and low, broad pyramidal. More similar is P. Mei F. Br. which has phal- anges 7 cm. long, 2.5 cm. wide, the six carpels with apices convex; stigmas 4 mm. in diameter, reniform, horizontal; endocarp walls 2.5-5 mm. thick. This occurs on Hivaoa, Marquesas Islands. The specific epithet is coined from the Latin, ferula, a splint; fero, to bear. It is formed in anal- ogy to the splints or vestigial digit bones in a horse’s leg, since the Pandanus phalanges often bear abortive marginal carpels, suggestive of the splint bones of a horse. Revision of the Genus Pandanus Stickman, Part 4 Disposition of Some Later Homonyms Harold St. John 1 Pandanus glauciferus St. John, nom. nov. P. glauciphyllus C. B. Robinson, Torrey Bot. Cl. Bull. 35: 64, 1908, non P. glaucophyllus Ridl., Asiat. Soc Str. Br. J. 41: 50, 1904. Under the 1956 International Code, Rec. 73G and Art. 73 Note 2, the epithet glauciphyllus, differing only in the connecting vowel, and i being an incorrect one, is an orthographic error. The epithet glauciphyllus must be treated as a later homonym and under Art. 64(2) must be rejected. Hence, a new name for it is provided. Pandanus compressus Martelli, Webbia 1: 363, 1905; 4(1): table 10, figs. 14-15, 1913. P, tetrodon Balf. /. ex Kanehira, Bot. Mag. Tokyo 52: 236-239, fig. 70, 1938, avow- edly by (Gaud.) Balf. /., based on Bar- rotia tetrodon Gaud., non P. tetrodon Ridl., R. Asiat. Soc. Str. Br. J. 68: 13-14, 1915. 1 Faculte des Sciences, Universite de Saigon, Viet- nam. Manuscript received April 1, I960. The genus Barrotia Gaud, was published in 1841 in the botanical atlas of the voyage of "La Bonite.” The new genus was not given a descrip- tion, and since it contained three species, the genus and its species are invalid and these can- not be adopted or transferred. Balfour in 1878 printed the combination P. tetrodon (Gaud.) Balf. f., but did not accept it himself. Hence, P. tetrodon Balf. /. ex Kanehira dates only from 1938 and is a later homonym of P. tetrodon Ridl. Kanehira attempted to solve the problem by renaming P. tetrodon Ridl. as P. singaporen- sis Kaneh. (1938), but this action was illegal, and Ridley’s species stands. For P. tetrodon Balf. /. ex Kanehira there is an available, legitimate name: P. compressus Martelli, based on a collection from the Solo- mon Islands. 327 Revision of the Genus Pandanus Stickman, Part 5 Pandanus of the Maidive Islands and the Seychelles Islands, Indian Ocean Harold St. John 1 The maldive islands form a western fringe of atolls off the coast of India and Ceylon. Their flora was summarized by J. C. Willis and J. S. Gardiner (1901), but Pandanus has not re- ceived particular attention. They recorded Pan- danus on many of the atolls, and described their utilization for mats, pillows, sails, and for food. Three species were listed: P. odoratissimus L. P. Lerarn Jones ex Fontana with the doubt- ful synonym (? P. Hornei Balf. /.), and P. sp. indet. The present writer considers the first to be a misidentification with the Ceylonese spe- cies; the second, P. Lerarn, is confined to the Andaman and Nicobar Islands, and the Maldive plant is doubtless the new P. Fosbergii St. John, and the third remains undetermined. Four addi- tional species of the affinity of P. odoratissimus are here described. Dr. F. R. Fosberg in 1956 observed and col- lected Pandanus on several of the islands. He identified (1957: 11) them as P. odoratissimus . His no. 36,916 is here described as P. maldive- cus St. John; and his no. 36,899 as P. Fosbergii St. John. Dr. W. D. Hartman was a member of the Yale University 1957 Expedition to the Indian Ocean. From his excellent collections there are here described several new species. Hartman also visited the Chagos Archipelago, but on his brief land excursions he found no Pandanus. SEYCHELLES ISLANDS The flora of these islands is well known and was well presented by Baker (1877). From the islands we know P. Balfourii Martelli, P. Hornei Bory, P. multispicatus Balf. /., and P. sechel- 1 B. P. Bishop Museum, Honolulu 17, Hawaii, U. S.A. Manuscript received May 20, I960. larum Balf. /. Of these, Hartman in 1957 col- lected all but the third, and his complete speci- mens allowed supplements to the previously published descriptions. For Aldabra Atoll, J. C. F. Fryer (1912: 402 ) recorded P. Vandermeeschii Balf. /. in Baker, a species now considered endemic to Mauritius. The Aldabra specimens are in need of further revision, and the writer has not seen them as yet. Pandanus OF THE MALDIVE ISLANDS Pandanus adduensis sp. nov. (sect. Pandanus) Fig. 12 NOM. VERN.: "karikayo.” DIAGNOSIS HOLOTYPI: Arbor 6-7 m. alta, corona late ovoidea, radicibus fulturosis evi- dentibus, foliis 150 cm. longis 6.3 cm. lads firme subcoriaceis ligulatis apice gradatim longe acuminatis apice ultimo non viso ad basem M-formatis, marginibus in regione basali cum aculeis 3-4 mm. longis 12-22 mm. distantibus adscendentibus forte subulatis cum basibus in- crassatis et apicibus obscure brunneis midnervio proximo infra cum aculeis 2.5-3 mm. longis 11-30 mm. distantibus reflexis crasse subulatis cum basibus incrassatis, in regione mediali mar- ginibus cum aculeis 2-3 mm. longis 8-22 mm. distantibus subulatis in 30° adscendentibus, midnervio proximo infra inermi; in regione apicali aculeis 0.4-0. 6 mm. longis crasse sub- ulatis in 45° adscendentibus eis marginalium 3-5 mm. distantibus illis midnervi 5-10 mm. distantibus adscendentibus, phalangibus 4. 5-4.9 cm. longis 3.4-4. 1 cm. latis 3-3.7 cm. crassis obovoideis basi cuneata apice valde convexo, parte Va supera libera lateribus laevibus lu- cidis in sicco brunneis fere planatis anguste Fig. 12. Pandanus adduensis, from the holotype. a, Phalange, lateral view, X 1; b, phalange, longitudi- nal section, X 1; G phalange, apical view, X 1; d, tree, X 1/100; e, leaf base, lower side, X If leaf middle, lower side, X 1; S> l ea f apex, lower side, X 1* 328 Page 30: Revision of Pandanus, 5. Maldives and Seychelles — St. John 329 i 2 cm i 2cm 330 PACIFIC SCIENCE, Vol. XV, July 1961 9 . l_L cm Page 32: Revision of Pandanus , 5. Maldives and Seychelles — St. John 331 '6-7-angulosis, suturis lateralibus nullis, sinibus apicalibus centralibus 3-4 mm. profundis V- formatis (phalangiis geminatis 5 — 5«5 cm. lads et carpellis 18-22), carpellis 9-13, apicibus hu- militer pyramidalibus eis centralibus valde ob- latis Mis marginalibus late distal! truncatis et etiam concavis, stigmatibus 2.5-4 mm. longis cordatis vel irregulariter ellipticis oblique in- clinatis centripetalibus, sutura proximal! V 5 -V 2 distancia ad fundum extenia, endocarpio paene submediali osseoso obscure brunneo marginibus lateralibus 2 mm. crassis, seminibus 8-10' mm. longis 3=5-5 mm. diametro oblique ellipsoideis, rnesocarpio apical! in quaque carpella cavernoso cum fibris longitudinalibus et membranis rne- dullosis, rnesocarpio basali fibroso. DIAGNOSIS OF HOLOTYPE: Tree 6-7 m. tall; crown broad ovoid; prop roots present; leaves 150 cm. long, 6.3 cm. wide, firm subcoriaceous, M -shaped in cross-section, ligulate, the apex gradually long acuminate, but the tip not seen, the margins near the base with prickles 3-4 mm. long, 12-22 mm. apart, stout subulate, broad based, ascending, dark brown tipped; the nearby midrib below with prickles 2.5-3 mm. long, 11-30 mm. apart, reflexed, stout subulate from an enlarged base; at the midsection the margins with prickles 2-3 mm. long, 8-22 mm. apart, subulate, ascending at 30°; the midrib unarmed; near the apex the margins and mid- rib below with ascending prickles 0.4-0 .6 mm. long, stout subulate, ascending at 45°, those of the margin 3-5 mm. apart, but those of the mid- rib 5—10 mm. apart; phalanges 4.5-4.9 cm. long, 3. 4-4.1 cm. wide, 3-3.7 cm. thick, obovoid with the base cuneate and the apex convex, up- per 14 free, the sides smooth, shining, brown when dry, almost plane, sharply 6-7-angled, lateral sutures none; central apical sinuses 3-4 mm. deep, V-shaped; (phalange twins 5-5.5 cm. wide and with 18-22 carpels); carpels 9- 13; carpel apices low pyramidal, the central ones strongly oblate, the marginal ones broad truncate and even concave distally from the stigma; stigmas 2.5-4 mm. long, cordate or ir- regularly elliptic, oblique, centripetal; proximal suture running l A~Vi way to valley bottom; endocarp slightly submedian, bony, dark brown, the lateral margins 2 mm. thick; seeds 8-10 mm. long, 3.5-5 mm. in diameter, obliquely ellipsoid; apical mesocarp in each carpel a cavity with longitudinal fibers and pithy partitions; basal mesocarp fibrous. HOLOTYPUS: Maldive Islands: Addu Atoll, Hitadu Islet, near shore, common, Oct. 13, 1957, W . D. Hartman 4 (Arn. Arb. ) . DISCUSSION : P. adduensis is a member of the section Pandanus . In that section its close rela- tive is an undescribed species from Papua, St. John 26,234 , which has the phalange upper third free, the apex low convex; carpel apices rounded pyramidal, the outer ones with an oblique outer plane surface reaching only to the stigma; stigmas 1-2 mm. long, ellipsoid; bark with papillae developing ascending rootlets. P. adduensis has the phalange with the upper quarter free, its apex convex; carpel apices low pyramidal, the marginal ones broad truncate and even concave distally from the stigma; stigmas 2.5-4 mm. long, cordate or irregularly elliptic; and the bark lacking ascending rootlets. The new epithet is made of Addu, the type locality; and - ensis , the Latin adjectival place ending. Pandanus Fosbergii sp. nov. (sect. Pandanus) Figs. 13-14 NOM. VERN.: "karikio.” DIAGNOSIS HOLOTYPI: Arbor 6-8 m. aka, foliis 3-4 ("ad saltern 4”) m. longis 11-13 cm. latis firmis spongioso-coriaceis minime AA-for- matis in sectio ligulatis sed ad apicem con- strictis in apice (acuto? ), basi amplexicaule exarmatis per 10 cm. deinde marginibus grosse subulato-serratis serris 3.5-5 mm. longis 10-25 mm. distantibus in apice brunneis, midnervio infra cum serris simulantibus 3 mm. longis 15- 30 mm. distantibus reflexis, in regione mediali marginibus cum serris 3-4 mm. longis 7-17 mm. distantibus grosse subulatis adpresse ad- scendentibus, midnervio infra salient! et cum dentis sumulantibus adscendentibus 4-8 cm. dis- tantibus, proxima apice marginibus subulato- FlG. 13. Pandanus Fosbergii, from holotype. a, Phalange, lateral view, i| 1; b, phalange, longitudinal section, X 1*, G phalange, apical view, X 1 \ d, e, stigmas, oblique view, X 4. 332 PACIFIC SCIENCE, Vol. XV, July 1961 serratis dentibus 1 mm. longis 1.5-3 mm. dis- tantibus adscendentibus, midnervio cum serris simulantibus sed 3-13 mm. distantibus, pedun- culo circa 30 cm. longo trigono folioso, bracteis longe subulato-acuminatis et cum spinis sim- ulantibus, syncarpio "oblongo scilicet 3 dm. longo 1.5 dm. diametro,” phalangibus 8.7 cm. longis 6.2 cm. latis 5.2 cm. crassis late cunei- form i bus sed in sectio transversali trigono, apice subconvexo, parte l A supera libera, suturis lat- eralibus nullis, lateribus laevibus gradatim con- vexis in sicco pallide brunneis obtuse 3-4-an- gulosis, sinibus centralibus apicalibus nullis suturis tantum lineis superficielibus tessellatis sed in post maturite cum rima lata, carpellis 6, apicibus subconvexis non protensis, stigmatibus 4-5 mm. longis suborbicularibus vel obreni- formbus centripetalibus tota elevatis obliquis, sutura proximali plerumque ad suturam extenta, endocarpio mediali 3-4 cm. longo osseoso ob- scure brunneo superficie interiori lucida, mar- ginibus lateralibus 2-3 mm. crassis, seminibus 2. 2-2 .4 cm. longis 11 mm. diametro anguste obovoideis, mesocarpio cum cavernis subdis- tinctis cum fibris et medulla firma persistenti, mesocarpio basali fibroso et carnoso et dicet esculenti. diagnosis OF holotype: Tree, up to 6-8 m. tall, large stemmed; leaves 3.4 "to at least 4” m. long, 11-13 cm. wide, firm, spongy coriace- ous, slightly M-shaped in cross-section, ligulate, but near the apex narrowed to the (acute?) tip, the base amplexicaul, unarmed for about 10 cm., then the margins coarsely subulate serrate, the teeth 3.5-5 mm. long, 10-25 mm. apart, brown-tipped, the midrib below with similar teeth 3 mm. long, 15-30 mm. apart, reflexed; at the midsection the margins with the teeth 3-4 mm. long, 7-17 mm. apart, heavy subulate, appressed ascending, the midrib below salient and with similar, ascending teeth 4-8 cm. apart, near the apex the margin with teeth 1 mm. long, 1.5-3 mm. apart, subulate-serrations, the midrib below with similar teeth but these 3-13 mm. apart; peduncle about 30 cm. long, trigon- ous, leafy bracted, the bracts similarly spiny, and long subulate acuminate; syncarp "oblong, at least 3 dm. X 1-5 dm., but none seen mature”; phalange 8.7 cm. long, 6.2 cm. wide, 5.2 cm. thick, broad cuneiform, but trigonous in cross section, the apex low convex, upper V6 free, lateral sutures none, the sides smooth, gently convex, pale brown when dried, with 3- 4 obtuse angles; apical central sinuses none, the suture lines merely superficial and tessellate; in post maturity a broad crack develops near each sutural region; carpels 6, their apices low con- vex, not projecting; stigmas 4-5 mm. long, suborbicular to obreniform, centripetal, wholly elevated and oblique; proximal crease running quite or nearly to the sinus line; endocarp me- dian, 3-4 cm. long, bony, dark brown, the inner walls polished and shining, the lateral walls 2- 3 mm. thick; seeds 2.2-2.4 cm. long, 11 mm. in diameter, narrowly obovoid; apical mesocarp of scarcely distinct caverns, with longitudinal fibers and a firm, persistent medullary tissue; basal mesocarp fibrous and fleshy, reported to be edible. HOLOTYPUS: Maidive Islands: Male Atoll, Kuda Bados Islet, occasional in coconut planta- tion, 1 m. alt., key picked up on ground nearby, April 1, 1956, F. R. Fosberg 36,899 (US). DISCUSSION: P. Fosbergii is a member of the section Pandanus, but the structure of the apex of the phalange is bizarre for that section, and at a glance more closely resembles the section Vin- sonia. This is due to the broad and obtuse apex of the phalange which is not lobed. The inter- carpellary apical sinuses are superficial and are mere tessellate lines, though in postmaturity wide cracks appear at or near the sinus lines. Diagnostic of the section Vinsonia are the flush, horizontal stigmas. In P. Fosbergii the stigmas are of another sort. They are large (4-5 mm. long) , strongly elevated, oblique and centripetal. Hence, under the present classification of the genus, the new species must be placed in the section Pandanus. Though not a close relative, the most similar appearing species is P. spiralis R. Br. of Austra- lia. This species has the phalanges 6 cm. long, the apex convex; carpels 9-20; stigmas 2-3 mm. long, hippocrepiform; seeds 4 mm. in diameter, very narrowly elliptic; and the apical mesocarp of separate caverns. P. Fosbergii has the pha- langes 8.7 cm. long, the apex low convex; car- pels 6; stigmas 4-5 mm. long, suborbicular; seeds 10 mm. in diameter, obovoid; and the apical mesocarp of confluent caverns. Page 34: Revision of Pandanus , 5. Maldives and Seychelles — St. John 333 This new species is named for its collector, Dr. F. Raymond Fosberg, of the U. S. Geological Survey. This species seems to be the one recorded (and doubtless misdetermined ) by J. C. Willis and J. S. Gardiner (1901: 104) as P. Let am Jones ex Fontana. That is a native of the distant Nicobar and Andaman Islands. It has massive phalanges roughly similar to those of the species here described, but the carpels are arranged in transverse, parallel lines, the stigmas are laterally directed, and the species belongs to a different section of the genus, Hombronia. They observed the plant, "a very large green stemmed Panda- nus . . . reaching a height of over 50 feet, found in the Hulule, Turadu, Goifurfehendu, Limbo Kandu, &c ( Gardiner ) , and known as Karikeo ( Keeva in Addu Atoll The tentative record of P. Hornet Balf. /. in Baker, by Willis and Gardiner (1901: 104), also seems to belong here. Pandanus Hartmanii sp. nov. (sect. Pandanus) Fig. 15 DIAGNOSIS HOLOTYPI: Arbor 5-6 m. alta circa 12 cm. diametro, ramis horizontalibus deinde adscendentibus griseis laevibus, radicibus fulturosis multis sublaevibus 1-2 m. longis circa 4 cm. diametro et frequenter in 6-10 dm. ex terra divisis in radicellis pluribus, foliis 120 cm. longis 5.7 cm. latis firme subcoriaceis ligu- latis in sectio late M-formatis longe attenuatis et in 10 cm. ex apice 2.5 mm. latis proxima basi marginibus cum spinis 3.5-5 mm. longis 8-18 mm. distantibus crassiter subulatis in 45° ad- scendentibus, midnervio cum spinis reflexis 2- 3 mm. longis 10-12 mm. distantibus crassiter subulatis, in sectio mediali marginibus cum spinis 3. 5-4.5 mm. longis 5-15 mm. distantibus adpressis crassiter subulatis, midnervio infra cum spinis 2-3 mm. longis 13-23 mm. distan- tibus adpressis et plerumque adscendentibus, proxima apice marginibus et midnervio infra cum spinis 0.6-0.8 mm. longis 7-14 mm. dis- tantibus, aculeis omnibus cum apicibus nigris, syncarpio solitario terminal! scilicet ( ex photo ) late ellipsoideo et 20 cm. long, 14 cm. diametro, phalangibus 4-4.3 cm. longis 3.2-4 cm. latis 2.8- 3.1 cm. crassis in sicco pallide bmnneis infra subluteis, lateribus planis vel cum sulcis vadosis paucis 4-6-angulosis, parte supera Vs libera, apice convexo, suturis lateralibus nullis, sinibus centralibus apicalibus 3-5 mm. profundis an- gustis, carpellis 12-14, plerumque 12, apicibus conicis sed valde oblatis et eis centralibus gra- datim minoribus interioribus plerumque Vi- 2 A tarn grandis quarn exterioribus, stigmatibus grandis prominentibusque centralibus elevatis sed subhorizontalibus interioribus 3-4 mm. longis 2-2.5 mm. latis ellipticis exterioribus 2— 3 mm. longis 3-4 mm. latis irregulariter cor- datis, endocarpio central! 1.7—2 cm. longo osse- oso brunneo lateribus 1-3 mm. crassis, semini- bus 11-14 mm. longis 4 mm. diametro late el- lipsoideis, mesocarpio apicali in quaque car- pella cavernoso cum membranis albis medullo- sis paucis, mesocarpio basali fibroso et carnoso. DIAGNOSIS OF HOLOTYPE: Tree 5-6 m. tall, about 12 cm. in diameter; branches horizontal, then ascending; trunk gray, smooth; prop roots 1-2 m. long, about 4 cm. in diameter, smooth, many of them at 6—10 dm. from ground divid- ing into a cluster of smaller rootlets; leaves 1.2 m. long, 5.7 cm. wide, firm subcoriaceous, lig- ulate, broad M -shaped in section, long attenuate and at 10 cm. from the tip only 2.5 mm. wide, near the base the margins with very stout subulate teeth 3.5-5 mm. long, 8-18 mm. apart, ascending at 45°, the midrib below with mostly reflexed prickles 2-3 mm. long, 10-25 mm. apart, stout subulate; at the midsection the marginal teeth 3-5 — 4.5 mm. long, 5-15 mm. apart, appressed, stout subulate and heavy-based; the midrib below with prickles 2-3 mm. long, 13-23 mm. apart, appressed, mostly ascending; near the apex the margins and midrib below with prickles 0.6-0.8 mm. long, 7-14 mm. apart; all prickles with darkish tips; syncarp solitary, terminal, apparently broad ellipsoid and 20 cm. long, 14 cm. in diameter; phalanges 4— 4.3 cm. long, 3.2-4 cm. wide, 2. 8-3.1 cm. thick, quadrate-cuneate, apex convex, upper Vs free, sides smooth, shining, light brown, 4-6-angled, the sides flat or with a few shallow furrows, the lower half when dry still yellowish; central apical sinuses 3-5 mm. deep; carpels 12-14, mostly 12; carpel apices conic but the apex much depressed, those towards the center grad- ually smaller, the central ones Vi~ 2 A as large as the outer ones; stigmas centripetal, large and 334 PACIFIC SCIENCE, VoL XV, July 1961 FIG. 14. Pandanus Foshergii, from holotype. a, Leaf base, lower side, | 1; leaf middle, lower side, X 1; c, leaf tip, lower side, X 1- tilOg Page 3 6: Revision of Pandanus, 5. Maldives and Seychelles — S t. John 335 prominent, the inner ones 3-4 mm. long, 2-2.5 mm. wide, elliptic, the outer ones 2-3 mm. long, 3-4 mm. wide, irregularly cordate, ele- vated, but more or less horizontal; endocarp central, 1.7-2 cm. long, bony, brown, the lateral walls 1-3 mm. thick; seeds 12-13 mm. long, broad ellipsoid; apical mesocarp with ellipsoid caverns mostly empty but with a few transverse white membranes; basal mesocarp fibrous. HOLOTYPUS: Maidive Islands, Haddumati Atoll, Hitadu Islet, common near shore, Oct. 10, 1957, W. D. Hartman 3 (Arn., Arb.). DISCUSSION: P. Hartmanii is a member of the section Pandanus. The closest relative seems to be P. tectorius Soland. var. spiralis (Blanco) Martelli, of the Philippines. It has the carpels 6-9, and their apices low pyramidal-convex; and the leaves 5 cm. wide. P. Hartmanii has the carpels 12-14, and their apices conic but the apex much depressed; and the leaves 5.7 cm. wide. This species is named for the anthropologist Dr. W. D. Hartman, member of the recent Yale University Expedition to the Indian Ocean. He collected all but one of the species of Pandanus known on the Seychelles Islands, and he col- lected excellent material of three new species on the Maidive Islands. His notes indicate the pres- ence of others there which were either not in fruit or the fruits of which were not attainable. Because of his excellent collections of these dif- ficult plants it is a pleasure to name this species for him. Pandanus Karikayo sp. nov. (sect. Pandanus) Figs. 16-17 NOM. VERN. "karikayo.” DIAGNOSIS HOLOTYPI: Arbor 4-5 m. alta scilicet 15 cm. diametro, corona ovoidea, cortice sublaevi, radicibus fulturosis evidentibus, foliis 100-125 (-300?) cm. longis 5.2 cm. latis ligu- latis subcoriaceis longe attenuatis ad basem V- formatis sed ultra planis in regione basali mar- ginibus cum aculeis 0.5-1 mm. longis 3-5 mm. distantibus salientibus, in regione 15 cm. ex basi cum aculeis 5-7 mm. longis 13-26 mm. distanti- bus subulatis in 45° adscendentibus, midnervio infra cum aculeis 4-4.5 mm. longis 1 5—30 mm. distantibus crassiter subulatis recurvatis, in sec- tio mediale marginibus cum aculeis 5-7 mm. longis 13-26 mm. distantibus adscendentibus, midnervio cum aculeis infra 3.5-4 mm. longis 1 5—30 mm. distantibus recurvatis, circa apice marginibus midnervioque cum aculeis 0.6-1 mm. longis 3-8 mm. distantibus crassiter subu- latis adpresse adscendentibus, aculeis paene totis pallidis, syncarpio unico terminali scilicet 18 cm. longis et 13 cm. diametro rubro late ellip- soideo, phalangibus multis 3. 8-4.4 cm. longis 2. 1-3 .4 cm. latis 2-2.6 cm. crassis obovoideo- cuneatis apice subconvexo parte V4 supera libera lateribus laevibus lucidis in sicco pallide brun- neis curvatis vel minime sulcatis 4-5-angulosis, suturis lateralibus nullis, sinibus apicalibus cen- tralibus 4-8 mm. profundis omnibus angustis vel paucis cum f undam latam, carpellis 10-14 et rare cum carpella laterali abortiva, apicibus conicis adscendentibus depresse subrotundatis omnibus aequalibus vel centralibus minoribus 2 A tarn grandibus quam marginalibus eis saepe conicis asymmetricalibus et dare divengentibus, stigmatibus 1.5-2. 5 mm. longis ellipsoideis ob- liquis elevatis prominentibus centripetalibus breve sulcatis, sutura proximali V$ distancia ad fundam extenta, endocarpio centrali osseoso ma- hogani-colorato marginibus lateralibus 1.5-2 mm. crassis, seminibus 10-11 mm. longis ellip- soideis, mesocarpio apicali in quaque carpella cavernoso fibroso et cum membranis subalbis medullosis, mesocarpio basali cum fibris forti- bus. DIAGNOSIS OF holotype: Trees 4-5 m. tall, about 15 cm. in diameter; trunk rather smooth; crown ovoid; prop roots diverging at about 40°; leaves 100-12,5 (-300?) cm. long, 5.2 cm. wide, subcoriaceous, ligulate, long attenuate, to- wards the base V-shaped, beyond it plane, near the base the margins with salient prickles 0.5— 1 mm. long, 3-5 mm. apart, but at 15 cm. up from the base the prickles 5-7 mm. long, 13— 26 mm. apart, subulate, ascending at 45°, the nearby midrib below with prickles 4-4.5 mm. long, 1 5—30 mm. apart, stout subulate, recurv- ing; at the middle the margins with prickles similar to those near the base but ascending at 25°-40°, the midrib below with prickles 4-4.5 mm. long, 15-30 mm. apart, stout subulate, recurving; near the apex the margins and mid- rib below with prickles 0.6-1 mm. long, 3-8 2cm FIG. 15. Pandanus Hartmann, from holotype. a, Phalange, lateral view, X 1; b, phalange, longitudinal section, X 1; c, phalange, apical view, XI \ d, tree, X 1/75; e, leaf base, lower side, X 1; f, leaf middle, lower side, X 1; g, leaf tip, lower side, X 1. 2cm Page 38: Revision of Pandanus, 3. Maldives and Seychelles — S t. John 337 Fig. 16. Pandanus Karikayo, from holotype. a. Phalanges, lateral view, X 1 ; b, phalange, longitudinal section, X 1; ^ phalange, apical view, X 1; 4 tree, X 1/80; e , leaf apex, lower side, X 4; f, leaf apex, upper side, X 1- 2cm 338 PACIFIC SCIENCE, Vol. XV, July 1961 0 1 i— i Fig. 17. Pandanus Karikayo, from holotype. a, Leaf mm. apart, stout subulate, closely ascending; prickles all pale; syncarp single, terminal, broad ellipsoid, apparently 18 cm. long, 13 cm. in diameter, "bright red”; phalanges numerous, 3. 8-4.4 cm. long, 2. 1-3.4 cm. wide, 2-2.6 cm. thick, obovoid-cuneate, apex slightly convex, up- per 14 free, sides smooth, shining, when dry light brown, curving or slightly channeled, 4-5- angled; lateral sutures none; apical central si- nuses 4-8 mm. deep, all narrow or a few broad based; carpels 10-14, rarely with a tiny, lateral, splintlike one also; carpel apices ascending, de- pressed, somewhat rounded conic, but the mar- ginal ones often conic, asymmetric and distinctly divergent, all subequal or the ones toward the center somewhat smaller, these about 34 as large as the marginal ones; stigmas 1.5-2. 5 mm. long, ellipsoid, short creased, prominent, raised, 2 cm — i base, lower side, X 1 ^ leaf middle, lower side, X 1. oblique, centripetal; proximal suture extending Vb way to the valley bottom; endocarp central, bony, mahogany-colored, the lateral margins 1.5- 2 mm. thick; seeds 10-11 mm. long, ellipsoid; apical mesocarp in each carpel forming a large cavern with strong longitudinal fibers and whitish pithy partitions; basal mesocarp with strong fibers. HOLOTYPUS: Maidive Islands, Fadiffolu Atoll, Kuredu Islet, common on shore and middle, Sept. 25, 1957, W. D. Hartman 1 (Arn. Arb. ) . DISCUSSION: P. Karikayo is a member of the section Pandanus. Its closest relative appears to be P. odoratissimus L. /., of the island of Ceylon, which has the phalanges with the upper l A free; the central apical sinuses 4 mm. deep; and the stigmas of the outer carpels horizontal. In P. Karikayo the phalanges have the upper 14 Page 40: Revision of Pandanus, 5. Maldives and Seychelles — St. John 339 free; the central apical sinuses 4-8 mm. deep; and the stigmas of the outer carpels divergent. The new specific epithet is the vernacular name of the species on Fadiffolu Atoll. Pandanus maldivecus sp. nov. (sect. Pandanus) Fig. 18 NOM. vern.: "karikayo.” DIAGNOSIS HOLOTYPI: Arbor 2-6 m. alta, trunco scilicet 12 cm. diametro cum foliis sub- persistentibus, radicibus fulturosis scilicet 3-6 dm. longis et laevibus, foliis 117 cm. longis 4 cm. latis ligulatis subcoriaceis longiter attenu- atis in sectio ad basem W-formatis ad mediam planatis, apice longe acuminato et in 10 cm. ex apice 2.5 mm. lato, ad basem marginibus cum aculeis 2-4.5 mm. longis 9-22 mm. distantibus crassiter subulatis adscendentibus, midnervio ad basem infra cum aculeis 2-3 mm. longis 1-3 cm. distantibus subulatis valde reflexis, in re- gione mediali marginibus cum aculeis 1.5-2. 5 mm. longis 8-17 mm. distantibus, midnervio infra cum aculeis 1.5-2 mm. longis 15-22 mm. distantibus subulatis adscendentibus, proxima apice marginibus cum dentibus 0.3-0. 5 mm. longis 3-7 mm. distantibus, aculeis omnibus cum apice nigro, syncarpio solitario terminali 14 cm. longo 11 cm. diametro late ellipsoideo cum circa 60 phalangibus eis 4.8-5 .2 cm. longis 2. 5-3 .4 cm. latis 2.2-2 .9 cm. crassis obovoideis cuneatis scilicet luteis in sicco pallide brunneis laevibus lucidis lateribus planatis vel rare va- doso-sulcatis 4-6-angulosis parte Va supera li- bera, apice convexo sinibus apicalibus centrali- bus 2-3 mm. profundis angustis vel latis, car- pellis 5-9 plerumque 6, apicibus valde depresso- pyramidalibus eis marginalibus clariter cum apice et stigmate divergenti, apicibus interiori- bus minoribus, centralibus 2 A tarn grandis quam marginalibus, stigmatibus 2. 5-3. 5 mm. longis ovatis vel rhombicis prominentibus elevatis cen- tripetalibus, sutura proximali Va distancia ad fundum extenta, endocarpio paene submediali osseoso obscure mahogani-colorato marginibus lateralibus 1.5-2 mm. crassis, seminibus 11-13 mm. longis 4 mm. diametro asymmetriciter el- lipsoideis, mesocarpio apicali cum cavernis cum membranis firmis albis completis, mesocarpio basali fibroso et carnoso. DESCRIPTION OF ALL SPECIMENS EXAMINED: Tree 2-6 m. tall, and about 12 cm. in diameter, the bark and roots appearing smooth; trunk usually more or less clothed with persistent, dead leaves; leaves 117 cm. long, 4-5 cm. wide, ligulate, subcoriaceous, towards the base W- shaped, further out plane, then long attenuate, the apex 10 cm. back only 2.5 mm. wide; near the base the margins with prickles 2-4.5 mm. long, 9-22 mm. apart, stout subulate, ascend- ing; the midrib the only prominent nerve, and near the base with prickles below 2-3 mm. long, 1-3 cm. apart, strong subulate, reflexed; near the middle the margins with prickles 1.5- 2.5 mm. long, 8-17 mm. apart; the midnerve below with prickles 1.5-2 mm. long, 15-22 mm. apart, reflexed; near the tip the margins and midrib below with prickles 0.3-0. 5 mm. long, 3-7 mm. apart; all the prickles dark-tipped; syncarp solitary, terminal, 14 cm. long, 11 cm. in diameter, broad ellipsoid, with about 60 pha- langes, these 4.8-5. 2 cm. long, 2. 5-3 .4 cm. wide, 2. 2-2 .9 cm. thick, obovoid, cuneate, apparently yellow, when dry pale brown, shining, the sides smooth, shining, plane or a few of them shal- low furrowed, 4-6-angled, upper Va free, the apex convex; central apical sinuses 2-3 mm. deep, narrow or broad; carpels 5-9, mostly 6; carpel apices much depressed pyramidal, the marginal ones with apex and stigmas distinctly diverging, the others decreasing in size towards the center, the central ones being about 2 A size of the marginal ones; stigmas 2. 5-3. 5 mm. long, ovate to rhombic, short creased, prominent, raised, centripetal; the proximal suture running Va way to valley bottom; endocarp slightly sub- median, bony, dark mahogany-colored, the lateral margins 1.5-2 mm. thick; seeds 11-13 mm. long, 4 mm. in diameter, asymmetric ellipsoid; apical mesocarp with a cavern in each carpel, these filled with white, firm, cross membranes; basal mesocarp fibrous and fleshy. HOLOTYPUS: Maidive Islands, South Malos- madulu Atoll, Dunikolu Islet, common, chiefly on the shore, Oct. 2, 1957, W. D. Hartman 2 (Arn. Arb.). SPECIMENS EXAMINED: Maidive Islands, Male Atoll, Male Islet, forming thicket back of sea- ward beach, 1-2 m. alt., nom. vern "bokio,” fruits red on ripening, not eaten or used for 2cm Fig. 18. Pandanus maldwecus, from holotype. a, Phalange, lateral view, X 1; h; phalange, longitudinal section, S pa t: c. phalange, apical view, XI \ d, tree, X 1/60; e, leaf base, lower side, X 1; f, leaf middle, lower side, X 1; g, leaf apex, lower side, X 1- 2cm Page 42: Revision of Pandanus, 5. Maldives and Seychelles — St. John 341 anything, April 12, 1956, F. R. Fosberg 36,916 (US). DISCUSSION: P. maldivecus is a member of the section Pandanus. Its closest relative seems to be P. tectorius Sol. var. sarawakensis Martelli of Borneo. This has the phalanges 6-6.5 cm. long, the upper Vl free, the sides flat or more or less concave; central apical sinuses apparently 4-5 mm. deep; and the stigmas 2-2.8 mm. long, indexed even on the marginal carpels. In con- trast, P. maldivecus has the phalanges 4.8-5 .2 cm. long, upper free, the sides plane and curved or shallowly furrowed; central apical sinuses 2-3 mm. deep; and the stigmas of the marginal carpels distinctly diverging. The new specific epithet is taken from the geographic name Maidive, to which is added a Latin adjectival suffix. Pandanus OF THE SEYCHELLES ISLANDS Pandanus Balfourii Martelli, Webbia 1: 361— 362, 1905; 4(1): 7, t. 4, figs. 4-7, 1913 ( sect. Pandanus ) Fig. 19 This species was briefly described, then later illustrated by Martelli, from collections from the Seychelles made by Horne and by Dupont. It is a well-marked species, and its status is not in question. However, the collection, data, and photos by Dr. W. D. Hartman, made in 1957, make possible an expanded and more complete description, including stems, leaves, etc. The supplementary descriptive details are as follows: Tree 5-6 m. tall, and at a point just above the roots 15-17.5 cm. in diameter, but below the first branches 20-23 cm. in diameter, ap- pearing smooth; crown semi orbicular; prop roots borne up to 6 dm. from the ground; leaves 125-145 cm. long, 4.6-4.S cm. wide, firm, strong, subcoriaceous, ligulate, AA-shaped in cross-section, the tip long, slender acuminate to a subulate tip, this at 10 cm. down only 2 mm. wide, at 20 cm. only 8 mm. wide, near the base the margins with prickles 3-3.5 mm. long, 3-10 mm. apart, stout, broad-based subu- late, salient ascending, the nearby midrib below beginning 12 cm. up with prickles 2.5-3 mm. long, 10-25 mm. apart, stout subulate, reflexed; at the midsection the margins with prickles 2— 2.5 mm. long, 2-5 mm. apart, appressed ascend- ing and more slenderly subulate, tipped with dark brown; the midrib below with prickles l. 5-2 mm. long, 7-15 mm. apart, slender subu- late, ascending-subappressed; near the apex the margins and midrib below with prickles 0.3- 0.5 mm. long, 2-3 mm. apart, stout subulate, broad based, appressed-ascending, brown tip- ped; phalanges 102; these 3.7-4 cm. long, 2- 4.3 cm. wide, 1.6-2. 5 cm. thick, upper 1/5 -Va free, apex low convex, sides gently convex or plane, shining, pale brown when dry; lateral sinuses none; lateral furrows none or slight, 4— 6- angled, lower half fibrous and fleshy and orange; central apical sinuses 1-2 mm. deep, shallow, broad V-shaped; carpels 5-10, their apices low, rounded pyramidal, green when ripe; stigmas 1-2 mm. long, suborbicular to cordate; endocarp dark brown, slightly supramedian, the lateral margins 1-2 mm. thick. SEYCHELLES ISLANDS: Mahe Island, opposite Souris Island, group on granite cliffs near shore, Nov. 9, 1957, W . D. Hartman 6 (Arn. Arb. ). Pandanus Hornei Balf. f., in Baker, FI. Mauritius and Seychelles 397, 1877; Martelli, Webbia 4(1): 17, t. 16, figs. 1-3, 1913 (sect. Dauphinensia ) , new sectional placement Figs. 20-21 NOM. vern.: "vacoa parasol.” This very distinct species has been known for nearly a century. The phalanges were later illus- trated by Martelli, but his figures did not clar- ify the remarkable stigmatic apparatus. Now, the recent collection and data by Dr. W. D. Hartman allow the furnishing of a supplemen- tary description and more adequate illustrations, as follows: Trees 15-20 m. tall, and the trunk at 3 m. from the ground 15-17.5 cm. in diameter, at 7- 10 m. up 20-22.5 cm. in diameter, tall straight, twice trichotomous, then dichotomous; prop roots 2.4-3 m. long, nearly vertical and seeming to hug the trunk; leaf (1 seen) 2.25 m. long, 10 cm. wide, subcoriaceous, gradually tapering to the elongate apex (but the tip lost), at base 1 -furrowed, further out M -plicate, at base the margin entire for 11 cm., then in- acm FIG. 19- Pandanus Balfourii, from Mahe I., Hartmai 6. a, Phalange, lateral view, XI \b, phalange, longi- tudinal section, X 1; c, phalange, apical view, X 1; d, tree, X 1/56; e, leaf base, lower side, X 1; f> leaf middle, lower side, X 1, and one prickle X 2; g, leaf apex, lower side, X 1. 2cm Fig. 20. Pandanus Hornet, from Praslin I., Hartman 7. a, Phalange, lateral view, XI \ b> phalange, longi- tudinal section, X 1; c, phalange, apical view, X 1; d, medium-size tree, X 1/300; e, large tree, X 1/333. 2 cm 2cm Fig. 21. Pandanus Hornet, from Praslin I., Hartman 7. a, Trunk, branches, and prop roots, b, leaf base, lower side, X 1; o, leaf middle, lower side, X 1; d, leaf apex, lower side, X 1- Page 46: Revision of Pandanus , 5. Maldives and Seychelles — ST. JOHN 345 creasingly coarsely aculeate, the prickles 3-4.5 mm. long, 3-10 mm. apart, stout subulate, up- ward curved, blackish; the nearby midrib un- armed; at midsection the marginal prickles 2 mm. long, 3-5 mm. apart, dark brown, subulate, upward curved; the midrib below with prickles 1 mm. long, 4-6 mm. apart, subulate, appressed ascending; near the tip the margins and mid- rib below with aculeate-serrations 0. 2-0.4 mm. long, 1-2 mm. apart, brown, appressed ascend- ing; pistillate inflorescences single, terminal; phalanges (4 seen) 12.2-13.2 cm. long, 5.3- 6.8 cm. wide, 5-6.3 cm. thick, cuneate oblanceo- loid, the apex rounded, upper free and green, glaucous, the lower 44 fleshy, reddish orange, the sides smooth, above slightly convex, below plane; apex with a central rounded rim and within a concavity 2 mm. deep, the surface of which is covered by several irregular, unequal, cordate elevations which at first glance would seem to be stigmas, but are not these; and in some there is an apical central crack 1 mm. deep; along the apical rim are two arcuate or paren- thesis-like, wavy ridges which are the two cen- tripetal stigmas which along the perimeter are 20-22 mm. long; indocarp in upper 2/5, bony, dark brown, the lateral margins at the thinnest parts 3-4 mm. thick; seeds 45-46 mm. long, 10-11 mm. in diameter; mesocarp continuous, of solid pith and few longitudinal fibers. SEYCHELLES ISLANDS: Praslin Island, Vallee de Mai, near stream, 500-600 ft. elev., Nov. 11, 1957, W. D. Hartman 7. He also recorded, "Another stand of these trees was later seen (and photographed) grow- ing along a stream near Newcombe Forest, Praslin Island. They were not quite as tall as the trees in the Vallee de Mai.” DISCUSSION: The crowded, cordate elevations in the apical concavity of the phalange are de- ceptively like stigmas. Only after careful study of the apex and of cut sections was it demon- strated that the actual stigmas were the two arcuate or parenthesis-like projections on the central apical rim. It would have been a pleasure to publish Hartman § excellent color photos of this species, but the cost forbids. Fig. 22. Pandanus sechellarum, from Mahe L, Hart- man 5 . a, Leaf base, lower side, X 1; b, leaf middle, lower side, X leaf apex, lower side, X 1. o i 2 cm 1-.. « I 346 PACIFIC SCIENCE, Vol. XV, July 1961 Pandanus multispicatus Balf. /., in Baker, FI. Mauritius and Seychelles 403, 1877; War- burg, Engler’s Pflanzenreich IV, 9: 67-68, 1900; Martelli, Webbia 4(1): 24, 1913 (sect. Micro stigma') SEYCHELLES ISLANDS: endemic. Martelli’s record of this species for Reunion Island was not confirmed by Vaughan and Wiehe (1953: 31). This species has never been illustrated. Pandanus sechellarum Balf. /., in Baker, FI. Mauritius and Seychelles 402, 1877; Mar- telli, Webbia 4 ( 1 ) : 30, t. 7, figs. 6^8, 1913 Fig. 22 nom. vern.: "vacoa marron.” This was collected by Hartman in 1957, but since he found only half-grown fruits, his col- lection furnished only vegetative characters to supplement the known ones, as follows: Prop roots present up to 6 dm. from ground; leaves 88-90 cm. long, 4-4.2 cm. wide, sub- coriaceous, dark green above, paler below, lig- ulate, tapering gradually to the acute apex, this at the point 10 cm. down 11 mm. wide, AA- shaped in section, the enlarged base entire, but beginning 10 cm. up the margins with prickles 1.5-2 mm. long, 7-16 mm. apart, acicular, appressed and seeming like a continuation of the margin since located above a concavity; the nearby midrib unarmed; at the midsection the teeth of the margins 1-1.5 mm. long, 11-26 mm. apart, acicular, appressed ascending; the midrib below with prickles 0.5-1 mm. long, 10-20 mm. apart, acicular, appressed ascending; near the tip the margins and midrib below with teeth 0.2 mm. long, 3-6 mm. apart, subulate, broad based, appressed ascending. SEYCHELLES ISLANDS: Mahe Island, opposite Souris Island, group of 6-8 trees among granite boulders near shore, shaded by coconut palms, Nov. 9, 1957, W. D. Hartman 5. REFERENCES Baker, John Gilbert. 1877. Flora of Mauri- tius and the Seychelles. London. 20 + 557 pp. Fosberg, F. Raymond. 1957. The Maidive Is- lands, Indian Ocean. Atoll Res. Bull. 58: 1-37. Fryer, J. C. F. 1902. The structure and forma- tion of Aldabra and neighbouring islands — with notes on their flora and fauna. Linn. Soc. Lond. Zool. Trans. 14(2): 397-442. Summerhayes, V. S. 1931. An enumeration of the angiosperms of the Seychelles Archipel- ago. Linn. Soc. Lond. Zool. Trans. 19(2): 261-299. Vaughan, R. E., and P. O. Wiehe, 1953. The genus Pandanus in the Mascarene Islands. Linn. Soc. Lond. Bot. J. 55(356): 1-33, map, figs. 1-3, pis. 1-4. Willis, J. C., and J. S. Gardiner. 1901. The botany of the Maidive Islands. R. Bot. Gdn. Peradeniya Ann. 1(2): 45-164. A Contribution to the Ecology of the Kermadec Islands V. J. Chapman 1 In 1956 and 1957, owing to the kindness of the Civil Aviation Department, to whom especial thanks are due, it was possible for two visits to be made to the Kermadec Islands. These visits were made under the aegis of the Botany De- partment, Auckland University, though on the first occasion the personnel included members from elsewhere. The first visit took place in May and June, 1956, and the members of the party were Dr. R. C. Cooper (botanist, Auck- land Institute and Museum), P. L. Bergquist (Botany Dept.), and J. S. Edwards (Zoology Dept.). This party was originally expected to stay only a fortnight but, owing to circum- stances beyond our control, they were there for a month. Transport facilities to the Kermadecs were generously provided by the Royal New Zealand Navy and return transport by the N. Z. Institute of Oceanography, to whom thanks are due. The second visit took place in October, 1957, when only one scientist could be accom- modated on the normal supply vessel. The late D. Knowlton went on this occasion, and in the course of the three days available he was able to achieve much valuable work. The original purpose of these visits was to set up permanent quadrats so that the regeneration of vegetation after the reduction of the goat population could be followed over a course of years. It Is therefore hoped that further visits may be made in the future. The first visit in 1956 coincided with the end of a drive to elimi- nate the goats from the island, when some 1,500 goats were shot. About a hundred more were shot in 1956-57. As a result of these activities the goat population was reduced to the point where it could reasonably be hoped that it might be kept in check. Another purpose of the visits was to study more intensively the algal flora of the Kermadecs. It is true that earlier collections and lists had been made by the Gepps (1911) 1 The University of Auckland, New Zealand. Manu- script received November 7, I960. and by Cotton (1912), but In view of the impor- tance of the Kermadecs in relation to sea tem- peratures and tropical influences, it was believed that a more intensive study would reveal the existence of more species of tropical affinities. This has indeed proved to be the case. Later, in another communication, it is intended to com- pare the flora of these islands with those from the North Cape region of New Zealand and Norfolk Island. So far as the terrestrial flora is concerned, ex- tensive lists were already in existence as a re- sult of visits by Cheeseman in 1887 and by Oliver in 1909 to the islands. These earlier workers had collected methodically and exten- sively and there were no outstanding additions to be made to the earlier lists. One feature of interest was the way in which the taro, Alocasia macrorrhiza, had spread over the island since its introduction some time after 1909. ALGAE In the list that follows, algae recorded also by the Gepps (1911) are noted by G and those also recorded by Cotton by C. Algae newly added to the New Zealand flora are marked by +• Those not collected in 1956 or 1957 are marked by *. MYXOPHYCEAE 2 Chlorococcales Anacystis thermalis (Menengh.) Dr. & Daily. Supralittoral fringe pool, Lava Pt. Anacystis montana ( Lightf . ) Dr. & Daily. On rotten log, Green Lake; supralittoral fringe pool, Lava Pt. Pleurocapsales EntO'pAysalis ponferta (Kiitz.) Dr. & Daily. Fleetwood Bluff, supralittoral fringe: on Derbesia 4-8 m. down, Meyer Is. 2 1 am grateful to Dr. F. Drouet for determinations of this group. 347 348 PACIFIC SCIENCE, Vol. XV, July 1961 Nostocales Lyngbya confervoides C. Ag. Fleetwood Bluff, supralittoral fringe; mid-littoral pool, Meyer I.; on Pterocladia, Lava Pt. Lyngbya semiplena (C. Ag.) J. Ag. Fishing Rock, sheltered supralittoral pool, also Hutchinson’s Bluff: on Pterocladia, Lava Pt. Phormidium cor mm ( Ag. ) Gom. On Pter- ocladia, Lava Pt. Phormidium autumnale (Ag.) Gom. Local- ity unknown. Hydrocoleum glutinosum (Ag.) Gom. Fleetwood Bluff, sheltered crevice (mid- littoral ) , also supralittoral fringe. Hapalosiphon laminosus Hansgirg. On fu- marole orifice. Tolypothrix tenuis Kiitz. Supralittoral fringe pool, Lava Pt. S cytonema hofmani Ag. On rotten log, Green Lake. Amphithrix violacea (Kiitz.) Born, et Flah. Fleetwood Bluff, supralittoral fringe. Plectonema nostocorum Bornet. Supralit- toral fringe pool, Lava Pt. Calothrix Crustacea Thur. Fishing Rock, sheltered supralittoral pool, also on wet concrete. CHLOROPHYCEAE Ulotrichales G Ulva lactuca var. rigida (Ag.) Le Jol. G * Ulva laetevirens Aresch. t Enter omorp ha kylinii Bliding. The record of this species gives the alga a great discontinuity. In view of the relatively recent recognition of this species, however, it is likely that a wider distribution will be found for it than at present recorded. Supralit- toral fringe, Hutchinson’s Bluff, on rocks always wet with spray. G # Enteromorpha comprCssa Grev. Enter omorpha prolifera (Muel.) J. Ag. f. crinata (Roth.) comb. nov. Young plants which probably belong to this species. Supralittoral fringe, Meyer I. Chaetophorales Trentepohlia jolithus Wallr. Oedogoniales Oedogonium sp. ) Neither fertile, Bulbochaete sp. \ Green Lake. Siphonocladales Lola tortuosa (Dillw. ) Chapm. Fishing Rock, sheltered supralittoral pool. Rhizoclonium hookeri Kiitz. Fleetwood Bluff, on Pohutakawa log. Rhizoclonium hier o glyphicum Kiitz. emend Stockm. Waterfall near Hut- chinson’s Bluff; Blue Lake. G C Cladophora prolifera ( Roth ) Kiitz. This is recorded as C. fusca Marten by both the Gepps and Cotton. For a discus- sion on this matter the reader is re- ferred to Chapman (1956). Deep mid-littoral pool, Lava Pt.; sublittoral, Meyer I. to 8 m. Cladophora fracta (Dillw.) Kiitz. var. lacustris (Kiitz.) Brand ex Heering. Waterfall near Hutchinson’s Bluff. Clad op h or op sis herpestica (Mont.) Chapm. Boat Cove, Fishing Rock, t Cladophoropsis membranacea Borg, t Cladophoropsis membranacea Borg. var. repens (J. Ag.) Phinney. I believe two plants have been confused here. There is a cushion-like Cladophora, C. repens, and there another plant which is clearly a Cladophoropsis in that the branches have no cells at the base. Su- pralittoral pool, Meyer I. Microdictyon umbilicatum (Velley) Za- nard. Only occurs elsewhere in New Zealand on Mayor I. (Chapman, 1956). Dasycladales f Acetabularia parvula S o 1 m s-L a u b a c h. Mid-littoral pool, Fishing Rocks. Siphonales Derbesia novae-zelandiae Chapm. This was probably the plant that the Gepps recorded as V aucheria sp. Pool in sub- littoral fringe, pool mid-littoral, Fish- ing Rocks; sublittoral, 4-8 m., Meyer I. Ecology of Kermadec Islands — CHAPMAN 349 Fig. 1. Bryopsis kermadecensis. a, Base of pinnule showing constriction and plug of protoplasm; b, apex of pinnule; c, plant (X 2); d, pinnule. t Bryopsis kermadecensis sp. nov. (Fig. 1). Plantis 1.5 cm. longis, axe primario distincto, simplici, 149-158 p dia., pariete 6-6.5 p crasso, partibus in- ferioribus nudis; pinnulis inferioribus caducis, pinnulis superioribus radiali- ter dispositis, basi constrictis, apici obtusis, 44-61 p dia. Plants 1.5 cm. long, main axis distinct, unbranched, 149-158 p diam., wall 6- 6.5 p thick, lower part bare below; lower pinnules deciduous, upper pin- nules radially arranged, constricted at the base, obtuse at the apex, 44—61 p diam. Type specimen in Lind. Herb. Auckl. Univ. This species is very close to B. der- besioides Chapm., but differs in its smaller size and the greatly constricted bases of the branches. The diameter of the pinnules is also more uniform and the apices obtuse rather than ta- pering. It appears to be smaller than other species of Bryopsis recorded from the Australian mainland. For the present, therefore, it has been re- garded as a new species, though ex- tensive collecting will be necessary be- fore any further decision can be made. On Corallina in sublittoral pool, Fish- ing Rocks. G * C odium dichotomum (Huds.) S. F. Gray f. novozelandicum Dellow. This is the plant recorded by Gepp as C. tomentosum . (See Chapman, 1956.) Codium adhaerens (Cabr.) Ag. var. convolutum Dellow. Sublittoral, One- rahi Bluff. f Caulerpa racemosa ( Forsk. ) J. Ag. var. peltata (Lmx.) Eubank, Mid-tide pool, Fishing Rocks. G Caulerpa racemosa (Forsk.) J. Ag. var. uvifera (Turn.) J. Ag. f. intermedia Web. van Bosse. Fishing Rock, mid- littoral pool: sublittoral, 12 m., and sublittoral fringe, Meyer I. t Caulerpa webbiana Mont. Deep pool, mid-tide, Fishing Rock with Corallina\ sublittoral, 12 m., Meyer I. PHAEOPHYCEAE Dictyotales G Dictyota prolificans A. & E. S. Gepp. In red turf, Hutchinson’s Bluff, mid-lit- toral. CG Taonia australasica (Kiitz.) J. Ag. Sub- littoral, 4—8 m., Meyer I. CG Pocockiella nigrescens (Sond.) Papenf. Pool, Boat Cove: in Corallina turf, sublittoral, Fishing Rocks. Padina fraseri (Grev.) J. Ag. (Lin- dauer, 1957). Hydroclathrus clathratus (Bory) Howe (Lindauer, 1957). C * Haliseris kermadecensis Cotton Sporochnales C * Perithalia capillaris J. Ag. Dictysiphonales Colpomenia sinuosa (Roth.) Derb. et Sol. Fishing Rocks. Fucales C * Sargassum sinclairii Hook, f . & Harv. I think some doubt attaches to this iden- tification. G Sargassum fissifolium (Mert.) C. Ag. Fishing Rock, mid-tide pool: also in the sublittoral fringe, Lava Pt. Carpophyllum phyllanthus, C. plumo- sum, C. maschalocarpum, Durvillea antarctica, and H ormosira banksii have all been recorded from the drift (Cotton, Gepp) but there is no evi- 350 PACIFIC SCIENCE, Vol. XV, July 1961 dence that they grow there. The record of Carpophyllum elongatum is erro- neous, as inspection of the material at Kew and the British Museum shows that the specimens are worn plants of C. mas chalo carpum (Chapman, in press ) . RHODOPHYCEAE Bangiales f Porphyra denticulata Levr. Previously only reported from Queensland. Sub- littoral fringe, Lava Pt., and supralit- toral fringe pool. Nemalionales G * Chantransia sp. t Nemalion helminth oides (Velley) Bat- ters. Very exposed rock faces, Fishing Rocks. G * Galax am a lax a Kjellm (as Brachycladia marginata Schm. in Gepp ) . t Galaxaura arborea Kjellm. Mid-littoral pool, Meyer I. t Galaxaura rudis Kjellm. Sublittoral, 16 m., Meyer I. G f Galaxaura fastigiata Dene. ( as G. lap- ides cens in Gepp ) . Sublittoral, 3 m., Meyer I. G Asparagopsis taxiformis (Delile) Col- lins & Hervey (= A. sandfordiana ) . Very common in sublittoral: extends into sublittoral fringe where exposure and wave action not great. C G Delis ea ffmbriata ( Lamour. ) Mont. Sub- littoral to 4-5 m., Meyer I. ( = D. pulchra, D. s errata ) . Chaetangium corneum J. Ag. f Chaetangium pulvinatum Levr. Turf, mid-littoral, Fishing Rocks. Gelidiales G Gelidium longipes J. Ag. Fishing Rock. Mid-tide pool. Gelidium crinale J. Ag. Turf, lower mid- littoral, Flutchinson’s Bluff. Gelidium pusillum Le Jol. Turf, lower mid-littoral, Hutchinson’s Bluff. Gelidium caulacantheum J. Ag. CG Pterocladia capillacea (Gmel.) Born, et Thur. Fishing Rock, Lava Pt.: sub- littoral, 4-8 m. Meyer I. Cryptonemiales G * Peyssonnelia rubra (Grev.) J. Ag. CG Corallina cuvieri Lmx. Boat Cove (sparse); mid-tide pool, Fishing Rock. G Corallina officinalis L. Mid-tide pool and sublittoral fringe, Fishing Rock; sublittoral, 4-8 m., Meyer I. Jania rubens Lmx. Mid-tide pool and sublittoral fringe, Fishing Rock. This appears different from the local /. mi- cr arthrodia. G * Amp hiro a anceps (Lmx.) Dene. GC * Cheilosporum elegans (Hook. f. & Harv.) Aresch. Schmitziella cladophorae Chapm. On Cladophora prolifera. Posliella farinosa. On Cladophora pro- lifera, Lava Pt. Gigartinales C # Plocamium cos tat um J. Ag. G C Plocamium brachiocarpum Kiitz. Plocamium angustum J. Ag. On Ptero- cladia, lower mid-littoral, Hutchinson’s Bluff. G * Gracilaria conferv oides (L.) Grev. Drift. Ceramiales G Martensia elegans Hering. Sublittoral fringe in Corallina turf, Boat Cove: mid-tide pool, Fishing Rocks: sublit- toral, 3 m„, Meyer I. Caloglossa leprieurii J. Ag. . G * Nitophyllum decumbens J. Ag. Drift. G * Euzoniella incisa ( J. Ag.) Falkbg. G * Spongoclonium brounianum (Hook. f. & Harv.) J. Ag. Microcladia novae-zelandiae J. Ag. Boat Cove, sparse. G * Lamencia forsteri (Mert.) Grev. Drift, f Ceramium codii ( Richards ) G. Mazo- yer. On C odium adhaerens. The plants were not fruiting but habit and meas- urements agreed. Onerahi Bluff. Centroceras clavulatum (C. Ag.) Mont. Epiphytic on Ptero cladia, Hutchin- son’s Bluff; sublittoral, 12 m., Meyer I. Ecology of Kermadec Islands — CHAPMAN 351 Total marine algae: Myxophyceae 12 Chlorophyceae 20 Phaeophyceae 10 Rhodophyceae 35 This is not an impressive list considering the variety of habitats available and the fact that the islands clearly lie in a zone where there is a mingling of cold and warm waters. BRYOPHYTA The mosses were kindly determined by the late Mr. G. O. K. Sainsbury and the liverworts by Mrs. Hodgson. Musci Fissidens pungens C. M. & Hpe. Fissidens oblongifolius H. f. & W. First record for the Kermadecs. Rhizogonium Plongiflorum (Mull.) Jaeg. Rhizogonium spinif or me (Hedw.) Bruch. "Differs from type in having male flowers not synoicous and the perichaetial bracts lengthened. It looks as if some people would consider the Kermadecs plant to be a form of R. longiflorum.” Isopterygium minutirameum (C. M.) Jaeg. Leucobryum candidum (Brid.) H. f. & W. Echinodium hispidum (H. f. & W.) Jaeg. Oxyrrhynchium compr es sif olium (Mitt.) Broth. "This is quite different from other allied forms (Eurhynchium) in New Zea- land.” Papillaria flaxicaulis (Taylor) Jaeg. A canth o cladium extenuatum (Brid.) Mitt. "Another first record” for the Kermadecs. Rhacopilum Ppacificum Hepaticae Phaeoceros sp. Chiloscyphus argutus Nees. Lophocolea heterophylloides Nees. Plagiochila sp. Asterella sp. Marchantia sp. Radula sp. REFERENCES Chapman, V. J. 1956. The marine algae of New Zealand, Part I. Myxophyceae and Chlorophyceae. Jour. Linn. Soc. Bot. 55(360): 333. In press. The marine algae of New Zea- land, Part II. Phaeophyceae. Nova Hedwigia. Cheeseman, T. F. 1888. On the flora of the Kermadec Islands. Trans. N. Z. Instit. 42:118. Cotton, A. D. 1912. Marine algae from the north of New Zealand and the Kermadecs. Kew Bull. Misc. Inf. 9: 256. Gepp, A., and E. S. Gepp. 1911. Marine algae from the Kermadecs. Jour. Bot. Lond., p. 17. Lindauer, V. W. 1957. A descriptive review of the Phaeophyceae of New Zealand. Trans. Roy. Soc. N. Z. 85(1): 61. Oliver, W. R. B. 1910. The vegetation of the Kermadec Islands. Trans. N. Z. Instit. 42: 118. Feasibility of a Lava-diverting Barrier at Hilo, Hawaii 1 C K. Wentworth, H. A. Powers, and J. P. Eaton 2 The SUBJECT of the value and possibility of protecting Hilo Harbor and vicinity from de- vastation by a lava flow from Mauna Loa is again being given thoughtful consideration by the residents of Hawaii. Those who must weigh the pros and cons of this matter need informa- tion, part of which can best be appraised by geologists and engineers. From the geologists’ appraisal should come answers to questions such as the following: How often might protection from a lava flow be needed? Is it physically possible to divert a lava flow with a man-made structure? What are the necessary dimensions of such a structure? Of what should it be built? What is its expected useful life? Various references to cost have been made. Some say that a barrier is justified, regardless of cost; others hold to a strict accounting of sup- posed risk against cost, amortization, and other factors. These opposed views are widely sep- arated. Many risks could be reduced by astro- nomical spending, but such spending may be beyond reasonable relation to contemporary life or even to capacity of the community to pay. Though opinions may differ greatly, the crite- rion of economic justification cannot be ignored altogether. Much has been written on the subject of a lava barrier for Hilo. The latest and most com- prehensive review and discussion is by Gordon A. Macdonald ( 1958). His greatest emphasis is laid on the matter of a barrier system to be con- structed across the slope above Hilo to divert the course of an approaching lava flow. He con- cludes that a system of barriers can divert the course of a lava flow. The conclusions reached in this report differ in this matter from those expressed by Mac- donald because different evaluations are made of the same few facts available for appraisal. 1 Publication authorized by the Director, U. S. Geo- logical Survey. Manuscript received October 24, I960. 2 U. S. Geological Survey, Hawaiian Volcano Ob- servatory, Hawaii National Park, Hawaii. Among the most important of these different evaluations, this report concludes that the mini- mum condition for the successful functioning of a diversionary system is the construction of a channelway adequate to conduct the lava flow along the chosen route behind the barrier sys- tem. An adequate channel may exceed 2 mi. in width with rock excavation in excess of a 400-ft. depth along the upslope margin, even with a barrier 60 ft. high along the downslope margin. Facts needed to design the channel system and to appraise the amount of funds that can pru- dently be invested in it are imponderable — facts such as the volume of flow to be expected and the probable frequency of hazard. In the face of such imponderables, a downslope diversionary system is unrealistic; it would seem prudent to rely on, and plan for, defensive actions that can be taken during an eruption, such as causing distributary flows at or near the vent. FORECASTING ACTIVITY The waxing and waning of volcanic activity shown in the geologic history of Mauna Loa makes it impossible to give a dependable predic- tion of the probable hazard to Hilo from lava flows. The possible hazards cover a great range: Hilo might be obliterated by another eruption from the same vicinity as the prehistoric erup- tions that formed the Halai Hills (see Fig. 1); or it is possible that no future lava flow will ever reach Hilo. Since Hawaiian oral history be- gan, perhaps about A.D. 1100, only one lava flow from Mauna Loa, that of 1881, reached the vicinity of Hilo. It is natural to predict future events on the premise that events of the best-known past will be repeated; in this instance, the history of Mauna Loa’s activity since 1843. How disas- trously wrong such a prediction can be was em- phasized by the eruption in I960 of the Kilauea lava flow in Puna. After the devastating flow in 1955, no further outbreak in that region was to 352 Lava Barrier at Hilo — Wentworth, Powers, and Eaton 353 be expected for several decades, based on the spacing of recent known eruptions: 1740?, 1793?, 1840, and 1955 (Macdonald, 1941). Forecasts of Mauna Loa activity based on an even shorter period of time may be equally wrong; in fact, an inspection of a longer record shows that the activity of Mauna Loa waxes and wanes in a manner that gives no useful basis for predicting the frequency of future activity. In the 180 years since 1780 there have been 20 lava flows from the flanks of Mauna Loa; in the preceding 600 odd years covered by Hawaiian oral traditions there apparently were no lava flows from Mauna Loa; and, representing activ- ity previous to A.D. 1100, 60 different ancient cinder cones can be found that indicate flank eruptions that took place over an unknown span of time. There is no geological basis for pre- dicting how long the present epoch of frequent eruption may last; it may continue or it may have run its course. The evidence for the dormancy of Mauna Loa during about 600 years covered by Hawaiian oral history is considered here in some detail, as it has not had the attention in the literature that it deserves. It consists of the evaluation of observations by early explorers and of geologists and evaluation of Hawaiian oral history and mythology. Members of Captain Cook’s expedition in 1778-79, particularly John Ledyard who at- tempted to climb the mountain, noted that Mauna Loa was a volcano and described features on the slopes ”... that had every appearance of past eruption and fire. . . . But there is no tradi- tion among the inhabitants of any such circum- stance” (Hitchcock, 1909: 61-62). Archibald Menzies, the botanist on one of Vancouver’s expeditions, climbed to the sum- mit crater of Mauna Loa in February, 1794; he contrasts "the Mountain” Mauna Loa with "the Volcano” Kilauea in his descriptions (Hitch- cock, 1909: 68-72). William Ellis, a British missionary who knew the Polynesian language, explored Hawaii in 1823 and queried the Hawaiians about volcanic activity. They had no oral history of lava flows from Mauna Loa but reported that Kilauea had been active from "time immemorial” and that some part of the lands of Kau and Puna had been devastated by a lava flow during the reign of every King (Hitchcock, 1909: 163-164). The United States exploring expedition under Captain Wilkes spent nearly a month on the summit of Mauna Loa in the winter of 1840-41, having traversed the northeast ridge in the as- cent (at that time, only one of the known his- toric eruptions had broken out from this re- gion). They reported that the whole area was of lava, chiefly of very ancient date ( Hitchcock, 1909: 83). A large area of the southwest ridge of Mauna Loa was explored by R. H. Finch of the U.S. Geological Survey during December, 1925. He observed, "The lava on the southwest flank of Mauna Loa may well be divided into two ages: recent (within the last 100 to 150 years, say), and old. Lava flows of various ages showing a uniform gradation in weathering between the oldest and newest flows are not to be found” (Finch, 1925: 90). There is thus some geologic evidence for a considerable period of dormancy of Mauna Loa, implied by the lack of mention of Mauna Loa flows in Hawaiian oral history. Moreover, recent seismological evidence that Kilauea’s lava rises from a zone about 60 km. beneath the region between the Kilauea caldera and Mauna Loa’s northeast rift zone raises the possibility that both volcanoes are fed from the same source and that when one is in a period of unusual ac- tivity the other erupts infrequently. Such alter- nation in activity between the two volcanoes over century-long intervals is suggested by the historic evidence quoted above. However, the Hawaiians were well aware of the fact that Mauna Loa was a volcano; many of their myths describing the activities of the demigods were explanations of volcanic features they found on the slopes of Mauna Loa. Pre- Hawaiian lava flows on the southwest slope are explained in the legend of "Na Pu'u o Pele” (Westervelt, 1916: 22-26); the lava flows that bank against the north slope of Mauna Kea were, to the Hawaiians, evidence of legendary conflicts between Pele and the snow-goddesses (Westervelt, 1916: 62); and the most recent lava flow in the forest south of Hilo was, to them, a record of the battle between Hi'iaka and Pana-'Ewa (Westervelt, 1916: 96-103). In 354 PACIFIC SCIENCE, VoL XV, July 1961 contrast, the lava island in Hilo Bay called Co- conut Island was fished up from the sea by the demigod Maui (Westervelt, 1916: 28), appar- ently not associated in Hawaiian minds with the demigods to whom they attributed volcanic phenomena. DIVERSION SYSTEM TECHNICALLY POSSIBLE All who have considered the problem have agreed that a solution to containing a lava flow does not lie in impounding lava behind a dam; the topography is not favorable and the total amount of lava that would need to be stored cannot be estimated. The solution is sought, therefore, in some manner of diverting the course of flow. A lava flow following a natural channel can be entirely diverted along a chosen alternate channel if one fundamental condition is met— -the artificial channel must be able to carry the lava away from the point of intercep- tion as rapidly as it is delivered there by the natural flow. The average gradient of such a diversion channel will, of necessity, be considerably less than the average gradient of the natural slope across which it is constructed. To offset the un- favorable loss of gradient, the built channel must offer less obstruction in its floor, such as irregularities and vegetation, and provide space for a greater cross-section of flow. It is not enough to consider that a cross-section of a diversion channel is adequate by allowing an added area to compensate for the reduced gradi- ent on the basis only of gravity flow of a liquid. Allowance must be made also for the capacity of the lava flow to transmit enough heat to maintain its liquidity. This differential term in the equation works against a wide flow, even though its greater width might sufficiently com- pensate reduced depth for water. If these condi- tions are met, the channel will direct the move- ment of the flow, and the barrier need only confine the downslope margin, not act as a dam across the flow. However, if lava behind the barrier is ponded to a considerable depth (50 ft. or so), the pos- sibility that it might inject itself through the barrier or its foundation cannot be overlooked. Such an engineering accident was responsible for the early failure of a barrier constructed dur- ing the I960 eruption of Kilauea. DESIGN OF THE DIVERTING SYSTEM The designer of a system of structures to di- vert flowing lava must know the probable maxi- mum rate of delivery of lava that can be ex- pected to enter the system. Here again, geologic experience cannot predict the probable require- ments, it can only point out the possible maxi- mum load. If the outbreak takes place within 10 or 15 mi., lava may be expected to enter the system at a rate of about 25,000,000 cu. yd/hr, based on the observations made on the Mauna Loa eruption of 1950, the most voluminous eruption that has been sufficiently documented (Finch and Macdonald, 1953). Should the de- signer anticipate the voluminous load from a nearby eruption? What are the data upon which to make the decision? The pre-Hawaiian lava flows that form the south shore of Hilo Bay (see Fig. 1) appear to have come from vents along the lower part of the northeast rift, according to current studies of recent air photographs and some reconnais- sance field identifications. The topographic ridge built by these and similar eruptions is the south boundary of the topographic trough that slopes into Hilo Bay. Any future eruption along this rift line below an altitude of about 3,500 ft. will lie on the south side of the ridge, and its lava flows thus would be directed away from Hilo Bay; an eruption along this zone above about 3,500' ft. will be more than 15 mi. from Hilo. Any source vent closer to Hilo than 15 mi. would have to break through the flank of Mauna Loa considerably to the north of the zone of old cinder cones that mark the lower part of the northeast rift. However, Stearns and Macdonald (1946: 70) reasoned that the vents in Hilo (Halai Hills) lie on a branch of the northeast rift, and Macdonald restated the supposition in 1958 (p. 259). An eruption on any part of this supposed branch of the rift zone will be in the trough leading to Hilo; such an eruption must be expected geologically, even though there are no existing vents along this line between Hilo (Halai Hills) and a point 22 mi. from Hilo at Lava Barrier at Hilo — WENTWORTH, POWERS, and EATON 355 Fig. 1 . Map of the vicinity of Hilo, Hawaii, showing historic and recent prehistoric lava flows from Mauna Loa that approached Hilo Bay. Shaded area represents present extent of the city of Hilo. an altitude of 6,800 ft. Even if it is assumed that the reasoning of Stearns and Macdonald is incorrect (and there is no compelling geologi- cal basis for such an assumption), and that the supposed branch of the rift zone does not exist, there remains strong geologic precedent for an outbreak through any flank area away from a known rift zone. Of the 72 known flank erup- tions, 18, or one-fourth, have broken through the mountain flank several miles away from any known rift zone. The eruption of 1877, above and in Kealakekua Bay (Hitchcock, 1909: 115), broke out as far from a rift zone as it is possible to be. There has been no eruption, in 356 PACIFIC SCIENCE, Vol. XV, July 1961 the past, within 15 mi. upslope from the prob- able site of a diversion structure, but there is no known geologic reason why an eruption may not break through in this area. A reasoned decision about the necessary barrier design cannot be made on such data; the decision must be based on other considerations. Another, and completely unrelated, problem of design for which no geologic or engineering solution is possible rises from certain charac- teristics of a lava flow (Wentworth, 1954). Every flow of lava inevitably sends out distribu- tary flows from time to time and from place to place along its course, as one way of responding to frequent large fluctuations in the amount and rate of eruption of lava at the source vent. Therefore, it may be expected that more than one flow of lava will enter the channel of the diversion system during any one eruption. Inas- much as mobile lava becomes immobile rock as soon as it cools slightly, a considerable amount of any lava that enters the channel system will solidify there and form an obstruction in the channel. Thus, any subsequent flow of lava that enters the channel system at a point upgrade will have to override this obstruction in order to keep on moving downgrade. If the channel system has been built with enough capacity, the overriding flow will be contained and the sys- tem will continue to function; if the system has too small a capacity at this point, the barrier wall of the channel will be overrun at the ob- stacle and the diversion system will fail to func- tion. At the designing stage of an adequate diver- sion system, it is obviously impossible to antici- pate the point at which a future first lava flow will enter the system, to estimate the magnitude of the obstruction that it will form, or to ap- praise the amount of lava that may have to pass over the obstruction. The designer can cope with this situation only by overdesigning the entire system. He can only guess how much to overdesign : — twofold ? — tenfold ? In considering design of barriers and diver- sion channels, the tendency of liquids adjacent to a dam to cause uplift pressure and to burrow through should be realized. To allow for such tendency is standard practice in designing dams, because some have failed in this way. Lava bar- riers have also failed in this way, as was recently observed in some instances at Kapoho. How- ever, in the case of a massive barrier built of well-compacted rock and soil, this is thought to be a very remote contingency because of the cooling effect. Lava might retain liquidity through tenuous openings for a distance of 200 or 300 ft. but would seem unlikely to do so through 1,000 ft. or more except in a pre-estab- lished tube. Such an accident is not entirely dis- missable, however. SAMPLE ESTIMATES OF DIVERSION CHANNEL DIMENSIONS We can neglect for the moment the impon- derable matter of overdesign and consider the dimensions required to convey two sample lava flows that may be assumed to move as simple flow units. The average natural gradient of the trough that leads to Hilo, which must be intercepted by the diversion system, is between 250 and 300 ft/mi. The diversion channel probably could be laid out with an average gradient of no more than 200 ft/mi. Estimates of the ve- locity of movement of lava flows on comparable low gradient can be made from published de- scriptions of previous flows. The hot, mobile lava near the vent of the 1954 eruption of Ki- lauea (Macdonald and Eaton, 1954) moved at rates not less than 400 yd/hr. A channel de- signed to move 25,000,000 cu. yd. of hot, mobile lava at this velocity would need to provide space for a flow cross-section of 63,000 sq. yd. If a containing barrier on the downslope margin of the channel were built high enough to give an average depth of flow of 20 yd. in the channel, the width of the channel would be 3,150 yd. (approaching 2 mi. wide), and the maximum rock excavation at the upslope margin would be greater than 400 ft. A different example: the relatively cool and viscous lava of the 1926 flow that destroyed the beach village of Hoopuloa (Hawaiian Volcano Observatory, 1926) moved at rates not less than 60 yd/hr. A similar relatively cool flow from a distant vent reaching the diversion system at a rate of 2,000,000 cu. yd/hr would require a channel cross-section of nearly 34,000 sq. yd. to Lava Barrier at Hilo— WENTWORTH, POWERS, and EATON 357 carry the load at 60 yd/hr velocity. Assuming an average depth of 20 yd., the width of channel required is about 1,700 yd. ( 1 mi.) and the up- slope would exceed 200 ft. These ' examples have neglected the overde- sign necessary to accommodate the transporta- tion of distributary flows. CHANGING THE MOVEMENT PATTERN OF LAVA FLOWS BY BOMBING It has long been understood by observers of Hawaiian lava flows that the course and progress of a flow can be radically altered by breaching the levee bank of the main feeding channel. Macdonald (1958) presents an excellent discus- sion and evaluation of the matter which need not be repeated here. He concludes that efforts to divert the flows by bombing should be made in the event of a threat to Hilo, but that a bar- rier system also should be constructed as insur- ance against failure of the bombing effort, par- ticularly in the event that a voluminous, fast moving flow would overrun the area before bombing could be carried out. However, it would seem from the discussion in previous par- agraphs that an artificial diversion system of dimensions adequate to cope with a voluminous, fast moving flow would be expensive beyond prudent economic justification. Thus, it would seem that the hazard of being overrun by lava is one that must be accepted and lived with, perhaps analogous to the acceptance of earth- quake harzards by Tokyo and cities in other earthquake areas. REFERENCES Finch, R. H. 1925. Expedition to the southwest rift of Mauna Loa. Hawaii. Vole. Obs. Bull. 13: 89-9E Finch, R. H., and G. A. Macdonald. 1953. Hawaiian volcanoes during 1950: The 1950 eruption of Mauna Loa. U. S. Geol. Surv. Bull. 99'6-B: 49-81. Hawaiian Volcano Observa- tory. 1926. Journal of Mauna Loa eruption. Monthly Bull. 14 (14): 31-50. Hitchcock, C. H. 1909. Hawaii and Its Vol- canoes. The Hawaiian Gazette Co., Ltd., Ho- nolulu. 306 pp. Macdonald, G. A. 1941. Lava flows in eastern Puna. Volcano Lett. no. 474: 1-3. 1958. Barriers to protect Hilo from lava flows. Pacif. Sci. 12(3): 258-277. Macdonald, G. A., and J. P. Eaton. 1954. The eruption of Kilauea volcano in May 1954. Volcano Lett. no. 524: 1-9. Stearns, H. T., and G. A. Macdonald. 1946. Geology and ground-water resources of the island of Hawaii. Hawaii Div. Hydrogr. Bull. 9, 363 pp. Wentworth, C. K. 1954. The physical be- havior of basaltic lava flows. J. Geol 62: 425-438. Wester velt, W. D. 1916. Hawaiian Legends of Volcanoes. Ellis Press, Boston. 205 pp. Variations in the Lava of the 1959 Eruption in Kilauea Iki 1 Gordon A. Macdonald and Takashi Katsura 2 In RECENT years the Hawaiian Islands have become of great importance in studies concerned with igneous petrogenic theory. This is partly because of the total absence in that region of sialic crust that might affect the formation of other rock types by its assimilation in rising magma. Partly, also, it is because of the exten- sive lateral and vertical exposure of successions of rocks, the structural and stratigraphic rela- tionships of which are now well known. These conditions have attracted to the area workers from many parts of the world, and progress in knowledge of Hawaiian rocks has been rapid. The knowledge is, of course, the accumulation of the findings of many workers, starting with the visit of J. D. Dana to the islands in 1840, as a member of the U. S. Exploring Expedition. Outstanding among recent workers are Hisashi Kuno of Tokyo University, P. Niggli of the University of Zurich, C. E.' Tilley of Cambridge University, H. S. Yoder of the Geophysical Lab- oratory of the Carnegie Institution of Washing- ton, and H. A. Powers of the United States Geological Survey. During I960 and 1961 two studies have largely confirmed, but also extended and some- what modified, the previous petrogenic picture. In April, I960, the junior author of this paper, on leave from the Tokyo Institute of Technol- ogy, commenced a series of chemical analyses of Hawaiian lavas at the University of Hawaii under a National Science Foundation grant to the senior author. At the end of March, 1961, approximately 150 new analyses, primarily of the hitherto largely neglected "primitive” lavas of the Hawaiian volcanoes, had been completed. These have helped fill important gaps in the basic knowledge of Hawaiian rocks. Study of them is continuing. During July, I960, a core hole was drilled in the crust of the recently erupted lava pool in Kilauea Iki crater of Kilauea volcano, and pene- 1 Hawaii Institute of Geophysics, Contribution 23. Manuscript received March 21, 1961. 2 University of Hawaii. trated into the molten lava beneath. The results of study of the drill-hole samples and samples of the earlier lava of the eruption are reported in this paper. Acknowledgments. The core hole in the crust of the Kilauea Iki lava pool was drilled for the Lawrence Radiation Laboratory of the Univer- sity of California in cooperation with the Uni- versity of Hawaii, and the chemical analyses of the core samples were done at the University of Hawaii for the Lawrence Radiation Labora- tory. The entire program is a part of the Law- rence Radiation Laboratory’s Plowshare Program for the development of peaceful uses of atomic energy. The results of other aspects of the in- vestigation will be published by members of the staff of the Lawrence Radiation Laboratory. The drilling was done by a crew in the em- ploy of Nat Whiton of Honolulu, under the general supervision of Walter Bennett and Donald E. Rawson of the Lawrence Laboratory. General scientific supervision was furnished by Macdonald. Thanks are due to the National Park Service for permitting the drilling for scientific purposes within Hawaii National Park, and to the Lawrence Radiation Laboratory for permitting us to publish the chemical analyses and other data on the core samples. The sample of Pele’s hair analyzed was col- lected and given to us by Mr. H. Ikawa, of the Department of Agronomy, University of Hawaii. Two analyses (S-l and S-2) in Table 1 are by J. H. Scoon of Cambridge University. We wish to express our thanks to A. T. Ab- bott, of the University of Hawaii, for critical reading of the manuscript. HAWAIIAN ROCK SUITES In the classical Mull Memoir (1924), E. B. Bailey and his associates distinguished three principal rock types, which they termed the plat- eau, central porphyritic, and central nonporphy- ritic types. In 1933, W. Q. Kennedy applied the names olivine basalt and tholeiite to the first and 358 Kilauea Iki, 1959 — Macdonald and Katsura last types, respectively. In 1935, H. A. Powers recognized the presence of both these types in the Hawaiian Islands, though he did not em- ploy the same terminology. Later Macdonald (1949^: 88) also pointed out that, although it had been thought by Kennedy to be absent from the oceanic areas, tholeiite is present in Hawaii. In 1955, Powers emphasized the essentially silica-saturated nature of the lavas of Mauna Loa and Kilauea. Although it had been antici- pated to some extent by Powers (1935), C. E. Tilley (1950) was the first to definitely point out the presence of two distinct rock series in Hawaii. These he termed the tholeiitic and al- kali olivine basalt series. Tholeiite has been defined (Tilley, 1950; Kuno et al, 1957) as a rock essentially satu- rated or slightly oversaturated with silica in which magnesian olivine bears reaction relation- ship to orthopyroxene and Ca-poor clinopyrox- ene. In contrast, alkali olivine basalt was defined as an undersaturated rock in which magnesian olivine and Ca-rich clinopyroxene undergo par- allel crystallization. The characteristic ground- mass pyroxene of tholeiite is pigeonitic, though the wollastonite content ranges from about 40 per cent to less than 10 per cent. That of alkali olivine basalt is Ca-rich augite. Hypersthene is present in some tholeiites. Recently Kuno (I960) has pointed out the wide distribution of basalts resembling the cen- tral porphyritic type of Mull in their richness in alumina, but differing from it in being essen- tially nonporphyritic. High-alumina basalt of this type has not been found in the Hawaiian Islands, though some rocks containing abundant phenocrysts of feldspar are moderately high in alumina. The predominant lavas of the great bulk of the visible part of the Hawaiian shield volcanoes contain scattered to moderately abundant phe- nocrysts of olivine, commonly as much as 5 mm. in diameter. These rocks have been called "oli- vine basalt” by Macdonald (1949^, b) . Typi- cally, however, the olivine phenocrysts are partly resorbed, and obviously were reacting with the remaining liquid at the time of consolidation of the rock. Chemical analyses show many of these rocks to be essentially saturated in silica, and the pyroxenes are lime-poor. They are thus typical 359 tholeiites. They grade into other rocks that are otherwise similar but contain normative olivine. The extreme of the latter group is picrite-basalt of oceanite type, which may contain more than 50 per cent olivine phenocrysts. These rocks are chemically undersaturated with silica, and even with the attainment of complete equilibrium on crystallization should contain modal olivine. As in the more silica-saturated types, however, phenocrystic olivine commonly has separated in excess of its stoichiometric proportion, and was undergoing reaction with the liquid at the time of consolidation, as is indicated by their partial resorption. The groundmass pyroxene is largely or entirely lime-poor augite and pigeonite. Thus, although they are undersaturated with silica and therefore not true tholeiites, these rocks conform with Tilley’s definition of tholeiite as a rock in which magnesian olivine bears a reaction rela- tionship to Ca-poor pyroxene, and the entire group from the nonporphyritic true tholeiites to the picrite-basalts of oceanite type may be termed the tholeiitic suite. Both Powers (1955) and Macdonald ( 1944) have pointed out that the variations within the tholeiitic suite can be largely accounted for by settling of olivine phenocrysts in the magma, though movement of minor amounts of pyrox- ene and plagioclase also probably are involved (Muir and Tilley, 1957; Macdonald, 1949 b: 1576). Locally differentiation has yielded small amounts of iron -rich basalt and granophyre (Kuno et al., 1957). The major tholeiitic part of the Hawaiian shields is succeeded by a relatively small amount of lavas of other types, including alkali olivine basalt, picrite-basalt of ankaramite type, hawai- ite (Macdonald, I960), mugearite, and trachyte. This group may be called the alkalic suite, be- cause of its relationship to alkali olivine basalt, and because the great majority of the members of the group contain a larger proportion of al- kalies than do members of the tholeiitic suite that contain the same amount of silica. The lavas of the alkalic suite constitute only a few per cent of the total bulk of the Hawaiian volcanic mountains. The nature of the transition from the tholeiitic to the alkalic suite will be dis- cussed in detail in a future report describing the results of the recent chemical analyses. 360 PACIFIC SCIENCE, Vol. XV, July 1961 A still later group of lavas, which may be called the nephelinic suite, consists characteris- tically of nepheline basalt and melilite-nepheline basalt, but includes also basanites, and alkali oli- vine basalts ("linosaites”) in which nepheline is present in the norm though not in the mode. The nephelinic suite is in general separated from the rocks of the other groups by a pro- found erosional unconformity (Stearns, 1946: 22 ). THEORIES OF ORIGIN OF HAWAIIAN ROCK SUITES There appears to be little or no question that the tholeiitic magma originates in the upper part of the earth’s mantle, probably at a depth of about 30 or 40 mi., and that variations within the suite are largely or entirely the result of crys- tal differentiation. The same degree of certainty does not extend to the alkalic suite. Macdonald has previously attributed the formation of the more alkalic members of the alkalic suite to dif- ferentiation of a parent magma corresponding approximately in composition to the average basalt of Kilauea (Macdonald, 1949^: 92; Stearns and Macdonald, 1946: 205), or to an average of basalts from all Hawaiian volcanoes (Macdonald, 19497?: 1569) . The former average corresponds with tholeiite only slightly under- saturated with silica. The latter average included alkali olivine basalts, and therefore is a little higher in alkalies and less saturated with silica than the Kilauean average. Calculations indi- cated that the alkalic rocks could be derived from either parent by crystal differentiation. To do so, however, it was necessary to hypothesize the separation of a large amount of pyroxene (both diopside and hypersthene) from the magma. Derivation of alkali olivine basalt per se was not considered, because it was not recog- nized as an independent rock type. The possibil- ity of other differentiation processes, such as al- kali transfer by volatiles, was also recognized. Tilley (1950: 44-45) also attributed the var- ious members of the alkalic rock suite to crystal differentiation of tholeiite, separation of hypers- thene in place of olivine producing the alkali olivine basalt. Powers (1955) agreed to the im- portance of the movement of olivine crystals in producing the variations among the tholeiitic basalts, but pointed out, as indeed he had earlier (Powers, 1935), that crystal differentiation alone is inadequate to produce alkali olivine basalt from a saturated tholeiite. It certainly is true that desilication of a magma by crystal differentiation can only result from the removal of crystals containing more silica than the magma. The removal of pyroxene from tholeiite can perpetuate a state of under- saturation in silica, but cannot bring it about. No mineral containing more silica than a satu- rated tholeiite magma is likely to form and sep- arate except during the very latest stages of crystallization. Provided, however, that a dis- tinctly undersaturated tholeiitic magma exists as a liquid, crystallization of pyroxene can not only perpetuate the undersaturation, but increase it. Murata (I960) has suggested that alkali olivine basalt is derived in this way from undersatu- rated tholeiitic magma. Kuno et al. (1957: 212) agree with Powers that crystal differentiation cannot produce alkali olivine basalt from tholeiite. Instead, they as- sume the existence of two independent primary basalt magmas, tholeiite and alkali olivine basalt, formed by partial melting of peridotite at differ- ent levels in the mantle, incongruent melting of pyroxene at the higher level supplying extra silica for the tholeiite. More recently, Kuno (I960) has hypothesized still a third primary magma, high-alumina basalt, produced by melt- ing in the mantle at a depth intermediate to the other two levels. For several reasons, the existence of two pri- mary basalt magmas in Hawaii appears improb- able. Chemical analyses demonstrate a complete intergradation of the two types in composition ( Figs. 1 and 3 ) . Furthermore, it seems unlikely that melting at a deeper level in the mantle would produce a magma richer in alkalies than would melting at a shallower level. If anything, the reverse would be expected. These arguments are far from conclusive but, particularly when it is remembered that the rocks of the alkalic suite comprise only a very small proportion of the total, they do suggest that in some way alkali basalt magma is produced from tholeiitic magma, rather than having a wholly independent origin. Kilauea Iki, 1959— Macdonald and Katsura KILAUEA IKI LAVA The pool of lava in Kilauea Iki crater, a pit crater immediately adjacent to the east edge of Kilauea caldera, was accumulated during the eruption that lasted from November 14 to De- cember 20, ,1959. The eruption consisted of 16 separate eruptive phases, from 2 to 167 hr. in length, separated by quiet periods of 8-101 hr. duration during which part of the lava flowed back into the vent from which It had issued (Macdonald, in press). Lava from fissures on the southwest wall of the crater about 300 ft. above the pre-emption floor poured down the crater wall and formed a pool that gradually deepened until its surface was above the level of the original vents. Temperature measurements up to nearly 1200° C. were obtained on the lava fountain at the vent (Richter and Eaton, I960). During eruptive phases lava was added to the crater fill partly by injection beneath the crust and partly by spreading over the previous crust. A new crust formed quickly on the molten lava as each successive outpouring covered the crater floor, but from time to time broke up and foundered during convective overturns in the liquid. The latter suggests that submerged crusts of previous eruptive phases had been largely destroyed (possibly accumulating as sunken fragments at the bottom ) , leaving a pool of melt that was essentially continuous from bot- tom to top. The final depth of the pool is ap- proximately 380 ft. In July, I960, 7 months after the end of the eruption, the drill hole already mentioned was sunk into the solid crust of the lava in Kilauea Iki crater. Continuous core samples were taken, and cutting samples were preserved from succes- sive drill runs to supplement the cores in inter- vals in which core recovery was poor. On July 25 the drill bit encountered very viscous semi- solid material at a depth of 19.1 ft., and at 19.5 ft. it entered underlying liquid lava and started to sink into it. The tools were immediately re- moved from the hole, and a sample of the liquid that had congealed in it was removed from the bit. Additional samples were taken the next day by thrusting the drill pipe, without a bit, down into the liquid and withdrawing it. By the morn- ing of July 28 liquid lava had risen in the hole to a level 18.7 ft. below the surface. 361 Eight samples from the drill hole, and one of the surface crust adjacent to it, have been analyzed chemically. They are listed in Table 1. Three analyses of lava extruded earlier in the eruption also are given. Of these, one is of Pele’s hair wafted from the crater during the first days of the eruption. The other two (S-l and S-2), by J. H. Scoon of Cambridge University, repre- sent lava poured into the crater and spatter thrown onto the crater rim during the first erup- tive period, between November 14 and 21 . Analyses 9 and 10 are of samples of the liquid lava that underlay the solidified crust. It will be noted that the two are not identical. Sample 10 is decidedly richer in alkalies. Sample 9 contains many more phenocrysts of olivine than sample 10, and is presumed to have come from a slightly lower level in the liquid. Column 1 of Table 1 is an analysis of Pele’s hair formed at the beginning of the eruption. Scattered phenocrysts of olivine were present in the Pele’s hair, but none were included in the sample analyzed. The sample consisted essen- tially wholly of glass, even microlites being al- most entirely absent. Therefore, the analysis represents magma that was completely liquid at the time of eruption. The norm indicates a notable degree of undersaturation of the liquid in respect to silica. It has generally been assumed that normative olivine in tholeiitic rocks is largely the result of addition to the magma of solid crystals of olivine that have sunk from higher levels. In the case of the Pele’s hair of analysis 1 the undersaturation is not, however, the result of inclusion of olivine crystals in the analyzed sample. The liquid phase of the magma was itself undersaturated. Yet its very low con- tent of alkalies and alumina mark it clearly as a member of the tholeiitic suite. It is close to the value suggested by Bowen (1928: 164) as the limit of undersaturation in completely liquid basaltic magmas, though less undersaturated than some rocks believed by Drever ( 1956) to have been derived from completely liquid magmas. In thin section the samples from the upper 17 ft. of the drill hole differ from each other almost entirely in the abundance of olivine phenocrysts, which range from about 10 per cent in the sample from 16.5 ft. to 40 per cent in that from 7.5 ft. The latter is a picrite-basalt of oceanite TABLE 1 Chemical Analyses of Lavas of the 1959 Eruption in Kilauea Iki 362 PACIFIC SCIENCE, Vol. XV, July 1961 res CN © 00 NT NT © r_l © CO res 00 CO 00 CN m res res © VO VO rH H VO CN Nr vo o ICS © o res r^i rH H o : CN rH NP 00 o^ oo WO ITS A 00 r-H r-H O res Cs CN © © © CN © © d CN O^ H d d A H NP © Np r— < r-H r-H © i — i CN CN T 1 1—1 CN NT res NO NT © Nf CN © NP ICS ICN © CO xp © NP 00 ITS vo 04 O' r— 1 Cs nP m CN CN NT 1 CN ICS O © vo CN i—l l-H o^ res 00 CN © CN 00 q vq A 00 cs ^ Cs © © O ICS l/A © vo O' r-H 00 m ICS CN vo O © o *-* o © rH CN rH 00 O d N N O © © D © © d CN res '-i NT CN r ICS res Cs © Cs NT CN © O Cn NT CN vo i-H NP rH res ON NP NP H I/O i/O O VD © np r-H CN VD 00 CN © © © m © © Cn vd © VO VO 04 res oo d e^4 res OI WO H np ON CN CN c/A O CN ICS ON res © © Nf v/A 00 CN 00 Cn NP CN Cs vo O VO © Cs O' irs vo 00 O' Cs np O rCN ICS res CN r-H CN rH H © CN 1—J CN 04 © c/A res Cs vq o © CN res r-H r-H © ics Cn r-H © © © CN © d © d rn CN © (N T-i ro NP vd A NP © np r-H rH rH © rH CN i — i rH rH 1—1 CO | s res 00 ics vo i — i © ,_i CO 00 res ICN ICN 00 o CN © vo Cn © CN © O vo vp © WO NP VO h© ICS °0 CN NT NT res res 1—1 CN © CN rH o! % CN O CN 00 l/A O^ O'- res 00 H o^ C\ res ICS © r-H dodoo H © © © CN © © Cs CN r-H d r> l/A rH d A d A A A d hp Cs CN 0^4 vo CN Cn 00 CN 00 © r- © vo © O' © O' CN © 04 © NP np vq res O' VO NP NP vq O res Nf CN res CN CN r^ rH rH Cs vq © © H O' o r WO VO 00 res res ICS A Cs r-H © i-h r-~ rH © © © r-H d © CN 1—1 H 00 A A rH CN CN vd A H A d np H CN CN CN 00 ICS vo vo © O CN © ^ res Cn 00 res O' res 00 © CN © O' ITS vo NO NP NP NP ICS CN CNJ NSJ H H CN r-H © rH H 1 — 4 VO NP rH vo O' res O 1^ res NP °q res res NP Cs H © A r-L r-H © © © H d d © rH CN Cs VO NP H © CN A A rH A © np i— i CN © rH rH 04 1—1 cn nt ICS Cs res © © 00 VO Nt ICN 00 CN CN 00 © res O NP © ON Np CO lH VO Np Cs H ICS h res Cs i r\ NT CN © CN CN I— 1 © CN ITS res vo vo A 00 xp VO 00 O' CN res CCS VO (N CN CN Nt 3 Cn H © © © CN © © © 04 CN ITS ON V0 oi rCS Np © res res NP © np rH © CN o- np CN its irs Cn res CO CN "ft 00 NP CO vo CN © NP VO o 00 © vo Xp H uo rH NP o> © Cs oo oo i r\ res r-H © CN CN H © CN H O' 00 O' res CN ITS H O' wo xp res CN VO CN rL ON NT CN r-H © O © 04 © d © CN res NP Cs vo CN © res Np i/O A np d np © CN 04 CN © © © CN ICS 00 CN CN e- np 00 CN VP VO NP © o CN © CN ON wo wo CN O' 00 'sf CN © res CN ICS © © O' CN 1— 1 CN res CO O' res l/A © res © Cs wq wo res vq 00 rO o Cs Cs L CN © © © cn © d © res 00 NP 04 O' Np Cn A A A A A d 'Sf r-H r— 1 © rH CN H i— i « CO rs ^ o + 1 e-s r wo ,en t CO O ctf V+H MH o V, v,< Q tj) Q H oqooo VJ a sS A ■ A 1 H IH M O i?3 an -3 hy o a il ap On N< N< Cs N* VO o cs so a hj A A A HH 0 i/ o Katsura i and 21 and 21 , •s-i c o H ^ S _r _Q u i §6 =!§| _ C - £ & OJ -H 15 2 ol (j ° ni c c - y rt rt S p „ „ a,j 3 aei ^ .2 .2 £ ^ oo 2 u ¥h'°’ ^ . .! H gg.*s j3 "S.4! i a > S4J2tS-S' a og g.^'^^ t ^^- ( cs(NN2W^ g rG «es vo • • w HH 5 ®^. G h4 §2 “ r- r- £ re • O "o •5 -g jg -S u o & &&.5 a a, o-S-S- d ITTT aTT „ A "G 03 03 O 13 -a •5 o s 6 S -« w 22Sti2^ <« U H V*H M_ G V*hUh “«J >2 W JU ^ O JO XI 4-. eS 2 GS&|o 2 2 g GGJ 3 1 S 6 * % o, J3 -g 2 2 _ _ v '> d .ET.ET — © cL I— i Li O O c/5 C/S N* l/S SO I'' Kilauea lid. 1959'— MACDONALD and KATSURA 363 type. It is believed to be the olivine-enriched lower part of the last lava to spread across the crater floor. No phenocrysts other than olivine are present in any of the samples. The ground- mass is intergranular to intersertal, and consists of calcic plagioclase, averaging about An 65 , monoclinic pyroxene, iron ore, and locally inter- stitial glass. The pyroxene ranges from augite with an optic axial angle of at least 50° to pigeonite with an angle close to 0°. No silica mineral was observed. Glass increases gradually in samples below 15 ft., suggesting that inter- stitial fluid in the rocks was chilled at the time of sampling. Samples 9 and 10, the congealed liquid picked up in the end of the drill pipe, are much alike except for greater abundance of olivine pheno- crysts in the former. Both consist predominantly of glass. Around the edge a zone 1-2 mm. thick consists of about 85 per cent pale brown glass with scattered microlites of feldspar, pyroxene, and magnetite. This appears to have resulted from quick chilling of the liquid against the steel pipe. The rest of the core consists of black to deep brown glass, containing phenocrysts of olivine up to 1.5 mm. long, and small grains of plagioclase and pyroxene, locally clotted to- gether in glomeroporphyritic texture. The larger olivines have (— )2V = 80°=±i. They show moderate skeletal development, and little or no signs of resorption. The larger pyroxene grains are lime-poor augite, with ( + )2V = 50° ±. Some smaller grains have ( + )2V = 30° ±. Small grains of magnetite are present, and the dark color of the glass appears to result from very finely dispersed iron ore. The proportion of glass is variable, but averages 50-60 per cent. All but one of the Kilauea Iki samples are quite definitely tholeiitic. The single exception is sample 10, which is transitional from the tholeiitic to the alkalic rocks. In Figure 2, in which total alkali content is plotted against silica, sample 10 lies just within the alkalic FeO Fig. 1. Diagram of compositional variations in tho- leiitic rocks of the Hawaiian Islands. Large dots repre- sent the Kilauea Iki samples, the numbers correspond- ing with those in Table 1. Small dots represent other rocks. F represents the position of the iron-enriched segregation veinlet described by Kuno et al. (1957), and G indicates the position of the granophyre from Palolo Quarry, Honolulu, described by the same authors. basalt field. On the basis of very quick inspec- tion of the analyses, Dr. Kuno has suggested (oral communication, March 2, 1961) that sample 10 represents an iron- and alkali-enriched segregation similar to the veinlet from a flow in the wall of Kilauea caldera described by him and his associates (1957: 187). Indeed, in Figure 1 the point representing the iron-enriched segregation veinlet (F) lies directly on the trend of the Kilauea Iki rocks. However, the veinlet is distinctly poorer in alumina and richer in silica than Kilauea Iki sample 10, and lies well within the tholeiite field (Fig. 2). The similar- ity of Kilauea Iki sample 10 to typical alkalic basalts strongly suggests the possibility of de- rivation of the alkalic basalts in general from undersaturated tholeiitic magma. TABLE 1 ( Cont .) * Four careful determinations of each component in analysis 10 were made. The reproducibility (precision of measurement) is sufficient for the present purpose. Cu, Zn, V, Ba, Sr, Cr, Mo, W, and Pb were detected qualitatively by means of X-ray, but not in significant amounts, and cannot therefore be responsible for the low total in this analysis. The latter is the result of an appreciable amount of graphite in the analyzed sample, which was not soluble in mixed solutions of H2SO4 and HF or HNO3 and HF. The graphite unquestionably came from the powdered graphite used to lubricate the joints of the drilling tools. The amounts of F, Cl, and C are being determined at the Tokyo Institute of Technology. 364 PACIFIC SCIENCE, Vol. XV, July 1961 W cn S pq £ J H Pm Pm S O <3 H nJ c z < < C3 B h « Q W Z < S o WD z o H 00 O Pm z a o o H <-> p £ a * 6 § 5 ef « < o x C/5 w ►4 W Pm H K H Pm O § E K CO O O Pm Pm a a o o o u xf u 00 Xf (XV (XV VO 00 03 13 oo d h ri oi 0 Xf r-l r-M 03 13 Xf O O r-j ND p rn a oo m d oi rd < Xf cH r-l rn 'O VO h V3 \0 (N 00 (N 6 oi CM XT H H (XV C\ Xf O iA iA (N CO (N O N oi Xf <-l rH (N ^ ON IA rn (N 3 "i r-H CV r-J oi 03 T- | p WO Xf 04 n h d h oi u xr m WO r- Xf 00 vo w 3 vd Xf d 00 oi WO Pi a p VO xf wo xf .25 04 <3 C/5 d << wd xf d 00 oi wo o p wo Xf cv VO WO w >-3 Calc wd xf cv oi “ (XV Pm vo (XV a • p p-H 00 OJ (XV oq <3 CO c3 d < ud Xf G\ oi 1— J rH r™< p p Xf Xf o^ wo Xf w 3 ir\ Xf 4 00 oi r— 1 00 Pm a VO to 04 xf co o- CV CO <3 c /5 ctf c •< xf Xf G\ oi r-H t " H u wo \q o- G\ wo xf 03 cxv w 3 oi Xf Cv r-H oi XF Pm WO a p r=H cv oq wo 04 04 <3 c/5 1 oi Xf oi cv t-H oi u Xf ITN VO o wo xf 20 04 w nJ 3 oi Xf - cv oi lO Pm a < 13 00 p Cv oq xf m .24 wo a < vd Xf oi cv oi o' Q, d ^ d o o uo <3 U /Z tZ H Pm S « 3 o O 6^2 T3 y Punta Entrada, Isla Mag- dalena, Baja Calif.; northern Gulf of Calif, south to Isla Ildefonso Dictyota binghamiae J. Agardh Dawson, 1954; Dawson, Neushul, Wildman, I960, I960 a Coos Bay, Oreg., to central Baja Calif. Dictyota concrescens Taylor Dawson, 1954 Cabo San Lazaro, Isla Magdalena, Baja Calif. Dictyota crenulata J. Agardh Dawson, 1954, 1954c, 1959 Isla San Benedict©, Revillagigedo Arch.; Isla San Diego, Gulf of Calif., to Puerto Cule- bra, Costa Rica Dictyota dichotoma (Hudson) Lamouroux Dawson, 1954, 1957, 1959 Gulf of Calif.; tropical Pacific Mexico and Central America ?; Galapagos Arch. Dictyota divaricata Lamouroux Dawson, 1954, 1957, 1957*, 1959 Isla Guadalupe, Baja Calif.; Revillagigedo Arch.; La Paz, Gulf of Calif., to Isla La Plata, Ecuador Dictyota flabellata ( Collins ) Setchell & Gardner Smith, 1944; Taylor, 1945; Doty, 1947; Daw- son, 1954, 1959, 1959c; Dawson, Neushul & Wildman, I960 Southern Calif, to Punta Santa Rosalia, Baja Calif.; northern Gulf of Calif, south to Isla Tortuga Dictyota friabilis Setchell Dawson, 1957*; Dawson, Neushul & Wild- man, 1960^ Islas San Benito, Baja Calif.; Golfo de Nicoya, Costa Rica ? Dictyota major Taylor Taylor, 1945 Galapagos Arch. Dictyota masonii Setchell & Gardner Dawson, 1954, I960* Isla Margarita, Baja Calif. ?; Isla Clarion, Revillagigedo Arch., Isla del Cano, Costa Rica ? Dilophus okamurai Dawson Dawson, 1954 Vicinity of San Lucas, Baja Calif. Dilophus pinnatus Dawson Dawson, 1954 Miramar, Nayarit Dictyopteris cokeri (Howe) Taylor Taylor, 1945 Galapagos Arch. Dictyopteris delicatula Lamouroux Dawson, 1954 Revillagigedo Arch.; Clipperton I. Dictyopteris diaphana Taylor Taylor, 1945 Galapagos Arch. Dictyopteris johnstonei Gardner Silva, 1957 (indicating probable identity with D. zonarioides) Santa Cruz I., Calif., to Islas San Benitos, Baja Calif. Dictyopteris membranacea (Stackhouse) Batters Dawson, 1954 Isla Guadalupe; Scammons Lagoon, Baja Calif. Dictyopteris repens ( Okamura ) Bprgesen Dawson, 1957, 1957*, 1959, 1959* Isla Carmen, Gulf of Calif.; Golfo de Nicoya, Costa Rica; Clipperton I. Dictyopteris zonarioides Farlow Dawson, 1954, 1959', 1959c; Dawson, Neu- shul & Wildman, I960 Southern Calif, to San Jose del Cabo, Baja Calif.; northern Gulf of Calif, south to Isla Ildefonso Benthic Algae— -Dawson 389 Taonia lermeb acker ae Farlow Dawson, 1954, 1959c; Dawson, Neushul, Wildman, I960, 1960« Lechuza P'oint, Calif., to Bahia Asuncion, Baja Calif. Syringoderma abyssicola (Setchell & Gardner) Levring Scagel, 1957 Northern Wash. Spatoglossum ecuadoreanum Taylor Taylor, 1945 Galapagos Arch. Spatoglossum howellii Setchell & Gardner Dawson, 1954; Dawson, Neushul, Wildman, 1960a Central Baja Calif.; Galapagos Arch. Spatoglossum lanceolatum Dawson Dawson, 1954 Ensenada de San Francisco, Sonora Spatoglossum sp. aff. S. schroederi (Mertens) J. Agardh Dawson, 1959 Isla San Pedro Nolasco, Gulf of Calif. Spatoglossum schmittii Taylor Taylor, 1945 Galapagos Arch. Spatoglossum subflabellatum Dawson Dawson, 1954 Ensenada de San Francisco, Sonora Glossophora galapagensis Taylor Taylor, 1945 Galapagos Arch. Zonaria lob at a C. Agardh Taylor, 1945 Galapagos Arch. Zonaria farlowii Setchell & Gardner Dawson, 1954, 1959c; Dawson, Neushul, Wildman, I960 Santa Barbara Co., Calif., to Isla Magdalena, Baja Calif. Pocockiella variegata (Lamouroux) Papenfuss Taylor, 1945; Dawson, 1954, 1954c, 1957, 1957$, 1959; 1959$ Rocas Alijos, Baja Calif.; Revillagigedo Arch.; southern Gulf of Calif.; Costa Rica; Gu- ayas, Ecuador; Galapagos, Arch.; Clipper- ton I. Padina caulescent Thivy Dawson, 1954, 1957$, 1959 Isla Monserrate, Gulf of Calif.; Isla Maria Magdalena, Nayarit; Golfo de Nicoya, Costa Rica Padina concrescens Thivy Taylor, 1945 Galapagos Arch. Padina crispata Thivy Dawson, 1954, 1957$, 1959 Southern Gulf of Calif, to Bahia Honda, Panama Padina durvillaei Bory Taylor, 1945; Dawson, 1954, 1957, 1957$, 1959; Dawson, Neushul, Wildman, I960 Punta Maria, Baja Calif., to Guayas, Ecuador; Galapagos Arch. Padina mexicana Dawson Dawson, 1954, 1959 Isla Turner to La Paz, Gulf of Calif. Padina tetrastromatica Hauck Dawson, 1954 Isla Maria Madre, Nayarit ? order CHORDARIALES fam. MYRIONEMATACEAE Myrionema attenuatum Setchell & Gardner Smith, 1944 Monterey Peninsula, Calif. Myrionema balticum f. calif ornicum Setchell & Gardner Smith, 1944 Monterey Peninsula, Calif. 390 PACIFIC SCIENCE, Vol. XV, July 1961 Myrionema balticum f. pedicellatum Setchell & Gardner Setchell & Gardner, 1925 Point Carmel, Calif. Myrionema compsonematoides Setchell & Gard- ner Scagel, 1957 Friday Harbor, Wash. Myrionema corunnae f. angulatum Setchell & Gardner Setchell & Gardner, 1925 San Francisco, Calif. Myrionema corunnae f. sterile Setchell & Gard- ner Smith, 1944 Carmel, Calif. Myrionema corunnae f. uniforme Setchell & Gardner Setchell & Gardner, 1925 San Mateo Co., Calif. Myrionema foecundum f. ramulosum Setchell & Gardner Scagel, 1957 Friday Harbor, Wash. Myrionema foecundum f. simplicissimum Set- chell & Gardner Scagel, 1957 Kodiak I., Alaska, to central Calif. Myrionema foecundum f. subulatum Setchell & Gardner Setchell & Gardner, 1925 San Francisco, Calif. Myrionema globosum f. affine Setchell & Gard- ner Scagel, 1957 Sitka, Alaska, to Point Carmel, Calif. Myrionema minutissimum Setchell & Gardner Setchell & Gardner, 1925 San Francisco, Calif. Myrionema obscurum Setchell & Gardner Setchell & Gardner, 1925 Moss Beach, Calif. Myrionema phyllophyllum Setchell & Gardner Setchell & Gardner, 1925 Sitka, Alaska Myrionema primarium Setchell & Gardner Scagel, 1957 Alaska to Carmel, Calif. Myrionema setiferum Setchell & Gardner Setchell & Gardner, 1925 Sitka, Alaska Myrionema strangulans Greville Scagel, 1957 Sitka, Alaska, to Carmel, Calif. Compsonema coniferum Setchell & Gardner Smith, 1944 Monterey Peninsula, Calif. Compsonema dubium Setchell & Gardner Smith, 1944 Monterey Peninsula, Calif. Compsonema fasciculatum Setchell & Gardner Smith, 1944 Pacific Grove, Calif. Compsonema fructuosum Setchell & Gardner Setchell & Gardner, 1925 Tomales Bay, Calif. Compsonema immixtum Setchell & Gardner Dawson, 1954 Isla Partida, Gulf of Calif. Compsonema intricatum Setchell & Gardner Smith, 1944 Carmel, Calif. Compsonema myrionematoides Setchell & Gard- ner Smith, 1944 Pacific Grove, Calif. Benthic Algae— DAWSON 391 Compsonema nummuloides Setchell & Gardner Setchell & Gardner, 1925 Moss Beach, Calif. Compsonema pusillum Setchell & Gardner Smith, 1944 Carmel, Calif. Compsonema r amnio sum Setchell & Gardner Smith, 1944 Carmel, Calif. Compsonema secundum Setchell & Gardner Smith, 1944 Moclips, Wash., to Carmel, Calif. Compsonema serpens Setchell & Gardner Smith, 1944 Monterey Peninsula, Calif. Compsonema sessile Setchell & Gardner Scagel, 1957 Meah Bay, Wash. Compsonema speciosum f. piliferum Setchell & Gardner Setchell & Gardner, 1925 Moclips, Wash. Compsonema sporangiiferum Setchell & Gard- ner Scagel, 1957 Neah Bay, Wash. Compsonema streblonematoides Setchell & Gardner Setchell & Gardner, 1925 Tomales Bay, Calif. Compsonema tenue Setchell & Gardner Setchell & Gardner, 1925 Sitka, Alaska Hecatonema clavatum Setchell & Gardner Setchell & Gardner, 1925 Tomales Bay, Calif. Hecatonema lawsonii Setchell & Gardner Setchell & Gardner, 1925 Uyak Bay, Alaska Hecatonema variabile Setchell & Gardner Smith, 1944 Monterey Peninsula, Calif. FAM. ELACHISTACEAE Halotbrix lumbricalis ( Kiitzing ) <. Reinke Dawson, 1954 Isla Guadalupe, Baja Calif. Elachistea fucicola (Velley) Areschoug Scagel, 1957 Sitka, Alaska, to Coos Bay, Oreg. Elachistea lubrica Ruprecht Setchell & Gardner, 1925 Prince William Sound to Wrangell, Alaska Gonodia johnstonii Setchell & Gardner Dawson, 1954 Isla San Marcos, Gulf of Calif. Gonodia mar chant ae Setchell & Gardner Dawson, 1954 La Paz, Gulf of Calif. fam. CORYNOPHLOEACEAE Petrospongium rugosum (Okamura) Setchell & Gardner Smith, 1944; Dawson, 1954, 1959c San Mateo Co., Calif., to Bahia Asuncion, Baja Calif. Leathesia difformis ( Linnaeus ) Areschoug Coe, 1932; Dawson, 1954; Scagel, 1957; Daw- son, 1959c Bering Sea to Bahia Asuncion, Baja Calif. Leathesia nana Setchell & Gardner Smith, 1944; Doty, 1947; Dawson, 1958, 1959c Or eg.; Monterey Peninsula; Carpinteria, Calif. fam. CHORDARIACEAE Eudesme vires cens (Carmichael) J. Agardh Scagel, 1957 Shumagin I. to Sitka, Alaska; La Jolla, Calif. 392 Haplogloia andersonii ( Farlow ) Levring Scagel, 1957; Dawson, 1954 Sitka, Alaska, to Cabo Colnett, Baja Calif. Haplogloia kuckuckii Kylin Scagel, 1957 Sitka, Alaska, to Friday Harbor, Wash. Chordaria dissessa Setchell & Gardner Scagel, 1957 Northern Wash. Chordaria flagelliformis (Muller) C. Agardh Setchell & Gardner, 1925 Bering Sea to Sitka, Alaska Chordaria gracilis Setchell & Gardner Setchell & Gardner, 1925 Unalaska, Alaska Saunder sella simplex ( Saunders ) Kylin Scagel, 1957 Cook Inlet, Alaska, to southern British Colum- bia Hetero chordaria ahietina (Ruprecht) Setchell & Gardner Okamura, 1933; Scagel, 1957 Bering Sea to Point Conception, Calif. fam. SPERMATOCHNACEAE Nemacystus brandegeei (Setchell & Gardner) Kylin Dawson, 1954, 1959 Scammon Lagoon, Baja Calif.; Gulf of Calif. order SPOROCHNALES fam. SPOROCHNACEAE Carpomitra costata (Stackhouse) Batters Dawson, Neushul, Wildman, 1960^ Vancouver I., British Columbia; Bahia Vis- caino, Baja Calif. Carpomitra luxurians Taylor Taylor, 1945 Galapagos Arch. PACIFIC SCIENCE, Vol. XV, July 1961 Sporochnus holleanus Montagne Taylor, 1945; Dawson, Neushul, Wildman, 1960a Isla Guadalupe; central Baja Calif.; Galapagos Arch. Sporochnus pedunculatus (Hudson) C. Agardh Dawson, 1954; Dawson, Neushul, Wildman, 1960a Isla Guadalupe; Scammon Lagoon, Baja Calif. Sporochnus rostratus Taylor Taylor, 1945 Galapagos Arch. order DESMARESTIALES fam. DESMARESTIACEAE Desmarestia farcta Setchell & Gardner Scagel, 1957 Northern Wash. Desmarestia plamentosa Dawson Dawson, 1954 Isla Angel de la Guarda, Gulf of Calif. Desmarestia foliacea Pease Scagel, 1957 Northern Wash. Desmarestia herhacea ( Turner ) Lamouroux Scagel, 1957; Dawson, 1959c; Dawson, Neu- shul, Wildman, I960 Kodiak Is., Alaska, to Isla Cedros, Baja Calif. Desmarestia intermedia Postels & Ruprecht Scagel, 1957 Bering Sea to Oreg. Desmarestia jordanii Gardner Gardner, 1940 Ventura, Calif. Desmarestia latifrons (Ruprecht) Kiitzing Doty, 1947 Coos Bay, Oreg., to Point Sur, Calif. Desmarestia latissima Setchell & Gardner Scagel, 1957 Northern Wash. Benthic Algae-- Dawson 393 Desmarestia ligulata (Lightfoot) Lamouroux Scagei, 1957 Southern British Columbia to northern Wash. Desmarestia linearis Gardner, in Smith Smith, 1944 Monterey Peninsula, Calif. Desmarestia media ( C . Agardh) Greville var. media Setchell & Gardner, 1925; Okamura, 1933 Atka I., to Unalaska, Alaska Desmarestia media var. tenuis Setchell & Gard- ner Scagei, 1957 Juneau, Alaska, to Puget Sound, Wash. Desmarestia mexicana Dawson Dawson, 1954 Isla Angel de la Guarda, Gulf of Calif. Desmarestia munda Setchell & Gardner Scagei, 1957; Dawson, Neushul, Wildman, I960 Northern British Columbia to Punta Pequena, Baja Calif.; Galapagos Arch. Desmarestia pacifica Setchell & Gardner Dawson, 1954 Santa Catalina L, Calif.; Isla Guadalupe, Baja Calif. ? Desmarestia tabacoides Okamura Dawson, 1950$; Dawson, Neushul, Wildman, I960 Santa Cruz I. to La Jolla, Calif. Desmarestia tropica Taylor Taylor, 1945 Galapagos Arch. Desmarestia vkidis (Muller) Lamouroux Scagei, 1957 Alaska to Carmel, Calif. ORDER DICT Y OSI PHON ALES FAM. STRIARIACEAE Stictyosiphon tortilis (Ruprecht) Reinke Scagei, 1957 Port Clarence, Alaska; Monterey Peninsula, Calif. fam. PUNCTARIACEAE Punctaria chartacea Setchell & Gardner Setchell & Gardner, 1925 Sitka, Alaska Punctaria expansa Setchell & Gardner Scagei, 1957 Southern British Columbia to northern Wash. Punctaria hesperia Setchell & Gardner Scagei, 1957 Victoria, British Columbia; Monterey; San Pedro, Calif. Punctaria latifolia Greville Setchell & Gardner, 1925 Metlacatla; Baranoff L, Alaska Punctaria lob at a (Saunders) Setchell & Gardner Setchell & Gardner, 1925 Prince William Sound to Sitka, Alaska Punctaria occidentalis Setchell & Gardner Smith, 1944 Monterey, Calif. Punctaria orbiculata Jao Scagei, 1957 San Juan L, Wash. Punctaria plantaginea Greville Saunders, 1901; Setchell & Gardner, 1925 Yakutat Bay, Alaska ? Halorhipis winstonii ( Anderson ) Saunders Smith, 1944 Monterey Peninsula, Calif. 394 PACIFIC SCIENCE, Vol. XV, July 1961 S or anther a ulvoidea Postels & Ruprecht f. ul- voidea Scagel, 1957; Dawson, 1958, 1959c Bering Sea to Government Point, Calif. Soranthera ulvoidea f. difformis Setchell & Gard- ner Scagel, 1957 Bering Sea to Cape Flattery, Wash. Myelophycus intestinale Saunders Okamura, 1933; Scagel, 1957 Atka I., Alaska, to Puget Sound, Wash. Ishige foliaceae Okamura Dawson, 1954; Setchell & Gardner, 1924 (as Poly opes sinicola) Northern Gulf of Calif. Phaeostrophion australe Dawson Dawson, 1958, 1959c Government Point, Calif. Phaeostrophion irregulare Setchell & Gardner Setchell & Gardner, 1925; Doty, 1947 Coos Bay, Oreg.; Bolinas, Calif. fam. SCYTOSIFHONACEAE S cy to siphon attenuatus (Foslie) Doty Doty, 1947 Coos Bay, Oreg. to central Calif. Scytosiphon bullosus Saunders Dawson, 1954; Scagel, 1957 Cook Inlet, Alaska, to central Calif.; central Baja Calif.; Gulf of Calif.; Galapagos Arch. Scytosiphon complanatus ( Rosenvinge ) Doty Doty, 1947 Cape Arago, Oreg., to Carmel, Calif. Scytosiphon lomentaria (Lyngbye) J. Agardh f. lomentaria Coe, 1932; Dawson, 1954; Scagel, 1957; Daw- son, 1959c Bering Sea to Islas San Benito, Baja Calif. Scytosiphon lomentaria f. cylindricus subf. mac- ulatus Setchell & Gardner Setchell & Gardner, 1925 San Francisco, Calif. Scytosiphon complanatus (Rosenvinge) Doty Doty, 1947 Cape Arago, Oreg., to Carmel, Calif. Petalonia debilis (C. Agardh) Derbes & Solier f. debilis Coe, 1932 (as Ilea fascia ); Dawson, 1954 (as Ilea fascia ); Scagel, 1957; Dawson, I960 a\ (see also Setchell & Gardner, 1925 for Ilea fascia f. caespitosa and Ilea fascia f. zosterifolia ) Unalaska, Alaska, to Isla Magdalena, Baja Calif. Endarachne binghamiae J. Agardh Dawson, 1954, 1959c Southern Calif, to Bahia Asuncion, Baja Calif. Colpomenia mollis Taylor Taylor, 1945 Isla Gorgona, Colombia Colpomenia ramosa Taylor Dawson, 1954 Isla Cedros, Baja Calif., to Puerto Parker, Costa Rica Colpomenia sinuosa (Roth) Derbes & Solier f. sinuosa Taylor, 1945; Coe, 1932; Dawson, 1954, 1957, 1959, 1959c; Scagel, 1957; Dawson, Neushul, Wildman, I960 Yakutat Bay, Alaska, to Puerto Parker, Costa Rica; Clipperton I.; Galapagos Arch. Colpomenia sinuosa f. expansa Saunders Setchell & Gardner, 1925 Avalon, Santa Catalina I., Calif. Colpomenia sinuosa f. tuber culata (Saunders) Setchell & Gardner Dawson, 1954; Scagel, 1957; Dawson, 1959 Unalaska, Alaska, to southern Baja Calif.; Gulf of Calif. Benthic Algae— DAWSON 395 Hydro clathms clathratus (Rory) Howe Coe, 1932; Dawson, 1954; Taylor, 1945; Dawson, 1959 La Jolla, Calif, to Scammon Lagoon, Baja Calif.; southern Gulf of Calif.; Guay as, Ecuador Rosenvingea intricata (J. Agardh) Rprgesen Dawson, 1954; Taylor, 1945; Dawson, 1959 Bahia Tepoca, Sonora, to Acapulco, Guerrero; Revillagigedo 1 Arch.; Guay as, Ecuador Rosenvingea orientalis (J. Agardh) Bprgesen Dawson, 1960» Bahia Potrero Grande, Costa Rica fam. CHNOOSPORACEAE Chnoospora implexa Hering, in J. Agardh Dawson, 1957, 1951b, 1959, 1959b Southern Gulf of Calif.; Golfo de Nicoya, Costa Rica; Clipperton I. Chnoospora minima (Hering) Papenfuss Papenfuss, 1956; Dawson, 1954, 1954c (both as C. paciftca) La Paz, Gulf of Calif., to San Agustin, Oaxaca; Isla San Benedict©, Revillagigedo Arch. Chnoospora pannosa J. Agardh Dawson, 1954 Isla Guadalupe, Baja Calif. FAM. DICTYOSIPHONACEAE Coilodesme bulligera Stroemfelt Okamura, 1933; Scagel, 1957 Aleutian Islands to Coos Bay, Oreg. Coilodesme calif ornica (Ruprecht) Kjellman Scagel, 1957; Dawson, Neushul, Wildman, I960, I960* Queen Charlotte Str., British Columbia, to central Baja Calif. Coilodesme corrugata Setchell & Gardner Setchell & Gardner, 1925 Santa Catalina L, Calif. Coilodesme cystoseirae (Ruprecht) Setchell & Gardner Setchell & Gardner, 1925 Kukak Bay; Yakutat Bay, Alaska Coilodesme polygnampta Setchell & Gardner Setchell & Gardner, 1925 Amaknak L; Unalaska, Alaska Coilodesme rigida Setchell & Gardner Setchell & Gardner, 1925; Dawson, 1959c; Dawson, Neushul, Wildman, I960, 1960^ Redondo, Calif., to Bahia Tortuga, Baja Calif. Coilodesme sitchensis Setchell & Gardner Setchell & Gardner, 1925 Sitka, Alaska Dictyosiphon foeniculaceus (Hudson) Greville Scagel, 1957 Bering Sea to Puget Sound, Wash. Dictyosiphon hippuroides (Lyngbye) Kiitzing Setchell & Gardner, 1925 Bering Sea to Unalaska, Alaska Dictyosiphon hispidus Kjellman Setchell & Gardner, 1925 Orca, Alaska Dictyosiphon sinicola Gardner Scagel, 1957 Southern British Columbia to northern Wash. Dictyosiphon tenuis Setchell & Gardner Setchell & Gardner, 1925 Golofin Bay, Alaska ORDER L AMIN ARI ALES fam. CHORDACEAE Chorda plum ( Linnaeus ) Lamouroux Scagel, 1957 Bering Sea to Puget Sound, Wash. ? 396 PACIFIC SCIENCE, Vol. XV, July 1961 fam. LAMINARIACEAE Laminaria complanata (Setchell & Gardner) Setchell Scagel, 1957 Friday Harbor, Wash. Laminaria cor data Dawson Dawson, 1950*2 Santa Catalina I., Calif. Laminaria cuneifolia J. Agardh f. cuneifolia Scagel, 1957 Bering Sea to Oreg. Laminaria cuneifolia f. amplissima Setchell & Gardner Scagel, 1957 Sitka, Alaska, to Cape Flattery, Wash. Laminaria cuneifolia f. angusta Setchell & Gard- ner Scagel, 1957 Southern British Columbia to northern Wash. Laminaria cuneifolia f. subsimplex Setchell & Gardner Scagel, 1957 Southern British Columbia to northern Wash. Laminaria dentigera Kjellman Setchell & Gardner, 1925 Aleutian Islands, Alaska Laminaria ephemera Setchell Scagel, 1957 Southern British Columbia to Oreg.; Monte- rey Peninsula, Calif. Laminaria farlowii Setchell Dawson, 1954, 1959c; Dawson, Neushul & Wildman, I960 Santa Cruz, Calif., to Bahia del Rosario, Baja Calif. Laminaria longipes Bory Okamura, 1933; Setchell & Gardner, 1925 Aleutian Islands, Alaska Laminara personata Setchell & Gardner Setchell & Gardner, 1925 Yakutat Bay to Sitka, Alaska Laminaria platymeris De la Pylaie Scagel, 1957 Bering Sea to northern Wash. Laminaria saccharina (Linnaeus) Lamouroux f. saccharina Scagel, 1957 Alaska to Coos Bay, Oreg. Laminaria saccharina f. linearis J. Agardh Scagel, 1957 Unga, Alaska, to Puget Sound, Wash. Laminaria saccharina f. membranacea J. Agardh Scagel, 1957 Alaska to Coos Bay, Oreg. Laminaria setchellii Silva Scagel, 1957 Northern British Columbia to southern Calif. Channel Islands Laminaria sinclairii ( Harvey in Hooker f. ) Far- low, Anderson & Eaton Scagel, 1957; Dawson, 1958, 1959c Southern British Columbia to Ventura Co., Calif. Pleurophycus gardneri Setchell & Saunders Scagel, 1957 Yakutat Bay, Alaska, to Coos Bay, Oreg. Cyamathere triplicata (Postels & Ruprecht) J. Agardh Scagel, 1957 Bering Sea to northern Wash. Costaria costata (Turner) Saunders Scagel, 1957 Shumagin I., Alaska, to San Pedro, Calif. Costaria mertensii J. Agardh Scagel, 1957 Bering Sea to Monterey, Calif. Benthic Algae — D awson 397 T halassiophyllum clathrus (Gmelin) Postels & Ruprecht Okamura, 1933; Scagel, 1957 Bering Sea to Straits of Juan de Fuca ? Agarum cribrosum ( Mertens ) Bory Scagel, 1957 Bering Sea to northern Wash. Agarum fmbriaium Harvey Scagel, 1957; Dawson, Neushul & Wildman, I960 Puget Sound, Wash., to southern Calif. Chan- nel Islands Hedophyllum bongardianum (Postels & Ru- precht) Yendo Miyabe & Nagai, 1932; Scagel, 1957 (as Hedophyllum sub sessile) Bering Sea to Coos Bay, Oreg. Hedophyllum sessile (C. Agardh) Setchell Okamura, 1933; Scagel, 1957 Aleutian Islands to Point Sur, Calif. Arthrothamnus bifidus (Gmelin) J. Agardh Setchell & Gardner, 1925 Aleutian Islands, Alaska fam. LESSONIACEAE Dictyoneurum calif ornicum Ruprecht Scagel, 1957 Vancouver I., British Columbia ?, to San Luis Obispo Co., Calif. Dictyoneuropsis reticulata (Saunders) G. M. Smith Smith, 1944; Silva, 1957 Fort Ross; Monterey Peninsula, Calif.; north- ern Channel Islands Nereocystis luetkeana (Mertens) Postels & Ru- precht Scagel, 1957 Shumagin I., Alaska, to San Luis Obispo Co., Calif. ( drift only to Santa Barbara ) Postelsia palmaeformis Ruprecht Scagel, 1957 Hope I., British Columbia to San Luis Obispo Co., Calif. Macrocystis integrifolia Bory Womersley, 1954; Scagel, 1957 x Sitka, Alaska, to Carmel, Calif. Macrocystis pyrifera (Linnaeus) C. Agardh Doty, 1947; Dawson, 1954, 1957, 1959c; North, 1959; Dawson, Neushul & Wild- man, I960 Sitka, Alaska, to Punta San Hipolito, Baja Calif., and sporadically to Isla Magdalena; Rocas Alijos Pelagophycus porra (Leman) Setchell Dawson, 1954; Dawson, Neushul & Wild- man, I960 Point Conception, Calif., to Islas San Benito, Baja Calif. Lessoniopsis littoralis ( Farlow & Setchell ex Tilden) Reinke Scagel, 1957 Sitka, Alaska, to Carmel, Calif. fam. ALARIACEAE Pterygophora calif ornica Ruprecht Dawson, 1954; Scagel, 1957 Southern British Columbia to Bahia del Ro- sario, Baja Calif. Alaria dolichorhachis Kjellman Okamura, 1933; Setchell & Gardner, 1925 Aleutian Islands, Alaska Alaria fistulosa Postels & Ruprecht f. fistulosa Okamura, 1933; Setchell & Gardner, 1925 Bering Sea to Wrangel, Alaska Alaria fistulosa f. platyphylla Setchell Setchell & Gardner, 1925 Bering Sea to southeastern Alaska Alaria fistulosa f. stenopylla Setchell Setchell & Gardner, 1925 Bering Sea to southeastern Alaska 398 PACIFIC SCIENCE, Vol. XV, July 1961 Alaria lanceolata Kjellman Setchell & Gardner, 1925 Bering Sea to Sitka, Alaska Alaria marginata Postels & Ruprecht Scagel, 1957 Northern British Columbia to Carmel, Calif. Alaria nana Schrader Scagel, 1957 Hope I., British Columbia to Carmel, Calif. Alaria pylaii ( Bory ) Greville Setchell & Gardner, 1925 Prince William Sound to Kodiak I., Alaska Alaria tenuifolia Setchell f. tenuifolia Okamura, 1933; Setchell & Gardner, 1925 Bering Sea to northern Wash. Alaria tenuifolia f. amplior Setchell & Gardner Scagel, 1957 Southern British Columbia to northern Wash. Alaria valida Kjellman & Setchell f. valida Scagel, 1957 Unga, Alaska, to Puget Sound, Wash. Alaria valida f. longipes Setchell & Gardner Scagel, 1957 Queen Charlotte Sound, British Columbia, to northern Wash. Eisenia arborea Areschoug Dawson, 1954, 1959c; Dawson, Neushul & Wildman, I960 Kyuquot Sound, Vancouver I., British Colum- bia; Redondo, Calif, to Isla Magdalena, Baja Calif. Eisenia desmaresti o ides Setchell & Gardner Dawson, 1954 Isla Guadalupe, Baja Calif. Eisenia galapagensis Taylor Taylor, 1945 Galapagos Arch. Eisenia masonii Setchell & Gardner Dawson, 1954 Isla Guadalupe, Baja Calif. Egregia laevigata Setchell subsp. laevigata Coe, 1932; Dawson, 1954, 1957 (as E. aus- tralis Hollenberg ms.), 1959c; Dawson, Neushul & Wildman, I960; Silva, 1957 Goleta, Calif., to Punta San Eugenio, Baja Calif.; Rocas Alijos Egregia laevigata subsp. borealis (Setchell) Silva Silva, 1957 Santa Cruz to Gaviota, Calif. Egregia menziesii (Turner) Areschoug subsp. menziesii Scagel, 1957 Northern British Columbia to Point Concep- tion, Calif. Egregia menziesii subsp. insularis Silva Silva, 1957 Northern Channel Islands, Calif. order FUCALES fam. FUCACEAE Fucus distichus Linnaeus, emend Powell Powell, 1957 (All species and subspecies of Fucus heretofore recognized from Pacific North America, with the exception of F. parksii, are considered by Powell as var- iants of E. distichus. Under subspecies edentatus he places F. edentatus De la Pylaie, F. furcatus C. Ag. (F. gardneri Silva), F. nitens Gard. and F. evanescens C. Ag., in part. Under subsp. evanescens he places F. evanescens C. Ag., in part. He considers F. memhranaceus Gard. not suf- ficiently distinct even for subspecies status. See Dawson, 1946^, for a listing of the described forms, also Scagel, 1957 (various northern forms); Dawson, 1958 (as F. furcatus) , 1959 c (as F. furcatus) Bering Sea to Government Point, Calif. Benthic Algae — DAWSON 399 Fucus parksii Gardner Gardner, 1940 Humbolt Bay; Eureka, Calif. Pelvetia fastigiata (J. Agardh) G. De Toni f. fastigiata Dawson, 1934; Scagel, 1957; Dawson, 1959c Horswell Channel, British Columbia; Coos Bay, Oreg., to Punta Baja, Baja Calif. Pelvetia fastigiata f. gracilis Setchell & Gardner Smith, 1944; Dawson, 1954 Monterey Peninsula; Channel Islands, Calif.; northern Baja Calif. Pelvetiopsis arborescens Gardner Gardner, 1940 Point Carmel, Calif. Pelvetiopsis limitata (Setchell) Gardner f. lim- itata Scagel, 1957 Hope I., British Columbia, to Carmel, Calif. Pelvetiopsis limitata f. lata Gardner Smith, 1944 Tomales Point to Monterey, Calif. Hesperophycus harveyanus (Decaisne) Setchell & Gardner Dawson, 1954, 1959c Santa Cruz, Calif., to Islas San Benito, Baja Calif. fam. SARGASSACEAE Cystophora brandegeei (Setchell & Gardner) Dawson Dawson, 1954, 1945 c La Jolla, Calif., to Isla Guadalupe, Baja Calif. Cystophora galapagensis (Piccone & Grunow in Piccone) comb. nov. Taylor, 1945 (as Blossevillea galapagensis ); Piccone, 1886: 40 (as Fucodium galapa- gense) Galapagos Arch. Cystoseira geminata C. Agardh Scagel, 1957 Bering Sea to northern Wash. , Cystoseira neglecta Setchell & Gardner Dawson, 1954 Santa Catalina I., Calif., to Desembarcadero de Miller, Baja Calif. Cystoseira osmundacea (Menzies) C. Agardh Doty, 1947; Dawson, 1954, 1959c; Dawson, Neushul & Wildman, I960. Seaside, Oreg., to Punta Abreojos, Baja Calif. Cystoseira setchellii Gardner Setchell & Gardner, 1925 Redondo to San Diego, Calif. Halidrys dioica Gardner Dawson, 1954, 1959c; Dawson, Neushul & Wildman, I960 Redondo, Calif., to Isla Asuncion, Baja Calif. Sargassum acinacif olium Setchell & Gardner Dawson, 1954 Puerto Libertad to Guaymas, Sonora Sargassum agardhianum Far low Dawson, 1954, 1959c, 1960a; Dawson, Neu- shul & Wildman, I960 Point Dume, Calif., to Punta Eugenio, Baja Calif. Sargassum, albemarlense Taylor Taylor, 1945; Dawson, 1957 Galapagos Arch. Sargassum brandegeei Setchell & Gardner Taylor, 1945; Dawson, 1954, 1959 Puerto Libertad to Isla San Pedro Nolasco, Gulf of Calif.; Galapagos Arch. Sargassum camouii Dawson Dawson, 1954 Bahia Viscaino, Baja Calif.; east central Gulf of Calif. Sargassum ecuadoreanum Taylor Taylor, 1945 Vicinity of Guayas, Ecuador 400 PACIFIC SCIENCE, VoL XV, July 1961 Sargassum galapagense Grunow Taylor, 1945 Galapagos Arch. Sargassum herporhizum Setchell & Gardner Dawson, 1954 Isla George; Isla San Pedro Martir, Gulf of Calif. Sargassum horridum Setchell & Gardner Dawson, 1954, 1959 Southern Gulf of Calif. Sargassum howellii Setchell & Gardner Dawson, 1954 Revillagigedo Arch. Sargassum johnstonii Setchell & Gardner f. johnstonii Dawson, 1954 Punta Penasco to Bahia Agua Verde, Gulf of Calif. Sargassum johnstonii f. gracile Setchell & Gard- ner Dawson, 1954 Punta Penasco to Guaymas, Gulf of Calif. Sargassum lapazeanum Setchell & Gardner Dawson, 1954 Guaymas to La Paz, Gulf of Calif. Sargassum liebmannii J. Agardh var. liebmannii Dawson, 1954, 1951b San Jose del Cabo, Baja Calif., to Bahia Honda, Panama Sargassum liebmannii var. nicoyana Grunow Dawson, 1951b Golfo de Nicoya, Costa Rica Sargassum macdougalii Dawson Dawson, 1954, 1959 Central and northern Gulf of Calif. Sargassum muticum (Yendo) Fensholt Scagel, 1957 Nanaimo, British Columbia, to Coos Bay, Oreg. Sargassum pacificum Bory f. pacificum Taylor, 1945; Dawson, 1957 Galapagos Arch. Sargassum pacificum f. congestum Setchell Taylor, 1945 Galapagos Arch. Sargassum pacificum f . rigidiusculum ( Grunow) Setchell Taylor, 1945 Galapagos Arch. Sargassum pacificum f. me gap hy Hum Taylor Taylor, 1945 Galapagos Arch. Sargassum palmeri Grunow Dawson, 1954 Santa Cruz ?; Santa Catalina I., Calif., to Islas San Benito, Baja Calif. Sargassum setifolium ( Grunow ) Setchell Taylor, 1945; Dawson, 1957 Galapagos Arch. Sargassum sinicola Setchell & Gardner Dawson, 1954, 1959 Punta Penasco to La Paz, Gulf of Calif.; Isla Guadalupe, Baja Calif. Sargassum skottsbergii Sjoestedt, prox. Taylor, 1945 Salinas, Ecuador Sargassum sonorense Dawson Dawson, 1960a Northern Gulf of Calif. Sargassum templetonii Setchell Taylor, 1945; Dawson, Neushul & Wildman, I960 a Isla Magdalena, Baja Calif.; Galapagos Arch. Sargassum vizcainense Dawson Dawson, 1954; Dawson, Neushul & Wild- man, 1960a Bahia Vizcaino and Isla Cedros to Punta Pequena, Baja Calif. Benthic Algae — DAWSON 401 Sargassum zacae Setchell Taylor, 1945 Galapagos Arch. . PHYLUM RHODOPHYCOPHYTA class RHODOPHYCEAE subclass B ANGIOPH Y CID AE order GONIOTRICHALES fam. GONIOTRICHACEAE Goniotrichutn cornu-cervi (Reinsch) Hauck Scagel, 1957 Friday Harbor, Wash.; Santa Cruz L, Calif. Goniotrichutn elegans (Chauvin) Zanardini Taylor, 1945 (as G . alsidii ); Dawson, 1954, 1957 h; Scagel, 1957 Friday Harbor, Wash., to Isla Gorgona, Co- lombia Goniotrichopsis sublittoralis G. M. Smith Smith, 1944 Monterey, Calif. order BANGIALES fam. ERYTHROPELTIDACEAF Smith or a naiadum (Anderson) Hollenberg Hollenberg, 1959; Dawson, Neushul & Wild- man, I960; Dawson, 1949$ (as Porphyra naiadum ) ; Scagel, 1957 (as Porphyra nai- adum ); Dawson, 1954 (as Porphyra nai- adum var. australis ), 1959c (as Porphyra naiadum ) Northern British Columbia to Isla Magdalena, Baja Calif. Erythropeltis discigera ( Berthold ) Schmitz Dawson, 1954 Isla Tiburon, Gulf of Calif. Erythrotrichia ascendens Dawson Dawson, 1954 Cabo Pulmo, southern Gulf of Calif. Erythrotrichia biseriata Tanaka Dawson, 1954, 1954c Islas San Benito; Punta Abreojps, Baja Calif.; Gulf of Calif, to Mazatlan, Sinaloa; Isla San Benedicto, Revillagigedo Arch. Erythrotrichia boryana (Montague) Berthold Dawson, 1954 Punta Baja to Bahia Asuncion, Baja Calif. Erythrotrichia calif ornica Kylin Smith, 1944; Dawson, 1954 Monterey, Calif.; Ensenada, Baja Calif.; Isla Tiburon, Gulf of Calif. Erythrotrichia carnea (Dillwyn) J. Agardh Smith, 1944; Dawson, 1954, 1957 b, 1959 h Monterey, Calif., to Golfo Duke, Costa Rica; Clipperton I. Erythrotrichia ciliaris (Carmichael) Batters Dawson, 1954 Punta Baja; Isla Cedros, Baja Calif. Erythrotrichia kylinii Gardner Scagel, 1957 Bering Sea to Puget Sound, Wash. Erythrotrichia parksii Gardner var. parksii Gardner, 1927 Eureka, Calif. Erythrotrichia parksii var. minor Gardner Scagel, 1957 Southern British Columbia to northern Wash. Erythrotrichia polymorpha Howe Taylor, 1945 Galapagos Arch. Erythrotrichia porphyroides Gardner Gardner, 1927 San Francisco, Calif. Erythrotrichia pulvinata Gardner Doty, 1947; Hollenberg, 1948; Dawson, 1954 Middle Bay, Oreg., to Bahia Asuncion, Baja Calif. 402 PACIFIC SCIENCE, Vol. XV, July 1961 Erythrotrichia tetraseriata Gardner Dawson, 1954 San Pedro, Calif.; Isla Magdalena, Baja Calif. Erythrotrichia welwitschii ( Ruprecht ) Batters Doty, 1947 South Bay, Oreg.; Monterey, Calif. Erythrocladia ectozoica Dawson Dawson, 1954 Scammon Lagoon, Baja Calif. Erythrocladia irregularis Rosenvinge Dawson, 1954; Scagel, 1957 Northern Wash, to Monterey, Calif.; Isla Turner, Gulf of Calif. Erythrocladia suhintegra Rosenvinge Dawson, 1954; Scagel, 1957 Northern Wash, to northern Baja Calif.; Gulf of Calif. fam. BANGIACEAE Bangia enteromorphoides Dawson Dawson, 1954 Mazatlan, Sinaloa Bangia f us copurpurea (Dillwyn) Lyngbye Dawson, 1954, 1959c; Scagel, 1957 Northern British Columbia to Costa Rica Bangia maxima Gardner Gardner, 1927 Bolinas, Calif. Bangia tenuis Gardner Scagel, 1957 Orcas I, Wash. Bangia vermicularis Harvey Sanborn & Doty, 1946; Doty, 1947 Coos Bay, Oreg., to Carmel, Calif. Porphyra abyssicola Kjellman Scagel, 1957 Southern British Columbia to northern Wash. Porphyra amplissima (Kjellman) Setchell & Hus Scagel, 1957 Amaknak I., Alaska, to northern Wash. Porphyra hollenbergii Dawson Dawson, 1954, 1959 Southern Gulf of Calif. Porphyra laciniata ( Lightfoot ) C. Agardh Setchell & Gardner, 1903 Amaknak I. to Sitka, Alaska Porphyra lanceolata (Setchell & Hus) G. M. Smith Scagel, 1957 Chehalis Bay, Wash, to Carmel, Calif. Porphyra miniata f. cuneiformis Setchell & Hus Scagel, 1957 Gulf of Alaska to Monterey, Calif. Porphyra nereocystis Anderson Scagel, 1957 Uyak Bay, Alaska, to San Pedro, Calif.? (prob- ably to San Luis Obispo Co. ) Porphyra occidentalis Setchell & Hus Sanborn & Doty, 1946; Doty, 1947 Coos Bay, Oreg.; Carmel Bay, Calif. Porphyra pendula Dawson Dawson, 1954 Isla Partida; Isla Patos, Gulf of Calif. Porphyra perforata J. Agardh f. perforata Okamura, 1933; Dawson, 1954, 1959c; Scagel, 1957 Aleutian Islands, Alaska, to Isla Magdalena, Baja Calif. Porphyra perforata f. segregata Setchell & Hus Scagel, 1957 Southern British Columbia to northern Calif. Porphyra pulchra Hollenberg Smith, 1944 Santa Cruz to Monterey, Calif. Benthic Algae — Dawson 403 Porphyra schizophylla Hollenberg Doty, 1947 Harris State Park; Otter Point, Oreg.; Pesca- dero Point, Calif. Porphyra thuretii Setchell & Dawson, in Dawson Doty, 1947; Dawson, 1954, 1959c Chetco Cove, Oreg., to Isla Magdalena, Baja Calif.; Gulf of Calif.; Puerto Parker, Costa Rica Porphyra tenuis sima (Stroemfelt) Setchell & Hus Setchell & Gardner, 1903 Shumagin I.; Yakutat Bay, Alaska Porphyra umbilicalis (Linnaeus) Kiitzing Scagel, 1957 (but with incorrect authorship) St. Paul I., Alaska, to southern British Colum- bia Porphyra variegata (Kjellman) Hus Scagel, 1957 Southern British Columbia to Monterey, Calif. Porphyrella calif ornica Hollenberg Hollenberg, 1945 Southern Calif. Porphyrella gardneri Smith & Hollenberg Hollenberg, 1945; Scagel, 1957 Northern British Columbia to Monterey, Calif. Porphyropsis coccinea (J. Agardh) Rosenvinge Scagel, 1957 Friday Harbor, Wash. Conch ocelis rosea Batters (probably represents a stage in the life cycle of one or more species of Porphyra ) Scagel, 1957 Northern Wash. subclass FLORIDEOPHYCIDAE ORDER NEMALIONALES fam. ACROCHAETIACEAE Acrochaetium amphiroae (Drew) Papenfuss Dawson, 1945 b (as Rhodochorton amphi- roae ); Doty, 1947; Dawson, 1954 Oreg. to Isla Cedros, Baja Calif. Acrochaetium angustum (Drew) Papenfuss Papenfuss, 1945 Santa Catalina I., Calif. Acrochaetium ascidiophilum Dawson Dawson, 1954 Bahia San Quintin, Baja Calif. Acrochaetium bonnemaisoniae (Batters) J. & G. Feldmann Dawson, 1954 Punta San Quintin, Baja Calif. Acrochaetium bornetii Papenfuss Dawson, 1954 San Pedro, Calif.; Punta Pequena, Baja Calif. Acrochaetium coccineum (Drew) Papenfuss Papenfuss, 1945 Marin Co.; San Francisco, Calif. Acrochaetium daviesii (Dillwyn) Nageli Dawson, 1954; Dawson, Neushul & Wild- man, 1960a Carmel, Calif., to Isla Clarion, Revillagigedo Arch. Acrochaetium densum (Drew) Papenfuss Papenfuss, 1945 San Francisco, Calif. Acrochaetium desmarestiae Kylin Scagel, 1957 Northern Wash, to Oreg. Acrochaetium dictyotae Collins Papenfuss, 1945 San Diego Co., Calif. 404 PACIFIC SCIENCE, Vol. XV, July 1961 Acrochaetium eastwoodae (Setchell & Gardner) Papenfuss Dawson, 1954 Isla Clarion, Revillagigedo Arch. Acrochaetium elegans (Drew) Papenfuss Papenfuss, 1945 La Jolla, Calif. Acrochaetium erythrophyllum Jao Scagel, 1957 Port Angeles, Wash. Acrochaetium gymnogongri (Drew) Papenfuss Papenfuss, 1945 San Francisco, Calif. Acrochaetium banco ckii (Dawson) Papenfuss Dawson, 1954 Isla Angel de la Guar da, Gulf of Calif. Acrochaetium inf es tans Howe & Hoyt Dawson, 1960# Puerto Culebra, Costa Rica Acrochaetium macounii (Collins) Hamel Scagel, 1957 Vancouver I., British Columbia, to Carmel Bay, Calif. Acrochaetium magnipcum (Drew) Papenfuss Papenfuss, 1945 La Jolla, Calif. Acrochaetium ohscurum (Drew) Papenfuss Papenfuss, 1945 Marin Co. to Carmel, Calif. Acrochaetium pacificum Kylin Dawson, 1954; Scagel, 1957; Dawson, Neu- shul & Wildman, 1960# Northern Wash.; Oreg.; Islas San Benito, Baja Calif.; vicinity of Guaymas, Sonora Acrochaetium penetrate (Drew) Papenfuss Taylor, 1945; Dawson, 1954 La Jolla, Calif.; Bahia San Quint in, Baja Calif.; Puerto Parker, Costa Rica Acrochaetium plumosum (Drew) G. M. Smith Papenfuss, 1945; Dawson, 1954 Tomales Bay, Calif., to Bahia Asuncion, Baja Calif. Acrochaetium punctatum Dawson Dawson, 1954 Bahia Bocochibampo, Sonora Acrochaetium rhizoideum (Drew) Jao var. rhizoideum Dawson, 1954; Scagel, 1957 Wash, to Isla Cedros, Baja Calif. Acrochaetium rhizoideum var. patens (Drew) G. M. Smith Smith, 1944 Carmel, Calif. Acrochaetium scinaiae Dawson Dawson, 1954 Santa Barbara I., Calif.; Gulf of Calif. Acrochaetium simplex (Drew) Papenfuss Papenfuss, 1945 Santa Monica, Calif. Acrochaetium sinicolum (Dawson) Papenfuss Dawson, 1954 Isla Turner, Gulf of Calif. Acrochaetium spiculiphilum Dawson Dawson, 1954 Bahia San Quintin, Baja Calif. Acrochaetium suhimmersum (Setchell & Gard- ner) Papenfuss Scagel, 1957 Northern Wash, to Carmel, Calif. Acrochaetium suhseriatum Bprgesen Dawson, 1959 h Clipperton I. Acrochaetium tenuis simum (Collins) Papenfuss Papenfuss, 1945 San Pedro, Calif. Benthic Algae — D awson 405 Acrochaetium thuretii var. agama (Rosenvinge) Dawson Dawson, 1945*7 San Pedro, Calif. Acrochaetium vagum (Drew) Jao Scagel, 1957 Northern Wash, to Moss Beach, Calif. Acrochaetium variabile (Drew) G. M. Smith Dawson, 1954; Scagel, 1957; Dawson, 1954c Northern Wash, to Isla San Martin, Baja Calif.; Mazatlan, Sinaloa; Isla San Bene- dicto, Revillagigedo Arch. Kylinia arcuata ( Drew ) Kylin Dawson, 1954; Scagel, 1957 Cape Flattery, Wash., to Moss Beach, Calif.; Isla Turner, Gulf of Calif. Kylinia eras sipes (Bprgesen) Kylin Dawson, 1954 Bahia Vizcaino, Baja Calif.; Isla Rasa, Gulf of Calif. ? Kylinia hirsuta (Drew) Kylin Dawson, 1949; Scagel, 1957 Vancouver, British Columbia; Santa Catalina I., Calif. Kylinia implicata ( Drew ) Papenfuss Papenfuss, 1947 Moss Beach, Calif. Kylinia microscopica ( Nageli ) Kylin Kylin, 1944; Papenfuss, 1945 (as Chroma- strum micro scopicum) La Jolla, Calif. Kylinia moniliformis ( Rosenvinge ) Kylin Scagel, 1957 Vancouver I., British Columbia Kylinia p or phyrae (Drew) Papenfuss Papenfuss, 1945 (as Chromastrum p or phy- rae) ; Dawson, 1954 San Francisco; Monterey, Calif.; Bahia San Quintin, Baja Calif. Kylinia secundata ( Lyngbye ) Papenfuss Dawson, 1954 Bahia Bocochibampo, Sonora Kylinia seriaspora Dawson Dawson, 1954 Isla Turner, Gulf of Calif. Rhodochorton concrescens Drew Papenfuss, 1945 Carmel Bay, Calif. Rhodochorton penicilliforme (Kjellman) Ro- senvinge Scagel, 1957 Spruce I., Alaska, to San Juan Co., Wash. Rhodochorton purpureum (Lightfoot) Rosen- vinge Dawson, 1954; Scagel, 1957 Bering Sea to Punta Eugenio, Baja Calif. Rhodochorton tenue Kylin Scagel, 1957 San Juan I., Wash.; Santa Cruz, Calif. Audouinella memhranacea (Magnus) Papenfuss Scagel, 1957 San Juan Co., Calif. fam. HELMINTHOCLADIACEAE Nemalion helminth oides (Velley) Batters Dawson, 1954; Scagel, 1957; Dawson, 1959c, 1960a Sitka, Alaska, to Isla Magdalena, Baja Calif. Nemalion pulvinatum Grunow Dawson, 1954 Bahia Agua Verde, Baja Calif. Nemalion virens J. Agardh Dawson, 1954 Cabeza Ballena, southern Baja Calif.; Pacific Mexico, probably Oaxaca Helminth or a saundersii Gardner Smith, 1944 Monterey, Calif. 406 PACIFIC SCIENCE, Vol. XV, July 1961 Helminth or a strict a Gardner Gardner, 1926 La Jolla, Calif. Dermonema frappieri (Montagne & Millardet) B0rgesen Dawson, 1954, 1954 b, 195 Ac, 1959 Isla San Benedicto, Revillagigedo Arch.; Isla Carmen, Gulf of Calif.; Mazatlan, Sinaloa Helminth ocladia calif ornica (J. Agardh) Kylin Dawson, 1954 Santa Barbara, Calif., to Punta Pequena, Baja Calif. Helminth ocladia gracilis Gardner Gardner, 1926 Santa Barbara, Calif. Cumagloia andersonii (Farlow) Setchell & Gardner Dawson, 1954; Scagel, 1957 Hope I., British Columbia, to Cabo Colnett, Baja Calif. Liagora abbottae Dawson Dawson, 1954 Punta Santa Rosalia, Baja Calif. Liagora calif ornica Zeh Dawson, 1945 d, 1954 Santa Catalina I., Calif., to Isla Guadalupe, Baja Calif. Liagora ceranoides Lamouroux f. ceranoides Taylor, 1945; Dawson, 1957 b Costa Rica Liagora ceranoides f. leprosa (J. Agardh) Ya- mada Dawson, 1954 Isla Guadalupe; Bahia Vizcaino, Baja Calif. Liagora farinosa Lamouroux f. farinosa Dawson, 1959 Southwestern Gulf of Calif. Liagora farinosa f. pinnatiramosa Yamada Dawson, 1954, 1957 Isla Guadalupe; Rocas Alijos, Baja Calif. Liagora magniinvolucra Dawson Dawson, 1954, 1959 Southwestern Gulf of Calif. Liagora orient alis J. Agardh Dawson, 1954 Bahia Vizcaino, Baja Calif. Liagora valida Harvey Dawson, 1957£ Golfo Duke, Costa Rica FAM. CHAETIANGIACEAE Gloiophloea confusa Setchell Dawson, 1954, 1959c, I960 #; Scagel, 1957; Taylor, 1945; Dawson, Neushul & Wild- man, I960; also see Levring, 1955 Vancouver I., British Columbia, to Punta Maria, Baja Calif.; Gulf of Calif.; Costa Rica; Galapagos Arch. Scinaia articulata Setchell Setchell, 1914#; Dawson, 1949 Santa Barbara I.; Santa Barbara, Calif. Scinaia complanata ( Collins ) Cotton Taylor, 1945 Puerto Culebra, Costa Rica; Galapagos Arch. Scinaia johnstoniae Setchell Taylor, 1945; Dawson, 1954; Dawson, Neu- shul & Wildman, I960 Southern Calif, to Costa Rica; Gulf of Calif.; Galapagos Arch. Scinaia latifrons Howe Taylor, 1945; Dawson, 1954, 1953, 1960# Southern Calif, to Puerto Guatulco, Oaxaca; Gulf of Calif.; Galapagos Arch. Scinaia minima Dawson Dawson, 1954 Isla Cedros, Baja Calif. Scinaia setchellii Taylor Taylor, 1945 Galapagos Arch. Benthic Algae— Dawson 407 Pseudoscinaia snyderae Setchell Setchell, 1914^; Dawson, \9A5b San Pedro to San Diego, Calif. W hidb eyella cartilaginea Setchell & Gardner Scagel, 1957 Whidbey L, Wash. Galaxaura angustifrons Kjellman Taylor, 1945 Galapagos Arch. Galaxaura arborea Kjellman Dawson, 1954, 1959 Bahia Santa Maria, Baja Calif.; southern Gulf of Calif. Galaxaura barbata Chou Taylor, 1945 Galapagos Arch. Galaxaura fastigiata Decaisne Dawson, 1954, 1959 Isla Guadalupe; Bahia Vizcaino, Baja Calif.; Gulf of Calif.; Isla Maria Madre, Nayarit Galaxaura filamentosa Chou Taylor, 1945; Dawson, 1954, 1951b Isla Clarion, Revillagigedo Arch.; Costa Rica; Galapagos Arch. Galaxaura intermedia Chou Taylor, 1945 Galapagos Arch. Galaxaura marginata Lamouroux Farlow, 1902 Galapagos Arch. Galaxaura oblongata (Ellis & Solander) Lam- ouroux Taylor, 1945 Esmeraldas, Ecuador Galaxaura ramulosa Kjellman Taylor, 1945; Dawson, 1957# Golfo Duke, Costa Rica; Isla Jicarita, Panama Galaxaura spathulata Kjellman Taylor, 1945 Galapagos Arch. Galaxaura squalida Kjellman Taylor, 1945; Dawson, 1954 Southeasternmost Baja Calif.; Bahia Honda, Panama Galaxaura stupocaula Chou Dawson, 1951b Costa Rica Galaxaura subfruticulosa Chou Dawson, 1954 Punta Palmilla, Baja Calif, del Sur; Isla Clar- ion, Revillagigedo Arch. Galaxaura veprecula Kjellman- Taylor, 1945; Dawson, 1954, 1951b , 1959 Southern Gulf of Calif.; Costa Rica; Gala- pagos, Arch. fam. BONNEMAISONIACEAE Bonnemaisonia nootkana (Esper) Silva Scagel, 1957 Northern British Columbia to Santa Rosa I., Calif. Bonnemaisonia geniculata Gardner Smith, 1944 Carmel Bay to Point Sur, Calif. Bonnemaisonia hamifera Hariot Feldmann & Feldmann, 1942; Dawson, 1954 (both as the gametophyte generation of Trailliella intricata)] Kylin, 1941 (as Aspa- ragopsis hamifera) ; Silva, \951a (as Bonne- maisonia intricata, the alternate generation of Trailliella intricata) Santa Rosa I., Calif., to Punta San Quintin, Baja Calif. Trailliella intricata Batters Scagel, 1957 (as sporophyte generation of Bonnemaisonia ) Northern Wash, to central Baja Calif. 408 PACIFIC SCIENCE, Vol. XV, July 1961 Asparagopsis taxiformis (Delile) Collins & Her- vey Dawson, 1954, 1957, 1959 (as the gameto- phyte generation of Falkenbergia hille- br audit) Isla Guadalupe; central Baja Calif.; Rocas Alijos; Revillagigedo Arch.; Gulf of Calif.; Isla Salango, Ecuador; Galapagos Arch. Asparagopsis svedelii Taylor Taylor, 1945 Galapagos Arch. Falkenbergia hillebrandii (Bornet) Falkenberg Dawson, 1954 (as the sporophyte generation of Asparagopsis taxiformis ) Isla Guadalupe; Bahia Vizcaino, Baja Calif.; Miramar, Nayarit; Puerto Marques, Guer- rero order GELIDIALES fam. GELIDIACEAE Gelidium arborescens Gardner Smith, 1944 Monterey to Cambria, Calif. Gelidium cartilagineum var. robustum Gardner Scagel, 1957; Dawson, 1954, 1959c; Dawson, Neushul & Wildman, I960 Southern British Columbia to Isla Magdalena, Baja Calif. Gelidium contortum Loomis Loomis, I960 San Francisco; Topanga, Calif. Gelidium coronadense Dawson Dawson, 1954 Islas Los Coronados, Baja Calif. Gelidium coulteri Harvey Sanborn & Doty, 1946; Doty, 1947; Dawson, 1954, 1959c Coos Bay, Oreg., to Isla Magdalena, Baja Calif. Gelidium crinale (Turner) Lamouroux var. crinale Dawson, 1954, 1959c; Scagel, 1957; Taylor, 1945 Southern British Columbia to Isla Magdalena, Baja Calif.; Gulf of Calif.; Mazatlan, Sina- loa; Galapagos Arch. Gelidium crinale var. luxurians Collins Gardner, 1927c; Dawson, 1954 Southern Calif, to Punta Baja, Baja Calif. Gelidium deciduum Dawson Dawson, 1954 Cabeza Ballena, near San Lucas, Baja Calif. Gelidium densum Gardner Gardner, 1927c; Dawson, 1959c Montecito; San Pedro, Calif. Gelidium distichum Loomis Loomis, 1949 Redondo, Calif. Gelidium flicinum Bory Taylor, 1945 Galapagos Arch. Gelidium gardneri Loomis Loomis, I960 Santa Monica Bay, Calif. Gelidium galapagense Taylor Taylor, 1945 Galapagos Arch. Gelidium banco ckii Taylor Taylor, 1945 (probably the G. serrulatum of Farlow, 1902) Galapagos Arch. Gelidium isabelae Taylor Taylor, 1945 Bahia Utria, Colombia; Galapagos Arch. Gelidium johnstonii Setchell & Gardner Dawson, 1954, 1959 Punta Santa Rosalia; Isla Magdalena, Baja Calif.; Gulf of Calif. Benthic Algae — D awson 409 Gelidium microdentatum Dawson Dawson, 1960^ San Bias, Nayarit Gelidium microphysa Setchell & Gardner Dawson, 1954 Isla Guadalupe; Isla Cedros, Baja Calif.; northern Gulf of Calif. Gelidium nudifrons Gardner Dawson, 1954; Dawson, Neushul & Wild- man, I960, I960 a Southern Calif, to Bahia Tortuga, Baja Calif. Gelidium papenfussii Loomis Dawson, 1954 San Francisco; Santa Monica, Calif.; Punta Baja; Punta Santa Rosalia, Baja Calif. Gelidium polystichum Gardner Dawson, 1949, 1954 San Pedro; Santa Catalina I., Calif.; Isla Gua- dalupe, Baja Calif. Gelidium pulchrum Gardner Kylin, 1941; Sanborn & Doty, 1946; Doty, 1947; Dawson, 1954, 1959c Coos Bay, Oreg., to Bahia Rosario, Baja Calif. Gelidium purpurascens Gardner Smith, 1944; Dawson, 1954, 1959c; Dawson, Neushul & Wildman, I960 Bolinas, Calif., to Cabo Colnett, Baja Calif. Gelidium pusillum (Stackhouse) Le Jolis var. pusillum Smith, 1944; Doty, 1947; Dawson, 1954, 1954c, 1957 b, 1959 Oreg. to Panama Gelidium pusillum var. cylindricum Taylor Taylor, 1945; Dawson, 1953 Pacific Mexico; Esmeraldas, Ecuador Gelidium pusillum var. pacifcum Taylor Taylor, 1945 Galapagos Arch. Gelidium pusillum var. minus culum Weber van Bosse Dawson, 1953 Pacific Mexico Gelidium pusillum var. mucronatum Dangeard Dawson, 1953 Pacific Mexico Gelidium pusillum var. conchicola Piccone & Grunow Dawson, 1953 Pacific Mexico Gelidium pusillum var. pulvinatum (C. Agardh) Feldmann Dawson, 1953 Pacific Mexico Gelidium ramuliferum Gardner Gardner, 1927c; Dawson, 1959c Santa Barbara; San Pedro, Calif. Gelidium sclerophyllum Taylor Taylor, 1945; Dawson, 1954, 1957& Isla Clarion, Revillagigedo Arch.; southern Baja Calif, to Golfo Duke, Costa Rica; Esmeraldas; Guayas, Ecuador Gelidium setchellii Gardner Gardner 1927c Marin Co., Calif. Gelidium sinicola Gardner Doty, 1947; Smith, 1944 South Bay, Oreg.; San Francisco Bay; Mon- terey, Calif. Gelidium ? sonorense Dawson Dawson, 1954 Near Guaymas, Sonora Gelidium undulatum Loomis Loomis, I960 Redondo, Calif. Gelidium venturianum Dawson Dawson, 1958 Mussel Shoals, Ventura Co., Calif. 410 PACIFIC SCIENCE, Vol. XV, July 1961 Pterocladia caloglossoides (Howe) Dawson Dawson, 1954; Dawson, Neushul & Wild- man, 1960$; Doty, 1947 (as Gelidium caloglossoides ); Smith, 1944 (as Gelidium caloglossoides ) Or eg. ?; Monterey Peninsula, Calif. ?; South- ern Calif, to Isla Asuncion, Baja Calif.; Barra de Navidad, Jalisco Pterocladia complanata Loomis Dawson, 1954 Anaheim Landing, Calif.; Punta Descanso; Punta Eugenio, Baja Calif. Pterocladia media Dawson Dawson, 1958 La Jolla, Calif. Pterocladia mcnabbiana Dawson Dawson, 1957& Golfo Dulce, Costa Rica Pterocladia musciformis Taylor Dawson, 1957^, 1960$ Bahia Carrizal, Colima; Golfo Dulce, Costa Rica Pterocladia parva Dawson Dawson, 1954 San Felipe, Gulf of Calif. Pterocladia pyramidale (Gardner) Dawson Dawson, 1954, 1954c, 1957, 1959 (as prob- ably including Gelidium decomp ositum ) , 1959c; Dawson, Neushul & Wildman, I960 Southern Calif, to Isla Magdalena, Baja Calif.; Isla Guadalupe; Rocas Alijos; Revillagi- gedo Arch.; central Gulf of Calif.; Gala- pagos Arch.; vicinity of Guayas, Ecuador Gelidiocolax microsphaerica Gardner Dawson, 1954; Fan & Papenfuss, 1959 Balboa, Calif.; Punta Descanso; Isla San Mar- tin, Baja Calif. fam. GELIDIELLACEAE Gelidiella acerosa (Forsskal) Feldmann & Hamel Dawson, 1954 Southern Gulf of Calif, to Central America Gelidiella adnata Dawson Dawson, 1954c Isla San Benedicto, Revillagigedo Arch. Gelidiella banco ckii Dawson Dawson, 1954, 1959; 1960$ Bahia Magdalena, Baja Calif.; Gulf of Calif.; Isla del Cano, Costa Rica; Isla del Rey, Panama Gelidiella liguiula Dawson Dawson, 1954 Southeasternmost Baja Calif. Gelidiella machrisiana Dawson Dawson, 1957^ Golfo de Nicoya, Costa Rica Gelidiella ? refugiensis Dawson Dawson, 1954 Isla Angel de la Guarda; Bahia San Carlos ?, Gulf of Calif. Gelidiella stichidiospora Dawson Dawson, 1954 Isla Cedros, Baja Calif. Gelidiella pannosa (Feldmann) Feldmann & Hamel Dawson, 1957^ (as G. tenuis sima) Costa Rica OF UNCERTAIN FAMILY POSITION W 'urdemannia miniata (Drapamaud) Feldmann & Hamel Dawson, 1954, 1957^ Puerto San Carlos, Gulf of Calif, to Panama Benthic Algae — Dawson 411 order CRYPTONEMIALES fam. DUMONTIACEAE Dudresnaya colombiana Taylor Taylor, 1945 Isla Gorgona, Colombia Cryptosiphonia woodii (J. Agardh) J. Agardh Scagel, 1957; Dawson, 1959c Unalaska, Alaska, to Venice, Calif. Thuretellopsis peggiana Kylin Scagel, 1957 Friday Harbor, Wash. Dumontia pliformis (Lyngbye) J. Agardh Setchell & Gardner, 1903 Alaska Baylesia plumosa Setchell Smith, 1944 Santa Cruz; Monterey, Calif. Pikea calif ornica Harvey Dawson, 1954, Scagel, 1957 Southern British Columbia to Punta Baja, Baja Calif. Pikea pinnata Setchell Scagel, 1957 Northern Wash, to Coronado, Calif. Farlowia compressa J. Agardh Scagel, 1957 Northern Wash, to Carmel, Calif. Parlowia mollis (Harvey & Bailey) Farlow & Setchell Scagel, 1957; Doty, 1947 (as including F. crassa)\ Dawson, 1959c (as F. crass a) Dixon Harbor, Alaska, to San Diego, Calif. Dilsea calif ornica (J. Agardh) Schmitz Scagel, 1957 Unga I., Alaska, to San Francisco, Calif. Dilsea integr a (Kjellman) Rosenvinge Setchell & Gardner, 1903 (as Sarcophyllis arctica ) Alaska Leptocladia binghamiae J. Agardh Setchell, 1912; Kylin, 1941; Dawson, 1954; Dawson, Neushul & Wildman, I960; Tay- lor, 1945 Santa Cruz, Calif., to Isla Magdalena, Baja Calif.; Galapagos Arch. Leptocladia conferta Setchell Smith, 1944; Doty, 1947 (as Pikea nootkana) Crescent City to Carmel Bay, Calif. Weeksia fryeana Setchell Scagel, 1957 Northern Wash. Weeksia howellii Setchell & Gardner Dawson, 1954 Isla Natividad, Baja Calif. Weeksia reticulata Setchell Smith, 1944; Dawson, 1945c (but the de- termination erroneous) Monterey Peninsula, Calif. Weeksia templetonii Setchell & Gardner Dawson, 1954 Santa Cruz I., Calif.; Isla Cedros, Baja Calif. Constantinea rosa-marina (Gmelin) Postels & Ruprecht Setchell & Gardner, 1903; Setchell, 1906 Western Aleutian Islands to Sitka, Alaska Constantinea simplex Setchell Scagel, 1957 Northern British Columbia to Carmel, Calif. Constantinea subulifera Setchell Scagel, 1957 Northern British Columbia to northern Wash. fam. RHIZOPHYLLIDACEAE Ochtodes crockeri Setchell & Gardner Taylor, 1945; Dawson, 1957 Galapagos Arch. 412 PACIFIC SCIENCE, Vol. XV, July 1961 fam. GLOIOSIPHONIACEAE Gloiosiphonia calif ornica (Farlow) J. Agardh Dawson, 1954; Scagel, 1957; Dawson, 1958 Sitka, Alaska, to Punta Baja, Baja Calif. Gloiosiphonia capillaris (Hudson) Carmichael Scagel, 1957 Southern British Columbia Gloiosiphonia verticillaris Farlow Scagel, 1957 Sitka, Alaska, to Carmel Bay, Calif. fam. ENDOCLADIACEAE Gloiopeltis furcata (Postels & Ruprecht) J. Agardh Scagel, 1957 Aleutian Islands, Alaska, to Oreg. Gloiopeltis minuta Kylin Kylin, 1941; Dawson, 1954 Santa Catalina I., Calif.; Islas San Benito; Punta Eugenio, Baja Calif. Endocladia muricata (Postels & Ruprecht) J. Agardh Okamura, 1933; Dawson, 1954, 1959c; Scagel, 1957 Aleutian Islands, Alaska, to Punta Santo To- mas, Baja Calif. fam. SQUAMARIACEAE Ethelia mexicana Dawson Dawson, 1954 Acapulco, Guerrero Peyssonelia clarionensis Taylor Taylor, 1945; Dawson, 1954 Isla Clarion, Revillagigedo Arch.; Galapagos Arch. Peyssonelia conchicola Piccone & Grunow Dawson, 1954, 1957& Bahia San Lucas, Baja Calif.; Golfo Dulce, Costa Rica Peyssonelia guadalupensis Dawson Dawson, 1954 Isla Guadalupe, Baja Calif. Peyssonelia mexicana Dawson Dawson, 1954 Cabo Colnett, Baja Calif.; Acapulco, Guerrero Peyssonelia pacifica Kylin Dawson, 1954; Scagel, 1957 San Juan I., Wash., to Isla Cedros, Baja Calif.; Mazatlan, Sinaloa Peyssonelia rubra war. oriental is Weber van Bosse Dawson, 1954, 1957, 1959; Dawson, Neushul & Wildman, I960 Southern Calif. Channel Islands to Miramar, Nayarit; Revillagigedo Arch.; Galapagos Arch. Peyssonelia squamaria (Gmelin) Decaisne Dawson, 1954 San Jose del Cabo, Baja Calif. Cruoriella dubyi ( Crouan & Crouan ) Schmitz Dawson, 1954, 1959c, l%0a La Jolla, Calif., to Isla Brincanco, Panama Cruoriella pssurata Dawson Dawson, 1954, 1957£, 1960a Guaymas, Sonora, to Isla del Rey, Panama Cruoriella banco ckii Dawson Dawson, 1954, 1960a Islas San Benito, Baja Calif., to Bahia San- tiago, Colima Cruoriella magdalenae Dawson Dawson, 1954 Punta Eugenio; Isla Magdalena, Baja Calif. Rhododermis elegans Crouan Smith, 1944; Hollenberg, 1948 Marin Co. to Monterey, Calif. Rhododermis georgii (Batters) Collins Doty, 1947 Cape Arago, Oreg. ? Benthic Algae — Dawson 413 Asymmetria expansa ( Setchell & Gardner ) Set- chell & Gardner Smith, 1944 Monterey, Calif. FAM. HILDENBRANDIACEAE Hildenbrandia canariensis var. dawsonii Ardre Ardre, 1959; Dawson, 1954 (as H. prototypus var. kerguelensis)\ Dawson, Neushul & Wildman, 1960^ Central Baja Calif. Hildenbrandia galapagensis Setchell & Gardner Setchell & Gardner, 1937 Galapagos Arch. Hildenbrandia occidentalis Setchell Taylor, 1945; Scagel, 1957; Dawson, Neu- shul & Wildman, 1960^ Northern British Columbia to Isla San Ge- ronimo, Baja Calif.; Galapagos Arch. Hildenbrandia prototypus Nardo Taylor, 1945; Dawson, 1954, 1954c, 1957£, 1959, 1959c; Scagel, 1957 (in part as H. rosea ) Alaska to Panama; Galapagos Arch. fam. CORALLINACEAE Archeolithothamnium crosslandi Lemoine Taylor, 1945 Galapagos Arch. Archaeolithothamnium howei Lemoine Lemoine, 1929; Dawson, 1960a Isla Coiba, Panama Archaeolithothamnium zonatosporum Foslie Foslie, 1906; Dawson, 1960*2 Long Beach, Calif. ? Sporolithon pacificum Dawson Dawson, 1960^ Isla del Cano, Costa Rica Melobesia galapagensis (Foslie) Taylor Taylor, 1945 Galapagos Arch. Melobesia marginata Setchell & Foslie Scagel, 1957; Dawson, 1959c, I960 Southern British Columbia to Bahia Vizcaino, Baja Calif.; southern Gulf of Calif.; Re- villagigedo Arch.; Costa Rica Melobesia mediocris (Foslie) Setchell & Mason Scagel, 1957; Dawson, 1959c, I960; Dawson, Neushul & Wildman, I960 Northern British Columbia to Isla Magda- lena, Baja Calif. Melobesia membranacea (Esper) Lamouroux Dawson, I960 Cabeza Ballena, Baja Calif.; Isla San Bene- dicto, Revillagigedo Arch. Melobesia polystromatica Dawson Dawson, I960 Bahia Tenacatita, Jalisco Lithothamnium aculeiferum L. Mason Dawson, I960 Duxbury Reef, Calif., to Scammon Lagoon, Baja Calif. Lithothamnium australe (Foslie) Foslie Dawson, I960 Isla Guadalupe, Baja Calif., to Bahia Honda, Panama. Gulf of Calif.; Revillagigedo Arch. ? Lithothamnium bisporum Foslie Dawson, 1949 Santa Catalina I., Calif. Lithothamnium calif ornicum Foslie Scagel, 1957; Dawson, 1959c, I960 Southern British Columbia to Isla Magdalena, Baja Calif. Lithothamnium cottoni Lemoine Taylor, 1945 Galapagos Arch. Lithothamnium fruticulosum (Kiitzing) Foslie Dawson, I960; Dawson & Beaudette, I960 Bodega Bay, Calif., to Bahia Uvita, Costa Rica; Islas Revillagigedos; southern Gulf of Calif. 414 PACIFIC SCIENCE, Vol. XV, July 1961 Lithothamnium giganteum L. Mason Dawson, I960; Dawson, Neushul & Wild- man, I960 Southern Calif, to central Baja Calif.; Bahia Petatlan, Guerrero; Corinto, Nicaragua ? Lithothamnium guadalupensis Dawson Dawson, I960 Isla Guadalupe, Baja Calif. Lithothamnium heteromorphum (Foslie) Foslie Dawson, 1957£ Golfo de Nicoya, Costa Rica ? ? Lithothamnium laeve (Stromfelt) Foslie Dawson, I960 Vicinity of Isla Cedros, Baja Calif. Lithothamnium lamellatum Setchell & Foslie Smith, 1944; Doty, 1947; Dawson, I960; Dawson, Neushul & Wildman, I960 Oreg. to central Baja Calif. Lithothamnium lenormandii (Areschoug) Foslie Dawson, I960 Throughout Pacific Baja Calif.; northern Gulf of Calif.; Bahia Santiago, Colima; Puerto Guatulco, Oaxaca Lithothamnium montereyicum Foslie Smith, 1944; Dawson, 1949; Mason, 1953; Dawson, I960 (as possibly a form of L. australe ) Monterey; Anacapa I., Calif. Lithothamnium muricatum (Foslie) G. De Toni Scagel, 1957 Port Renfrew, British Columbia Lithothamnium pad f cum (Foslie) Foslie Scagel, 1957; Dawson, I960; Taylor, 1945 Vancouver I., British Columbia, to La Jolla, Calif.; Isla Grande, Guerrero ?; Galapagos Arch. ? Lithothamnium phy mat odeum Foslie Scagel, 1957 Whidbey I., Wash., to Pacific Grove, Calif. Lithothamnium pocillum Lemoine Taylor, 1945 Galapagos Arch. Lithothamnium volcanum Dawson Dawson, I960 Santa Cruz I., Calif.; Cortez Bank; Isla Mag- dalena, Baja Calif. Polyporolithon conchatum (Setchell & Foslie) L. Mason Scagel, 1957 Southern British Columbia to Cambria, Calif. Polyporolithon par cum (Setchell & Foslie) L. Mason Scagel, 1957 San Juan I., Wash., to Carmel, Calif. Polyporolithon re clinatum (Foslie) L. Mason Scagel, 1957 Vancouver I., British Columbia, to La Jolla, Calif. Clathromorphum compactum (Kjellman) Foslie Mason, 1953 St. Michael, Norton Sound, Alaska Clathromorphum circumscriptum ( Stromfelt ) Foslie Mason, 1953 Aleutian I. to Unalaska, Alaska Hydrolithon arenicolum Dawson Dawson, I960 Scammon Lagoon, Baja Calif. Hydrolithon conicum Dawson Dawson, I960 Isla Socorro, Revillagigedo Arch. Hydrolithon reinholdii (Weber van Bosse & Foslie) Foslie Dawson, I960 Revillagigedo Arch. Hydrolithon setchellii (Foslie) Setchell & L. Mason Dawson, I960 Redondo to La Jolla, Calif.; Isla Guadalupe, Baja, Calif. Benthic Algae—DAWSOM 415 Fo sheila farinas a (Lamouroux) Howe Dawson, I960 Isla Guadalupe; southern Baja Calif.; Islas Tres Marias to Acapulco, Guerrero Fosliella (?) minuta Taylor Taylor, 1945 Bahia Honda, Panama; Guayas, Ecuador Fosliella paschalis (Lemoine) Setchell & Gard- ner Dawson, 195%, I960’ Isla Guadalupe, Baja Calif.; Isla Ildefonso; Punta Palmilla, Gulf of Calif.; Clipperton I. Goniolithon tessellatum (Lemoine) Setchell & L. Mason Setchell & Mason, 1943; Dawson, 1960$ Islas Contreras, Panama; Galapagos Arch. Dermatolithon canescens (Foslie) Foslie Dawson, 1957 b, I960 Isla Guadalupe; Bahia Vizcaino, Baja Calif.; central Gulf of Calif.; Mazatlan, Sinaloa; Costa Rica Dermatolithon corallinae (Crouan & Crouan) Foslie Dawson, I960 Santa Catalina L, Calif.; Cabeza Ballena, Baja Calif.; Bahia Tenacatita, Jalisco Dermatolithon dispar (Foslie) Foslie Scagel, 1957; Dawson, I960; Dawson, Neu- shul & Wildman, I960 Northern Wash, to La Jolla, Calif.; Cabeza Ballena, Baja Calif.; near Guaymas, Sonora Dermatolithon pustulatum (Lamouroux) Foslie f. pustulatum Dawson, 1957 Galapagos Arch. Dermatolithon pustulatum f. ascripticium (Fos- lie) Foslie ex De Toni Dawson, 1959c (as D, ascripticium) , I960; Taylor, 1945 (as Lithophyllum mutabile ) Coos Bay, Oreg., to Bahia Ballena, Costa Rica; Revillagigedo Arch.; Galapagos Arch. Dermatolithon saxicolum (Lemoine) Setchell & L. Mason Setchell & Mason, 1943 Isla Cocos, Costa Rica Dermatolithon veleroae Dawson Dawson, I960 Bahia Agua Verde, Baja Calif. Tenarea ere eta Lemoine Taylor, 1945 Galapagos Arch. Porolithon castellum Dawson Dawson, 1960$ Isla del Cano, Costa Rica Porolithon cocosicum Lemoine Lemoine, 1929 Isla Cocos, Costa Rica Porolithon oncodes (Heydrich) Foslie Dawson, 1959 'b, 1960$ Isla del Cano, Costa Rica; Clipperton I. Porolithon marshallense Taylor Dawson, 1959 b Clipperton I. ? Porolithon sonorense Dawson Dawson, I960 Bahia Vizcaino, Baja Calif.; northern Gulf of Calif. Mesophyllum laxum Lemoine Taylor, 1945 Galapagos Arch. Lithophyllum alternans Lemoine Taylor, 1945 Galapagos Arch. Lithophyllum amplostratum Taylor Taylor, 1945 Galapagos Arch. Lithophyllum (?) coibense Lemoine Lemoine, 1929 Isla Coiba, Panama 416 PACIFIC SCIENCE, Vol. XV, July 1961 Lithophyllum complexum Lemoine Taylor, 1945 Galapagos Arch. Lithophyllum decipiens Foslie Scagel, 1957; Dawson, 1957 b, 1959c, I960 Northern British Columbia to Guaymas, So- nora; Revillagigedo Arch.; Costa Rica ?; Islas Perlas, Panama Lithophyllum sp. aff. L. detrusum Foslie Dawson, I960 Isla Cedros, Baja Calif. Lithophyllum diguetii (Hariot) Heydrich Dawson, I960 Isla Angel de la Guarda; vicinity of La Paz, Gulf of Calif. Lithophyllum (?) divaricatum Lemoine Taylor, 1945; Dawson, I960 Galapagos Arch. Lithophyllum farlowii Heydrich Taylor, 1945 Galapagos Arch. Lithophyllum (?) fetum Foslie Lemoine, 1929 Isla Cocos, Costa Rica Lithophyllum frutescens (Foslie) Lemoine f. frutescens Taylor, 1945 (possibly the Lithothamnium calcar eum of Harvey, 1847 ) Galapagos Arch. Lithophyllum frutescens f. galapagense Foslie Taylor, 1945 Galapagos Arch. Lithophyllum grumosum (Foslie) Foslie Smith, 1944; Doty, 1947; Dawson, I960 Capa Arago, Oreg. ?, to Bahia Vizcaino, Baja Calif.; Isla San Benedicto, Revillagigedo Arch. Lithophyllum hancockii Dawson Dawson, I960 Isla Espiritu Santo, Gulf of Calif. Lithophyllum imitans Foslie Dawson, I960; Dawson, Neushul & Wild- man, I960 Moss Beach, Calif., to central Baja Calif.; Gulf of Calif.; Revillagigedo Arch.; Puerto Gua- tulco, Oaxaca Lithophyllum intermedium Foslie Taylor, 1945 Galapagos Arch. Lithophyllum lichenare L. Mason Scagel, 1957; Dawson, I960 Kanaka Bay, Wash.; Farnsworth Bank, so. Calif.; Punta Banda, Baja Calif. Lithophyllum lithophylloides Heydrich Dawson, I960 Gulf of Calif.; Isla Clarion, Revillagigedo Arch. Lithophyllum (?) lividum Lemoine Lemoine, 1929 Isla Cocos, Costa Rica Lithophyllum margaritae ( Hariot ) Heydrich Dawson, I960 Vicinity of La Paz, Gulf of Calif. Lithophyllum moluccense var. gemino stratum Taylor Taylor, 1945 Galapagos Arch. Lithophyllum neofarlowii Setchell & L. Mason Scagel, 1957; Dawson, I960 San Juan Islands, Wash., to Cambria, Calif.; Bahia Asuncion, Baja Calif. Lithophyllum pallescens ( Foslie ) Heydrich Dawson, I960 Southern Gulf of Calif.; Islas Contreras, Pa- nama Lithophyllum proboscidium Foslie var. probos- cidium Dawson, I960 Bodega Bay, Calif., to Punta Santa Rosalia, Baja Calif.; vicinity of Guaymas, Sonora; Barra de Navidad, Jalisco; Acapulco, Guer- rero Benthic Algae — DAWSON 417 Lithophyllum proboscideum var. delicatulum Dawson Dawson, I960 Isla Guadalupe, Baja Calif. Lithophyllum (?) propinquum var. cocosica Lemoine Lemoine, 1929 Isla Cocos, Costa Rica Lithophyllum reesei Dawson Dawson, I960 Punta Banda, Baja Calif. Lithophyllum rileyi Lemoine Taylor, 1945 (as probably equal to the Lithothamnium eras sum of Piccone, 1889) Galapagos Arch. Lithophyllum samoense Foslie Dawson, I960 Isla Cedros, Baja Calif. Lithophyllum sancti-georgei Lemoine Taylor, 1945 Galapagos Arch. Lithophyllum trichotomum (Heydrich) Lemoine Dawson, 1957, I960 Revillagigedo Arch.; Gulf of Calif, to Isla Jicaron, Panama; Galapagos Arch. Lithophyllum veleroae Dawson Dawson, I960 Isla Espiritu Santo; vicinity of Santa Rosalia, Gulf of Calif. Lithophyllum whidbeyense Foslie Scagel, 1957 Southern British Columbia to northern Wash. Heteroderma corallinicola Dawson Dawson, I960 Guaymas, Sonora Heteroderma gibbsii (Foslie & Setchell) Foslie Dawson, I960 Gulf of Calif. Heteroderma lejolisii (Rosanoff) Foslie Dawson, I960 Bahia San Quint in, Baja Calif. Heteroderma minutula Foslie Dawson, 1959 h, I960 Isla Guadalupe; Bahia Tortuga, Baja Calif.; Puerto Guatulco, Oaxaca, to Punta Na- ranja, Panama; Clipperton I. Heteroderma nicholsii Setchell & L. Mason Dawson, I960 Bolinas, Calif., to Punta Abreojos, Baja Calif. Heteroderma parvicarpa Dawson Dawson, I960 Scammon Lagoon, Baja Calif. Heteroderma subtilissima ( Foslie ) Foslie Dawson, I960 Isla Guadalupe, Baja Calif.; southwestern- most Gulf of Calif. Litholepis accola Foslie Dawson, I960 Isla Guadalupe, Baja Calif.; Isla Clarion, Re- villagigedo Arch. Litholepis fertilis (Lemone) Setchell & L. Mason Setchell & Mason, 1943 Isla Coiba, Panama Litholepis sonorensis Dawson Dawson, I960 Isla Espiritu Santo, Gulf of Calif. Choreonema thuretii (Bornet) Schmitz Taylor, 1945; Hollenberg, 1948; Dawson, 1949, I960 Southern Calif.; Isla Guadalupe to Bahia Viz- caino, Baja Calif.; southwestern Gulf of Calif.; Revillagigedo; Costa Rica; Gala- pagos Arch. Corallina bathybentha Dawson Dawson, 1949 Southern Calif. Channel Islands 418 PACIFIC SCIENCE, Vol. XV, July 1961 Corallina berterii Mont. ? Farlow, 1902; Lemoine, 1929 (as Arthrocar- dia sp. ) Galapagos Arch. Corallina delicatulus (Doty) comb. nov. Doty, 1947: 167 (as Joculat or delicatulus) Cape Arago, Or eg. Corallina gracilis Lamouroux var. gracilis Dawson, 1945 b, 1954, 1959c; Dawson, Neu- shul & Wildman, I960 Southern Calif, to Isla Cedros, Baja Calif. Corallina gracilis var. verticillata Dawson Dawson, 1954 Santa Catalina I., Calif. ?; Isla Guadalupe, Baja Calif. Corallina janioides Dawson Dawson, 1954 Isla Guadalupe, Baja Calif. Corallina officinalis Linnaeus var. officinalis Taylor, 1945 Galapagos Arch. Corallina officinalis var. ch Hen sis (Harvey) Kiitzing Scagel, 1957; Dawson, 1954, 1959c; Dawson, Neushul & Wildman, I960 Northern British Columbia to Isla Magdalena, Baja Calif.; Guayas, Ecuador Corallina pilulifera Postels & Ruprecht Scagel, 1957 (as to Alaska records only) Alaska Corallina pinnatifolia (Manza) Dawson var. pinnatifolia Dawson, 1954, 1958, 1959c; Taylor, 1945 (as Joculator pinnatif olius ) Carpinteria, Calif., to Isla Magdalena, Baja Calif.; Galapagos Arch. Corallina pinnatifolia var. digitata Dawson Dawson, 1954, 1959 Throughout the Gulf of Calif. Corallina polysticha Dawson Dawson, 1954, 1959 Santa Catalina I., Calif., to Bahia Vizcaino, Baja Calif. Corallina vancouveriensis Yendo var. Vancouver - iensis Dawson, 1954, 1959c, 1960^; Scagel, 1957; Dawson, Neushul & Wildman, I960 Southern British Columbia to Punta Lobos, near Todos Santos, Baja Calif. Corallina vancouveriensis var. densa Yendo Dawson, 1953 Northern part of the range of the species. Corallina vancouveriensis var. aculeata (Yendo) Dawson Dawson, 1954. Throughout the range of the species. Corallina vancouveriensis var. ly cop odioides (Taylor) Dawson Dawson, 1954 Southern Calif, and Baja Calif. Serraticardia macmillani (Yendo) Silva 1957 Scagel, 1957; Silva, 1957 Vancouver I., British Columbia; Fort Ross; Pescadero Point, Calif. Cheilosporum planiusculum (Kiitzing) Yendo Yendo, 1902 Port Renfrew, Vancouver L, British Columbia Cheilosporum frondescens (Postels & Ruprecht) Yendo f. frondescens Yendo, 1902 Port Renfrew, Vancouver I., British Columbia Cheilosporum frondescens f. maxima Yendo Yendo, 1902 Port Renfrew, Vancouver I., British Columbia Cheilosporum frondescens f. intermedia Yendo Yendo, 1902 Port Renfrew, Vancouver L, British Columbia Benthic Algae — ID AW SON 419 Cheilospomm frondescens f. polymorpha Yendo Yendo, 1902 Port Renfrew, Vancouver L, British Columbia Pachyarthron ctetaceum (Postels & Ruprecht) Manza Scagel, 1957 Unalaska, Alaska, to northern Wash. Litbotknx aspergillum J. E. Gray Dawson, 1954, 1959c; Scagel, 1957; Dawson, Neushul & Wildman, I960' Vancouver I., British Columbia, to Isla Mag- dalena, Baja Calif. Bos siella calif omica (Decaisne) Silva Scagel, 1957 Southern British Columbia to Monterey, Calif. Bossiella corymbifera (Manza) Silva Scagel, 1957 Vancouver I., British Columbia, to Point Lo- bos, Calif. Bossiella coop eri ( Dawson & Silva ) Silva Silva, 1957 Southern Calif. Channel Islands to Punta Baja, Baja Calif. Bossiella dichotoma (Manza) Silva var. dicho- toma Dawson, 1954, 1959c, 196(Ez; Scagel, 1957 Southern British Columbia to Punta Thurloe, Baja Calif. Bossiella dichotoma var. garcCneri (Manza) Dawson Dawson, 1960^, 1954 (as Bossea gardneri ); Silva, 1957 (as Bossiella gardneri ); Daw- son, 1959c (as Bossiella dichotoma- gard- neri complex) ; Dawson, Neushul & Wild- man, I960 (as Bossiella gardneri) Oreg. to Punta Maria, Baja Calif. Bossiella insularis (Dawson & Silva) Silva Silva, 1957; Dawson, 1958, 1959c Southern Calif. Channel Islands; Port Hue- neme, Calif., to Bahia Asuncion, Baja, Calif. Bossiella interrupta (Manza) Silva Silva, 1957 Pacific Grove, Calif. Bossiella ligulata (Dawson) Silva Silva, 1957; Dawson, 1958, 1959c Ventura Co., Calif.; Isla Guadalupe, Baja Calif. Bossiella orbigniana (Decaisne) Silva Doty, 1947 (as Bossea orbigniana ) ; Dawson, 1954 (as Bossea orbigniana ); Silva, 1957; Dawson, 1959c; Dawson, Neushul & Wild- man, I960 Cape Arago, Oreg. to Isla Cedros, Baja Calif. Bossiella p achy clada (Taylor) Silva Dawson, 1954 (as Bossea p achy clada ) ; Silva, 1957 Santa Cruz L, Calif., to Isla Cedros, Baja Calif. Bossiella plumosa (Manza) Silva Scagel, 1957; Dawson, 1959c Vancouver I., British Columbia, to Govern- ment Point, Calif. Bossiella sagitata (Dawson & Silva) Silva Silva, 1957; Dawson, 1958, 1959c Southern Calif. Channel Islands; Port Hue- neme, Calif.; Islas Coronados; Punta Eu- genio, Baja Calif. Jania adhaerens Lamouroux, prox. Taylor, 1945 Esmeraldas, Ecuador Jania arborescens Yendo, prox. ? Taylor, 1945 Guay as, Ecuador Jania capillacea Harvey Taylor, 1945; Dawson, 1954, 1957, 1951b, 1959, 1959b Isla Cedros, Baja Calif., to Guayas, Ecuador; Galapagos Arch. Jania decussate -dichotoma (Yendo) Yendo Dawson, 1954, 1959 La Jolla, Calif.; southern Gulf of Calif. 420 PACIFIC SCIENCE, Vol. XV, July 1961 Jania longiarthra Dawson Dawson, 1954, 1957*, 1959, 1959* Southern Gulf of Calif.; Golfo de Nicoya, Costa Rica; Clipperton I. Jania mexicana Taylor Dawson, 1954 Mazatlan, Sinaloa, to Salina Cruz, Oaxaca Jania natalensis Harvey Dawson, 1954, 1958, 1959c Lechuza Point, southern Calif., to Isla Cedros, Baja Calif. Jania suhpinnata Dawson Dawson, 1954 Bahia de La Paz, Gulf of Calif. Jania tenella Kiitzing var. tenella Dawson, 1954, 1954c, 1957, 1957 *, 1959, 1959*, 1959c Southern Calif, to Costa Rica; Clipperton I. Jania tenella var. zacae Dawson Dawson, 1954, 1957* Bahia Vizcaino, Baja Calif.; Isla Ildefonso, Gulf of Calif.; Costa Rica Jania ungulata Yendo, prox. Taylor, 1945 Galapagos Arch.; Guayas, Ecuador Amphiroa annulata Lemoine var. annulata Dawson, 1954, 1957, 1957*, 1959 Central Baja Calif.; Southern Gulf of Calif, to Costa Rica; Galapagos Arch. Amphiroa annulata var. pinnata Dawson Dawson, 1954, 1957* Cape district, Baja Calif.; Golfo Dulce, Costa Rica? Amphiroa beauvoisii Lamouroux Dawson, 1957* Costa Rica Amphiroa brevianceps Dawson Dawson, 1954; Dawson, Neushul & Wild- man, 196(k Punta San Hipolito, Baja Calif.; Guaymas ?, Punta Frailes ?, Gulf of Calif.; Salina Cruz, Oaxaca Amphiroa compressa Lemoine Taylor, 1945 (as probably equal to the A. dilatata of Farlow, 1902) Galapagos Arch. Amphiroa crossiandii Lemoine Dawson, 1954, 1957 Rocas Alijos, Baja Calif.; Isla Socorro, Revil- lagigedo Arch. ?; Tangola-Tangola, Oaxaca; Isla Gorgona, Colombia Amphiroa dimorpha Lemoine var. dimorpha Dawson, 1954, 1954c, 1957*, 1959; Dawson, Neushul & Wildman, I960, 1960^; Taylor, 1945 Central Pacific Baja Calif, to Central Amer- ica; Gulf of Calif.; Islas Revillagigedos; Galapagos Arch. Amphiroa dimorpha var. digitiforme Dawson Dawson, 1959 Isla Carmen; Isla Ildefonso, Gulf of Calif. Amphiroa drouetii Dawson Dawson, 1954, 1959 Central and southern Gulf of Calif. Amphiroa foliacea Lamouroux Dawson, 1954 Punta Verlero, Bahia Vizcaino, Baja Calif.; Isla Maria Magdalena, Nayarit Amphiroa epiphlegnoides J. Agardh, in Harvey Harvey, 1862 Fuca Strait, Wash. Amphiroa franciscana Taylor var. frauds cana Dawson, 1954 Isla Guadalupe, Baja Calif.; Isla Maria Mag- dalena, Nayarit; Esmeraldas; Guayas, Ecua- dor Benthic Algae-— Dawson 421 Amphiroa franciscana var. robust a Dawson Dawson, 1954 Bahia Agua Verde, Gulf of Calif.; Acapulco, Guerrero Amphiroa galapagensis Taylor Taylor, 1945 Galapagos Arch. Amphiroa magdalenensis Dawson Dawson, 1954, 1959 Isla Guadalupe; Isla Magdalena, Baja Calif.; southern Gulf of Calif. Amphiroa mexicana Taylor Dawson, 1954 Morro de Petatlan, Guerrero; Salina Cruz, Oaxaca Amphiroa minutissima Taylor Dawson, 1957& Golfo Duke, Costa Rica Amphiroa peruana Areschoug Taylor, 1945 Galapagos Arch. Amphiroa subcylindrica Dawson Dawson, 1954, 1959 Gulf of Calif.; Barra de Navidad, Jalisco Amphiroa taylorii Dawson Dawson, 1954, 1954c, 1957 b, 1959 Revillagigedo Arch.; southern Gulf of Calif.; Isla Maria Magdalena, Nayarit; Costa Rica Amphiroa tuberculosa (Postels & Ruprecht) Endlicher Yendo, 1902 Port Renfrew, British Columbia Amphiroa vanbosseae Lemoine Taylor, 1945 Galapagos Arch. Amphiroa zonata Yendo Taylor, 1945 (as A. peninsularis) ; Dawson, 1954, 1959c; Dawson, Neushul & Wild- man, I960 Santa Catalina I., Calif., to Isla Magdalena, Baja Calif.; Gulf of Calif.; Isla Maria Madre, Nayarit; Esmeraldas, Ecuador Arthrocardia tuberculosa Weber van Bosse Lemoine, 1929 (as equal to the Amphiroa orbigniana of Farlow, 1902) Galapagos Arch. Calliarthron chielosporioides Manza Smith, 1944; Dawson, 1954; Dawson, Neu- shul & Wildman, I960 Bolinas, Calif., to Isla Cedros, Baja Calif. Calliarthron regenerans Manza Scagel, 1957; Dawson, Neushul & Wildman, 1960, I960* Vancouver I., British Columbia, to Isla San Geronimo, Baja Calif. Calliarthron setchelliae Manza Smith, 1944 Moss Beach; Monterey, Calif. Calliarthron schmittii Manza Dawson, 1949; Scagel, 1957 San Juan I., Wash.; Tanner Bank; San Diego, Calif. fam. DERMOCORYNIDACEAE Dermocorynus occidentalis Hollenberg Dawson, 1954 Los Angeles Co., Calif., to Punta Banda, Baja Calif. fam. CRYPTONEMIACEAE Grateloupia abreviata Kylin Hollenberg, 1948; Dawson, 1954, 1959c Southern Calif. Grateloupia avalonae Dawson Dawson, 1949 Catalina I., Calif. 422 PACIFIC SCIENCE, Vol. XV, July 1961 Grateloupia dactylifera Dawson Dawson, 1954 Near Guay mas; Isla Espiritu Santo, Gulf of Calif. Grateloupia filicina (Lamour.) C. Agardh Dawson, 1954 Northern Baja Calif, to San Bias, Nayarit; Gulf of Calif. Grateloupia hancockii Dawson Dawson, 1954 Kino; vicinity of Guay mas, Gulf of Calif. Grateloupia howeii Setchell & Gardner Dawson, 1954, 1959; Dawson, Neushul & Wildman, I960 Islas San Benito to Isla Magdalena, Baja Calif.; Puerto Libertad, Sonora, to Mazat- lan, Sinaloa Grateloupia (?) johnstonii Setchell & Gardner Dawson, 1954 Isla Angel de la Guarda, Gulf of Calif. Grateloupia maxima ( Gardner ) Kylin Doty, 1947; Dawson, 1954; Dawson, Neu- shul & Wildman, I960* Capa Arago region, Oreg., to Isla Asuncion, Baja Calif. Grateloupia multiphylla Dawson Dawson, 1954 Islas San Benito to Isla Magdalena, Baja Calif. Grateloupia pinnata (Postels & Ruprecht) Set- chell Scagel, 1957 Sitka, Alaska, to northern Wash. Grateloupia prolongata J. Agardh Dawson, 1954, 1958, 1959, 1959c Gaviota, Calif., to southern Mexico Grateloupia schizophylla Kiitzing Dawson, 1954, 1959c; Dawson, Neushul & Wildman, I960; Scagel, 1957 (as G. cali- fornica ) Southern British Columbia to Isla Magda- lena, Baja Calif. Grateloupia setchellii Kylin Doty, 1947 Oreg.; Monterey, Calif. Grateloupia (?) squarrulosa Setchell & Gard- ner Dawson, 1944 Isla Smith, Gulf of Calif. Grateloupia versicolor (J. Agardh) J. Agardh var. versicolor Dawson, 1954, 1954c, 1959 Isla Magdalena, Baja Calif., to Salina Cruz, Oaxaca; southern Gulf of Calif. Grateloupia versicolor var. prostrata Dawson Dawson, 1954c Isla San Benedicto, Revillagigedo Arch. Grateloupia violacea (Setchell & Gardner) Daw- son Dawson, 1961 Northern Gulf of Calif. Cryptonemia angustata (Setchell & Gardner) Dawson Dawson, 1954, 1958; Dawson, Neushul & Wildman, I960; Dawson, I960* La Jolla, Calif., to southern Baja Calif.; Bahia Tepoca, Sonora; Isla Socorro, Revillagi- gedo Arch.; Bahia Culebra, Costa Rica Cryptonemia borealis Kylin Scagel, 1957 Southern British Columbia to Oreg. Cryptonemia decolorata Taylor Dawson, 1954 Isla Maria Magdalena, Nayarit Cryptonemia guaymasensis ( Dawson ) Dawson Dawson, 1954, I960* Isla Patos to Bahia San Lucas, Gulf of Calif.; Bahia Culebra, Costa Rica Cryptonemia obovata J. Agardh Kylin, 1925; Scagel, 1957; Dawson, 1954, 1959c; Dawson, Neushul & Wildman, I960, I960* Prince William Sound, Alaska, to Bahia Tor- tuga, Baja Calif. Benthic Algae— Dawson 423 Cryptonemia ovalifolia Kylin Scagel, 1957 Northern Wash, to Pacific Grove, Calif. Cryptonemia peltata Dawson, Neushul & Wild- man Dawson, Neushul & Wildman, 1960$ Punta Eugenio, Baja Calif. Cryptonemia veleroae ( Dawson ) Dawson Dawson, 1954 Guaymas, Sonora Aeodes (?) ecuadoreana Taylor Taylor, 1945 Galapagos Arch. Aeodes gardneri Kylin Scagel, 1957 Southern British Columbia to San Pedro, Calif. Carpopeltis bushiae (Farlow) Kylin Kylin, 1941; Dawson, 1954, 1959c (all as Polyopes bushiae ); Dawson, 1959; Daw- son, Neushul & Wildman, I960; 1960$ Southern Calif, to Punta Abreojos, Baja Calif. Carpopeltis clarionensis (Setchell & Gardner) Dawson Dawson, 1954 (as Polyopes clarionensis ) ; Dawson, 1959 Isla Clarion, Revillagigedo Arch. Carpopeltis divaricata Okamura Dawson, Neushul & Wildman, 1960$ Punta San Hipolito; Isla Magdalena, Baja Calif. Carpopeltis stella-polaris Dawson Dawson, 1959 Isla Ildefonso, Gulf of Calif. Pachymenia saxicola Taylor Taylor, 1945 Galapagos Arch. Halymenia abyssicola Dawson Dawson, 1954 Isla Partida, Gulf of Calif. Halymenia actinophysa Howe Dawson, 1954, 1960$ Throughout Gulf of Calif.; Revillagigedo Arch.; Bahia Salinas, Costa Rica Halymenia agardhii G, DeToni Dawson, 1954; Dawson, Neushul & Wild- man, 1960$ Isla Asuncion, Baja Calif.; Isla Maria Mag- dalena, Nayarit Halymenia bifida Dawson Dawson, 1954 Bahia de San Lucas, Baja Calif. Halymenia calif ornica Smith & Hollenberg Dawson, 1954; Scagel, 1957 Northern Wash.; Monterey Peninsula, Calif.; Punta Maria, Baja Calif. Halymenia me gasp ora Dawson Dawson, 1954 Canal de San Lorenzo, Gulf of Calif. Halymenia santamariae Taylor Taylor, 1945 Galapagos Arch. Halymenia utriana Taylor Taylor, 1945 Bahia Utria, Colombia Prionitis abbreviata Setchell & Gardner var. abbreviata Taylor, 1945; Dawson, 1954 (as Zanardinula abbreviata) , 1959 Throughout the Gulf of Calif.; Galapagos Arch. Prionitis abbreviata var. guaymasensis (Daw- son) Dawson Dawson, 1954 (as Zanardinula abbreviata var. guaymasensis ) , 1959 Throughout the Gulf of Calif, south to Isla San Pedro Nolasco 424 PACIFIC SCIENCE, Vol XV, July 1961 Prionitis acroidalea ( Setcheli & Gardner ) comb, nov. Setcheli & Gardner, 1924: 781 (as Gratelou- pia acroidalea ); Dawson, 1954 (as Za- nardinula acroidalea ) Kina, Sonora, to La Paz, Baja Calif.; Galapa- gos Arch. Prionitis albemarlensis Taylor Taylor, 1945 Galapagos Arch. Prionitis andersoniana Eaton ex J. Agardh Sanborn & Doty, 1946; Doty, 1947; Dawson, 1945^, 1954 (as Zanardinula andersoni- ana ), 1959c; Dawson, Neushul & Wild- man, I960 Coos Bay, Oreg., to Punta Maria, Baja Calif. Prionitis australis (J. Agardh) J. Agardh Smith, 1944; Papenfuss, 1944 (as Zanardi- nula australis ) Monterey, Calif. Prionitis cornea (Okamura) Dawson Smith, 1944 (as Prionitis linearis ) ; Dawson, 1954 (as Zanardinula cornea ); Papenfuss, 1944 (as Zanardinula linearis ); Dawson, 1958, 1959c; Dawson, Neushul & Wild- man, I960 San Mateo Co., Calif., to Isla San Roque, Baja Calif. Prionitis delicatula ( Taylor ) comb. nov. Taylor, 1945: 210 (as Prionitis fiiliformis f. delicatula ); Dawson, 1954 (as Zanardi- nula delicatula ) Central Pacific Baja Calif, south to Isla Mag- dalena Prionitis filiformis Kylin Doty, 1947 (as Zanardinula filiformis ); Daw- son, Neushul & Wildman, I960* Cape Arago, Oreg., to central Calif.; Punta Abreojos, Baja Calif. Prionitis galapagensis Taylor Taylor, 1945 Galapagos Arch. Prionitis banco ckii Taylor Taylor, 1945 Galapagos Arch. Prionitis lanceolata (Harvey) Harvey Dawson, 1954 (as Zanardinula lanceolata ); Scagel, 1957; Dawson, 1959c; Dawson, Neushul & Wildman, 19'60 : Southern British Columbia to Isla Cedros, Baja Calif. Prionitis lyallii Harvey Scagel, 1957; Dawson, 1954 (as Zarnardinula lyallii var. ? ) Southern British Columbia to Carmel Bay, Calif.; vicinity of Guaymas, Sonora ? Prionitis mexicana Dawson Dawson, 1954 (as Zanardinula mexicana ) Southern Baja Calif. Prionitis vizcainensis ( Dawson ) Dawson, Neu- shul & Wildman Dawson, 1954 (as Zanardinula vizcainensis)^ Dawson, Neushul & Wildman, 1960# Bahia Vizcaino, Baja Calif. Lobocolax deformans Howe Doty, 1947; Dawson, 1954 Oreg. to central Baja Calif. fam. KALLYMENIACEAE Kallymenia latiloba Taylor Taylor, 1945 Galapagos Arch. Kallymenia multiloba Taylor Taylor, 1945 Galapagos Arch. Kallymenia oblongifructa Setcheli Scagel, 1957 Sitka; S el do via, Alaska, to Puget Sound, Wash. Kallymenia ornata (Postels & Ruprecht) J. Agardh Scagel, 1957 Yakutat Bay, Alaska, to Vancouver i, British Columbia Benthic Algae — D awson 425 Kallymenia pacifica Kylin Kylin, 1956 (as Callymenia ) ; Dawson, Neu- shul & Wildman, 1960a La Jolla, Calif.; Punta Eugenio, Baja Calif. Kallymenia (?) pertusa Setchell & Gardner Dawson, 1954 (as Callymenia) Isla San Pedro Martir, Gulf of Calif. Kallymenia reniformis (Turner) J. Agardh f. reniformis Scagel, 1957 Southern British Columbia Kallymenia reniformis f. cuneata J. Agardh Setchell & Gardner, 1903 Unalaska; Morzhovoi Bay, Alaska Kallymenia setchellii Taylor Taylor, 1945 Galapagos Arch. Pugetia fragilissima Kylin Scagel, 1957 Southern British Columbia to Monterey, Calif. Callophyllis aero car pa Setchell Dawson, 1954^ Santa Cruz to Santa Barbara, Calif. Callophyllis crassifolia Setchell & Swezy Dawson, 1954^ San Mateo Co. to Carmel, Calif. Callophyllis crenulata Setchell Scagel, 1957 Northern British Columbia to Pacific Grove, Calif. Callophyllis dissecta Setchell & Swezy Dawson, 1954 San Pedro, Calif., to Punta San Quintin, Baja Calif. Callophyllis edentata Kylin Scagel, 1957 Hope I., British Columbia, to Oreg. Callophyllis filicina Setchell & Swezy Dawson, 1954^ Southern Calif.; Islas San Benito, Baja Calif. Callophyllis firma (Kylin) Norris Scagel, 1957 Northern British Columbia to Monterey, Calif. Callophyllis flabellulata Harvey Scagel, 1957 Northern British Columbia to Coos Bay, Oreg. Callophyllis gardneri Setchell Dawson, 1954^ Santa Rosa I., Calif., to Bahia Vizcaino, Baja Calif. Callophyllis heanophylla Setchell Scagel, 1957 Southern British Columbia to Coos Bay, Oreg. Callophyllis (?) ligulata Taylor Taylor, 1945 Isla Gorgona, Colombia; Galapagos Arch. Callophyllis marginifructa Setchell & Swezy Scagel, 1957; Dawson, 1954; Dawson, Neu- shul & Wildman, I960, 1960a Northern Wash, to Sacramento Reef, Baja Calif. Callophyllis megalocarpa Setchell & Swezy Dawson, 1954; Scagel, 1957 Northern Wash, to Isla Magdalena, Baja Calif. Callophyllis ohtusifolia J. Agardh Dawson, 1954, 1959c Santa Cruz, Calif., to Punta Santa Rosalia, Baja Calif. Callophyllis odonthalioides Setchell Dawson, 1954^ Santa Cruz to San Diego, Calif. Callophyllis oregona Doty Doty, 1947 North Bay and Brookings, Oreg. 426 PACIFIC SCIENCE, Vol. XV, July 1961 Callophyllis phyllohaptera Dawson Dawson, 1954 Punta Descanso to Isla Magdalena, Baja Calif. Callophyllis pinnata Setchell & Swezy Dawson, 1954^ Oreg. to Punta Baja, Baja Calif. Callophyllis plumosa Setchell & Swezy Dawson, 1954^, 1959c San Pedro to La Jolla, Calif. Callophyllis stenophylla Setchell Dawson, 1949, 1954^ Bolinas; Santa Cruz I., Calif. Callophyllis thompsonii Setchell Scagel, 1957 Northern Wash. Callophyllis violacea J. Agardh var. violacea Doty, 1947; Dawson. 1954^, 1959c; Dawson, Neushul & Wildman, I960, 1960^ Oreg.; Santa Barbara, Calif., to Isla Asuncion, Baja Calif. Callophyllis violacea var. epiphytica Dawson Dawson, 1954, 1957 Central Baja Calif.; Rocas Alijos Callocolax fungiformis Kylin Scagel, 1957 Northern Wash. Callocolax glob ulo sis Dawson Scagel, 1957 Northern Wash.; San Diego, Calif. Erythrophyllum delesserioides J. Agardh Scagel, 1957 Hope I., British Columbia, to San Luis Obispo Co., Calif. Erythrophyllum gmelini ( Grunow ) Y endo Okamura, 1933; Setchell & Gardner, 1903 (as Callymenia gmelini) Atka to Agattu I., Aleutian Islands, Alaska Erythrophyllum splendens Doty Doty, 1947 Cape Arago, Or eg. Euthora cristata (Linnaeus) J. Agardh Setchell & Gardner, 1903 Shumagin I. to Cook Inlet, Alaska Euthora jruticulos a (Ruprecht) J. Agardh Okamura, 1933; Scagel, 1957 Bering Sea to northern Wash. fam. CHOREOCOLACEAE Choreocolax polysiphoniae Reinsch Dawson, 1954; Scagel, 1957 Sitka, Alaska, to Cabo Colnett, Baja Calif. order GIGARTINALES fam. CRUORIACEAE Cruoria pacifca Kjellman Hollenberg, 1948 Southern Calif. Cruoria profunda Dawson Dawson, 1961 Cortez Bank, off U.S.-Mexico boundary Haematocelis rubens J. Agardh Dawson, 1954, 1961 La Jolla, Calif., to Isla Magdalena, Baja Calif. Cruoriopsis mexicana Dawson Dawson, 1954, 1957&, 1960a, 1961 Islas Coronados, Baja Calif.; San Felipe, Gulf of Calif.; Golfo Duke, Costa Rica Petrocelis anastomosans Dawson Dawson, 1961 Salina Cruz, Oaxaca Petrocelis ascendens Dawson Dawson, 1961 Bahia Rosario, Baja Calif. Benthic Algae-— DAWSON 427 Petrocelis frauds cana Setchell & Gardner Dawson, 1954, 1959 'c; Scagel, 1957; Dawson, 1961 Northern Wash, to Los Angeles Co., Calif.; Cabo Colnett, Baja Calif. Petrocelis haematis Hollenberg Hollenberg, 1943 Southern Calif. Petrocelis middendorffii (Kuprecht) Kjellman Scagel, 1957 Bering Sea to Whidbey Is., Wash. fam. NEMASTOMACEAE Scbizymenia epipkytica (Setchell & Lawson) Smith & Hollenberg Smith, 1944; Papenfuss, l94Aa Monterey Peninsula, Calif. Scbizymenia pacifica (Kylin) Kylin Scagel, 1957; Dawson, 1954, 1959c, 1961 Unga L, Alaska, to Isla Magdalena, Baja Calif. Predaea masonii (Setchell & Gardner) J. De Toni Dawson, 1954; Feldmann, 1942; Dawson, 1961 Isla Guadalupe, Baja Calif.; Isla Clarion, Re- villagigedo Arch.; southern Gulf of Calif. Predaea subpeltata Dawson Dawson, 1961 Puerto Guatulco, Oaxaca Platoma fanii Dawson Dawson, 1961 Isla San Esteban, Gulf of Calif. fam. SEBDENIACEAE Sebdenia rubra Taylor Taylor, 1945 Galapagos Arch. fam. SOLIERIACEAE Turner ella mertensiana ( Postels & Ruprecht) Schmitz Scagel, 1957 Shumagin L, Alaska, to northern Wash. Eucbeuma uncinatum Setchell & Gardner Dawson, 1961 Isla Angel de la Guarda, Gulf of Calif., to Mazatlan, Sinaloa Agardhiella tenera (J. Agardh ) Schmitz Dawson, 1959c (as A. coulteri ) , 1961 Queen Charlotte Str., British Columbia, to Revillagigedo Arch.; Northern Gulf of Calif.; Galapagos Arch. Gardneriella tuberifera Kylin Hollenberg, 1948; Dawson, 1961 Monterey Peninsula, Calif., to Punta Baja, Baja Calif. Opuntiella calif ornica ( Farlow ) Kylin Scagel, 1957 Unga L, Alaska, to San Diego, Calif. Reticulobotrys catalinae Dawson Dawson, 1961 Santa Catalina L, Calif.; Isla Guadalupe, Baja Calif. Sarcodiotheca dichotoma ( Howe ) Dawson Dawson, 1961 Puerto Refugio to La Paz, Gulf of Calif. Sarcodiotheca divaricata Taylor Taylor, 1945 Galapagos Arch. Sarcodiotheca furcata (Setchell & Gardner ) Kylin Scagel, 1957; Dawson, 1961 Southern British Columbia to Islas Tres Marias Mexico; Puerto Culebra, Costa Rica; Galapagos Arch. Sarcodiotheca linearis Setchell & Gardner Dawson, 1961 Isla Magdalena to San Lucas, Baja Calif. Sarcodiotheca taylorii Dawson Dawson, 1961 Punta Frailes, Baja Calif.; Islas Tres Marias 428 PACIFIC SCIENCE, Vol. XV, July 1961 Sarcodiotheca tenuis Taylor Taylor, 1945; Dawson, 1961 (as probably equal to S. divaricata ) Galapagos Arch. fam. RHABDONIACEAE Catenella repens ( Lightfoot ) Batters Taylor, 1945 Galapagos Arch.; Esmeraldas, Ecuador Catenella impudica (Montagne) J. Agardh Taylor, 1945 Bahia Buenaventura, Colombia fam. RHODOPHYLLIDACEAE Rhodophyllis dichotoma f. setacea Kjellman Setchell & Gardner, 1903 Unalaska, Alaska FAM. HYPNEACEAE Hypneocolax stellaris f. orientalis Weber van Bosse Dawson, 1961 Bahia Magdalena, Baja Calif. Hypnea boergesenii Tanaka Dawson, 1960*2 Isla Brincanco, Panama Hypnea cervicornis J. Agardh Taylor, 1945; Dawson, 1957, 1959, 1961 Isla Guadalupe, Scammon Lagoon, Bahia Santa Maria, Baja Calif.; San Felipe to Bahia San Lucas, Gulf of Calif.; Guayas, Ecuador; Galapagos Arch. Hypnea johnstonii Setchell & Gardner Dawson, 1958, 1961 Newport Harbor, Calif.; southern Baja Calif.; Gulf of Calif.; Isla San Benedicto; Revilla- gigedo Arch.; to San Bias, Nayarit Hypnea marchantae Setchell & Gardner Taylor, 1945 Isla Gorgona, Colombia; La Libertad, Ecuador Hypnea pannosa J. Agardh Dawson, 1954c, 1951b, 1961 Isla Pond, Gulf of Calif., to Islas Secas, Pa- nama; Galapagos Arch. Hypnea spicifera ( Suhr ) Harvey ? Dawson, 1961 Mazatlan, Sinaloa Hypnea spinella ( C. Agardh ) Kutzing Dawson, 1959 b, 1961 Guaymas, Sonora, to Bahia Petatlan, Guer- rero; Clipperton I. Hypnea valentiae (Turner) Montagne Dawson, 1961 Santa Barbara, Calif., to Guayas, Ecuador Hypnea variabilis Okamura Dawson, 1961 Catalina I.; La Jolla, Calif., to Isla San Roque, Baja Calif. FAM. PLOCAMIACEAE Plocamium coccineum (Hudson) Lyngbye var. coccineum Farlow, 1902 Galapagos Arch. Plocamium coccineum var. pacificum (Kylin) Dawson Dawson, 1957, 1959c; Scagel, 1957; Dawson, Neushul & Wildman, I960, 1960*2 (all as P. pacificum)', Dawson, 1961 Southern British Columbia to Rocas Alijos and Revillagigedo Arch. Plocamium oreganum Doty Scagel, 1957 Southern British Columbia to Sonoma Co., Calif. Plocamium tenue Kylin Scagel, 1957; Doty, 1947 Sitka, Alaska, to South Bay, Oreg. Benthic Algae — DAWSON 429 Plocamium violaceum Farlow Scagel, 1957; Dawson. 1959c, 1961 Southern British Columbia to Islas San Beni- to, Baja Calif. Plocamiocolax pulvinata Setchell Scagel, 1957; Dawson, 1961 Northern Wash, to Cabo Colnett, Baja Calif.; Isla Guadalupe fam. SPHAEROCOCCACEAE Caulacanthus indicus Weber van Bosse Dawson, 1951b Golfo Dulce, Costa Rica ? Caulacanthus ustulatus Kiitzing Dawson, 1961 Isla Cedros to Isla Margarita, Baja Calif.; Isla Turner, Gulf of Calif. Taylorophycus laxa ( Taylor ) Dawson Dawson, 1961 Isla Magdalena, Baja Calif. fam. GRACILARIACEAE Gelidiopsis tenuis Setchell & Gardner Dawson, 1954, 1957£, 1959, 1961 Guaymas, Sonora, to Costa Rica; Revillagi- gedo Arch. Gelidiopsis variabilis ( Greville ) Schmitz Dawson, 1954, 1959, 19 60a, 1961 Guaymas, Sonora, to Bahia Santiago, Colima Gracilaria ascidiicola Dawson Dawson, 1961 Puerto Escondido, Gulf of Calif. Gracilaria brevis Taylor Taylor, 1945 Guayas, Ecuador Gracilaria cerrosiana Taylor Dawson, 1961 Isla Cedros to Punta Abreojos, Baja Calif. Gracilaria crispata Setchell & Gardner Dawson, 1951b, 1961 Isla Tiburon, Sonora, to Costa Rica; Revilla- gigedo Arch. Gracilaria ecuadoreanas (Taylor) Dawson Dawson, 19497? Galapagos Arch. Gracilaria (?) linearis Kylin Smith, 1944; Dawson, 1949^ (as not of the genus Gracilaria) Monterey Peninsula; Point Sur, Calif. Gracilaria marcialana Dawson Dawson, 1961 Bahia Agua Verde; Isla Espiritu Santo, Gulf of Calif. Gracilaria pachydermatica Setchell & Gardner Dawson, 1961 Isla Cedros to Isla Magdalena, Baja Calif.; throughout the Gulf of Calif. Gracilaria ramisecunda Dawson Dawson, 1961 Southern Gulf of Calif. Gracilaria rubrimembra Dawson Dawson, 1961 Vicinity of Guaymas, Sonora Gracilaria rugulosa Montagne ? Farlow, 1902 Galapagos Arch. Gracilaria skottsbergii Taylor Dawson, 19497? Galapagos Arch. Gracilaria spinigera Dawson Dawson, 1961 Guaymas to Puerto Escondido, Gulf of Calif.; Isla Margarita, Baja Calif. Gracilaria subsecundata Setchell & Gardner Dawson, 1961 Throughout the Gulf of Calif, to Bahia To- polobampo, Sinaloa 430 PACIFIC SCIENCE, Vol. XV, July 1961 Gracilaria symmetrica Dawson Dawson, 1961 Bahia Chamela, Jalisco; Bahia Chacahua, Oaxaca; Bahia Santa Elena, Costa Rica Gracilaria tepocensis ( Dawson ) Dawson Dawson, 1961 Northern Gulf of Calif.; Costa Rica Gracilaria textorii (Suringar) J. Agardh var. textorii Dawson, 1961 Vicinity of Punta Eugenio; Punta Pequena, Baja Calif.; Gulf of Calif.; Galapagos Arch. Gracilaria textorii var. cunninghamii (Farlow) Dawson Dawson, 1961; Dawson, Neushul, & Wild- man, I960 (as G. cunninghamii) Santa Barbara, Calif., to Isla Magdalena, Baja Calif. Gracilaria turgida Dawson Dawson, 1961 Newport Harbor, Calif.; Bahia Magdalena, Baja Calif. Gracilaria veleroae Dawson Dawson, 1961 San Diego Co., Calif; northern Gulf of Calif, to Acapulco, Guerrero; Revillagigedo Arch. Gracilaria verrucosa ( Hudson ) Papenf uss Taylor, 1945 (as G. confervoides in part); Scagel, 1957; Dawson, 1961 Southern British Columbia to central Gulf of Calif.; Guayas, Ecuador ? Gracilariopsis under sonii ( Grunow ) Dawson Dawson, 1961; Kylin, 1941 (as Gracilaria under sonii ) Santa Barbara Co., Calif., to Punta Maria, Baja Calif. Gracilariopsis claviformis Dawson Dawson, 1961^ Near Santa Cruz I., Calif. Gracilariopsis costaricensis Dawson Dawson, 1957£, 1961 Bahia Tenacatita, Jalisco; Costa Rica Gracilariopsis megaspora Dawson Dawson, 1961 Bahia San Carlos, Sonora Gracilariopsis panamensis (Taylor) Dawson Dawson, 1949 h Puerto Culebra, Costa Rica; Isla Taboga, Pa- nama; Galapagos Arch. Gracilariopsis rhodotricha Dawson Dawson, 1961 Revillagigedo Arch. Gracilariopsis robusta (Setchell) Dawson Dawson, 1949& Monterey Peninsula, Calif. Gracilariopsis sjoestedtii (Kylin) Dawson Scagel, 1957; Dawson & Beaudette, I960; Dawson, 1961 Vancouver I., British Columbia, to Bahia Salinas, Costa Rica Gracilariophila gardneri Setchell Dawson, 1945c, 1961 Santa Monica to La Jolla, Calif.; Bahia Boco- chibampo, Sonora Gracilariophila oryzoides Setchell & Wilson Doty, 1947; Dawson, 1961 Coos Bay, Oreg., to Bahia Rosario, Baja Calif. fam. DICRANEMACEAE Dicranema rosaliae Setchell & Gardner Dawson, 1961 Throughout the Gulf of Calif, to Puerto Par- ker, Costa Rica FAM. PHYLLOPHORACEAE Phyllophora calif ornica (J. Agardh) Kylin Dawson, 1959c, 1961 San Francisco; Santa Barbara, Calif., to Punta San Hipolito, Baja Calif. Phyllophora clevelandii Farlow Dawson, 1961 Bolinas, Calif., to Socorro, Baja Calif. Benthic Algae — DAWSON 431 Phyllophora suhmaritimus Dawson Dawson, 1961 Cortes Bank, Calif.; Bahia Tortuga, Baja Calif. Petroglossum pacifcum Hollenberg Hollenberg, 1943, 1945; Doty, 1947; Kylin, 1956 (as Phyllophora pacifcum ) Coos Bay, Or eg., to La Jolla, Calif. Petroglossum parvum Hollenberg Dawson, 1959c, 1961 Laguna Beach, Calif., to Isla Cedros, Baja Calif. Ahnfeltia concinna J. Agardh Dawson 1954c, 1961; Scagel, 1957 Southern British Columbia to Oaxaca, Mex- ico; Isla San Benedicto, Revillagigedo Arch. Ahnfeltia durvillaei (Bory) J. Agardh var. durvillaei Taylor, 1945 Galapagos Arch. Ahnfeltia durvillaei var. implicata (Kiitzing) Howe Taylor, 1945 Galapagos Arch. Ahnfeltia plicata (Hudson) Fries Okamura, 1933; Scagel, 1957; Dawson, 1959c, 1961 Bering Sea to Punta Baja, Baja Calif. Ahnfeltia svensonii Taylor Dawson, 1961 Southern Gulf of Calif.; Galapagos Arch. Gymnogongrus ? carnosus Setchell & Gardner Dawson, 1961 Isla San Pedro Martir, Gulf of Calif. Gymnogongrus crustiforme Dawson Dawson, 1961 Salina Cruz, Oaxaca Gymnogongrus griff it hsiae var. galapagensis Piccone & Grunow Taylor, 1945 Galapagos Arch. Gymnogongrus guadalupensis Dawson Dawson, 1961 Isla Guadalupe; near Cabo San Lucas, Baja Calif. Gymnogongrus johnstonii ( Setchell & Gardner) Dawson Dawson, 1961 Gulf of Calif, to Salina Cruz, Oaxaca; Bahia Salinas, Costa Rica; Revillagigedo Arch. Gymnogongrus leptophyllus J. Agardh Doty, 1947; Dawson, 1959c, 1961; Dawson, Neushul & Wildman, I960 Oreg. to Punta Baja, Baja Calif. Gymnogongrus linearis (Turner) J. Agardh Doty, 1947 Cape Perpetua, Oreg., to San Luis Obispo Co., Calif. Gymnogongrus martinensis Setchell & Gardner Dawson, 1961 Isla San Martin to Isla Magdalena, Baja Calif. Gymnogongrus melanothrix Grunow Taylor, 1945 Galapagos Arch. Gymnogongrus norvegicus (Gunner) J. Agardh Scagel, 1957 Southern British Columbia to northern Wash. Gymnogongrus platyphyllus Gardner Scagel, 1957; Dawson, 1959c, 1961 Southern British Columbia to Punta Santa Rosalia, Baja Calif. Gymnogongrus serenei Dawson Dawson, 1961 Cabo San Lucas, Baja Calif.; Isla San Bene- dicto, Revillagigedo Arch. 432 PACIFIC SCIENCE, Vol. XV, July 1961 Gymnogongrus smithii Taylor Taylor, 1945 Galapagos Arch. Gymnogongrus vermicularis J. Agardh Taylor, 1945 Galapagos Arch. Steno gramme interrupta (C. Agardh) Montagne Dawson, 1959c, 1961; Scagel, 1957 (as S. calif ornica ) ; Kylin, 1956 (as S. calif or - nica ) ; Dawson, Neushul & Wildman, I960 Southern British Columbia to Cabo San Lucas, Baja Calif.; Revillagigedo Arch,; Galapa- gos Arch. FAM. GIGARTINACEAE Chondrus (?) albemarlensis Taylor Taylor, 1945 Galapagos Arch. Chondrus canaliculatus Greville Farlow, 1902 Galapagos Arch. Chondrus crispus (Linnaeus) Lyngbye Setchell & Gardner, 1903 Bering Sea to Unalaska, Alaska Chondrus (?) hancockii Taylor Taylor, 1945 Galapagos Arch. Gigartina agardhii Setchell & Gardner Smith, 1944 Fort Ross to La Jolla ?, Calif. Gigartina armata J. Agardh var. armata Dawson, 1951*2, 1961 Southern Calif, to Isla San Martin, Baja Calif. ? Gigartina armata var. echinata (Gardner) Daw- son Dawson, 1961 Santa Catalina I.; La Jolla, Calif.; Isla Gua- dalupe, Baja Calif. Gigartina asperifolia J. Agardh Setchell & Gardner, 1933; Dawson, 1958, 1959c, 1961 Santa Barbara, Calif., to Punta Santa Rosalia, Baja Calif. Gigartina calif ornica J. Agardh Smith, 19 44; Sanborn & Doty, 1946; Doty, 1947; Dawson, 1959c, 1961 Coos Bay, Oreg., to near Punta Maria, Baja Calif. Gigartina canaliculata Harvey Doty, 1947; Dawson, 1959c, 1961; Dawson, Neushul & Wildman, I960 Coos Bay, Oreg., to Isla Magdalena, Baja Calif. Gigartina chauvinii (Bory) Montagne Taylor, 1945 Galapagos Arch. Gigartina corymhifera ( Kiitzing ) J. Agardh Scagel, 1957 (in part as G. hinghamii) ; Daw- son, 1961 Northern Wash, to San Quintin, Baja Calif. Gigartina dichotoma Gardner Setchell & Gardner, 1933 Duxbury Reef to Point Sur, Calif. Gigartina exasperata Harvey & Bailey Scagel, 1957 Southern British Columbia to northern Calif. Gigartina harveyana (Kiitzing) Setchell & Gardner Doty, 1947; Dawson, 1959c, 1961 Oreg. to near Punta Maria, Baja Calif. Gigartina intermedia Suringar Dawson, 1961 Isla Tortuga, Gulf of Calif.; Mazatlan, Sina- loa Gigartina jardinii J. Agardh Setchell & Gardner, 1933 Monterey; Santa Barbara, Calif. Gigartina johnstonii Dawson Dawson, 1961 Northern Gulf of Calif, south to Guaymas Benthic Algae — DAWSON 433 Gigartina latissima ( Harvey ) Eaton Scagel, 1957 Table I., British Columbia, to northern Wash. Gigartina leptorhynchos J. Agardh f. lepto- rhynchos Dawson, 1959c 1961 Santa Cruz, Calif., to Isla Cedros, Baja Calif. Gigartina leptorhynchos f. latissima Dawson Dawson, 1949, 1961 Point Arguello, Calif.; Isla San Martin, Baja Calif. Gigartina leptorhynchos f. cylindrica Dawson Dawson, 1949, 1954, 1961 Southern Calif, and northern Baja Calif. Gigartina lessonii J. Agardh Farlow, 1902 Galapagos Arch. Gigartina macdougalii Dawson Dawson, 1961 Northern Gulf of Calif, south to Isla Tiburon Gigartina mamillosa (Goodenough & Wood- ward) J. Agardh Scagel, 1957 Southern British Columbia to Seal Rocks, Oreg. Gigartina multidichotoma Dawson Dawson, 1961 Santa Catalina I.; La Jolla, Calif.; Punta Des- canso, Baja Calif. Gigartina papillata (C. Agardh) J. Agardh f. papillata Scagel, 1957 (in part as G. cristata ) ; Daw- son, 1959c 1961; Hollenberg, 1948 Northern Wash, to Punta Baja, Baja Calif. Gigartina papillata f. cristata Setchell Scagel, 1957; Dawson, 1959c (both as G. cristata) , Dawson, 1961 Northern Wash, to Bahia Rosario, Baja Calif. Gigartina pectinata Dawson Dawson, 1961 Northern Gulf of Calif. Gigartina sen at a Gardner Setchell & Gardner, 1933; Dawson, 1959c, 1961 Balboa, Calif., to Isla Magdalena, Baja Calif. Gigartina sitchensis Ruprecht Scagel, 1957 Sitka, Alaska, to Whidbey I., Wash. Gigartina spinosa ( Kiitzing ) Harvey Coe, 1932; Smith, 1944; Dawson, 1959c 1961 Monterey, Calif., to Punta Baja, Baja Calif. Gigartina stellata ( Stackhouse ) Batters Scagel, 1957 West Coast of Vancouver I., British Columbia Gigartina tepida Hollenberg Dawson, 1959, 1961 Balboa Harbor, Calif.; Bahia San Quintin, Baja Calif.; northern Gulf of Calif, south to Topolobampo, Sinaloa Gigartina turneri Setchell & Gardner Setchell & Gardner, 1933; Dawson, 1945 h Southern Calif. Gigartina unalas chensis (Ruprecht) Ruprecht Scagel, 1957 Bering Sea to northern Wash. Gigartina volans (C. Agardh) J. Agardh Doty, 1947; Dawson, 1959c, 1961; Dawson, Neushul & Wildman, I960 Coos Bay, Oreg., to near Punta Maria, Baja Calif. Rhodoglossum affine (Harvey) Kylin Scagel, 1957; Dawson, 1959c 1961 Southern British Columbia to Isla San Roque, Baja Calif. Rhodoglossom americanum Kylin Smith, 1944; Dawson, 1945 b, 1958, 1959c 1961 Bolinas, Calif., to near Punta Maria, Baja Calif. 434 PACIFIC SCIENCE, Vol XV, July 1961 Rhodoglossum cobinae Dawson Dawson, 1961 Punta Maria to Isla Magdalena, Baja Calif. Rhodoglossum coriaceum Dawson Dawson, 1958, 1959c, 1961 Santa Cruz I.; Ventura, Calif., to near Punta Maria, Baja Calif. Rhodoglossum difjusum Dawson Dawson, 1961 Isla Angel de la Guarda, Gulf of Calif. Rhodoglossum digital um Dawson Dawson, 1961 Isla Partida; Isla Rasa, Gulf of Calif. Rhodoglossum hancockii Dawson Dawson, 1961 Northern Gulf of Calif. Rhodoglossum latissimum J. Agardh Scagel, 1957 Northern British Columbia to northern Wash. Rhodoglossum linguiforme Dawson Dawson, 1961 Arroyo Hondo, southern Calif., to Punta Cono ( ? ) , Baja Calif. Rhodoglossum parvum Smith & Hollenberg Smith, 1944; Hollenberg, 1948 Monterey; southern Calif. Rhodoglossum toseum (Kylin) G. M. Smith Smith, 1944; Dawson, 19 59c, 1961 Monterey Peninsula, Calif.; Punta Descanso to Punta Baja, Baja Calif. Iridaea agardhiana (Setchell & Gardner) Kylin Smith, 1944; Sanborn & Doty, 1946; Doty, 1947 (all as Iridophycus flaccidum ); Ky- lin, 1941; Dawson, 1959c (as Iridaea flac- cidum ) Cape Arago, Oreg., to Government Point, Calif. Iridaea calif ornica J. Agardh Kylin, 1941 Santa Barbara, Calif. Iridaea cor data (Turner) Bory Scagel, 1957 Banks L, British Columbia to Oreg. Iridaea coriaceum (Setchell & Gardner) Scagel Scagel, 1957 Northern Wash, to Carmel, Calif. Iridaea fulgens ( Setchell & Gardner ) Papenf uss Setchell & Gardner, 1937; Doty, 1947 (both as Iridophycus fulgens ) ; Papenfuss, 1958 Cape Arago, Or eg.; Trinidad, Humbolt Co., Calif. Iridaea furcata (Setchell & Gardner) Papenfuss Papenfuss, 1958 Sitka, Alaska Iridaea heterocarpa Postels & Ruprecht Scagel, 1957; Dawson, 1959 'c Northern British Columbia to Government Point, Calif. Iridaea linear e ( Setchell & Gardner ) Kylin Smith, 1944; Doty, 1947 (both as Iridophy- cus line are ) ; Kylin, 1941; Dawson, 1959c Cape Mears, Oreg., to Ventura, Calif. Iridaea oregana (Doty) Papenfuss Papenfuss, 1958 Brookings, Oreg. Iridaea parksii (Setchell & Gardner) Papenfuss Papenfuss, 1958 Trinidad, Calif. Iridaea parvula (Kjellman) Papenfuss Okamura, 1933 (as Iridaea laminarioides var. parvula ); Papenfuss, 1958 Aleutian Islands to Kodiak L, Alaska Iridaea reediae ( Setchell & Gardner ) Papenfuss Dawson, 1958, 1959c; Papenfuss, 1958 BushnelFs Beach, San Luis Obispo Co., and Solromar, Ventura Co., Calif. Iridophycus sanguineum Setchell & Gardner Kylin, 1941; Dawson, 1945c (both as prob- ably identical with Iridaea calif ornica); Doty, 1947 Cape Arago, Oreg., to Pacific Grove, Calif. Benthic Algae— DAWSON 435 Iridaea sinicola ( Setchell & Gardner ) Papenfuss Papenfuss, 1958 San Francisco Bay, Calif. Iridaea splendens (Setchell & Gardner) Papen- fuss Doty, 1947 (as Iridophycm splendens ); Pa- penfuss, 1958; Dawson, 1958, 1959c Cape Kiawandi, Greg.; Monterey Peninsula and Government Point, Calif. Iridaea whidbeyanum (Setchell & Gardner) Scagel Scagel, 1957 Southern British Columbia to Oreg. Besa papillaeformis Setchell Smith, 1944 Lands End, San Francisco; Monterey, Calif. order RHODYMENIALES fam. RHODYMENIACEAE Bauch ea (?) crispa Taylor Dawson, 1954 Isla Maria Magdalena, Nayarit Bauchea fryeana Setchell Scagel, 1957 Southern British Columbia to northern Wash. Bauchea galapagensis Taylor f. galapagensis Taylor, 1945 Galapagos Arch. Bauchea galapagensis f. pygmaea Taylor Taylor, 1945 Galapagos Arch. Bauchea laciniata J. Agardh var. laciniata Dawson, 1954; Scagel, 1957; Sparling, 1957; Dawson, Meushul & Wildman I960, 1960a; Hollenberg, 1948 (as F, media ) Southern British Columbia to Isla Asuncion, Baja Calif. Bauchea laciniata f. pygmaea Setchell & Gardner Kylin, 1941; Dawson, 1949$ (both as F. pyg- maea)] Dawson, 1954; Dawson, Neushul & Wildman, 1960$; Sparling, 1957 (as F. laciniata) Throughout the southern part of the range of the species Bauchea mollis Howe Dawson, 1954 La Paz, Baja Calif. Bauchea rhizophylla Taylor Taylor, 1945 Galapagos Arch. Bauchea sefferi Howe Dawson, 1954 La Paz, Baja Calif. Baucheocolax attenuata Setchell Scagel, 1957; Sparling, 1957 Northern Wash, to Monterey Peninsula, Calif. Gloioderma conjuncta (Setchell & Gardner) Dawson Dawson, 1954 (as Estebania conjuncta ); Dawson, 1959 Northern Gulf of Calif, south to Isla Tortuga Sciadophycus stellatus Dawson Dawson, 1954; Dawson, Neushul & Wild- man, I960 Southern Calif, to southern Baja Calif. Leptofauchea pacifca Dawson Dawson, 1954 Isla Cedros, Baja Calif. Bryella gardneri ( Setchell ) Kylin Scagel, 1957 Southern British Columbia to Oreg. Botryocladia beaudettei Dawson Dawson, 1960$ Bahia Potrero Grande, Costa Rica Botryocladia chiajeana (Meneghini) Kylin ? Dawson, 1957 Dawson, Neushul & Wildman, 1960 $ Near Punta Eugenio; Rocas Alijos, Baja Calif. 436 Botryocladia banco ckii Dawson Dawson, 1954 Bahia Agua Verde, Gulf of Calif. Botryocladia neushulii Dawson Dawson 1958; Dawson, Neushul & Wildman, I960 Goleta, Calif., to Bahia San Quintln, Baja Calif. Botryocladia pseudodichotoma (Farlow) Kylin var. pseudodichotoma Scagel, 1957; Dawson, 1954; Dawson, Neu- shul & Wildman, I960, I960 a Vancouver I., British Columbia, to Isla Maria Magdalena, Nayarit; Isla Socorro, Revilla- gigedo Arch.; Galapagos Arch. Botryocladia pseudodichotoma var. datilensis Dawson Dawson, 1954 Isla Tiburon; near Guaymas, Gulf of Calif. Botryocladia tenuissima Taylor Taylor, 1945 Galapagos Arch. Botryocladia uvarioides Dawson Dawson, 1954, 1959 Southern Gulf of Calif. Halosaccion glandiforme ( Gmelin ) Ruprecht Scagel, 1957 Westernmost Aleutian Islands, Alaska, to Monterey Peninsula, Calif. Halosaccion ramentaceum (Linnaeus) J. Agardh Setchell & Gardner, 1903 Sannak L to Cook Inlet, Alaska Halosaccion tilesii Kjellman Setchell & Gardner, 1903 Kukak Bay to Wrangel, Alaska Drouetia coales cens ( Farlow ) ' G. De Toni G. De Toni, 1938; Taylor, 1945 (as Herpo- phyllon coales cens ) ; Dawson, 1949 Galapagos Arch. PACIFIC SCIENCE, Vol. XV, July 1961 Drouetia rotata Dawson Dawson, 1949; Dawson, Neushul & Wild- man, I960 (partly as D. peltata by error ) , 1960# Southern Calif. Channel Islands to Baja Calif. Rhodymenia arb ores cens Dawson Dawson, 1954; Dawson, Neushul & Wild- man, I960, 1960# Southern Calif, to Isla Magdalena, Baja Calif. Rhodymenia attenuata Dawson Smith, 1944; Dawson, Neushul & Wildman, I960, 1960# Monterey, Calif., to Isla Asuncion, Baja Calif. Rhodymenia calif ornica Kylin Taylor, 1945; Dawson, 1954, 1949'#, 1951#, 1959, 1959c; Scagel, 1957; Sparling, 1957; Dawson, Neushul & Wildman, I960 Southern British Columbia to Isla Magdalena, Baja Calif.; Isla Carmen, Gulf of Calif.; Galapagos Arch. Rhodymenia dawsonii Taylor Dawson, 1954; Dawson, Neushul & Wild- man, 1960# Bahia Tortuga, Isla Magdalena, Baja Calif. Rhodymenia decumbens Taylor Taylor, 1945 Galapagos Arch. Rhodymenia divaricata Dawson Taylor, 1945; Dawson, 1954 Northern Gulf of Calif.; Galapagos Arch. Rhodymenia flabellifolia Bory Taylor, 1945; Dawson, 1941 Galapagos Arch. Rhodymenia hancockii Dawson Dawson, 1954, 1959 Isla Angel de la Guarda; Bahia Agua Verde, Gulf of Calif. Rhodymenia lobata Dawson Smith, 1954; Dawson, 1954 Monterey, Calif.; near Isla Isabel, Nayarit Benthic Algae — Dawson 437 Rhodymenia lobulifera Dawson Smith, 1944; Dawson, 1959c Bolinas to San Pedro, Calif. Rhodymenia pacifica Kylin Scagel, 1957; Sparling, 1957; Dawson, 1959c; Dawson, Neushul & Wildman, I960, l%0a Northern British Columbia to Punta Cabras, Baja Calif. Rhodymenia palmata (Linnaeus) Greville f. palmata Okamura, 1933; Scagel, 1957 Aleutian Islands, Alaska, to northern Wash. Rhodymenia palmata f. mollis Setchell & Gard- ner Scagel, 1957 Agattu I., Alaska, to Pacific Grove, Calif. Rhodymenia palmata f. sarniensis (Mertens) J. Agardh Setchell & Gardner, 1903 Kodiak Is. to Orca, Alaska Rhodymenia palmetta (Esper) Greville Taylor, 1945 Galapagos Arch. Rhodymenia palmettiformis Dawson Dawson, 1945, 1954, 1959c; Dawson, Neu- shul & Wildman, I960 La Jolla, Calif., to Punta Baja, Baja Calif. Rhodymenia pertusa (Posteis & Ruprecht) J. Agardh Scagel, 1957; Sparling, 1957 Alaska to Coos Bay, Oreg. Rhodymenia rhizoides Dawson Dawson, 1941, 1945, 1959c Lechuza Point to San Diego, Calif. Rhodymenia rosea Dawson Dawson, 1954 Isla Angel de la Guarda, Gulf of Calif. Rhodymenia stipitata Kylin Scagel, 1957 British Columbia; Puget Sound, Wash. Rhodymeniocolax botryoidea Setchell Scagel, 1957; Sparling, 1957 Northern Wash, to San Diego, Calif. fam. LOMENTARIACEAE Binghamia calif or nica J. Agardh Silva, 1952 Santa Barbara to Santa Monica, Calif. Binghamia forkii ( Dawson ) Silva Dawson, 1945; Hollenberg, 1948 (both as Binghamiella) ; Dawson, 1954, 1957 (as Binghamiella by error) ;* Dawson, Neushul & Wildman, I960, I960* Laguna Beach, Calif., to Isla Asuncion; Rocas Alijos, Baja Calif. Lomentaria bailey ana (Harvey) Farlow Dawson, 1954 Bahia Vizcaino, Baja Calif.; Revillagigedo Arch.; Puerto Culebra, Costa Rica Lomentaria caseae Dawson Dawson, 1946 Del Mar, Calif. Lomentaria catenata Harvey Dawson, 1954 Punta Santa Rosalia; Isla Magdalena, Baja Calif.; northern Gulf of Calif. Lomentaria hakodatensis Yendo Dawson, 1954, I960*; Dawson, Neushul & Wildman, 1960^ Isla Guadalupe; Islas San Benito; Laguna de San Ignacio, Baja Calif.; northern Gulf of Calif, to Puerto Culebra, Costa Rica Champia parvula ( C. Agardh ) Harvey Taylor, 1945; Dawson, 1954, 1954c, 1957 b; 1959 Revillagigedo Arch.; Gulf of Calif, to Golfo de Nicoya, Costa Rica; Esmeraldas, Ecua- dor 438 PACIFIC SCIENCE, Vol. XV, July 1961 Gastroclonium coulteri (Harvey) Kylin Scagel, 1957; Dawson, 1954, 1958c; Dawson, Neushul & Wildman, I960 Nootka Sound, British Columbia, to Bahia Asuncion, Baja Calif. Coeloseira compress a Hollenberg Dawson, 1954; Dawson, Neushul & Wild- man, 1960a Pacific Grove, Calif., to Isla Asuncion, Baja Calif. Coeloseira pacifica Dawson Dawson, 1954 Northern Gulf of Calif, south to Guaymas Coeloseira parva Hollenberg Dawson, 1954 Redondo, Calif., to Punta Santa Rosalia, Baja Calif. order CERAMIALES fam. CERAMIACEAE Crouania attenuata (Bonnemaison) J. Agardh Dawson, 1954, 1957 Isla Guadalupe; Bahia Vizcaino; Rocas Alijos, Baja Calif. Lejolisia colomhiana Taylor Taylor, 1945 Isla Gorgona, Colombia Gymnothamnium elegans (Schousboe) J. Agardh Hollenberg, 1948; Dawson, 1954; Taylor, 1945 Southern Calif.; Isla Clarion, Revillagigedo Arch.; Galapagos Arch. Antithamnion alternans Gardner Gardner, 1921b Cook Inlet, Alaska Antithamnion asymmetricum Gardner Gardner, 1927c Sitka, Alaska Antithamniun baylesiae Gardner Doty, 1947 South Slough, near Charleston, Oreg.; Mon- terey Peninsula, Calif. Antithamnion breviramosus Dawson Dawson, 1954, 1957, 1957*, 1959, I960* Santa Catalina I., Calif., to Isla del Rey, Pa- nama; southern Gulf of Calif. Antithamnion defectum Kylin Scagel, 1957 Northern British Columbia to Monterey, Calif. Antithamnion dendroideum Smith & Hollen- berg Smith, 1944 Monterey, Calif. Antithamnion densiusculum Gardner Scagel, 1957 Southern British Columbia to Pacific Grove, Calif. Antithamnion dumontii Dawson Dawson, I960* Isla del Cano, Costa Rica, to Isla Jicaron, Pa- nama Antithamnion floccosum (Muller) Kleen Scagel, 1957 Southern British Columbia to northern Wash. Antithamnion gardneri G. De Toni G. De Toni, 1935; Doty, 1947 (as A. tenuis- simum ) Coos Bay, Oreg.; La Jolla, Calif. Antithamnion glanduliferum Kylin Scagel, 1957; Dawson, 1954d; Dawson, Neu- shul & Wildman, 1960a Southern British Columbia to Isla Asuncion, Baja Calif. Antithamnion kylinii Gardner Scagel, 1957; Dawson, 1954 Victoria, British Columbia, to Punta San Hipolito, Baja Calif. Benthic Algae — D awson 439 Antithamnion mcnahhii Dawson Dawson, 1959 Bahia Agua Verde, Gulf of Calif. Antithamnion nigricans Gardner 'Scagel, 1957 Vancouver I., British Columbia Antithamnion occidental Kylin Taylor, 1945; Scagel, 1957 Alaska to La Jolla, Calif.; Galapagos Arch. Antithamnion pacificum (Harvey) Kylin Dawson, 1954; Scagel, 1957 Yakutat Bay, Alaska, to La Jolla, Calif.; north- ern Gulf of Calif. Antithamnion pulchellum Gardner Gardner, 1927c Santa Monica, Calif. Antithamnion pygmaeum Gardner Doty, 1947; Dawson, 1949; Dawson, Neushul & Wildman, 1960^ Oreg. to Punta Eugenio, Baja Calif. Antithamnion secundatum Gardner Gardner, 1927c San Diego, Calif. Antithamnion setaceum Gardner Gardner, 1927c San Diego, Calif. Antithamnion simulans Gardner Gardner, \921b Sitka, Alaska Antithamnion sublittorale Setchell & Gardner Dawson, 1954 San Jose del Cabo, Baja Calif. Antithamnion suhulatum (Harvey) J. Agardh Scagel, 1957 Southern British Columbia to Monterey, Calif. Antithamnion uncinatum Gardner Scagel, 1957 Northern Wash, to Carmel, Calif. Antithamnion veleroae Taylor Taylor, 1945 Galapagos Arch. Platythamnion heteromorphum (J. Agardh) J. Agardh Doty, 1947 Oreg. to San Pedro, Calif. Platythamnion pectinatum Kylin var. pectinatum Dawson, 1954; Scagel, 1957 Hope I., British Columbia, to Cabo Colnett, Baja Calif. Platythamnion pectinatum var. laxum Taylor Dawson, 1954 Isla Clarion, Revillagigedo Arch. Platythamnion reversum (Setchell & Gardner) Kylin var. reversum Scagel 1957 Northern Wash, to Oreg. Platythamnion reversum var. laxum Taylor Taylor, 1945 Galapagos Arch. Platythamnion tepocensis Dawson Dawson, 1954 Bahia Tepoca, Gulf of Calif. Platythamnion villosum Kylin Dawson, 1954; Scagel, 1957; Dawson, Neu- shul & Wildman, 1960a Sitka, Alaska, to Bahia Tortuga, Baja Calif. Ceramium affine Setchell & Gardner var. affine Taylor, 1945; Dawson, 1954 Isla Guadalupe, Baja Calif.; Isla Angel de la Guarda, Gulf of Calif.; Guayas, Ecuador Ceramium affine var. peninsularis Dawson Dawson, 1954 Corona del Mar, Calif.; Scammon Lagoon, Baja Calif.; Punta Frailes, Gulf of Calif. 440 PACIFIC SCIENCE, Vol. XV, July 1961 Ceramium avalonae Dawson Dawson, 1954, 1960*2 Santa Catalina L, Calif.; Isla Guadalupe, Baja Calif.; Puerto Culebra; Bahia Potrero Grande, Costa Rica Ceramium calif ornicum J. Agardh Dawson, 1954, 1959c; Scagel, 1957 Southern British Columbia to Bahia Magda- lena, Baja Calif. Ceramium camouii Dawson Dawson, 1954, 1957 Punta Banda; Bahia Vizcaino; Rocas Alijos, Baja Calif.; Gulf of Calif. Ceramium caudatum Setchell & Gardner Dawson, 1954, 1959 Balboa Harbor; Santa Catalina I., Calif.; Isla Guadalupe Baja Calif.; Gulf of Calif. Ceramium clarionense Setchell & Gardner Dawson, 1954, 1957 Balboa Harbor, Calif., to Isla Cedros; Rocas Alijos, Baja Calif.; Isla Jorge, Gulf of Calif, to Miramar, Nayarit; Revillagigedo Arch. Ceramium codicola J. Agardh Dawson, 1954; Scagel, 1957 Sitka, Alaska, to Bahia Tortuga, Baja Calif. Ceramium eatonianum (Farlow) J. B. De Toni Doty, 1947: Dawson, 1954, 1959c Coos Bay, Oreg. to Isla Magdalena, Baja Calif. Ceramium equisetoides Dawson Dawson, 1954, 1951b Balboa Harbor, Calif.; Bahia Tepoca, Sonora to Mazatlan, Sinaloa; Golfo de Nicoya, Costa Rica Ceramium evermannii Setchell & Gardner Dawson, 1954 Isla Guadalupe, Baja Calif. Ceramium fmbriatum Setchell & Gardner Dawson, 1954, 1959 Punta Banda to Cabeza Ballena, Baja Calif.; Isla Angel de la Guarda, Gulf of Calif, to Acapulco, Guerrero Ceramium gardneri Kylin Doty, 1947; Dawson, 1950<7 Coos Bay, Oreg., to Santa Cruz I. and Ventura, Calif. Ceramium gracillimum var. byssoideum (Har- vey) G. Mazoyer Dawson, 1950*7, 1954 (both as C. masonii ), 1956, 1957&, 1959; Taylor, 1945 (as C. byssoideum ) Corona del Mar, Calif., to Cabeza Ballena, Baja Calif.; Gulf of Calif.; Costa Rica; Galapagos Arch. Ceramium hamatispinum Dawson Dawson, 1954 Miramar, Nayarit Ceramium hoodii Taylor Taylor, 1945 Galapagos Arch. Ceramium horridum Setchell & Gardner Dawson, 1954 Gulf of Calif. Ceramium howellii Setchell & Gardner Setchell 8c Gardner, 1937; Dawson, 1957 Galapagos Arch. Ceramium marsh aliens e Dawson Dawson, 1951b Golfo de Nicoya, Costa Rica Ceramium mazatlanense Dawson Dawson, 1954, 1954c, 1957£, 1959 b Kino, Sonora, to Salina Cruz, Oaxaca; Golfo Duke, Costa Rica; Isla San Benedicto, Re- villagigedo Arch.; Clipperton I. ? Ceramium nakamurai Dawson Dawson, \951b Golfo de Nicoya, Costa Rica Ceramium obesum Dawson Dawson, 1954 Bahia Agua Duke, Isla Tiburon, Sonora Benthic Algae— -D awson 441 Ceramium ornatum Setchell & Gardner Dawson, 1954 Isia Guadalupe, Baja Calif. Ceramium pacificum (Collins) Kylin Scagel, 1957; Dawson, 1954, 1959 c\ Dawson, Neushul & Wildman, I960 Northern British Columbia to Punta Maria, Baja Calif. Ceramium paniculatnm Okamura Dawson, 1954, 1959 Guaymas, Sonora, to Mazatlan, Sinaloa Ceramium personatum Setchell & Gardner Taylor, 1945; Dawson, 1954 Isla Guadalupe; vicinity of Punta Santa Ro- salia, Baja Calif.; Puerto Parker, Costa Rica ? Ceramium procumbens Setchell & Gardner Hollenberg, 1948; Dawson, .1954, 1959, 1960^ Southern Calif, to Punta Santa Rosalia, Baja Calif.; Gulf of Calif.; Bahia Ballena, Costa Rica Ceramium recticorticum Dawson Dawson, 1954 Guaymas, Sonora; Mazatlan, Sinaloa Ceramium rubrum (Hudson) C. Agardh Scagel, 1957 Bering Sea to northern Wash. Ceramium serpens Setchell & Gardner Dawson, 1949*, 1954, 1957 Gulf of Calif.; Clipperton I. ?; Galapagos Arch. ? Ceramium sinicola Setchell & Gardner var. j ini- cola Dawson, 1954, 1957, 1959 Southern Calif, and Pacific Baja Calif.; Rocas Alijos; Isla San Benedicto, Revillagigedo Arch.; northern Gulf of Calif. Ceramium sinicola var. interrupta (Setchell & Gardner) Dawson Hollenberg, 1948; Dawson, 1954, 1959 Same range as the species Ceramium sinicola var. johnstonii ( Setchell & Gardner) Dawson Dawson, 1954 Same range as the species Ceramium strictum Harvey Dawson, 1950c (as probably referable to C. gardneri) ; Scagel, 1957 Southern British Columbia Ceramium taylorii Dawson Taylor, 1945 (as C. fasti giatum ) ; Dawson, 1954, 1954c, 1957 b, 1959 Laguna Beach, Calif., to Bahia Magdalena, Baja Calif.; Gulf of Calif, to Acapulco, Guerrero; Isla San Benedicto, Revillagi- gedo Arch.; Golfo de Nicoya, Costa Rica; Galapagos Arch. ? Ceramium templetonii Setchell & Gardner Setchell & Gardner, 1937; Dawson, 1957 Galapagos Arch. Ceramium v a gab unde Dawson Dawson, 1954c (as Ceramium sp.) , 1957*, 1957 b Isla San Benedicto, Revillagigedo Arch.; Golfo Duke, Costa Rica Ceramium viscainoense Dawson Dawson, 1954 Punta Santa Rosalia, Baja Calif. Ceramium zacae Setchell & Gardner Dawson, 1954, 1957, 1959; Dawson, Neu- shul & Wildman, 1960a Southern Calif, to Rocas Alijos; Punta Abre- ojos, Baja Calif.; Bahia Agua Verde, Gulf of Calif. Centroceras helium Setchell & Gardner Dawson, 1954 Guaymas, Sonora 442 Centro cer as clavulatum (C Agardh) Montagne var. clavulatum Dawson, 1954, 1954c, 1957, 1957&, 1959, 1959c; Taylor, 1945 Santa Cruz, Calif., to Panama; Galapagos Arch. Centroceras clavulatum var. inerme (Kiitzing) Piccone Dawson, 1957 Galapagos Arch. Centroceras minutum Yamada Dawson, I960* Isla Brincanco, Panama Micro cladia borealis Ruprecht Okamura, 1933; Scagel, 1957 Aleutian Islands, Alaska, to San Simeon, Calif. Microcladia calif ornica Farlow Smith, 1944 San Francisco to San Diego, Calif. Microcladia coulteri Harvey Dawson, 1954, 1959c, Scagel, 1957; Dawson, Neushul & Wildman, I960 Vancouver L, British Columbia, to Punta Santa Rosalia, Baja Calif. Neoptilota asplenioides ( Esper ) Kylin Kylin, 1956; Scagel, 1957 (as Ptilota aspleni- oides ) Yakutat Bay, Alaska, to Puget Sound, Wash. Neoptilota calif ornica (Ruprecht) Kylin Kylin, 1956; Scagel, 1957 (as Ptilota Cali- fornia ) Northern British Columbia to San Diego, Calif. Neoptilota densa (C. Agardh) Kylin Kylin, 1956; Dawson, 1954 (as Plumaria densa); Smith, 1944; Dawson, 1945& both as Ptilota densa ) Tomales Bay, Calif., to Punta Baja, Baja Calif. Neoptilota hypnoides ( Harvey ) Kylin Kylin, 1956; Scagel, 1957 (as Ptilota hyp- noides ) Sitka, Alaska, to San Luis Obispo Co., Calif. PACIFIC SCIENCE, Vol. XV, July 1961 Ptilota filicina (Farlow) J. Agardh Kylin, 1956; Scagel, 1957; Dawson, 1954 (as Plumaria filicina); Dawson, 1949*, 1959c ( both as Ptilota filicina ) Bering Sea to Cabo Colnett, Baja Calif. Ptilota pectinata (Gunner) Kjellman Kylin, 1956; Scagel, 1957 Bering Sea to Puget Sound, Wash. Ptilota tenuis Kylin Kylin, 1956; Scagel, 1957 Northern British Columbia to northern Wash. Spyridia filamentosa (Wulfen) Harvey Kylin, 1941; Taylor, 1945; Dawson, 1954, 1957b Southern Calif, to Panama W rangelia argus (Montagne) Montagne Dawson, I960* Punta Naranja, Panama Callithamnion acutum Kylin Scagel, 1957 Northern Wash. Callithamnion arborescens Gardner Gardner, 1927 h Sitka, Alaska Callithamnion biseriatum Kylin Scagel, 1957 Northern British Columbia to Monterey, Calif. Callithamnion hisporum Gardner Scagel, 1957 Northern Wash. Callithamnion breviramosum Gardner Gardner, 1927c La Jolla, Calif. Callithamnion byssoides Arnott Dawson, 1954, I960* Isla Guadalupe, Baja Calif. Benthic Algae — DAWSON 443 Callithamnion calif ornicum Gardner Smith, 1944 Monterey Peninsula, Calif. Callithamnion ecmdoreanum Taylor Taylor, 1945 Galapagos Arch. Callithamnion endovaginum Setchell & Gardner Kylin, 1941; Dawson, 1944, 1954 La Jolla, Calif.; Guaymas, Sonora Callithamnion epiphyticum Taylor Taylor, 1945 Galapagos Arch. Callithamnion laxum Setchell & Gardner Scagel, 1957 Southern British Columbia and northern Wash.; Greg. Callithamnion marshallense Dawson Dawson, 196(H Bahia Uvita, Costa Rica Callithamnion pacijicum Taylor Dawson, 1954 Isla Socorro, Revillagigedo Arch. Callithamnion paschale Bprgesen Dawson, 1957, 1959 Isla San Benedicto, Revillagigedo Arch. ?; Gulf of Calif. Callithamnion pikeanum Harvey var. pikeanum Scagel, 1957; Dawson, 1959 c Vallenar Point, Alaska, to Point Dume, Calif. Callithamnion pikeanum var. pad f cum (Har- vey) Setchell & Gardner Scagel, 1957 Shumagin L, Alaska, to Greg.; Point Mugu, Calif. Callithamnion ramosissimum Gardner Gardner, 1927c; Dawson, 1949 Santa Catalina L, La Jolla, Calif. Callithamnion mpicolum Anderson f. mpicolum Smith, 1944; Dawson, 1954 Fort Ross, Calif., to Isla Magdalena, Baja Calif. Callithamnion mpicolum f. pygmaeum Collins in Collins, Holden & Setchell Collins, Holden & Setchell, 1911 (P.B.A. 1797); Dawson, 1949; Taylor, 1945 (as possibly C. mpicolum f . pygmaeum ) Santa Catalina L, Calif.; Galapagos Arch. Callithamnion socorriense Taylor Dawson, 1954 Isla Socorro, Revillagigedo Arch. Callithamnion squarrulosum Harvey Harvey, 1853; J. Agardh, 1876 Golden Gate ( San Francisco ) Calif. Callithamnion varispiralis Dawson Dawson, 1954 Santa Catalina L, Calif.; Bahia Vizcaino, Baja Calif. Haloplegma mexicanum Taylor Dawson, 1954 Isla Maria Magdalena, Nayarit Pleonosporium ahysicola Gardner Scagel, 1957 Near Friday Harbor, Wash. Pleonosporium complanatum Taylor Taylor, 1945 Galapagos Arch. Pleonosporium dasyoides (J. Agardh ) J. B. De Toni Dawson, 1954, 1959c Tomales Bay, Calif., to Punta Descanso, Baja Calif. Pleonosporium globulifemm Levring Dawson, 1957^, 1960a Puerto Guatulco, Oaxaca; Golfo de Nicoya, Costa Rica 444 PACIFIC SCIENCE, Vol. XV, July 1961 Pleonosporium kobayashii Okamura Okamura 1933 Atka I., Aleutian Islands, Alaska Pleonosporium polycarpum Gardner Gardner, 1921b La Jolla, Calif. Pleonosporium pygmaeum Gardner Gardner, 1921b La Jolla, Calif. Pleonosporium squarrosum Kylin var. squar- rosum Scagel, 1957 Southern British Columbia to Northern Wash. Pleonosporium squarrosum var. ob ov atum Gardner Scagel, 1957 Sidney, Vancouver I., British Columbia Pleonosporium vancouverianum (J. Agardh) J. Agardh Scagel, 1957 Northern British Columbia to Monterey, Calif. Grifjithsia ( ? ) anthericephala Dawson Dawson, 1954 Isla Guadalupe, Baja Calif. Grifjithsia multiramosa (Setchell & Gardner) Taylor var. multiramosa Dawson 1945£, 1954 San Diego, Calif.; Isla Guadalupe; Scammon Lagoon; San Ignacio Lagoon, Baja Calif.; Gulf of Calif. Grifjithsia multiramosa var. balboensis Hollen- Berg Hollenberg, 1945 Balboa Harbor, Calif. Grifjithsia multiramosa var. minor Taylor Dawson, 1954 Punta Gorda, Baja Calif, del Sur Grifjithsia pacifca Kylin Scagel, 1957; Dawson, 1954; Dawson, Neu- shul & Wildman, I960 Southern British Columbia to southern Baja Calif.; northern Gulf of Calif.; Revillagi- gedo Arch. ?; Galapagos Arch. Grifjithsia tenuis C. Agardh Dawson, 1954, 1959 Scammon Lagoon, Baja Calif.; Punta Penasco, Sonora, to Mazatlan, Sinaloa Spermoth amnion phycophilum Taylor Taylor, 1945 Galapagos Arch. Spermothamnion snyderae Farlow Dawson, 1954, 1959c; Dawson, Neushul & Wildman, I960 Bolinas, Calif., to Isla Magdalena, Baja Calif.; northern Gulf of Calif. Tiffaniella saccorhiza (Setchell & Gardner) Doty & Menez Doty & Menez, I960; Dawson, 1954 (as Spermothamnium saccorhiza ) Isla Guadalupe; Bahia Vizcaino, Baja Calif. fam. DELESSERIACEAE Caloglossa leprieurii (Montagne) J. Agardh Taylor, 1945; Dawson, 1951b Golfo Dulce, Costa Rica; Bahia Buenaven- tura, Colombia; Galapagos Arch.; Bahia San Francisco, Ecuador Branchioglossum undulatum Dawson Dawson, 1949, 1960^; Dawson, Neushul & Wildman, I960, 1960^ Santa Rosa I., Calif., to Punta Asuncion, Baja Calif.; Bahia Santiago, Colima Branchioglossum woodii (J. Agardh) Kylin Dawson, 1954, 1957, 1959; Scagel, 1957 Vancouver L, British Columbia, to Bahia Viz- caino, Baja Calif.; Rocas Alijos; Gulf of Calif. Benthic Algae— DAWSON 445 Hypoglossum abyssicolum Taylor Dawson, 1954 Islas Las Tres Marias, Nayarit; Puerto Cub ebra, Costa Rica; Isla Gorgona, Colombia; Galapagos Arch. Hypoglossum. attenuatum Gardner Dawson, 1954 Isia Guadalupe; San Ignacio Lagoon, Baja Calif.; northern Gulf of Calif. Hypoglossum retusum Dawson Dawson, 1954; Dawson, Neushul & Wild- man, 1960$ Isla Cedros; near Punta Eugenio, Baja Calif. Hemineura (?) howellii Taylor Taylor, 1945 Galapagos Arch. Membranoptera denticulata (Montagne) Kylin Kylin, 1924; Setchell & Gardner, 1903 (as Pteridinm spinulosum ) Bering Sea to Morzhovoi Bay, Alaska Membranoptera dimorpha Gardner Scagel, 1957 Queen Charlotte L, British Columbia, to Coos Bay, Oreg. Membranoptera multiramosa Gardner Doty, 1947 Oreg. to Monterey, Calif. Membranoptera platyphylla (Setchell & Gard- ner) Kylin Scagel, 1957 Southern British Columbia to Oreg. Membranoptera setchellii Gardner Gardner, 1926 Morzhovoi Bay, Alaska Membranoptera spatulata Dawson Dawson ,1954 Isla Patos, Sonora Membranoptera tenuis Kylin Scagel, 1957 Northern Wash. Membranoptera weeksiae Setchell & Gardner Scagel, 1957 Northern Wash.; San Francisco to San Diego, Calif. Pantoneura baerii (Ruprecht) Kylin Kylin, 1924 Sitka, Alaska Pantoneura juergensii (J. Agardh) Kylin Kylin, 1924 Bering Sea to Unalaska, Alaska Holme sia calif ornica (Dawson) Dawson Scagel, 1957 Northern Wash.; San Diego, Calif. Schizoseris pygmaea Dawson Dawson, 1954, 1959 Northern Gulf of Calif. Delesseria decipiens J. Agardh Scagel, 1957 Prince William Sound, Alaska, to Carmel, Calif. Delesseria banco ckii Taylor Taylor, 1945 Galapagos Arch. Platysiphonia clevelandii ( Farlow ) Papenfuss Scagel, 1957 Northern Wash, to San Diego, Calif. Platysiphonia parva Silva & Cleary Silva & Cleary, 1954; Dawson, 1954 Isla Guadalupe, Baja Calif. Taenioma perpusillum J. Agardh Dawson, 1954, 1959, 1960$ Gulf of Calif.; Isla Brincanco, Panama Erythroglossum calif ornicum ( j. Agardh) J. Agardh Doty, 1947; Dawson, 1958 Oreg.; Monterey to Ventura, Calif. Erythroglossum intermedium ( J. Agardh ) Kylin Scagel, 1957 Vancouver I., British Columbia 446 PACIFIC SCIENCE, Vol. XV, July 1961 Sorella delicatula (Gardner) Hollenberg var. delicatula Hollenberg, 194$ San Pedro, Calif. Sorella delicatula var. calif ornica Hollenberg Hollenberg, 194$ Southern Calif. Sorella divaricata (Setchell & Gardner) Hollen- berg Hollenberg, 194$; Smith, 1944 (as Erythro- glossum divaricata ) Pacific Grove, Calif. Sorella pinnata Hollenberg Dawson, 1954 Southern Calif.; Islas San Benito, Baja Calif.; Gulf of Calif. Grinnellia lanceolata Dawson Dawson, 1954, 1959c Punta Gorda, Baja Calif, del Sur Polyneura latissima (Harvey) Kylin Scagel, 1957; Dawson, 1954, 1959c; Dawson, Neushul & Wildman, I960, 1960^ Northern British Columbia to Punta Eugenio, Baja Calif. Polyneurella banco ckii Dawson Dawson, 1954 Isla Angel de la Guarda, Gulf of Calif. Nienburgia andersoniana (J. Agardh) Kylin Doty, 1947; Dawson, 1954, 1959c; Dawson, Neushul & Wildman, I960 Oreg.; Santa Cruz, Calif., to Isla Magdalena, Baja Calif. Nienburgia borealis (Kylin) Kylin Scagel, 1957 Southern British Columbia to northern Wash. Phycodrys ambigua Gardner Gardner, Y)21a Sitka, Alaska Phycodrys bullata Gardner Gardner, 1927^ Agattu I., to Sitka, Alaska Phycodrys elegans Setchell & Gardner Setchell & Gardner, 1937; Taylor, 1945 Galapagos Arch. Phycodrys pule hr a Taylor Taylor, 1945 (as probably equal to P. elegans ) Galapagos Arch. Phycodrys setchellii Skottsberg Doty, 1947; Dawson, 1954, 1959c Oreg. to Isla Cedros, Baja Calif. Pseudophy codry s rainosukei Tokida Okamura, 1933 Atka I., Aleutian Islands, Alaska Polycoryne gardneri Setchell Scagel, 1957 Northern Wash, to Monterey, Calif. Polycoryne phycodricola Dawson Dawson, 1954 Isla Cedros, Baja Calif. Haraldia pro strata Dawson, Neushul & Wild- man Dawson, Neushul & Wildman, 1960*2 Islas San Benito, Baja Calif. Anisocladella pacipca Kylin Smith, 1944; Hollenberg, 1948; Dawson, 1954, 1959c; Dawson, Neushul & Wild- man, I960 Santa Cruz, Calif., to Bahia Asuncion, Baja Calif. Myriogramme caespitosa Dawson Dawson, 1949; Dawson, Neushul & Wild- man, I960, 1960^ Santa Rosa I., Calif., to Isla Magdalena, Baja Calif. Myriogramme divaricata Dawson Dawson, 1954 Isla Angel de la Guarda, Gulf of Calif. Benthic Algae — D awson 447 Myriogramme hollenbergii Kylin Smith, 1944; Dawson, 1954, 1958 Monterey; La Jolla, Calif.; Isla Magdalena, Baja Calif. Myriogramme kylinii Taylor Taylor, 1945 Galapagos Arch. Myriogramme osorioi Dawson Dawson, 1954 Isla Patos; Isla Partida, northern Gulf of Calif. Myriogramme pulchra Gardner Scagel, 1957 Northern Wash. Myriogramme repens Hollenberg Hollenberg, 1945 Point Vicente, Calif. Myriogramme spectabilis ( Eaton ) Kylin Dawson, 1954; Scagel, 1957 Northern Wash, to Socorro, Baja Calif. Acrosorium (?) fragile Taylor Taylor, 1945 Galapagos Arch. Acrosorium papenfussii Taylor Taylor, 1945 Galapagos Arch. Acrosorium uncinatum (Turner) Kylin Kylin, 1941; Dawson, 1954, 1959c; Dawson, Neushul & Wildman, I960 Southern Calif, to Bahia Santa Maria, Baja Calif. Nitophyllum divaricatum Taylor Taylor, 1945 Galapagos Arch. Nitophyllum galapagense Taylor Taylor, 1945 Galapagos Arch. Nitophyllum mirabile Kylin Scagel, 1957 Southern British Columbia to northern Wash. Hymenena cuneifolia Doty Doty, 1947 Coos Bay, Oreg Hymenena flab elliger a (J. Agardh) Kylin Scagel, 1957 Southern British Columbia to Carmel, Calif. Hymenena kylinii Gardner Doty, 1947 Oreg.; San Francisco to Monterey, Calif. Hymenena multiloba (J. Agardh) Kylin Doty, 1947 Oreg. to Carmel, Calif. Hymenena setchellii Gardner Scagel, 1957 Hope I., British Columbia, to Carmel, Calif. Hymenena smithii Kylin Doty, 1947 Oreg.; Carmel, Calif. Cryptopleura brevis Gardner Doty, 1947 Oreg.; Moss Beach, Calif. Cryptopleura corallinara (Nott) Gardner Dawson, 1954, 1957, 1959c; Dawson, Neu- shul & Wildman, 1960a San Diego, Calif., to Isla Magdalena; Rocas Alijos, Baja Calif. Cryptopleura crispa Kylin Kylin, 1941; Doty, 1947; Dawson, 1954, 1959c; Dawson, Neushul & Wildman, 1960 Oreg.; Ventura Co., Calif., to Isla Magdalena, Baja Calif. Cryptopleura dichotoma Gardner Gardner, 1927 San Pedro, Calif. 448 PACIFIC SCIENCE, Vol XV, July 1961 Cryptopleura lobulifera (J. Agardh ) Kylin Smith, 1944; Dawson, 1949, 1954, 1959c Tomales Bay, Calif., to Bahia Santa Maria, Baja Calif. Cryptopleura ruprechtiana (J. Agardh) Kylin Scagel, 1957 Sitka, Alaska, to northern Calif. Cryptopleura spatulata Gardner Gardner, 1927; Dawson, 196(k Santa Barbara; San Pedro, Calif.; Islas Todos Santos, Baja Calif. Cryptopleura violacea (J. Agardh) Kylin Scagel, 1957; Dawson, 1954, 1959c; Dawson, Neushul & Wildman, I960 Southern Vancouver I., British Columbia, to near Punta Maria, Baja Calif. Botryoglossum jarlowianum (J. Agardh) J. B. De Toni Scagel, 1957 Dawson, 1954, 1959c; Dawson, Neushul & Wildman, I960, 1960a Puget Sound, Wash., to Islas San Benito, Baja Calif. Gonimophyllum skottsbergii Setchell Scagel, 1957 Friday Harbor, Wash., to San Diego, Calif. FAM. DASYACEAE Dasya abyssicola Dawson Dawson, 1949 San Clemente Is., Calif. Dasya calif or nica Gardner Dawson, 1945 b, 1954 Southern Calif, to Isla Magdalena, Baja Calif. Dasya eastwoodae Setchell & Gardner Dawson, 1954 Isla Guadalupe, Baja Calif. Dasya pedicellata C. Agardh Dawson, 1954, 1957, 1959 Rocas Alijos; central Baja Calif.; northern Gulf of Calif. Dasya sinicola ( Setchell & Gardner ) Dawson Dawson, 1959 Southern Gulf of Calif. Dasya stanfordiana Farlow Taylor, 1945; Dawson, 1954 Isla Guadalupe, Baja Calif.; Islas Tres Marias, Nayarit; Galapagos Arch. Rhodoptilum densum (G. M. Smith) comb, nov. Smith, in Smith & Hollenberg, 1943; 217; Smith, 1944; Hollenberg, 1948 (all as Dasyopsis densa ) Santa Cruz, Calif., to southern Calif. Rhodoptilum plumosum (Harvey & Bailey) Kylin Kylin, 1956; Scagel, 1957 (as Dasyopsis plu- mosa) Queen Charlotte Sound, British Columbia, to northern Calif. Pogonophorella calif ornica (J. Agardh) Silva Hollenberg, 1948; Dawson, 1945 b (both as Pogonophora) , 1954, 1959c; Dawson, Neu- shul & Wildman, I960 Santa Barbara, Calif., to Bahia Asuncion, Baja Calif. Heterosiphonia asymmetria Hollenberg Hollenberg, 1945 Corona del Mar, Calif. Heterosiphonia densiuscula Kylin Scagel, 1957 Northern Wash. Heterosiphonia erecta Gardner, em. Setchell & Gardner Dawson, 1949, 1954, 1957; Taylor, 1945 Santa Monica, Calif., to Rocas Alijos; Isla Magdalena, Baja Calif.; Guayas, Ecuador Heterosiphonia laxa Kylin Scagel, 1957 Southern British Columbia to northern Wash. Heterosiphonia wurdemannii var. laxa Bprgesen Dawson, 1957&, 1959 Isla Espiritu Santo, Gulf of Calif.; Golfo de Nicoya, Costa Rica Benthic Algae — Dawson 449 fam. RHODOMELACEAE Poly sip honia acuminata Gardner Hollenberg, 1942^; Smith, 1944 Monterey to La Jolla, Calif, Poly sip honia bajacali Hollenberg Hollenberg, 1961 Isla Guadalupe, Baja Calif. Polysiphonia heaudettei Hollenberg Hollenberg, 1961 Isla Guadalupe; Scammon Lagoon, Baja Calif.; Isla Socorro, Revillagigedo Arch.; Isla Grande, Guerrero; Bahia Potrero Grande, Costa Rica Polysiphonia bifurcata Hollenberg Taylor, 1945; Hollenberg, 1961 Scammon Lagoon, Baja Calif.; Playa Blanca, Costa Rica; Galapagos Arch. Polysiphonia brodiaei (Dillwyn) Greville Hollenberg, 1944; Smith, 1944 Sausalito to Santa Monica, Calif. Polysiphonia confusa Hollenberg Hollenberg, 1961; Dawson, 1959c (as P. in- conspicua ) Corona del Mar, Calif., to Desembarcadero de Miller, Baja Calif. Polysiphonia decussata Hollenberg Hollenberg, 1961; Taylor, 1945 Topanga Canyon, Calif., to near Punta Malar- rino, Baja Calif.; Isla Guadalupe; Gala- pagos Arch. Polysiphonia ferulacea Suhr Dawson, 1957 Clipperton I. Polysiphonia flaccidissima Hollenberg var. flac- cidissima Hollenberg, 1961 Laguna Beach, Calif., to Panama Polysiphonia flaccidissima var. smithii Hollen- berg Hollenberg, 1942^; Smith, 1944 San Francisco to San Diego, Calif. Polysiphonia guadalupensis Setchell & Gardner Hollenberg, 1961 Isla Guadalupe, Baja Calif. Polysiphonia banco ckii Dawson Hollenberg, 1961 Cabeza Ballena to San Jose del Cabo, Baja Calif, del Sur Polysiphonia hendryi Gardner var. hendryi Doty, 1947; Hollenberg, 1961 Cape Arago, Oreg., to Isla Cedros, Baja Calif. Polysiphonia hendryi var. gardneri ( Kylin ) Hollenberg Scagel, 1957; Dawson, 1959c (both as P, col- linsii) ; Hollenberg, 1961 British Columbia to Cabo San Lucas, Baja Calif. Polysiphonia hendryi var. compacta (Hollen- berg) Hollenberg Hollenberg, 1961 Southern Calif, and northern Baja Calif. Polysiphonia hendryi var. deliquescens ( Hollen- berg) Hollenberg Scagel, 1957 (as P. collinsii var. deliques- cens) >; Hollenberg, 1961 Prince William Sound, Alaska, to Oreg. Polysiphonia hendryi var. luxurians (Hollen- berg) Hollenberg Scagel, 1957 (as P. collinsii var. luxurians ; Hollenberg, 1961 Northern Wash. Polysiphonia homia Setchell & Gardner Hollenberg, 1961 Isla Guadalupe, Baja Calif. Polysiphonia howei Hollenberg Taylor, 1955; Hollenberg, 1958^ Bahia de Panama, Panama; Bahia Cobita, Colombia 450 PACIFIC SCIENCE, Vol. XV, July 1961 Polysiphonia indigena Hollenberg Hollenberg, 1958^, 1944 (as P. dichotoma ) Santa Cm 2 ; San Diego, Calif. Polysiphonia johnstonii Setchell & Gardner var. johnstonii Dawson, Neushul & Wildman, 1960a; Hob lenberg, 1961 Santa Catalina L, Calif., to Topolobampo, Sinaloa; Gulf of Calif. Polysiphonia johnstonii var. concinna (Hollen- berg) Hollenberg Hollenberg, 1961 La Jolla, Calif., to Mazatlan, Sinaloa; Gulf of Calif. Polysiphonia macounii Hollenberg Scagel, 1957 Southern British Columbia Polysiphonia masonii Setchell & Gardner Hollenberg, 1961 Isla Guadalupe, Baja Calif. Polysiphonia mollis Hooker & Harvey Dawson, 1959c; Scagel, 1957 (both as P. sny- derae); Hollenberg, 1961; Dawson, 1957; 1957 b, 1959$; Dawson, Neushul & Wild- man, I960 Southern British Columbia to central Mexico Polysiphonia nathanielii Hollenberg Hollenberg, 1961 Santa Monica, Calif., to Playa Rosarita, Baja Calif. Polysiphonia pacified var. pacifica Doty, 1947; Dawson, 1959c; Hollenberg, 1961 Sitka, Alaska, to Islas Coronados, Baja Calif. Polysiphonia pacifica var. delicatula Hollenberg Hollenberg, 1961 Monterey, Calif., to Bahia Cahuacan, Chi- apas; central Gulf of Calif. Polysiphonia pacifica var. determinata Hollen- berg Scagel, 1957 Alaska to central Calif. Polysiphonia pacifica var. distans Hollenberg Scagel, 1957 Southern British Columbia to Santa Cruz, Calif. Polysiphonia pacifica var. disticha Hollenberg Hollenberg, 1942*; Scagel, 1957; Segi, 1951 (as P. abscissa) Vancouver I., British Columbia, to central Calif. Polysiphonia pacifica var. gracilis Hollenberg Scagel, 1957 Southern British Columbia to central Calif. Poly sip honia^paniculata Montague Scagel, 1957; Dawson, 1959c; Hollenberg, 1961; Dawson, Neushul & Wildman, 1960a Port Holmes, British Columbia, to Punta Baja, Baja Calif.; northern Gulf of Calif. Polysiphonia savatieri Hanot Hollenberg, 1961; Dawson, 1943* (as P minutissima ) Santa Catalina L, Calif., to Isla Guadalupe; Punta Banda, Baja Calif. Polysiphonia senticulosa Harvey Scagel, 1957 Southern British Columbia to Monterey, Calif. Polysiphonia simplex Hollenberg Hollenberg, 1961; Dawson, 1959c Southern Calif, to Costa Rica; Gulf of Calif.; Revillagigedo Arch. Polysiphonia sonorensis Hollenberg Hollenberg, 1961 Guaymas; Empalme, Sonora Polysiphonia subtilissima Montagne (fide Tseng) Dawson, 1957*, 1959 b Costa Rica; Clipperton I. Benthic Algae— Dawson 451 Polysiphonia urceolata (Lightfoot) Greville Okamura, 1933; Segi, 1951; Scagel, 1957 Aleutian L, Alaska, to northern Wash. Tayloriella dictyums (J. Agardh) Kylin Kylin, 1956; Dawson, 1954 (as Bryocladia dictyums ) Bahia Petatian, Guerrero; Pochetti, Oaxaca ? Bryocladia borealis Dawson Dawson, 1954 Campo Malarrimo, Bahia Vizcaino, Baja Calif. Ophidocladus calif ornica (Hollenberg) Kylin Kylin, 1956; Dawson, I960*; Hollenberg, 1943 (as Rbodosiphonia ) Santa Barbara to San Diego, Calif. Digenia simplex (Wulfen) C. Agardh Dawson, 1954, 1957$, 1959 Gulf of Calif, to Costa Rica ■ Bryothamnion pacificum Taylor Dawson, 1954 Bahia Vizcaino, Baja Calif.; Is la Maria Mag- dalena, Nayai.it Bostrycbia binderi Harvey Taylor, 1945; Dawson, 1957$ Golfo Duke, Costa Rica; Galapagos Arch. Bostrichia calliptera (Montagne) Montagne Taylor, 1945 Panama; Choco, Bahia Buenaventura; Isla Gorgona, Colombia; Galapagos Arch. Bostrycbia radicans Montagne Taylor, 1945; Dawson, 1954, 1957$; Post 1955 (as B. radicans f. moniliforme ) Guaymas, Sonora; Barra de Navidad, Jalisco; El Salvador; Golfo Duke, Costa Rica; Choco, Bahia Buenaventura, Colombia; Galapagos Arch. Bostrycbia simplicins cula Harvey ex J. Agardh Dawson, 1957$ Golfo de Nicoya, Costa Rica Bostrycbia tenella J. Agardh Taylor, 1945 Galapagos Arch. Lophosipbonia macra (Harvey) Falkenberg Dawson, 1954c Isla San Benedicto, Revillagigedo Arch. Lophosipbonia mexicana Dawson Dawson, 1954 Isla Cedros, Baja Calif.; Isla Angel de la Guarda, Gulf of Calif.; Isla Clarion, Re- villagigedo Arch. Lophosipbonia reptabunda (Suhr) Kylin Kylin, 1956; Dawson, 1957$, 1959c (pos- sibly confused with Ophidocladus ); Scagel, 1957 (as L. obscura) ; Hollenberg, 1958 a (as L. obscura) Southern British Columbia to northern Wash.; southern Calif. ?; Costa Rica ? Lophosipbonia scopulorum (Harvey) Womer- sley Dawson, 1957$, 1959, 1959c, 1954 (as L. villum ) ; Scagel, 1957 (as L. villum) Southern British Columbia to Isla Magdalena, Baja Calif.; Gulf of Calif. Brogniartella mucronata ( Harvey ) Schmitz Dawson, 1954; Kylin, 1956 (as probably a species of Micropeuce ) Isla Maria Magdalena, Nayarit Veleroa subulata Dawson Dawson, 1954 Bahia Tepoca, Sonora Pterosipbonia arctica Setchell & Gardner Okamura, 1933; Scagel, 1957 Aleutian L, Alaska, to Puget Sound, Wash. Pterosipbonia baileyi (Harvey) Falkenberg Dawson, 1954, 1959c; Dawson, Neushul & Wildman, I960 Crescent City, Calif., to Isla San Roque, Baja Calif. 452 PACIFIC SCIENCE, VoL XV, July 1961 Pterosiphonia bipinnata (Postels & Ruprecht ) Falkenberg var. bipinnata Okamura, 1933; Scagel, 1957 Aleutian I., Alaska, to San Pedro, Calif. Pterosiphonia bipinnata var. robusta (Gardner) Doty Scagel, 1957 Sitka, Alaska, to Oreg. Pterosiphonia calif or nica Kylin Kylin, 1941; Hollenberg, 1948 La Jolla, Calif. Pterosiphonia dendroidea ( Montagne ) Falken- berg Taylor, 1945; Scagel, 1957; Dawson, 1954, 1959c; Dawson, Neushul & Wildman, 1960 Northern British Columbia to San Jose del Cabo, Baja Calif.; Galapagos Arch. Pterosiphonia gracilis Kylin Scagel, 1957 Southern British Columbia to Coos Bay, Oreg. Pterosiphonia pennata (Roth) Falkenberg Dawson, 1954 Bahia Tepoca, Sonora Levringiella gardneri ( Setchell ) Kylin Dawson, 1945c; Hollenberg 1948 (both as Stromato carpus gardneri ) ; Kylin, 1956 Santa Monica to La Jolla, Calif. Pterochondria pygmaea ( Setchell ) Hollenberg Dawson, 1954; Dawson, Neushul & Wild- man, I960 Southern Calif, to Bahia Asuncion, Baja Calif. Pterochondria woodii (Harvey) Hollenberg Scagel, 1957; Dawson, 1959c; Dawson, Neu- shul & Wildman, I960 Vancouver, British Columbia, to Isla Asun- cion, Baja Calif. Herposiphonia grandis Kylin Scagel, 1957 Southern British Columbia to Coos Bay, Oreg. Herposiphonia rigida Gardner var. rigida Scagel, 1957 Southern British Columbia to Santa Monica, Calif. Herposiphonia rigida var. laxa Setchell & Gard- ner Dawson, 1949, 1954 Santa Catalina I., Calif.; Isla Guadalupe, Baja Calif. Herposiphonia secunda (C. Agardh) Ambronn Taylor, 1945; Hollenberg, 1948; Dawson, 1957, 1957 b, 1959, 1959*, 1959c Southern Calif, to Isla Jicarita, Panama; southern Gulf of Calif.; Clipperton I. Herposiphonia spinosa Dawson Dawson, 1959 Isla Partida, Baja Calif, del Sur Herposiphonia subdisticha Okamura Dawson, 1954; Scagel, 1957 (both as H. parva) ; Dawson, 1954, 1951b , 1959, 1960a Northern Wash, to Isla Guadalupe; Islas San Benito, Baja Calif.; Gulf of Calif.; Golfo de Nicoya, Costa Rica Herposiphonia tenella (C. Agardh) Nageli Hollenberg, 1948; Dawson, 1954, 1954c, 1957£, 1959 Southern Calif, to Bahia Vizcaino, Baja Calif.; Kino, Sonora to Golfo de Nicoya, Costa Rica; Isla San Benedict©, Revillagigedo Arch. Herposiphonia verticillata ( Harvey ) Kylin Coe, 1932; Dawson, 1954, 1959c San Francisco, Calif., to Punta Santa Rosalia, Baja Calif. Amplisiphonia pacifica Hollenberg Dawson, 1954; Scagel, 1957; Dawson, Neu- shul & Wildman, I960, I960* Northern British Columbia to near Punta Eugenio, Baja Calif. Benthic Algae — -Dawson 453 Jantinella vermcaeformis (Setchell & McFad- den) Kylin Dawson, 1954 (in part as J, sinicola ) , 1959; Setchell & Gardner, 1930 (as ], sinicola ) San Pedro, Calif., to Bahia Vizcaino, Baja Calif.; Isla Clarion, Revillagigedo Arch.; southern Gulf of Calif. Cbondria arcuata Hollenberg Hollenberg, 1945 Corona del Mar; Laguna Beach, Calif. Chondria calif arnica ( Collins ) Kylin Dawson, 1946 (as C. cus cutoides ) , 1954, 1954 d, 1959, 1959c; Dawson, Meushul & Wiidman, I960 La Jolla, Calif., to Puerto Culebra, Costa Rica Cbondria clarionensis Setchell & Gardner Dawson, 1954, 1959 (as possibly equal to C. calif ornica ) Isla Guadalupe, Baja Calif., Isla Clarion, Re- villagigedo Arch. Chondria dangeardii Dawson Dawson, 1960^ Isla del Rey, Panama Chondria dasyphylla (Woodward) C. Agardh Dawson, 1954, 1959 Bahia Gonzaga to Bahia Agua Verde, Gulf of Calif. Chondria decipiens Kylin Smith, 1944; Dawson, 1945 b, 1954, 1958, 1959, 1959c Santa Cruz, Calif., to Cabo Colnett, Baja Calif. Isla San Pedro Molasco, Gulf of Calif. ? Chondria flexicaulis Taylor Taylor, 1945 Galapagos Arch. Chondria lancifolia Okamura Dawson, 1957, 1951b Rocas Alijos, Baja Calif.; Golfo de Micoya, Costa Rica ? Chondria nidifica Harvey Kylin, 1941; Dawson, 1954, 1959c; Dawson, Meushul & Wiidman, I960 La Jolla, Calif., to Bahia Asuncion, Baja Calif. Chondria oppositiclada Dawson Dawson, 1946 La Jolla, Calif. Chondria pacifica Setchell & Gardner Dawson, 1954, 1958 Arroyo Hondo, Santa Barbara Co., Calif., to Bahia Tortuga, Baja Calif. Chondria platyclada Taylor Taylor, 1945 Dawson & Beaudette, I960 Costa Rica; Isla Plata, Ecuador Chondria repens Bprgesen Dawson, 1951b Golfo Dulce, Costa Rica Chondria secundata ( J. Agardh) J. B. De Toni J. B. De Toni, 1903 Santa Barbara, Calif. Lauren cia clarionensis Setchell & Gardner Dawson, 1954 Isla Clarion, Revillagigedo Arch. Laurencia congesta Taylor Taylor, 1945 Galapagos Arch. Laurencia crispa Hollenberg Smith, 1944 Monterey Peninsula, Calif. Laurencia decidua Dawson Dawson, 1954c Isla San Benedicto, Revillagigedo Arch. Laurencia densissima Setchell & Gardner Setchell & Gardner, 1937 Galapagos Arch. Laurencia diegoensis Dawson Hollenberg, 1948; Dawson, 1954, 1959c; Dawson, Meushul & Wiidman, I960 Carpinteria, Calif., to Punta Baja, Baja Calif. 454 PACIFIC SCIENCE, VoL XV, July 1961 Laurencia estebaniana Setchell & Gardner Dawson, 1954 Isla Smith; Isla San Esteban, Gulf of Calif. Laurencia gardneri Hollenberg Smith, 19 44 Monterey Peninsula, Calif. Laurencia hancockii Dawson Dawson, 1954 Bahia Agua Verde, Gulf of Calif. Laurencia humilis Setchell & Gardner Dawson, 1954 Isla Clarion, Revillagigedo Arch. Laurencia intricata Lamouroux Dawson, 1954 Bahia Petatlan, Guerrero Laurencia johnstonii Setchell & Gardner Dawson, 1954 Northern Gulf of Calif. Laurencia lajolla Dawson Dawson, 1958 La Jolla, Calif. Laurencia masonii Setchell & Gardner Dawson, 1954 Isla Guadalupe, Baja Calif. Laurencia medio cris Setchell & Gardner Setchell & Gardner, 1937 Galapagos Arch. Laurencia multibulb a Dawson, Neushul & Wild- man Dawson, Neushul & Wildman, I960* Off Punta Hughes and Punta Entrada, Isla Magdalena, Baja Calif. Laurencia obtusa (Hudson) Lamouroux var. gracilis Harvey Taylor, 1945 Galapagos Arch. Laurencia obtusiuscula Setchell & Gardner var. obtusiuscula Dawson, 1949^, 1954, 1959; Dawson, Neu- shul & Wildman, 1960a Near Punta Eugenio, Baja Calif.; Gulf of Calif, from Guaymas to San Jose del Cabo Laurencia obtusiuscula var. corymbifera Setchell & Gardner, prox. Taylor, 1945 Galapagos Arch. Laurencia obtusiuscula var. laxa Setchell & Gard- ner Dawson, 1959, 1954 (as L. obtusiuscula) Southern Gulf of Calif. Laurencia oppositoclada Taylor Taylor, 1945 Galapagos Arch. Laurencia pacifica Kylin Smith, 1944; Dawson, 1954, 1959c Pacific Grove, Calif., to Isla Magdalena, Baja Calif. Laurencia paniculat a (C. Agardh) J. Agardh Dawson, 1954 Bahia Gonzaga; Guaymas to Bahia Agua Verde, Gulf of Calif. Laurencia papillosa var. pacifica Setchell & Gard- ner Dawson, 1954, 1959 Isla Guadalupe; southern Gulf of Calif. Laurencia peninsularis Taylor Dawson, 1954 Cabo San Lazaro, Baja Calif. Laurencia richardsii Dawson Dawson, 1954c Isla San Benedicto, Revillagigedo Arch. Laurencia scrippsensis Dawson Dawson, 1954, 1954c La Jolla, Calif., to Isla Magdalena, Baja Calif.; Isla San Benedicto, Revillagigedo Arch. Benthic Algae— Dawson 455 Lamencia sinicola Setchell & Gardner Dawson, 1954, 1959 Southern Gulf of Calif. Lamencia synderae Dawson Dawson, 1954, 19 : 54^ Santa Catalina L; La Jolla, Calif.; Mas San Benito; Scammon Lagoon; Bahia Vizcaino, Baja Calif. Lamencia spectabilis Postels & Ruprecht Scagel, 1957; Dawson, 1959c Sitka, Alaska, to nothern Santa Barbara Co., Calif. Lamencia splendens Hollenberg Smith, 1944; Dawson, 1954, 1959c; Dawson, Neushul & Wildman, I960 Santa Cruz, Calif., to Punta Baja, Baja Calif. Lamencia subdisticba Dawson, Neushul & Wild- man Dawson, Neushul & Wildman, 1960^ Islas San Benito, Baja Calif.; Isla Jicaron, Panama Lamencia sub opposita (J. Agardh) Setchell Dawson, 1944^, 1954; Dawson, Neushul & Wildman, I960, I960** La Jolla, Calif., to Punta Eugenio, Baja Calif. Lamencia tmbinata Setchell & Gardner Setchell & Gardner, 1937 Isla San Martin, Baja Calif.; Galapagos Arch. Lamencia voragina Taylor Dawson, 1954 White Friars I., Guerrero Erythrocystis saccata (J. Agardh) Silva Smith, 1944 (as Ricardia saccata); Silva, 1952; Dawson, 1954, 1954c Pacific Grove, Calif., to Isla Guadalupe, Baja Calif.; Isla San Benedicto, Revillagigedo Arch. Janczewskia gardneri Setchell & Gardner Dawson, 1954; Scagel, 1957 Vancouver L, British Columbia, to Cabo Col- nett, Baja Calif. Janczewskia lappacea Setchell Kylin, 1941; Dawson, 1954 San Pedro, Calif., to Cabo Colnett, Baja Calif. Janczewskia moriformis Setchell Setchell, 1914 Santa Monica, Calif. Janczewskia solmsii Setchell & Guernsey Dawson, 1954 b, 1954 Redondo to San Diego, Calif.; Isla Guadalupe, Baja Calif. Rhodomela larix (Turner) C. Agardh Okamura, 1933; Scagel, 1957; Dawson, 1958, 1959c Bering Sea to Government Point, Calif. Rhodomela lycopodioides (Linnaeus) C. Agardh Okamura, 1933; Scagel, 1957 Aleutian L, Alaska, to northern Wash. Rhodomela macrantha ( Kiitzing ) Setchell Tokida, 1949 Alaska; British Columbia Odonthalia aleutica (Mertens) J. Agardh Okamura, 1933 Aleutian I., Alaska Odonthalia dentata ( Linnaeus ) Lyngbye Setchell & Gardner, 1903; Tokida, 1949 Alaska to Victoria, British Columbia Odonthalia floccosa ( Esper ) Falkenberg Scagel, 1957 Southern British Columbia to Pismo Beach, Calif. Odonthalia kamtschatica (Ruprecht) J. Agardh Scagel, 1957 Alaska to Puget Sound, Wash. Odonthalia lyallii (Harvey) J. Agardh Okamura, 1933; Scagel, 1957 Alaska to northern Wash.; Oreg. ? Odonthalia oregana Doty Doty, 1947 Curry Co., Oreg. 456 PACIFIC SCIENCE, Vol. XV, July 1961 Odonthalia washingtoniensis Kylin Scagel, 1957 Hope I., British Columbia, to Coos Bay, Oreg. REFERENCES Agardh, J. G. 1852. Species genera et ordines algarum. C. Gleerup, Lund. Vol. 2, pt. 2: 337-720. 1876. Species genera et ordines algarum. T. O. Weigel, Lund. Vol. 3, pt. 1; Epicrisis systematis floridearum, 724 pp. 1899. Analecta algologica: observationes de speciebus minus cognitis earumque dis- position, continuatio V. Actis Coc. Physiog. Lundensis, N. S. 10: 1-160. Ardre, F. 1959. Un interessant Hildenbrandtia du Portugal. Rev. Algol., N. S. 4(4): 227- 237, 3 pis. Coe, W. R. 1932. Season of attachment and rate of growth of sedentary marine organisms at the pier of the Scripps Institution of Oceanography, La Jolla, Calif. Scripps Instn. Oceanogr. Bull. 3(3): 37-86, pis. 1-6. Collins, F. S., I. Holden, and W. A. Set- CHELL. 1895-1919. Phycotheca Boreali- Americana. Exsiccati Fasicles 1-46 and A-E. Priv. printed, Malden, Mass. Dawson, E. Y. 1941. A review of the genus Rhodymenia with descriptions of new species. Allan Hancock Pacif. Exped. 3: 123-180, 13 pis. 1944. The marine algae of the Gulf of California. Allan Hancock Pacif. Exped. 3: 189-454, 47 pis. 1944^. Some new Laurenciae from southern California. Madrono 7(8): 233— 240, 3 pis. 1944 A A new parasitic red alga from southern California. Torrey Bot. Cl. Bull. 77(6): 655-657, 1 pi. 1945. Notes on Pacific Coast marine algae, I. S. Calif. Acad. Sci. Bull. 43(3) : 95- 101, 1 pi. 1945^. Some new and unreported sub- littoral algae from Cerros Island, Mexico. S. Calif. Acad. Sci. Bull. 43(3): 102-112, 3 pis. 1945A An annotated list of the marine algae and marine grasses of San Diego County, California. S. Diego Soc. Nat. Hist. Occ. Pap. (7): 1-87. — 1945c. Notes on Pacific Coast marine algae, II. S. Calif. Acad. Sci. Bull. 44(1): 22-27, 2 pis. 1945 d. Notes on Pacific Coast marine algae, III. Madrono 8(3): 93-97, 1 pi. 1945c. Marine algae associated with up- welling along the northwestern coast of Baja California, Mexico. S. Calif. Acad. Sci. Bull. 44(2): 57-71. 3 pis. 1946. New and unreported marine al- gae from southern California and northwest- ern Mexico. S. Calif. Acad. Sci. Bull. 44(3): 75-91, 20 figs. 1946^. A guide to the literature and distributions of the marine algae of the Pacific coast of North America. S. Calif. Acad. Sci. Mem. 3(1): 1-134. 1949 . Contributions toward a marine flora of the southern California Channel Is- lands, I— III. Allan Hancock Fdn. Occ. Pap. 8: 1-57, 15 pis. — 1949<^. Resultados preliminares de un reconocimiento de las algas marinas de la costa Pacifica de Mexico. Rev. Soc. Mex. Hist. Nat. 9: 215-255. 1 map. 1949A Studies of northeast Pacific Gracilariaceae. Allan Hancock Fdn. Occ. Pap. 7: 1-104, 25 pis. — 1950. Notes on some Pacific Mexican Dictyotaceae. Torrey Bot. Cl. Bull. 77(2): 83-93. 1950^. Notes on Pacific Coast marine algae, IV. Amer. J. Bot. 37: 149-158, 6 figs. 1950A A giant new C odium from Pa- cific Baja California Torrey Bot. CL Bull. 77(4): 298-300, 1 pi. 1950c. Notes on Pacific Coast marine algae, V. Amer. J Bot. 37: 337-344. 1950 d. A review of Ceramium along the Pacific Coast of North America with spe- cial reference to its Pacific Mexican repre- sentatives. Farlowia 4(1) : 113-138, 4 pis. 1950c. A note on the vegetation of a new coastal upwelling area of Baja California. J. Mar. Res. 9(2) : 65-68, 3 figs. Benthic Algae — D awson 457 1951. On the status of the brown alga, Dictyota binghamiae J. G. Agardh. Wasmann J, Biol 8(3): 267-269. 1951*. A further study of upwelling and associated vegetation along Pacific Baja California, Mexico. J. Mar. Res. 10(1): 39- 58, 6 figs. 1953. Marine red algae of Pacific Mex- ico, Part 1. Bangiales to Corallinaceae subf. Corallinoideae. Allan Hancock Pacif. Exped. 17(1) : 1-239, 33 pis. 1953*. Resumen de las investigaciones recientes sobre algas marinas de la costa Pa- cifica de Mexico, con una synopsis de la literatura, sinonimia y distribucion de las especies descritas. Rev. Soc. Mex. Hist. Mat. 13: 97-197. — — 1954. Resumen de las investigaciones recientes sobre algas marinas de la costa Pa- cifica de Mexico, con una synopsis de la literatura, sinonimia y distribucion de las especies descritas, reprinted with corrections, index pagination and addenda i-x from Rev. Soc. Mex. Hist. Nat, 13: 97-197. 1954^. Marine red algae of Pacific Mex- ico, Part 2. Cryptonemiales (cont. ) Allan Hancock Pacif. Exped. 17(2): 250-398, 44 pis. — 1954 A Notes on tropical Pacific marine algae. S. Calif. Acad. Sci. Bull. 53(1): 1-7, 3 figs. — — 1954c. The marine flora of Isla San Benedicto following the volcanic eruption of 1952-53. Allan Hancock Fdn. Occ. Pap. 16: 1-25, 5 pis. 195 4d. Notes on Pacific Coast marine algae, VI. Wasmann J. Biol. 11 (3): 323— 351. 6 figs. — - 1956. Some marine algae of the south- ern Marshall Islands, Pacif. Sci. 10(1): 25- 66, 66 figs. — — - 1957. Notes on eastern Pacific insular marine algae. Los Angeles Co. Mus. Contr. Sci. (8) : 1-8, 4 figs. - — — 1957*. An annotated list of marine al- gae from Emwetok Atoll, Marshall Islands. Pacif. Sci. 11(1): 92-132, 31 figs. 1 957 A. Marine algae of the Pacific Costa Rican gulfs. Los Angeles Co. Mus. Contr. Sci. (15): 1-28, 4 figs. - — — 1958. Notes on Pacific Coast marine algae, VIE S. Calif. Acad. Sci. Bull. 57(2): 65-80, 5 pis. — — 1959. Marine algae from the 1958 cruise of the Stella Polaris in the Gulf of California. Los Angeles Co. Mus. Contr. Sci. (27): 1-39, 9 figs. 1959*. William H. Harvey’s report on the marine algae of the United States north Pacific Exploring Expedition of 1853-56. (E. Y. Dawson, ed.) Pacif. Nat. 1(5): 1-40, 10 pis. — 195 9 A Some algae from Clipperton Is- land and the Danger Islands. Pacif. Nat. 1(7): 1-8, 1 fig. — 1959c. A primary report on the benthic marine flora of southern California, pp. 109- 264. In: Oceanographic Survey of the Con- tinental Shelf area of Southern California. Publ. no. 2, State [California] Water Pollu- tion Control Board. Multilith 560 pp. Sacra- mento. — I960. Marine red algae of Pacific Mex- ico, Part 3. Cryptonemiales, Corallinaceae subf. Melobesioideae. Pacif. Nat. 2(1): 1- 125. 50 pis. I960*. New records of marine algae from Pacific Mexico and Central America. Pacif. Nat. 1(20): 31-52, 7 figs. — 1961. Marine red algae of Pacific Mex- ico, Part 4. Gigartinales. Pacif. Nat. 2(5): 189-343. 63 pis. 1961*. A new sublittoral Gracilariopsis from southern California. Res. Coun. Israel Bull.: in press. Dawson, E. Y., and P. T. Beaudette. I960. Field notes from the 1959 eastern Pacific cruise of the Stella Polaris. Pacif. Nat. 1(13): 1-24, 16 figs. Dawson, E. Y., M. Neushul, and W. D. Wildman. I960. Seaweeds associated with kelp beds along southern California and northwestern Mexico. Pacif. Nat. 1(14): 1- 81. 43 pis. 1960^. New records of sublittoral marine plants from Pacific Baja California. Pacif. Nat. 1(19): 1-30, 4 pis. De Toni, G. 1935. Noterelle di nomenclature algologica. V. V 'Antith amnion tenuis simum Gardner 1927. Priv. printed, Brescia. 458 PACIFIC SCIENCE, VoL XV, July 1961 1936. Noterelle di nomenclature algol- ogica, VII. Primo elenco di Floridee omonime. Priv. printed, Brescia. 1938. A note on phycological nomen- clature. Rhodora 40(469): 27. De Toni, J. B. 1897-1924. Sylloge algarum omnium hucusque cognitarum. 4 (Sylloge Floridearum) (1): I-XJ, I-LXI, 1-388. 1897; (2): 387-77 6. 1900; (3) 775-1522. 1903; (4): 1523-1973. 1905; 6 (Sylloge Floridearum) (5 Additimenta) : I-XI, 1-767. Priv. printed, Patavii, 1924. Doty, M. S. 1947. The marine algae of Oregon, I— II. F arlowi* 3: 1-65, 159-215, 14 pis. Doty, M. S., and E. G. Menez. I960. Tiffani- ella, a new genus in the Ceramiales. Amer. Micr. Soc. Trans. 79(2) : 135-144, 14 figs. Fan, K. C. 1959- Studies on the life histories of marine algae, I. Codiolum petrocelidis and Epongomorpha coalita. Torrey Bot. Cl. Bull. 86 ( 1 ): 1 - 12 . Fan, K. C., and G. F. Papenfuss. 1959. Algal parasites occurring on members of the Geli- diales. Madrono 15(2): 33-38, 10 figs. Farlow, W. G. 1902. Algae. In: Robinson, Flora of the Galapagos Islands, pp. 89-104. Amer. Acad. Arts Sci. Proc. 38(3). Feldmann, J. 1942. Remarques sur les Nemas- tomacees. Soc. Bot. Fr. Bull. 89(4-6): 104- 113. Feldmann, J., and G. Feldmann. 1942. Recherches sur les Bonnemaisoniacees et leur alternance de generations. Ann. Sci. Nat. Bot. ser. 11,3: 75-175,2 6 figs. Feldmann-Mazoyer, G. 1942. A propos de quelques Spermothamnion a polysporanges. Soc. Hist. Nat. N. Bull 33: 15-18. Fensholt, Dorothy. 1955. An emendation of the genus Cystophyllum ( Fucales ) . Amer. J. Bot. 42: 305-322, 51 figs. 1951. A revision of the genus Cysto- phyllum. Summaries of Doctoral Diss. North- western Univ. 1951: 560-563. Foslie, M. H. 1906. Algologiske Notiser. Kgl. Norske Vidensk. Selsk. Skr. 1906(2): 1-28. 1929. Contributions to a monograph of the Lithothamnia. 60 pp., 75 pis. Aktietry- kkeriet, Trondhjem. Gardner, N. L. 1926. New Rhodophyceae from the Pacific Coast of North America, I. Calif. Univ. Publ. Bot. 13: 205-226, 7 pis. 1927. New Rhodophyceae from the Pa- cific Coast of North America. II. Calif. Univ. Publ. Bot. 13: 235-272, 12 pis. 1927^. New Rhodophyceae from the Pacific Coast of North America. III. Calif. Univ. Publ. Bot., 13: 333-368, 13 pis. 1927 b. New Rhodophyceae from the Pacific Coast of North America. IV. Calif. Univ. Publ. Bot., 13: 373-402, 11 pis. 1927c. New Rhodophyceae from the Pacific Coast of North America. V. Calif. Univ. Publ. Bot., 13: 403-434, 10 pis. 1927^. New Rhodophyceae from the Pacific Coast of North America. VI. Calif. Univ. Publ. Bot., 14: 99-138, 17 pis. 1927c. New species of Gelidium on the Pacific Coast of North America. Calif. Univ. Publ' Bot., 13: 273-318, 19 pis. 1940. New species of Melanophyceae from the Pacific Coast of North America. Calif. Univ. Publ. Bot., 19: 267-286, 6 pis. GRUNOW, A. 1915. Additamenta ad cogni- tionem Sargassorum. Verh. Kaiserl. Zool.-bot. Ges. Wein 65: 328-448; 66: 1-48, 136-185. Hamel, G. 1931-39. Pheophycees de France, xlvii 432 pp., 10 pi. Impr. Wolfe, Paris. — 1939. Sur la classification des Ectocar- pales. Bot. Not. 1939: 65-70. Harvey, W. H. 1847. Nereis Australis 2. Reeve Bros., London, vii, 124 pp., 50 pis. 1853. Nereis Boreali- Americana, Part II. Rhodospermae. Smithson. Contr. Knowl. 3(5): 1-258, 24 pis. 1862. Notice of a collection of algae made on the northwest coast of America, chiefly at Vancouver’s Island, by David Lyall. Linn. Soc. Bot. J. 6: 157-177. Hollenberg, G. J. 1940. New marine algae from southern California, I. Amer. J. Bot. 27(10): 868-877, 17 figs. 1941. Culture studies of marine algae, II. Hapterophycus canaliculatus S. & G. Amer. J. Bot. 28(8): 676-683, 17 figs. 1942. Phycological notes — - I. Torrey Bot. Cl. Bull. 69: 528-538, 15 figs. Benthic Algae — DAWSON 459 — 1942^ An account of the species of Polysipbonia on the Pacific Coast of North America, L Oligosiphonia. Amer. J. Bot. 29: 772-785, 21 figs. — 1943. New marine algae from southern California, II. Amer. J. Bot. 30: 571-579, 16 figs. — 1944. An account of the species of Poly- sipbonia on the Pacific Coast of North Amer - ica. II. Polysipbonia. Amer. J. Bot. 31: 474- 483, 12 figs. 1945. New marine algae from southern California. III. Amer. J. Bot. 32: 447-451, 9 figs. — — — 1948. Notes on Pacific Coast marine algae. Madrono 8(5): 155-162. — 1958. Observations concerning the life cycle of Spongomorpha coalita (Ruprecht) Collins. Madrono 14(8): 249-251. — — — 1958^. Phycological notes, II. Torrey Bot. Cl. Bull. 85(1): 63-69, 1 fig. — - 1959. Smith or a, an interesting new al- gal genus in the Erythropeltidaceae. Pacif. Nat. 1(8): 1-12, 1 fig. — - 1 96 1 . Part 5, the genus Polysipbonia. In: E. Y. Dawson, Marine red algae of Pa- cific Mexico. Pacif. Nat. 2(6): 345-375. 7 pis. Ho.pp.AUGH, K. W. 1930. A taxonomic study of the species of the genus Vaucheria collected in California. Amer. J. Bot. 17: 329-347, 4 text figs., pis. 24-27. Jao, C. 1937. New marine algae from Washing- ton. Mich. Acad. Sci. Pap. 22: 99-116, 3 pis. Kylin, H. 1924. Studien iiber die Delesseria- ceen. Lunds Univ. Arsskr, N. F. 20(6): 1- 111, 80 figs. — 1925. The marine red algae in the vi- cinity of the biological station at Friday Har- bor, Wash. Lunds Univ. Arsskr., N. F. 21 (9) : 1-87, 47 figs. — I 932 . Die Florideenordnung Rhody- meniaies. Lunds Univ. Arsskr., N. F. 27 ( 11 ) : 1-48, 8 figs., 20 pis. — - 1935. Zur Nomenclature einigen De lesseriaceen. Forh. Kgl. Fysiogr. Sallsk. Lund 5(23): 1-5. — 1940. Die Phaeophyceenordnung Chor darkles. Lunds Univ. Arsskr., N. F. 36(9) : 1-67, 30 figs., 8 pis. - — — — 1941. Californische Rhodophyceen. Lunds Univ. Arsskr, N. F. 37(2): 1-51, 7 figs, 13 pis. ..... 1944 . Die Rhodophyceen der Schwedis- chen Westkuste. Lunds Univ. Arsskr, N. F. 40(2): 1-104, 32 pis. — 1956. Die Gattungen der Rhodophy- ceen. xv + 673 pp, 458 figs. CWK Gleer- ups, Lund. LEMOINE, Mme. P. 1929 . Les Corallinacees de 1’Archipel des Galapagos et du Golfe de Pa- nama. Arch. Mus. Hist. Nat. 6(4): 47-88, 35 figs, 4 pis. Levring, T. 1940. Die Phaeophyceengattungen Chlanidophora, Distromium und Sr y in go - derma. Kgl. Fysiogr. Sallsk. Lund 10(20) : 1- 11 . — — 1955. Contributions to the marine al- gae of New Zealand. Ark. Bot. 3(11): 407- 432, 14 figs. Loomis, Nina. 1949. New species of Gelidium and Pterocladia with notes on the structure of the thalli in these genera. Allan Hancock Fdn. Publ. Occ. Pap. (6): 1-28, 10 pis. — I960. New species of Gelidium and Pterocladia from the Pacific Coast of the United States and the Hawaiian Islands. Al- lan Hancock Fdn. Publ. Occ. Pap. (24) : 1- 35, 12 pis. Manza, A. V. 1937. Some north Pacific species of articulated corallines. Nat. Acad. Sci. Proc. 23: 561-567. — ■ 1940. A revision of the genera of articu- lated corallines. Philipp, j. Sci. 71: 239-316, 20 pis. North, W. J. 1959. Experimental ecology and IMR kelp bibliography. In: The effects of waste discharges on kelp. Univ. Calif. Inst. Mar. Resour. ref. 59—1 1 (multilith) . I960. Experimental ecology, in kelp in- vestigations program, quarterly progress re- port. Univ. Calif. Inst. Mar. Resour. ref. 60- 3. Ohmi, H. 1955. Contributions to the knowl- edge of the Gracilariaceae from Japan. Hok- kaido Univ. Bull. Fac. Fish. 5(4): 320-331, 6 pis. Ok AMUR A, K. 1933. On the algae of Alaska collected by Y. Kobayashi. Records Oceanogr. Wks. Jap. 5(1): 85-97, 2 pis. 460 PACIFIC SCIENCE, Vol. XV, July 1961 Papenfuss, G. 1944. Notes on algal nomen- clature, III. Miscellaneous species of Chloro- phyceae, Phaeophyceae and Rhodophyceae. Farlowia 1(3): 337-346. 19 AAa. Marine algae of the Monterey Peninsula. A review. Farlowia 7(7): 226- 231. — 1944 A Structure and taxonomy of Tae- nioma , including a discussion on the phy- logeny of the Ceramiales. Madrono 7(7): 193-214, 2 pis. 1 fig. 1945. Review of the Acrochaetium- Rhodochorton complex of the red algae. Calif. Univ. Publ. Bot. 18(14): 299-334. 1947. Further contributions toward an understanding of the Acrochaetium-Rhodo- chorton complex of the red algae. Calif. Univ. Publ. Bot. 18(19): 433-447. — 1950. Review of the genera of algae described by Stackhouse. Hydrobiologia. 2(4): 181-208. — 1955. Classification of the algae pp. 115-224, In: A century of progress in the natural sciences 1853-1953. Calif. Acad. Sci., San Francisco. — - 1956. Notes on South African Marine algae, IV. J. S. Afr. Bot. 22(2): 65-77. — 1958. Notes on algal nomenclature, IV. Taxon 7(4): 104-109. PlCCONE, A. 1886. Alghe del viaggio di cir- cumnavigazione della Vettor Pisani. 97 PP-, 2 pis. 1st. Sordo-Muti, Genova. 1889. Nuove alghe del viaggio di cir- cumnaviazione delle Vettor Pisani. R. Accad. Lincei Iv, 6: 9-63. Post, Erica. 1955. Weitere Datum zur Ver- breitung des Bostrychietum iv. Arch. Protis- tenk. 100(3): 351-377, pis. 11-15. Powell, H. T. 1957. Studies in the genus Fucus L., II. Distribution and ecology of forms of Fucus distichus L. emend. Powell in Bri- tain and Ireland. J. Mar. Biol. Assoc. U. K. 36: 663-693, 4 pis. Proskauer, J. 1950. On Prasinocladm . Amer. J. Bot. 37(1): 59-65. Sanborn, E. L, and M. S. Doty. 1946. The marine algae of the Coos Bay-Cape Arago Region of Oregon. Oreg. St. Monogr. Bot. (8): 1-52, 4 pis. Saunders, De A. 1898. Phycological Memoires. Calif. Acad. Sci., Proc, 3rd ser., (Bot.) 1: 147-168, 21 pis. — 1901. The algae. Papers from the Har- riman Alaska Exped. XXV. Wash. Acad. Sci., Proc. 3: 391-486, 20 pis. SCAGEL, R. F. 1957. An annotated list of the marine algae of British Columbia and north- ern Washington (inch keys to genera). Nat. Mus. Can. Bull. (150): i-vi, 1-289, 1 fig. Schmidt, O C. 1935. Pringsheimia Reinke jetzt Brings keimiella v. Hoehn. Hedwigia, 74: 29. SETCHELL, W. A. 1906. A revision of the genus Constantinea. Nuova Notarisia 17: 162-173. 1912. Algae novae et minus cognitae. Calif. Univ. Publ Bot. 4: 229-268, 7 pis. — — 1914. Parasitic Florideae, L Calif. Univ. Publ. Bot. 6: 1-34, 6 pis. — - 1914a, The Scinaia assemblage. Calif. Univ. Publ. -Bot. 6: 79-152, 7 pis. — — — - 1923. Parasitic Florideae, II. Calif. Univ. Publ. Bot. 10: 393-396. 1923a A revision of the west North American species of Callopbyllis . Calif. Univ. Publ. Bot. 10: 397-401. — — — - 1933. Some early algal confusions, II. Calif. Univ. Publ. Bot. 17: 187-254, 20 pis. — - — - 1937. Report on the sargassums. The Templeton Crocker Expedition of the Cali- fornia Academy of Sciences, 1932, no. 34. Calif. Acad. Sci. Proc. iv, 22(5): 127-158, 6 pis. — - — 1940. Fucus cordatus Turner. Nat. Acad. Sci. Proc. 26(11): 643-651, 2 figs. Setchell, W. A., and E. Y. Dawson. 1941. Bingbamia, the alga, vs. Bingbamia , the cac- tus. Nat. Acad. Sci. Proc. 27(8): 376-381, 1 ph Setchell, W. A., and N. L. Gardner. 1903. Algae of Northwestern America. Calif. Univ. Publ. Bot. 1: 165-418, 21 pis. • — — — 1920. The marine algae of the Pacific Coast of North America. Part 2, Chlorophy- ceae. Calif. Univ. Publ. Bot. 8: 139-381, 21 pis. — 1924. The marine algae. Expedition of the California Academy of Sciences to the Gulf of California in 1921. Calif. Acad. Sci. Proc. iv, 12: 695-949, 77 pis. Benthic Algae— Dawson 461 1925. The marine algae of the Pacific Coast of North America, Part 3, Melanophy- ceae. Calif. Univ. Publ. Bot. 8: 383-898, 20 pis. 1930- Marine algae of the Revillagigedo Islands Expedition of 1925. Calif. Acad. Sci. Proc. iv, 19: 109-215, 15 pis. — — — - 1933. A preliminary survey of Gigar- tina , with special reference to its Pacific North American species. Calif. Univ. Publ. Bot. 17: 255-340, 20 pis. 1937. Iridophycus in the northern hem- isphere. U. S. Nat. Acad. Sci. Proc. 23: 169- 174. — ■ — - 1937^. A preliminary report on the al- gae. The Templeton Crocker Expedition of the California Academy of Sciences. Calif. Acad. Sci. Proc. iv. 2(2): 65-98, 1 fig., 23 pis. Setchell, W. A., and L. R. Mason. 1943. New or little-known crustose corallines of Pacific North America. U.S. Nat. Acad. Sci. Proc. 29: 92-97. — — 1943^. Goniolithon and Neogonioli- thon\ two genera of crustaceous coralline al- gae. Nat. Acad. Sci. Proc. 29: 87-92. Silva, P. C. 1951. The genus C odium in Cali- fornia, with observations on the structure of the walls of the utricles. Calif. Univ. Publ. Bot. 25(2) : 79-114, pis. 1-6, 32 text figs. 1952. A review of nomenclatural con- servation in the algae from the point of view of the type method. Calif. Univ. Publ. Bot. 25(4): 241-324. 1953. The identity of certain fuci of Exper. Wasmann J. Biol. 11(2): 221-232. 1957. Notes on Pacific marine algae. Madrono 14(2): 41-51. 1957^. Remarks on algal nomenclature. Taxon 6(5): 141-145. Silva, P. C, and Ann Cleary. 1954. The structure and reproduction of the red alga, Platysiphonia. Amer. J. Bot. 41(3): 251- 260, 37 figs. Smith, G. M. 1944. Marine Algae of the Mon- terey Peninsula, California, ix -\- 622 pp., 98 pis. Stanford Univ. Smith, G. M., and B. J. Hollenberg. 1943. On some Rhodophyceae from the Monterey Peninsula, California. Amer. Jour. Bot. 30: 211-222, 30 figs. Sparling, Shirley. 1957. The structure and reproduction of some members of the Rhody- meniaceae. Calif. Univ. Publ. Bot. 29(3): 319-396, pis. 48-59, 15 text figs. Taylor, W. R. 1939. Algae collected on the Presidential cruise of 1938. Smithson. Contr. Knowl. misc. ser. 98(9) : 1-18, 2 pis, 14 figs. I945 Pacific marine algae of the Allan Hancock Pacific Exped. 12*: 1—528. 100 pis. 3 figs. 1952. Notes on Vaucheria longicaulis Hoppaugh. Madrono 11(7): 274-277, figs. 1 - 11 . Tokida, J. 1949. Notes on some new or little known marine algae (4). Jour. Japan. Bot. 23(5-6): 69-71. Womersley, H. B. S. 1954. The species of Macrocystis with special reference to those on the southern Australian coasts. Calif. Univ. Publ. Bot. 27(2): 109-132, pis. 1-8, 1 map. Yendo, K. 1902. Corallinae verae of Port Ren- few. Minn. Bot. Stud. 2(40): 711-720, 6 pis. Zeh, W. 1912. Neue Arten der Gattung Lia- gora. Berl. Notizbl. bot. Gart. 5(49): 268- 273. Fishes Killed by the 1950 Eruption of Mauna Loa, Part V Gonostomatidae Marion Grey 1 Among the deep-sea fishes collected at the surface during the Mauna Loa lava flow into the sea in 1950 were 30 small specimens belong- ing to the family Gonostomatidae. Of the 5 genera and 5 species represented, 2 of the gen- era ( 1 of them new ) and all of the species ( 1 or 2 of them new ) were hitherto unknown from waters around the Hawaiian Islands. Gosline et al. (1954) described the lava flow and the methods of collecting. I am indebted to Dr. Gosline for permission to report on these speci- mens, and to Dr. Rolf Bolin for relinquishing his prior claim to them. Gono stoma atlanticum Norman MATERIAL examined: 1 specimen, 59.5 mm. in standard length, collected off the Mauna Loa lava flow, Hawaii, by Moore et al ., June 3, 1950. One specimen, 47 mm. in standard length, collected off the Mauna Loa lava flow, Hawaii, by Yamaguchi, June 6, 1950. Three specimens, 58.5, 54, and 49 mm. in standard length, collected off the Mauna Loa lava flow, Hawaii, by Gosline et al ., June 6, 1950. Counts and measurements given in order of diminishing size, largest specimen first. Dorsal rays 17, 17, 18, 18, 16. Anal rays 28, 29, 29, 28, 28. Pectoral rays 10. Ventral rays 6. Gill rakers on first arch 1 1 + 6, 11 + 6, ? + 6, — , 11 + 7. Photophores: BR 9; IV 16; VAV 5; AC 19 (par- tially lost in smallest specimen); IC 40; OA 13. Per cent of standard length (59.5, 58.5, 54, 49, 47 mm.): depth 16.8, 17.1-17.9, 16.7-17.6, ca. 18.3, 18.0; head 24.3, 25.6, 25.0, 24.5-25.5, 25.5; snout 4.2, 4.27-5.12, 4.62, — 4.25-5.3; orbit 4.2, 4.27, 3.5-4.62, — , 4.25; interorbital width at center of eye 3.36, 3.42-4.27, 3.5-4.62, — , ca. 4.25; upper jaw 18.5, 19.6-20.4, 20.4, 1 Chicago Natural History Museum, Chicago, Illi- nois. Manuscript received December 19, I960. 19.4-20.4, 20.2-21.2; premaxillary 3.36, 4.27- 5.1, 4.62, — , 5.3; toothed portion of maxillary 14.3, 15.4, 15.7, — , 15.9; tip of snout to dorsal origin 58.0, 58.9, 58.4=59.2, 58.1-59.1, 59.5; to anal origin 56.3, 57.2, 56.5-57.4, 56.1, 56.4; to ventral bases 46.2, 48.6, 46.2, 47.8, 46.7; dis- tance between anal origin and caudal base 41.1 — 42.0, 42.7, 41.6, 42.7, 42.5; between last anal ray and caudal base ca. 10.1, ca. 10.2, 10.2, - — , 10.6; between last dorsal ray and caudal base ca. 23.5, ca. 22.2, 22.2-23.1, — , ca. 23.4; between inner insertion of ventral and origin of anal ca. 8.4, 9.4, 7 .4-8. 3, — , ca. 7.45; least depth of cau- dal peduncle 5.8-6.7, 5.97-6.83, ca. 5.5,— ,6.38; dorsal base 16.8, ca. 17.1, 17.6, — , ca. 17.0; anal base 31.9, 32.4, 33.3, — , ca. 32.0; pectoral length — -, 17.1, 16.7, — , 12.75; ventral length ca. 7.56, 6.83, 7.4, —, 6.38. Skin partially lost in all specimens. Only a few scales remaining, mostly over photophores, but well-preserved scale pockets indicating that back and tail, at least, were fully clothed with large, thin, cycloid scales. Pectoral fins reaching as far as 12th or 13th IV photophore. Ventral fins reaching anus. Most specimens with 13 long teeth on maxillary (only 9 or 10 in 2 speci- mens reported by me in I960 from the Marshall Islands) and apparently no posterior pterygoid teeth. Three largest specimens with 3 palatine teeth, increasing in size posteriorly. A small re- flector present behind ORB in only 1 specimen (47 mm.). Only two largest with small glands visible below OA. All 5 specimens with a single pale yellowish infracaudal gland at base of pos- terior procumbent caudal rays; possibly a sec- ond one originally existed anterior to it. Two well-developed supracaudals with the outward appearance of large photophores, being sur- rounded on sides and lower surface by blackish brown pigment. Color blackish brown, head and abdomen darker, cheeks silvery or iridescent and with black puncticulations, peritoneum black, oper- 462 Gonostomatidae-— Grey 463 cular linings brown, inside of mouth pale ante- riorly and brown posteriorly. Largest specimen ( abdomen damaged ) , a fe- male with large ovaries. Counts and measurements are in close agree- ment with those found in published descriptions of this species. The body depth is possibly some- what greater in Hawaiian specimens and the count, in one, of 1 1 + 7 gill rakers is unique. These specimens represent the second Pacific record of G. atlanticum. The first capture was in the Marshall Islands area (Grey, I960). The species is probably more widely distributed in the central portions of the Pacific Ocean than is known at present. Cyclothone sp. MATERIAL EXAMINED: 3 specimens, standard length ca. 20.5, ca. 23, and ca. 23.5 mm., col- lected off the Mauna Loa lava flow, Hawaii, by Moore et al., June 3, 1950. Specimens all in poor condition; following characters common to all: first vav very close to ventral base, anus directly below it; VAV evenly spaced; color pale brownish, abdomen darker, myomeres outlined in darker pigment; on 1 specimen a shred of skin with a few black spots remaining, its original position on body not determinable; 2 narrow vertical brown bars at extreme end of caudal peduncle, 1 above mid- line and 1 below mid-line; dashes of internal pigment at end of caudal peduncle; a series of 6 internal brown spots along base of dorsal fin and 18 along anal base; ventral portion of body between ventral and anal fins, below muscula- ture, entirely colorless and transparent, the VAV photophores situated in this region; branchi- ostegal membrane with a narrow brown line at base and a broken line of brown pigment on edge, otherwise colorless; a bar of brown pig- ment curving down from pectoral base and ex- tending forward on isthmus. Specimen ca. 23.5 mm.: BR 8; VAV 4, possibly only 3 ( fourth at first anal ray ) . IV, AC, and OA all lost or damaged. Maxillary teeth subequal, becoming gradually larger posteriorly, first one not larger than those immediately behind it. Specimen ca. 23 mm.: BR 8, iv 13. vav 4, possibly only 3 (fourth at first anal ray). AC and OA mostly lost. Dorsal rays probably 12, anal rays probably 19. Specimen ca. 20.5 mm.: BR 7. IV and VAV mostly lost. AC 9 or 10, last one on caudal base. Last two VAV present, last at anal origin; if this photophore belongs to the AC series, the total number is 10. Gill rakers on first arch 9+1 + 3 = 13, only 1 in angle. Vomer tooth- less. Palatines and pterygoids each with 3 micro- scopic teeth. Ventral fins undamaged, reaching to or slightly past anal origin. Both the poor condition of the Hawaiian specimens and the uncertain taxonomic status of species of the genus Cyclothone Goode and Bean prevent positive identification of these specimens. They are allied to C. signata Garman and C. alba Brauer ( pale coloration, a single gill raker in the angle of the first arch, no vomerine teeth) and are possibly identical with C. alba. However, C. alba was described with an AC count of 12 or 13, similar to that of C. signata , and the AC number is only 9 or 10 on the only Hawaiian specimen on which these photophores are preserved. The total number of gill rakers on the first arch is similar in all 3 forms, 13 in the 1 Hawaiian specimen on which they can be counted, 14—15 in C. signata and C. alba. C. alba and the Hawaiian specimens have 7-8 BR pho- tophores, in contrast to the 9 or 10 of C. signata. Both the first VAV photophore and the anus of the Hawaiian specimens are extremely close to the ventral bases. In C. alba these were figured by Brauer (1906: 80, fig. 30) as being placed somewhat more posteriorly. However, in speci- mens examined from the Atlantic (Florida), probably identical with C. alba, the first VAV is also closer to the ventral bases than shown in Brauer’s figure and this character may be a vari- able one. Although it is no surprise to learn that a light- colored Cyclothone inhabits Hawaiian waters, the only species of the genus hitherto reported from the area have been the dark-colored forms C. atraria Gilbert and C. canina Gilbert. Araiophos, NEW GENUS type SPECIES: Araiophos gracilis, new species. Eye normal, large. Snout shorter than orbit. Interorbital width at center of eye less than diameter of orbit or length of snout. Mouth 464 PACIFIC SCIENCE, Vol. XV, July 1961 Fig. 1. Araiophos gracilis, holotype, standard length 34 mm. moderate, oblique; edge of premaxillary straight, its angle oblique; toothed edge of maxillary slightly convex, reaching beyond middle of eye but not as far as its posterior margin. Premaxil- lary about half as long as toothed portion of maxillary. Angle, of preopercle almost vertical. All teeth minute, not numerous, uniserial in up- per jaw; lower jaw with an outer row of 3-4 teeth anteriorly. Presence or absence of teeth on vomer, palatines, pterygoids, and tongue not de- terminable. Gill rakers on first arch 15 + 3 = 18 ( 1 specimen ) , 2 in angle. Minute ( micro- scopic) clusters of spines on inner edge of first gill arch. Presence or absence of pseudobranch- iae not determinable. No evidence of scales. Anus about half-way between ventral bases and anal origin or slightly nearer the former. Head and trunk about equal in length to tail or slightly shorter. Origin of dorsal fin well behind middle of body length. Anal origin and ventral bases well ahead of dorsal origin. Adipose fin present. ORB 1, in front of eye, close to pre- maxillary. OP 1 (lower posterior), level with end of maxillary. SO absent. BR (6). No addi- tional photophores on head. Body with a single row of photophores; photophores present on isthmus. IV ( 2 ) on isthmus, + (3) + 4+ (2) = 11. vav (4-5). ac (2) + 3-4 + (2) = 7-8. IC 22-23. No additional photophores and no luminous tissue on body as far as known. Fin rays: dorsal 13-14, anal 28-29, pectoral (15?) 16-17, ventral 6. Branchiostegal rays 8, no spines at bases. Number of vertebrae un- known. Araiophos differs distinctly from all other maurolicid genera in the reduced number of photophores. Only in the BR count of 6 and the vav count of 4 or 5 is it similar to others ( Val- enciennellus Jordan and Evermann also has 4 or 5 VAV and several genera have 5 ) . The small number of photophores might be a juvenile character and the absence of photophores be- tween the isthmus and the posterior end of the pectoral base (where in some species they are known to develop late), the absence of 2 of the OP, and the small size of the ORB, give sup- port to the possibility that all of the specimens studied are young. On the other hand, although the specimens range in standard length from a little over 21 mm. to 34 mm., the number and relative size of the photophores are identical in all. If more were to develop, on the body at least, one would expect to find the smallest spec- imen differing somewhat from the largest. The presence of gonads is also significant and the obvious implication is that Araiophos provides another example of neoteny among fishes. Even were the specimens still juvenile they could not be shown to belong to any known genus. Only in Maurolicus Cocco is the dorsal fin situated so far behind the middle of the body length; and the Hawaiian specimens cannot belong to Mau- rolicus, in which the developmental stages are well known and which has acquired all of the meristic characters of the adult at a length of about 20 mm. Nor is it likely that the Hawaiian specimens represent a metamorphosis stage of some known genus; as far as we know the de- velopment of maurolicid genera is direct and none are known to undergo a metamorphosis Gonostomatidae — G rey 465 stage with changes in body form or fin positions. An attempt to modify the diagnosis of one of the known genera to accept this new form has been unsuccessful. Several characters align it with Tboropbos Bruun and Neopbos Myers: relative positions of dorsal and anal origins; number of gill rakers; elongate body form; iv photophores on isthmus straight, none curving upward toward pectoral base; VAV not reaching anal origin. The lack of OA photophores, if not a juvenile character, is an additional point of resemblance between the new genus and Neo- pbos, which has only 1 OA. However, in both Tboropbos and Neopbos the angle of the gape anteriorly is much more acute than that of Arai- opbos, and in Neopbos and Tboropbos the gape is wider, the maxillary reaching to or beyond the posterior margin of the eye (only slightly past the middle of the eye in Araiopbos) . In fin ray counts Araiopbos is closest to Argyripnus Gilbert and Cramer, and to Maurolicus; and only in the latter, among maurolicid genera, is the dorsal origin situated well behind the middle of the body length as it is in AraiopJoos. Derivation of name: araios, Gr., 'few’; pbos, Gr., 'light.’ Araiopbos gracilis , new species Figs. 1-3 HOLOTYPE: Standard length 34 mm., collected off the Mauna Loa lava flow, Hawaii, by Gosline et al., June 6, 1950. PARATYPES: 9 specimens, standard length 30.5, 24, and 23 mm. (3 specimens); length from tip of snout to base of caudal fin 31.5, 28, 27.5, and 26.5 mm. (5 specimens, lower jaws broken), and ca. 21 mm. from nape to base of caudal fin ( 1 specimen, head lacking), collected off the Mauna Loa lava flow, Hawaii, by Moore et al., June 3, 1950. Dorsal rays 13-14 (holotype 14). Anal rays 28-29 (holotype 29). Pectoral rays 16-17 (holo- type 16, possibly only 15). Ventral rays 6. Branchiostegal rays 8. Gill rakers on first arch 15 + 3 = 18 (holotype only), 2 in angle. Measurements of holotype expressed in per cent of the standard length (34 mm.), fol- lowed in parentheses by similar measurements of 2 specimens 31.5 mm. from tip of snout to caudal base and 30.5 mm. in standard length: depth 13.2 (12.7, 13.1-14.7); head ca. 22.1 ( — , 22.9); snout 5.87 ( — , 6.55); orbit 8.8 Fig. 2. Araiopbos gracilis, holotype, ventral view. 466 PACIFIC SCIENCE, Vol. XV, July 1961 FIG. 3 . Araiophos gracilis . paratypes, standard length 30.5 and ca. 23 mm. (7.87, 6.55-8.19); interorbital width at center of eye 4.41 (4.73, ca. 4.9); upper jaw 13.2; pre- maxillary ca. 4.41; toothed portion of maxillary 8.8; distance between tip of snout and dorsal origin 57.3 ( 55.2, 57.3 ), anal origin 48.5 (46.6- 47.3, ca. 49.0, and in a specimen 28 mm. long from snout to caudal 48.2-50.0), ventral base ca. 44.1 ( ca. 39.4, 44.1); distance between first anal ray and base of middle caudal rays 51.4 (52.0, 49.0, and in a specimen 28 mm. long from snout to caudal 50.0), last anal ray and base of middle caudal rays ca. 11.7 (11.0, 14.7— 16.4), last dorsal ray and base of middle caudal rays 27.9 ( ca. 28.4, ca. 26.2), last dorsal ray and adipose fin 4.41 (6.3, — ), ventral bases and anal origin 5.87 (4.73, 4.9); least depth of caudal peduncle 5.87 (6.3, 4.9-6.55); dorsal base 11.7 (14.2?, 11.5-13.1); anal base 38.2 (39.4, 31.1-32.8); adipose base 2.94 (6.3, — , and in a specimen 23 mm. in standard length 4.34-6.52). Specimens all damaged to some extent, espe- cially in head and abdominal regions; both eyes of holotype lost, this specimen otherwise in fair condition. Body elongate, slender, compressed but with ventral surface flat between the second single abdominal IV photophore and the VAV group. No scales or scale pockets remaining. Pectoral fins of holotype reaching half-way between pec- toral and ventral bases; bases broadly peduncu- late. Ventral fins of holotype reaching anal ori- gin. Adipose fin with a relatively long base; very fragile and easily lost, leaving no trace ( ab- sent on 2 specimens, tearing loose on a third, possibly incomplete on holotype); its origin above a vertical from the third single AC pho- tophore, above end of anal fin. Specimens too small and fragile to determine presence or absence of pseudobranchiae. Teeth rather sparse, visible only under mag- nification, their presence or absence on vomer, palatines, pterygoids, or tongue impossible to determine but 1 specimen with a few minute teeth on roof of mouth anteriorly. Photophores as in generic diagnosis. ORB minute, close to premaxillary. The single OP double, relatively large. Only (2) small IV on isthmus posteriorly, well separated from abdom- inal IV (no photophores between isthmus and posterior end of pectoral base); abdominal iv commencing just behind pectoral base, organs of groups (first 3 and last 2 organs) small, the 4 single organs much larger, well separated from one another. VAV occupying only anterior half of space between ventrals and anal, the individ- ual organs very small; anus below last, thus about half-way between ventral bases and anal Gonostornatidae — G rey 467 origin or slightly nearer the former; number usually (4), two specimens with (4) on one side and (5) on the other, holotype with (4) on both sides. First group of AC (two organs) above fifth to seventh anal rays, last group ( two organs ) behind anal fin but remote from caudal base; single organs much larger than individual organs of groups; number of single organs usually 3, 3 specimens with 3 on one side and 4 on the other, holotype with 3 on both sides. Visible inside flattened belly of most speci- mens are two long, narrow, whitish or yellowish masses with the appearance of testes, extending forward to third or fourth single IV photophore. In addition to these structures at least 4 speci- mens (24, 27.5, 27.5, 31.5 mm.) also with ovaries; eggs minute except in 24 mm. speci- men, in which the ovaries are relatively large and contain eggs of different sizes. Color in alcohol yellowish, abdomen darker. Minute brown puncticulations along mid-line on tail, and outlining myomeres of tail. A nar- row vertical brown bar at extreme end of caudal peduncle. Some specimens with a short series of minute brown spots along anal base between first AC group and second single AC. Sparse brown pigment on top of head. A few minute brown spots on caudal rays, fins otherwise color- less. Inner bronzy iridescence and a few rela- tively large black chromatophores visible through abdominal wall laterally. Danaphos oculatus ( Garman ) MATERIAL EXAMINED: 7 specimens, 2 31.5 and ca. 39 mm. in standard length, others slightly shorter but too damaged to measure, col- lected off the Mauna Loa lava flow, Hawaii, by Moore et al., June 3, 1950. Two specimens, standard length 36.5 and ca. 36.5 mm., collected off the Mauna Loa lava flow, Hawaii, by Gosline et al., June 6, 1950. Cne specimen, standard length 33 mm., col- lected off the Mauna Loa lava flow, Hawaii, by Yamaguchi, June 6, 1950. Dorsal rays 6 (4 specimens). Anal rays (23?) 24-25 (6 specimens). Pectoral rays ca. 18 and ca. 16 (2 specimens), upper 7 or 8 rays longer and much heavier than others, lowermost rays small and fine. Gill rakers on first arch 10-11 -T 1 +2 = 13-14 (2 specimens), only 1 in angle. Measurements of 2 specimens expressed in per cent of the standard length (36.5 and 33 mm.): depth 23.3, 21.2; head 23.3-24.6, 24.2; snout 5.47, 4.54; orbit (both vertical and hori- zontal) 8.2, 7.55; interorbital width too narrow to be measured accurately; upper jaw — , 15.1- 16.6; premaxillary 8. 2-9.6, 7.55; toothed por- tion of maxillary 9.6-10.95, 9.06; tip of snout to dorsal origin 34.2, 34.8; to anal origin 46.5, 45.4-46.9; to ventral base ca. 41.0, 37.8-39-3; distance between first anal ray and base of mid- dle caudal rays 52.0, 56.0-57.5; between last anal ray and base of middle caudal rays 8. 2-9.6, 7.55; between last dorsal ray and middle caudal rays ca. 60.2, 53.0; least depth of caudal pedun- cle ca. 6.85, 7.55; dorsal base ca. 4.1, ca. 4.54; anal base ca. 43.8, 46.9; length of pectoral fin 24.6, 24.2-25.7; length of ventral fin 8.2-9.6, 9.06-10.6. A few large, round, thin scales remaining on several specimens. Anus below third VAV. slightly nearer anal fin than ventral bases. Dor- sal base very short. Anal origin close behind a vertical from last dorsal ray. No adipose fin. Pectorals, when complete, extending as far as third VAV photophore. Ventrals, when complete, reaching slightly past anal origin. Ventral bases below dorsal fin. Photophores: No upper OP. BR (6). IV com- plete on only 2 specimens; (3) + (4) on isth- mus, organs of first group smaller than those of second group; abdominal group 11, in 1 speci- men the first one separate, connection of others not determinable; abdominal series of a second specimen separated thus: 1 + (2) + (8), the gland joining the (2) obvious on outer surface of body, the (8) seen through damaged abdo- men to be joined to one another by a narrow gland, and each organ also joined to its counter- part on other side of body; total IV 17-18. VAV (5) in all 10 specimens. AC (3) + 15 + (4) + 1 = 23 in 5 specimens, (4) + 15 + (4) + 1 = 24 in 1 specimen, (3) + 14 + ( 4 ) + 1 = 22 in 1 specimen, (3) + 16 + (4) + 1 = 24 in 1 specimen, and 1 specimen with (3) + l6+(4) + 1 = 24 on one side, (3) + 17 + (4) + 1 = 25 on other side; damaged in 1 specimen. OA (2) + 4 = 6, complete on only 1 specimen. 468 PACIFIC SCIENCE, Vol. XV, July 1961 In alcohol, color of tail and back whitish, abdomen and head, excepting the colorless snout, blackish. Body wall along most of anal base colorless and transparent. End of caudal peduncle blackish. A series of about 28 black spots running from nape almost to caudal base. Two specimens contain ovaries, the eggs larger in one than in the other. D. oculatus has not been reported previously from the Hawaiian Islands although it is known to inhabit adjacent parts of the North Pacific, as well as the eastern Pacific from off Lower California to Monterey Bay. Argyripnus atlanticus Maul Figs. 4, 5 MATERIAL EXAMINED: 2 specimens, stand- ard length 27 or 27.5 mm. and 35.5 mm., col- lected off the Mauna Loa lava flow, Hawaii, by Moore et al ., June 3, 1950. Smaller specimen in poor condition; head, tail, and abdomen severely damaged; adipose fin, latter part of anal fin, last group of AC photophores, and isthmus IV mostly lost; ab- dominal IV damaged. Dorsal rays 12. Anal rays ca. 12 in front of AC group of five photophores. Pectoral fins short and fleshy, rays discernible under magnification but not clearly enough to be counted. Gill rakers on first arch 19 + 5 = 24 on left side, 19 + 6 = 25 on right side, 2 in angle. Depth 5-5.5 mm. Head and trunk shorter than tail, distance between snout and anal origin ca. 12-12.5 mm. and between first anal ray and base of middle caudal rays 15 mm. Anal origin below about fourth dorsal ray. BR (6). IV on isthmus (6) on one side, apparently straight, not curving upward posteriorly; abdominal IV (10), hanging loose from body, possibly in- complete. First VAV-AC group (14), probably not fully developed, first 6 or 7 in front of anal fin and larger than remainder. Middle group of AC (5). OA lost on one side of body, (2) on other side, remainder probably still undeveloped. Color in alcohol yellowish; upper half of body with a double series of small black chromato- phores from nape to tail and a third, incom- plete, row running posteriorly from above anal origin. Larger specimen, 35.5 mm., in fairly good condition although difficult to measure, being bent. A few large, thin, cycloid scales remain- ing on back and sides. Dorsal rays 11. Anal rays 15 + 10 == 25, a short space filled with mem- brane below third and fourth organs of the AC group of (5), separating the two groups of rays. Pectoral rays 18. Gill rakers on first arch 19 + 6 = 25, 2 in angle. Measurements in millimeters: depth 8; head 11; snout 2-2.5; orbit ca. 4; interorbital width at center of eye 2-2.5; upper jaw 7-7.5; pre- maxillary 3—3.5; toothed portion of maxillary 4-4.5; tip of snout to dorsal origin ca. 17, to anal origin 15.5, to ventral base ca. 12.5-13; distance between first anal ray and base of mid- dle caudal rays 18.5—19, last anal ray and base of middle caudal rays ca. 6.5, last dorsal ray and base of middle caudal rays ca. 15; least depth of caudal peduncle 2.5-3; dorsal base 3.5-4; anal base ca. 11.5; pectoral length 6.5-7; adipose base 1.5; distance between VAV-AC group of photophores and group of (5) 3.5 FIG. 4. Argyripnus atlanticus, standard length 35.5 mm. GonostomatIdae—--GKEY 469 Fig. 5. Argyripnus atlanticus, standard length 35.5 mm., ventral view. and between group of (5) and posterior AC group 3.5-4. Premaxillary more than half as long as toothed portion of maxillary. Maxillary reach- ing a vertical from posterior margin of eye. Head and trunk shorter than tail. Anal origin slightly in advance of dorsal origin, which is above a vertical from about third anal ray. Adi- pose fin well developed, its origin above a verti- cal from just behind Ac group of ( 5 ) . Pectoral fin long, nearly reaching anal origin. Ventrals inserted well in advance of dorsal origin, ends of rays broken. BR (6). IV (6) + (11) = 17, first organ of abdominal group directly below last organ of isthmus group, which curves upward posteriorly; individual organs of first group slightly larger than those of abdominal group. VAV-AC (21) + (5) + (17) = 43, first 7 in front of anal fin and larger than remainder, upward curve over anal fin low and gradual. IC 60. OA ( 2 ) + 3 = 5 on left side, no damage apparent, probably not fully developed; ? + 4 on right side, first organs lost but gland remaining (possibly contained 2-3 photophores ) . Color in alcohol yellowish; diffuse brown pig- ment on caudal peduncle and nape; 3 irregular series of small black chromatophores from nape to caudal. These two little fishes are considerbly smaller than any specimens of the genus Argyripnus previously reported and provide the first scant evidence of developmental changes. The larger one possibly possesses all adult characters but it seems more likely that the OA are still in- complete and that a few of the organs of the first VAV-AC group have not yet developed. One or 2 additional organs may also appear in the last group of AC, which in this little fish begins directly behind the anal fin. In all other speci- mens of Argyripnus seen or figured there are 2 to 4 of these photophores above the end of the anal fin. Characters of the smaller specimen suggest that the middle group of 5 AC develops relatively early and that of the anterior VAV-AC group the organs in front of the anal fin appear first. In the last group of AC the posterior organs are apparently first to appear. The OA develop late, probably being still incomplete on the 35.5 mm. specimen. If the isthmus IV are actually straight on the small damaged fish, as they appear to be, the displacement upward of the posterior organs would occur after the species reaches a 470 PACIFIC SCIENCE, Vol. XV, July 1961 standard length of about 27 mm. and before it attains 35 mm. An additional juvenile char- acter of the smaller specimen is the short, fleshy pectoral fin. The larger of the 2 Hawaiian specimens dif- fers from all others of the genus Argyripnus in two rather important characters, the position of the anal origin in front of the dorsal origin, and the abdominal count of 11 IV. The IV pho- tophores number (6) + (10) = 16 in almost all known specimens of the genus, the only ex- ceptions being found on 2 western Atlantic spe- cimens in which the counts are (6/7) + (10) = 16/17 and (6) + (9/10) = 15/16. Both of the Hawaiian specimens are otherwise very close to, and probably identical with, A. atlanticus Maul, a species hitherto known only from the North Atlantic. In Pacific species (A. ephip- piatus Gilbert and Cramer, A. iridescens Mc- Culloch, and an unrecorded specimen from the Philippines) the^anal origin is situated below the end of the dorsal fin. In A. atlanticus it is be- low the anterior rays of the dorsal and the anal origin of the younger Hawaiian specimen is similar. The fact that the two little Hawaiian fishes differ from one another in this character suggests that the anal position may be somewhat variable. In meristic characters these specimens differ from A. atlanticus only in the number of photophores in the first VAV-AC group and, as noted above, these photophores are probably not fully developed on the Hawaiian fishes. If a few more photophores are still to appear in this series, both the VAV-AC and the IC counts would be within the range of the same counts of A. atlanticus. Evidence that a few more photo- phores actually are present in the adult is shown by the fact that the distance between the AC groups is proportionately greater than in any other specimens of the genus known ( see Table 3). These specimens are also like A. atlanticus and unlike other species in the lower and less abrupt curve in the VAV-AC series above the front of the anal fin. Their more slender bodies can certainly be attributed to their youth and the relative lengths of the trunk and tail prob- ably also change with age. Both of the young Hawaiian fishes have a proportionately longer tail, shorter trunk, and shorter distance between the snout and ventral bases than are found in older specimens from either the Atlantic or Pa- cific oceans. TABLE 1 Meristic Characters of Argyripnus Species SPECIMENS STAND- ARD LENGTH DOR- SAL RAYS ANAL RAYS PEC- TORAL RAYS VEN- TRAL RAYS GILL RAKERS ON FIRST ARCH A. atlanticus, juv. Hawaii 21-21 A 11-12 19+5-6=24-25 A. atlanticus, juv. Hawaii 35.5 11 15+10=25 18 _ 19+6=25 A. atlanticus, type 1 eastern Atlantic 56 12 — =27 19 7 17+7=24 A. atlanticus western Atlantic 55-71 11-12 13-1 5 + (8)9-10=22-27 (16)17-19 6-7 17-19+6-7=22-26 A, ephippiatus Hawaii 72 12 12+12=24 15 6? 13-14+5=18-19 A. ephippiatus, type Hawaii 75 11 11 + 11=22 15 _ 14+5 = 19 A. ephippiatus Hawaii 80 11-12 15 _ 13+5-6=18-19 A. iridescens 2 Australia.... 90-135 12-14 - = 24-25 16-17 7 16+?=? Argyripnus sp. Philippines 82 12 13 + 12=25 17 7 12+4=16 1 Data from Maul, 1952: 56. 2 From McCulloch, 1926: 169, and Norman, 1930: 299. Gonostomatidae- — G rey 471 TABLE 2 Photophore Counts of Argyripnus Species VAV+ STAND- FIRST POSTE- VAV 4 ARD GROUP RIOR AC +AC SPECIMENS LENGTH IV OF AC GROUP TOTAL IC OA A. atlanticus, juv. Hawaii 27-27.5 - (14) „ (2) A. atlanticus, juv. Hawaii ........... 35.5 (6)+(ll)=17 (21) (17) 43 60 (2)+3 = 5 A, atlanticus, type 1 eastern Atlantic...... A. atlanticus 56 (6)+(10)=16 (28) (18) 51 67 7 western Atlantic 55-71 (6/ 7)+(9/10)= (24-28) (16-18) 46-51 62-67 (3-4) +3-4=7 15-17 3 A. ephippiatus Hawaii 72 (6)+(10)=l6 (20) (14) 39 55 (5)+2=7 A. ephippiatus, type Hawaii 75 (6)+(10)=l6 (19) (14/15) 38/39 55/56 (5)+2=7 A. ephippiatus Hawaii 80 — (5)+2=7 A. iridescens 2 Australia 90-135 «5)+(10) = l6 (20-21) (12-14) 37-40 53-56 7 Argyripnus sp. Philippines 82 (6)+(10)=l6 (18) (12) 35 51 (5/ 6)+(l/2) = 7 1 From Maul s 1952: 56. 2 See footnote to Table 1. 3 iv usually (6) + (10) = l6; 7 in isthmus group on one side of 1 specimen only and 9 in abdominal group on one side only of a second specimen. 4 Including middle AC group, which is always (5). SPECIES OF Argyripnus Meristic characters and some body propor- tions of a number of specimens of Argyripnus are shown in Tables 1-3. In addition to the young specimens reported here the following have been examined: A. ephippiatus Gilbert and Cramer, holotype, USNM no. 47708, "Al- batross” sta. 3472; and two specimens, USNM no. 126079, Hawaii, "Albatross,” 1902, exact data lacking (probably the specimens reported in 1905 by Gilbert). Argyripnus sp., USNM no. 135402, Philippines, "Albatross” sta. 5542, vicinity of northern Mindanao, 8° 48' 30" N., 123° 35' 30" E., 200 fathoms (366 m.), 1 speci- men, hitherto unreported. A. atlanticus Maul, "Oregon,” western Caribbean Sea, 1 specimen (Grey, I960: 67); and 13 hitherto unreported specimens taken off Puerto Rico ("Oregon” sta. 2644, 2645, 2646) and north of the Bahamas "Combat” sta. 235 ) . Specimens of Argyripnus are too scarce in museum collections to allow a determination of the number of species contained in the genus but are numerous enough to show that at least 2 distinct species exist. It is equally clear that each of these 2 forms is variable but the extent and limits of the variation are not determinable. The eastern Atlantic form of A. atlanticus , rep- resented by a single specimen, differs from west- ern Atlantic specimens in having a few more anal rays, more posteriorly situated dorsal and anal fins, and a smaller mouth. A. atlanticus also inhabits the Pacific at Hawaii as shown above. The second distinct form of the genus has been found so far only in the Pacific, at Ha- waii, the Philippines, and Australia. As shown in Tables 1-3 these Pacific specimens show variation that may indicate specific distinction. A. ephippiatus Gilbert and Cramer, from Ha- waii, has fewer pectoral rays than other Pacific specimens. A. iridescens McCulloch, from Aus- tralia, is deeper bodied than other Pacific speci- 472 PACIFIC SCIENCE, Vol. XV, July 1961 TABLE 3 Proportions of Argyripnus Species Expressed in Per Cent of Standard Length SPECIMENS STANDARD LENGTH DEPTH HEAD SNOUT ORBIT UPPER JAW A. atlanticus, juv. Hawaii 27-27.5 18.0-20.4 — — A. atlanticus, juv. Hawaii 35.5 22.5 30.9 5.62-7.02 ca. 11.3 A. atlanticus, type 1 eastern Atlantic. ... 56 26.3 30.0 — — A. atlanticus western Atlantic 55-71 23.2- ca. 28.7- ca. 5. 0-6.8 10.5 to 19.0 to ca. 27.6 ca. 31.9 12.1-12.9 20.4-21.2 A. ephippiatus Hawaii.... 72 — _ — — — A. ephippiatus , type Hawaii 75 26.0 ca. 32.0 6.65-7.32 12.0 23.3 A. ephippiatus Hawaii. ... 80 25.6 26.2 ca. 6.86 13.7 22.5-25.1 A. iridescens 2 Australia,.. 90-135 29.6 30.6 6.8 11.65 — Argyripnus sp. Philippines 82 26.2 34.7-35.3 6.7-7 .3 12.8 23.2 FIRST LAST LAST ANAL ANAL DORSAL SNOUT TO SNOUT TO SNOUT TO RAY TO RAY TO RAY TO DORSAL ANAL VENTRAL CAUDAL CAUDAL CAUDAL SPECIMENS ORIGIN ORIGIN BASE BASE BASE BASE A. atlanticus, juv — 44.5-46.2 54.5 or — — or 55.5 43.6-45.5 A. atlanticus, juv. ca. 47.8 43.6 ca. 35.2- 52.0-53.5 ca. 18.3 ca. 42.2 36.6 A. atlanticus eastern Atlantic 49-0 55.0 43.6 — — — A. atlanticus western Atlantic 46.3-49.0 3 46.3-47.2 ca. 39-6- 48.6 to ca. 16.4- 39.4 to to ca. 52.7 43.1 52.5-53.5 18.7 43.2-44.0 A. ephippiatus (72) Hawaii — — — — — — A. ephippiatus, type Hawaii 41.4 54.5 46.0 48.6 17.3 ca. 43.4 A. ephippiatus (80) Hawaii 45.0 55.6 44.4 48.7 14.4 ca. 41.8 A. iridescens Australia — — — — — — Argyripnus sp. Philippines 46.4 ca. 55.5 46.4 46.4 15.2 ca. 41.5 FIRST VAV-j- LEAST AC GROUP MIDDLE DEPTH OF TO AC GROUP CAUDAL DORSAL ANAL ADIPOSE PECTORAL MIDDLE TO LAST SPECIMENS PEDUNCLE BASE BASE BASE LENGTH AC GROUP AC GROUP A. atlanticus, juv. Hawaii — — — — — — — A. atlanticus, juv. Hawaii 7.02-8.45 9.84-11.3 ca. 32.3 4.22 18.3-19-7 9.84 9-84-11.3 A. atlanticus eastern Atlantic 11.2 — — — 23.2 — — A. atlanticus western Atlantic 7.04-9.32 11.9-13.2 30.0-33.9 ca. 5.8- 18.9-24.4 3.79 to 6.9-9.06 5 10.5 5. 3-6.0 A. ephippiatus (72) Hawaii — — — — — 2.78 7.65 A. ephippiatus, type Hawaii 8.65 14.0 ca. 31.3 ca. 8.0 — — 8.0 A. ephippiatus (80) Hawaii 9.36 10.85 33.1 10.0 25.0 ca. 4.37 ca. 8.25 A. iridescens Australia.... — — — — 24.3 — — Argyripnus sp. Philippines 9.14 12.2 31.5 4 21.9 3.05 7.33 1 Calculated from measurements given by Maul, 1952: 56. 2 Calculated from measurements given for a single specimen by McCulloch, 1926: 169. 3 42.7 and 43.0-43.9 in 2 specimens 55 and 57 mm. in standard length. * Upper surface of tail damaged, adipose fin lost. 3 lu.0-10.9 in smallest specimen (55 mm.). Gonostomatidae — G rey 473 mens and may have more gill rakers, although information on the total number of rakers on the first arch of this species is lacking. The sin- gle specimen of Argyripnus from the Philip- pines is unique in a few characters. Because of these unique features and because it is the only specimen to have been caught between Hawaii and Australia, the specimen is not identified to species at the present time, although it was originally labeled " Argyripnus iridescens ” and is catalogued under that name in the collection of the U. S National Museum. A few minor errors appeared in the’ original description of A. ephippiatus (Gilbert and Cramer, 1896: 4 14 ) and may be corrected as follows: dorsal rays 11 (not 10); BR (6) (not 5 ) ; AC in the most posterior group (15) on the left side and (14) on the right side; adipose fin present and undamaged. The smallest of the 3 specimens of A. ephip- piatus examined is bent and cannot be measured accurately. The photophores of this specimen and of the holotype are intact, but many of those of the largest specimen are lost ( abdomen and ventral portion of tail damaged). In all 3 specimens the tail is a little shorter than the trunk; the anal origin is below the end of the dorsal fin; there is a distinct gap, filled with membrane, between the two groups of anal rays in the holotype and the smallest specimen; the unbroken pectoral fin of the largest fish ex- tends to a point half-way between the ventral and the anal fins. The pseudobranchiae are well developed; on the inner edge of the first gill arch is a series of groups of minute spines and 1 or 2 slightly enlarged spines on the upper limb near the angle. The maxillary reaches the posterior margin of the orbit; 1 or 2 very small teeth are present on each side of the vomer; and on each palatine, anteriorly, are a few still smaller teeth. A. ephippiatus differs from A. atlanticus in having fewer gill rakers and fewer photophores in the ventral series. It is unique in its lower pectoral count and in having 1 or 2 slightly enlarged spines on the inner edge of the upper limb of the first gill arch. The specimen from the Philippines has the tail a little shorter than the trunk; the anal origin is below the end of the dorsal fin; there is a short gap between the two groups of anal rays, situated below the third organ of the mid- dle AC group of 5; the pectoral fin extends slightly beyond the ventral base; the back is damaged and the adipose fin is lost. The pseudo- key TO species OF Argyripnus la. Anal origin below first few dorsal rays, or slightly anterior to dorsal origin. Gill rakers on first arch 17-19 + 5-7 = 22-26. vav + Ac (24-28) + (5) + (16-18) = 46-51. IC 62-67 A. atlanticus Maul North Atlantic, Hawaii lb. Anal origin below posterior half of dorsal fin. Gill rakers on first arch 12—16 + 4-6 = 16-19 (20 + ?). 2 vav + ac (18-21) + (5) + (12-15) = 35-40. ic 51-56. 2a. Gill rakers on lower limb of first arch 13-16. First group of VAV + AC (19-21). IC 53-56. Head 2.6 to ca. 3.8 times in standard length. 3a. Pectoral rays 16-17. Depth 3. 1-3.4 times in standard length. Orbit 2.5-27 times in head length . A. iridescens McCulloch Australia 3b. Pectoral rays 15. Depth 3.8-3 -9 times in standard length. Orbit 1.9-2.25 times in head length A. ephippiatus Gilbert and Cramer Hawaii 2b. Gill rakers on lower limb of first arch 12. First group of VAV + AC (18). IC 51. Head 3.85 times in standard length (depth 3.8 times in standard length. Orbit 27- 275 times in head length) . Argyripnus sp. Philippines 2 A. iridescens with 16 on lower limb; number on upper limb, and total count, unknown. 474 PACIFIC SCIENCE, Vol. XV, July 1961 branchiae are well developed; on the inner edge of the first gill arch is a series of minute spine- clusters but no enlarged spines are present. The maxillary extends to the posterior margin of the orbit; 1 tooth is present on each side of the vomer; 1 minute tooth can be seen anteriorly on each palatine. This specimen differs from A. atlanticus in the same characters that distin- guish A . ephippiatus from that species. It is unique in its low gill raker and photophore counts and its relatively large head. A description of the western Atlantic speci- mens of A. atlanticus is being published else- where. REMARKS ON MAUROLICID GENERA The discovery of 2 young specimens of Ar~ gyripnus and of a new maurolicid genus re- quires a partial revision of my key to gonosto- matid genera published in I960. Until relation- ships within the family are better understood, a number of monotypic genera are maintained, however reluctantly, although some of them may prove to be synonymous. For example, Thoro- phos, Neophos, and Araiophos may represent 3 species of a single variable genus, but until the characters separating them can be shown to be of specific rather than of generic value it seems advisable to place them in separate genera. Larger series of specimens, the discovery of post- larval stages, and possibly an understanding of their ecology should one day provide the basis for a more definitive classification. In the mean- while Neophos remains separate from Thoro- phos principally on the basis of the biserial pre- maxillary teeth; the only other genera in the family with 2 rows of teeth on the premaxillary bones are Yarrella , Triplop h os, and Polymetme, Neophos also differs from most maurolicids in having a body form similar to that of several nonmaurolicid genera; only Araiophos, among maurolicids, is similarly slender-bodied anteri- orly and thus atypical of the group with which it is classified. Araiophos is unique in the form and arrangement of the IV photophores, which consist of 2 minute grouped organs on the isth- mus, and on the abdomen four large, well-sep- arated organs preceded by a group of 3 tiny photophores and followed by a group of 2 small ones. The abdominal IV show no such differenti- ation in any other genus in the family. The 2 genera Sonoda and Ar gyripnus may also prove to be synonymous. Sonoda paucil- ampa Grey (1961), a species found recently in the western Atlantic, is in some respects inter- mediate between the 2 genera; and the dis- covery at Hawaii of an Argyripnus with the anal origin anterior to the dorsal origin, as it is in Sonoda, further narrows the gap between the 2 genera. Material used in the preparation of the fol- lowing key, in addition to that reported here and in I960, included a single damaged speci- men of Thorophos euryops Bruun from "Dana” sta. 3736-v, 9° 17' N., 123° 58' E., 28 June 1929, 1000 m. of wire out; and 6 specimens of Neophos nexilis Myers, USNM no. 151400, from off Mindanao, Philippines, "Albatross” sta. 5516, 8° 46' N., 123° 32' 30" E., 9 August 1909, 175 fathoms (320 m.). KEY TO MAUROLICID GENERA OF GONOSTOMATIDAE la. AC composed mostly of separate photophores, more or less evenly spaced, usually one or two groups of 2-4 small organs included ( always at least 3 separate organs ) . 2a. Anal origin in advance of dorsal origin. Ventral bases well ahead of dorsal origin. IV on isthmus straight, none curving upward posteriorly. Total number of AC 7-15. ic 22-37. Dorsal rays 8-14. Anal rays 28-38. Gill rakers on first arch 13-15 + 3-5 = 18-19, 2 in angle. 3a. Maxillary reaching to or beyond posterior margin of eye. Anus closer to anal fin than to ventral bases. Dorsal rays 8-11. Anal rays 31-38. OP 3. so present. Number of IV on isthmus 6, IV present below pectoral base, total number of IV 17. AC present above the most anterior anal rays, total number 13-15. IC 35-37. OA present. (AC mostly single.) Gonostomatidae — G rey 475 4a. Premaxillary teeth biserial. Lower jaw teeth biserial in anterior half. Dorsal origin about in middle of body length. No adipose fin. Dorsal rays 8-9. Anal rays 38. IV on isthmus 1+ (2) + (3) =6. VAV l+(3) + l = 5. OA 1 - Neophos Myers 4b. Premaxillary teeth uniserial. Lower jaw teeth uniserial? 3 Dorsal origin slightly behind middle of body length. Adipose fin present. Dorsal rays 11. Anal rays 31. IV on isthmus 3 + (3) =6. VAV ( 5 ) . OA (2) + 5 = 7 - Thorophos Bruun 3b. Maxillary reaching slightly past middle of eye. Anus half-way between ventral bases and anal origin or slightly nearer ventral bases. Dorsal rays 13-14. Anal rays 28-29- OP 1. SO absent. Number of IV on isthmus (2), no IV below pec- toral base, total number of IV 11. AC absent above the most anterior anal rays, total number 7-8. IC 22-23. No OA. (Dorsal origin well behind middle of body length. Adipose fin present.) Araiophos, new genus 2b. Anal origin behind or below last dorsal ray. Ventral bases below dorsal fin. IV on isthmus curving upward posteriorly. Total number of AC 22-26. IC 45-49. Dorsal rays 6. Anal rays 24-25. Gill rakers on first arch 10-12 + 1 + 2 = 13-15, only 1 in angle. (Dorsal origin well in advance of middle of body length. Anus about half-way between ventral bases and anal fin or slightly nearer anal. Adipose fin present or absent. SO absent. Total number of IV 18, (3) -f (4) on isthmus. AC present above the most anterior anal rays. Vertebrae 38.) Danaphos Bruun lb. AC composed of 2 to 5 groups of 2 or more photophores each ( 1 separate organ present anteriorly only in Maurolicus ) . ( Two gill rakers in angle of first arch. ) 5a. AC in 3 to 6 groups of 2 to 4 small photophores each. IV (3) + (4) on isthmus, (16-17) on abdomen, total 23-24. VAV (4-5). Gill rakers on first arch 12 + 2- 3 ™ 14—15. (Anal origin below or slightly in advance of dorsal origin. Adipose fin present. Anus closer to anal origin than ventral bases. No SO. No AC above the most anterior anal rays. Vertebrae 32-33?.) V alenciennellus Jordan and Evermann 5b. AC in 2 or 3 groups of 5 or more photophores each. IV (6, rarely 7) on isthmus, (10-13) on abdomen, total 16-19. VAV (6-8), or joined to anterior group of AC ( Argyripnus ) . Gill rakers on first arch 12-22 + 3-8 = 16-30. 6a. Anus about half-way between ventral bases and anal fin or nearer ventrals. Dorsal origin about in middle of body length. SO absent. Lower posterior OP greatly enlarged. IV (6, rarely 7) + (10, rarely 9 or 11) = 16 (rarely 15 or 17 ) . Total number of OA 6-7. 7a. No adipose fin. VAV (7-8), well separated from AC. AC in 2 long groups or 3 short groups, absent above the most anterior anal rays. Dorsal rays 8-9. (Vertebrae 40 and ?.) Sonoda Grey 7b. Adipose fin well developed. VAV grouped with photophores above the most anterior anal rays, this group numbering (18-28) in adult. Dorsal rays 10-14. (Vertebrae 47 and ?.) Argyripnus Gilbert and Cramer 6b. Anus near anal fin. Dorsal origin well behind middle of body length. SO present. Lower posterior op of normal size, iv (6) + ( 12-13) = 18-19. Total number of OA 9 (rarely 10). (Adipose fin present. AC present above the most anterior anal rays, in 2 groups of (14-18) and (7-9) photophores, preceded by a sin- gle elevated organ. Vertebrae 32-33.) Maurolicus Cocco ' 5 Premaxillary and lower jaw bones lost from only specimen of Thorophos examined. 476 PACIFIC SCIENCE, Vol. XV, July 1961 REFERENCES Brauer, A. 1906. Die Tiefseefische, 1. Sys- tematischer Teil. Wiss. Ergebn. Deutschen Tiefsee Exp. Valdivia 15(1): 1-432, 18 pis., 176 figs. Gilbert, C. H. 1905. The aquatic resources of the Hawaiian Islands, Part II. Section II. The deep-sea fishes. U.S. Fish Comm. Bull, for 1903, 23(2): i-ix, 577-713, pis. 66-101. Gilbert, C. H., and F. Cramer. 1896. Report on the fishes dredged in deep water near the Hawaiian Islands, with descriptions and fig- ures of twenty-three new species. U.S. Nat. Mus. Proc. 19: 403-435, pis. 36-48. Gosline, W. A., V. E. Brock, H. L. Moore, and Y. Yamaguchi. 1954. Fishes killed by the 1950 eruption of Mauna Loa, I. The origin and nature of the collections. Pacif. Sci. 8(1): 23-27, figs. 1-3. Grey, M. i 960. A preliminary review of the family Gonostomatidae, with a key to the genera and the description of a new species from the tropical Pacific. Mus. Comp. Zool. Bulk 122(2): 57-125, figs. 1-3. Maul, G. E. 1952. Monografia dos peixes do Museu Municipal do Funchal. Additions to previously revised families. Mus. Mun. Fun- chal BoD 6(16): 51-62, figs. 13-17. McCulloch, A. R. 1926. Report on some fishes obtained by the F.I.S. "Endeavour” on the coasts of Queensland, New South Wales, Victoria, Tasmania, south and southwestern Australia, Part 5. Biol. Res. Endeavour 5: 157-216, pis. 43-56, figs. 1-4. Norman, J. R. 1930- Oceanic fishes and flat- fishes collected in 1925-1927. Discovery Rep. 2: 261-370, pi. 2, figs. 1-47. Amorphous Mineral Colloids of Soils of the Pacific Region and Adjacent Areas 1 Yoshinori Kanehiro 2 and Lynn D. Whittig 3 The presence of amorphous mineral colloids in soils and geologic formations is not as un- common as was first believed in the early years following the acceptance of the clay mineral concept. In the early reports the occurrence of amorphous material was associated with only a few rare and isolated clay materials. Because amorphous colloids are not the major component in most soils and their presence may be found in relatively low concentrations, if found at all, their detection has been difficult. Moreover, whereas crystalline clay minerals are relatively uniform in composition, the amorphous mate- rials exhibit a varying degree of composition and poor degree of crystallinity, further adding to the difficulty in their identification. Often their presence has been suggested only because mineral allocations of crystalline materials failed to add up to 100 per cent. In recent years im- provement in the use of techniques such as X-ray diffraction, infrared absorption, electron microscopy, and surface area determination, has made it possible to make significant progress in the study of amorphous colloids. Much of the research dealing with amorphous mineral colloids in soils has been conducted by soil scientists working in the Pacific region or in its adjacent areas. The leadership in this field definitely belongs to this group of researchers. It is the object of this paper to review and discuss the contributions of these workers in order to obtain a better perspective of this very important fraction of soils. These investigators 1 This paper is based on part of a joint report by members of the Clay Mineralogy Work Group of the Western Soil and Water Research Committee. Manu- script received October 3, I960. 2 Department of Agronomy and Soil Science, Uni- versity of Hawaii, Honolulu. 3 Department of Soils and Plant Nutrition, Univer- sity of California, Davis, California. have pointed out that the amorphous consti- tuents make up a sizeable fraction in many soils occurring in Hawaii, Japan, New Zealand, Ore- gon, and other Pacific areas. These amorphous mineral colloids play a prominent role in soil formation and also impart certain distinctive and unique properties to the soil. Thus, a re- view of this nature appears justified. NOMENCLATURE OF AMORPHOUS COLLOIDS The isolation and description of amorphous colloids have been difficult because of the great variability in materials. Moreover, early sam- ples classified as "amorphous” were actually found to be finely crystalline with modern X-ray diffraction methods. Stromeyer and Hausmann first used the name allophane to describe amor- phous material lining cavities in marl in 1816. Since that time many related materials have been called allophane and this term has become associated with amorphous constituents of clay. Ross and Kerr (1934) described allophane as essentially an amorphous solid solution of silica, alumina, and water having no definite atomic structure, and they applied the term allophane to a great number of amorphous clay materials regardless of their composition. They studied five specimens of allophane, all essentially hy- drous aluminum silicates, and found that SiO L > ranged from 25 to 34 per cent, AI 0 O 3 from 30 to 36 per cent, and H 2 0 from 31 to 38 per cent. The New Zealand workers (Fieldes et al., 1952, 1954; Birrell and Gradwell, 1956) have used the term amorphous colloidal hydrous ox- ides apart from the term allophane in their description of amorphous clays. With allophane, which is considered to be one of the most im- portant amorphous minerals, Fieldes (1955, 1956) has preferred to recognize three distinct 477 478 PACIFIC SCIENCE, Vol. XV, July 1961 forms: allophane A, allophane B, and the in- termediate form, allophane AB. In classifying the clay minerals Grim (1953) has included only the allophane group under the amorphous clay minerals. Brown (1955) in his proposed nomenclature has divided the amorphous min- erals into oxides, silicates, and phosphates. In this system, allophane is included in the silicates. OCCURRENCE OF AMORPHOUS COLLOIDS Kelley and Page (1943) in their mineralog- ical investigation encountered two soils from Naalehu and South Point on the island of Ha- waii that exhibited very high cation exchange capacities, 120 m.e. and 88 m.e. per 100 g., respectively. They reported that differential thermal analysis showed pronounced endother- mic peaks at 160° C. for these two soils in addition to showing weak X-ray diffraction pat- terns. These investigators, therefore, concluded that the high cation exchange properties were related to the presence of considerable amor- phous material. Included in this study were soils from Vale, Oregon, and the Mojave Desert, which also gave very indistinct X-ray lines and showed low temperature breaks, inferring the presence of amorphous material. Dean ( 1947 ) in his D.T.A. study of a num- ber of Hawaiian soils derived from ash and lava found that many of these soils contained almost no crystalline clay minerals. In addition some showed almost no hydrous oxides. It was previously shown by Ayres ( 1943 ) that some of these same soils possess very high inorganic cation exchange capacities. Dean concluded that it was possible that some of these soils contain alterations of the kaolin minerals. Tanada (1950) divided Hawaiian soils into five groups on the basis of chemical analyses and dehydration studies. He obtained similar high cation exchange capacity values for the two soils, Naalehu and South Point, that Kelley and Page ( 1943 ) had previously reported. How- ever, Tanada did not draw any conclusions re- garding the cause of such high values. Tamura, Jackson, and Sherman (1953) em- ployed X-ray, chemical, thermal, and infrared techniques and found up to 30 per cent allo- phane in the less than 0.2 micron fraction of two hydrol humic latosols from the island of Hawaii. The authors noted that the allophane found in the subsoil of one of these soils was very similar to allophane from Woolwich, Eng- land (Kerr, 1951). Gibbsite and goethite were reported to make up the bulk of the remaining clay fraction. They also investigated the low humic latosols and reported that the dominant minerals are of the kaolin family. Up to 10 per cent allophane was found to occur in the clay fraction of this group of soils. In a subsequent paper (1955) the same authors reported on a humic ferruginous lato- sol from the island of Maui which showed that almost 30 per cent of the clay fraction in the subsoil was composed of allophane. The occurrence of allophane in some soils of northwestern Oregon was suggested by Whittig et al. (1957). These soils, members of the Cascade and Powell series, contained relatively high percentages of alkali-soluble silica and alumina. The amorphous alumino-silicate in these soils was formed by weathering of aeolian volcanic ash. In earlier work, Whittig (1954) reported the occurrence of a more stable form of allo- phane in two humic ferruginous latosols of Ha- waii. The allophane of these soils had a rela- tively low cation exchange capacity (of the order of 10 m.e. per 100 g.) and resisted solu- tion in boiling Na 2 COa solution. More recently Bates (1961) described the presence of mineral gels in Hawaiian soils which are mixtures of aluminum, iron, silica, and ti- tanium compounds. The gel material is very reactive chemically and gives rise to inorganic and organomineral complexes in the colloid fraction. Matsusaka and Sherman (I960) have re- ported that the iron hydroxide and oxide of the amorphous mineral colloid fraction of Hawai- ian lateritic soils will form strongly magnetic iron oxides on dehydration. This may help ex- plain the magnetic properties of weathered ferruginous geological formations. In Japan Sudo (1954), Sudo and Ossaka (1952), and Aomine and Yoshinaga (1955) have pointed to allophane as the dominant con- stituent of Ando soils which are formed from volcanic ash. These soils are characterized by a Amorphous Mineral Colloids — Kanehiro and WHITTIG 479 low bulk density, a high organic carbon con- tent, and low base saturation. These properties are attributed to the preponderance of allo- phane. The Ando soils and related types are found associated with the Pacific ring of vol- canic activity. These Japanese workers have found that the fine clay fraction of the Ando soils is characterized by being amorphous to X-rays and possesses medium-to-high cation ex- change capacities and high phosphate- and ethylene glycol-retention values. The New Zealanders have also worked ex- tensively on the identification of amorphous con- stituents. In 1952 Birrell and Fieldes (1952) and Birrell (1952) identified the presence of amorphous material, principally allophane, in soils derived from rhyolitic and andesitic ash. The allophane was found to be present mainly in the clay fraction although it was inferred that it was present to some extent in the silt fraction. These soils were characterized by a high water-holding capacity, high shrinkage, and irreversible drying, characteristics that are strik- ingly common to many other Pacific region soils dominated by allophane. Birrell ( 1952 ) also pointed out that these soils had a waxy appear- ance and were greasy to the feel, yet they were not unusually sticky. He further noted that as- sociated with nonreversible drying, liquid and plastic limit values were much greater for un- dried soils than for dried soils. Later reports, especially by Fieldes and his co-workers ( 1955, 1956, 1957, 1955), have con- firmed that allophane and other amorphous con- stituents dominate many New Zealand soils de- rived from volcanic ash and, in some cases, basaltic parent materials. These workers utilized electron microscopy, differential thermal analy- sis, and infrared absorption extensively in iden- tifying the presence of amorphous constituents. PEDOGENIC SIGNIFICANCE OF AMORPHOUS COLLOIDS Ross and Kerr (1934) described allophane as an amorphous hydrous aluminosilicate having no definite chemical composition and that it is commonly associated with halloysite. They were careful to point out that it is not a microscopic mixture of amorphous silica and alumina. Kerr (1951) offered confirmatory evidence by ab- sorption spectra that allophane is not a mixture of alumina and silica. Tamura et al. (1953) assigned allophane to weathering stage 1 1 or the gibbsite stage in the weathering sequence of clay-size minerals as presented by Jackson et al. ( 1948). They noted that the trend for increased gibbsite with in- creased rainfall is very marked in passing from the low humic latosols to the hydrol humic latosols. With this increase in gibbsite is an associated increase in allophane. A mechanism for the transition of alumina and silica through allophane to kaolin was pro- posed by Tamura and Jackson (1953). The steps are as follows: (1) amorphous hydrous alumina crystallizes to a gibbsite structure; ( 2 ) with partial dehydration, hydroxyls in the gibb- site octahedra are replaced by oxygens of the silica tetrahedra; ( 3 ) this process occurs in the presence of silica solutions and continues through entrance of silica between gibbsite sheets, resulting in a cross-linking of silicated octahedral sheets of alumina which corresponds to allophane; (4) kaolinite is formed from allophane on completion of unidirectional bond- ing through alternate wetting and drying in an acid medium where enough silica is available. The stable, nonreactive form of allophane re- ported by Whittig (1954) as a constituent of some humic ferruginous latosols of Hawaii was considered to be a weathering product of hal- loysite. Electron micrographs of clay fractions of these soils revealed a transition from well- developed halloysite rod structures to spherical, X-amorphous allophane particles. It was sug- gested that partial removal of silica from the rigid halloysite rods by leaching allowed the rods to curl up in a direction perpendicular to their original curvation. Allophane formed in this way possessed properties quite different from those of the more labile allophane described by Tamura and Jackson ( 1953) and would occupy a lower position in the weathering sequence of Jackson et al. (1948). More recently Bates (I960) suggested that the development of allophane is a logical stage in the weathering of certain Hawaiian volcanic ash and also in the matrix of rock. In other cases, he indicated that allophane is an inter- 480 PACIFIC SCIENCE, Vol XV, July 1961 mediate stage in the weathering sequence of halloysite to gibbsite. Bates also reported the formation of gibbsite crystals upon dehydration of amorphous Fe-Al gels. This observation sup- ports Sherman (1957), who reported that crys- talline gibbsite aggregates formed when the soils of the Hydrol Humic Latosol group were air dried. These soils have a high content of amorphous mineral colloids which contain a sub- stantial amount of gel material. Some of the Japanese workers (Sudo, 1954; Sudo and Ossaka, 1952) conclude that allo- phane precedes halloysite in the weathering sequence from ash to allophane to halloysite. Aomine and Yoshinaga (1955) have also em- phasized that the clay fraction of the volcanic ash soils of Kyushu and Hokkaido formed un- der similar well-drained conditions is pre- dominantly allophane, regardless of differences in temperature, weathering time, vegetation, and ash origin. The New Zealanders have also tried to pro- perly position the amorphous materials in the weathering sequence. Fieldes and S win dale (1954) have prepared a flow sheet tracing the mechanism of silicate minerals weathering. They have proposed that the nature of the clay constituents of any soil can be predicted if its parent material and weathering stage are known. The amorphous materials occupy a great role in this flow sheet in that it is thought that the primary silicates (aside from the micas) can- not form layer silicates without first passing through an amorphous stage. Clays derived from rhyolitic and andesitic ash pass through the weathering sequence from amorphous hydrous oxides through allophane to meta-halloysite and kaolin. It is believed that many of the Hawaiian soils derived from andesitic ash follow this same sequence. In an earlier paper Fieldes et al. (1952) em- phasized that the amorphous hydrous oxides played more than a brief transitory role. They reported some soils of the lower Cook Islands which showed high cation exchange capacity values. These soils were all low in silica, and also allophane was not found to be a constitu- ent in them. They attributed the cation ex- change capacity mainly to the amorphous hy- drous oxides. In later papers Fieldes (1955, 1956, 1957) reported enough fundamental differences in allo- phane to warrant recognizing three types: allo- phane A, allophane B, and the intermediate type, allophane AB. Based mainly on infrared absorption data, it was found that silica is linked with alumina to form allophane A while some silica is discrete as amorphous hydrous silica in allophane B. Fieldes could offer no satisfactory explanation as to why co-precipitation and link- ing of alumina and silica occur to only a limited extent in allophane B. He did not want to state that allophane B is simply a mixture of amor- phous alumina and silica. Differential thermal analysis shows that a high temperature exo- therm between 850° and 1000° C. is strong in allophane A, not present in allophane B, and weakly developed in the intermediate form, allophane AB. Fieldes (1955) has presented a weathering sequence of clays derived from rhy- olitic and andesitic ash: allophane B — allophane AB- — allophane A — meta-halloysite — -kaolinite. He has stated that in this sequence the stable form is meta-halloysite and progress towards this stable form through allophane A is consist- ent with the mechanism proposed by Tamura and Jackson (1953). The structure consisting of hydrous alumina octahedra randomly cross- linked by silica tetrahedra and called allophane by Tamura and Jackson would hence correspond to allophane A as proposed by Fieldes. There is a growing consensus among investi- gators in this field that amorphous colloids may play a very important role in soil formation and in establishing properties of many soils of the continental United States as well as in the Pacific islands. Because of their noncrystalline nature, identification of allophane and other amorphous constituents is at present very difficult, and at best very unreliable, by standard methods of analysis unless they be present as predominant components of their system. In view of the fact that transition from primary silicates (with the exception of micas) to the secondary layer sil- icates must include some solution and reprecip- itation, it is reasonable to suspect that amor- phous colloids exist, at least as a transition stage, in most of our soils. The extent of their pres- ence is masked in most of our mineralogical studies by the crystalline components present. Amorphous Mineral Colloids — Kanehiro and Whittig 481 The fact that amorphous colloids greatly affect properties of soils where they are dominant con- stituents emphasizes the need for more critical examination of our soils for evidence of their presence. According to Kanehiro and Sherman (1956), it is well established that aliophane has low bulk density, high water-holding capac- ity, and, in some cases at least, a high cation ex- change capacity. In addition it has been observed that aliophane has a strong aggregation effect on soils, a very high phosphate fixing capacity, and the ability in some cases to fix organic com- pounds so as to render nitrogen available with difficulty to microorganisms and higher plants (T. Sudo, private communication ) . Early researches of Burgess and McGeorge (1926) and Burgess (1929) in Arizona sug- gested that amorphous alumina, silica, or alumi- nosilicate may be formed quite readily in soils by application of alkaline solutions. Kerr (1928) further postulated that amorphous aluminosili- cates may form even in slightly acid soils as a result of local hydrolysis of feldspars, solution of alkaline silicates and aluminates, and subse- quent co-precipitation. The Pacific region and its adjacent areas, with their recent volcanic materials offer a splendid opportunity for investigation of the presence of, the properties imparted by, and the mechanisms of formation of, such amor- phous constituents in soils. SUMMARY The work on the amorphous mineral colloids of soils of the Pacific region and its nearby areas is described. The amorphous colloids, especially aliophane, dominate the clay fraction of many soils de- rived from volcanic ash, as well as some rocks, in the Pacific area. Soils dominated by these amorphous colloids have many distinct and unique properties. The position of the amor- phous colloids, especially of aliophane, in the weathering sequence remains to be fully clari- fied. For many reasons, the identification of amor- phous colloids is often difficult; however, re- cent improvements in instrumental techniques have greatly facilitated this identification. REFERENCES Aomine, S., and N. Yoshinaga. 1955. Clay minerals of some well-drained volcanic ash soils in Japan. Soil Sci. 79: 349-358. Ayres, A. S. 1943. Soils of the high rainfall areas of the Hawaiian Islands. Hawaii Agr. Expt. Sta. Tech. Bui. 1. Bates, T. F. I960. Rock weathering and clay formation in Hawaii. Mineral Industries 29: 1,4-6. BiRRELL, K. S. 1952. Some physical properties of New Zealand volcanic ash soils. Proc of the First Australia-New Zealand Conf. on Soil Mech. and Foundation Eng. Birrell, K. S., and M. Fieldes. 1952. Alio- phane in volcanic ash soils. Jour, of Soil Sci. 3: 156-166. BIRRELL, K. S., and M. Gradwell. 1956. Ion- exchange phenomena in some soils containing amorphous mineral constituents. Jour, of Soil Sci. 7: 130-147. Brown, G. 1955. Report of the day minerals group subcommittee on nomenclature of day minerals. Clay Min. Bui 2: 294-300. Burgess, P. S. 1929. The so-called "build-up” and "break-down” of soil zeolites as influenced by reaction. Arizona Agr. Exp. Sta. Tech. Bui. 28: 101-135. Burgess, P. S., and W. T. McGeorge. 1926. Zeolite formation in soils. Science 64: 652- 653. Dean, L. A. 1947. Differential thermal analysis of Hawaiian soils. Soil Sci. 63: 95-105. Fieldes, M. 1955. Clay mineralogy of New Zea- land soils, Part 2. Aliophane and related min- eral colloids. N. Z. Jour, of Sci. and Tech. 37: 336-350. — — — - 1956. Clay mineralogy of New Zealand soils, Part 3. Infrared absorption spectra of soil days. N. Z. Jour, of Sci. and Tech. 38: 31-43. — 1957. Clay mineralogy of New Zealand soils, Part 4. Differential thermal analysis. N. Z. Jour, of Sci. and Tech. 38: 533-570. 482 PACIFIC SCIENCE, Vol. XV, July 1961 Fieldes, M., and L. D. Swindale. 1954. Chem- ical weathering of silicates in soil formation. N. Z. Jour, of Sci. and Tech. 36: 140-154. Fieldes, M., L. D. Swindale, and J. P. Rich- ardson. 1952. Relation of colloidal hydrous oxides to the high cation-exchange capacity of some tropical soils of the Cook Islands. Soil Sci. 74: 197-205. Fieldes, M., and K. I. Williamson. 1955. Clay mineralogy of New Zealand soils, Part 1. Electron micrography. N. Z. Jour, of Sci. and Tech. 37: 314-335. Grim, R. E. 1953. Clay mineralogy. McGraw- Hill Book Co., Inc., New York. 384 pp. Jackson, M. L., S. A. Tyler, A. L. Willis, G. A. Bourbeau, and R. P. Pennington. 1948. Weathering sequence of clay-size minerals in soils and sediments, I. Fundamental general- izations. Jour, of Phys. and Colloid Chem. 52: 1237-1260. Kanehiro, Y., and G. D. Sherman. 1956. Effect of dehydration-rehydration on cation exchange capacity of Hawaiian soils. Soil Sci. Soc. Amer. Proc. 20: 341-344. Kelley, W. P., and J. B. Page. 1943. Criteria for the identification of the constituents of soil colloids. Soil Sci. Soc. Amer. Proc. 7: 175-181. Kerr, H. W. 1928. The nature of base exchange and soil acidity. Jour. Amer. Soc. Agron. 20: 309-335. Kerr, P. F. 1951. Reference Clay Minerals. Amer. Pet. Inst. Proj. 49. Columbia Univer- sity, New York. Matsusaka, Y., and G. D. Sherman. 1961. Magnetism of iron oxide in Hawaiian soils. Soil Sci.: in press. Ross, C. S., and P. F. Kerr. 1934. Halloysite and allophane. U.S. Geol. Surv. Prof. Paper 185-G: 135-148. Sherman, G. D. 1957. Formation of gibbsite aggregates in latosols developed on volcanic ash. Science 125: 1243-1244. Stromeyer, F., and J. F. L. Hausmann. 1816. Gotti ngische Gelehrte Anzeigen. 2: 125. Sudo, T. 1954. Clay mineralogical aspects of the alteration of volcanic glass in Japan. Clay Min. Bui. 2: 96-106. Sudo, T., and J. OssAKA. 1952. Hydrated hal- loysite from Japan. Jap. Jour, of Geol. and Geog. 22: 215-229. Tamura, T., and M. L. Jackson. 1953. Struc- tural and energy relationships in the forma- tion of iron and aluminum oxides, hydrox- ides, and silicates. Science 117: 381-383. Tamura, T.,~M. L* Jackson, and G. D. Sher- man. 1953. Mineral content of low humic, humic and hydrol humic latosols of Hawaii. Soil Sci. Soc. Amer. Proc. 17: 343-346. — 1955. Mineral content of a latosolic brown forest soil and a humic ferruginous latosol of Hawaii. Soil Sci. Soc. Amer. Proc. 19: 435-439. Tan ADA, T. 1950. Certain properties of the inorganic colloidal fraction of Hawaiian soils. Jour, of Soil Sci. 2: 83—9 6. Whittig, L. D. 1954. Crystalline and amor- phous weathering products in some soils of temperate and tropical origin. Thesis, Univ. of Wisconsin Library. Whittig, L. D., V. J. Kilmer, R. C. Roberts, and J. G. CADY. 1957. Characteristics and genesis of Cascade and Powell soils of north- western Oregon. Soil Sci. Soc. Amer. Proc. 2 1 : 226 - 232 . NOTE Daylight Surface Occurrence of Myctophid Fishes Off the Coast of Central America Franklin G. Alverson 1 The CAPTURE off the coast of Central America of Benthosema pterota (Alcock, 1891), by Cali- fornia-based tuna clippers, on three known occa- sions, at the surface during daylight hours is worthy of note. Myctophids are ordinarily be- lieved to be bathypelagic in nature and asso- ciated with the twilight zone of the sea. The fact that myctophids make nocturnal migrations to the surface has long been known and many spec- imens have been taken at the surface during the hours of darkness. The first daylight capture of Benthosema pte- rota was recorded 13 mi. west of Cape Blanco, Costa Rica, on March 14, 1952, by the tuna clip- per "Anna M.” The fish were taken with a dip net at 1300 on a clear day from the fishing racks of the vessel. The water was clear and the surface temperature was 82.5 °F. The skipper, Ted Sorensen, reported three "balls” of reddish- brown bait at the surface with yellowfin tuna ( Neothunnus macropterus ) feeding voraciously on them. The fish, as described by Sorensen, were in tight elliptical aggregations measuring about 10 ft. in length and 3 ft. in depth. The tuna feeding upon the myctophids refused the fish ( Cetengraulis mysticetus) used by this ves- sel for live bait when it was offered to them. A second capture of the species at the surface during daylight hours was reported by a crew member of the tuna clipper "Sun Traveler”; the capture was made on March 15-20, 1954, 65 mi. southwest of Cano Island, Costa Rica. Mr. Joseph Silva observed the fish at 10:00. They also appeared as a reddish-brown "ball” being fed on by skipjack ( Katsuwonus pelamis ) , yel- 1 Inter-American Tropical Tuna Commission, La Jolla, California. lowfin tuna, and sea birds. Four other small schools were seen in the same locality. The clip- per left the immediate area at this time and re- turned at 16:00 the same day. At this time, the schools of myctophids were still on the surface and when the vessel drew close they took shelter under the stern, evidently for protection from the yellowfin and skipjack. Specimens were easily captured with a dip net from the racks on the stern of the vessel. A third occurrence of this species on the sur- face during daylight hours was noted on April 7, 1955, at 13° 19' N. and 91° 16' W. by the crew of the tuna clipper "Commadore.” The fish were again captured with a dip net from the stern of the vessel while it was fishing for yellowfin. The fish again appeared as reddish- brown "balls” on the surface. The table below gives the size range and mean standard length for the three samples of Benthosema. Although only three samples have been ob- tained by the author, apparently the sighting by fishermen of myctophid schools on the sur- face during daylight hours in the spring of the year is not uncommon. Mr. Edward Silva, man- aging owner of the "Santa Rosa,” says that fish fitting the description of myctophids have been taken by clippers and attempts made to use them as tuna bait. Flowever, the fish scaled easily, were weak, and were unsatisfactory for bait. Mr. Manuel Luz, former skipper of the "Saratoga’” said he has seen many small "balls” of bait in the waters off Costa Rica during the spring of the year that fitted the same descrip- tion. He also noted that the fish made poor bait because they scaled easily and were too weak for retention in the bait wells. NO. DATE OF CAPTURE NO. MEASURED MEAN STANDARD LENGTH (mm.) RANGE IN STANDARD LENGTH (mm.) 1 March 14, 1952 49 39.0 35-47 2 March 15-20, 1954 43 46.3 38-54 3 April 7, 1955 50 35.2 32-40 483 News Note The Society for Economic Botany, an in- ternational scientific organization, invites your participation and welcomes your membership. The objective of the Society is the promotion and stimulation of scientific studies of plants useful to man. The methods of achieving this objective are annual meetings, technical confer- ences, and publication of Economic Botany, the official journal of the Society. Present members represent nearly all scientific disciplines: anthropology, pharmacology, geog- raphy, agriculture, chemistry, physics, and all fields of botanical science. The only prerequisite for membership is an interest in the study of the past, present, or future use of plants to man. Membership dues are $7.50 (U.S.) annually, and includes one year’s subscription to Economic Botany. The journal publishes monographs, re- view articles, and original manuscripts on all subjects relating to the use of plants and plant products. Inquiries and dues can be sent directly to the Treasurer, Richard M. Klein, The New York Botanical Garden, Bronx Park, New York 58, N. Y. 484 should not be fastened together in any way, and should be mailed flat. Inserts should be either typed on separate sheets or pasted on proper page, and point of insertion should be clearly indicated. Original copy and one carbon copy of manuscript should be submitted. The author should retain a car- bon copy. Although due care will be taken, the editors cannot be responsible for loss of manuscripts. Introduction and Summary. It is desirable to state the purpose and scope of the paper in an introductory paragraph and to give a summary of results at the end of the paper. Dictionary Style. It is recommended that authors fol- low capitalization, spelling, compounding, abbrevia- tions, etc., given in Webster’s New International Dic- tionary (unabridged), second edition; or, if desired, the Oxford Dictionary. Abbreviations of titles of pub- lications should, if possible, follow those given in World List of Scientific Periodicals. Examples of the usual abbreviations are to be found in Pacific Science. Footnotes. Footnotes should be used sparingly and never for citing references (see later). When used, footnotes should be consecutively numbered by supe- rior figures throughout the body of the paper. Foot- notes should be typed in the body of the manuscript on a line immediately below the citation, and sepa- rated from the text by lines running across the page. Citations of Printed Sources. All references cited should be listed alphabetically by author at the end of the paper, typed double-spaced. References to books and to papers in periodicals should conform to the following models : Batzo, Roderick L., and J. K. Ripkin. 1849. A Treatise on Pacific Gastropods. Rice and Shipley, Boston, vii + 326 pp., 8 figs., 1 map. Crawford, David L. 1920^. New or interesting Psyllidae of the Pacific Coast (Homop.). Proc. Hawaii. Ent. Soc. 4(1) : 12-14. 1920A Cerotrioza (Psyllidae, Homoptera). Proc. Hawaii. Ent. Soc. 4(2) : 374-375. Rock, Joseph F. 1916. The sandalwoods of Hawaii; a revison of the Hawaiian species of the genus Santalum. J. Aust. Inst. Agric. Sci. 12(3): 1-43, 13 pis. In the text, sources should be referred to by author, date, and page, as follows: "It was noted (Rock, 1916: 18) that . . .” or "Rock (1916: 21-24) says . . .” Quotations. Quoted matter of fewer than five printed lines (about 200 characters) should be given in the text in the usual form, using double quote marks. Longer quotations should be set flush with left mar- gin. The author is responsible for the accuracy of quoted material. Numbers. Decimals, measurements, money, percent- ages, time; enumerations in which any figure is 10 or over; and isolated enumerations of 10 and over should be given in Arabic figures, rather than spelled out, except when the number begins a sentence. ILLUSTRATIVE MATTER Only the minimum number of illustrations required to supplement the text will be accepted by the editors. Reproduction costs of illustrations in excess of the number allowed by the editors will be paid by the author. Illustrations should accompany manuscript, on sep- arate sheets. Often more valuable than a photograph is a good line drawing, which may be reproduced by the zinc etching process. Figures and Graphs. Copy for figures and graphs should always be drawn large enough to allow for at least one-third reduction by the engraver. Copy should consist of carefully prepared line drawings in one color only, drawn in India ink on plain white draw- ing paper or tracing cloth. Co-ordinate paper with lines in light blue (a color which will not show in a photograph) may be used; but co-ordinates which should appear in the finished graph must be drawn in India ink. If original figures may not be conven- iently submitted with manuscript, duplicate rough sketches or photographic prints may be furnished to aid the editors in their decisions. It is strongly urged that an indication of scale be incorporated as a part of all drawings in which mag- nification and size are critical considerations. Photographs. Photographs should be chosen for clarity in portraying essential information. They should be printed for contrast, on glossy paper, and should be sent unmounted. They should be identified with serial number written in soft pencil on the back to corre- spond with list of captions. Illustrations will be returned to the author. Tables. Tabular matter should be kept to a minimum. Each table, prepared to conform with Pacific Science style, should be typed on a separate page, and its posi- tion indicated on the manuscript. Captions. Readily identifiable captions for figures, graphs, photographs, and other illustrative matter should be supplied on a separate page. PROOF Proof should be corrected immediately and returned at once to Robert Sparks, assistant to the editors. Authors are reminded that the editors will allow only a minimum number of corrections on galley proof. Additions to the printed text and changes in style and content are not permitted. All queries on proof should be answered. If cor- rected proof is not received within four weeks after being sent to the author, author’s changes cannot be accepted. REPRINTS Reprints or separates should be ordered on the form provided and returned with author’s proof. All correspondence concerning separates must be directed to the printer, Star-Bulletin Printing Company, Inc., 420 Ward Avenue, Honolulu 14, Hawaii. w v . OCTOBER 1961 NO. 4 <0 <5‘ C | 111 VOL. XV PACIFIC SCIENCE A QUARTERLY DEVOTED TO THE BIOLOGICAL AND PHYSICAL SCIENCES OF THE PACIFIC REGION W. L. KLAWE Larvae, Juveniles, and Spawning of Bonito ROBERT F. SCAGEL Benthonic Algae in Queen Charlotte Strait D. J. WOODLAND New Species of Pranesus from the Great Barrier Reef OLGA HARTMAN New Pogonophora from the Eastern Pacific MAXWELL S. DOTY Possible Invader of the Marine Flora of Hawaii ROBERT BIERI Food of the Pelagic Coelenterate, Velella lata R. WINKLER and BERNARD E. TILTON Interspecific Differences in California Sea Hares PETER J. R. HILL and BENJAMIN C. STONE Vegetation of Yanagi Islet, Truk HAROLD ST. JOHN Revision of the Genus Pandanus Part 6. New Species from Queensland Part 7. New Species from Papua and the Solomon Islands J. S. WATSON Feral Rabbit Populations on Pacific Islands E. W. JAMESON, JR. Relationships of the Red-backed Voles of Japan E. S. HOBSON, F. MAUTIN, and E. S. REESE Two Shark Incidents at Eniwetok Atoll BENJAMIN C. STONE Oceanic Dispersal of Pandanus pistillaris UNIVERSITY OF HAWAII PRESS BOARD OF EDITORS O. A. BUSHNELL, Editor-in-Chief Department of Microbiology, University of Hawaii Robert Sparks, Assistant to the Editors Office of Publications and Information, University of Hawaii Agatin T. Abbott Department of Geology and Geophysics University of Hawaii Thomas S. Austin Bureau of Commercial Fisheries, Hawaii Area (U. S. Fish and Wildlife Service) Honolulu, Hawaii Albert J. Bernatowicz Department of Botany University of Hawaii L. H. Briggs Department of Chemistry University of Auckland Auckland, New Zealand Ai Kim Kiang Department of Chemistry University of Malaya, Singapore Colin S. Ramage Department of Geology and Geophysics University of Hawaii PAUL J. SCHEUBR Department of Chemistry University of Hawaii Donald W. Strasburg Bureau of Commercial Fisheries, Hawaii Area (U. S. Fish and Wildlife Service) Honolulu, Hawaii Albert L. Tester Department of Zoology and Entomology University of Hawaii Miklos F. Udvardy Department of Zoology University of British Columbia Vancouver, Canada Thomas Nickerson, Managing Editor Office of Publications and Information, University of Hawaii SUGGESTIONS TO AUTHORS Contributions to Pacific biological and physical science will be welcomed from authors in all parts of the world. (The fields of anthropology, agriculture, engineering, and medicine are not included.) Manu- scripts may be addressed to the Editor-in-Chief, PACIFIC SCIENCE, University of Hawaii, Honolulu 14, Hawaii, or to individual members of the Board of Editors. Use of air mail is recommended for all communications. Manuscripts will be acknowledged when received and will be read promptly by members of the Board of Editors or other competent critics. Authors will be notified as soon as possible of the decision reached. Manuscripts of any length may be submitted, but it is suggested that authors inquire concerning possi- bilities of publication of papers of over 30 printed pages before sending their manuscripts. Authors should not overlook the need for good brief papers, presenting results of studies, notes and queries, com- munications to the editor, or other commentary. PREPARATION OF MANUSCRIPT It is requested that authors follow the style of Pacific Science described herein and exemplified in the journal. Manuscript Form. Manuscripts should be typed on one side of standard-size, white bond paper and double-spaced throughout. Pages should be consecu- tively numbered in upper right-hand corner. Sheets (Continued on inside back cover) PACIFIC SCIENCE A QUARTERLY DEVOTED TO THE BIOLOGICAL AND PHYSICAL SCIENCES OF THE PACIFIC REGION VOL. XV OCTOBER 1961 NO. 4 Previous issue published July 12, 1961 CONTENTS PAGE Notes on Larvae , juveniles, and Spawning of Bonito (Sarda) from the Eastern Pacific Ocean. W . L. Klatve ... . 487 The Distribution of Certain Benthonic Algae in Queen Charlotte Strait, British Columbia, in Relation to Some Environmental Factors. Robert F. Scagel ... 494 ' Description of a New Species of Pranesus (Atherinidae: Pisces) from the Capricorn Group, Great Barrier Reef. D. J. Woodland 540 New Pogonophora from the Eastern Pacific Ocean. Olga Hartman 542 Acanthophora, a Possible Invader of the Marine Flora of Hawaii. Maxtvell S. Doty 547 Post-larval Food of the Pelagic Coelenterate, Velella lata. Robert Fieri 553 Interspecific Differences in the Reaction to Atropine and in the Histology of the Esophagi of the Common California Sea Hares of the Genus Aplysia. Lindsay R. Winkler and Bernard E. Tilton .. 557 The Vegetation of Yanagi Islet, Truk, Caroline Islands. Peter J. R. Hill and Benjamin C. Stone 56 1 Revision of the Genus Pandanus Stickman, Part 6. New Pandanus Species from Queensland, Australia. Harold St. John 563 Revision of the Genus Pandanus Stickman, Part 7 . Netv Species from Borneo, Papua, and the Solomon Islands. Harold St. John 576 Feral Rabbit Populations on Pacific Islands. J. S. Watson 591 Relationships of the Red-backed Voles of japan. E. W. Jameson, Jr 594 Tivo Shark Incidents at Eniwetok Atoll, Marshall Islands. E. S. Hobson, F. Mautin, and E. S. Reese 605 Pandanus pistillaris in the Caroline Islands: An Example of Long-Range Oceanic Dispersal. Benjamin C. Stone 6 10 INDEX ... 615 Pacific Science is published quarterly by the University of Hawaii Press, in January, April, July, and October. Subscription price is $4.00 a year; single copy, $1.25. Check or money order payable to University of Hawaii should be sent to University of Hawaii Press, Honolulu 14, Hawaii, U. S. A. Printed by Star-Bulletin Printing Company, Inc., 420 Ward Avenue, Honolulu 14, Hawaii. ■ft Notes on Larvae, Juveniles, and Spawning of Bonito ( Sarda ) from the Eastern Pacific Ocean W. L. Klawe 1 Two species OE bonito occur along the Pacific coast of the Americas (Hildebrand, 1946) 2 Sarda chiliensis ranges from southern British Columbia to northern Chile; the other species, S. orientalis , is found from Baja California to the Galapagos Islands and northern Peru. These species are of considerable importance to both commercial and sport fisheries of certain coun- tries (Walford, 1937; Berdegue, 1956). The bonito fishery in Peru, especially for S. chilien- sis, is particularly important. It contributes sig- nificantly to the economy, being valuable to the domestic market as well as for export. The fol- lowing table gives the commercial catches of bonito, expressed in thousands of metric tons, during the past few years : OFF PERU OFF CALIF. WORLD TOTAL 1950 31.8 0.3 61.6 1951 51.2 0.4 76.9 1952 50.3 1.0 76.1 1953 44.4 1.4 83.8 1954 52.8 1.1 114.4 1955 71.8 0.1 173.1 1956 83.4 0.1 189-9 1957 58.6 0.1 145.5 1958 66.2 2.4 163.5 Despite the commercial importance of bo- nitos, knowledge of their biology is relatively meager. Little is known of the early life his- tory and spawning, and descriptions of the eggs, larvae, and juveniles of Sarda are limited to a few reports. The only descriptions of young 1 Inter- American Tropical Tuna Commission, La Jolla, California. Manuscript received September 26, I960. 2 It is assumed that S. lineolata (Girard) ~S. chilien- sis (Cuvier and Valenciennes) and that S. velox Meek and Hildebrand = S: orientalis (Temminck and Schle- gel). S. chiliensis and S. sarda from the Atlantic Ocean and adjacent seas may be synonymous. A detailed dis- cussion on the relationship of the species of the genus Sarda is provided by Godsil (1955). bonitos from the eastern Pacific Ocean are those of S. chiliensis by Barnhart ( 1927 ) and Orton (1953*, 1953 h). Descriptions of eggs and young from other parts of the world seas are limited to the reports listed in Table 1. The information on spawning of bonito is scant. According to Yildoso (1955), S. chilien- sis reproduces in Peruvian coastal waters from October to March with the peak of spawning occurring from December to February. From Barnhart’s work (1927) it is' evident that off La Jolla, California, S. chiliensis spawns in June. Walford (1937) stated that in the northern latitudes this species spawns during the late spring and summer close to shore to as far out as 80 mi. Observations on spawning bonito in other parts of the world are limited to the reports on S. sarda tabulated below: SPAWNING AREA TIME REFERENCE Mediterranean.. May-jun. Sanzo, 1932 Black Sea Apr.-Aug. Borcea, 1939, 1933 Vodyanitsky, 1936 Malyatsky, 1940 Muman, 1955 Atlantic, off Morocco.. Jun —Jul. Furnestin et ah, 1958 off Dakar Feb --Mar Frade and Post'd., 1955 U.S. coast Jul. Sette, 1943 Jun. Bigelow and Schroeder, 1953 During the course of collecting biological material and other scientific data from commer- cial fishing vessels and research ships operating in waters of the eastern Pacific Ocean, staff members of the South Pacific Fishery Investiga- tions of the U. S. Bureau of Commercial Fisher- ies, Scripps Institution of Oceanography, and the Inter-American Tropical Tuna Commission have captured a variety of larval and juvenile fish, including young of Sarda. The data on the collections of young bonito made by these or- ganizations with some other records, kindly put 487 488 PACIFIC SCIENCE, Vol. XV, October 1961 Fig. 1 . Localities of capture of young Sarda in eastern Pacific Ocean. at the Commissions disposal, are given in Table 2. The geographical distribution of these catches is shown in Figure 1. The author wishes to thank Mr. Gerald V. Howard and Dr. Elbert H. Ahlstrom of the U. S. Bureau of Commercial Fisheries for their helpful criticism and suggestions on the prep- aration of this manuscript. The manuscript upon its completion in 1959 was made available to Mr. Leo Pinkas of the California Department of Fish and Game. Thanks are ’extended to Sra. Aurora Chirinos de Vildoso of the Peruvian Pesca y Caza for her cooperation in exchange of various data. Notes on Bonito — Klawe 489 TABLE 1 SPECIES MONTH COLLECTED PLACE SIZE OR STAGE REFERENCE S', sarda.... Jun. Mediterranean 7.2 mm. Ehrenbaum, 1924 S. sarda....... Jul. Mediterranean 26.5, 32 mm.* de Buen, 1930 S. sarda............... Jun. Mediterranean eggs and early larvae Sanzo, 1932 S. sarda............... Jun. Black Sea eggs and early larvae Vodyanitsky, 1936 S' sarda Mar. Gulf of Mexico 64, 67 mm. Klawe and Shimada, 1959 S. sarda............... Feb. Western Atlantic 34 mm. Klawe, in press S. orientalis Apr. Japan 170 mm. Kishinouye, 1923 S. orientalis.. Sep. Japan 230 mm. Kishinouye, 1923 * The 17.5 mm. individual represented by de Buen (1930) in figure 1 and originally identified by him as Sarda sarda has 39 vertebrae and was reidentified by him as Auxis thazard (de Buen, 1932) DESCRIPTION AND IDENTIFICATION OF LARVAE AND JUVENILES Two larvae of total lengths 2.9 and 3.5 mm. captured in a plankton tow off Baja California (Table 2) have been previously identified as those of Sarda by Alhstrom. An illustration of the larger specimen is given in Figure 2. These larvae have the usual characteristics of larval scombroid fishes. When the material was com- pared with Sanzo’s ( 1932 ) and Vodyanitsky ’s (1936) descriptions and illustrations of larvae of S. sarda hatched in the laboratory, the sim- ilarity was obvious, although their specimens were less advanced in development. Sanzo and Vodyanitsky report that their lavae measured about 4.3 mm. However, Sanzo and Vodyanit- sky were working with fresh material, while the planktonic larvae caught off Baja California had been preserved in formalin. Barnhart’s (1927) observations concerning larval bonito are based on material hatched in the laboratory from eggs collected in a plank- ton net off La Jolla, California. One of the striking differences between his description and that of S. sarda by Sanzo and Vodyanitsky and that of S. cbiliensis by Orton (1953*, 1953#) is the absence of melanophores in Barnharts material. Barnhart credits the newly hatched bo- nito with a few yellow chromatophores; on the other hand, Sanzo and Vodyanitsky show that in addition to the yellow pigment there are melanophores on the eggs just prior to hatch- ing and the larvae are also provided with nu- merous melanophores. Barnhart based his de- scription upon material which probably had been preserved in Bonin’s solution, which has the property of fading melanine according to Orton (1955). Fig. 2. A 3-5 mm. Sarda sp. caught on August 12, 1951, off Baja California. 490 PACIFIC SCIENCE, Vol. XV, October 1961 TABLE 2 Records of Capture of Developing Eggs and Young Sarda from the Eastern Pacific Ocean DATE GENERAL LOCALITY LATI- TUDE LONGI- TUDE HOW CAPTURED SIZE, 1 mm. NO. SPECIES REFERENCE OR COLLECTOR Jun. 30, 1927 off La Jolla, plankton eggs p chiliensis Barnhart, Calif. net 1927 May 17-18, off La Jolla, night light, post- 1 chiiiensis Hubbs 2 1947 Calif. dip net larva Feb. 20, 1951 off Pt. Lobos, p 125-130 3 chiliensis Vildoso, 1955 Peru Aug. 5, 1951 off Baja Calif., 25° 35' 113° 56' night light, 42 1 chiliensis Fish & Wild- Mexico N. W. dip net life Service 3 Aug. 12, 1951 off BallenasBay, 26° 29.5' 113° 29.2' plankton 2.9, 3.5 2 p Fish & Wild- Baja Calif., N. W. net life Service 3 Mexico Jan. 3, 1956 off Pt. Negra, bait net 143-164 3 chiliensis M.V. Peru "Corsair’ 1 Dec. 10, 1957 Independencia 14° 14' 76° 12' bait net 128 1 chiliensis B.M. Chatwin Bay, Peru S. W. & P. Boylan 4 Dec. 18, 1957 off Ilo, Peru bait net 70-135 3 chiliensis Arnold Neves 1 Dec. 31,1957 off Pt. Pichalo, 19° 35' 70° 16' night light, 37 1 chiliensis B.M. Chatwin Chile S. W. dip net & P. Boylan 4 Jan. 3, 1958 off Pt. Dos 24° 30' 70° 49' night light, 39 1 chiliensis E. Brinton 5 Reyes, Chile S. W. dip net Jan. 3, 1958 off Pt. Dos 24° 36' 7i° or night light, 36-44 3 chiliensis E. Brinton’’ Reyes, Chile S. w. dip net Jan. 14, 1958 SW off Fraile 13° 14.8' 77° 55.5' night light, 34 1 chiliensis E. Brinton 5 Pt., Peru S. W. dip net Feb. 1958 off Chimbote, bait net - 89-129 3 chiliensis Arnold Neves' 1 Peru May 6, 1958 Almejas Bay, bait net 103 1 chiliensis M.V. Baja Calif., "Lou Jean” 1 Mexico Feb. 1, 1959 Sama Cove, bait net 111 1 chiliensis M.V. Peru "Norman- die” 4 Mar. 15, 1959 off Ilo, Peru p 135-160 3 chiliensis P. Talledo 6 Mar. 16, 1959 off Barranca, p 173, 199 2 chiliensis Vildoso' 1 Peru 1 Direct distance from tip of snout to tip of shortest median caudal ray. 2 Field-book data of C. L. Hubbs, Scripps Institution of Oceanography, La Jolla, Calif. 3 In the collection of the South Pacific Fishery Investigations of the U.S. Fish and Wildlife Service, La Jolla, Calif. 4 In the collection of the Inter-American Tropical Tuna Commission, La Jolla, Calif. 5 In the collection of the Scripps Institution of Oceanography, La Jolla, Calif. 0 In the collection of the laboratory of Pesca y Caza, Lima, Peru. The planktonic specimens caught off Baja California have between 4$ and 45 myomeres, the actual number being difficult to establish. The mouth is large and contains a set of large teeth. There are three spines along the posterior edge of the preoperculum. The unpaired fins show little development and are still represented by the median fin fold. The embryonic pec- torals can be distinguished but there is no in- dication of the pelvics. The pigmentation of the head consists of distinct chromatophores over the area of the brain and an aggregation of what appears to be a group of chromatophores at the symphysis of the pectoral girdle. There Notes on Bonito— Klawe 491 Fig. 3. A 42 mm. Sarda chiliensis caught on August 5, 1951, off Baja, California. Drawing by George Mattson of the U. S. Bureau of Commercial Fisheries. are concentrations of chromatophores in various parts of the peritoneum but especially along its dorsal portion. A series of chromatophores can also be seen along the mid-ventral line extending almost to the tip of the urostyle. Identification of the two specimens, as to which of the two species they represent ( S . chiliensis or S. orientalis ) , is impossible with- out a complete developmental series. Other specimens of Sarda available from the eastern Pacific Ocean are several individuals over 30 mm. in length caught off Chile and Peru and two individuals caught off Baja Cali- fornia. On the basis of gill raker counts (see next paragraph), all individuals have been identified as S. chiliensis. A 43 mm. individual caught off Baja California is illustrated in Fig- ure 3. The general appearance of these juveniles resembles that of Euthynnus sp. However, sep- aration between these fishes is possible, as the outline of the first dorsal fin is concave in Euthynnus and convex in Sarda when the spines are erect. Upon closer examination the two genera can be separated easily, as the meristic counts differ considerably. The densely pig- mented pelvic fins shown in Figure 3 are characteristic of small Sarda . It is worthwhile to mention that the first dorsal fin in small S. Fig. 4. A 160 mm. Sarda chiliensis caught on January 3, 1956, off northern Peru. 492 PACIFIC SCIENCE, Vol. XV, October 1961 chiliensis is either dark with a lighter area in the posterior region of that fin or is uniformly dark. The bodies of all juvenile S. chiliensis examined are striped vertically. This also ap- plies to other species of Sarda (Kishinouye, 1923; de Buen, 1930; Furnestin et al., 1958; Klawe and Shimada, 1959). A 160 mm. in- dividual of S. chiliensis is illustrated in Figure 4. The vertical striations still persist. According to VildosA|(1955), these vertical bars are re- placed by-pblique markings characteristic of S. chiliensis at an approximate length of 320 mm. It should be mentioned that adult Sarda chilien- sis, when captured, briefly exhibit dusky ver- tical bands quite like those on preserved juve- niles, although not as intense. The body shape of the specimen illustrated in Figure 4 is like that of an adult. The pigmentation of the pelvic fins at this size is confined only to the more basal portion. As the gill rakers are one of the characteristics used for separating adults of S. chiliensis from S' orientalis, it is important to establish at what length they develop and at what length the final complement is attained. The gill raker counts for adult fish are 7 to 10 + 12 to 19 for S. chiliensis 3 and 2 to 3 + 6 to 10 for S. orientalis . Some idea can be obtained from the following counts made on several smaller S'. chiliensis caught off Peru and Chile. SIZE, mm. NO. OF GILL RAKERS SIZE, mm. NO. OF GILL RAKERS 34 7 + 1 + 15 89 7 + 1 + 17 36 7 + 1 + 15 94 9+1 + 16 37 4+1 + 16 106 7+1 + 15 39 5 + 1 + 14 111 8 + 1 + 17 41 7 + 1 + 15 128 9+1 + 17 44 7+1 + 16 129 7 + 1 + 15 70 7 + 1 + 18 135 | 7 + 1 + 15 CONCLUSIONS All evidence indicates that spawning of S. chiliensis takes place in the warmer season off California, Baja California, Peru, and northern 3 Author examined 25 specimens of S. chiliensis caught off San Diego, California, on September 22, 1959, ranging in length from 330 mm. to 380 mm. The gill raker counts for this group of fish were 7 to 9+1 + 15 to 18. Chile. It appears that it should be possible to distinguish between juveniles of S. chiliensis and S. orientalis by means of gill raker counts even at a relatively small size. REFERENCES Barnhart, P. S. 1927. Pelagic fish eggs off La Jolla, California. Bull. Scripps Instn. Oceanogr. Tech. Ser. 1(8): 91-92. Berdegue A., Julio. 1956. Peces de importancia comercial en la costa nor-occidental de Mex- ico. Comision para el Fomento de la Piscicul- tura Rural. 345 pp. Bigelow, H. B, and W. C. Schroeder. 1953. Fishes of the Gulf of Maine. U.S. Fish. Bull. 53(74): 1-577. Borcea, I. 1929. Observations sur les poissons migrateurs dans les eaux Roumaines de la Mer Noire. Ann. Sci. Univ. Jassy 15: 656— 750. 1933. Nouvelles observations sur les migrations et sur le periode de ponte des especes de poissons migrateurs de la Mer Noire. Ann. Sci. Univ. Jassy 17: 503-564. Buen, Fernando de. 1930. Estados larvarios y juveniles de la Sarda sarda (Bloch). Inst. Esp. Oceanogr. Trab. (3): 3-32. 1932. Formas ontogenicas de peces (Nota primera). Inst. Esp. Oceanogr. Notas Ser. 2, (57): 1-38. Ehrenbaum, E. 1924. Scombriformes. Rep. Danish Oceanogr. Exped. 1908-10 Medit. In: Biology 2(8): 1-42. Frade, F., and E. PosTEL. 1955. Contribution a l’etude de la reproduction de Scombrides et Th on ides de 1’Atlantique Tropical. Rapp. Cons. Explor. Mer 137: 33-35. Furnestin, J., J. Dardignac, C. Maurin, A. Vincent, R. Coupe, and H. Boutiere. 1958. Donnees nouvelles sur les poissons du Maroc Atlantique. Rev. Trav. Inst. Peches Marit. 22(4): 379-493. Notes on Bonito — Klawe 493 Ggdsil, H, G 1955. A description of two spe- cies of bonito Sarda orientals and S. chilien- sis and a consideration of relationship within the genus. Fish Bull. Sacramento 99, 43 pp. Hildebrand, Samuel F. 1946. A descriptive catalogue of the shore fishes of Peru. U.S. Nat. Mus. Bull 139, 530 pp. Kishinouye, Kamakichl 1923. Contributions to the comparative study of the so-called scrombroid fishes. Tokyo Coll. Agric. J. 8(3): 293-475. Klawe, W. L. Young scombroids from the waters between Cape Hatteras and Bahama Islands. Bull. Mar. Sci. Gulf Caribb.: in press. Klawe, W. L., and Bell M. Shxmada. 1959. Young scombroid fishes from the Gulf of Mexico. Bull. Mar. Sci. Gulf Caribb. 9(1) : 100-115. MalyaTSKY, S. M. 1940. Materiali po ekologi naselenia pelagiali Chernogo moria. Trudy Nov. Biol. Sta. 2(3): 237-258. Numann, Wilhelm. 1955. Die Pelamiden des Schwarzen Meeres, des Bosporus, der Mar- mara und der Dardanellen ( Sarda sarda). Publ. Hydrobiol. Res. Inst. IJniv. Istanbul Ser. B, 3(2-3): 75-127. Orton, Grace L. 1953^. The systematics of vertebrate larvae. System. Zool. 2(2): 63-75. — — — 1953A Development and migration of pigment cells in some teleost fishes. J. Morph. 93(1): 69-99. — 1955. Early developmental stages of the California barracuda, Sphyraena argentea Girard. Calif. Fish Game 41(2): 167-176. SANZO, L. 1932. Uova e primi stadi larvali di Pelamys sarda C. V. R. Com. Talassogr. ItaL Mem. 188: 1-9. SETTE, O. E, 1943. Biology of the Atlantic mack- erel ( Scomber scombrus ) of North America, Part I. Early life history, including the growth, drift, and mortality of egg and larval popula- tions. U. S. Fish. Bull. 50(38): 149-237. Vildoso, Aurora Chirinos de. 1955. Estudio preliminar sobre el "bonito” Sarda chilensis (Cuvier & Valenciennes). Direcc. Pesqueria Caza Peru, Pesca Caza (6): 1-38. VODYANXTSKY, V. A. 1936. Nabliudenia nad pelagicheskimi iaitsami ryb Chernogo moria. Trudy Sevastopol. Biol. Sta. 5: 3-43. Walford, L. A. 1937. Marine game fishes of the Pacific Coast from Alaska to the Equator. Contr. Santa Barbara Mus. Nat. Hist., Univ. Calif. Press. 205 pp. The Distribution of Certain Benthonic Algae in Queen Charlotte Strait, British Columbia, in Relation to Some Environmental Factors Robert F. Scagel 1 In comparison with the progress in our knowl- edge of most groups of plants — especially con- cerning their life histories and distributions — the advances made in marine phycology and marine ecology have been relatively slow. The limited access to living material or to the facil- ities to maintain the larger marine algae in the living condition for a prolonged period of time, the difficulties of collection-— -particularly in the subtidal zone — and the lack of any extensive direct economic importance until recent years have all contributed to this slow progress. How- ever, in spite of these difficulties there has been a considerable amount of interest in the marine algae, including a number of studies of their ecology. Although this interest has been fairly widespread in a number of countries, until re- cently there has been little activity in the field of marine ecology relating to the benthonic algae on the Pacific Coast of North America and nothing of a comprehensive nature has been published for this area. It is an anachronism that this should be so in a region which received such prominence some 50 years ago through the efforts of a pioneer in the field, the late William Albert Setchell (1893, 1917, 1935). Knowledge of the effect of temperature on the world-wide distribution of plants both hor- izontally and vertically had developed gradually over a period of many years. However, it was only during the last hundred years that the at- tention of phycologists was brought to a con- sideration of the reasons for the observed dis- tributions of the marine algae. The historical development of this trend of thought and in- vestigation has been reviewed by Setchell (1917). Starting over 50 years ago, through a series of papers from 1893 to 1935, Setchell made a noteworthy attempt to explain the world-wide distribution of marine algae, espe- 1 Department of Biology and Botany, and Institute of Oceanography, University of British Columbia, Vancouver 8, Canada. Manuscript received March 31, I960. dally of members of the Laminariales on the Pacific Coast of North America, on the basis of latitudinal and seasonal temperature distribu- tions. The physical data available during this early period were limited, but many of the prin- ciples set forth by Setchell concerning the dis- tributions of marine algae are as sound now as when they were first proposed. Except for more precise knowledge of the physical and chemical factors of the environment and the distribu- tions of the algae concerned, much of Setchell’s ecological work can still be used as a good foundation for further study. Although it was largely a two-dimensional approach to the marine environment, Setchell’s work made a significant contribution to the development of marine algal ecology. Lamouroux (1825, 1826) had suggested the possibility that temperature stratification in the sea might account for the vertical distribution of the marine algae and had considered the effect of tides on intertidal zonations, but this trend to analyze the vertical distribution of the marine algae was not generally taken up in de- tail until much later. Coleman (1933) was one of the first to emphasize the use of tide levels to account for the vertical distribution of the marine algae in the intertidal zone. In a study in Oregon, on the Pacific Coast of the United States, Doty ( 1946) has given further evidence for the relationship between the vertical dis- tributions of marine algae and critical tide levels. A number of lists of marine algae have been published and attempts have been made not only to relate the floras of one area to another, such as that by Okamura (1926, 1932) in the North Pacific, but also to account in a general way for distributions on the basis of ocean cur- rents, such as that by Isaac (1935) in the area around South Africa and by Tokida (1954) in the region of northern Japan. However, there soon followed a decided shift to intertidal studies of regional areas, such as that by Feldmann (1937) in the Mediterranean and Chapman 494 Benthonic Algae — ScagEL 495 131 ° W. 125° W. ALASKA BRITISH COLUMBIA MAINLAND 54° N NORTH PACIFIC OCEAN 49° N 49° N. 130* W. FIG. 1. Map of geographical features of the coast of British Columbia. 496 PACIFIC SCIENCE, Vol. XV, October 1961 ( 1950) and his students in New Zealand. Some attempts have also been made to describe uni- versal features of intertidal zonation through- out the world ( Stephenson and Stephenson, 1949). At the same time there has been a tend- ency to place greater emphasis on the inter- relationships between the various organisms. Many of these intertidal studies have been of great value as an initial descriptive stage of investigation, and there is a need for further descriptive studies of this type in new and un- described regions. However, the variety of sys- tems of nomenclature and terms that have been proposed by marine ecologists to describe zona- tion, associations, and other ecological concepts has frequently only complicated the descriptive study rather than succeeded in explaining the observed phenomena. This has led to some con- fusion in terminology. It is a debatable point whether there can be such a thing as a universal system of classification beyond a generalized scheme, such as that proposed by Ekman (1935), and it is questionable whether some of the sys- tems proposed can contribute further to progress in marine algal ecology even in regional studies without simplification or clarification. There have been a number of recent comprehensive papers dealing with various aspects of marine ecology which make it unnecessary to dwell at length on a review of the trends that have been followed more recently in marine algal ecology and the results that have been attained ( Gislen, 1929, 1930; Feldmann, 1937, 1951; Fischer- Piette, 1940; Chapman, 1946, 1957; Doty, 1957; Hartog, 1959). Although the shift in emphasis to the inter- relationship of organisms was an important one, in some instances this approach has been re- sponsible for excluding adequate concurrent studies of the physical and chemical aspects of the environment. It is for this reason that a case may be made for reassessing the status of marine algal ecology, and a critical evaluation of the steps to be taken to further its progress is timely. Perhaps what may be called a three-dimensional or an oceanographic approach can be used to analyze more precisely various factors in the marine environment and the relationship of these factors to the benthonic algae. Steps in this direction have been made more recently Fig. 2. Map of Queen Charlotte Strait showing depth contours. Benthonic Algae — SCAGEL 497 SALINITY %, Fig. 3. T-S diagrams for stations in Queen Charlotte Strait and adjacent areas in June, 1953. with some measure of success by Dawson (1945, 1951) in Baja California, Doty (1946) in Ore- gon, and Womersley (1956) in Australia. The physical environment of the sea imposes problems of considerable magnitude which, in contrast to many land environments, presents formidable obstacles such as complex tidal and circulation patterns. In an effort to manipulate or simulate some aspects of this environment, both in the field as well as in the laboratory, the more generally available resources are soon taxed. Thus the methods that have been used in marine algal ecology have, in many instances of neces- sity, been crudely quantitative, less than ideal experimentally, frequently encumbered by ter- minology as a result of limited concurrent phys- ical and chemical data, and sometimes they have failed to establish clearly the objectives being sought. As a result most of the efforts in marine algal ecology have been descriptive rather than functional in nature. As in most oceanographic work there is value in an approach from the grosser aspects to the particular. To the ocean- ographer the most complicated physical or chem- ical situation to explain may be the smallest unit of the environment with which he is faced. This is partly a problem of instrumentation. However, it is usually much easier to recognize significant discontinuities in properties, such as temperature, salinity, and even plankton dis- tributions, over extensive areas of the ocean than in restricted or local regions. It is also easier to use such information in describing dynamic processes. Hence, it is suggested that more attention should be given to studies of the general distribution of various physical and chemical properties in the marine environment in an attempt to set up some workable hypo- 498 PACIFIC SCIENCE, Vol. XV, October 1961 theses to account for observed distributions of marine algae. In this way we may hope to ex- plain and account for biological phenomena rather than be satisfied by a description of the phenomena or by devising terms to describe them which do nothing more than give names to dynamic aspects of marine ecology much in need of logical explanation. With the recent increased activity in oceanography in the Pacific we may now hope for more abundant and usable data on some of the more general oceanographic properties of the North Pacific. In specific cases, particularly in more restricted areas, the ecol- ogist will be forced to turn more attention to obtaining in situ physical and chemical data before further progress can be made. One can arbitrarily start by summarizing all the factors in the marine environment as geo- logical, physical, chemical, and biological. The way in which these are considered may be some- what a matter of interpretation. Salinity, for example, may be considered directly, from a chemical standpoint, or indirectly as a physical factor responsible for changes in density and thus contributing to the pattern of circulation. Likewise, the nature of the substratum may be considered indirectly as a geological factor or directly as a physical or mechanical factor re- stricting or permitting establishment of ben- thonic organisms because of particle size. There has been much written on some of these aspects of ecological study in special cases, but it is suggested that, in a general over-all reassess- ment of the environment, an attempt be made to proceed from this more general position to the particular. This approach may initially lead only to the erection of further hypotheses, since the indirect or direct nature of the action of Fig. 4. T-S diagrams for stations in Queen Charlotte Strait and adjacent areas in May, 1956. Benthonic Algae— SCAGEL 499 Fig. 6. Map of station positions in Queen Charlotte Strait in May, 1956. 500 PACIFIC SCIENCE, VoL XV, October 1961 Fig. 8. Salinity distribution in Queen Charlotte Strait in June, 1953, at depths of 10 and 20 m. Benthonic Algae — SCAGEL 501 FlG. 10. Salinity distribution in Queen Charlotte Strait in May, 1956, at depths of 10, 20, and 50 m. 502 PACIFIC SCIENCE, Vol. XV, October 1961 any particular factor in the environment may ultimately be established only by experimental work either in the field or in the laboratory. In some regions where the algal flora is well known taxonomically the descriptive aspect can and has been undertaken and the time has come when further progress in such an area can only be expected by undertaking experimental field and laboratory studies. In some regions, even where the flora may be well known, inadequate physical and chemical data militate against fur- ther progress and even the advancement of tenable hypotheses. Only when such hypotheses are put forward, based on a correlation of the observational data, can one anticipate and justify embarking on an experimental field and lab- oratory approach in an attempt to solve prob- lems relative to the distribution of marine algae. This is the approach that is outlined here in studies on marine organisms and, particularly on the marine algae, being carried out on the coast of British Columbia, some aspects of which will be considered here in detail. As a step in the direction of increasing our knowledge of the entities which comprise the tools of the algal ecologist and of completing this descrip- tive phase of the study of the marine benthonic algae, an annotated check list has been com- pleted for the coast of British Columbia and northern Washington (Scagel, 1957). Based on this list, further studies are now in progress to augment the existing data on the marine flora of British Columbia, not only on distributions but also on life histories, growth, reproduction, and seasonal aspects. These fundamental studies are basic to all other aspects of ecological re- search, especially when an attempt is made to use specific organisms as indicators of ocean- ographic conditions. Not only does this descriptive phase require an adequate consideration of the taxonomic as- pects, but also a complete description of the other factors in the environment is needed. With increased activity recently and currently in gen- eral oceanographic studies of the North Pacific by a number of organizations on the Pacific MONTHS Fig. 11. Monthly salinities at Pine Island for the period 1942-52. Benthonic Algae — SCAGEL 503 MONTHS JFMAMJJAS OND Fig. 12. Monthly salinities at Pulteney Point, Malcolm Island, for the period 1954—55. Coast of Canada, in Japan, and in the United States, there has resulted a considerable body of scientific knowledge but it is still far from adequate for the ecologist, particularly for one interested in coastal dynamics. Circulation pat- terns, temperature, salinity, oxygen, and in some areas phosphate, silicate, and nitrate distribu- tions are fairly well known in a broad and gen- eral sense, but at the present time these ocean- ographic data are known in sufficient detail for few restricted areas. Knowledge of the distribu- tion of other chemical constituents and to a large extent even of the plankton composition, distribution, and activity is almost completely lacking. Much more information is needed in order to tackle many problems relating to spe- cific distributions and to set up field and lab- oratory studies to test hypotheses. It is already apparent that much can be done experimentally both in the field and in the laboratory with the benthonic algae. Many of the problems encountered by the ecologist dealing with large marine algae present unique culture problems both in the field as well as in the lab- oratory. Some studies of growth and reproduc- tion, particularly of some of the larger Lami- nariales, have been done in this region both in the field (Scagel, 1948) as well as in the lab- oratory. Although the size of manyaotf the cold- water marine algae adds special problems, at least some of the stages can be carried out to the point where transplant experiments can be made from the laboratory into the sea for fur- ther study. Transplant experiments of natural populations of juvenile stages, at least of these larger marine algae, even in the case of Ma- cro cystis, are quite feasible. The successful use of the experimental ap- proach in the laboratory is primarily dependent on having facilities for maintaining tempera- ture and light control, although the size of plants may again present certain special prob- lems. Cultures of Laminariales have been main- tained in controlled-environment tanks at the 504 PACIFIC SCIENCE, Vol. XV, October 1961 University of British Columbia for as long as a year, during which the complete sexual gen- erations were grown and the young sporophytes reached a length of 14 in., well past the stage where secondary morphological characteristics had developed to a point permitting positive identification. These studies have permitted in- disputable identification of the sporophytes to genus, and in some cases to species. The study of cultures in this group suggests that much of the early work on gametophytes in the Lami- nariales and in fact even on the early sporo- phytes may be in some question. In most of the early studies reported in the literature, plants were not grown long enough to establish beyond doubt the characteristic secondary morphological features of the sporophytes of the genera from which zoospores were intially obtained. In the presence of contaminating zoospores of other species which can soon supplant the original species under study, there is no other way of establishing that the same species or even the same genus in the Laminariales was obtained in the sporophyte generation succeeding the gametophyte generations in culture. It would be remiss not to mention much of the worthwhile physiological work that has been done on marine algae and other organisms. However, there is a need for a great deal more physiological work, particularly of the type done by Gail (1918, 1919, 1922) in an attempt to relate physiological processes more specifically and directly to the environment and ecological problems encountered in the field. In physio- logical studies there is frequently a tendency to proceed more and more deeply into special as- pects of the physiological behaviour or the bio- chemistry of an organism under artificial condi- tions. Although this information is very often of great value there is a very real need to project back to the field and attempt to explain be- haviour under the conditions existing in the natural environment. The statistical approach, as illustrated by Berquist’s (1959) revealing study of Hormosira, has also been little used as yet. Almost all of the quantitative aspects of the productivity of the benthonic algae in this area have related to species of economic interest (anon., 1947, 1948^; Scagel, 1948; Hutchinson, 1949). These studies have dealt largely with harvestable quantities and distributions, and have contributed little to an evaluation or an explanation, in terms of oceanographic factors, of the causes for this production. Obviously the ideal of a functional interpre- tation in the ecology of marine benthonic algae is dependent on an adequate and balanced knowl- edge of all of the foregoing aspects — of the qualitative and quantitative features of both the organisms and the environment. Many, much needed data are still lacking. An attempt to fol- low this line of investigation has been pursued on the coast of British Columbia and in a some- what more restricted area at the north end of Vancouver I. in Queen Charlotte Strait. Fur- ther detailed work is in progress in the Strait of Juan de Fuca at the south end of Vancouver I. between Vancouver I. and northern Washing- ton. The study in Queen Charlotte Strait forms the major part of this paper. JUNE 16 HOUR OF DAY (PSD 0600 0800 1000 1200 1400 1600 1800 2000 Fig. 13. Fluctuations in temperature, salinity, and oxygen at various depths near Malcolm Island at sta- tion 19 (1953). Benthonic Algae — S cagel 505 HOUR OF DAY (PS.T) JUNE 17 JUNE 18 Fig. 14. Fluctuations in temperature, salinity, and oxygen at various depths near Malcolm Island at station 20 (1953). The more detailed physical and chemical data presented in this paper for the Queen Charlotte Strait area were obtained during two cruises, one (53/1) on the C.G.M.V. "Cancolim II” in 1953 (anon., 1955^) and the other (56/1) on the CN.A.V. "Ehkoli” in 1956 (anon., 1956^). The stations occupied during these two cruises are indicated in Figures 5 and 6, respec- tively. Further less detailed observations were made on a cruise (57/6) in May, 1957, on the CN.A.V. "Clifton” (anon., 1958). In analyzing the seasonal characteristics and annual fluctua- tions of temperature and salinity that occur, reference has been made to the data obtained in the area from the daily records (anon., 1944, 1946, 1948 b, 1949, 1950, 1951, 1952 b, 1953 a, 1955c, 1956c) and data reports of the Hecate Strait Project (anon., 1955 b, 1955 d, 1955c, 195 6b) of the Pacific Oceanographic Group. Meteorological data have been obtained from the Meteorological Observations in Canada (anon., 1953 b), and tidal data were taken from the Pacific Coast Tide Tables (anon., 1952*). Chart data have been taken from charts of the Canadian Hydrographic Office and the British Admiralty. Some additional physical and chem- 506 ical (anon., 1954; Pickard and McLeod, 1953) and geological data (Dawson, 1880, 188 li, 1881 b, 1888, 1897; Bostock, 1948) have been referred to in the literature. Earlier biological observations and collections, which form part of the background for this paper, were made in the area in 1946 (anon., 1947, 1948^) and 1947 (Scagel, 1948). The results of a study of the phytoplankton and zooplankton collections which were also made during the same cruises will be presented in a subsequent paper. GENERAL ECOLOGICAL CHARACTERISTICS OF COASTAL REGION The Pacific Coast of Canada is ideally suited to a study of benthonic organisms and the effect of oceanographic factors on their distribution both in the intertidal and the sub tidal zones. Although the coast of British Columbia (Fig. 1) is only about 600 mi. long, proceeding di- rectly from the Strait of Juan de Fuca to Dixon Entrance, if all its various ramifications are in- cluded there is a coastline estimated at about 16,900 mi. in length. The tidal amplitude in this region is great, ranging from about 11 ft. at the southern boundary to nearly 26 ft. at the northern boundary. As a result of thorough mix- ing in the coastal region the upper zone in this area, except for a few local anomalies, is char- acteristically rather uniform in temperature at any one period and fluctuations occur within narrow limits. The annual range in temperature of the seawater near the surface is from about 6° to 18°C On the other hand, because of the runoff from large rivers, especially through the long mainland inlets, there are conditions rang- ing from practically fresh water at one extreme to full ocean salinity of about 34 %o at the other. Throughout the coast the physical nature of the substratum, ranging from mud and sand at one extreme to solid rock at the other, deter- mines to a large extent the organisms which are found in a specific area. However, a com- parison of the flora and fauna on various types of bottom is possible in a number of regions which are otherwise oceanographically rather similar. This permits a correlation of the dis- tribution of a wide variety of plants and animals with other physical and chemical factors of the environment. The oceanographic conditions PACIFIC SCIENCE, Vol. XV, October 1961 characteristic of the coast provide an ideal area in which to study the distribution of marine ben- thonic organisms particularly in relation to salinity over a rather extensive geographic area. GEOLOGICAL CHARACTERISTICS General Coast Features The Coast Mountains of British Columbia, which run along the whole length of the prov- HOUR OF DAY (PS.T) Fig. 15. Fluctuations in temperature, salinity, and oxygen at various depths near the Klucksiwi River at station 21 (1953). OXYGEN(mg/L) SALINITY (% c ) TEMP (°C) Benthonic Algae — Scagel 507 HOUR OF DAY (PST) FIG. 1 6. Fluctuations in temperature, salinity, and oxygen at various depths near the Klucksiwi River at station 22 ( 1953 ). 12 1 1 ' 10 9 8 32 31 30 16 15 14 13 12 I I 10 9 8 7 6 ?IG. » 3 ). HOUR OF DAY (PS.T.) JUNE 20 JUNE 21 * JUNE 2200 2400 0200 0400 0600 0800 1000 1200 1400 1600 1800 2000 2200 2400 0200 \ \ / / / / / / / / \/ STATION 23 (1953) QUEEN CHARLOTTE STRAIT — 0 maters 3 metsrs ----- 10 meters x- — — •*: 20 miters — 25.5“3Q meters . Fluctuations in temperature, salinity, and oxygen at various depths near Deer Island at station Benthonic Algae — Scagel 509 ince with an average width of about 100 mi., constitute the mainland coast. Although not as rugged as elsewhere in the Cordillera, the west- ern side of the Coast Mts. rises from the sea precipitously to summits in places exceeding 8,000 or 9,000 ft. Several rivers, which rise in the plateau country to the eastward, flow com- pletely across this range to the Pacific, where the lower parts of their valleys, as well as those of many streams originating in the mountains themselves, continue in the extensive system of fjords along the mainland of British Columbia (Fig. 1). These sediment-filled systems of valleys with steep slopes and numerous deltas can be traced in some places even through the coastal archipelago, which represents a partly sub- merged margin of the Coast Mountains. West of the Coast Mountains, and in a partly sub- merged condition, lies another chain, the In- sular Mountains, of which Vancouver I. and the Queen Charlotte Islands are projecting ranges. This outer chain stands on the edge of the continental platform with the great depths of the Pacific seaward from it. Between these MONTHS Fig. 18. Monthly temperatures of seawater at Pine Island for the period 1942-52. 510 PACIFIC SCIENCE, Vol. XV, October 1961 12 II 10 M MONTHS M J J N o o or 3 < 8 (T TEMPERATURE (°C.) MALCOLM I. (PULTENEY POINT) AUG. 1954- DEC. 1955 (MONTHLY) MAXIMA MEAN MAXIMA — GRAND MEAN MEAN MINIMA X — x- — x MINIMA Fig. 19. Monthly temperatures of seawater at Pulteney Point, Malcolm Island, for the period 1954-55. two ranges lies the Coastal Trough, part of a great depression extending intermittently north- westward from the Gulf of California through the Puget Sound- W illiamette lowland and on into Alaska. In British Columbia this great val- ley is largely submerged and comprises the ex- tensive areas of the Strait of Georgia, Queen Charlotte Strait, Queen Charlotte Sound, and Hecate Strait ( Fig. 1 ) . The Coast Mountains consist largely of Mes- ozoic rocks ranging from Triassic to early Ter- tiary and are composed of principally granitic rocks with some included masses of Mesozoic and Palaeozoic strata. The rocks of the Van- couver I. Ranges are composed conspicuously of masses of Triassic and Jurassic lava and vol- canic products, lesser contemporaneous sedi- ments with subordinate amounts of later granitic rocks, and marine and continental Cretaceous TEMPERATURE (C°) 10 10 10 10 10 10 10 Fig. 20. Temperature profiles at station 19 (1953) near Malcolm Island from bathythermograph traces. Benthonic Algae — Scagel 511 TEMPERATURE (°C) 10 10 10 10 10 10 10 10 10 10 10 10 sediments which have participated in its fold- ing. Horizontal Miocene beds occur along some parts of the shore. General Features of Queen Charlotte Strait Along the mainland portion of Queen Char- lotte Strait (Figs. 1, 3) the rocks are chiefly Jurassic and intrusive, composed of granodiorite, chiefly quartz diorite. Parts of the island groups near the entrance to the Strait, including Hope I. and Nigei I., include some intrusive rocks sim- ilar to those on the mainland side, but the major part of the north end of Vancouver I. is com- posed of Triassic to Jurassic rocks. In this latter region argillites, lavas, tuffs, breccia, sandstones, and limy siltstones are common. A smaller por- 512 PACIFIC SCIENCE, VoL XV, October 1961 tion of the east coast of Vancouver I. from Hardy Bay southward almost to Johnstone Strait is Cretaceous (chiefly Upper Cretaceous), with shales and sandstones. The rocks are heavily glaciated throughout almost the whole coastal area. The Queen Char- lotte Strait area was heavily glaciated during the Pleistocene, which fact is particularly evident in the vicinity of Deer I., where northwestern slopes are comparatively rough in contrast to the grooved and polished vertical or near ver- tical faces on southeastern parts. Well-stratified deposits of clays, silts, and sands occur, partic- ularly toward the east end of the Strait and along the Vancouver I. side. Cormorant I., Harwood I., and Malcolm I. are also examples of these deposits. In some places cliffs of these deposits border the shoreline, with extensive accumula- tions of boulders at the base. In such regions the boulders, which occur in great abundance along the beaches, are probably erratics derived from morainic material. Although the detailed geology of Queen Char- lotte Strait region is not well known supratidally, it is even less well known subtidally. The gen- eral features described, however, indicate the wide variety of substrata available for the at- tachment of benthonic organisms, particularly in the intertidal and shallower subtidal zones. This variety in the physical nature of the sub- TEMPERATURE (°C) 10 10 iO 10 10 10 10 10 10 10 10 10 10 10 FIG. 22. Temperature profiles at station 21 (1953) near Malcolm Island from bathythermograph traces. TEMPERATURE (°C) Benthonic Algae — Scagel 513 3 05 OJ 6 6 H CW) H±d3Q Fig. 24. Temperature profiles at station 23 (1953) near Deer Island from bathythermograph 514 PACIFIC SCIENCE, Vol. XV, October 1961 FlG. 25. Diagrammatic representation of regions of marine habitat (modified after Ekman, 1935). stratum is apparent not only geographically in the area but also vertically. In the deeper water soft mud bottoms predominate but sand and gravel are also found, and in the shallower areas sand, mud, gravel, pebbles, boulders, and solid rock are all found to varying extent, particularly along the Vancouver I. side of the Strait. In the shallower regions, however, a solid rock or bouldery substratum predominates. The nature of this rocky substratum in Queen Charlotte Strait is equally varied. There are igneous as well as sedimentary and metamorphosed sub- strata. There are basalts, dolerites, trachytic rocks, hard sandstones or quartzites, shales, con- glomerates, argillites, fine-grained to crystalline, commonly cherty limestones mixed with feld- spathic rocks, and dioritic and granitic frag- ments are also common. The preceding brief summary of the general geological features of the area indicates the present status of published knowledge (Bo- stock, 1948; Dawson, 1880, 1881*, 1881 b, 1888, 1897) concerning the Queen Charlotte Strait region. Bottom Topography of Queen Charlotte Strait Although soundings are still incomplete for the area, a study of the bottom topography ( Fig. 2) in Queen Charlotte Strait indicates exten- sive shallows, particularly along the Vancouver I. side of the Strait and around Malcolm I. In this region an abundant and varied inter- tidal and subtidal flora and fauna are supported. In the central part of the Strait and between Nigei and Vancouver islands (Fig. 2), there are deeper channels exceeding 100 fathoms. The water in these deeper channels is not continu- ous, however, with the deep waters of the main- Benthonic Algae — SCAGEL 515 land inlets and Johnstone Strait, and exhibits physical and chemical properties quite distinct from the latter (Figs. 3, 4). PHYSICAL CHARACTERISTICS Hydrographic Conditions in Queen Charlotte Strait The salinity distribution near the surface (Figs. 7-10) in Queen Charlotte Strait suggests a general circulation in a counter-clockwise fash- ion. The runoff from the mainland inlets along the north shore and at the east end of the Strait, particularly from Knight Inlet at the east end, contributes large volumes of fresh water which tends to move seaward at the surface, mixing as it progresses along the north shore into Queen Charlotte Sound with the deeper more saline water below. The more saline water from the open ocean and Queen Charlotte Sound moves into the Strait centrally as well as along the Van- couver I. side of the Strait and along the north side of Malcolm I. The intrusion of high salin- ity water along the deep channels in the central part of the Strait is also apparent. This general pattern of salinity distribution, with fluctuations to varying degrees near the surface, is pro- nounced in the upper zone to a depth of at least 20 m. (Figs. 7-10). Although strong winds may assist in the movement of water near the surface there is clear evidence at times of the movement of water against the wind and there are strong tidal currents throughout the region. The cur- MONTHS Fig. 26. Monthly air temperatures at Bull Harbour, B. C, for 1953. 516 PACIFIC SCIENCE, VoL XV, October 1961 MONTHS rent velocities involved vary considerably but at times may reach at least 80' cm/sec from near the surface to a depth of 20 m., and in deeper portions of the Strait, although generally much less, they may attain as much as 57 cm/sec at a depth of 125 m. An analysis of surface salinity data over a 10-year period (Fig. 11) from Pine L Light- house, which is near the entrance to Queen Charlotte Strait, indicates a salinity maximum of about 33 %o and a minimum of about 30 %o, with an annual mean of 31.75 %o. Although this station is not characteristic of the Strait it- self, the data available probably give a reason- able approximation of the annual salinity fluc- tuations for the Strait. Insufficient data are avail- able from Pulteney Point on Malcolm I. re- analyze the seasonal fluctuations more precisely in the central region of the Strait, but for the period available (Fig. 12) a range of about 28 %o to 32 %o with an annual mean of 21.50 %o Is indicated. Seasonal data indicate that near the surface the distribution of salinity throughout the year follows the same general pattern, decreasing to- ward the mainland and being higher along the Vancouver I. side of the Strait. At any one point, however, there is a general decrease in salinity in time from the maximum In April to a minimum in midsummer, when the maximum runoff from the mainland Inlets occurs. The winter salinity may be somewhat modified near Benthonic Algae— SCAGEL 517 the surface during the period of maximum pre- cipitation, which may reduce salinity near the surface. The minor extent of fluctuations that occur in the salinity distribution in the upper 20 m. is indicated by data taken at a number of an- chor stations (Figs. 13—17). These fluctuations are greatest at or near the surface and, in cer- tain instances, as near the Klucksiwi River ( Fig. 15), show the influence of fresh-water inflow of a more localized nature. A comparison of the te.mperatu.te -sail n ity characteristics of various parts of the Strait and the connecting bodies of water by means of T-S diagrams (Figs. 3, 4) indicates the discreteness of the water masses typical of Johnstone Strait, mouth of Knight Inlet, and Queen Charlotte Sound. The T-S diagrams for Queen Charlotte Strait indicate a characteristically intermediate condition between these extremes in properties of temperature and salinity for the greater part of the Strait. An analysis of surface temperature data over a 10-year period (Fig. 18) from Pine I. Light- house, which is near the entrance to Queen Char- lotte Strait, indicates a temperature maximum of about 12 °C and a minimum of about 5°C, with an annual mean of 8.6°C. Although this station, as already indicated, is not characteris- tic of the Strait itself in all respects, it prob- ably gives a reasonable approximation of the annual temperature fluctuations. Insufficient data are available from Pulteney Point on Mal- colm I. to analyze more precisely the seasonal MONTHS JFMAMJ J A S 0 N D Fig. 28. Monthly air temperatures at Alert Bay, B. C, for 1953. RELATIVE HUMIDITY (%) „ RELATIVE HUMIDITY (%) 518 PACIFIC SCIENCE, Vol. XV, October 1961 MONTHS MONTHS FIG. 30. Monthly relative humidity at Port Hardy, B. C., for 1953. PRECIPITATION (inches) RELATIVE HUMIDITY (%) Benthonic Algae — SCAGEL 519 MONTHS Fig. 31. Monthly relative humidity at Alert Bay, B. C., for 1953. MONTHS JFMAMJJASOND Fig. 32. Monthly precipitation at Bull Harbour, B. C, for 1953. 520 fluctuations in the central region of the Strait, but for the period available (Fig. 19) a range of from about 5° to 1 1°C with an annual mean of 8.2 5 °C. is indicated. The temperature distribution with depth, as shown by bathythermograph records taken at shallow anchor stations, indicates a well-mixed region near the surface along Vancouver I. and Malcolm I. (Figs. 20-24). Except for slight anomalies the water is generally almost isother- mal, with temperatures during this period of ob- servation seldom above 11°C at the surface and falling to not less than 8°C at a depth of 20 m. As indicated from the anchor stations, there are no marked changes with time in this general picture of the vertical distribution of tempera- ture, and at any one depth the fluctuations in- dicated were less than 2 C° (Figs. 13-17). However, there is a slight difference (usually not more than 1 C.°) in temperature at the sur- PACIFIC SCIENCE, Vol. XV, October 1961 face at most times of the year, with the water along the Vancouver I. side being warmer than that along the mainland side of the Strait. This is a result of the inflow of colder, less saline water from the mainland inlets. Seasonal data indicate that in general during the summer months the temperature at the sur- face is seldom above 10 °C. and at a depth of 20 m. is seldom above 9°C. During the winter months it is seldom below 7.5 °C. at the surface but may be about 7°C. at a depth of 20 m. This picture of the vertical temperature distribution is somewhat modified during the winter as a result of surface cooling of the water down to a depth of about 20 m. and a temperature in- version has been observed in January which dis- appears by April. During this early period the water at the depth of 20 m. and, in places, even to a depth of 400 m., may be warmer than that in the upper 20 m. by as much as 1 C.° and MONTHS J FMAMJJ A S O N D Fig. 33. Monthly precipitation at Hardy Bay, B. C, for 1953. CLOUD COVER (%) PRECIPITATION (Inches) Benthonic Algae — S cagel 521 MONTHS MONTHS FlG. 35. Monthly cloud cover at Bull Harbour, B. C, for 1953. CLOUD COVER (%) CLOUD COVER (%) 522 PACIFIC SCIENCE, Vol. XV, October 1961 MONTHS FIG. 3 6. Monthly cloud cover at Hardy Bay, B. C, for 1953. MONTHS J FMAMJ JASOND Fig. 37. Monthly cloud cover at Alert Bay, B. C., for 1953. Benthonic Algae- — SCAGEL 523 then fall to a lower temperature again, beneath this warmer upper layer, down to between 6° and 7°C. in the deeper regions. Thus the environment presented to the inter- tidal and immediate subtidal zones (see Fig. 25 for terminology) appears to be a relatively stable one as far as the temperature of the sea- water is concerned. Meteorological Conditions The intertidal region, however, during periods of exposure, is subjected to a varying degree to meteorological conditions, particularly fluctua- tions in temperature and precipitation, which must be considered in assessing the environment of organisms in this region. A comparison of the meteorological data (air temperatures, pre- cipitation, mean relative humidities, and mean cloud cover) available for the coast of British Columbia, particularly from Bull Harbour (Hope L) , Hardy Bay, and Alert Bay, gives some picture of the meteorological conditions at the northeast end of Vancouver L (Figs. 26-37). Along the coast, air from the maritime Pacific Ocean is usually present and results in mild winters and cool summers. Holding a high mois- ture content, this air does not become extremely hot or cold. However, occasional outbreaks of continental air (polar) from the interior of the 1353 Fig. 38. Monthly summary of tidal features at Hope Island for 1953 (see Figure 39 for significance of lines). 524 PACIFIC SCIENCE, Vol. XV, October 1961 18 16 14 12 LjJ u- 8 6 4 2 0 Fig. 39- Summary of tidal features for 1953. 1953 (JANrDEG ) HHHW continent bring cold periods during the winter, although generally the Coast and Cascade moun- tains provide considerable protection. Along the outer coast, the maritime conditions, which are present almost continually, result in high pre- cipitation, prolonged cloudiness, and small ranges in temperature- — -the typical conditions prevailing in the vicinity of Hope L (Figs. 26, 29, 32, 35) and Hardy Bay (Figs. 27, 30, 33, 36), despite the fact that these points are some- what on the lee side of Vancouver I. Along the inner coast and the lee side of Vancouver I., where there is some protection from the maritime influence, precipitation and cloudiness are somewhat reduced and ranges in tempera- ture are somewhat increased, as at Alert Bay (Figs. 28, 31,34, 37). The number of frost-free days in the whole Queen Charlotte Strait region averages 200-250 in a year. In both outer and inner regions there are no months with all temperatures ranging below 0°C In the outer coastal region 4 to 5 months have temperatures above 10 °C, and in the inner coastal region 5 to 6 months have temperatures above 10 °C In the outer coastal region, depending upon the locality, the mean monthly air temperatures for January are 1.7° to 4.4° C, and for July 13.3°C., and the mean daily temperatures are — 1 1° to 1.7 °C and 15.6° to 18.9°C for the same months, respec- tively. In the inner coastal region, depending upon the locality, the mean monthly air tem- peratures for January are 1.7° to 3.3 °C., and for July 15.6° to 18.3°C., and the mean daily temperatures are —1.1° to 0°C and 21.1° to 23.9°C, respectively, for the same months. Although some restricted regions of British Columbia on Vancouver I. average as much as 264 in. of rain, in the Queen Charlotte Strait region the annual rainfall averages between 40 and 60 in. at the inner end (including Alert Bay and Malcolm I.), with between 3 10 per cent as snow, and between 60 and 100 in. at the outer end (including Hope I. and Port Hardy ) , with less than 5 per cent as snow. The period of minimum rainfall is in the summer, 1953 JUNE JULY AUGUST M = Midnight PACIFIC STANDARD TIME Fig. 40. Tidal features for the periods of greatest exposure during summer months of 1953. Benthonic Algae — SCAGEL 525 1953 NOVEMBER M 19 M 20 M 21 M 22 M 23 M with between 2-3 in. on the average falling during August in the Strait, and the maximum rainfall occurs in the winter, with between 9-13 in. on the average in December. Near the entrance of the Strait, especially along the north side of Hope I., conditions of heavy surf generally prevail. Even when there is no sea a heavy swell is common. Farther down in the Strait to the south and east there is an area protected by scattered islands and reefs near the entrance, from the heavy swell from the open ocean, but this whole area is sub- ject to strong westerly winds in the summer months and to even stronger southeasterly winds during the winter months, so that the Strait is generally subjected to strong wave action and wind mixing. During the period mid-Septem- ber to mid-May the southeasterly winds are predominent and blow at speeds frequently up to 40 m.p.h. and occasionally higher. During the period mid-May to mid-September the west- erly winds predominate and blow at speeds up to 25 m.p.h. Particularly during this latter period, however, there may be considerable calm periods, especially during the morning hours, followed by strong seas which reach a peak about 1600 hrs. and then drop rapidly to rela- tive calm by 2000 hrs. Tidal Characteristics The tidal amplitude in the vicinity of Hope I. is usually about 17 ft. and the highest on record (37 years) was almost 19 ft. (December, 1941). An extensive intertidal flora and fauna is exposed during low tide periods in this zone. Continuous tidal data are recorded since 1949 at Alert Bay, near station 8 (Fig. 6). At the north end of Vancouver I. the tides are semi- diurnal and only moderately declinational, and thus springs and neaps are distinguishable ( Figs. 526 PACIFIC SCIENCE, Vol. XV, October 1961 Fig. 42. Diagram showing number of times each month during 1953 various levels in the intertidal zone were exposed. 38-41). Twice a month there are two tides a day which are about equal, and in the intervals between there is a much greater inequality in the height of any two successive low waters than between the two high waters of the same day. During the summer months (Fig. 40) the low- est low waters occur during daylight hours and during the winter months (Fig. 41) the lowest low waters occur late at night or early in the morning. The number of times during the year when each of the various levels recognized (Fig. 39) are exposed is indicated in Figure 42, and the number of days of continuous exposure of the Benthonic Algae — SCAGEL 527 upper portions of the intertidal zone are shown in Figure 43. The number of times (and per cent) these levels are exposed (Fig. 44) and submerged (Fig. 45) are also presented in an attempt to indicate possible critical levels. The tide levels are referred to in feet above or below the datum (zero point) which, for the coast of British Columbia, is the level of lowest normal tides. Chemical Characteristics During the summer months, when plant pro- duction is at its peak in Queen Charlotte Strait, the surface zone may be supersaturated with oxygen; values in excess of 15 mg/1 are fre- quently encountered. The maximum values gen- erally prevailing near the surface and to a depth of 20 m. are between 7 and 10 mg/1. During the summer months the values are somewhat higher than in the winter. The maximum con- centration of oxygen occurs at a depth between 2 and 5 m. (Figs. 46, 47), rather than right at the surface during the summer, and is related to the region of maximum phytoplankton activ- ity. The water throughout most of the Strait has a higher oxygen content near the surface (Figs. 46, 47) than in Queen Charlotte Sound or in the adjacent connecting mainland chan- nels. Although marked fluctuations occur locally in the upper 10 m., the general picture is more stable (Figs. 13-17) at greater depth. Phosphate concentrations (PO4-P) are not available for the winter months, but for the summer, during which minimum amounts are probably reached, the values present (Fig. 48) were between 0.5 and 2.0 mg.-at. per liter in the upper 20 m. The minima were generally in the upper 10 m. and most of the minima for the Strait and Queen Charlotte Sound were between 1.0 and 1.5 mg.-at. per liter. BIOLOGICAL CHARACTERISTICS Biological observations, extending over the length of the coast, indicate that there is a high degree of uniformity in the populations of many benthonic plants and animals extending from the Strait of Juan de Fuca to Dixon Entrance. This would be anticipated under such relatively uniform conditions of temperature. In attempt- ing to correlate the distribution of some of these organisms with salinity characteristics, as well as other oceanographic factors, there are several areas on the coast which could be used for pur- poses of this study. Although some supporting observations have been made in the Strait of Juan de Fuca and Dixon Entrance, this paper is restricted largely to a consideration of the? vicinity of Queen Charlotte Strait near the north end of Vancouver I. Horizontal Distribution of Organisms in Queen Charlotte Strait Biological observations have been made throughout the area although a more intensive study and collection has been undertaken at Hope I., Deer I., and in the vicinity of the Keogh River, the Klucksiwi River, and Malcolm I. These areas present a transition from Hope I., where the highest salinities are encountered, to the north and east sides of the Strait, where lowest salinities are found, with Deer I. and Malcolm I. being intermediate between these extremes. The distributions in the Strait of the more conspicuous organisms (Table 1) observed during this study are illustrated in Figure 49- Although both the marine algae and the in- vertebrate animals have been observed, the em- phasis in this study is on the more conspicuous marine algae. Some organisms in the area are more cos- NUMBER OF DAYS CONTINUOUS EXPOSURE (1953) JANUARY fEBRUARY AUGUST SEPTEMBER , OCTOBER NOVEMBER DECEMBER pniTTT FTTTTTTTTTTn nr 1 FIG. 43. Diagram showing number of days various levels in the intertidal zone were subjected to contin- uous exposure. 528 PACIFIC SCIENCE, Vol. XV, October 1961 TABLE 1 List of the More Conspicuous Species of Marine Benthonic Organisms in Queen Charlotte Strait 1. Polyneura latissima (Harvey) Kylin 2. Chthamalus dalli Pilsbry 3. Tegula funebralis (Adams) 4. Acmaea instabilis (Gould) 5. Balanus carlo sus (Pallas) 6. Endocladia muricata (Harvey) J. Agardh 7. Gloiopeltis furcata (Postels et Ruprecht) J. Agardh 8. Prionitis lanceolata Harvey 9. Prionitis lyallii Harvey 10. Erythrophyllum deles serioides J. Agardh 1 1 . Petrocelis franciscana Setchell et Gardner 12. Hildenbrandia occidentalis Setchell 13. Opuntiella calif ornica (Farlow) Kylin 14. Plocamium pacificum Kylin 15. Gigartina papillata (C. Agardh) J. G. Agardh 16. Gigartina sitchensis Ruprecht 17. Iridaea cordata (Turner) Bory 18. Iridaea heterocarpa Postels et Ruprecht 19- Halosaccion glandiforme (Gmelin) Ruprecht 20. Rhodymenia palmata (Linnaeus) Greville f. palmata 21. G astro clonium coulteri (Harvey) Kylin 22. Micrpcladia borealis Ruprecht 23. Ptilota asplenioides (Esper) C. Agardh 24. Ptilota calif ornica Ruprecht 25. Ptilota hypnoides Harvey 26. Hymenena setchellii Gardner 27. Hymenena flabelligera (J. Agardh) Kylin 28. Cryptopleura ruprechtiana (J. Agardh) Kylin 29. Polysiphonia collinsii Hollenberg var. collinsii 30. Pterosiphonia bipinnata (Postels et Ruprecht) Falkenberg var. bipinnata 3 1 . haurencia spectabilis Postels et Ruprecht 32. Rhodomela larix (Turner) C. Agardh 33. Odonthalia floccosa (Esper) Falkenberg 34. Odonthalia washingtoniensis Kylin 3 5 . Prasiola meridionalis Setchell et Gardner 36. Agardhiella coulteri (Harvey) Setchell 37. Saunder sella simplex (Saunders) Kylin 38. Heterochordaria abietina (Ruprecht) Setchell et Gardner 39- Desmarestia intermedia Postels et Ruprecht 40. Desmarestia herbacea Lamouroux 4 1 . Desmarestia media var. tenuis Setchell et Gardner 42. Desmarestia munda Setchell et Gardner 43. S or anther a ulvoidea Postels et Ruprecht f. ulvoidea 44. Myelophycus intestinale Saunders 45. Scytosiphon lomentaria (Lyngbye) J. Agardh f. lomentaria 46. Coilodesme bulligera Stroemfelt 47. Laminaria cuneifolia J. Agardh f. cuneifolia 48. Laminaria saccharina (Linnaeus) Lamouroux f. saccharina 49. Laminaria setchellii Silva 50. Pleurophycus gardneri Setchell et Saunders 51. Agarum fimbriatum Harvey 52. Agarum cribrosum Bory •53. Hedophyllum sessile (C. Agardh) Setchell a. smooth form b. bullate form 54. Postelsia palmaeformis Ruprecht 55. Lessoniopsis littoralis (Farlow et Setchell) Reinke 56. Pterygophora calif ornica Ruprecht 57. Egregia menziesii (Turner) Areschoug subsp. menziesii 58. Fucus evanescens C. Agardh f. evanescens 59- Fucus gardneri Silva f. gardneri 60. Pelvetiopsis limitata (Setchell) Gardner f. limit at a 61. Cystoseira geminata C. Agardh 62. Smithora naiadum (Anderson) Hollenberg 63. Porphyra perforata j. Agardh f. perforata 64. Farlowia mollis (Harvey et Bailey) Farlow et Setchell 65. Dilsea calif ornica (J. Agardh) O. Kuntze 66. Gloiosiphonia calif ornica (Farlow) J. Agardh 67. Nereocystis luetkeana (Mertens) Postels et Ruprecht 68. Macrocystis integrifolia Bory 69. Alaria nana Schrader 70. Alaria marginata Postels et Ruprecht 7 1 . Alaria tenuifolia Setchell f . tenuifolia 72. Alaria valida Kjellman et Setchell f. valida 73. Bangia f us copurpurea (Dillwyn) Lyngbye 74. Bossiella plumosa (Manza) Silva 75. Bossiella calif ornica (Decaisne) Silva 76. Calliarthron regenerans Manza 77. Calliarthron schmittii Manza 78. Callithamnion pikeanum Harvey var. pikeanum 79. Callophyllis edentata Kylin 80. Callophyllis firma Kylin 81. Cladophora trichotoma (C. Agardh) Kutzing 82. Spongomorpha coalita (Ruprecht) Collins 83. Codium fragile (Suringar) Harlot 84. Codium setchellii Gardner 85. Coilodesme calif ornica (Ruprecht) Kjellman 86. Constantinea simplex Setchell 87. Constantinea subulifera Setchell 88. Corallina officinalis var. chilensis (Harvey) Kutzing 89- Costaria costata (Turner) Saunders 90. Costaria mertensii J. Agardh 91. Cryptosiphonia woodii J. Agardh 92. Cumagloia under sonii (Farlow) Setchell et Gard- ner 93. Cymathere triplicata (Postels et Ruprecht) J. Agardh 94. Halicystis ovalis (Lyngbye) Areschoug 95. Pylaiella litoralis (Lyngbye) Kjellman 96. Leathesia difformis (Linnaeus) Areschoug 97. Haplogloia under sonii (Farlow) Levring Benthonic Algae — S cagel 529 TABLE 1 ( continued ) 98. Ahnfeltia concinna J. Agardh 99- Rhodoglossum latissimum J. Agardh 100. Iridaea lineare (Setchell et Gardner) Kylin 101. Ahnfeltia plicata (Hudson) Fries 102. Schizymenia pacifica Kylin 103. Rhodoglossum affine (Harvey) Kylin 104. Flustrella corniculata (Smitt) 105. Mytilus calif ornianus Conrad 106. Mitella polymerus (Sowerby) 107. Pisaster ochraceus (Brandt) 108. Balanus glandulus (Darwin) 109- Styela montereyensis (Dali) 110. Haliotis kamtschatkana Jones 111. Strongylocentrotus drohachiensis (Muller) 112. Strongylocentrotus purpuratus (Stimpson) 113. Strongylocentrotus francis canus (Agassiz) 114. Mytilus edulis Linnaeus 115. Littorina planaxis Phillippi 116. Dictyoneurum calif ornicum Ruprecht 117. Phyllospadix scouleri Hooker 118. Zostera marina L. var. marina 119- Balanus nubilus Darwin 120. Ulva latissima L. 121. Rhizoclonium riparium (Roth) Harvey 122. Porphyrella gardneri Smith et Hollenberg 123. Desmarestia ligulata (Lightfoot) Lamouroux 124. Amplisiphonia pacifica Hollenberg 125. Rhodymenia pertusa (Postels et Ruprecht) J. Agardh 126. Pterochondria woodii (Harvey) Hollenberg mopolitan in their distribution, particularly in their tolerance to extreme dilution. Extending throughout the Strait (Fig. 49) are forms such as Alaria tenuifolia Setchell f. tenuifolia, Cym- athere triplicata (P. and R.) J. Ag., Cost aria costata (Turn.) Saunders, C. mertensii J. Ag., Laminaria saccharina ( L. ) Lamour. f . saccharina, Nereocystis luetkeana (Mert. ) P. and R., Por- NUMBER OF TIMES EXPOSED (1953) 0 200 400 600 800 1000 1200 1400 10 20 30 40 50 60 70 80 90 100 NUMBER OF TIMES EXPOSED (%) Fig. 44. Diagram showing number (and per cent) of times tidal condition caused exposure of various regions in the intertidal zone. phyra perforata J. Ag. f. perforata, Rhodomela larix (Turn.) C. Ag., 0 donthalia floccosa (Esper) Falk., Mytilus edulis Linnaeus, Lit- torina planaxis Philippi, and Strongylocentrotus drohachiensis (Muller). Restricted to the region of highest salinity, as at Hope I., are Postelsia palmaeformis Rupr., Lessoniopsi* littoralis (Farl. andSetch.) Reinke, Laminaria setchellii Silva, Pelvetiopsis limit at a (Setchell) Gardner f. limitata, Dilsea calif or - nica (J. Ag.) O. Kuntze, Erythrophyllum deles- serioides J. Ag., Iridaea lineare (S. and G.) Kylin, Hymenena s etchellii Gardner, Ptilota asplenioides (Esper) C. Ag., P. calif ornica Rupr., P. hypnoides Harvey, Mitella polymerus (Sow- erby ) , and Flustrella corniculata ( Smitt ) . A smooth form of Hedophyllum sessile (C. Ag.) Setch., Pleurophycus gardneri Setch. and Gardner, and Styela montereyensis Dali are also present in regions of highest salinity, but are somewhat less restricted in their distribution and extend into the Strait as far as Deer L A few organisms extend still farther into the Strait but only slightly beyond Deer I. Among these are Alaria nana Schrader and A. marginata P. and R. Still others extend from Hope I. throughout the Deer I. region and as far as Malcolm L, but not as far as the east and north sides of the Strait. Among these are Mytilus calif ornianus Conrad, Strongylocentrotus pur- puratus (Stimpson), Macro cystis integrifolia Bory, Egregia menziesii (Turner) Aresch. subsp. menziesii, the typical bullate form of Hedophyl- lum sessile (C. Ag.) Setch., Alaria valida 530 PACIFIC SCIENCE, Vol. XV, October 1961 NUMBER TIMES SUBMERGED (1953) 5 10 15 20 NUMBER TIMES SUBMERGED (%) FIG. 45. Diagram showing number (and per cent) of times tidal condition caused submergence of vari- ous regions in the intertidal zone. (Kjellm. and Setch.) f. valida, Constantinea sim- plex Setch., and H alio tis kamtschatkana Jones. Isolated populations of the abalone ( Haliotis kamtchatkana) , which have been noted farther eastward in Johnstone Strait and which may be related to local oceanographic features, pre- sent something of an anomaly in the general distribution. An exception to the general dis- tribution described in this group is that of Macrocystis integrifolia. Although Macrocystis occurs in regions of high salinity and extends as far down the Strait as the north side of Mal- colm I. and the south side of Numas I., it does not occur in the most exposed areas where Postelsia and Mitella are encountered. Vertical Distribution of Organisms The vertical distributions with reference to tide levels, based on data obtained using an Abney level, are presented for some of the more conspicuous organisms at Hope I. (Fig. 50). Comparisons are made also for some of these (Fig. 51) at Hope L, Deer L, and near the Klucksiwi River, to indicate the effect exposure to surf has on the vertical distribution of some organisms. By comparing the vertical distributions of organisms in Queen Charlotte Strait (Fig. 50) with tidal data presented with respect to emer- gence and submergence (Figs. 42-45), a num- ber of limiting levels are fairly apparent. Near the top of the intertidal zone is a region (HHHW to HLHW) which is rarely submerged (Fig. 43) for more than a few hours on each of a few days at any period of the year and which OXYGEN (mg/L.) FIG. 46. Distribution of oxygen with depth at vari- ous stations in Queen Charlotte Strait and adjacent regions in June, 1953. Benthonic Algae— Scagel 531 0 2 (mg./L.) o 5 10 15 FIG. 47. Distribution of oxygen with depth at vari- ous stations in Queen Charlotte Strait and adjacent regions in May, 1956. may be continuously exposed for as long as 4 months during the summer period. Those or- ganisms that can tolerate such conditions are rare and in this zone one finds chiefly Littorina, which is capable of moving sufficiently to extend into lower less extreme zones when necessary. In the region below (HLHW to MHHW), continuous exposure to the air may last for periods ranging from a few days to almost 2 weeks, and these periods occur at least twice a month, but the rest of the time this zone is submerged at least once a day. In the region below (MHHW to MLHW), continuous exposure to the air is rarely for more than a few days at a time, and for a few months during the winter the region is submerged at least once a day. Although the upper limit of a number of the organisms ( Fig. 50) in this region of the intertidal zone between 13.1 and 17.4 ft. varies to some extent with the organism, it is apparent that in this general region an upper boundary is probably deter- mined directly or indirectly by the degree of exposure to climatic conditions. The precise tide level, if there is one, at which this boundary occurs is not clear from the data available. The effect of surf in the exposed environment may also cause some variation in the upper limit of the vertical distribution. Since the greatest change in degree of exposure within this upper region (13.1-17.4 ft.) occurs between MLHW and MHHW, a critical level is suggested at this point for a considerable number of conspicuous organisms. All portions of the intertidal zone below MLHW are submerged at least once a day for various periods. The greatest change in condi- tions of submergence (Fig. 45) in the region of the middle intertidal zone (8.2-13.1 ft.) occurs between HHLW and LLHW, and it is at this point where another critical level is sug- gested, in some instances as the upper limit and in others as near the lower limit of the vertical distribution of certain organisms (Fig. 50). In the lower intertidal zone (0-8.2 ft.) the most extreme change in conditions of exposure (Fig. 44) and submergence (Fig. 45) occurs between MLLW and MHLW, and this is again reflected both at the upper and lower limits in the vertical distribution of certain organisms (Fig. 50). Another critical level (Fig. 50) occurs in the region between LLLW and LHLW, where another region of marked change (Fig. 45 ) exists in conditions of submergence. Although the upper limits in most instances appear to be relatively sharp, there is less con- sistency in this respect concerning lower limits. This suggests that other factors, perhaps com- petition for space or predation, may be responsi- ble for limiting distribution, particularly down- ward in some instances. DISCUSSION Knowing as we do from experimental work the responses of some organisms to environ- mental factors, it seems reasonable to anticipate DEPTH (meters) 532 PACIFIC SCIENCE, Vol. XV, October 1961 P0 4 “P (mg-at/L.) FIG. 48. Distribution with depth of phosphate-phosphorous at various stations in Queen Charlotte Strait and adjacent regions in May, 1956. Benthonic Algae — SCAGEL 533 NUMBERS REFER TO SPECIES LISTED in HORIZONTAL DISTRIBUTION TABLE I SPECIES HOPE 1. DEER 1. KEOGH R. KLUCKSIWI RIVER MALCOLM ISLAND APPR. INLETS SPECIES HOPE 1. DEER I. KEOGH R. KLUCKSIWI RIVER MALCOLM ISLANO APPR. INLETS 116 10 24 23 60 55 54 65 94 126 122 106 104 100 99 103 81 49 26 8 50 25 56 109 98 101 124 77 21 530 117 112 105 76 74 86 13 92 69 66 75 35 80 72 3 125 119 34 37 38 51 52 53b 7 11 12 14 15 19 27 31 85 88 79 ' 91 97 102 78 no 42 46 57 61 113 47 68 45 70 64 4 15 17 36 22 84 89 90 93 95 107 121 j 123 83 87 96 43 44 62 73 82 1 5 6 9 j 2 8 2 9 32 33 39 40 41 30 2 16 20 | 63 1 71 1 48 1 58 1 59 j 108 118 120 115 114 67 Fig. 49- Horizontal distribution of marine benthonic organisms in Queen Charlotte Strait. that the distribution of marine organisms in time and space can be explained on the basis of oceanographic features — provided the geologi- cal, physical, chemical, and biological factors of the environment and their interaction can be adequately assessed. Any attempt to oversimplify such a many-sided and complicated study would be an avoidance of reality. However, in such a study one expects to make assumptions based on the knowledge available, in some instances of necessity by extrapolation, and to pass through a descriptive phase in an attempt to 534 PACIFIC SCIENCE, Vol. XV, October 1961 correlate observed distributions on the basis of and in relation to other factors in the environ- ment. These comparisons may be largely qualita- tive in the first instance, but the intent is that they be not only qualitative but also quantitative in the final analysis and, as in all branches of science, eventually permit predictions. At best, however, only a descriptive treatment can be at- tempted at this point and can only point the way to further studies and hypotheses. There have been many attempts to describe the zonation of marine organisms on the shore, and a great jumble of appalling confusion in ecological terminology has evolved to the point where some magic significance is sometimes associated with the terms and units involved. Although the need for this descriptive phase of the study (with a minimum of terminology) is recognized at an early stage of study or in a new area under investigation, once it has served VERTICAL DISTRIBUTION TIDE LEVEL LIST OF SPECIES 51 • 116 - 56 • 79 ■ 93 • 89 ■ 109 104 70- 50- 68 - 67- 83 119 - 86 9 55 117 ■ 49 - 25 24 10 * * B r~ ^ ^ IX i n x r i i x i xx ^ £ s? $ $ 100 45 57 ■ 65 107 22 88 - 69 18 78 74 15 53 26 29 I - 38 54 17 30 106 105 32 6 108 5 63 60 7 16 60 II 73 115 Fig. 50. Vertical distribution of some marine benthonic organisms at Hope Island. Benthonic Algae — SCAGEL 535 HOPE 1. VERTICAL DISTRIBUTION DEER 1. KLUCKSIWI R. TIDE LEVEL r~ r~ 2 2 x r* x r* x r~ X X r- 1-2 r- x f- 2 X X X 1“ X LIST OF SPECIES r- r r n r~ 3= 3g 2? «£ £ r~ x x x 5? * X 2E X X ^ Si LEVEL (feet) ENDOCLADIA MURICATA GLOIOPELT1S FURCATUS HETEROCHORD ARIA ABIETINA PETROCELIS FRANCISCANA EQREGIA MENZIESII PORPHYRA PERFORATA 01 4^ w ro _L O Fig. 51. Comparison of vertical distribution of some benthonic organisms at several points in Queen Char- lotte Strait. its initial purpose little refinement in this method of approach seems conducive to an ex- planation of the causal factors. From this point it becomes necessary to look at the problem from a new perspective involving a detailed study of the organisms concerned— an under- standing of their life histories, rate of growth, reproduction, and various physiological require- ments and tolerances in relation to the environ- ment. These problems may be and, as already indicated, have been approached to some extent by actual field studies and experiments as well as by laboratory studies under controlled condi- tions. It is with this philosophy in mind that this study has been approached to the extent possible from the existing data, initially from the stand- point of the oceanographer with an analysis of the environmental factors and their relation to the organisms. The success of these preliminary efforts both in the field as well as in the labora- tory supports the conviction that the approach is a useful and instructive one. Differences which sometimes appear striking or significant on a broad scale are frequently less apparent and confused in a local area. The distribution of the genus Macrocystis is an interesting case in point. Although the global pattern of dis- tribution of this genus is rather clearly estab- lished (Setchell, 1932; Wornersley, 1954) on the basis of temperature distributions and hence follows the pattern of distribution of some of the cold water currents of both southern and northern hemispheres, the distribution of M. integrifolia in British Columbia follows a dis- tinct salinity distribution. As yet, however, it cannot be conclusively stated in the latter in- stance that salinity itself is the causal factor. It still remains to be established whether salin- ity in terms of an osmotic relationship, or some parallel factor associated with open ocean water of high salinity, provides a causal mechanism for the distribution of M. integrifolia on this coast. It is clear that there is a horizontal distribu- tion of organisms, including Macrocystis integri- folia, in Queen Charlotte Strait which follows closely the pattern of salinity distribution in the 536 Strait. In turn this reflects the circulation within the area. It would be premature to say that salinity is directly responsible for the observed distributions of all the organisms encountered. But one may say that the distribution reflects the dependence on high salinity water which is characteristic of the open ocean and exposure. In some instances it may be salinity that is a direct causal factor. On the other hand, the open coast has organisms associated with surf condi- tions. It has been suggested that the high oxygen requirement of certain organisms is met only in such an exposed environment. However, the distributions and concentrations of oxygen in the sea in this area do not directly support this argument. The oxygen content of the waters within the sheltered Strait is as high or higher than in the surf in the exposed regions. This is particularly true in the central part of the Strait when there is a heavy bloom of phytoplankton, at which time the water may be supersaturated with oxygen to as much as 175 per cent. Like- wise, although it is known that many marine algae have a high inorganic phosphate require- ment, there is no evidence that this nutrient is ever limiting in this area within the zone occu- pied by the benthonic algae. There is a great need for further knowledge of the presence, dis- tribution, amounts, and availability of many more dissolved inorganic as well as organic sub- stances and perhaps even of growth substances. There is also need for a study of the quantitative aspects of removal, the rate of removal of such substances, and precise requirements for growth and reproduction in the micro-environment of the marine benthonic algae. The restriction of certain organisms to surf conditions suggests that constant movement of water is required to provide nutrients and gases which may be rapidly exhausted from the immediate or micro- environment of the individual fixed alga, or in the case of the sessile marine invertebrates, such as Mitella polymerus, to provide particulate food. It may be that lowering the concentration or removal by dilution or by water movement of some toxic substances which may accumulate above a certain concentration in the micro- environment is just as significant as the avail- ability of others. PACIFIC SCIENCE, Vol. XV, October 1961 SUMMARY A detailed study of the horizontal and ver- tical distributions of marine benthonic organ- isms in Queen Charlotte Strait has been limited so far to the more conspicuous algae and in- vertebrates encountered. The relationship of these distributions to the salinity distribution indicates that more intensive study of the flora and fauna in this area, as well as elsewhere on the coast, will provide further supporting evi- dence indicating not only the effect of ocean- ographic variables on the distribution and pro- duction of marine benthonic organisms but also the possibility of using such organisms as in- dicators of oceanographic conditions both in time and space. The relationship of the vertical distribution of some of the organisms to certain tide levels indicates the response of the different organisms to varying degrees of exposure and submergence. This oceanographic approach, both qualita- tively and quantitatively, has given a broad understanding of some of the possible factors which are likely to be responsible for the ob- served distributions. However, before there can be a clear understanding and explanation of the fundamental relationships between the organ- isms and their environment in this region, as well as an understanding of the interrelation- ships and interaction among the organisms themselves, additional field observations, field experiments, and laboratory experiments must be undertaken. ACKNOWLEDGMENTS Financial assistance provided by the Defence Research Board of Canada in support of this research (DRB 9520-14) over a period of sev- eral years is gratefully acknowledged. The author would also like to thank the Defence Research Board for its assistance in providing ship trans- portation through the Pacific Naval Laboratory, Esquimalt, and the Royal Canadian Navy for the purpose of collecting data on several cruises. In addition, assistance provided by the Na- tional Research Council, the Joint Committee on Oceanography, the Fisheries Research Board of Canada, and the University of British Columbia is gratefully acknowledged. Benthonic Algae — Scagel 537 REFERENCES Anon. 1944. Observations of Sea Water Tem- perature and Salinity on the Pacific Coast of Canada. Fish. Res. Bd. Can., Nanaimo. Pacif. Oceanogr. Group. Vol. 5, 1942, 1943. 1946. Observations of Sea Water Tem- perature and Salinity on the Pacific Coast of Canada. Fish. Res. Bd. Can., Nanaimo. Pacif. Oceanogr. Group. Vol. 6, 1944, 1945. — 1947. A Report on the Location of Ma- rine Plants of Economic Importance in British Columbia Coastal Waters, Part 1. Brit. Colum- bia Res. Council Tech. Bull. 4. 5 pp., 8 figs. — 1948^. Marine Plants of Economic Im- portance in British Columbia Coastal Waters, Part 2. Brit. Columbia Res. Council Tech. Bull. 10. 9 pp., 9 figs., 7 tables. 1948A Observations of Sea Water Tem- perature and Salinity on the Pacific Coast of Canada. Fish. Res. Bd. Can., Nanaimo. Pacif. Oceanogr. Group. Vol. 7, 1946, 1947. — 1949. Observations of Sea Water Tem- perature and Salinity on the Pacific Coast of Canada. Fish. Res. Bd. Can., Nanaimo. Pacif. Oceanogr. Group. Vol. 8, 1948. 1950. Observations of Sea Water Tem- perature and Salinity on the Pacific Coast of Canada. Fish. Res. Bd. Can., Nanaimo. Pacif. Oceanogr. Group. Vol. 9, 1949. 1951. Observations of Sea Water Tem- perature and Salinity on the Pacific Coast of Canada. Fish. Res. Bd. Can., Nanaimo. Pacif. Oceanogr. Group. Vol. 10, 1950. 1952^. Tide Tables for the Pacific Coast of Canada, 1953. Surveys and Mapping Branch, Dept. Mines and Tech. Surveys, Ot- tawa. 1952 b. Observations of Sea Water Tem- perature and Salinity on the Pacific Coast of Canada. Fish. Res. Bd. of Can., Nanaimo. Pacif. Oceanogr. Group. Vol. 11, 1952. — 1953^. Observations of Sea Water Tem- perature and Salinity on the Pacific Coast of Canada. Fish. Res. Bd. of Can., Nanaimo. Pacif. Oceanogr. Group. Vol. 12, 1952. 1953 b. Monthly Record Meteorological Observations in Canada, January— December, 1953. Dept. Transport Meteorological Div., Ottawa. 1954. British Columbia Inlet cruises, 1953. Inst. Oceanogr. Univ. Brit. Columbia Data Rept. 3. 27 pp. 1955^. C.G.M.V. "Cancolim II” Survey of British Columbia Coast, 1953. Inst. Oceanogr. Univ. Brit. Columbia Data Rept. 4. 72 pp. 195 5 A Data Record, Current Measure- ments, Hecate Project, 1954. Joint Comm. Oceanogr., 1955, Canada. 74 pp. 1955c. Observations of Sea Water Tem- perature and Salinity on the Pacific Coast of Canada. Fish. Res. Bd. Can., Nanaimo. Pacif. Oceanogr. Group. Vol. 14, 1954. 195 5 A Physical and Chemical Data Record, Hecate Project, 1954, Queen Char- lotte Sound, Hecate Strait, Dixon Entrance. Joint Comm. Oceanogr., 1955, Canada. 99 pp. 1955c. Physical and Chemical Data Record, Hecate Project, with Appendix 1, Current Observations, 1955, Queen Charlotte Sound, Hecate Strait, Dixon Entrance. Joint Comm. Oceanogr., 1955. 107 pp. 1956^. Queen Charlotte Strait, 1956. Inst. Oceanogr. Univ. Brit. Columbia Data Rept. 9. 20 pp. 1956 A Physical and Chemical Data Record, Dixon Entrance, Hecate Strait, Queen Charlotte Sound, 1934, 1937, 1938, 1951. Joint Comm. Oceanogr., 1956, Canada. 56 pp. 1956c. Observations of Sea Water Tem- perature and Salinity on the Pacific Coast of Canada. Fish. Res. Bd. Can., Nanaimo. Pacif. Oceanogr. Group. Vol. 15, 1955. 1958. Indian Arm Cruises, 1957; Queen Charlotte cruise, 1957. Inst. Oceanogr. Univ. Brit. Columbia Data Rept. 12. 35 pp. BERQUIST, P. L. 1959. A statistical approach to the ecology of Hormosira banksii . Bot. Mar. 1(1/2): 22-53, 23 figs., 11 tables. 538 PACIFIC SCIENCE, Vol. XV, October 1961 BOSTOCK, H. S. 1948. Physiography of the Ca- nadian Cordillera, with Special Reference to the Area North of the Fifty-fifth Parallel. Can. Dept. Mines and Resources Geol. Sur- vey Mem. 247. ix + 106 pp., 31 pis., 1 fig. Chapman, V. J. 1946. Marine algal ecology. Bot. Rev. 12: 628-672. 1950. The marine algal communities of Stanmore Bay, New Zealand. Studies in inter- tidal zonation, 1. Pacif. Sci. 4(1): 63-68, 3 figs., 1 table. 1957. Marine algal ecology. Bot. Rev. 23: 320-350, 2 figs. Coleman, J. 1933. The nature of intertidal zonation of plants and animals. Jour. Mar. Biol. Assoc. U. K. 18: 435-47 6, 15 figs, 7 tables. Dawson, E. Y. 1945. Marine algae associated with upwelling along the northwestern coast of Baja California, Mexico. Bull. South. Calif. Acad. Sci. 44(2) : 57-71, pis. 20-22. 1951. A further study of upwelling and associated vegetaxion along Pacific Baja Cali- fornia, Mexico. Jour. Mar. Res. 10(1): 39- 58, 6 figs, 1 table. Dawson, G. M. 1880. Report on the Queen Charlotte Islands. Geol. Survey Can. Montreal. Pp. 1B-101B, 9 figs, 2 maps, 2 pis. 188D. Sketch of the geology of British Columbia. Geol. Mag. Decade 2, 8(4): 1- 19, 2 figs. 1881 A The superficial geology of Bri- tish Columbia and adjacent regions. Quart. Jour. Geol. Soc. 1881 (May) : 272-285, 1 fig. 1888. Report on an exploration in the Yukon District, N.W.T, and adjacent north- ern portion of British Columbia. Can. Geol. Survey Rept. 1887: B1-B183, 7 figs, 1 map, 5 pis. 1897. The Physical Geography and Geol- ogy of Canada. Handbook of Canada. Rowsell and Hutchinson, Toronto. 28 pp. Doty, M. S. 1946. Critical tide factors that are correlated with the vertical distribution of marine algae and other organisms along the Pacific Coast. Ecology 27: 315-328, 6 figs. 1957. Rocky intertidal surfaces. In: J. W. Hedgpeth ( ed. ) , Treatise on Marine Ecology and Paleoecology. Vol. 1, Ecology. Geol. Soc. Am. Mem. 67. Pp. 535-585, 18 figs, 2 tables. Ekman, S. 1935. Tiergeographie des Meeres. Leipzig, xii -f 542 pp, 244 figs. Feldmann, J. 1937. Recherches sur la vegeta- tion marine de la Mediterranee. La Cote des Alberes. Rev. Algol. 10: 1-339, 1 map, 20 pis, 25 figs, 10 tables. 1951. Ecology of marine algae. In: G. M. Smith ( ed. ) , Manual of Phycology. Chro- nica Botanica, Waltham. Pp. 313-334. Fischer-Piette, E. 1940. Sur quelques progres recents, et sur les methodes et tendances actuelles, en bionomie intercotidale ( sub- stratum rocheux, regions temperees). Con- tribution a letude de la repartition actuelle et passee des organismes dans la zone neritique. Mem. Soc. Biogeogr. 7: 393-434. Gail, F. W. 1918. Some experiments with Fucus to determine the factors controlling its vertical distribution. Publ. Puget Sound Biol. Sta. 2: 139-151, 6 tables, 1 chart. 1919. Hydrogen ion concentration and other factors affecting the distribution of Fucus. Publ. Puget Sound Biol. Sta. 2: 287 - 306, pis. 51-52, 4 tables. 1922. Photosynthesis in some of the red and brown algae as related to depth and light. Publ. Puget Sound Biol. Sta. 3: 177— 193, pis. 31-33, 9 tables. GlSLEN, T. 1929 . Epibioses of the Gullmar Fjord, Part I. Kristin. Zool. Sta. 1877-1927. Skr. Svensk. Vetensk. Akad. 3: 1-123. — 1930- Epibioses of the Gullmar Fjord, Part II. Kristin. Zool. Sta. 1877-1927. Skr. Svensk. Vetensk. Akad. 4: 1-380. HARTOG, C. D. 1959. The epilithic algal com- munities occurring along the coast of the Netherlands. Wentia 1: 1-241, 27 figs, 43 tables. Hutchinson, A. H. 1949. Marine plants of economic importance of the Canadian Pacific coastal waters. Proc. Seventh Pacif. Sci. Congr. 5: 62-66. Benthonic Aigae — SCAGEL 539 ISAAC, W. E. 1935. The distribution and zona- tion of marine algae on the coasts of South Africa. Brit. Assoc. Adv. Sci. Rept. Ann. Meet. 1935: 455. Lamouroux, J. V. F. 1825. Distribution geo- graphique des productions aquatiques. Hy- drophytes des eaux salees. Diet, class. d’Hist. Nat. de Bory de Saint-Vincent 7: 245-251. 1826. Memoire sur la geographic des plantes marines. Ann. Sci. Nat. 7: 60-82. OKAMURA, K. 1926. On the distribution of marine algae in Japan. Proc. Third Pacif. Sci. Congr. 1: 958-963, 1 fig. — 1932. On the nature of the marine algae of Japan and the origin of the Japan Sea. Bot. Mag. Tokyo 41: 588-592. Pickard, G. L., and D. C. McLeod. 1953. Sea- sonal variation of temperature and salinity of surface waters of the British Columbia Coast. Jour. Fish. Res. Bd. Can. 10(3) : 125— 145, 5 tables, 8 figs. SCAGEL, R. F. 1948. An Investigation on Marine Plants near Hardy Bay, B. C. Prov. Dept. Fish. 1. 70 pp., 11 tables, 26 figs. 1957. An Annotated List of the Marine Algae of British Columbia and Northern Washington. Bull. Nat. Mus. Can. No. 152. 286 pp., 1 fig. SETCHELL, W. A. 1893- On the classification and geographical distribution of the Lamin- ariaceae. Trans. Conn. Acad. Arts Sci. 9: 333- 375. 1917. Geographical distribution of the marine algae. Science 45: 197-204. 1932. Macro cystis and its holdfasts. Univ. Calif. Publ. Bot. 16: 445-492, pis. 33- 48. 1935. Geographic elements of the marine flora of the North Pacific Ocean. Amer. Nat. 69: 560-577. Stephenson, T. A., and A. Stephenson. 1949. The universal features of zonation between tide-marks on rocky coasts. Jour. Ecol. 37(2): 289-305, pi. 8, 4 figs. Tokida, J. 1954. The marine algae of Southern Saghalien. Mem. Fac. Fish. Hokkaido Univ. 2(1): 1-264, 15 pis., 4 figs., 5 tables. WOMERSLEY, H. B. S. 1954. The species of Macro cystis with special reference to those on Southern Australian coasts. Univ. Calif. Publ. Bot. 27: 109-132, pis. 1-8, 1 map. 1956. The marine algae of Kangaroo Island, IV. The algal ecology of American River Inlet. Austral. Jour. Mar. Freshwater Res. 7(1): 64-87, 7 pis, 4 figs. Description of a New Species of Pranesus (Atherinidae: Pisces) from the Capricorn Group, Great Barrier Reef D. J. Woodland 1 During 1956-7 Richard J. Slack -Smith, De- partment of Zoology, University of Queensland, and the author investigated the fish fauna of Heron Island, Capricorn Group. A check list of the recorded species is to be published conjointly ( Pap. Dep. Zool. Univ. Qd. ) . The fish described here is one of a number of new species collected at the island. SUBFAMILY TAENIOMEMBRADINAE Schultz GENUS Pranesus Whitley Pranesus capricornensis, sp. nov. HOLOTYPE: 93 nim. standard length, from reef flat, Heron L, Aug. 1956, collected by R. J. Slack-Smith, Queensland Museum no. 1/8201. faratypes: Queensland Museum nos. 1/8202 -5; 4 specimens, 69, 78, 85, and 88 mm.; 85 and 88 mm. fish taken along with holotype, 69 and 78 mm. fish collected by Woodland, Mar. 1957, reef flat, Heron I. DESCRIPTION: Dorsal rays ¥11 ( V -VI.t ) 1,1,8 (9); anal I,i,12 (to 14); pectoral I,i,15; ventral 1,5; caudal i,8 + 7,i. Scales from upper limit of gill opening to root of caudal fin 47 ( 46 ) ; me- dian predorsal scales 21 (20); scales from ori- gin of first dorsal to midline of belly 6 l /j. Gill rakers on first right gill arch 5 (6) + l + 19 (to 22). Depth 5.0 ( 4.8-5. 1 ) ; head 4.2 (4.0-4.2); snout to first dorsal origin 1.75, to centre of anus 1.9, to anal origin 1.5 (1.4); all in standard length. Snout 4.5 (42-4.6); eye 2.4 ( 2.2— 2.4) ; tip of snout to rear of maxilla 2.8 (to 3.2 ) ; least depth of caudal peduncle 3.2 (to 3.4) ; postorbi- tal length of head 2.2 (2. 1-2.3 ) ; distance be- tween dorsal origins 1.5 ( 1.4) ; length of longest ray of pectoral fin 1.3 (1.2); interorbital space 1 Department of Zoology, University of Queensland, Brisbane. Present address: Zoology Department, Uni- versity of New England, Armidale, N. S. W. Manu- script received December 20, I960. 2.8 (2. 6-2. 8) ; all in length of head. Least depth of caudal peduncle in its length 2.4 (2. 0-2 .4). Lateral band broad, its greatest width 1.5 in postorbital length of head ( counts and measure- ments from holotype, with ranges exhibited in the four paratypes included in brackets ) . Premaxilla with front margin slightly convex; teeth minute on dentary, vomer, and palatines, on premaxilla giving a shagreen-like finish to outer surface; air bladder and body cavity end- ing bluntly in front of anal origin; bony edge of preopercle with concavity near lower poste- rior corner; gill rakers long and slender, longest l A as long again as diameter of pupil; maxilla reaching to halfway between vertical lines through anterior margins of orbit and pupil; centre of anus posterior to tips of depressed ventrals, 1.5 vertical scale rows distant from tips; vertical line through anus passing through fourth scale in front of first dorsal origin; anal fin origin under third scale in front of second dorsal origin; posterior margins of scales smooth; ascending premaxillary process short, broad based, not entering interorbital space (Fig. la); mandible without an abrupt elevation poste- riorly on each ramus (Fig. lb). The sequence employed in listing descriptive characters is after Schultz et d. (1953). The formula for recording fin counts is that now in use in many taxonomic works: upper case Ro- man numerals for spines or, as in this instance, nonstriated rays; lower case Roman numerals for striated, unbranched rays; and Arabic numerals for branched rays. DIAGNOSIS: P. capricornensis cannot be dis- tinguished from its congeners on any single fea- ture. Combination of the following characters is diagnostic: the broad lateral band; the scale row count from upper limit of gill opening to caudal fin base; and the position of the anus relative to the tips of the depressed ventrals and first dorsal fin origin. 540 Prune sus capricornensis n. sp. — WOODLAND 541 COLOUR IN LIFE: Above lateral band, green- ish; below band, silvery; lateral band silvery, but distinguishable from silver of belly by its more intense mirror-like finish; running from dorsal edge of pectoral base the full length of the up- per limit of the lateral band a narrow (5.0 in width of lateral band ) intense blue band termi- nating at root of caudal fin; a small blue spot on upper half of caudal fin scaly sheath; bluish crescent along upper 2 A of base of pectorals; dorsals, caudal, and distal half of pectoral fins faintly dusky, other fins pale; peritoneum black. COLOUR IN FORMALIN: Above lateral band, area lying under margin of each scale with dark pigmentation; lateral band black with a sugges- tion of the blue of the narrow band in its upper limits; area below lateral band pale; scaly sheath at base of caudal fin with two dark patches de- rived by anastomosing of lateral band; narrow bluish crescent on pectoral base persisting; col- our of fins as in life. ECOLOGY: Heron I. is a coral atoll 50 mi. from the Australian mainland (23° 27' S., 151° 57' E. ). In the cooler months (Mar.-Aug.) im- mense schools of atherinids are common over the reef flat when it is flooded at high tide. All individuals taken from these schools proved to be a new species (P. capricornensis ) . This fish was always seen swimming at depths of 1-3 ft. At low tides and during the warmer months it probably lives in open waters. The little tuna, Euthynnus alleteratus affinis (Can- tor), which normally restricts its movement to open water, will venture on to the reef flat to feed on P. capricornensis. Examination of alimentary canals of two adult fishes revealed various planktonic Crustacea, but the bulk of ingested material was unidentifiable. Females in spawning condition were recorded in Aug. 1956. ACKNOWLEDGMENT: I would like to thank Richard J. Slack-Smith for donating -specimens. REFERENCE Schultz, L. P., E. S. Herald, E. A. Lachner, A. D. Welander, and L. P. Woods. 1953. Fishes of the Marshall and Marianas islands. U. S. Nat. Mus. Bull. 202, 1: 1-685. New Pogonophora from the Eastern Pacific Ocean Olga Hartman 1 Continuing surveys, since 1955, of faunas in offshore areas of southern California, conducted by the Allan Hancock Foundation, using the facilities of the M/V "Velero IV,” have resulted in finds of several representatives of pogonopho- rans. One, belonging to the genus Siboglinum Caullery, is believed to represent an undescribed species. Other kinds, from different places, are briefly reported below, to indicate the wide range of occurrences in deep ocean bottoms off southern California. All come from depths ex- ceeding 950 m. and in latitudes south of 33° N. The much larger and far better explored shal- lower depths have yielded none, despite the ex- amination of thousands of grab samples from shelf, slope, basin, and canyon depths of the shelf lands between Point Conception, California, to south of the Mexican border. All specimens are deposited in the collections of the Allan Hancock Foundation. I am indebted to the Administration of the Foundation for per- mission and time to study these interesting ma- terials. The cost of collecting the materials was defrayed by funds from the National Science Foundation and the Allan Hancock Foundation. Captain G. Allan Hancock has generously pro- vided for any deficiencies in the running costs of the ship. The scientific and operating crews of the "Velero IV” took and processed the col- lections. Dr. Gosta Jiigersten, Uppsala, Sweden, examined specimens and verified their generic affinities. The illustrations were prepared by Anker Petersen. I am indebted to all of these people for their help and interest. GENUS Siboglinum Caullery, 1914 TYPE S. weberi Caullery, 1944 Siboglinum veleronis, new species Figs. 1-9 COLLECTION: More than 100 specimens were taken by the "Velero IV,” sta. 7049, May 7, I960, 1 Allan Hancock Foundation, University of Southern California, Los Angeles, California. Manuscript re- ceived January 27, 1961. 14.85 mi. 264° true from Point La Jolla, 32° 49' 37" N., 117° 35' 12" W., in 97 6 m, from the axis of La Jolla canyon near its convergence with the San Diego trench. The Campbell grab, weighing about 900 lb. with a capacity of about 5 cu. ft., recovered 2.51 cu. ft. of gray sand and green mud. In addition to the Siboglinum speci- mens, the sample yielded the kinds and numbers of animals listed below. DIAGNOSIS: Individual tubes measure 50-65 mm. long by 0.13 mm. across. Others are longer or narrower, having a length-to-width ratio of 450-720 to 1. Each tube is usually slender, cylin- drical, pale to dark straw-colored, and crossed by alternating dark brown and pale bands. The brown ones are typically paired ( Fig. 2 ) , so that both of a pair are nearly twice as long as one pale band. An occasional irregularity in this pat- tern results when the two dark ones are so close together as to appear fused, or are incomplete on one side to form a broader than usual, in- complete, pale band. The greater length of the tube is crossed by these alternating cross bars; only the distal anterior and posterior ends ( Fig. 1) are pale and usually collapsed. The animal within lies usually some distance within the banded region; none has been found partly or wholly out of the tube. With the aid of a stereo- scopic microscope the specimen can be seen through the wall of the tube and oriented with respect to the anterior and posterior regions, the tentacle in front directed forward, and the an- nular or uncinial band marking the approximate middle region of the body. However, dissection is necessary to study the animal more completely. Slitting the tube lengthwise has been found least destructive to the soft-bodied specimen; by us- ing a razor-sharp, finely pointed knife and in- serting it into the distal end of the tube, then extending the cut lengthwise for the length of the contained individual, one can then lift it out intact from the shell of the tube. An entire animal measures 15 mm. long to the annular, or uncinial, region which is near the middle, making the total length about 30 542 New Pogonophora — -H artman 543 mm. The body is nearly uniform in width but is widest in the mesosomal region where it meas- ures 0.065 mm.; the postannular region grad- ually becomes slenderer and tapers posteriorly. The tentacle is less than half as wide as the body and extends forward for a length nearly two and a half times that of the mesosomal length. Its insertion marks the ventral side of the body as herein interpreted, as well as the posterior end of the first body region or protosoma. The three body regions are not sharply separated from one another. The separation between the first and second regions is behind the tentacular insertion and is vaguely indicated by a faint line extend- ing obliquely forward around the body to the middorsum. The second region, or mesosoma, is about four times longer and is separated from the third region, or metasoma, by a transverse groove located just in front of the paired gland- ular papillae (see Figs. 5, 6). The third region is much the longest and comprises most of the length of the body. The tentacle is extended forward nearly straight in some specimens and is loosely to closely coiled in others, or only a short distal end is coiled. Most of the length has lateral, filiform branches or pinnules, inserted in two nearly reg- ular rows along the ventrolateral edge (Fig. 4); those nearer the distal end are more dispersed, whereas those nearer the base (Fig. 3 ) are closer together and tend to be curled toward the mid- ventral line. Each pinnule is very slender and longer than the tentacle is thick. These features agree with those described for other species of the genus having pinnules; only two ( S . inermis Ivanov and S. atlanticum Southward and South- ward) are said to lack them. The protosoma or anteriormost region of the body is short, subconical, tapers forward, and lacks markings; its posterior end is indicated on the ventral side of the body by the insertion of the tentacle. This fusion is firm, so that the two are not easily severed. A mouth, if present, would be expected at the posterior base of this ten- tacle, for which the principal function seems to be that of food gathering. I have seen no oral aperture or any indication of an alimentary tube but would not exclude the possibility of a length- wise transport tube of some kind in an animal of such great length. The second body region, the mesosoma, is considerably longer than the first and is cylindri- cal in shape; near its anterior end it is obliquely crossed by the bridle or frenulum (Ivanov) or girdle (some other authors). It is separated from the third region, or metasoma, by a trans- verse groove. A pair of circular pores is located at the sides, just behind the frenulum (Fig. 6); they are believed to be the external apertures of the coelomoducts for the second segment. The anterior end of the metasoma is charac- terized by the presence of ventrolateral paired glandular papillae; they are so arranged that the two members of a pair are separated from suc- cessive ones by a space about equal to that of the distance between the brown bars on the tube. It thus seems obvious that these papillar glands are directly concerned with tube secre- tion. This region is followed by the much longer goniadial region, characterized by sparsely and irregularly dispersed epithelial papillae. About halfway along the length of the body two trans- verse series of uncinial rows indicate the posi- tion of the annular region; for this reason the goniadial region in front is called the preannu- lar, and the same one behind is the postannular, region. The annulae are of particular interest because they bear hard, chitinized platelets, or unicini, in transverse series. Those of the anterior row are incomplete and number about 38, while those in the posterior row are more numerous, numbering about 44. All uncini are of one kind, and terminate distally in a single row of denti- cles (Fig. 8), numbering 9—13- Each uncinus is distally recurved so that those of successive un- cini tend to overlap. Seen from the end the den- ticles form an arcuate row (Fig. 9) . The basal or embedded stems are not chitinized but taper rapidly to very tenuous strands, and are so pro- longed that all those of one transverse series are brought together as a fascicle and carried ob- liquely forward to attach to the body wall. They may function to anchor the animal within the tube at selected places. The comparable uncini in some other species, S. atlanticum and S. iner- mis, have been shown as having multiple rows of crenulations (see Southward and Southward, 1958: 629, 631). S. veleronis was recovered from a quantitative sample covering a surface area less than a half 544 PACIFIC SCIENCE, Vol. XV, October 1961 New Pogonophora — Hartman 545 a square meter. The animals contained in the mud, without tubes, weighed about 6.4 g. (moist weight). They included the following: Siliceous sponge spicules, scattered through the debris Anemone, small white, 1 Crustaceans (not identified) ostracod, 1; amphipod, 1; cumacean, 2; gnathid isopod, 2 Mollusks: more than 100 living small gastro- pods, pelecypods, and 3 small solenogasters (not identified) Echinoderms, identified by Fred Ziesenhenne Ophiacantha normani Lyman, 2 Ophiura kojoidi J. F. McClendon, 22 Polychaetes: with 22 species and more than 200 specimens Ancistrosyllis tentaculata Treadwell, 3 Axiothella sp., 6 Aricidea aciculata Hartman, 3 A. uschakowi Zachs, 32 ampharetid, juveniles, 4 Brada glabra Hartman, 12 Chaetozone ? gracilis (Moore), 2 Chaetozone sp., fragments, 2 Cos sura Candida Hartman, juveniles, 2 Glycera Pcapitata Oersted, juveniles, 2 Haploscoloplos elongatus (Johnson), 4 Heteromastus Pflobranchus Berkeley, 4 Maldane cristata Treadwell, 13 (the largest and most conspicuous in the sample, con- tained in thick, mudwalled tubes with lateral vents and branches) Myriochele ? gracilis Hartman, juvenile, 1 Ninoe fusca Moore, 12 Nothria sp., juvenile, 1 Paraonis gracilis oculata Hartman, 77 + Pilar gis hamata Hartman, 1 sabellid tube, 1 Tereb ellides sp., juvenile, 1 Tharyx monilaris Hartman, 10 + T. tesselata Hartman, 4 Pogonophora Siboglinum veleronis, new species, 100 + The total number of species and specimens is thus 30+ and 428 + . SYSTEMATIC AFFINITIES: S. veleronis, the eleventh species to be described in the genus Siboglinum, differs from others in that the tube has double bands of dark brown bars, and the uncini of the annular region have 9-13 teeth in a single transverse row. The proportional lengths of the first and second body regions, shown in Figures 5 and 6, differ from those of other species. It is noteworthy that all but two of the known species occur in far northern latitudes. Six come from the northwestern end of the Pacific Ocean; three others are from the northeastern part of the Atlantic. The first species of the genus, S. iveberi Caullery, came from the Malay trench in great depths, and the present species is the first to be described from the western hemisphere. These species, with their distributions are as follows : S. tueberi Caullery, 1944, IndoPacific area, in 462-2060 m., in fine sand and mud. S. ekmani Jagersten, 1956, Skagerack, north- western Europe, in 487-650 m., clay. 5'. caulleryi Ivanov, 1957, Ochotsk Sea, in 90- 200 m., and other northwestern Pacific areas in depths of 23-8100 m., in the Kurile trench. S. cinctulum Ivanov, 1957, northwest Pacific, in 3420 m., in the Kurile trench. S. pellucidum Ivanov, 1957, southwestern part of the Bering Sea in 1740-4820 m. S. minutum Ivanov, 1957, Bering Sea in 3740-3840 m., and Kurile-Kamchatka trench in 5540 m. S. fedotovi Ivanov, 1957, southwestern part of the Bering Sea in 3330-3940 m. S. plumosum Ivanov, 1957, Ochotsk Sea east of Japan, in 124-318 m. FIGS. 1-9: Siboglinum veleronis, new species, from "Velero IV” sta. 7049- 1, Entire tube con- taining animal, X 22; 2, part of tube showing paired dark alternating with pale bands, X 89; 3, part of tentacle from near base, showing pinnules curled toward midventrum, X 228; 4, another part of tentacle from another individual, with pinnules nearer tip of tentacle, X 228; 5, animal with tube partly dissected away, tentacle directed forward, in left lateral view, X 25.5; 6, enlarged anterior regions including protosoma, mesosoma with frenulum and pore of coelomoduct, and part of glandu- lar metasoma, in left lateral view, X 83; 7, an uncinus from annular zone, seen from denticled side, X 5,720; 8, an uncinus seen from opposite side, X 5,720; 9, an uncinus seen from top, showing arcuate arrangement of denticles, X 5,720. 546 PACIFIC SCIENCE, Vol. XV, October 1961 S. atlanticum Southward and Southward, 1958, from the northeastern Atlantic, in 600- 1400 m. S. inermis Southward and Southward, 1958, from the same locality as the preceding. S. veleronis, n. sp., from La Jolla canyon, southern California, in 976 m. In addition to S. veleronis, the collections of the Allan Hancock Foundation contain repre- sentatives of three other pogonophores. One is mentioned earlier (Hartman and Barnard, I960: 283) coming from West Cortes, East Cortes, and Long basins and from the San Diego trench. The collar segments of the tube measure 2.9 mm. across and 3.3 mm. long; each is drab dark brown, cylindrical, paler at one end where it flares and is frayed. In size and shape they most nearly resemble those shown for Galath ealinum (Kirkegaard, 1956: 81). Another kind, a Siboglinum sp., has just been received from a trawl station, "Velero IV,” sta. 7231, off San Eugenio Pt., Lower California, Mexico, in 4800 m. This organism inhabits tubes which, in size and color, resemble those of S. veleronis, but the alternating brown bands are single, not double, and range from four to six times as long as the pale ones, or only a lit- tle longer. The protostoma has a pair of small red eyespots at the sides — the first known in- stance of eyes in pogonophores. The ventral side of the tentacle appears villous, due to paired rows of long, closely set pinnules. The uncini are in two transverse, single rows and are set far apart, the distance being about equal to four times that of the body width. A fourth species is S. ekmani Jagersten, from the Skagerack, Denmark, a gift from the Zoo- logical Museum at Copenhagen. REFERENCES Caullery, M. 1914. Sur les Siboglinidae, type nouveau d’invertebres, recueilli par l’expedi- tion du Siboga. C.R. Acad. Sci. Paris, tom. 153: 2014-2016. 1944. Siboglinum Caullery 1914, type nouveau d’invertebres, d’affinites a preciser. Siboga Exped., Mongr., 25 bis, pp. 1-26. Hartman, O., and J. L. Barnard. I960. The benthic fauna of the deep basins off southern California. Allan Hancock Pacific Exped., 22: 217-284. Hyman, L. H. 1959. The Invertebrates: Smaller Coelomate Groups. Vol. V. McGraw Hill Book Co., N.Y. 783 pp., 239 figs. Ivanov, A. V. 1957. Neue Pogonophora aus dem nordwestlichen Teil des Stillen Ozeans. Zool. Jbr., Abt. Syst. 85: 430-500, 59 figs. Jagersten, G. 1956. Investigations on Sibogli- num ekmani, n. sp., encountered in Skagerak, with some general remarks on the group Pogonophora. Zool. Bidr. Uppsala, 31: 211 — 252. Kirkegaard, J. 1956. Pogonophora. Galath eali- num bruuni n. gen. n. sp., a new representa- tive of the class. Galathea Report 2: 79-83, 2 figs. Southward, E. C, and A. J. Southward. 1958. On some Pogonophora from the north- east Atlantic, including two new species. Ma- rine Biol. Assoc. Plymouth, Jour. 37: 627- 632, figs. 1-3. Acanthophora, a Possible Invader of the Marine Flora of Hawaii 1 Maxwell S. Doty 2 In THE FALL OF 1952, a small algal fragment was brought to the author from Pearl Harbor 3 by Mr. Charles Cutress. In April, 1953, a much larger piece of the alga was brought in by Mr. Spencer Tinker, who had found it washed ashore near the Waikiki Beach Laboratory of the Uni- versity of Hawaii. Both collections were readily determined as representing a species of Acan- thophora Lamouroux (1813: 132), a rhodo- phytan genus. This genus is distinct and clearly recognizable among the floral elements occur- ring in this part of the world. A search of the literature and the herbaria available revealed no Pacific records of this genus from the Ha- waiian Islands or, with one exception, east of the Western Caroline and Marianas islands north of the equator. Other collections during succeeding years, and field observations as well, revealed a huge in- crease in the abundance of the species in Ha- waii during the next few years. In May, 1953, Dr. D. W. Strasburg found this alga "in abun- dance” at Keehi Lagoon, between Pearl Harbor and the Port of Honolulu on the leeward side of Oahu. Later the same month, a dense growth of the alga was found by the author ( numbered 10774) and Dr. E. Y. Dawson at Hauula, north of Honolulu, on the windward side of the is- land of Oahu. Parts of this collection are the earliest collections from Hawaii represented in both the Bishop Museum and the University of California herbaria. From that time on, Acan- thophora has been so common on the leeward 1 This work was made possible by a research time grant from the Graduate Research Committee of the University of Hawaii and facilitated by Office of Naval Research contract 2591 (00). Contribution number 154 of the Hawaii Marine Laboratory. Manuscript received January 23, 1961. 2 Botany Department, University of Hawaii, Hono- lulu 14, Hawaii. 3 According to Mr. Mikihiko Oguri, this algal col- lection probably came from West Loch, between Lau- launui Island and the northwestern shore. side of Oahu that it has not often been preserved as an herbarium specimen. Finally, during June, 1956, Dr. Otto Degener collected and sent in a specimen (his no. 24105) from Mokuleia on the windward shore of the island of Oahu, northwestward from Honolulu, collected by himself, Miss Marie Neal, and Dr. Constance Hartt, with the annotation ”... ubiq- uitous some distance within the reef; observed very rare here last year. This is first time aggres- sive marine alga threatening native kinds.” Cer- tainly it appears to be replacing (crowding out) elements of the native flora. Degener, who has paid close attention to the reef population at this particular site, feels the alga probably was not there until, at most, 2 years before this collection was made. Not only have frequency of observance and density of standing crop increased, but the dis- tribution has been that of a progressive en- circling of Oahu, one of the few islands of the Hawaiian Group where Acanthophora has been found. The alga has spread to Kauai, another of the Hawaiian Islands. Mr. Jan Newhouse has passed on local stories that the alga was not found around Kauai until about 1954 or 1955, and the observation that it is now ubiquitous. This genus was recorded from Kauai by Kohn and Helfrich ( 1957 : 243 ) . Their mention of the genus was based on observations of Newhouse about November, 1956, and was not included among the algae they collected and had identi- fied by Dawson at the time their earlier work (Helfrich and Kohn, 1955) was completed in October and November of 1955. It has been found neither among our earlier extensive col- lections from Kauai nor by Newhouse among his, but during August I960, Dr. Charles La- moureux and Mr. Tadayuki Kato made a collec- tion (Lamoureux coll. no. 1542) of Acantho- phora on Kauai at Poipu Beach. Specific, though in part cursory, searches dur- ing I960 for Acanthophora on the islands of 547 548 PACIFIC SCIENCE, Vol. XV, October 1961 Hawaii, Lanai, and Maui variously by Robert K. S. Lee, Mikihiko Oguri, Warren Wilson, and the author have resulted in no traces of this alga being found with one exception. The ex- ception is a collection made by the author (numbered 19431), Mrs. Meng Sung Doty and Mr. Lee along the north shore of Lanai in No- vember, I960, where the alga was washed onto the muddy sand beach in abundance, free or attached to shells, coral, or even rocks up to 2 lb. in weight. Unfortunately, the Kawaihae area on Hawaii and the shores of Molokai have not been specifically searched for this genus, but our collections made during earlier years from these places do not include Acanthophora. In the light of the ability and persistent vigor of the algologists who have at one time or another concerned themselves with the marine algae of the Hawaiian Islands, e.g., Drs. W. A. Setchell, Josephine Tilden, G. F. Papenfuss, 4 I. A. Abbott, and earlier, the Misses Minnie Reed and Marie Neal, and Mrs. Nina H. Loomis, it seems unlikely that this alga would have been overlooked had it been consistently present. It is a conspicuous alga. The older Polynesians in Hawaii seem to have had no name for Acan- thophora. If pressed for a Polynesian name now- adays, the common man professing native acu- men will apply local names such as manauea, the name widely used formerly for species of other genera, such as Gracilaria (now usually referred to by the Japanese name, ogo). Identifying the Hawaiian alga has led to a consideration of the differences purported to exist between the several species reported in the Pacific. Many variants can be found in the mate- rial that has been available for this study from both the Atlantic and the Pacific, but for the present it is felt that the many forms found might best be treated as variants of one species. Acanthophora spicifera 5 (Vahl) Boergesen (1910) is the name for this species having priority insofar as we know. 4 In correspondence, Dr. Papenfuss tells us that neither he nor Setchell found this genus in Hawaii, and that the only Hawaiian specimens in the Univer- sity of California herbarium are duplicates of the Hauula collections sent in by Dawson and mentioned above. 5 Basonym —Fucus spiciferus Vahl, 1802. Among the most common names 6 considered here as having been applied to the taxon A. spicifera, as found in the Pacific, is A. orientalis J. Agardh (1863). In describing A. orientalis as a new species, J. Agardh listed the Marianas Islands, of which Guam (13° N., 145° E.) 7 is one, as the source of one of the two collections he had seen. The other collection was probably from Manila Bay ( 14° N., 121° E.) in the Philippines. Safford (1905: 30-32) says that in the Marianas, the islands Guam, Rota ( 14° N., 145° E.) and Tinian (15° N., 146° E.) were visited by the Freycinet expedition. The material of this expedition from these islands is believed to be the source of one of the two collections Agardh reported. In his text, Safford (1905: 177 f.) lists A. orientalis from Guam and we presume this to be based on the Freycinet record, since Safford also says Dumont d’Urville collected several new species of algae on Guam. We ourselves have seen no specimens from Guam, despite a search through the several col- lections, now in our possession, which were made there by Mr. Ernani Menez in I960. Except for the reports from Hawaii, the genus is not known to occur in the Pacific east of the Marianas other than in the Ponape region, where it has been reported by Yamada (1944: 44) as A. muscoides (L.) Bory from Ant (7° N., 158° E.), an atoll 8 mi. to the southwest of Ponape. It is common about the large subcontinental or continent-related islands of the far western Pacific and, as A. spicifera, according to Womer- sley ( 1958) , in northern Australia. The genus is reported (Kanda, 1944: 749) from Palau (7° N., 134° E.) as A. orientalis. As A. thierii Lamx. the genus is recorded from the Admiralty Islands (probably 2° S., 147° E.) and Tonga- tabu (21° S., 175° W.) by Dickie (1875*: 238, 235, resp.) and from Torres Straits (10° S, 143° E.) by Dickie (1876: 447). Acan- thophora is common in the warmer part of the Atlantic, and Lamouroux (1813) believed the genus to be circumequatorial. It seems entirely possible that this species 6 The only similar species not mentioned otherwise here appears to be A. aoki Okamura, 1934. 7 The approximate latitude and longitude in degrees is given for the convenience of those interested in the location of the places named. Acanthophora in Hawaii — Doty 549 could have arrived in Hawaii from the west on a ship bottom, i.e., a man-made facilitation of the oceanic drift method of dispersal. This would be the carrying of an organism "up- stream.” 8 The progressive increase in abun- dance around the island over several years’ time, and, recently, what appears to be a leveling off in abundance, is considered evidence of an in- troduction into the Honolulu-Pearl Harbor area. The Honolulu-Pearl Harbor shore area is the part of the state having the greatest traffic with regions of the world where Acanthophora has been known as a common component of the flora for a long time. This area is not climatically extreme for the state. However, since Honolulu has been in contact with the East and West via ship for centuries, it seems likely that some re- cent unusual occurrence may have implemented this transport. The three following events have come to our attention and are considered in this regard. First, the recent warming of North Pacific waters (e.g., since the low temperature year of 1955 at Christmas Island, 2° N., 157° W.) would not, it seems, be accountable for the fol- lowing reasons: first, the warm temperatures did not begin until after the alga was well known in Hawaii, and, secondly, there has been little abnormality of water temperatures in the Hawaiian area itself. A second unsual event that may have led to the introduction is the increase in traffic be- tween the Honolulu-Pearl Harbor area and the Far East during World War II, 1941-45, and during the Korean Police Action, 1950-53. The idea is that a number of small introductions at nearly the same time might have provided to- gether a sufficiently large inoculation for the species to become established. From the rate of spread we tend to exclude the first period. The second is more timely. There is no special evi- dence that would lead one to choose this latter as the probable period, though such changes in oceanic traffic have been held to account for 8 This is the customarily postulated direction of mi- gration in deriving the populations of Hawaii, as sum- marized by Zimmerman (1948), and Gosline and Brock (I960). Ladd (I960) has emphasized a some- what different possible mode of origin for the popu- lations of the Pacific islands. the distribution of barnacles in some cases else- where in the world. A heavily fouled barge, the "Yon 146,” towed to Pearl Harbor from Guam, provides an exam- ple of the third, and more specific, type of event that may have led to the establishment of Acan- thophora in Hawaii. The idea here is that one heavily "fouled bottom” could have provided a sufficiently large inoculation for the species to become established. Upon arriving in Pearl Har- bor February 3, 1950, this "fuel oil barge (non- self-propelled)” was placed in a dry dock. The Pearl Harbor dry docks are about 12 km. by water and 8 km. in a straight line from the place Cutress collected the first material of Acanthophora found in Hawaii. The dry dock is about 30 km. by water from the place Tinker first found this species of Acanthophora. Fish and gastropod collections were made both from the growths, often 3-8 in. thick, on the barge and from the small pools left under the barge in the otherwise dry dry dock. They were made, at least in part, on April 10, 1950, and variously by Tinker, George Campbell, and Kenneth A. Wong. This vessel, 200 ft. long, 56 ft. in beam, concrete-hulled, under different descriptive names, has been mentioned as a pos- sible means of fish introduction by several au- thors, e.g., by Gosline and Brock (I960: 26), who have studied collections made on it. The same vessel is reported by Edmondson (1951: 183, 212) as having brought in invertebrates which have become established, 9 such as the brachyuran crab, Schizophrys asp ere, common to the far western Pacific. Chapman and Schultz 9 From accounts of long-time residents of the area, known distribution of the species, and information in the literature, it seems to me almost certain that Coty- lorhizoides pacijicus (Mayer) and Cassiopea medusa (Light), both Rhizostomae, were accidentally intro- duced to the Pearl Harbor area during the 1941-45 period. The type locality of these two medusae is the Philippine Islands. In Hawaii both were restricted to Pearl Harbor until about 1950. About this time, Cassiopea appeared in Honolulu Harbor and the Ala Wai Canal. Later, 1953-54, Cotylorhizoides appeared in Kaneohe Bay. Until the time of my departure (De- cember, 1955) neither medusa was known from the other Hawaiian Islands, Line Islands, Marshalls, Gil- berts, etc. Both forms undoubtedly came to Hawaii as scyphistoma on ships or the like, as neither are medu- sae of the open ocean. — C. E. Cutress, June, I960. 550 PACIFIC SCIENCE, Vol. XV, October 1961 ( 1952 ) concluded there was no evidence among the fish records they obtained that species had been brought to Hawaii in the fouling popula- tion on that barge. Dr. C M. Burgess, who, along with Campbell and Tinker, provided much of the specific information on these events, told the author that the species of the molluscan genus Cypraea brought in on this vessel, the "Yon 146,” did not become established in Ha- waii, and this has been affirmed by Dr. Alison E. Kay, who is a student of this genus. In commenting upon Tripterygion hemimelas Kner & Steindachner and Ecsenius hawaiiensis Chapman & Schultz, two blennioid fishes re- ported to have been collected from the pools of water in the dry dock holding the "Yon 146,” Strasburg (1956: 245 f.) notes that the speci- mens of T. hemimelas were similar to a species from the Samoan Islands (14° S., 171° W.) . The service record of this barge, as far as the author has been able to trace it, indicates that it was tied up at Apra (13° N., 145° E.), Guam, from 1945 to the date it was towed to Pearl Harbor, a period of about 4 years. Thus is does not seem likely that it would have been directly the means by which a Samoan fish would have been introduced into Hawaiian waters. Whether T. hemimelas occurs in Guam or not is not know# to the present author. Possibly the blennioid fish, Omobranchus elongatus ( Peters ) , was brought to Hawaii (Strasburg, 1956: 257) from the Samoan area along with chunks of reef rock bearing living specimens of the giant clam, Tridacna. At least for the present, this splendid possible avenue of introduction is discounted; though Acantho- phora has been reported (Reinbold, 1896) as A. orientalis from Upolu (14° S., 171° W.), Western Samoa. Individual ships have been cited previously as the means by which algae have been intro- duced into the Central Pacific. Dickie (1875^: 33) published a note to the effect that Ulva latissima Linnaeus was introduced to Mangaia (22° S., 158° W.) in the Cook Islands when a whaling ship from the Antarctic was wrecked there on the reef in 1852. There is the pos- sibility, however, that the wrecked ship merely provided a favorable habitat, in which habitat an ulvoid alga interpreted by Dickie as repre- sentative of this specific taxon appeared. In form the ulvoid algae, of the larger benthic algae, are among the most plastic in respect to environmental conditions. Kohn (1959: 81) records Acanthophora (using the binomial A. orientalis) from Kaneohe Bay, Oahu, Hawaii, where it was the substratum upon which the eggs of Conus quercinus were found attached in February, 1956. C. quercinus has been recorded for Hawaii for many years (e.g., Bryan, 1915: 454), but the alga for only a few years. If C. quercinus is very host-specific in its egg-case depositing, this observation of Kohn s could be taken to imply long presence of Acanthophora in Hawaii. Though the alga is independent of the mollusc, if the mollusc is restricted to the algal species for egg-case de- position, the alga would probably have had to be here first and it would have taken many years for the mollusc to develop modified egg- case depositing habits including Acanthophora as a host. Our impression 10 is, however, that egg-case attachment by molluscs is not very specific, substratum-wise, and therefore, that there is no implication in Kohn’s record that Acanthophora grew in the islands, say, in 1915. After considering the types of events described above, we feel that it is most likely that A. spicifera arrived recently in Hawaii via the fouled bottom of a ship. Aside from the ordinary ship traffic, similar opportunities for introduction by vessels other than the "Yon 146” are known. One of these opportunities is provided in the case of a similar vessel, the "Yogn 41,” which was towed from Subic Bay (15° N., 120° E.) in May, 1947. This "gas- oline barge (non-self-propelled)” was dry- docked in February, 1950, in Pearl Harbor. While Subic Bay is unknown phycologically, A. spicifera is common in the general area, e.g., in Manila Bay a few miles to the south it is abundant. However, from the timing of the events and from the rate of spread after the first specimens were found, it is believed that 10 A letter received since from Prof. J. M. Oster- gaard supports this impression concerning the specific- ity of egg-case deposition by Conus and tells us of his "finds” of C. quercinus in the Honolulu Harbor area as dead shells in 1902 and 1905 and as living shells in 1915. Acanthophora in Hawaii — -Doty 551 an earlier vessel would be much less likely a source of the introduction of A. spicifera than the "Yon 146.” 11 REFERENCES Agardh, J. G. 1863. Species genera et ordines algarum ... 2. Part 8: 701-1278. Boergesen, F. 1910. Some new or little known West Indian Florideae, II. Botanisk Tids- skrift 30. 177-207. Bryan, W. A. 1915. Natural History of Ha- waii. . . . Privately printed, The Hawaiian Gazette Co., Ltd., Honolulu. 596 pp. Chapman, Wilberg M., and Leonard P. Schultz. 1952. Review of the fishes of the blennioid genus Ecsenius, with descriptions of five new species. Proc. U. S. National Mus. 102 (3310): 507-528. Dickie, G. 187 5^. Notes on algae from the Is- land of Mangaia, South Pacific. Jour. Linn. Soc. 15: 30-33. 1875£. Contributions to the botany of the expedition of H.M.S. "Challenger” Algae, chiefly Polynesian. Jour. Linn. Soc. 15: 235-246. — 1876. Notes on algae collected by H. N. Moseley, M. A., of H.M.S. "Challenger,” chiefly obtained in Torres Straits, Coasts of Japan, and Juan Fernandes. Jour. Linn. Soc. 15: 446-455. 11 Though we do not know what other biological consternation may be associated with this vessel, it may cause little more for Hawaiian biologists, for it is reported to have been sunk in Subic Bay on Octo- ber 16, 1955. Edmondson, Charles Howard. 1951. Some Central Pacific crustaceans. Occ. Pap. Bishop Mus. 20(13): 183-243. Gosline, William A., and Vernon E. Brock. I960. Handbook of Hawaiian Fishes. Univer- sity of Hawaii Press, Honolulu, ix -f 372 pp., illus. Helfrich, Philip, and Alan J. Kohn. 1955. A survey to estimate the major biological effects of a dredging operation by the Lihue Plantation Co., Ltd., on North Kapaa Reef, Kapaa, Kauai. Preliminary Report. Mime- ographed, privately distributed. 31 pp. (An expansion of one aspect of this work was published by Kohn and Helfrich, see below. ) Kanda, Tiyoiti. 1944. Ecological studies on marine algae from Kororu and adjacent is- lands in the south sea islands. Palao Tropical Biol. Sta. Studies 11(4): 733-800. Kohn, Alan J. 1959. The ecology of Conus in Hawaii. Ecological Monog. 29(1): 47-90. Kohn, Alan J., and Philip Helfrich. 1957. Primary organic productivity of a Hawaiian coral reef. Limnol. Oceanogr. 2(3): 241-251. Ladd, Harry S. I960. Origin of the Pacific is- land molluscan fauna. Amer. Jour. Sci., Brad- ley Volume 258A: 137-150. Lamouroux, J. V. F. 1813. Essai sur les genres de la famille des Thalassiophytes non arti- culees. Ann. Mus. Hist. Nat. Paris 20: 21— 47, 115-139, 267-293, pi. 7-13. Okamura, K. 1934. leones of Japanese Algae. Vol. VII. p. 5, pi. 318, figs. 15-17. Reinbold (Itzehoe), D. 1896. Meeresalgen. In: F. Reinecke, Die Flora der Samoa-Inseln. Botanische Jahrbiicher 23: 266-275. 552 PACIFIC SCIENCE, VoL XV, October 1961 Safford, William Edwin. 1905. The Useful Plants of the Island of Guam; with an In- troductory Account of the . . . Island. U. S. Dept. Agr., Contrib. U. S. Nat. Herb. 9. 416 pp. Strasburg, Donald W. 1956. Notes on the blennioid fishes of Hawaii with descriptions of two new species. Pacif. Sci. 10(3): 241- 267. Vahl, Martin. 1802. Endeel krytogamiske Planter fra St. Croix. Skrivter af Naturhis- torie-Selskabet 5(2). (Not seen.) WOMERSLEY, H. B. S. 1958. Marine algae from Arnheim Land, North Australia. Records of the American- Australian Expedition to Arn- hein Land. 3(4): 139-161. Yam ADA, Yukio. 1944. A list of the marine algae from the atoll of Ant. Sci. Pap. Inst. Algological Res. Hokkaido Univ. 3: 31-45. Zimmerman, Elwood C. 1948. Insects of Ha- waii, Vol. 1. Introduction. University of Ha- waii Press, Honolulu, xx + 206 pp. Post-larval Food of the Pelagic Coelenterate, Velella lata Robert Bieri 1 Huge, unexpected swarms of Velella that have drifted onto the coasts of France, England, the United States, and other countries have stimulated many of the studies on the "purple sail.” Although recent reports (Savilov, 1958; Bieri, 1959) have indicated some of the reasons for the apparently haphazard appearance and disappearance of these swarms, we have little or no idea of the effect of such sporadic inva- sions on the local fauna. The size of these popu- lations is indicated in several published reports, of which Woltereck’s (1904) is typical. This particular swarm when cast onto the beach at Villefranche formed a mound Vi m. wide, Vj m. high, and fully 1000 m. long. Such huge populations must have a considerable effect on the community of organisms in the sea beneath them. This paper records some quantitative data on the food of Velella. Some possible effects of pre- dation by Velella on associated zooplankton are suggested. MATERIAL AND METHODS In the present study, 137 specimens were used. Ninety-nine of these were obtained by dip-net between 0915 and 1000 at 32° 4l' N., 121° 04' W. on May 10, 1950. These specimens were sorted into five size-groups and preserved in formalin. On March 29, 1954, 38 specimens were collected in a special surface net between 1400 and 1500 at 32° 40' N., 118° 16' W, and were also preserved in formalin. In the laboratory all gonozooids were cut from the specimens and examined at 12 X mag- nification. Those parts containing visible food were removed and dissected, and the food items 1 Antioch College, Yellow Springs, Ohio. The early stages of this study were carried out at the Scripps Institution of Oceanography, La Jolla, California, and the Lamont Geological Observatory, Palisades, New York. Manuscript received July 29, I960. were identified. The main central gastrozooids were dissected separately. Details of the weight determinations are given in Bieri and Krinsley (1958). RESULTS AND DISCUSSION Only three brief comments on the food of Velella appear in the literature. Huxley (1858) found copepod remains in the gonozooids, while Lebour (1947) reported a young "macerated” Velella with its "stomach” (main central gas- trozooid) full of harpacticoid copepods. Totten (1954) reported a calanoid copepod and crus- tacean remains in the gonozooids. No quantita- tive data are available. In the present study most of the food was found in the gonozooids. Only 33 per cent of the specimens examined had food in the main central gastrozooids. Digestion occurs in both the gonozooids and the gastrozooid, as is evi- denced by the exoskeletons of crustaceans found in both places. The food in the gastrozooid was somewhat larger on the average than that in the gonozooids. Possibly the gastrozooid ingests the organisms caught by the gonozooids that are too large for them to ingest. However, the gastrozooid also ingests small items. In the 99 specimens taken off California in May 1950, fish eggs (mostly those of jack mac- kerel, Trachurus symmetricus 2 ) made up 48 per cent of the total food by count. Euphausiid eggs made up 7 per cent of the food items. On the other hand, in 38 specimens taken off California in March 1954, euphausiid eggs made up 78 per cent of the food items and fish eggs 3 per cent by count. The fish eggs had an average diameter of 1.1 mm., the euphausiid eggs a mean diameter of 0.41 mm. Thus a fish egg has nearly 17 times 2 Dr. E. H. Ahlstrom, U. S. Fish and Wildlife Serv- ice, La Jolla, Calif., kindly identified the fish eggs. 553 554 PACIFIC SCIENCE, Vol. XV, October 1961 FIG. 1. Ventral side of Velella killed and fixed in process of eating three large and three smaller fish eggs. Smaller eggs, marked by vertical lines, have diameter of 1.1 mm. Gastrozooid surrounding central egg has broken and appears as thin film above egg. Darkest egg is not completely ingested. the volume of a euphausiid egg. This means that in the Velella taken in March of 1954, fish eggs almost equalled the volume of euphausiid eggs in the diet of Velella, while in the May 1950 sample fish eggs were 120 times the vol- ume of euphausiid eggs. The ventral side of a Velella taken in the process of eating six fish eggs is shown in Figure 1. When the Velella were collected in March 1954, a surface net tow was taken at the same time. In Table 1 the per cent composition of the plankton is compared to the per cent com- position of the food of Velella. Euphausiid metanauplii and copepods were 10 times as common in the plankton as in Velella food. On the other hand, euphausiid eggs were nearly 8 times as common in the food of Velella as in the surface zooplankton. Larvaceans, which made up 10 per cent of the food by count, formed less than 0.1 per cent of the plankton. If the plankton sample is representative of the food that was available to this particular Velella population (change with time or micro-vertical distribution differences may mean the plankton sample is not representative), it appears that motile organisms such as copepods and nauplii are not caught as effectively as are weaker swim- ming zooplankton, such as the larvaceans or the nonmotile eggs of fish and invertebrates. Never- theless, comparatively large and active organ- isms such as larval fish and adult euphausiids are occasionally caught. In Figure 2 the per cent composition by count of the food of the 99 specimens taken in May 1950 is shown as a function of mean length of Velella. Apparently there is no marked food selection by the different sizes of Velella although the calanoid copepods (3) increase steadily in importance when Velella surpasses a length of 30 mm. This is another indication that motile organisms are not caught as effec- tively as are nonmotile objects. Besides jack mackerel eggs and the items listed in Table 1, the following items were found in the gono- Food of Velella lata — -Bieri 555 FIG. 2. Per cent composition, by count, of Velella food as function of total length of Velella. Shaded area is crustacean food. 1, Euphausiid eggs, 2, barnacle cyprids, 3, calanoid copepods, 4, other crustacean food. There appears to be no marked food selectivity by the different sizes, although copepods increase in impor- tance in diet at sizes above 30 mm. zooids and main central gastrozooids of Velella but were not studied quantitatively: larval fish, including a saury; chaetognaths and their eggs; barnacle nauplii, probably of Lepas; siphono- phores; larvae of decapod Crustacea; adult my- sids; copepod eggs and nauplii; larval and adult euphausiids; corycaeid copepods; pteropods. Velella also were found to have eaten the dia- tom Coscinodiscus and the medusae of other Velella. One specimen had eaten a fish scale. Specimens cast upon the beach were found to have ingested sand into the gastrozooid as well as into the gonozooids. Thus it appears that Velella is a carnivore, feeding on anything it can catch, generally weakly swimming or non- motile zooplankton. In Figure 3 the mean number of fish eggs Fig. 3- Amour t of food caught as function of length of Velella. Although number of fish eggs in- creases rapidly with length, weight of food caught per gram of wet body weight decreases after length of about 40 mm. caught by various sizes of Velella is shown as curve A. The number increases rapidly with in- creasing length of Velella. The amount of food in each size group of Velella was too small to allow reliable weight determinations. Therefore, the wet weight of food present was estimated by calculating the volume of the fish eggs, assuming their density to be one, and assuming that the crustacean food on the average was one-third the weight of the fish eggs. On the basis of these assumptions, the weight of food caught per gram wet body weight of Velella is shown as a function of the length of Velella (Fig. 3, curve B) . At lengths greater than about 40 mm. the Velella increase in weight more rapidly than they increase the amount of food caught. TABLE 1 Food of Velella Compared to Associated Zooplankton BY COUNT % OF Velella FOOD (38 specimens) % OF PLANKTON ( 0-Vi m. net tow) Euphausiid eggs 78 10 Euphausiid metanauplii 0.5 52 Copepods 3 34 Anchovy eggs 3 3 Other fish eggs 0.5 <0.1 Larvacea 11 <0.1 Barnacle cyprids 0.5 <0.1 Hyperiid amphipods 0.5 <0.1 Cladocera ( Evadne ) 1.0 <0.1 Emerita larvae 1.5 <0.1 Parasitic copepods, Caligus 0.5 <0.1 556 PACIFIC SCIENCE, Vol. XV, October 1961 CONCLUSIONS The data given above indicate that Velella is essentially carnivorous, feeding more or less in- discriminately on zooplankton primarily within the size range 0.2-10 mm. Motile organisms are not caught as effectively as nonmotile forms. Velella may be an important predator on fish eggs, which make up a major part of its diet. Euphausiid eggs are also an important part of its food. Off the California coast the seasonal appearance of Velella on the surface is corre- lated with the spring spawning of pelagic fish, such as hake, saury, sardine, and jack mackerel. The relatively greater abundance of suitable food during the spring months may be one of the factors responsible for the seasonal cycle of reproduction and growth in Velella. REFERENCES Bieri, R. 1959. Dimorphism and size distribu- tion in Velella and Physalia. Nat. 184: 1333— 1334. Bieri, R., and D. H. Krinsley. 1958. Trace elements in the pelagic coelenterate, Velella lata. J. Mar. Res. 16(3): 246-254. Huxley, T. H. 1858. The Oceanic Hydrozoa. Ray Society, London. 141 pp. Lebour, M. V. 1947. An interesting young Velella in the Plymouth plankton. J. Mar. Biol. Assoc. 26: 548-550. S AVILOV, A. I. 1958. Pleuston of the western Pacific. Doklady Akad. Nauk SSSR 122(6): 1014-1017. fin Russian.} Totten, A. K. 1954. Siphonophora of the In- dian Ocean, together with systematic and biological notes on related specimens from other oceans. Discovery Repts. 27: 1-162. Woltereck, R. 1904. Ueber die entwicklung der Velella aus einer in der Tiefe vorkom- menden Larve. Zool. Jahrb. Supp. VII, Fest- schrift A. Weismann: 347-372. Interspecific Differences in the Reaction to Atropine and in the Histology of the Esophagi of the Common California Sea Hares of the Genus Aplysia Lindsay R. Winkler and Bernard E. Tilton 1 During a study of the effects of certain cho- linergic agents on the tissues of Aplysia, it was noted that the esophagi of the two California species (A. calif ornica and A. vac carta) reacted divergently to atropine. This is of interest to both the taxonomy and physiology of the genus as well as potentially to a better understanding of the mode of action of atropine. Other drugs commonly known to show activity on muscle tissue were also tested on the two species to determine if any other interspecifically diver- gent reactions existed. These pharmacological reactions will be reported later. Botazzi (1898) observed the periodic con- tractions of the esophagus of the European Aplysia and made a thorough study of its phys- iology. The physiology of the crop was studied extensively by von Briicke (1905). Straub (1907) reported the reaction of the heart of an unspecified species of Aplysia to muscarine and noted the lack of atropine antagonism in the heart ventricle. Hogben (1924) studied the re- action of the crop of an European Aplysia to epinephrine. MATERIALS AND METHODS Small- to medium-size specimens of Aplysia (Neaplysia) calif ornica Cooper and Aplysia {Aplysia) vaccaria Winkler were collected dur- ing the summer months at Lunada Bay, Palos Verdes, Los Angeles Co., California. The ani- mals were packed in wet Pelvetia fastigata and transported to the laboratory where they were maintained in a 10-gal. salt water aquarium until needed. A Cole-Parmer model all-plastic impeller-type pump drew the water from the aquarium through a filter containing glass wool and activated charcoal and pumped it serially 1 Department of Pharmacology, Loma Linda Uni- versity School of Medicine, Loma Linda, California. Manuscript received January 13, 1961. through two 5 gal. carboys maintained in a re- frigerator. It was thus possible to keep the water clean and cooled to approximately the tempera- ture of the intertidal environment of these ani- mals. Parsley obtained in the local market was eaten in quantity by A. calif ornica but was re- fused by A. vaccaria. Consequently all speci- mens of the latter were used as soon as prac- ticable. Animals were sacrificed by incising the entire length of the foot, turning the animal inside out and removing the esophagus after tying it at both ends. The esophagi were suspended from a plastic holder in conventional baths using 30 ml. of sea water. The movable end of the esoph- agus was ligated to a Grass force-displacement transducer, which was connected to a Grass am- plifier-recorder. To identify the two types of excursions noted in the tracings, cross sections of esophagus tis- sue were tied on opposite sides of the ring thus formed so as to obtain tracings of the contrac- tions of the circular muscle with little or no effect from the longitudinally oriented tissue. At- ropine was used in 1: 1000 solution ( 1 mg/ml) in all cases. For histological examination both longitudi- nal and cross sections were made using standard techniques. These were stained with eosin and hematoxylin. Sections were also made of all the noticeably differentiated areas of the digestive tract. EXPERIMENTAL RESULTS The normal pattern of contraction as re- corded from the isolated esophagus consists pri- marily of comparatively rapid short excursions which vary in frequency and amplitude. These may be interspersed with occasional contractions of greater amplitude, especially in A. calif ornica (Fig. la). Since the latter contractions were ab- 557 558 PACIFIC SCIENCE, Vol. XV, October 1961 FIG. 1. The normal tracing of the esophagus of A. calif ornica is indicated in a and the effect produced by 0.2 mg. of atropine solution added to the 30 ml. bath is shown at the arrow. The depressing of the circular contractions of a ring of esophagus by 0.2 mg. atropine is shown in b. Tracings c and d show two types of normal tracings obtained from A. vaccaria esophagus and the effect of 0.4 mg. atropine added at arrow. sent when the rings of esophagus were used (Fig. lb), they are attributed to longitudinally oriented muscle in contrast to a primary circular muscle response. For some unknown reason the rapidity of contraction was increased when the esophagus was arranged in rings. The deep, in- termittent excursions are not as pronounced in A. vaccaria (Fig. 1 c) and may be absent (Fig. Id). The response of circular muscle, however, is consistently more rapid and more pronounced than is the case in A. calif ornica. In A. calif ornica even very small doses (0.05 mg. in some cases) of atropine produce an im- mediate and unfailing but transitory contraction of the longitudinal muscle (Fig. la). The esoph- agus of Aplysia vaccaria, however, does not react to atropine except in comparatively large doses of 0.4 ml. or more (in a 30 ml. bath), in which case the excursions of the circular muscle are depressed (Fig. Id). The circular muscle of A. californica, as demonstrated by the muscular ring preparations, is noticeably depressed by at- ropine in even smaller doses than that required to depress the circular type muscle in A. vaccaria (Fig. lb). This depression was not noticed in the usual tracings of A. californica muscle since it was obscured by the longitudinal contractions. Other muscle-active drugs with specific reac- tions of interest in themselves (which are to be reported later) did not produce divergent results between the two species. In an attempt to gain an insight into possi- ble reasons for the divergence of reaction be- tween the two species, histological sections were made and stained with eosin and hematoxylin. Strongly eosinophilic bundles of coarse, cylin- drical muscle strands were observed in the esophagus of A. californica (Fig. 2a, b) . These strands appear coarsely striated in some prep- arations (Fig. 2b) and coarsely granular in others, the differences possibly representing problems in killing and fixing. The nuclei are larger and more sparse than those of A. vaccaria ( Fig. 2c, d) and of the surrounding more conven- tional muscle of the present species. Not only are the nuclei three times as large but they tend to be arranged across the muscle strands, thus appearing rectangular in section. This muscle is Differences in Sea Hares — Winkler and Tilton 559 contrasted with the less eosinophilic, more un- dulatory muscle of the circular (Fig. 2a) and A. vaccaria type muscle (Fig. 2c, d) , which is very heavily nucleated. While strands of a type of muscle which seems morphologically to be somewhat similar to the former type of muscle appear sparsely in the crop, and to even a lesser extent in the area between the gizzard and "stomach” of the digestive tract in A. calif ornica, none of the eosinophilic muscles observed in other parts of the digestive tract of A. vaccaria possessed the distinctive tubular shape with the vacuolation, striation or granulation, and lack of undulatory characteristics. DISCUSSION AND CONCLUSIONS While it might be tempting to assume that the histological differences represent the imme- diate cause of the divergent pharmacological reactions observed, they are better interpreted as visible, easily demonstrable, morphological dif- ferences which parallel and are closely associated with neurohumoral and even biochemical dif- ferences which are themselves the underlying mechanistic causes. This is especially true since the effect of atropine on conventional tissue sys- tems is thought to be on the neuromuscular re- ceptors. The contractatory response observed in A. calif ornica appears to be the transitory result of stimulation, since the mechanical stimulation of washing will produce similar contractions. However, the sensitivity to this chemical stimu- lation is very great, being as low as 0.05 mg. atropine placed in the 30 ml. bath — a concen- tration of 1:6x10 s . Such a sensitivity represents a very delicately balanced chemical system. FIG. 2. A. calif ornica: a, cross section; b, longitudinal section. A. vaccaria: c, longitudinal section; d, cross section. The midsection of the esophagi were used for cross sections. 560 Moreover, the mechanism is of special interest since it is unique among known muscle prep- arations in that it is stimulated rather than de- pressed by atropine. The drug reaction and histological difference accentuates the divergence present within the genus. A. vaccaria belongs to a large and wide- spread subgenus ( Aplysia ), while A. calif ornica (subgenus Neaplysia ) is unique to the Calif or- nias. When it becomes possible to study com- paratively the members of the two other sub- genera 2 ( Varria and Pruvataplysia ), a taxonomic generalization may be possible. However, Bot- tazzi ( 1898) in his physiological study of the esophagus indicated that A. limacina (almost certainly A. fasciata of subgenus Varria ) was much more atonic than the close relative of A. vaccaria ( subgenus Aplysia ) with which he also worked. Since it is thought that A. calif ornica is an offshoot of this subgenus Varria , and since its esophagus is far from atonic by any inter- pretation, it may be that the musculature of the former has diverged considerably from its fore- runners. It would also seem quite probable that more comparative studies of the tissues of the species contained in other genera of animals would re- veal occasional divergences of at least equal magnitude. Such a possibility cannot safely be overlooked in experimental biology in any of its phases, and accentuates the need for accurate taxonomy before proceeding with studies which may be of a nontaxonomic nature. 2 Specimens of esophagus tissue from these subgen- era were kindly supplied from existing museum mate- rial on hand by Dr. N. B. Hales but, unfortunately, the conventional preservation method employed by the collectors was not adequate for any type of conclu- sions. PACIFIC SCIENCE, Vol. XV, October 1961 SUMMARY 1. The normal pattern of esophageal contrac- tions in Aplysia calif ornica consists of short rapid contractions attributed to circular muscle, interspersed at intervals by deep excursions at- tributed to the longitudinal musculature. These secondary contractions are less pronounced and less regular in A . vaccaria. 2. Atropine causes marked contraction of the esophagus in A. calif ornica, even in low con- centration, but in A. vaccaria it depresses circu- lar activity in higher concentrations. 3. The esophagus of A. calif ornica has longi- tudinal bundles of cylindrical muscle appearing coarsely striated or granular. These have not been found in A. vaccaria , and are suggested as being associated with the mechanistic cause for the differences in reaction noted. REFERENCES Bottazzi, Phillip. 1898. Contributions to the physiology of unstriated muscular tissue, Part IV. The action of electrical stimuli upon the oesophagus of Aplysia depilans and Aplysia limacina . Jour. Physiol. 22: 481-506. Cooper, j G. 1863. On new and rare mollusca inhabiting the coast of California. Proc. Calif. Acad. Sci. 3: 57. Hogben, Lancelot T. 1924. Studies on inter- nal secretion, III. The action of pituitary ex- tract and adrenaline on contractile tissues of certain invertebrata. J. Exp. Biol. 1 : 487-500. Straub, Walther. 1907. Zur chemischen Kinetic der Muscarinwirkung. Pfluger § Ar- chiv. 119. VON Brucke, E. Th. 1905. Zur Physiologic der Kropfmuskulatur von Aplysia. Pfliiger’s Ar chiv. 108. The Vegetation of Yanagi Islet, Truk, Caroline Islands Peter J. R. Hill 1 and Benjamin C. Stone 2 Yanagi Islet, located midway between Moen and Dublon in Truk lagoon (7° 24' N.; 151° 53' E.) is a small islet roughly 30 ft. high, about 800 ft. long and 300 ft. wide, with scarcely any strand but with volcanic rock pave- ment on all shores, and on the south side, numerous loose volcanic boulders. Much of the volcanic rock of which the islet is formed ap- pears to be a breccia resembling conglomerate. The reef surrounding the islet is rather large, with small storm-deposited rocks on the north- east sides. There is a smallboat pass on the west side and another pass, probably safe for small craft, on the north side. During World War II, the Japanese tunnelled through the islet from a cave on the south side to the north side; another Japanese cave on the south side does not run all the way through the islet. A small-gauge rail- road ran from the tunnel into the lagoon, for loading operations. The track is now rusty and twisted. The vegetation, despite the obvious prior use of the islet for military purposes, is relatively undisturbed in a few small areas. The Trukese do not live on the islet, but visit it occasionally for fish and shell-fish, and perhaps for mangoes. There are only a few coconut palms and a few of their seedlings. The vegetational zonation of the islet appears to comprise three areas : ( 1 ) a fringe area, more or less encircling the islet at sea level. (2) An open, rock-strewn extremity on the east, with a herbaceous cover, which slopes toward the east and terminates abruptly as a low cliff. This area is similar to the Tunnuk-Penniasene and Me- chitiu lava flows, but here there is no polygonal- patterned basalt. (3) An inner upland, with a developed tree-cover and fairly good soil. There 1 Pacific Islands Central School, Ponape. 2 Department of Botany, University of Hawaii. Pres- ent address: Department of Botany, U. S. National Museum, Smithsonian Institution, Washington, D. C. Manuscript received December 16, I960. is in fact a transitional area of low-crowned shrubs between zones 2 and 3; it is not known whether this is a natural attenuation in plant size, accompanied by increasing dominance of the herbaceous cover, which has persisted, or whether it is a result of human disturbance. However, the area contains a U. S. Navy Survey bench-mark on a concrete base (without further information ) . The most interesting features of the islet are the complete absence of coralline rock and soil, and the absence of most of the halophytic species of littoral plants which are so common elsewhere in Truk, a lack which is no doubt correlated with the nature of the substrate. The plants of Yanagi are by no means unique, and may be commonly found in Truk, and indeed through much of Micronesia. They are not re- stricted to volcanic soils, for some of them, at least, are known from the atolls nearby as well. But the strand species such as Scaevola Taccada, M esserschmidia argentea, Terminalia samoensis, and Soulamea amara are notably absent. Most of the species which occur on Yanagi also occur on the reef islets of Truk in coralline soil; but the lack of these littoral plants, which might rea- sonably be expected to grow there, is the curious and interesting feature of Yanagi Islet. VEGETATION ZONES: Since the islet is so small, altitudinal considerations are of no ac- count above sea level. The fringe area (1) seems to differ from the inner upland area ( 3 ) chiefly because of the exposure of the former and its hindrances to undisturbed plant growth, partly because of subjection to change from wind and waves, and to the concomitant salt spray, and partly because of the abundant volcanic boulders but thin, scanty soil. The grassland ( 2 ) and the intermediate area of shrubs may be the result of disturbance, or, again, may be the result of hard, thin soil with numerous embedded rocks, combined with exposure to wind and illumina- tion. Salt spray is apparently minimized by the 561 562 PACIFIC SCIENCE, Vol. XV, October 1961 reef, which lies many feet from the islet’s shore. No rainfall records are available; however, the inner forest with its better development of soil- and tree-cover no doubt accumulates moisture and retains it longer than the grassland and fringe area, with their porous volcanic boulders. 1, Fringe Area. A, Upper story: *Desmodium umbellatum (L.) DC., common, the trunk to 10 cm. diameter; *Colubrina asiatica (L. ) Brongn., common, shrubby; *Clerodendron inerme (L.) Gaertn., common scandent shrub; * Callicarpa candicans (Burm.) Hochr., scattered; *Barring- tonia asiatica (L.) Kurz, scarce, only small plants seen; * Guettarda speciosa L., only on south side; * Allophylus timorensis (DC.) BE, on south side; * Ficus sp., on south side; *Premna obtusifolia R. Br„, on south side only; *W edelia biflora (L.) DC. ex Wight, scandent shrub; *Morinda citrifolia L. — B, Vines: Abrus precatorius L.; *lpomoea digitata L.; *Passiflora foetida L.; Dios corea sp.; *lpomoea pes-caprae L., only one plant seen, in a small area; * Denis elliptica (Roxb. ) Bentham. — C, Ground cover: * Micro - sorum scolopendria (Burm.) Copel.; V ernonia cinerea (L. ) Less.; Echinochloa colonum (L.) Link; *'Tacca leontopetaloides (L.) Ktze., scat- tered and rare; *Digitaria pruriens ( Erin .) Biise. — D. Epiphytes: * Asplenium sp.; *Davallia solida (Forst. ) Sw.; *Dendrobium sp. 2, Grassland. A, Herbaceous cover, extreme eastern end of islet: *Digitaria pruriens (Trin.) Biise; P asp alum dilatatum (?); Chrysopogon aciculatus (Retz.) Trin.; Paspalum conjugatum Berg.; Cyperus sp.; *Fimbristylis atollensis St. John; Emilia sonchifolia DC.; P hyllanthus amarus Schum. & Thonn.; *Portulaca samoensis v. Poelln.; * Micro sorum scolopendria (Burm.) Copel., a dwarf form. — B, Transitional zone, with shrubs; merging with 3: Desmodium sp. (seedlings); Eledychium coronarium Koen. & Retz.; *Barringtonia asiatica (L.) Kurz, one seedling; *Tacca leontopetaloides (L. ) Ktze.; Abrus precatorius L.; *Premna obtusifolia R. Br.; *Passiflora foetida L.; Abelmoschus sp.; Capsicum frutescens L.; Cocos nucifera L.(a few young trees); *Morinda citrifolia L.; *Colub- rina asiatica (L. ) Brongn.; * Canavalia micro- carpa (DC.) Piper. * Species marked with an asterisk are indigenous. 3, Inner Upland Forest Area. — A, Trees: Mangifera indica L.; Musa balbisiana X acumi- nata; Carica Papaya L.; *? Eerminalia catappa L.; # Ficus tinctoria Forst.; Erythrina variegata L.; Cocos nucifera L. — B, Shrubs: *Colubrina asiatica (L. ) Brongn.; *Morinda citrifolia L.; *Glochidion ramiflorum Forst.; *Polyscias grandifolia Volkens. — C, Ground cover: *Op- lismenus compositus (L.) Beauv.; Paspalum con- jugatum Berg.; P. dilatatum ( ? ) . — D, Vines: * Canavalia microcarpa (DC.) Piper; Abrus pre- catorius L.; * Piper fragile Benth. — E, Epiphytes: *Davallia solida (Forst.) Sw.; * Microsorum scolopendria (Burm.) Copel. All plants mentioned are represented by speci- mens in possession of the senior author. animals PRESENT: The terrestrial animals, excluding insects and other invertebrates, were noted and identified by William R. Newman. Two species of lizard were fairly common, Emoya cyanura and E. boetgeri. The birds noted were Aplonis opacus angus, the Micronesian starling; Demigretta sacra sacra , the reef heron ( dark phase ) ; Myzomela cardinalis major, the cardinal honey-eater; Collocalis inquieta ruken- sis, the Carolines swiftlet; and Gygis alba, the fairy tern. SUMMARY The small volcanic islet in Truk Lagoon called Yanagi is remarkable for the lack of coralline development and the lack of halophytic vegeta- tion which are common elsewhere on Truk. The vegetation appears to comprise three zones, which are called the fringe area, the grassland area, and the upper inland forest. The summit is about 30 ft. above sea level, and the islet is sur- rounded by a reef. Though occupied during the war, the vegetation is now fairly recovered from disturbance. A list of the plants occurring in each vegetation zone, and a brief note on some animals observed, is given. REFERENCE Mayr, E. 1945. Birds of the South-West Pacific. Macmillan, N.Y. Revision of the Genus Pandanus Stickman, Part 6 New Pandanus Species from Queensland, Australia Harold St. John 1 This part of the author’s revision of the genus Pandanus is a collection of new species, all from the tropical northeastern corner of Queensland. They were all collected by L. J. Brass. Pandanus sphaericus sp. nov. (sect. Australibrassia) Figs. 23-24 NOM. VERN.: "maia.” DIAGNOSIS HOLOTYPI: Arbor 6-7 m. aka, trunco gracile "eo et ramis cum spinis brevibus conicis, ramis reflexis, radicibus fulturosis paucis 2-3 cm. longis,” foliis 1.5 m. longis 3 cm. latis subcoriaceis gladiatis in apice subulato longiter attenuatis, puncto 10 cm. ex apice 1.5 mm. lato, basi inermi, sed ex 5 cm. marginibus cum sub- ulato-serris pallidis 0.5-1 mm. longis 2-4 mm. distantibus, midnervio inermi, in regio mediali marginibus cum serris 0. 5-0.7 mm. longis 1.5-3 mm. distantibus apicibus subulatis, midnervio infra cum serris 0.2-0. 3 mm. longis 2-7 mm. distantibus adscendentibus, proxima apice mar- ginibus et midnervio infra cum serris 0.2-0. 3 mm. longis 2-3 mm. distantibus adscendentibus apicibus subulatis, pedunculo 26 cm. longo cur- vato cum bracteis foliosis paucis, syncarpio 12 cm. diametro subgloboso pendente circa 15 pha- langibus ferentibus, eis 4.3-4.6 cm. longis 3.9- 4.3 cm. latis 3. 1-3.2 cm. crassis suborbicularibus sed subcompressis 4-5-angulosis in sicco pallide brunneis exlucidis laevibus, lateribus valde cur- vatis parte 2 A supera libera, apice truncato, su- turis lateralibus nullis, sinibus centralibus api- calibus 4-5 mm. profundis anguste V-formatis, carpelis 9-11, apicibus anguste pyramidalibus sed cum regione proximali truncato et dimidio proximali duro brunneo lucido ad stigmatam dirigito, stigmatibus 1.5 mm. longis ellipticis ad ovatis sulcatis in apice latere proximo, suturis proximis Vi-Va ad fondam extentis, endocarpio 1 Botanist, B. P. Bishop Museum, Honolulu, Hawaii. Manuscript received August 15, I960. submediali 2.5-3 cm. longo osseoso pallide muris lateralibus 4-5 mm. crassis, seminibus 13-17 mm. longis 4 mm. diametro ellipsoidis, mesocarpio supero caverna unica 1 cm. longa cum fibris multis traversis formanto, mesocarpio basali fibroso et carnoso. DIAGNOSIS OF holotype: Tree 6-7 m. tall; trunk slender, it "and the branches studded with short conical thorns; branches down-turned; prop roots few, 2-3 dm. long”; leaves 1.5 rn. long, 3 cm. wide, subcoriaceous, rapier-like and tapering from the base to the long subulate tip which 10 cm. down is only 1.5 mm. wide, the base unarmed, but beginning 5 cm. from the base the margins with subulate serrations, pale, 0.5-1 mm. long, 2-4 mm. apart; the nearby mid- rib unarmed; at the midsection the margins with subulate-tipped serrations 0. 5-0.7 mm. long, 1.5-3 mm. apart; the midrib below with ascend- ing serrations 0.2-0. 3 mm. long, 2-7 mm. apart; near the tip the margins and midrib below with subulate-tipped ascending serrations 0.2-0. 3 mm. long, 2-3 mm. apart; peduncle 26 cm. long, curving, with a few leafy bracts; syncarp pend- ent, 12 cm. in diameter, subglobose, of about 15 phalanges, these 4. 3-4.6 cm. long, 3-9— 4.3 cm. wide, 3- 1-3.2 cm. thick, suborbicular, but some- what compressed, 4-5 -angled, when dried light brown, dull, smooth, the 4-5 sides strongly curv- ing, upper 2 A free; the apex truncate, lateral sutures none; central apical sinuses 4-5 mm. deep, narrow V-shaped; carpels 9-11, the apices narrow pyramidal with a proximal truncation and its inner half leading to the stigma hard, brown, shining; stigmas 1.5 mm. long, elliptic to ovate, creased, at tip of steep proximal face of apex; below it the proximal suture running Vj-Va way to valley bottom; endocarp subme- dian, 2.5-3 cm. long, bony, pale throughout, the lateral margins 4-5 mm. thick; seeds 13-17 mm. long, 4 mm. in diameter, ellipsoid; upper mesocarp a single cavity 1 cm. long, with many strong longitudinal fibers; basal mesocarp fi- brous and fleshy. 563 •uoT o -U4D 8 Fig. 23. Pandanus sphaericus, from holotype. a. Habit, X 1/70; b, syncarp, X Va\ o, phalange, lateral view, X 1; d, phalange, longitudinal median section, X 1; e, phalange, apical view, X 1; /, carpel apex and stigma, oblique view, X 4. Page 50: Revision of Pandanus, 6. Queensland — St. John 565 HOLOTYPUS: Australia, Queensland, Cape York Peninsula, abundant locally in rain forest behind the coastal sand dunes, 20 m. alt., May 9, 1948, L J. Brass 18,715 (bri). DISCUSSION: P. sphaericus is a member of the section Australibrassia and there also is its near- est relative, P. cochleatus St. John, which is dis- tinguished by having the phalanges pyriform; the carpels 12-16; stigma below an overhanging cochlea te rim; prop roots absent; and leaves having near the base marginal teeth 1.3-2 mm. long. It also occurs on the Cape York Peninsula. P. sphaericus differs by having the phalanges suborbicular; carpel 9-11; stigma on proximal face of carpel apex without an overhang; prop roots few; and leaves near base with marginal teeth 0.5-1 mm. long. The new epithet is the Latin adjective, sphaer- icus, spherical, in reference to the shape of the phalange and of the syncarp. Pandanus ferrimontanus sp. nov. (sect. Pandanus ) Figs. 25-26 DIAGNOSIS HOLOTYPI: Arbor 5-7 m. alta 10- 12 cm. diametro in corona laxa ramosa, radici- bus fulturosis nullis, foliis 133-160 cm. longis 5.8 cm. lads subcoriaceis glaucis supra infraque gladiformatis gradatim ex basi in apice subulato longe acuminato diminuentibus, in puncto 10 cm. ex apice 3 mm. latis, marginibus ex basi per 29 cm. inermibus turn cum aculeis 2-3 mm. longis 6-21 mm. distantibus subulatis adscend- entibus apice brunneo, midnervio inermi, in sec- tio mediali margine unica cum aculeis 1-2 mm. longis 2-6 mm. distantibus adpresse adscend- entibus subulatis cum basi crassa, margine al- tera cum aculeis paucis simulantibus sed minori- bus vel in partibus inermibus, midnervio infra cum aculeis paucis remotis minutis adscenden- tibus, in regione apicali marginibus et mid- nervio infra cum aculeis 0.2-0. 3 mm. longis 3-5 mm. distantibus subulatis validis adscenden- tibus, pedunculo 40 cm. longo excedenti obscure trigono folioso, syncarpio 16 cm. longo 14 cm. diametro terminali solitario ovoideo-globoso cum circa 45 phalangibus, eis 4. 5-4.7 cm. longis 4.2 cm. latis 3. 4-3 =5 cm. crassis late obovoideis subcompressis, apice rotundato, parte V 3 supera libera, lateribus cum 5-6 angulis majoribus et carpela quoque cum 2-4 angulis et valleculis angustis et turn lateribus phalangiorum proxi- morum interobseratis, in sicco lateribus papil- losis et parte supera rubro-brunnea, suturis lat- eralibus angustis in dimidia supera, sinibus api- calibus centralibus 3. 5-4.5 mm. profundis fondis angustis sinuosis et lateribus interobseratis, car- pelis 8-12 plerumque 9-10, apicibus subaequali- bus semiorbicularibus pallide brunneis sed lat- eribus brunneo-rimosis, regione obliqua brunnea concava ex stigmate distalia, stigmatibus 1—1.5 mm. longis ellipticis vel truncatis obscure brun- FlG. 24. Pandanus sphaericus, from holotype, lower side of leaf, X 1- a, Near base; b, middle; c, apex. 16 cm. FIG. 25. Pandanus ferrimontanus , from holotype. a, Syncarp, X Va\ b, phalange, lateral view, X 1; c, phalange, longitudinal median section, X 1; ^ phalange, apical view, X 1; ^ carpel apex with stigma, apical view, X 4. Page 52: Revision of Pandanus, 6. Queensland — S t. John 567 neis obliquis sulcatis centripetalibus, sinu prox- imo profundo Vz vel omnino distancia ad fon- dam extento, endocarpio 22-25 mm. longo su- pramediali osseoso stramineo lateribus laterali- bus 8-10 mm. crassis, interiore cavernarum seminarum brunneo, seminibus 18-20 mm. longis 3-5 mm. diametro irregulariter ellipsoi- deis, mesocarpio apicali in carpella quoque cavernam cum fibris paucis et membranis stra- mineis transversis medullosis formanti, meso- carpio basali fibroso et carnoso. DIAGNOSIS OF holotype: Tree 5-7 m. tall, 10-12 cm. in diameter, branched into an open crown; prop roots none; leaves 133-160 cm. long, 5.8 cm. wide, subcoriaceous, glaucous above and below, sword-like, gradually tapering from near the base to the long acuminate, subulate tip which 10 cm. down is 3 mm. wide, the basal margins unarmed for 29 cm., then with prickles 2-3 mm. long, 6-2 1 mm. apart, subulate, ascend- ing, brown-tipped; the nearby midrib unarmed; at the midsection one margin is with prickles 1-2 mm. long, 2-6 mm. apart, broad based subulate, appressed ascending, the other margin with a few similar but smaller prickles or in- termittently with none; the midrib below with a few remote, minute, ascending prickles; near the apex the margins and midrib below with prickles 0.2-0. 3 mm. long, 3-5 mm. apart, stout subulate, ascending; peduncle more than 40 cm. long, obscurely trigonous, leafy bracted; syncarp solitary, terminal, 16 cm. long, 14 cm. in diam- eter, ovoid-globose, the color unknown, of about 45 phalanges, these 4. 5-4.7 cm. long, 4.2 cm. wide, 3. 4-3. 5 cm. thick, broadly obovoid, some- what flattened, the apex rounded, upper l A free, the sides with 5-6 major angles and each carpel with 2-4 sharp secondary ridges and valleys so that adjacent phalanges fit as if mortised, when dry the sides papillose and the upper part red- dish brown; lateral sutures narrow but extend- ing half way down; central apical sinuses 3-5- 4.5 mm. deep, the bottoms narrow and sinuous mortised; carpels 8-12, mostly 9-10, the apices subequal, semiorbicular, the sides with dark brown cracks, elsewhere pale brownish, with an oblique distal, brown concavity leading to the stigma; stigma 1-1.5 mm. long, elliptic or trun- cate elliptic, creased, dark brown, oblique, cen- tripetal; proximal sinus deep, running Vz way r— — r— — i o 1 cm. FIG. 26. Pandanus ferrimontanus, from holotype, lower side of leaf, a, Near base, XI \ b, middle, X 1; d, apex, X 1; c, marginal serrations at middle, X 4. Fig. 27. P andanus oblatus, from holotype. a, Syncarp, X 2/5; b, phalange, lateral view, X 1; ^ phalange, longitudinal median section, X 1; d, phalange, apical view, X 1; e, apex of carpel and stigma, apical view, X 4. Page 54: Revision of Pandanus, 6. Queensland — St. John T T 569 to all the way to the valley bottom; endocarp supramedian 22-25 mm. long, bony, stramine- ous except for the dark brown, shining lining of the seed cavities, lateral walls 8-10 mm. thick; seeds 18-20 mm. long, 3-5 mm. in diameter, irregularly ellipsoid; apical mesocarp in each carpel forming a cavern with a few strong, lon- gitudinal fibers, and transverse, stramineous, medullary membranes; basal mesocarp fibrous and fleshy. HOLOTYPUS: Australia, Queensland, Cape York Peninsula, Iron Range, scattered or gre- garious in groves in savanna forest, 20 m. alt., June 24, 1948, L. ]. Brass 19,312 (bri). DISCUSSION: The flowers were not seen by the collector. This new species is a member of the section Pandanus , to which most of the Australian species belong. None of the previ- ously described species are close relatives, so no contrast is needed here. The new specific epithet is formed from the Latin, ferrum, iron; montanus, of the mountain, in reference to the name of the type locality. Pandanus oblatus sp. nov. (sect. Pandanus ) Figs. 27-28 DIAGNOSIS HOLOTYPI: Arbor 12-14 m. alta, trunco "crasso, corona ramosa, radicibus fultu- rosis multis longis spinosis,” foliis 1.6 m. longis 9.2 cm. latis crassiter coriaceis infra subglaucis, in section valde M-formatis, gladiformatis gra- datim in apice subulato trigono valide diminu- entibus, in regione 10 cm. ex apice 5-6 mm. latis, basi amplexicauli et integri sed ex 7 cm. marginibus cum aculeis 2—3 mm. longis 3-6 mm. distantibus subulatis validis luteis subad- scendentibus, midnervio infra inermi, in regione mediali marginibus cum aculeis 1.5-3 mm. lon- gis 3-5 mm. distantibus subulatis validis arcuato- adscendentibus, midnervio infra salienti acuito et cum aculeis 2.5-3 mm. longis 10-20 mm. distantibus simulantibus adscendentibus; in re- gione apicali marginibus et midnervio infra subulato-serratis serris 0.8- 1.2 mm. longis 2-3 mm. distantibus, pedunculis 15 vel plus cm. longis obtuse trigonatis folioso-bracteatis, syn- carpio 18 cm. longo 13 cm. diametro cylindrico sed apicibus ambis rotundatis cum circa 92 pha- langibus, eis 4.7-5. 1 cm. longis 2.9-3. 5 cm. latis 2. 4-2 .7 cm. crassis obovoideis in sicco rubro- o 1 s cm. Fig. 28. Pandanus oblatus, from holotype, lower side of leaf, X !■ ^ Near base; b, middle; c, apex. 570 PACIFIC SCIENCE, Vol. XV, October 1961 brunneis 4-6-angulosis, lateribus papillosis sub- curvatis vel paene planatis parte l A supera li- bera, apice rotundato, suturis lateralibus nullis, sinibus apicalibus centralibus 2-4 mm. profun- dis late V-formatis, carpelis 5-12 sed in pha- langibus dimidiae superae plerumque 6—7 et in eis inferis plerumque 10—11, apicibus centrali- bus oblato-semiorbicularibus, eis marginalibus oblato-pyramidalibus, stigmatibus 1-1.8 mm. longis suborbicularibus ad ovalibus sulcatis ob- scuris apicalibus centripetalibus plerumque obli- quis, sinu proximali lato Vl distancia ad fondam extento, endocarpio mediali osseoso obscure brunneo excepta marginibus pallidis, lateribus lateralibus 4-5 mm. crassis, seminibus 16 mm. longis 4 mm. diametro ellipsoideis, mesocarpio apicali in carpela quoque caverna cum mem- branis pallidis medullosis formanti, mesocarpio basali fibroso et carnoso. diagnosis of HOLOTYPE: Tree 12-14 m. tall; trunk "thick, crown branched; stilt roots many, long, and prickly”; leaves 1.6 m. long, 9.2 cm. wide, thick coriaceous, below slightly glaucous, in section sharply M-shaped, sword shaped, gradually narrowed to the stout, trigo- nous, subulate apex, this 10 cm. down is 5-6 mm. wide, at very base amplexicaul and entire, but from 7 cm. up the margins with prickles 2-3 mm. long, 3-6 mm. apart, stout subulate, yellow, slightly ascending, the nearby midrib unarmed below; at midsection the margins with prickles 1.5-3 mm. long, 3-5 mm. apart, stout subulate, arcuate ascending; the midrib below sharp, salient, with prickles 2.5-3 mm. long, 10-20 mm. apart, similar, ascending; near the apex the margins and midrib below subulate- serrate, the teeth 0.8-1. 2 mm. long, 2-3 mm. apart; peduncle 15 or more cm. long, obtusely trigonous, leafy bracted; syncarp 18 cm. long, 13 cm. in diameter, cylindric but with the ends rounded, of about 92 phalanges, these 4.7-5. 1 cm. long, 2.9-3. 5 cm. wide, 2.4-27 cm. thick, obovoid, when dried reddish brown, 4-6-angled, the sides minutely papillose, gently curving or nearly plane, upper Va free, the apex rounded; lateral sutures none; central apical sinuses 2-4 mm. deep, wide V-shaped; carpels 5—12, those of the upper half mostly 6-7, those of the lower mostly 10-11, the central apices oblate-semior- bicular, the marginal ones oblate-pyramidal; stigmas 1-1.8 mm. long, suborbicular to oval, creased, dark, apical, centripetal, mostly oblique; proximal sinus wide, running l /i way to the val- ley bottom; endocarp median, bony, dark brown except for the pale outer margins, the lateral walls 4-5 mm. thick; seeds 16 mm. long, 4 mm. in diameter, ellipsoid; upper mesocarp in each carpel forming a cavern nearly free of fibers but with pale medullary transverse membranes; basal mesocarp fibrous and fleshy. HOLOTYPUS: Australia, Queensland, Cape York Peninsula, Iron Range, occasionally in gul- lies in rain forest, 20 m. alt., June 24, 1948, L. ]. Brass 19,309 (bri). DISCUSSION: The new species is a member of the section Pandanus. In this the closest relative seems to be P. odoratissimus L. f. var. novo- guineensis ( Martelli ) St. John, which as the phalanges 5.7-6 cm. long, 3-3.5 cm. wide, the lateral sinuses distinct in the upper half; apical central sinuses 5-6 mm. deep; carpel apices semiorbicular; and the endocarp supramedian. On the other hand, P. oblatus has the phalanges 4.7-5. 1 cm. long, 2. 9-3. 5 cm. wide; lateral si- nuses none; apical central sinuses 2-4 mm. deep; apices of the marginal carpels oblate- pyramidal; and the endocarp submedian. The epithet is the Latin participle, oblatus, given the modern scientific meaning, depressed. This refers to the oblate-pyramidal carpel apices. Pandanus somersetensis sp. nov. (sect. Pan- danus ) Figs. 29-30 DIAGNOSIS HOLOTYPI: Arbor 5-8 m. alta 20- 30 cm. diametro supra media ramosa, cortice armato cum muriculis, radicibus fulturosis 6-15 cm. longis pluribus, foliis 93 cm. longis 6.8-7.2 cm. latis coriaceis glaucis gladiformibus sed sen- sing diminentibus ad apicem acutum in puncto 10 cm. ex apice 7 mm. latis in base marginibus integris sed ex loco 7-10 cm. e base in uno latere cum aculeis 1.5-2. 3 mm. longis 3-8 mm. distantibus crassiter subulatis stramineis ad- scendentibus sed in latere altero aculeis 0.5-1 mm. longis 3-15 mm. distantibus midnervio exarmato fere ubique, foliis in section mediale integris et in parte apicale marginibus integris sed nervo mediale infra cum aculeis 0.3 mm. longis paucis remotis, pedunculis ca. 25 cm. Fig. 29- Pandanus somersetensis, from holotype. a, Habit, X 1/100; b, syncarp, X 3/16; c, phalange, lateral view, X 1; d, phalange, apical view, X 1; e, apex of carpel and stigma, apical view, X 4. 1 cm. o 2 cm. o to cm. 572 PACIFIC SCIENCE, Vol. XV, October 1961 6 1 5 cm. C b Fig. 30. Pandanus somersetensis, from holotype, lower side of leaf, XI ■ a > Near base; b, middle; c, apex. longis foliosis, syncarpiis terminalibus singulari- bus maximis eorum 22 cm. longis 13 cm. diam- etro late ellipsoideis subtriangularibus cum ca. 70 phalangibus eis 5-5.5 cm. longis 4-5.1 cm. latis 3. 2-4. 5 cm. crassis anguste obpyramidali- bus 4-6-angulosis in sicco brunneis laevibus et sublucidis lateribus paene curvatis vel planis parte Va supera libera, apice lato depresse con- vexo, suturis lateralibus raris sed paucis et solo in parte quarta supera evidentis, suturis apicali- bus centralibus 0.5-1 mm. profundis sed ple- rumque tarn vadosis quam paene imperceptis sunt et suturis solum tessellatis, carpelis 7-9, apicibus inequalibus quod carpelis lateralibus in latere exteriore distentis sunt sed apicibus omnibus veris brunneis subaequalibus et sub- planis sed eis centralibus minime depresso- conicis, stigmatibus 1.5-3 mm. longis ovato- deltoideis ad ellipticis horizontalibus centripe- talibus minime elevatis, sutura proxima ad dimi- diam intervallam ad basem producta, endocar- pio mediale osseoso subalbo in latere 2-3 mm. crasso, seminibus 15-16 mm. longis 5 mm. diametro oblique ellipsoideis, mesocarpio su- pero cavernoso et cum fibris longitudinalibus crassis et membranis albis, mesocarpio infero fibroso et carnoso sed minimo. DIAGNOSIS OF holotype: Tree 5-8 m. tall, 20-30 cm. in diameter, branching in 2’s or 3’s, beginning halfway up, the crown semiorbicular; bark armed with scattered, short, sharp bosses; prop roots several, 6-15 cm. long; leaves 93 cm. long, 6. 8-7. 2 cm. wide, coriaceous, glaucous, sword-like but tapering gradually from the base to the acute tip, this 10 cm. down 7 mm. wide, at the very base the margins entire, but begin- ning 7-10.5 cm. above it having on one side ascending prickles 1.5-2 .2 mm. long, 3-8 mm. apart, stout subulate, stramineous, but on the other border 0.5-1 mm. long, 3-15 mm. apart, the midrib smooth almost throughout; at the middle the margins entire; near the tip the margins entire and only the midrib below with a few remote ascending prickles 0.3 mm. long; Page 58: Revision of Pandanus, 6. Queensland — S t. John 573 peduncles about 25 cm. long, leafy bracted; syn- carps terminal, single, the largest 22 cm. long, 13 cm. in diameter, broadly ellipsoid, obscurely 3 - angled, the color unknown, of about 70 pha- langes; phalanges 5-5.5 cm. long, 4-5.1 cm. wide, 3. 2-4. 5 cm. thick, narrowly obpyramidal, 4- 6-angled, when dried light brown, smooth and somewhat shiny, the sides gently curving or plane, free in upper Va , the apex broad, low con- vex, lateral carpel sutures mostly none, but a few perceptible down Va way, the central apical sutures 0.5-1 mm. deep, but mostly so shallow as to be almost imperceptible and the sutures merely making a tessellate pattern; carpels 7-9, the apices somewhat unequal as the lateral ones have considerable expansion on the outer side, but the brown colored, actual apex is subequal on all and is almost flat, though the central ones have a perceptible depressed conic elevation; stigmas 1.5-3 mm. long, ovate-deltoid to ellip- tic, horizontal, centripetal, slightly raised, below it the proximal suture on the carpel apex from very short to Vi as long as the distance to the inter carpellary suture; endocarp median, bony, whitish, that surrounding the outer seeds 2-3 mm. thick; seeds 15—16 mm. long, 5 mm. in diameter, obliquely ellipsoid; upper mesocarp cavernous, with very stout longitudinal fibers and delicate, white, medullary membranes; lower mesocarp fibrous and fleshy, but of small extent. HOLOTYPUS: Australia, Queensland, Cape York Peninsula, Newcastle Bay, 2Vz mi. S. of Somerset, commonly on rocky coastline and be- side small freshwater streams in coastal sand dune area, May 13, 1948, L. J. Brass 18,792 (BRI). This new species certainly belongs in the sec- tion Pandanus, even though the stigmas are sub- horizontal. None of the published species in that section are so similar that it would be of value to state the contrasting differences. The new epithet is geographic, indicating the type locality of the species. Pandanus truncatus sp. nov. (sect. Pandanus) Figs. 31-32 DIAGNOSIS HOLOTYPUS: Arbor 6-8 m. alta, laxe ramosa, foliis 1.43 m. longis 6 cm. latis in base coriaceis ligulatis infra minus glaucis ex base gradatim diminuentibus ad apicem subula- tum in puncto 10 cm. ex apice 3 mm. latis in base exarmatis sed ex puncto 15 cm. ex base marginibus cum aculeis 1.5-2 mm. longis 3—10 mm. distantibus aciculatis rectis adscendentibus in apice brunneis, in parte mediale marginibus et midnervio exarmatis, ad apicem marginibus midnervioque infra cum serris remotis 0.2-0. 3 mm. longis, syncarpio terminale singulare sub- pendente, pedunculo 20 cm. longo triangulare bracteato, syncarpio 18 cm. longo 13 cm. diam- etro late ellipsoideo cum ca. 56 phalangibus 4.5-5 cm. longis 3. 1-4.1 cm. latis 2.9-3. 3 cm. crassis late obovoideis, apice depresse convexo, p 3 parte supera libera, lateribus cum 6-8 an- gulis majoribus et 2-3 -plo angulis et vallibus minoribus in sicco pallide brunneis laevibus et lucidis, suturis lateralibus fortibus Vi vel omnia distancia extensis, sinibus centralibus apicalibus 3-4 mm. profundis angustis sinuatisque, carpel- lis 7-11, apicibus subaequalibus pyramidali- semiorbicularibus omnibus marginalibus et paucis centralibus cum area plana vel minime concava obscure brunnea terminale, stigmatibus centripetalibus 1-1.5 m. longis suborbicularibus vel obcordatis valde inclinatis, sutura proximali et stigmate dimidia vel omnia distancia ad vadum extensa, endocarpio mediale osseoso albo solum in parte supera extensa sed in lateribus et parte infera diminuenda in muris 0.5 mm. crassis, seminibus 23-25 mm:* longis 4-6 mm. diametro ellipsoideis, mesocarpio supero pluri- cavernoso cum fibris validis et membranis me- dullosis pallidis, mesocarpio infero fibroso et carnoso sed in base toto carnoso. DIAGNOSIS OF HQLOTYPE: Tree 6-8 m. tall, branched into a very open crown; leaves 1.43 m. long, 6 cm. wide near the base, coriaceous, ligulate, somewhat glaucous below, tapering gradually from the very base up to the subulate tip which 10 cm. down is 3 mm. wide, the base unarmed, but beginning 15 cm. up the margins with spines 1.5-2 mm. long, 3-10 mm. apart, straight acicular, ascending, brown-tipped; in the median region the margins and midrib un- armed; near the apex the margins and midrib below with a few remote serrations 0.2-0. 3 mm. long; syncarp terminal, single, almost pendent, borne on a triangular, leafy bracted peduncle about 20 cm. long, the syncarp 18 cm. long, 13 cm. in diameter, broadly ellipsoid, of about 56 phalanges, these 4.5-5 cm. long, 3. 1-4.1 cm. wide, 2.9-33 cm. thick, broadly obovoid, the Fig. 31. Pandanus truncatus, from holotype. a, Habit, X 1/100; h, syncarp, X 5/18; c, syncarp, longi- tudinal section, X 5/18; d, phalange, lateral view, X 1- - phalange, longitudinal median section, X 1; /• pha- lange, apical view, X 1; ^ apex of carpel and stigma, oblique view, X 4. Smm. Page 60: Revision of Pandanus, 6. Queensland — -S t. John 575 r 1 ' 1 1 i o 5 cm. a Fig. 32. Pandanus truncatus, from holotype, lower side of leaf, §| l. a, Near base; b, middle; c, apex. apex low convex, upper l A free, the sides with 6- 8 major angles and with 2-3 -times that num- ber of secondary sharp ridges, so that the adja- cent phalanges fit as if mortised, when dried light brown, smooth and shiny, lateral carpel sutures well marked Vi way or nearly all the way down, the central apical sinuses 3-4 mm. deep, the bottoms narrow and sinuous; carpels 7- 11, the apices subequal pyramidal-semiorbic- ular, all the marginal and some of the central ones with a dark brown flat or slightly concave surface below the proximal edge of which is the steeply oblique, centripetal stigma 1-1.5 m. long, suborbicular to obcordate, and below it the proximal crease running Vz way or all the way to the bottom; endocarp median, bony, white, extensive only at the upper end, diminishing on the sides and lower part to thin sheets only 0.5 mm. thick; seeds 23-25 mm. long, 4-6 mm. in diameter, ellipsoid; upper mesocarp of several caverns with strong longitudinal fibers and pale medullary membranes; lower mesocarp fibrous and fleshy, but the actual base almost wholly fleshy. HOLOTYPUS: Australia, Queensland, Cape York Peninsula, Lockerbie, 10 mi. WSW of Somerset, common and in places forming small groves on sandy soil in savanna-forest, 30 m. alt., 1948, L. /. Brass 18,552 (bri). DISCUSSION: The closest relative, P. Brookei Martelli, has the leaves near the base with mar- ginal teeth 2-3 mm. long; syncarp cylindric, 43 X 21.5 cm., with about 66 phalanges the sides of which are plane or gently curving; apices of the marginal carpels broad concave on a broad pyramidal base; central apical sinuses 4-6 mm. deep, the lines straight or gently curved; stigmas horizontal, 2-3 mm. long; and the endocarp 2 cm. long. On the other hand, P. truncatus has the leaves with marginal teeth near the base with teeth 1.5-2 mm. long; pha- lange sides sharply ridged and mortised with neighbors; syncarp broadly ellipsoid, 18 X 13 cm., with about 56 phalanges; apices of margi- nal carpels with oblique, truncate planes exte- rior of the stigmas; central apical sinuses 3-4 mm. deep, the lines sinuous; and the stigmas steeply inclined, 1-1.5 mm. long. The new epithet is the Latin participle, trun- catus, cut off, or flat ended, in allusion to the shape of the carpel apices. Revision of the Genus Pandanus Stickman, Part 7 New Species from Borneo, Papua, and the Solomon Islands Harold St. John 1 This seventh part of the series contains de- scriptions of miscellaneous new species from is- lands in the tropical Pacific. Pandanus Andersonii sp. nov. (sect. Acro- stigma ) Figs. 33-34 NOM. VERN.: "surong irit.” DIAGNOSIS HOLOTYPI: Pedunculo ad 10 cm. longo 2.5 cm. diametro obtuse trigono ebrac- teato erecto, syncarpiis 1-3 spicatis in inflores- centia erecta 15 cm. longa evidenta, syncarpiis 11.5-14 cm. longis 13.5-14 cm. diametro late ellipsoideis vel subglobosis obtuse trigonatis cum drupis numerosissimis, drupis 35-40 mm. longis 6-7 mm. latis 5-6 mm. crassis fusiformi- bus 5-6-angulatis corpore 18-21 mm. longo oblongo-ellipsoideo pileo quam corporem lat- iore 13-17 mm. longo in stylo sensim dimi- nuenti, parte principali brunnea ex squamis late deltoideis lucidis dense congregatis erectis ex axile perpendicularibus, drupis in fasciculis dehiscentibus, et pileis in laminis grandibus dehiscentibus, stylo 5-7 mm. longo subulato ar- cuato stigmatam distalem ferento, stigmate 4-5 mm. longo sublineari in fissura, endocarpio in Va parte infera albo cartilagineo lateribus 0.2 mm. crassis, mesocarpio apicali cavernoso ellip- soideo 8-9 mm. longo, mesocarpio basali 5 mm. longo fibroso et carnoso. DESCRIPTION OF ALL SPECIMENS EXAMINED: "Roots ramify below water table; rhizome at surface level.” Stem 20—25 cm. long, 1.2 cm. in diameter, assurgent, forking and forming dense thickets; leaves 2.6-6.8 m. long, 2.6-2.8 cm. wide, coriaceous, ligulate, 3 -nerved and plicate, AA-shaped in cross section, the under side slightly paler, abruptly narrowed to the heavy subulate tip which 10 cm. down is 13 mm. wide, the base entire and amplexicaul, but beginning 9 1 Botanist, B. P. Bishop Museum, Honolulu, Hawaii. Manuscript received September 13, I960. cm. up the margins with stout serrations 2-2.5 mm. long, 5-13 mm. apart, pale; the midrib below beginning 7-8 cm. up with reflexed prickles 4-6 mm. long, 7-17 mm. apart, stout arcuate subulate, becoming brown; at midsec- tion the margins with serrations 1-1.5 mm. long, 5-10 mm. apart, pale below, the tips brownish; the midrib unarmed; near the tip the margins and midrib below with serrations 0.5-1 mm. long, the marginal 2-3 mm. apart, those of the midrib 3-8 mm. apart; on the upper side the two secondary ribs with serrations about 0.5 mm. long, 9-30 mm. apart; pistillate inflores- cence ascending, not concealed; peduncle up to 10 cm. long, and 2.5 cm. in diameter, obtusely trigonous, bractless, erect; heads solitary or 2-3 in a spike 15 cm. long; syncarps 11.5-14 cm. long, 13.5-14 cm. in diameter, broadly ellipsoid to subglobose, obtusely trigonous, bearing very numerous drupes (several hundred); drupes 35- 40 mm. long. 6-7 mm. wide, 5-6 mm. thick, fusiform, 5-6-angled, the body 18-21 mm. long, oblong-ellipsoid; the pileus the widest part, lance conic, 13-17 mm. long, narrowing grad- ually into the style, the body appearing rough and brown, from the dense outgrowth of closely packed, brown, shining (perhaps viscid) broad triangular scales, erect from the surface and all perpendicular to the drupe axis; drupes shed- ding in bundles and the pileuses cohering, early shedding in great sheets; style 5-7 mm. long, subulate, arcuate, glabrous, bearing a distal stigma 4-5 mm. long, almost linear in an open crack on the curved style; endocarp in lower V4, white, cartilaginous, the walls 0.2 mm. thick; apical mesocarp an empty cavern, 8-9 mm. long, ellipsoid; basal mesocarp 5 mm. long, fibrous and fleshy. HOLOTYPUS: Sarawak, Lawas District, Kayan- geran Forest Reserve, in "Sempilor” ( Dacry - dium Beccarii var. subelatum) forest, 10 ft. alt., Nov. I960, /. A. R. Anderson (fruit only) (SARF). 576 o n 5 cm, ~r - — — r r r O 5 10 cm. L _j i i i __L_ l_ i _j „ i _ I FIG. 33. Pandanus Andersonii St. John, from holotype. a, Habit, X 1/40; b, drupe, lateral view, X 1; c, d, drupes after shedding pileus, lateral view, X 1; ft drupe, longitudinal median section, X 1; f, g, pileus, lateral view, X 1; b, pileus and stigma, lateral view, X 4; i, style and stigma of basal drupe, lateral view, X 4; j, scales of pileus, basal view, X 100; k, base of leaf, lower side, X 1; l, leaf middle, lower side, X 1; m, leaf, slightly beyond the middle, upper side, X 1;«» leaf apex, lower side, X 1- 578 PACIFIC SCIENCE, Vol. XV, October 1961 SPECIMENS EXAMINED: Sarawak, Loba Ka- bang Protected Forest, abundant, ground flora in "alan” ( Shorea albida) type forest, found in wetter areas and associated with "bakong” ( Hanguana malayana) , May 1 6 , 1954, J. A. R. Anderson S.2,81 5 ( sarf ) . DISCUSSION: P. Andersonii is a member of the section Aero stigma and there one finds its closest relative, P. brevifolius Martelli of Bor- neo. This has the syncarp solitary, 5 cm. long, 4 cm. in diameter; drupes 15-16 mm. long, 3 mm. wide; pileus abruptly contracted to the style; and the endocarp in the lower V 3 . On the other hand, P. Andersonii St. John has the syncarps Fig. 34. Pandanus Andersonii St. John, in foreground, in peat swamp forest of Shorea albida, Baram River, Sarawak. (Photo by J. A. R. Anderson.) Page 64: Revision of Pandanus, 7. Papua and the Solomons — S t. John 579 1- 3 in a spike and 11.5-14 cm. long, 13.5-14 cm. in diameter; drupes 35-40 mm. long, 6-7 mm. wide; pileus tapering into the style; and the endocarp in the lower Va. The new species is named in honor of its col- lector, Mr. J. A. R. Anderson, Forest Research Officer of the Sarawak government. Pandanus fruticosus sp. nov. (sect. Acro- stigma) Figs. 35-36 DIAGNOSIS HOLOTYPI: Frutex 15.5 m. altus ramosus, ramulis foliosis 6-9 cm. diametro, cor- tice rubrescenti-brunneo lucido cum cicatricibus pallidis cinctis, foliis 51-58 cm. longis 8-10 mm. latis subcoriaceis ligulatis M-formatis longe diminuentibus in apice subulato 10 cm. ex apice 3.5 mm. lato, marginibus in basi integris sed ex 4 cm. marginibus cum dentibus validis subu- latis 0.5-1. 3 mm. longis 1-3 mm. distantibus nigris in 45° adscendentibus, midnervio exar- mato, marginibus in sectio mediale cum serris nigris 0.5 mm. longis 1-2 mm. distantibus, mid- nervio infra cum serris nigris 2-6 mm. longis 2- 6 mm. distantibus, circa apicem marginibus et midnervio infra cum serris acutis 0.4-0. 5 mm. longis 0.5-1. 5 mm. distantibus, inflores- centia foeminea solitaria terminale cum bracteis plurimis 16 mm. latis amplectentibus, syncarpio subsessile 3 cm. longo 2 cm. diametro ellip- soideo, drupis multis 9-10 mm. longis 2-2.5 mm. diametro fusiformis, corpore 4-5 mm. longo, pileo obliquiter conico 4 (-5) mm. longo ad stylum tenuato, stylo subulato cum linea stigmata proxima 2.5-3 mm. longa, endo- carp io supramediale lateribus 0.1 mm. crassis pallidis, semine 3.5 mm. longo 1.6 mm. diam- etro late elliptico, mesocarpio infero cavernoso. DESCRIPTION OF HOLOTYPE: Shrub 15.5 m. tall, forking; leafy branchlets 6-9 cm. in diam- eter, the bark reddish brown, shining, ringed by pale leaf scars; leaves 51-58 cm. long, 8-10 mm. wide, subcoriaceous, ligulate, M-shaped in cross section, long tapering to a subulate apex which at 10 cm. back from the tip is 3.5 mm. wide, margins at the very base entire, but from about 4 cm. up the margins with stout subu- late teeth, 0.5-1. 3 mm. long, 1-3 mm. apart, wholly black, ascending at 45°, the midrib un- armed; at the midsection the margins black serrate, the teeth 0.5 mm. long, 1-2 mm. apart, the midrib below appressed, subulate, black ser- rate, the teeth 2-6 mm. apart; near the apex the margins and midrib below sharply serrate, the teeth 0.4-0. 5 mm. long, 0.5-1. 5 mm. apart; pistillate inflorescence solitary, terminal, sur- rounded by several bracteal leaves 16 mm. wide around the spike, but otherwise like the normal foliage leaves; syncarp subsessile 3 cm. long, 2 cm. in diameter, ellipsoid, with numerous drupes; drupes 9-10 mm. long, 2-2.5 mm. in diameter, fusiform, the body 4-5 mm. long, the pileus oblique conic, 4 (-5) mm. long, taper- ing into the subulate style which is linear stig- matic for 2.5-3 mm. on the proximal side; en- docarp from middle to apex of body, the wall 0.1 mm. thick, pale; seed 3.5 mm. long, 1.6 mm. in diameter, broady ellipsoid; lower mesocarp cavernous. HOLOTYPUS: North Borneo, Keningau Dis- trict, 76V2 mi. on path from Tenom to Ranau (81/2 mi. NE. of Tambunan), 2,100 ft. alt., Aug. 27, 1954, G. H. S. Wood & J. Wyatt-Smith A 4,4.30 ( bri ) . DISCUSSION: The new species P. fruticosus is in the section Aero stigma. Its nearest relative is the Malayan species P. aurantiacus Ridl. which has the leaves 3.8 cm. wide; inflorescence of 5 heads; syncarp 6.2 cm. long, 2.5 cm. in diam- eter; drupes 18 mm. long; and the endocarp 12 mm. long. On the other hand, P. fruticosus has leaves 8-10 mm. wide; the syncarp solitary; drupes 9-10 mm. long; and the endocarp 3.7 mm. long. The new epithet is the Latin adjective, fruti- cosus, shrubby, given with reference to the plant habit. Pandanus ruber sp. nov. (sect. Micro stigma) Figs. 37-38 DIAGNOSIS HOLOTYPI: Arbor 3-5 m. alta ramosa, trunco cum aculeis adscendentibus ar- mato, radicibus fulturosis paucis aculeatis, foliis 172 cm. longis 10.2 cm. latis coriaceis in sectio M-formatis ligulatis acutis in sectio 10 cm. ex apice 4 cm. latis in base integris sed ex puncto 9 cm. e base cum aculeis 1-1.6 mm. longis 2-7 mm. distantibus salientibus stramineis, mid- nervio inerme, in sectio mediale marginibus cum aculeis 1-1.5 mm. longis 3-6 mm. distanti- P B ,T H O Fig. 35. Pandanus fruticosus St. John, from holotype. Habit, X 1. Fig. 36. Pcmdanus fruticosus St. John, from holotyp?. a , Syncarp, lateral view, X 1; b, drupe, lateral view, X 10; drupe, longitudinal median section, X 10; d, pileus and stigma, lateral view, X 10; e, pileus and stigma, apical view, X 10; /, leaf base, lower side, X 1; ^ leaf middle, lower side, X 1; ^ leaf margin, at middle, X 10; i, leaf apex, lower side, X 1. Fig. 37. Pandanus ruber St. John, from holotype. a, Syncarp, median section, X 1; ^ drupes, apical view, XI \ c, drupe, lateral view, X 1; d, drupe, apical view, X 4; e, f, drupes, lateral view, X 4; g, drupe, longi- tudinal median section, X 4. Page 68: Revision of Pandanus, 7. Papua and the Solomons — S t. John 583 bus proxime adpresse-adscendentibus, midnervio cum aculeis 0.3-1 mm. longis 3-11 mm. distanti- bus valide aciculatis adpresse adscendentibus subbrunneis, proxime apice marginibus unice- vel duplo-serratis serris 1.5-2. 1 mm. longis 2-5 mm. distantibus stramineis, supra nervis secun- dariis cum aculeis paucis eis 0.8-1 mm. longis, infra nervo mediale saliente et cum aculeis si- multantibus 1.5-2 mm. longis 3-10 mm. distan- tibus, syncarpiis solitariis plerumque 42 cm. longis 10.5 cm. diametro cylindricis sed sub- triangulatis, receptaculo aurantiaco-rubro, dr ti- pis numerosissimis in sectio mediale 13-15 mm. longis 3-4.5 mm. latis 2.5-3 mm. crassis unicar- pellatis obscure rubris 4-6-angulatis parte Va supera libera, pileo 3-4 mm. longo anguste pyramidale asymmetrico et arcuato corpore ob- longo in base subcuneato, stigmatibus 0.8-1. 5 mm. longis et latis suborbicularibus brunneis verticalibus proximalibus infra petaso terminale, endocarpio basale 9 mm. longo osseoso obscure brunneo lucido 0.4-0. 5 mm. crasso, mesocarpio apicale caverno unico. DIAGNOSIS OF holotype: Tree, 3-5 m. tall, branched; trunk and branches armed with up- turned prickles; prop roots few, prickly; leaves 172 cm. long, 10.2 cm, wide, coriaceous, M- shaped in cross section, ligulate, acute, 4 cm. wide at 10 cm. from the tip, at the very base the margins entire, but from 9 cm. up prickly, the prickles 1-1.6 mm. long, 2-7 mm. apart, spread- ing, straw-colored, the nearby midrib unarmed; at the middle point the margins with prickles 1-1.5 mm. long, 3-6 mm. apart, closely ap- pressed ascending, those of the nearby midrib 0.3-1 mm. long, 3-11 mm. apart, stout acicular appressed ascending, brownish near the tip the margin simply or doubly subulate-serrate, the teeth 1.5-2. 1 mm. long, 2-5 mm. apart, straw- colored, above the ribs of the two secondary pleats with a few ascending prickles 0.8-1 mm. long, below the salient midrib with similar prickles 1.5-2 mm. long, 3-10 mm. apart; syn- carps solitary, averaging 42 cm. long, 10.5 cm. in diameter, cylindric but somewhat 3 -sided; re- ceptacle, orange-red; with myriad drupes, those at midsection 13-15 mm. long, 3-4.5 mm. wide, 2.5-3 mm. thick, 1 -celled, dark red, 4-6-angled, upper Va free, the pileus 3-4 mm. long, nar- rowly pyramidal, asymmetric and arcuate to the apex, body almost oblong, but slightly cuneate below; stigmas 0.8-1. 5 mm. long and wide, suborbicular, brown, vertical on proximal side of apex, below the apical cap; endocarp basal, 9 mm. long, bony, dark brown, shiny, the walls 0.4-0.5 mm. thick; apical mesocarp a single cavern. HOLOTYPUS: Papua, Central Division, Bella Vista, 1,450 m. alt., planted by the natives in gullies and ravines close to their gardens, Nov. 1933, L. J. Brass 5,463 (bri); and isotype (A). DISCUSSION: The closest relative is P. Holl- rungii Warb. of Deutsch Neu-Guinea, now Aus- tralian New Guinea. It differs by having: the leaves 2.5 m. long, 5 cm. wide; syncarp 35 X 4-7 cm., sessile, spathe-surrounded; drupes 11- 13 mm. long, 2.5—3 mm. wide, the apex semi- orbicular-pyramidal; stigmas apical, subhorizon- tal, 1.5 mm. wide; and the endocarp slightly submedian. P. ruher differs by having: the leaves 1.7 m. long, 10.2 cm. wide; syncarp 42 X 10.5 cm.; drupes 13-15 mm. long, 3-4 mm. wide, 3 mm. thick, the apex narrowly pyramidal, asym- metric and arcuate; stigmas 0.8-1. 5 mm. wide, vertical on proximal side below overhanging apical cap; and the endocarp basal. The collector recorded that, "the leaves of mature trees always much damaged by in.- sects. . . .” Also that the trees were cultivated. It is doubtless also wild in the forest. The several similar species with cylindric, red syncarps yield on boiling a fruit sauce important in the diet of the New Guinea natives. This plant was identified as P. conoideus Lam. by Merrill and Perry (Journ. Arn. Arb. 20: 175, 1939). Lamarcks species is similar in its sub- terminal stigmas, but the drupes are 16 mm. long, the pileus is broader, and the endocarp is only submedia, and the leaves are narrower and with horrid spines. It is based on P. ceramicus Rumph ( 1743) from Amboina, Ceram, and the Indonesian islands. The epithet, ruber , is the Latin adjective meaning red, here chosen because of the red color of the fruits. Pandanus yuleensis sp. nov. ( sect. Pandanus ) Figs. 39-40 DIAGNOSIS HOLOTYPI: Frutex ad 6.6 m. alta, cortice radicibus crassis adscendentibus ad 12 584 PACIFIC SCIENCE, Vol. XV, October 1961 Fig. 38. Pandanus ruber Sc. John, from holotype. a, Leaf base, lower side, X 1; b, leaf middle, lower side, X 1; c, leaf apex, lower side, X 1; d. leaf margin, near apex, X 3; e, leaf apex, upper side, X 1- Page 70: Revision of Pandanus, 7. Papua and the Solomons — S t. John 585 Fig. 39- Pandanus yuleensis St. John, from holotype. a, Phalange, lateral view, X f ^ phalange, longitu- dinal median view, X 1; c, phalange, apical view, X 1; d, carpel, apical view, X 3; e, stem with ascending rootlets, lateral view, X 1- 586 PACIFIC SCIENCE, Vol. XV, October 1961 mm. longis 3-10 mm. distantibus obtectis, foliis 1.6 m, longis 5.3 cm. latis gladiformis vel ligu- latis sed paulatim ex base in apice longo subu- lato diminuentibus coriaceis in sectio transverso late M-formatis griseo-viridibus infra apparente glaucis apice deltoideo et 10 cm. ex puncto 1 mm. lato in base marginibus integris sed ex puncto 16 cm. e base cum dentibus 1.5-2 mm. longis 3-12 mm. distantibus subulatis paene cur- vatis in base brunneis midnervio inerme, in me- dio in uno latere cum dentibus 1-1.5 mm. longis 3-15 mm. distantibus adpresse adscendentibus in puncto brunneis, latere altero inerme, mid- nervio infra cum dentibus 0.3 mm. longis 10- 30 mm. distantibus paucis brunneis validis subu- latis adpresse adscendentibus, ad apicem mar- ginibus midnervioque infra cum serris paucis 0.2 mm. longis remotis brunneis, syncarpiis scilicet terminalibus solitariisque griseo-viridibus ad 12.5 cm. diametro, phalangibus 4.4—47 cm. longis 2.2-3 cm. latis 1.9-2 cm. crassis late cu- neatis apice subconvexo vel truncato compressis parte l A supera libera lateribus 5-6-angulosis subcurvatis vel paene planis laevibus lucidis in sicco brunneis, suturis lateralibus nullis, sinibus apicalibus centralibus 3-3.5 mm. profundis, car- pellis 4-6 apicibus subaequalibus conico-subor- bicularibus eis marginalibus plerumque cum petaso truncato parvo et sub margine proximo eo stigma est, stigmatibus 1-2 mm. longis ovali- bus vel obdeltoideis sillonatis centripetalibus, sutura proxima ex stigmate dimidia vel omnina distancia ad vadum extento, endocarpio mediale 25-27 mm. longo osseoso obscure brunneo la- teribus 2.5-8 mm. crassis, seminibus 15-20 mm. longis 4 mm. diametro, mesocarpio apicale ca- vernoso cum fibris longitudinalibus et mem- branis medullosis, mesocarpio basale fibroso et carnoso. DESCRIPTION OF ALL SPECIMENS EXAMINED: Tree up to 6.6 m. tall; bark gray, with abundant ascending, stout, adventitious roots up to 1 2 mm. in length but having no further development, placed 3-10 mm. apart; leaves 1.6-1.96 m. long, 5.3-6 cm. wide, sword-shaped or ligulate but tapering gradually from the base to the long subulate tip, coriaceous, in section broad M- shaped, gray-green, below apparently somewhat glaucous, the apex triangular and 10 cm. back only 1 mm. wide, at base the margins entire, but c FIG. 40. Pandanus yuleensis St. John, from holo- type. a, Leaf base, lower side, XI \ b, leaf middle, lower side, X 1; c. leaf apex, lower side, X 1. Page 72: Revision of Pandanus, 7. Papua and the Solomons — S t. John 587 beginning 10-16 cm. up, with teeth 1.5-2 mm. long, 3-12 mm. apart, subulate, gently curved, brown based, the midrib unarmed; at the middle with teeth on one side 1—1.5 mm. long, 3-15 mm. apart, brown tipped, subulate, appressed ascending, but the other margin unarmed, the adjacent midrib below with a few teeth 0.3 mm. long, 10-30 mm. apart, brown, stout subulate, appressed ascending; near the apex the margins and midrib below with a few, remote, brown serrations 0.2 mm. long; syncarps probably ter- minal and solitary, "gray green, up to 12.5 cm. in diameter,” phalanges 4.4-4.7 cm. long, 2.2-3 cm. wide, 1.9-2 cm. thick, broad cuneate, the apex subconvex or truncate, compressed, upper l A free, the sides 5-6-angled, gently curving or nearly plane, smooth, shining, when dried brown, lateral carpel sutures none, central apical sinuses 3-3.5 mm. deep; carpels 4-6, the apices subequal, conic-semiorbicular, the marginal ones usually with a small truncate cap in the proximal lee of which is the inclined stigma, stigmas 1-2 mm. long, oval to obdeltoid, creased, centripetal, below it the proximal crease on the carpel apex running from half way to all the way to the bot- tom; endocarp median, 25-27 mm. long, bony, dark brown, the lateral walls 2.5-8 mm. thick; seeds 15-20 mm. long, 4 mm. in diameter; apical mesocarp cavernous, traversed by longi- tudinal fibers and with delicate medullary mem- branes; basal mesocarp fibrous and fleshy. Papua: Yule Island, Jan. 16, 1955, /. S. Womersley & N. W . Simmonds 3,099 (BRI). SPECIMENS EXAMINED: Papua, Yule I., July/ Aug., 1918, C. T. White 770 (bri), consisting of one leaf only. DISCUSSION: The most similar kind is P. odoratissimus L. f. var. novo-caledonicus (Mar- telli) St. John, which has the phalanges 5.5 cm. long, the upper half free; carpels 7-9, the outer 2-3 -times the larger; apical central sinuses 2- 3.5 mm. deep; stigmas apical, horizontal; and the endocarp l A the length of the phalange. On the contrary, P. yuleensis has the phalanges 4.4- 4.7 cm. long, the upper third free; carpels 4-6, subequal; apical central sinuses 3-3.5 mm. deep; stigmas steeply oblique, placed below the trun- cate apex; and the endocarp more than half the length of the phalange. The new specific epithet is an adjective made from the name of the type locality, Yule, by adding the Latin place suffix, -ensis. P and anus magnicavernosus sp. nov. (sect. Mammillarisia ) Figs. 41-42 DIAGNOSIS HOLOTYPI: Arbor 8 m. alta erecta ramosa cum radicibus fulturosis, foliis 1.8-1. 9 m. longis 9.5-10 cm. latis pallide viridibus mol- liter subcoriaceis subplanis in base midnervioque luteo-brunneis midnervio et nervis binis secun- dariis fortibus sed dentis innoxis apice acuto base paene dilatata et cum marginibus merri- branaceis per 25 cm. inermibus, marginibus ul- tra cum serris fortibus 1.5-2 mm. longis 3-10 mm. distantibus stramineis sed apicibus brun- neis, midnervio inerme, in sectio mediale mar- ginibus cum serris 0.5 mm. longis 2-10 mm. distantibus subulatis valde adpressis, midnervio infra cum aculeis simulantibus subulatis sed fortioribus et paene brevioribus, nervis secun- dariis inermibus, ad apicem marginibus cum serris %ubulatis 0. 5-0.9 mm. longis 1-2 mm. distantibus pallidis adscendente adpressis, nervis secundariis supra scabris cum serris minutis ir- regularibus adscendentibus, midnervio infra cum aculeis 0.3-1 mm. longis subulatis adscendenti- bus aggregatis, pedunculis 90 cm. longis gracili- bus triangularibus, syncarpio solitario penduloso 24 cm. diametro globoso glauco, phalangibus 2-3 (rare 1 ) -carpellatis numerosis 7. 5-8. 5 cm. longis 17-35 mm. latis 14-21 mm. crassis cunei- formis 5-7-angulosis carnosis parte supera 1/6 libera apice depresse conico plurianguloso cum 1-3 extremitatibus, sinibus apicalibus centrali- bus 1-2 mm. profundis, stigmatibus 1.5-2 mm. longis cordatis apicalibus sulcatis quando binis centripetalibus quando tribus eo centrali ad la- terem dirigito, lateribus phalangiorum planis, endocarpio in tertia infera 24-30 mm. longo osseoso pallide brunneo lateribus 3-4 mm. cras- sis, seminibus 10-12 mm. longis 3—4 mm. dia- metro ellipsoideis, mesocarpio supero cavernam unical dimidiam quam longam quam syncarpiam formante sed sectio breve cum fibris et mem- branis medullosis, mesocarpio infero fibroso et carnoso. DIAGNOSIS OF HOLOTYPE: "Tree, 8 m. tall, erect, branched near top; trunk supported on Fig. 41. Pandanus magnicavernosus St. John, from holotype. a, Phalanges, apical view, X 1; b, phalanges, lateral view, X 1; ^ phalange, lateral view, X 1; d, phalange, longitudinal median section, X 1; ^ phalange and stigmas, apical view, X 4. Fig. 42. Pandanus magnicavernosus St. John, from holotype. a, Leaf base, lower side, XI \ b, leaf middle, lower side, X 1; c, leaf margin at middle, X 4; d, leaf apex, lower side, X 1; e, leaf margin near apex, X 4. 590 PACIFIC SCIENCE, Vol. XV, October 1961 stilt roots; leaves 1.8-1. 9 m. long, 9.5-10 cm. wide, pale green, glaucous beneath, midrib and basal part of leaf yellow brown,” soft coriaceous, nearly flat, the midrib and two secondary ribs strong, the teeth inoffensive, the apex abruptly acute, the base slightly dilated and with a mem- branous unarmed margin for 25 cm., beyond that the margins with stout serrations 1.5-2 mm. long, 3-10 mm. apart, stramineous but brown tipped, the adjacent midrib unarmed; the margins at midsection with the teeth 0.5 mm. long, 2-10 mm. apart, subulate but closely ap- pressed, the midrib below with similar but stouter and slightly shorter subulate prickles, the secondaries unarmed; near the tip the mar- gins with subulate prickles 0. 5-0.9 mm. long, 1-2 mm. apart, ascending-appressed, pale, the secondaries above scabrous with minute, irreg- ular, ascending serrations, below the midrib crowded with ascending subulate prickles 0.3-1 mm. long; fruit on slender 3-angled peduncle 90 cm. long; syncarp solitary, pendulous, 24 cm. in diameter, globose, glaucous; phalanges 2-3 (rarely 1) -celled, numerous, 7. 5-8. 5 cm. long, 17-35 mm. wide, 14-21 mm. thick, wedge- shaped, 5-7-angled, fleshy, upper 1/6 free, the apex depressed conic, with 1-3 tips, several angled, apical central sinuses 1-2 mm. deep and the crease between adjacent phalanges narrow, 1 mm. deep; stigmas 1.5-2 mm. long, cordate, apical, creased, if two then centripetal, if three then the central one facing laterally, phalange sides plane; endocarp in lower third, 24-30 mm. long, bony, pale brown, the lateral walls 3-4 mm. thick; seeds 10-12 mm. long, 3-4 mm. in diameter, ellipsoid; upper mesocarp one huge cavern occupying upper half of phalange except a short apical part with longitudinal fibers and transverse medullary membranes; lower meso- carp fibrous and fleshy. holotypus: Solomon Islands, San Cristoval I., Star Harbour, rain forest on coastal hills, one tree seen, Oct. 28, 1932, L. J. Brass 3,120 (BRl) . DISCUSSION: This is a striking tree, with am- ple leaves and huge, conspicuous fruits. It fits into the section Mammillarisia, but at present has no known close relatives. The specific epithet is formed from the Latin, magnus , large; cavernosas, cavernous, in allusion to the cavernous upper mesocarp. Feral Rabbit Populations on Pacific Islands J. S. Watson 1 In their monograph, Thompson and Worden (1956: 7-22) discuss the world distribution of the European rabbit ( Oryctolagus cuniculus L.), but make no mention of several colonies estab- lished on islands of the tropical and subtropical Pacific. It is worth drawing attention to these both to complete the picture and because of the light they throw on the great adaptability of this species. laysan island (25° 46' N.; 171° 49' W.): A low, sand and coral island about 2 mi. long by 1 mi. wide, in the northern half of the Hawaiian chain. Groves of sandalwood trees, thickets of bushes, and fan palms formerly grew on the island, which supported a vast albatross rookery and five endemic species of land birds. The guano deposits of Laysan were exploited be- tween 1892 and 1904, and the manager of the works, Mr. M. Schlemmer, introduced various breeds of domestic rabbits, including the large white domestic English rabbit, to the island in about 1903 (Dill and Bryan, 1912; E. H. Bryan, 1942). The island was later set aside as a bird sanctuary. Professor Homer R. Dill led a scien- tific expedition from Iowa State College to Lay- san in 1911. The expedition found that al- though the rabbits had killed many bushes and nearly exterminated several plant species, they had on the whole done less damage than might have been expected from their numbers. The extermination of the rabbits, however, was rec- ommended as they were likely to eat out the vegetation which would result in the disappear- ance of the insects on which a number of en- demic bird species were dependent (Dill and Bryan, 1912). To accomplish this an expedition 1 Formerly of the Animal Ecology Division, New Zealand Department of Scientific and Industrial Re- search, Wellington, N. Z. Manuscript received January 4, 1961. J. S. Watson died August 12, 1959, after complet- ing the first draft of this paper. Additional informa- tion on rabbits in the Hawaiian Islands has since been obtained and included in the text. — K. Wodzicki. of four men was sent to the island for 3 months in 1912-13; unfortunately, they were inade- quately equipped to deal with the problem and, although over 5,000 rabbits were shot, there were so many petrel burrows and other cover that without poison it was impossible to elimi- nate them in so short a time (Bailey, 1956). In 1923 the Tanager Expedition visited Laysan; the island by then had been reduced to a barren waste of sand with a few stunted trees, only 4 of the 26 species of plants recorded from the island were found ( Christophersen and Caum, 1931); and there were a few hundred rabbits present. These were shot, the last ones being hunted out individually. The endemic warbler ( Acrocepha - lus familiaris Rothschild) had vanished; the last three Laysan honeyeaters ( Himatione sanguinea fraithii Rothschild) died during a sand storm while the expedition was on the island; and the Laysan rail ( Porzanula palmeri Frohawk) died out shortly afterwards (Wetmore, 1925). Exter- mination of the rabbits was completed and no sign of them was seen in 1936, when the island was found to be recovered in vegetation. While circling over Laysan in an aeroplane in 1949, Bailey (1956) found that the vegetation had staged a remarkable recovery and there were concentrations of black-footed albatrosses ( Di - omedea nigripes Forster ) and Laysan albatrosses (D. immutabilis Rothschild). lisianski ISLAND (26° N., 174° W. ): An- other low sand and coral island about 1V4 mi. long by 44 mi. wide, in the Hawaiian chain, some 115 mi. west of Laysan. Rabbits from Lay- san liberated there some time after their intro- duction on Laysan in 1903 had destroyed the vegetation by 1913, when only a few living but many dead rabbits were seen (Elschner, 1915). When the Tanager Expedition arrived in 1923 the only signs of rabbits were their bleached and weathered bones, and the vegetation was start- ing to come back. The rabbits, having stripped the vegetation, had apparently died of starvation (Wetmore, 1925). 591 592 ISLETS IN MAIN HAWAIIAN GROUP: Accord- ing to Mr. D. N. Woodside (personal commu- nication), rabbits were introduced to several of the small islets of volcanic origin (10 to 70 acres) lying off the islands of the main Hawai- ian group at unrecorded dates before or about 1915. At present only three of the islets are populated by rabbits: Manana ("Rabbit Is- land") (22° N, 158° W.), off the coast of Oahu; Lehua (22 Q N., 160° W.), off the coast of Niihau; and Molokini (21° N, 156° W.), situated in the channel between Maui and Ka- hoolawe. These tuff-cone islets are rather arid, suffering summer drought and occasionally miss- ing sufficient winter rains to alleviate the dry- ness. The rabbit populations are subject to vio- lent fluctuations, "die-offs” occurring during the droughts, which perhaps enables the vegetation to survive. There are no predators on these is- lets. The rabbits on Lehua and Molokini appear to be a mixture of domestic breeds, being of every colour; the Manana rabbits closely resem- ble the American cottontail in colour and ear length and it is conceivable that these rabbits are in fact cottontails ( Sylvilagus spp.). PHOENIX ISLAND (4° S., 171° W.) : A small coral island about 44 mi. long in the Phoenix group. There was an American guano company working here in the late 1860’s and about this time domestic rabbits were liberated (Bryan, 1942). Lister (1891) found them fairly plentiful in 1889 and they were still fairly numerous in 1924, but apparently were doing no damage to the vegetation (Bryan, 1942). A survey party, which in 1937 caught a number of rabbits there in an unsuccessful attempt to liberate them in the Gilbert Islands, found them in very poor condition, squatting and allowing themselves to be caught after running about a hundred yards. There is no fresh water on the island (Maude and Maude, 1952). PHILLIP ISLAND: A small island about half a mile long of decomposed basalt rising to 900 ft., lying off Norfolk Island (29° S., 168° E.), originally covered with trees and thick vegeta- tion. Pigs which were introduced at an early date destroyed much of the undergrowth and rooted up the soil, starting erosion. Rabbits were introduced subsequently (Laing, 1915) and the island thronged with them in 1865 and their principal food seemed to be the bark of trees PACIFIC SCIENCE, Vol. XV, October 1961 (Brenchley, 1873). By 1912 the island was fast becoming a complete desert; there was practi- cally no soil, only a few isolated trees; hardly a seedling was to be seen and the only grass was in clefts in the rock near the beach (Laing, 1915). Rabbits were still present in 1943 and a single tree was conspicuous in its isolation (Mr. I. L. Baumgart, personal communication). The present situation on Phillip Island is in- dicated in a recent letter to Dr. K. Wodzicki from Mr. B. a’B Marsh, Agricultural Officer, Norfolk Island. Mr. Marsh made a hurried visit to the island on 4 March 1961 and saw rabbit scratchings, droppings, and a few burrows, the last being under the roots of Lagunaria Patersonii and another unidentified tree. The visit, being in the middle of the day, was not well timed for seeing rabbits, but, as a very rough guess, Mr. Marsh thought there might be between two and four dozen rabbits on the island. The animals had been eating the leaves of Lagunaria which seemed their main food. Vegetation consists of six Norfolk Island pines ( Araucaria excelsa) , several dozen Lagunaria trees, a few unidentified trees with large shiny leaves, a reed known locally as Mo-oo and a few grasses and weeds. The reed grows in a few areas of 1/10 to Vi acre, usually in flat areas where there is enough soil to retain water, but over perhaps 80 per cent of the is- land there is no vegetation or soil. Rabbits do not eat the reed or the unidentified trees; the latter are the only trees that appear healthy. All trees have their roots exposed to a depth of 1-6 ft.; the pine trees are not actively grow- ing but are setting seeds; the leaves of the Lagunaria are confined to the main branches so that the trees look as though recovering from a fire. The topography of the island is steep with V-shaped erosion gullies at frequent intervals; run-off of water is extremely rapid and com- plete and carries extraordinary quantities of solid material. The effective rainfall is probably about 5 to 10 in. per annum overall, but the steeper parts retain practically no rain; the is- land was quite dry only two days after heavy rain. All soil has gone except from a few flat areas and screes, and removal of rabbits will Feral Rabbits on Pacific Islands — WATSON 593 not now alter the island except in a very minor way. In March 1953 myxomatosis was introduced, but further introductions were abandoned due to the difficulty of landing. It is now unlikely that any further eradication work will be at- tempted. DISCUSSION These islands provide an interesting contrast. On Phoenix Island, almost on the equator, con- ditions must be only just within the level of tolerance for the rabbit species, and factors other than food presumably hold the population down to a level where it does not seriously affect the vegetation. Rabbits have been on Phillip Island for nearly a hundred years and must have reached a state of equilibrium with the vegeta- tion, most of which has been destroyed, but some plant species must survive that are both resistant to rabbit grazing and yet sufficiently palatable to support the existing rabbit popula- tion. On Lisianski Island no such equilibrium was reached, and it was thought that rabbits would similarly have gone from Laysan Island had they not been exterminated (Wetmore, 1925). Presumably the rate of increase of the rabbits quickly produced a large population, and the vegetation was destroyed to such an extent that it was unable to recover in time to main- tain even a greatly reduced number of rabbits. Manana, Lehua, and Molokini islets are all subject to droughts during which the rabbit populations are drastically reduced and presum- ably the vegetation can recover sufficiently to survive. REFERENCES Bailey, A. M. 1956. Birds of Midway and Lay- san islands. Denver Mus. Nat. Hist. Piet. Ser. 12: 1-130. Brenchley, J. L. 1873. Jottings During the Cruise of H.M.S. Curasao Among the South Sea Islands in 1865. Longmans, London. 487 pp. Bryan, E. H. 1942. American Polynesia and the Hawaiian Chain. Honolulu. 253 pp. Christophersen, E., and E. L. Caum. 1931. Vascular plants of the Leeward Islands, Ha- waii. B. P. Bishop Mus. Bull. 81: 1-41. Dill, H. R., and W. A. Bryan. 1912. Report of an expedition to Laysan Island in 1911. U. S. Dept. Agr. Biol. Surv. Bull. 42: 1-30. Elschner, C. 1915. The Leeward Islands of the Hawaiian Group. Honolulu. 69 pp. (Not seen, cited by Christophersen and Caum, 1931.) Laing, R. M. 1915. A revised list of the Nor- folk Island flora, with some notes on the spe- cies. Trans. N. Z. Inst. 47 : 1-39. Lister, J. J. 1891. Notes on the birds of the Phoenix Islands (Pacific Ocean). Proc. Zool. Soc. Lond. ? : 289-300. Maude, H. C, and H. E. Maude. 1952. The colonization of the Phoenix Islands. J. Poly. Soc. 61: 62-89. Thompson, H. V., and A. Worden. 1956. The Rabbit. Collins, London. 240 pp. Wetmore, A. 1925. Bird life among lava rock and coral sand. Nat. Geogr. Mag. 48: 76-108. Relationships of the Red-backed Voles of Japan E. W. Jameson, Jr. 1 Eastern Asia is particularly rich in genera and species of microtine rodents, some of which are more or less intermediate between Microtus on the one hand and Clethrionomys on the other. Where species of only these two genera are con- cerned, there is no problem of identity because there are differences in the color, in the struc- ture of the palate, and in the skull in general. The age of the specimen changes strikingly the condition of the cheek teeth, in which there are some of the most important generic features. Molars of immature specimens of Clethrion- omys , for example, are rootless, and resemble in this respect the molars of adult individuals of Microtus. In eastern Asia there are a number of species which resemble the species of Clethrion- omys in the palate, which terminates in a shelf, and are like Microtus s. str. in having rootless molars. For these forms which combine some of the structural characteristics of Clethrionomys and Microtus there are a number of generic names. A reviewer naturally wonders if his material contains mature individuals, and this doubt has caused some disagreement as to the proper generic allocation of red-backed voles of the Far East. The red-backed voles in Japan have been discussed by Hinton (1926), whose conclusions have been accepted by Eilerman (1941) and by Ellerman and Morrison-Scott (1951). Hinton (1926: 259-262) judged that several different forms were all based on immature specimens and placed bedfordiae, andersoni, and niigatae as synonyms of smithi . He very carefully ex- plained the pitfalls in separating voles on a small series and he emphasized the early age at which voles breed (and are apparently adult). In the Japanese forms in question, the age of the speci- mens is of great importance because these names were proposed for individuals in which the mo- 1 Department of Zoology, University of California, Davis, California. Manuscript received February 10, 1961 . lars are rootless, and Hinton’s conclusion was that they are all immature specimens. His cau- tion is certainly justified but can be carried to misleading extremes, for by combining several distinct forms as one, it is not difficult to show that a given character, in this case the condition of the molar roots, is indeed remarkably variable. There are remarks in Hinton’s appraisal of the situation that raise some doubts as to its appli- cation to the specimens he had at hand. For example, in reference to the small vole called smithi, he stated (1926: 260) that the type was a young male and added that . . by accident the fifty-three specimens of the series collected in Hondo, Kiushiu, and Shikoku, by M. P. An- derson in the following year are all young too.” It is possible, of course, to sample a population and obtain a series in which immature speci- mens predominate; but it is incredible that such an experienced collector as Anderson would have preserved 53 specimens of smithi from at least three widely separated localities and fail to include a single adult. Hinton added: Later on Mr. Anderson collected thirteen in Hokkaido. The majority of these are adult, some even old; in size, skull form, and tooth pattern they are strikingly different from the material referred by Thomas to E. — smithii. These were therefore described as a new species, E. ( Crase - omys) bedfordiae. But two specimens of this original series of E. bedfordiae are immature, and these are not distinguishable from the ma- terial upon which E. smithii was founded. These two presumably immature specimens could possibly belong to Clethrionomys rutilus, which is not uncommon in some parts of Hok- kaido; and, in this region, rutilus is a small bright short-tailed form somewhat like smithii. On the same page he dismissed the form de- scribed as E. ( C. ) andersoni with the comment, . . specimens, however, are merely large ado- lescents, intermediate in age between the adult material upon which E. bedfordiae was estab- lished and the immature material referred to 594 Red-backed Voles of Japan— JAMESON 595 as E. smithii.” Lastly, "Anderson’s E. niigatae also has been founded upon an adolescent animal. . . Apparently Hinton attached no importance to the occurrence of immature smithi on Kyu- shu, Shikoku, and Honshu, and the absence of adult smithi everywhere but on Hokkaido. It is surprising that he did not consider that these forms with rootless molars might actually be adults, for on the mainland of Asia and on Tai- wan are genera of red-backed voles in which the molars are rootless in the adults. Hinton recog- nized Eothenomys, Anteliomys, and Aschizomys , genera in which the molars are rootless; but failed to suggest any similarities between smithi and the species of these genera. The classification of these voles by Japanese mammalogists differs considerably from the re- cent presentation of Ellerman (1941) and of Ellerman and Morrison-Scott (1951). In the latter work the authors were apparently un- aware of the very careful study on Japanese murid rodents by Tokuda ( 1941 ) . In his monograph, Tokuda ( 1941 ) followed Oldfield Thomas in general in the classification of the Japanese red-backed voles. He consid- ered smithi as a separate species in the genus Clethrionomys, bedfordiae was retained as a sub- species of Clethrionomys rujocanus, and ander- soni was judged to be a distinct species but closely allied to rujocanus . Tokuda (1941: 51) followed Hanaoka in placing Craseomys niiga- tae as a synonym of Clethrionomys andersoni. Tokuda did not suggest at this time that smithi had affinities except in the genus Clethri- onomys. The first to suggest a different position for smithi was Imaizumi (1949), who placed it in the genus Eothenomys . Subsequently, Tokuda (1955) presented evidence for considering smithi a species of Anteliomys. Imaizumi (1957), considering Anteliomys a synonym of Eothenomys , continued to use the latter name, and described a second species, Eothenomys kageus, from the northern part of Honshu. The most recent classification of the red- backed voles in Japan is by Imaizumi (I960). In this arrangement there are Eothenomys: smithi and kageus; and three species of Cle- thrionomys: rutilus, rujocanus, and sikotanensis; in contrast to Tokuda, Hinton, and other recent students, Imaizumi maintained niigatae as a species distinct from andersoni, and placed both in Aschizomys. During 1952 I spent 10 months in Japan and made a small collection of voles. In 1958 and 1959 I was able to trap additional speci- mens of all but one ( Clethrionomys sikotanen- sis) of the forms of red-backed voles in Japan, and from these specimens and the excellent dis- cussions of Tokuda and Imaizumi, it is appar- ent that the original descriptions of Oldfield Thomas and the recent researches of Tokuda and Imaizumi more accurately express the rela- tionships of these voles. The work of Hinton was executed apparently with little knowledge of the geologic history of the Japanese archi- pelago and without the advantage of having studied these Japanese voles in the field. The arrangement given below is, in most respects, like that accepted by Japanese mammalogists today. Because of the stature of Hinton’s mono- graph, his conclusions have been followed by European and American zoologists, and reiter- ated, in regard to the rodents in question, by Ellerman (1941) and Ellerman and Morrison- Scott (1951). The environmental distributions and economic importance of these voles is dis- cussed by Ota and Jameson (in press). GENUS Clethrionomys Tilesius Originally all the red-backed voles in Japan were placed in Clethrionomys. The Japanese C. rutilus mikado exists only on the island of Hok- kaido. In general appearance and in most details rutilus closely resembles the other species of Clethrionomys s. str. The skull is light with rounded contours, the palate is abnormal (dif- fering from Microtus ) , the cheek teeth are rel- atively light and are rooted in the adult. The mammae are in four pairs. The species in this genus are characterized by other features but the above are sufficient to separate rutilus from the other species of microtine rodents in Japan. The species rujocanus has long been known as a rather aberrant species of Clethrionomys. Miller (1898^: 360) treated in some detail the distinctive characters of this vole: So divergent is the animal that it may well be questioned whether it is to be regarded as a true 596 PACIFIC SCIENCE, Vol. XV, October 1961 Evotomys (= Clethrionomys). Its heavy and ap- parently imperfectly rooted teeth more closely resemble those of many species of typical Mi- crotus than they do the weak, perfectly rooted teeth of true Evotomys . The relationship of the root of the lower incisor and the posterior lower molar, while not typical of either genus are clearly suggestive of Microtus rather than Evotomys . The palate structure, on the other hand, appears to agree with that of Evotomys. Also, in a key to the Arctic species of red- backed voles Miller (1898^: 359) separated rufo- canus by the comment: "Teeth large and heavy as in Microtus (never perfectly rooted?). . . In characterizing rufocanus, Hinton (1926: 245) also emphasized its approach to the species of Microtus and on page 215 pointed out that in- dividuals look mature "long before the molars show the slightest sign of rooting.” To distin- guish rufocanus from the other species of Cle- thrionomys Miller ( 1900) created the subgenus Craseomys with the following characters: Skull as strongly angular as in Microtus, the postorbital processes well developed; teeth rel- atively as large as in Microtus, the molar row about equal to the diastema; roots of molars developed late in life; root of posterior lower molar lying in a distinct capsule on the lingual side of incisor root. Soon after Thomas ( 1905) placed the Japanese forms bedfordiae and andersoni in the subgenus Craseomys. Later (1907), he used Craseomys as a genus for Craseomys regulus ( = Clethri- onomys rufocanus regulus) from Korea and noted that not even the oldest in the series of 18 specimens showed any trace of roots on the molars. Aschizomys lemminus was described as a new genus and species by Miller (1898£) on the basis of a single specimen and was characterized in these terms: "Palate as in Clethrionomys. Molars small and weak, as in Clethrionomys, but teeth growing from a persistent pulp as in Microtus) which strongly displaces root of large posterior lower molar. Plantar tubercles six. Number of mammae unknown.” The illus- trations accompanying the original description indicate other features common to both lemmi- nus and rufocanus: the encapsulated roots of the upper second molar and the lower third molar and the reduced lateral tubercles on the rostrum. The upper third molar of lemminus is longer than in rufocanus and there is little doubt that the two species are separate. The third upper molar of Clethrionomys niigatae of northern Honshu is long and similar to lemminus. In the original description of Aschizomys Miller recom- mended that rufocanus was allied to lemminus and later Imaizumi (1957: 199) suggested that Clethrionomys niigatae might well be placed in Aschizomys. Also, Hinton (1926: 43) stated that lemminus seemed to be very close to rufocanus. However, as Miller suggested ( 1898 ^: 359), the molars of rufocanus form roots very late in life; it is quite possible that rufocanus and lemminus are alike in this respect as well, and that a large series of Aschizomys lemminus would probably contain a few individuals old enough to have partly rooted molars. Granting this supposition, a reasonable arrangement would be to place rufocanus with lemminus in Aschizomys . In this case Craseomys Miller, 1900 will be- come a synonym of Aschizomys Miller 1898. Recent authors have regarded Craseomys a syn- onym of Clethrionomys; perhaps because they considered it unnecessary to retain a subgenus which contained but a single species. In placing rufocanus, niigatae, andersoni, and a new spe- cies from Honshu together with lemminus all in Aschizomys, one must decide the proper posi- tion of this group. Although Miller preferred to call it a genus, he did consider that Aschizomys could be placed with Eothenomys and Ante- liomys. Eothenomys combines characters of Mi- crotus and Clethrionomys; and Aschizomys (in- cluding rufocanus and andersoni) bridges the gap between Eothenomys and Clethrionomys. As pointed out by Ellerman and Morrison- Scott (1951: 670), Russian authors place As- chizomys as a subgenus of Alticola. The two groups are certainly very close; but, if I am correct in presuming that very old specimens of Aschizomys lemminus will tend to develop roots on the molars, then Aschizomys is more appropriately placed with Clethrionomys. Inasmuch as a few old individuals of rufo- canus and niigatae have partly rooted molars, it seems best to place Aschizomys as a subgenus under Clethrionomys, and the forms known in Japan can be separated by the key below. Red-backed Voles of Japan — JAMESON 597 KEY to SPECIES OF Clethrionomys IN japan 1. Molars developing long roots in adults; postorbital crests poorly developed or absent; second upper molar and third lower molar not encapsulated in adults: (subgenus Clethrionomys ) I . rutilus Molars rootless except in very old adults; postorbital crests well developed; second upper molar and third lower molar usually encapsulated in adults (subgenus Aschizomys ) 2 2 . Third upper molar with three inner salient angles (Fig. 1); tail less than one-half the body length; on Hokkaido rufocanus bedfordiae Third upper molar with more than three inner salient angles or loops although some pro- jections may be irregular or rounded (Fig. 1); on Honshu... 3 3. Salient angles of all cheek teeth rounded or irregular, and triangles often open (Fig. 1); auditory bullae relatively small; in low elevation evergreen broad leaved forests of ex- treme southern Honshu ..imaizumii Salient angles more or less pointed, and triangles usually closed; auditory bullae relatively large; in coniferous forests of the northern half of Honshu.. 4 4. Rostrum relatively long; anterior palatine foramina arise from about level of anterior molar alveolae; upper molar tooth row more than 6.0 mm niigatae Rostrum shorter; anterior palatine foramina arise from a point distinctly anterior to the level of anterior molar alveolae; upper molar row less than 6.0 mm... andersoni Clethrionomys rufocanus bedfordiae (Thomas, 1905) Evotomys bedfordiae Thomas, 1905, Ab- stracts, Proceedings, Zoological Society of London, no. 23. (Holotype from Shinshi- notsu, near Sapporo, Hokkaido.) This is the form of rufocanus which occurs on the northern island, Hokkaido. It is a large, richly colored species with a tail of moderate length. It is difficult to compare bedfordiae with the other named subspecies: the illustrations of rufocanus shanseius (of the adjacent mainland) in Hinton (1926) are made from specimens which perhaps are not rufocanus and may not even belong in Clethrionomys. Presumably the nearest relatives live on the island of Sakhalin and the adjacent mainland. Clethrionomys siko- tanensis (Tokuda, 1935) seems to be quite dis- tinct from, although allied to, rufocanus; siko- tanensis is known from the Southern Kuriles and may be more closely related to lemminus. The skull of bedfordiae resembles the illus- tration of C. r. rufocanus in Hinton (1926: fig. 80 ) ; the depicted specimen is from an un- specified locality. Hinton’s illustration of C. r. shanseius (1926: fig. 83) is from the type series of Caromys inez; shanseius (from China) should be closer to bedfordiae than is typical rufocanus (from Sweden), but there is little doubt that Hinton’s illustration of shanseius represents a form at least specifically distinct from rufocanus. The enamel pattern (Fig. 1) is typical of C. r. bedfordiae and the third upper molar is especially characteristic. In 22 adults examined, 2 have distinctly rooted molars; in 8, the pulp cavities are more or less closed off, in- dicating that roots would probably have formed later in life. The anterior palatine foramina are expanded anteriorly and quite narrow poste- riorly and resemble C. r. rufocanus in this re- spect. The tail is less than one-half the body length. C. rufocanus bedfordiae is the most abundant microtine mouse on Hokkaido and is almost always more common than C. rutilus. In the ab- sence of any competing species of Microtus ( which does not occur on Hokkaido) , rufocanus is sometimes a common inhabitant of open meadows and is of considerable economic im- portance (Ota and Jameson, in press). 598 PACIFIC SCIENCE, Vol. XV, October 1961 Clethrionomys rutilus J 5mm Clethrionomys Clethrionomys niigatae rufocanus Clethrionomys Clethrionomys imaizumii Eothenomys kageus Eothenomys smithi Fig. 1 . Enamel patterns of Japanese red-backed voles. Red-backed Voles of Japan — JAMESON 599 Clethrionomys andersoni (Thomas, 1905) Evotomys andersoni Thomas, 1905, Abstracts, Proceedings, Zoological Society of London, no. 28. (Holotype, from Tsunagi, Iwate- ken, Honshu.) This form is sometimes considered to be a subspecies of rufocanus, to which it is closely related. C. andersoni is easily separable from C. rufocanus bedfordiae by the longer third upper molar, which has four inner salient angles. C. andersoni is close also to C. niigatae, which is sometimes considered to be the same. Examina- tion of specimens of the type series of both andersoni and niigatae indicated that the two are distinct and that adults can be identified by the above key. C. andersoni is an inhabitant of coniferous forests in northern Honshu, in Aomori-ken, Fu- kushima-ken, and Iwate-ken, according to Imai- zumi (I960: 134). Clethrionomys niigatae (Anderson, 1909) Craseomys niigatae Anderson, 1909, The An- nals and Magazine of Natural History, vol. 4, ser. 8: 317. (Holotype from Akakura, Niigata-ken, Honshu. ) For a very long period C. niigatae had been considered to be a synonym of C. andersoni; but, after studying specimens from the type series of both species, I agree with Imaizumi ( I960) that the two are distinct species. In the nine adults examined, one specimen has closed pulp cavities and incipient roots. This is in contrast to the specimens examined by Imai- zumi (1957 and I960); the series in the Na- tional Science Museum in Tokyo all possess rootless molars. C. niigatae lives in the higher elevations of central Honshu. In Nagano-ken, it is found at approximately 1900 m. and higher among boulders both in virgin forests of fir and spruce and in rather open tangles of wild rasp- berries and currants. Clethrionomys imaizumii, new species TYPE: Adult male, skin and skull. Collected 13 Feb. 1959, Nachi Falls, 300 feet elevation, Wakayama-ken, Honshu, Japan; E. W. Jameson, Jr., no. 1083. RANGE: Probably in broadleaved forests at low elevations in extreme southern Wakayama- ken (Kii Peninsula). DIAGNOSIS: A rather large, brightly colored, long-tailed species of the subgenus Aschizomys. Dorsum Rood’s Brown; venter Sayal Brown; tail bicolor, thinly haired, as in rufocanus and an- dersoni (color from Ridgway, 1912). Skull with rather small auditory bullae. Molars with angles rounded. MEASUREMENTS ( in mm. ) : Holotype ( and paratype): Total length, 194 (184); tail, 67 (61); hind foot, 22 (21). Skull: condylobasilar length, 29.1 (skull of paratype damaged); zyg- omatic breadth, 15.6 (15.1); interorbital breadth, 4.3 (4.4); lambdoidal breadth 12.5 ( — ); alveolar length of upper molar row, 6.9 (6.3 ); diastema, 8.3 (7.7). This species is most nearly like C. niigatae, but differs strikingly in the color and dentition. The molars are rootless with open dentine spaces in the two adult males seen. The enamel pattern is quite different, however: the triangles are rounded and indistinct and frequently open (Fig. 1 ) . The auditory bullae are markedly smaller. This species inhabits the luxuriant hardwood forests of the southern part oFthe Kii Peninsula in Wakayama-ken, the southernmost part of Honshu. Specimens were first collected by Dr. R. Kano; and Dr. Yoshinori Imaizumi kindly directed me to collecting localities. These four forms of the subgenus Aschizomys are separable on external characters, and these features are substantiated by constant dental and other cranial morphology. The relative proportions of the tail and body lengths are rather different for rufocanus, on the one hand, and for andersoni, niigatae, and imaizu- mii on the other hand. C. niigatae and C. imaizumii are long tailed. In nine specimens of niigatae from the upper slopes of Yatsuga- take, the tail is from 58 to 66 mm. in actual length, and the tail length is from 51 to 62 per cent of the body length. In a series of 22 adults of C. rufocanus bedfordiae from various localities in Hokkaido, the tail is from 34 to 56 mm., and the tail length is from 28 to 46 per cent of the body length. It is quite possible that large series of both species would show some 600 PACIFIC SCIENCE, Vol. XV, October 1961 overlap in this character but the difference is nevertheless real and fairly constant. C. imaizu- mii is a long-tailed animal like niigatae. The general fascies of the enamel patterns of C . ru- focanus bedfordiae, C. niigatae , and C. under - soni are angular (Fig. 1), but under soni and niigatae are distinct in having four inner salient angles in the third upper molar, whereas rufoca- nus bedfordiae has three. The enamel pattern of imaizumii is quite different in having the angles rounded, and there is a tendency for the loops and triangles to remain open ( Fig. 1 ) . The four species differ also in the form of the anterior palatine foramina. The hind foot of rufocanus bedfordiae is more densely furred than that of niigatae , imaizumii , and andersoni , but ail spe- cies are alike in possessing six plantar tubercles between which there are tiny projections. Imaizumi (1957) intimated that niigatae might be genetically distinct from rufocanus bedfordiae because the molars are rarely rooted in bedfordiae and., seemed never to be rooted in niigatae. In one specimen of niigatae examined by me, the pulp cavities are closed and there are Incipient roots. Later (I960) Imaizumi placed niigatae and andersoni in Aschizomys and kept rufocanus in Clethrionomys ; but such an ar- rangement does not indicate the nearness of these three species. As will be pointed out later in this paper, andersoni and niigatae almost cer- tainly emigrated from Hokkaido from a stock close to rufocanus. Imaizumi’s suggestion, rea- sonable as it may seem, simply emphasizes the weakness of the presence or absence of molar roots as a generic character in this case. Clethrionomys rutilus mikado (Thomas, 1905) Evotomys mikado Thomas, 1905, Abstract, Proceedings, Zoological Society of London, no. 23, p. 19 . ( Holotype from Aoyama, Hokkaido. ) This bright red little vole is quite unlike any other species in Japan. The color, rounded en- amel pattern (Fig. 1), well-developed molar roots, and small size separate rutilus from rufo- canus, the only other vole in Hokkaido. The molar row is rather short, as in Eothenomys smithi and E. kageus, and the enamel patterns are similar, but the angles are less rounded. In C. rutilus there are four' pairs of mammae. One specimen was examined. C. rutilus mikado is a rather uncommon dweller of the forests of Hokkaido and is not known to occur elsewhere In Japan. GENUS Eothenomys Miller The other genus of red-backed voles in Japan is Eothenomys ; most of the species occur in China and one ( melanogaster , the type of the genus) is found also on Taiwan. These species are quite close to Clethrionomys in general and some of them are perhaps related to the species of the subgenus Aschizomys In particular. In Eothenomys the color Is highly variable, but some species are reddish and so resemble the species of Clethrionomys, with which they have been confused on more than one occasion. Eothenomys differs in possessing rootless molars and only two or three pairs of mammae. Even in the oldest specimens there is no closure of the pulp cavities. The skull of Eothenomys is light and delicate and tends to be rounded; and the palate terminates in a shelf, as in Clethrionomys. key TO species OF Eothenomys IN japan 1. Mammae four kageus Mammae six smithi Eothenomys smithi (Thomas, 1905 ) Evotomys smithi Thomas, 1905, Annals and Magazine of Natural History, series 7, vol. 15, p. 493. (Holotype from Kobe, Hon- shu.) In Japan, the species smithi was described from specimens from Kobe, on the island of Honshu; smithi is also a common species on Shikoku and Kyushu, and a shorter-tailed sub- species ( okiensis ) was described from the is- land of Dogo in the Oki Group. The species smithi was described by Thomas in 1905, who at that time placed it in Evotomys ( = Clethri- onomys); and, noting some differences from the morphology of most species of that genus, he erected the subgenus Ehaulomys. Thomas stated then that smithi showed characters of Evotomys , Eothenomys, and Anteliomys. Externally smithi is similar to species of Clethrionomys , except that there are two or three pairs of mammae Red-backed Voles of Japan — JAMESON 601 instead of four pairs. The form of the skull resembles that of C. rutilus and the other species of Clethrionomys s. str. (not including rufoca- nus, andersoni, imaizumii, and niigatae ) : the skull is delicate and rounded and the postorbital projections are barely developed. The enamel pattern tends to be rounded rather than angular; Thomas noted that the closed triangles tend to be little broader than long ( Fig. 1 ) . Phaulomys stands apart from Clethrionomys because the molars are rootless in the adult. Thomas also pointed out the encapsulated root of the second upper molar which projects into the orbital fossa; this invariably occurs in microtine rodents when the molars are not rooted and is simply another way of stating this condition. Since the description of smithi, many hundred of speci- mens have been collected and the molars are known to be rootless even in old adults. (Hin- ton’s [1926} observation to the contrary re- sulted from his confusing Clethrionomys rufo- canus bedfordiae and perhaps C. rutilus mikado with smithi .) Recent studies by Japanese workers (e.g., Tokuda, 4955, and Imaizumi, 1957 ) indicated that smithi is allied to the Chi- nese species of Anteliomys (a synonym or sub- genus of Eothenomys ) , and the proper name for this species is Eothenomys smithi. Nineteen specimens were examined. Eothenomys kageus Imaizumi, 1957 Eothenomys kageus Imaizumi, 1957, Bulletin, National Science Museum (Tokyo), vol. 3, no. 3, p. 204. (Holotype from Yamura- machi, Minamitsuru-gun, Yamanashi-ken, Honshu. ) Imaizumi ( 1957 ) named the form from the northeastern part of Honshu as E. kageus. From the detailed description of kageus, it is apparent that kageus and smithi are very close. E. kageus has two pairs of mammae whereas in smithi there are three pairs. The size, color, and body measurements of E. kageus are very close to those of E. smithi. The enamel patterns of the two species are also very similar (Fig. 1 ) . The original description of E. kageus included a dif- ference in the posterior angle of the zygomatic arch; but the explanatory drawing (Imaizumi, 1957: fig. 3) indicated that the angle measured on E. kageus was different from that taken from E. smithi, and the data are not comparable. The differences in the bacula of E. kageus and E. smithi may be due to individual variation: spe- cimens of both species are extremely variable and there seem to be no constant differences in the bacula of the two named forms (Fig. 2). Six specimens were examined. Future studies may reveal that E. kageus is a subspecies of E. smithi. Eothenomys smithi and E. kageus are similar in habits and habitat. They dwell in forested re- gions from sea level to elevations of 2500 m. or more, but generally do not occur together with Clethrionomys niigatae, imaizumii, or an- dersoni. On Yatsugatake, E. kageus occurs up to about 2000 m., above which level C. niigatae is found; but on the eastern slope of Ontake, where C. niigatae does not occur, E. smithi ex- tends well above 2000 m. (Tokuda, 1950); and on Mt. Fuji (where C. niigatae is absent) E. kageus is found to the upper limit of the forest (Imaizumi, 1944). On Kyushu, E. smithi is the only red-backed vole, and it lives in wooded areas from at or near sea level to the highest peaks, becoming more abundant at the higher elevations. On Shikoku, where Microtus montebelloi is not found, E. smithi may move from the forest to relatively open grassy or brushy areas (Ota and Jameson, in press). Eothenomys does not occur on Hokkaido. The bacula of the red-backed voles in Japan ( Fig. 2 ) are rather variable within a given spe- cies, and provide rather poor taxonomic char- acters at the specific level. Some of the smaller bacula are obviously from young animals but even examples of the same size may be remark- ably dissimilar. There are differences in both size and shape of the shaft and its base as well as in the three prongs. There seem to be no distinctions between the bacula of Eothenomys kageus and E. smithi unless the shaft in kageus is slightly more slender. There are probably no differences between the bacula of Clethrionomys niigatae and C. rufocanus bedfordiae, but C. imaizumii has relatively larger prongs which are rather divergent, and in this respect resembles those of some species of Microtus. Most bacula of these voles have a slight keel on the median tyne, a character of the bacula of both Clethri- onomys and Microtus. No bacula of C. ander- soni were available. 602 PACIFIC SCIENCE, Vol. XV, October 1961 Clethrionomys rufocanus FIG. 2. Bacula of Japanese species of Eothenomys and Clethrionomys ( Aschizomys ). Red-backed Voles of Japan — JAMESON 603 GEOGRAPHIC ORIGIN OF RED-BACKED VOLES IN JAPAN In the Pleistocene and earlier the present ar- chipelago of Japan was connected to the main- land in the south, between Kyushu and the Korean Peninsula, and in the north, between Hokkaido, Sakhalin, and adjacent Siberia. At approximately the same time, changes resulted in the separation of Kyushu and Korea, and also the formation of the Tsugaru Strait be- tween Hokkaido and Honshu. Consequently, the islands of Kyushu, Shikoku, and Honshu no longer received immigrants from the continent, and they were isolated from the effects of faunal movements on the mainland. Today, the fauna of the old islands of Kyushu, Shikoku, and Hon- shu is, in many respects, quite different from that of the immediately adjacent mainland, and has its closest affinities to the southwest in China. This faunal difference has been men- tioned as early as Thomas’ 1905 paper. On the other hand, Hokkaido and Sakhalin remained part of the mainland and were subjected to the faunal changes that affected southeastern Siberia and Korea. There are no land mammals in Hokkaido that are more than subspecifically distinct from those of nearby Siberia and Korea. The important and extensive movements of mammals in this general area during the Pleis- tocene (see Simpson, 1947: 643) changed the fauna of Hokkaido, replacing with more recent forms the original fauna that now remains in the older Japanese islands. Eothenomys smithi probably entered Japan on the old route between Korea and Kyushu via a connection which remains today as the islands of Tsushima and Iki. E. smithi moved on from Kyushu to Shikoku and is now the only micro- tine rodent known from the latter island. This vole moved also to Honshu and now occupies a large part of that island, at least to southern Nagano-ken. E. smithi, or its progenitor, at one time must have occurred in Korea although the genus is not there now. E. kageus may have come from the north via Sakhalin and Hok- kaido, although its occurrence is now confined to the northern half of Honshu south to where it abuts the geographic range of E. smithi. The changes which resulted in the extinction or the emigration of Eothenomys from Korea and Si- beria apparently affected Hokkaido as well. Ac- tually, it seems likely that E. smithi and E. kageus are offshoots from a single ancestor which differentiated slightly along the coast from Korea to Siberia so that different subspe- cies entered Kyushu and Hokkaido. This could account for the great similarity between these two forms that now occupy adjacent but not overlapping geographic areas. Similarly, the species of Clethrionomys came to Japan by two routes. C. imaizumii, which is now known only from the low elevations of Wakayama-ken, may have its relatives among one of the many kinds of red-backed voles described from China. This is conjecture, but this vole is now associated with a semitropical flora of broad-leaved hardwoods. This species, moreover, is morphologically quite distinct from C. under soni, C. niigatae, and C. rufocanus bed- fordiae, which are found in forests of fir ( Abies spp.) and spruce ( Picea spp. ) . C. niigatae and C. andersoni are almost certainly arrivals via Hokkaido. Although niigatae and andersoni are close relatives of C. rufocanus bedfordiae, they may have- been derived from an earlier stock. Hokkaido and Honshu were separated since be- fore the Pleistocene, and niigatae and andersoni were immune to the immigrants of C. rufocanus bedfordiae. Not only are the three species close morphologically, but they share three species of fleas which are parasites of C. rufocanus in Hok- kaido: Catallagia striata, Megabothris sokolovi, and Rhadinopsylla alphabetica. One ( Catallagia striata ) occurs also in Siberia and another (. Megabothris sokolovi) in the Kuriles. The occurrence of C. rutilus needs little ex- planation. It is obviously a rather recent ar- rival in Hokkaido. If it ever extended to Hon- shu, it did not persist, and it may well have entered Japan after the separation of Hokkaido from Honshu. SUMMARY There are seven species of red-backed voles in Japan. Clethrionomys ( Clethrionomys ) rutilus occurs in Hokkaido. The species andersoni, nii- gatae, and imaizumii (from Honshu) and rufo- canus (from Hokkaido) are considered to rep- resent Aschizomys, which is placed as a sub- genus of Clethrionomys. Eothenomys smithi in- 604 PACIFIC SCIENCE, Vol. XV, October 1961 habits Kyushu, Shikoku, and Honshu, and E. kageus occupies the northern half of Honshu. ACKNOWLEDGMENTS The specimens discussed above were collected with the help of Dr. Tatsuo Udagawa, Mr. Ko- hei Sakaguti, Dr. Ryo Tanaka, and Dr. Kasio Ota, with whom I made many excursions. To these gentlemen I am most grateful. Dr. Richard van Gelder ( American Museum of Natural His- tory), and Dr. John E. Hill (British Museum [Natural History]), generously loaned compara- tive material. Dr. David H. Johnson graciously read the manuscript. REFERENCES Allen, Glover M. 1940. Natural History of Central Asia, vol. XI, part 2. The mammals of China and Mongolia. American Museum of Natural History. Part 2: 621-1350. Ellerman, J. R. 1941. The Families and Genera of Living Rodents. British Museum (Natural History). Vol. II, 690 pp. Ellerman, J. R., and T. C. S. Morrison- SCOTT. 1951. Checklist of Palaearctic and In- dian Mammals, 1758 to 1946. British Mu- seum (Natural History). 810 pp. Hinton, M. A. C. 1926. Monograph of the Voles and Lemmings ( Microtinae ) , Living and Extinct. British Museum (Natural His- tory). 488 pp. Imaizumi, Yoshinori. 1944. [Mammals of the Narusawa-mura District on the northern slope of Mt. Fuji.] Shizen Kagaku to Hakubutsu- kan 18(1): 1-10. [In Japanese.] 1957. Taxonomic studies on the red- backed vole from Japan, Part I. Major divi- sions of the vole and descriptions of Eothen- omys with a new species. Bulletin, National Science Museum (Tokyo), no. 40: 195-216. I960. The Mammals of Japan. Hoiku- sha, Osaka. 196 pp. Miller, Gerrit S., Jr. 1896. Genera and Sub- genera of Voles and Lemmings. North Amer- ican Fauna no. 12. 78 pp. 1898^. Description of a new genus and species of microtine rodent from Siberia. Proc. Acad. Nat. Sciences of Philadelphia 50: 368-371. 1898 A Notes on Arctic red-backed mice. Proc. Acad. Nat. Sciences of Philadel- phia 50: 355-367. 1900. Preliminary revision of the Euro- pean red backed mice. Proc. Wash. Acad. Sci. 11: 83-109. Ota, Kasio, and E. W. Jameson, Jr. In press. Ecological relationships and economic im- portance of Japanese microtinae. Ecology. Ridgway, Robert. 1912. Color Standards and Color Nomenclature. Author, Washington, D. C. 43 pp. + 53 color plates. Simpson, George Gaylord. 1947. Holarctic mammalian faunas and continental relation- ships during the Cenozoic. Bulletin, Geolog- ical Society of America 58(7): 613-687. Thomas, Oldfield. 1905. On some new Jap- anese mammals presented to the British Mu- seum by Mr. R. Gordon Smith. Annals and Magazine of Natural History 1 5 ( ser. 7): 487-495. 1905. The Duke of Bedford’s zoolog- ical exploration in eastern Asia, I. List of mammals obtained by Mr. M. P. Anderson in Japan. Proceedings, Zoological Society of London 2 (of 1905): 331-363. 1907. The Duke of Bedford’s zoolog- ical exploration in eastern Asia. Proceedings, Zoological Society of London ( 1906) : 863. 1908. The Duke of Bedford’s zoolog- ical exploration in eastern Asia, X. List of mammals from the Provinces of Chih-li and Shan-si, N. China. Proceedings, Zoological Society of London ( 1908) : 635-646. Tokuda, Mitosi. 1941. A revised monograph of the Japanese and Manchou-Korean Muri- dae. Transaction, Biogeographical Society of Japan 4 (1): 1-156. 1950. [Field mice and shrews on Mts. Ontake and Yatsugatake, with special refer- ence to "sympatric” and "allopatric” mam- malian species found in these districts.] Doo- butsu Gaku Zasshi 59(9): 210-213. [In Japanese.] 1953. Small mammals from Hokkoda (Aomori Prefecture) with special reference to "allopartic” shrew-moles in this district and other districts of Japan. Ecological Re- view 13(3): 129-134. 1955. Congeneric species of voles found in Japan and Yunnan. Bulletin, Biogeograph- ical Society of Japan 16-19: 388-391. Two Shark Incidents at Eniwetok Atoll, Marshall Islands 1 E. S. Hobson , 2 F. Mautin , 3 and E. S. Reese 2 The two shark incidents described in this paper can be be considered as unprovoked shark attacks (Gilbert et al. I960: 324). The impor- tance of documenting these incidents is that they were observed by persons with some experience in the study of animal behavior. The great need of precise and detailed observational informa- tion on shark attacks has been stressed in recent reports on the subject (Coppleson, 1938: ix; Gilbert et al, I960: 323; Tester, I960: 181). It is the belief of the authors that well-docu- mented reports of shark incidents will contrib- ute significantly to a clearer understanding of the stimulus situations in which shark attacks on man can be expected. The authors wish to thank Dr. Albert L. Tester, who read the manuscript and whose Of- fice of Naval Research contract Nonr-2756(00) Project NR 104503 enabled Mr. Hobson and Mr. Mautin to participate in the shark research program at the Eniwetok Marine Biological Lab- oratory, and the Atomic Energy Commission, whose support enabled Dr. Reese to work at the laboratory. The authors are also indebted to Dr. R. W. Hiatt, director of the laboratory, who made the facilities available. FIRST INCIDENT At noon on September 1, I960, Hobson, Mautin, and Reese were engaged in spear fish- ing activities on a reef about 25 ft. below the surface and approximately 200 yd. from shore on the lagoon side of Parry Island, Eniwetok Atoll, Marshall Islands. Dr. Richard A. Booloo- tian of the University of California was in a 16-ft. skiff powered by a 35-hp. outboard motor which was idling nearby. The sea was calm, there was very little wind, and the sky was clear. Underwater visibility was in excess of 100 ft. 1 Contribution no. 158, Hawaii Marine Laboratory. Manuscript received November 9, I960. 2 Department of Zoology and Entomology, Univer- sity of Hawaii. 3 781 5th Avenue, New York City. Table 1 indicates the nature of the equipment that was being used. The spatial relationship of the divers to one another and to the skiff above the reef at the start and the end of the encounter is shown in Figure 1. Hobson speared a 25-lb. grouper ( Plectro - pomus sp.), which swam, trailing the spear, beneath a large coral boulder (rock 1, Fig. 1). Almost immediately three grey sharks ( Car- charhinus menisorrah, Fig. 2), approximately 5-7 ft. in length, appeared from the deep water of the lagoon and began to circle the rock. Hob- son observed ( 1 ) the sharks were swimming in an excited manner, and ( 2 ) they paid no atten- tion to him floating quietly on the surface. Sud- denly the grouper broke cover and dashed to a second large coral boulder (rock 2, Fig. 1). The three sharks immediately followed and began to circle the second rock. Reese, who was directly above rock 2, no- ticed the sharks and was also impressed with the excited appearance of their swimming. Mean- while, Hobson swam towards rock 2 to inform Mautin and Reese of the speared fish and the excited sharks in the area. At about this time Mautin observed the sharks and noted their ex- cited state. All three began to swim toward the skiff, making a conscious effort not to produce an undue amount of commotion on the surface. There were now four or five sharks swimming in an excited manner on the reef below. Mautin, who was unarmed and farthest from the boat, was swimming strongly in an effort to join the others. Both Hobson and Reese observed that Mautin’s swim-fins were breaking the surface of the water resulting in clouds of bubbles being carried beneath the surface at each down stroke. The following sequence of events was very rapid. Almost simultaneously with the above ob- servation, Mautin had reached a position almost above rock 2 ( see Fig. 1 ) . Suddenly one of the sharks rose from the bottom and swam very fast in a direct line towards Mautin. Mautin saw the approaching shark and rolled on his right side 605 606 PACIFIC SCIENCE, Yol. XV, October 1961 TABLE 1 Physical Characteristics of Divers and Nature of Equipment Used at Time of First Incident HOBSON 1 MAU- TIN REESE Weight 180 195 175 Height 5' 10" 6'3- 5 ' 9 - Skin color brown brown brown Bathing suit khaki none gray & white Swim fins light blue yellow black Mask dark blue white dark blue Snorkel green white black Spear Hawaiian sling none Hawaiian sling with 6 -ft. with 6 -ft. free shaft free shaft 1 Same equipment in second incident. in order to face the onrushing shark. When the shark was very close he kicked it violently on the snout with his swim-fins. The shark veered away, circled once or twice behind Mautin and, upon the approach of Hobson and Reese, swam away. All three swimmers were now quiet in the water. Immediately after the first shark left the bot- tom, a second shark began to swim toward the surface on the same course as the first shark. However, after approaching approximately half- way, it returned to the reef floor. It is suggested that the cessation of commotion on the surface Fig. 1. Spatial relationship of divers to one another and to skiff above the coral reef at start of incident (white triangles) and at time of shark attack (black triangles ) . Letters beside triangles refer to diver which they represent, H — Hobson, M = Mautin, and R := Reese. Arrows indicate direction of movement of sharks. and the presence of three swimmers resulted in the first shark swimming away and the second shark returning to the bottom. Both Hobson and Reese were approximately 20-30 ft. from Mautin at the time the first shark attacked; however, there was no indica- tion that the shark directed any attention toward them. It appeared that the shark was orienting to the agitation at the water’s surface caused by Mautin’s strong swimming. This observation supports the suggestion by Tester (I960: 183) and others that sharks are attracted by unusual commotion. A second possibility is that the shark was attracted to the bright yellow swim- fins which Mautin was wearing, but Hobson, on the basis of unpublished data, feels that this is unlikely. Mautin gained the impression that the shark came for him believing he was the wounded fish, and that competition from the other sharks resulted in the direct and swift attack not pre- faced by the usual cautious investigation. The four to five sharks were still swimming in the same excited manner in the vicinity of rock 2 when the divers left the water. SECOND INCIDENT A second incident occurred the following day, September 2. Again the time was approximately noon, the sky was clear, and the surface of the lagoon was calm with underwater visibility in excess of 100 ft. Fish were being speared in shallow water along the beach at Sand Island adjacent to the deep water channel leading into the lagoon from the east. Hobson was in the water, and John C. Kay, a graduate student at the University of Hawaii, was standing off the beach in a 16-ft. skiff with the outboard motor idling. Hobson speared a 10-lb. parrot fish (Scar- idae) which carried the spear shaft toward deeper water. About 35 yd. from the beach, at the edge of the reef, the bottom drops abruptly from approximately 15 ft. to over 100 ft. Grey sharks are common along this drop off. Hobson pursued the fish along the bottom, thus avoiding the surface commotion which had apparently attracted the attention of the sharks the day be- fore. He overtook the injured fish on the bot- tom at a depth of approximately 15 ft. on the edge of the drop off. When the trailing spear Shark Incidents at Eniwetok — Hobson, Mautin, and Reese 607 shaft was grasped, the fish struggled free. At the same instant a 6-ft. grey shark (Car char hinus menisorrah ) was upon Hobson, swimming at a slightly upward inclination directly at his face. The mouth of the shark was open, and move- ment of the jaws was distinctly recalled. Instinc- tively Hobson projected his right arm, catching the onrushing shark under the head. This, com- bined with a twisting, ducking motion, diverted the shark’s forward rush over Hobson’s left shoulder. The shark turned abruptly and circled so close that the spear which was still in the left hand could not be brought into play. As the shark circled, Hobson turned with it, push- ing it away several times with his open hand until the shark was circling at a distance of ap- proximately 4 ft.- — enough room to bring the spear into use. The blunt end, which happened to be the end toward the shark, was used as a prod, and after a few jabs the shark opened its circle to approximately 10 ft. The shark appeared to be rapidly losing its aggressiveness. It seemed that apprehension toward the diver was now displacing the attack response which had as- serted itself seconds before. As the boat ap- proached, the shark fled into the adjacent deep water. The entire incident lasted only a few seconds and took place entirely on the bottom. The shark was recognized by a deformation of the dorsal fin as one which had been involved in many of the behavior experiments conducted during the summer. It was, therefore, very fa- miliar with the sight of humans in the water. There had been no apparent hesitation involved in its attack. Presumably its approach was made from deep water, coming over the edge of the reef which was within 10 ft. of the attack site. An approach from any other direction would have been observed. Although the use of bare hands in warding off a shark has generally been discouraged ( Gil- bert et al., I960: 326), this was the only re- course in the present situation. The effectiveness of the hands in this case was no doubt largely due to the relatively small size of the shark. No lacerations of the hands occurred. Fig. 2. Carcharhmus menisorrah, the species which was involved in both incidents, is common in the la- goons of most Pacific atolls, attaining a length of about 7 ft. (Photo: E. Hobson.) 608 PACIFIC SCIENCE, Vol. XV, October 1961 TABLE 2 Summary of Similarities and Differences Between the Two Incidents FIRST INCIDENT SECOND INCIDENT Activities at time of incident spear fishing spear fishing Dead or injured fish present yes: 1 yes: 1 Skin diving or SCUBA skin diving skin diving Species of shark Carcharhinus menisorrab Carcharhinus menisorrab Number of sharks in area 3-5 1 Size of sharks (length) 5-7 ft. 6 ft. Date Sep. 1, I960 Sep. 2, I960 Time of day approx, noon approx, noon Condition of sky clear, few clouds clear, few clouds Condition of sea cajm calm Location of incident lagoon, Eniwetok lagoon, Eniwetok Distance from shore 200 yd. 35 yd. Near deep water yes yes Underwater visibility 100 ft. + 100 ft. -!- Depth of water 25 ft. 15 ft. Depth at which incident occurred - surface 1 5 ft., bottom Water temperature approx. 85° F. approx. 85° F. Nature of bottom coral reef coral reef Number of persons in water 3 1 Nature of approach direct, fast direct, fast Area of body approached feet head Persistency of shark moderate considerable DISCUSSION The two incidents are summarized in Table 2. The authors believe that neither the coloration of the swimmer’s equipment, the condition of the water, -nor the time and nature of the day are significant. There are four major differences between the two incidents which are believed to be significant: (1) the number and relative positions of persons in the water, ( 2 ) the num- ber of sharks in the vicinity, ( 3 ) the site of the incidents, the first on the surface, the second on the bottom, and (4) the location of the diver relative to the wounded fish. There are three major similarities between the two incidents which are believed to be significant: (1) both occurred in the vicinity of, and subsequent to, the spearing of a fish; (2) both occurred near deep water; and (3) both involved the same species of shark, which attacked without hesita- tion. The last point may be explained by two facts. First, throughout the summer, Hobson, Mautin, and others noticed that C. menisorrab was a far more aggressive shark than the other two species which are common in the lagoon (C. melanop- terus and Triaenodon obesus) . This view is sup- ported by Harry (1953: 48), who reports that the natives of the Tuamotus fear this species and claim it will attack man. Second, both incidents occurred within approximately a mile of where shark behavior experiments were conducted throughout the summer. Hobson recognized the shark in the second incident as having been a participant in these experiments, and it is prob- able that the sharks Involved in the first incident had also had considerable experience with seeing human beings in the water. It is the belief of Hobson, Mautin, and others that the sharks be- came progressively bolder towards humans dur- ing the course of the summer. Thus it may be significant that both incidents occurred at the end of the summer. It appears that one can generalize to this ex- tent. Some species of sharks, for example C. menisorrab , are attracted to an area where there are injured fish. Apparently stimuli originating with the injured fish release an excited, highly motivated pattern of exploratory behavior In the shark. At such times many sharks seem to be particularly reactive to any unnatural distur- bances created by humans in the general area. Shark Incidents at Eniwetok — Hobson, Mautin, and Reese 609 This was illustrated by the response to the com- motion at the surface of the water in the first incident. Furthermore, it seems they may also attack a human simply because he is located in an area of high concentration of the postulated stimulating factor, such as occurred in the sec- ond incident. In both incidents, visually directed thrusts at the sharks using hands, feet, or spear proved effective in momentarily warding off the attacking shark. Eibl-Eibesfeldt and Hass (1959: 746) and others have also noted the effectiveness of a pointed shaft as a shark de- terrent. CONCLUSIONS 1. Sharks may be expected to appear wherever spear-fishing activities are undertaken, particu- larly in those areas where aggressive species are known to occur. 2. When sharks appear under these circum- stances they may attack a human located within the immediate area of the stimulating factor as well as humans associated with an unnatural amount of activity in the general area. 3. Sharks which have become familiar with the sight of humans in the water may be more likely to attack humans than those which have not. 4. The conclusion reached by others — that vis- ually directed blows, preferably using a pointed instrument (such as a spear shaft), are effective in warding off sharks — is confirmed. REFERENCES Coppleson, V. M. 1958. Shark Attack. Angus and Robertson, Sydney, xi + 266 pp., 32 pis. Eibl-Eibesfeldt, L, and H. Hass. 1959. Erfah- rungen mit Haien. Z. Tierpsych. 16(6) : 733— 746, 13 pis. Gilbert, P. W., L. P. Schultz, and S. Springer. I960. Shark attacks during 1959. Science 132(3423): 323-326, 1 fig. Harry, R. R. 1953. Ichthyological Field Data of Rarioa Atoll, Tuamotu Archipelago. Atoll Research Bulletin 18: 48. Tester, A. L. I960. Fatal shark attack, Oahu, Hawaii, December 13, 1958. Pacific Sci. 14(2): 181-184, 1 chart. Pandanus pistillaris in the Caroline Islands: An Example of Long-Range Oceanic Dispersal Benjamin C. Stone 1 In CONJUNCTION with the current revision of the genus Pandanus (St. John, I960; St. John and Stone, in sched.) it seems appropriate to mention the factors of dispersal operative in the genus. Since a number of species of Pan- danus are littoral in habitat, and because at least one (if not several) species is found on nearly every tropical atoll in the Pacific, it has rightly been assumed that oceanic dispersal of the buoy- ant fruits is responsible for the distribution. There is no question that throughout vast island groups in Micronesia, Melanesia, and Polynesia Pandanus has long been part of the indigenous vegetation. There are, in fact, some indications that oceanic dispersal accounts for the farthest limits of the genus, in West Africa at one ex- treme, Polynesia at the other, the Bonin Islands in the north, and Australia in the south. The actual number of species involved is not yet known, but species of section Pandanus ( section Keura of earlier authors; see St. John, I960) play a prominent role, especially in the Pacific region. Certain species of other sections of the genus also are primarily ocean-distributed (for example, Pandanus duhius Spreng., and Panda- nus tetrodon ( Gaud. ) Balf i. ) . In general, fruits of these species which are buoyant in seawater exhibit anatomical structures which are pre- sumably adaptations to permit, or prolong, flo- tation. Brown (1931) has discussed the tissues present in fruits of certain Polynesian species, and uses the apt term aerenchyma to designate the light, pithy, upper mesocarp of such fruits. In addition, the seeds are encased in a highly 1 Department of Botany, U. S. National Museum, Smithsonian Institution, Washington 25, D. C. This work was carried out while the author was Research Assistant (under N. S. F. Grant 1834) at the Univer- sity of Hawaii, Dr. H. St. John, principal investigator. Grateful acknowledgment is expressed to Dr. St. John for his aid, and to Dr. M. S. Doty and Dr. T. F. Aus- tin, of the University of Hawaii, and Y. Sinoto of the B. P. Bishop Museum, for their criticism of this con- tribution. Manuscript received January 9, 1961. sclerified osseous endocarp which, except for the minute apertures at either end, are nearly imper- vious to water. Fruits of this type are capable of remaining afloat for a considerable period. No experiments or observations seem to be availa- ble, but it would appear quite probable that a period of 6 months or 1 year in the ocean would not be an exaggerated estimate of the time in which floating fruits could be dispersed by currents. The vast areas encompassed by some of these species or groups of species with buoyant fruits, and the considerable taxonomic difficulties en- countered in the groups, have not heretofore permitted any definite observations to be made with respect to direction of dispersal. Over a period of many years, and because of numerous observations, the knowledge of the currents in the Pacific is now in a position to be of value in problems of oceanic dispersal; but when grave doubt exists as to the nature (and number) of entities involved, little can be said except the obvious, that Pandanus species of littoral areas are probably ocean dispersed. The specimen discussed below, however, be- cause of its unique character, is a reasonably ac- curate indication of one route of oceanic disper- sal. Because it represents a species which is a member of section Intraob tutus St. John, it is at once distinguishable from the widely dispersed members of section Pandanus . The specimen was collected by Dr. Ryozo Kanehira, late professor at Kyushu University, Fukuoka, Japan; the label reads, 'Mokil Atoll; March 12, 1937, Kanehira no. 4203; nom. ver- nac. "arowan.” Mokil Atoll is an isolated atoll roughly 110 mi. east of Ponape, at 6° 40' N., 159° 45' E., consisting of three islands on a cir- cular reef. In April 1957 the present writer visited Mokil, spending the day of the 16th bo- tanizing on the major islet. During this period, 15 numbers of Pandanus were collected (all of section Pandanus) and many vernacular names 610 Oceanic Dispersal of Pandanus ■ — STONE of plants. The name mentioned by Kanehira "arowan,” is a general term used to designate any wild (i.e., nonedible) Pandanus or its fruit. The word is apparently of Marshallese origin, where the preferred spelling is "edwaan” ( the d rolled like an rr) . The other two islets of Mokil were not seen, so there is no corroborative infor- mation regarding the Kanehira collection. There seems no reason, however, to doubt Kanehira’s data, and it is assumed that the specimen is in- deed from Mokil. The specimen consists of several phalanges, or keys, of the fruit. The configuration of the carpels which, fused, make up each phalange, determine the section of the genus, and the pe- culiar "focussed” pattern, as pointed out by St. John (I960) in his discussion of section Intra- oh tutus, is apparent. Although somewhat smaller than Martelli’s description and figure indicate, the Kanehira specimen appears to be referable to Pandanus pistillaris Martelli. This specks,, and in fact the section itself, is known only from Melanesia, specifically, from the Bismarck Archi- pelago. Similar species, and a closely related sec- tion ( Lateriobtutus St. John) occur in New Guinea. In December 1957 the author observed spe- cies of these sections — in particular, P. pistil- laris — in their native habitats in New Ireland and New Britain. The trees are characteristic of lowland areas, although they may occur at alti- tudes of 300 m. or more. Frequently they are found along streams, and not uncommonly, near the coast, especially (as near Kavieng, New Ire- land) on limestone. Phalanges which had fallen from these trees were occasionally seen as drift along the stream estuaries and along adjacent beaches around Kavieng and northwestern New Ireland. In this area, a considerable variation in fruit size was observed, much more than is indi- cated by Martelli’s description; because of this, and the above factors, the Kanehira specimen, without much hesitation, can be placed as P. pistillaris. If this is established, it is interesting to con- sider the possibilities for drift. One obstacle is our meagre knowledge of the full, natural range of P. pistillaris; there seems, however, to be lit- tle question that, if it is not actually endemic in the Bismarck Archipelago, it is restricted to that 611 general area. From a study of the excellent charts provided by Schott (1939) and by Wyrtki (1957), it may be seen that a strip about 5 to 7° north of the equator and east of the 140° meridian, during both seasons of the year, is characterized by a more or less constant eastward flow, which in northern winter emanates from the Mindanao-Morotai region, and in northern summer emanates also from a strong current moving westward and slanting northward along the upper edge of the Melanesian area, but re- versing its course in the Mindanao-Morotai re- gion. The northern summer current spans a wider course along the equator, reaching nearly all of the Caroline Islands ( except perhaps Palau and Yap) and proceeding possibly beyond the Marshalls. During this season, the route of a drifting object starting from New Britain or in that vicinity would be westward along the north- ern coast of New Guinea as far as Morotai, thereafter more or less directly eastward through Micronesia. During northern winter, counter currents forming a strong oceanic stream along the northern coast of New Guinea would pre- vent such a route of drift, while several minor whorl systems in the area bounded by the Ad- miralty Islands on the west and by the Solomons on the east would make the drift pattern irreg- ular and unpredictable, possibly even somewhat southward. The actual number of miles and the elapsed time at sea must remain a matter of con- jecture. No information is available as to the length of time over which floating phalanges may be viable; but indirect evidence (cf. Ed- mondson, 1941, for data on viability of coco- nuts after ocean flotation) appears to indicate that distances of 1,000 mi. or more are possible. Thus it may be seen that the route of drift of an object afloat near the Bismarck Archipelago might, by the devious route outlined, terminate in Mokil Atoll or others of the Caroline Islands. Corroboration of such a route is available. Both Riesenberg ( 1959) and Sinoto ( I960) have re- ported New Guinea canoe prows washed ashore in the Marshall Islands. Riesenberg (personal communication) has photographs of an entire canoe in the Marshalls which apparently drifted along this same route. Sinoto describes a New Guinea canoe prow which drifted to Nukuoro Atoll. 612 PACIFIC SCIENCE, Vol. XV, October 1961 FIG. 1 . Pandanus pistillaris Martelli. Phalanges in lateral view, natural size, and in top view. Drawn from Kanehira 4203 from Mokil Atoll, Caroline Islands. Del. JS.'S.T 4 &u. Oceanic Dispersal of Pandanus — Stone It is postulated that the collection of Panda- nus pistillaris in Mokil Atoll is a further exam- ple of such a route of drift. The Kanehira speci- mens (two of the several phalanges) are here illustrated ( Fig. 1 ) . REFERENCES Brown, F. B. H. 1931. Flora of south-eastern Polynesia, I. Monocotyledons. Bishop Mus. Bull. 84: 1-194, pis. 1-35. Edmondson, C. H. 1941. Viability of coconut seeds after floating in the sea. Occas. Pap. Bishop Mus. 16(12): 293-304. Martelli, U. 1914. Le Specie e varieta nova di Pandanus menzionate nelle Enumerazione delle Pandanaceae. Webbia 4(2): 399-435, pis. 1-43. 613 Riesenberg, S. H. 1959. A New Guinea canoe prow found in the Marshall Islands. Jour. Polynesian Soc. 68(1): 45-46. St. John, H. I960. Revision of the genus Pan- danus Stickman, I. Key to the sections. Pacific Sci. 14(3): 224-241. Schott, G. 1935. Geographic des Indischen und Stillen Ozeans. C. Boysen Verlag, Hamburg. 413 pp., 37 table-maps, 1 folding map. 1939. Die aquatorialen Stromungen des westlichen Stillen Ozeans. Annal. Hydrogra- phie und Maritimen Meterologie, for May, pp. 247-257, tables 26-27. Sinoto, Y. 1959. Drifting canoe prows. Jour. Polynesian Soc. 68(4): 354-355. Wyrtki, K. 1957. Die Zirkulation an der Ober flache der siidostasiatischen Gewasser. Deutsch. Hydrographisch. Zeitschrift 10(1) : 1-3, tables 1-2. Index to Volume XV Author Index Abbott, Donald P. : The Ascidians of Point Barrow, Alaska, Part I. Suborder Phlebobranchia ( EnterOgona) , 1 37 — 143 Alvarino, Angeles: Two New Chaetognaths from the Pacific, 67—77 Alverson, Franklin G. : Daylight Surface Occurrence of Myctophid Fishes Off the Coast of Central America, 483 Armitage, Kenneth B.: Studies of the Biology of Polychoerus carmelensis (Turbellaria: Acoela), 203—210 Bergman, R. A. : The Anatomy of Coluber radiatus and Coluber melanurus, 144—154 Bergquist, Patricia R. : A Collection of Porifera from Northern New Zea- land, with Descriptions of Seventeen New Spe- cies, 33-48 Bieri, Robert: Post-larval Food of the Pelagic Coelenterate, Vel- ella lata, 553-556 Brownlie, G. : Studies on Pacific Ferns. Part III. The Lindsaeoid Ferns, 64—66 Part IV. The Pteridophyte Flora of Pitcairn Is- land, 297-300 Chakravarti, Diptiman and Ronald Eisler : Strontium-90 and Gross Beta Activity in the Fat and Nonfat Fractions of the Liver of the Coconut Crab ( Birgus latro) Collected at Rongelap Atoll during March 1958, 155-159 Chapman, V. J. : A Contribution to the Ecology of the Kermadec Islands, 347-351 Cooke, William Bridge: Fungi from Raroia in the Tuamotu Archipelago, 186-188 Dawson, E. Yale: A Guide to the Literature and Distributions of Pacific Benthic Algae from Alaska to the Gala- pagos Islands, 370—461 Doty, Maxwell S. : Acanthophora, a Possible Invader of the Marine Flora of Hawaii, 547-552 Doty, Maxwell S., and Isabella A. Abbott: Studies on the Helminthocladiaceae (Rhodophyta) : Helminth ocladia, 56— 63 Dumbleton, L. J. : The Aleyrodidae ( Hemiptera-Homoptera ) of New Caledonia, 114-136 Grey, Marion: Fishes Killed by the 1950 Eruption of Mauna Loa, Part V. Gonostomatidae, 462-476 Flora, Charles J.: The Species Commonality Index: A Method for Comparing Habitats, 307-308 Hartman, Olga: New Pogonophora from the Eastern Pacific Ocean, 542-546 Hill, Peter J. R., and Benjamin C. Stone: The Vegetation of Yanagi Islet, Truk, Caroline Islands, 561-562 Hobson, E. S., F. Mautin, and E. S. Reeve: Two Shark Incidents at Eniwetok Atoll, Marshall Islands, 605-609 Jameson, E. W., Jr.: Relationships of the Red-backed Voles of Japan, 594-604 Johnson, Martin W. : On Zooplankton of Some Arctic Coastal Lagoons of Northwestern Alaska, with Description of a New Species of Eurytemora, 311-323 Kanehiro, Yoshinori, and Lynn D. Whittig: Amorphous Mineral Colloids of Soils of the Pacific Region and Adjacent Areas, 477-482 Kay, Alison: A New Opisthobranch Mollusc from Hawaii, 112— 113 King, Judith E., and R. J. Harrison: Some Notes on the Hawaiian Monk Seal, 282-293 KLAWE, W. L. : Notes on Larvae, Juveniles, and Spawning of Bonito ( Sarda ) from the Eastern Pacific Ocean, 487-493 Kohn, Alan J.: Studies in Spawning Behavior, Egg Masses, and Larval Development in the Gastropod Genus Conus, Part I. Observations on Nine Species in Hawaii, 163—179 DE LAUBENFELS, M. W. : Porifera of Friday Harbor and Vicinity, 192-202 Macdonald, Gordon A., and Takashi Katsura : Variations in the Lava of the 1959 Eruption in Kilauea Iki, 358-369 Martin, W. E. : Life Cycle of Mesostephanus appendiculatus (Ciurea, 1916) Lutz, 1935 (Trematoda: Cyathocotyli- dae), 278-281 Plucknett, D. L., and B. C. Stone: The Principal Weedy Melastomaceae in Hawaii* 301-303 615 616 PACIFIC SCIENCE, Vol. XV, October 1961 Noffsinger, T. L.: Leaf and Air Temperature under Hawaiian Condi- tions, 304-306 Randall, John E. : A Contribution to the Biology of the Convict Sur- geonfish of the Hawaiian Islands, Acanthurus triostegus sandvicensis, 215—272 Reish, Donald J.: A New Species of Micronereis (Annelida, Poly- chaeta) from the Marshall Islands, 273-277 Roe, Sister Margaret James : A Taxonomic Study of the Indigenous Hawaiian Species of the Genus Hibiscus , 3-32 Sakagami, Shoichi F. : An Ecological Perspective of Marcus Island, with Special Reference to Land Animals, 82-104 Scagel, Robert F. : The Distribution of Certain Benthonic Algae in Queen Charlotte Strait, British Columbia, in Re- lation to Some Environmental Factors, 494—539 Schafer, Rita D. : Effects of Pollution on the Free Amino Acid Con- tent of Two Marine Invertebrates, 49—55 St. John, Harold: Revision of the Genus Pandanus Stickman. Part 2. Pandanus in Western Australia and Notes on the Section Foullioya, 180—185 Part 3. A New Species from Oeno Island, Tua- motu Archipelago, 324-326 Part 4. Disposition of Some Later Homonyms, 327 Part 5. Pandanus of the Maidive Islands and the Seychelles Islands, Indian Ocean, 328-346 Part 6. New Pandanus Species from Queensland, Australia, 563-575 Part 7. New Species from Borneo, Papua, and the Solomon Islands, 576-590 Subject Aaptos aaptos, 44 Acanthophora spicifera, 547-551 Acanthurus triostegus sandvicensis , 215—272 Adiantum hispidulum, 298 Adocia parietaloides, n. sp., 37 venustina, n. sp., 37 Aleurocanthus brevispinosus, n. sp., 115—116 multispinosus, n. sp., 116—118 nudus, n. sp., 118 spinothorax, n. sp., 118-120 Aleyrodidae of New Caledonia, 114-136 algae, benthic, of Pacific, guide to literature of, 370- 561 algae, benthonic, in Queen Charlotte Strait, 494-539 amino acid content of two marine invertebrates, 49-55 Angiopteris chauliodonta, 297 Anthostomella sp., 186 Stone, Benjamin C. : Pandanus pistillaris in the Caroline Islands: An Example of Long-range Oceanic Dispersal, 610— 613 Sund, Paul N. : Two New Species of Chaetognatha from the Waters off Peru, 105-111 Uchida, Richard N. : Hermaphroditic Skipjack, 294-296 Watson, J. S.: Feral Rabbit Populations on Pacific Islands, 591 — 593 Wells, John W.: Notes on Indo- Pacific Scleractinian Corals, Part 3= A New Reef Coral from New Caledonia, 189— 191 Wentworth, G K., H. A. Powers, and J, P. Eaton: Feasibility of a Lava-diverting Barrier at Hilo, Hawaii, 352-357 Winkler, Lindsay R. : Preliminary Tests of the Toxin Extracted from Cali- fornia Sea Hares of the Genus Aplysia , 211-214 Winkler, Lindsay R., and Bernard E. Tilton: Interspecific Differences in the Reaction to Atropine and in the Histology of the Esophagi of the Common California Sea Hares of the Genus Aplysia , 557-560 Woodland, D. J. : Description of a New Species of Pranesus (Ather- inidae: Pisces) from the Capricorn Group, Great Barrier Reef, 540-541 Wood wick, Keith H„ : Polydorarickettsi, a New Species of Spionid Poly- chaete from Lower California, 78-81 Index Aplysia calif ornica, 211—214; 557-560 vac carta, 211-214; 557-560 Araiophos, n. gen., 463-465 gracilis , n. sp., 465-467 Argyripnus atlanticus, 468-470 ephippiatus, 471 irridescens, 473 sp., 473 Arthressa evansi, n. sp., 112-113 Arthrostemma latifolium, 301, 303 Ascidia callosa , 137—139 ascidians of Alaska, 137—143 Asplenium nidus, 299 obtusatum, 299 Athyrium polyantbes, 299 atropine, effect of, upon Aplysia spp., 557— 560 Auricularia mesenterica, 186 Axiamnon ere eta, 41 Index to Volume XV 617 Bantamia merleti, n. sp., 189—191 Bemisia cordylinidis, n. sp., 120—121 Benthos ema pterota, 483 beta activity in coconut crab, 155—159 Biemna rhabderemioides , n. sp., 40 Birgus latro, 155—158 bonito, notes on larvae, juveniles, and spawning, 487— 493 Borneo, new species of Pandanus from, 576-590 Burtonanchora lacunosa, 195—197 Callyspongia ramosa, 36 Carcharhinus menisorrah, 605— 609 Centropages abdominalis, 317 Cerithidea calif ornica, 278—281 Chelysoma inaequale, 140-142 macleayanum, 139—140 Choanites sub ere a var. lata, 201 Clethrionomys, 595-597 andersoni, 599 imaizumii, n. sp., 599-600 nii gat ae, 599 rufocanus bedfordiae, 597 rutilus mikado, 600 Clidemia hirta, 301, 303 Cliona celata, 44, 201 muscoides, 44 Coluber melanurus, 152—154 radiatus, 144-152 Conus abbreviatus, 164-166 catus, 166-169 imperialis, 169 leopardus, 169—170 lividus, 170-171 pennaceus, 171—173 quercinus, 173-174 rattus, 174-175 vitulinus, 175-176 corals, scleractinian, from Indo-Pacific, 189—191 Corticella novae-zealandiae, n. sp., 45 Cyathea cumingii, 298 Cyclosorus parasiticus, 299 Cyclothone sp., 463 Danaphos oculatus, 467-468 Davallia solida, 298 Dialeurodes dothioensis, n. sp., 121-122 dumbeaensis, n. sp., 122—123 psychotriae, n. sp., 123-124 Dicranopteris linearis, 297 Doodia media, 299 Dothioia, n. gen., 124 bidentatus, n. sp., 124-126 Dysidea cristagalli, n. sp., 33 Ectyodoryx parasitica, 197 Eniwetok Atoll, shark incidents at, 605-609 Eurytemora foveola, n. sp., 317-321 pacifica, 317 Euthenomys kageus, 601 smithi, 600—601 Foullioya, 184-185 Galaxea fascicularis, 191 Gomenella, n. gen., 126 multipora, n. sp., 126-127 reflexa, n. sp., 127-129 Gonostoma atlanticum, 462-463 Great Barrier Reef, new Pranesus from, 540-541 Halichondria moorei, n. sp., 40 panicea, 41 Haliclona clathrata, 35 glabra, n. sp., 35 heterofibrosa, 35 isodictyale, n. sp., 34 permollis, 193-194 petrosioides, 36 tenacior, n. sp., 34 Haliotis cracherodii, 49—55 Hawaiian monk seal, 282-293 Helicomyces roseus, 187 Helicosporium lumbricoides, 187 Helix aspersa, 211-214 H elminth o cladia rhizoidea, n. sp., 60— 63 simplex, n. sp., 56-60 Heterocentron subtriplinervium, 301, 303 Hibiscus, taxonomic study of indigenous Hawaiian species, 3-32 Hibiscus Arnottianus, 26-30 f. parviflora, 30-31 f. punaluuensis , 31 Brackenridgei, 9-11 var. kauaiana, 12 var. mokuleiana, n. var., 12-13 var. molokaiana, 11—12 immaculatus, n. sp., 22—24 kahili, 21 Kokio, 14-17 var. pukoonis, 17—18 Newhousei, n. sp., 22 Saintjohnianus, n. sp., 18-21 waimeae, 24—26 Youngianus, 6—8 Hymeniacidon perlevis, 41 Hypoxylon chrysoconium, 186 investiens, 186 lophon chelifer calif orniana, 198 Ircinia novae zealandiae, n. sp., 33 Isociella incrustans, n. sp., 42 Japan, red-backed voles of, 594-604 Katsuwonus pelamis, 294-296 Kermadec Islands, ecology of, 347-351 Kilauea Iki, eruption of 1959, 358-369 lava-diverting barrier, feasibility of, 352-357 Leucopogonella, n. gen., 129 apectenata, n. sp., 129-130 pallida, n. sp., 130-131 618 simila, n. sp., 131-132 sinuata, n. sp., 132 Leucosolenia nautilia, 192 Limnocalanus grimaldi, 316—317 johanseni, 317 Lindsaea dimorpha, 64 lapeyrousii, 64 nervosa, 65 Lissodendorynx firma, 194—195 noxiosa, 195 Loxoscaphe gibberosum, 299 Maidive Islands, Pandanus of, 328-346 Marcus Island, ecological perspective of, 82-104 Medinilla magnifca, 301 Melastoma decemfidum, 301—303 malabathricum, 301-303 Me so stephanus appendiculatus, 278-281 Microciona coccinea, n. sp., 38 heterospiculata, 38 rubens, n. sp., 38 Micronereis eniwetokensis, n. sp., 273-277 Monachus schauinslandi, 282-293 Mycale adhaerens, 198-199 rara, 39 myctyophid fishes, daylight surface occurrence of, 483 Nephrolepis bis err at a, 298 hirsutula, 298 new combination: Phymatodes pitcairnense, 300 Schizoloma prolongata , 65 new genus: Araiophos, 463—465 Dothioia, 124 Gomenella, 126 Leucopogonella, 129 new name: Pandanus glauciferus, 327 New species: Adocia parietalioides, 37 venustina, 37 Aleurocanthus brevispinosus, 115-116 multispinosus, 116-118 nudus, 118 spinithorax, 118-120 Araiophos gracilis, 465—467 Art hr ess a evansi, 112—113 Bantamia merleti, 189—191 Bemisia cordylinidis , 120-121 Biemna rhabderemioides , 40 Clethrionomys imaizumii, 599-600 Corticella novae-zealandiae , 45 Dialeurodes dothioensis, 121-122 dumbeaensis, 122—123 psychotriae, 123-124 Dysidea cristagalli, 33 Eurytemora foveola, 3 1 7-3 2 1 Gomenella multipora, 126—127 reflexa, 127-129 PACIFIC SCIENCE, Vol. XV, October 1961 Halichondria moorei, 40 Haliclona glabra, 35 isodictyale, 34 tenacior, 34 Helminthocladia rhizoidea, 60-63 simplex, 56—60 Hibiscus immaculatus, 22-24 Newhousei, 22 Saintjohnianus, 1 8—2 1 Ircinia novae zealandiae, 33 Isociella incrustans, 42 Leucopogonella apectenata, 129-130 pallida, 130-131 simila, 131-132 sinuata, 132 Microciona coccinea, 38 rubens, 38 Micronereis eniwetokensis, 273-277 Orchamoplatus perdentatus, 133-134 Pandanus adduensis, 328-331 Andersonii, 576—579 convexus, 180—182 feruliferus, 324—32 6 Fosbergii, 331—333 fruticosus, 579 Hartmanii, 333-335 Karikayo, 335-339 kimberleyanus, 182—184 magnicavernosus , 587—590 maldivicus, 339—341 oblatus, 569-570 ruber, 579-583 somersetensis, 570—573 sphaericus, 563— 565 truncatus, 573-575 yuleensis, 583-587 Parabemisia reticulata, 134-135 Phorbas intermedia, 36 Polydora rickettsi, 7 8-8 1 Polymastia fusca, 43 Pranesus capricornus , 540—541 Rhabderemia stellata, 41 Sagitta bierii, 67-71 euneritica, 12—16 peruviana, 105—108 popovicii, 108 Siboglinum veleronis, 542—546 Stylissa stipitata, 199—200 Suberites cupuloides, 42 Syringella amphispicula, 200 T etraleurodes submar ginata, 135-136 new variety: Hibiscus Brackenridgei var. mokuleiana, 12-13 oceanic dispersal of Pandanus pistillaris, 610—613 Oidium curtisii, 187 Ophlitaspongia pennata, 198 Orchamoplatus perdentatus, n. sp., 133-134 Oryctolagus cuniculus, 591-593 Oxyspora paniculata, 301 Index to Volume XV 619 Pachygrapsus crassipes, 49-55 Pandanus adduensis, n. sp., 328-331 Andersonii, n. sp., 576-579 Balfourii, 341 compressus, 327 convexus, n. sp., 182-184 ferromontanus, n. sp., 565—569 feruliferus, n. sp., 324—326 Fosbergii, n. sp., 331-333 fruticosus, n. sp., 579 glauciferus, nom. nov., 327 Hartmanii, n. sp., 333-335 Hornet, 341-345 Karikayo, n. sp., 335-339 kimber ley anus, n. sp., 180-182 magnicavernosus, n. sp., 587—590 maldivecus, n. sp., 339—341 multispicatus, 346 oblatus, n. sp., 569-570 pis t Maris, 610-613 ruber, n. sp., 579-583 secheilamm, 346 somersetensis, n. sp., 570-573 sphaericus, n. sp., 563—565 truncatus, n. sp., 573—575 yuleensis, n. sp., 583-587 Phorbas intermedia, n. sp., 36 Phymatodes pitcairnense, n. comb., 300 Pitcairn Island, pteridophyte flora of, 297-300 Plakina monolopha, 47 trilopha, 47 Polychoerus carmelensis, 203-210 Polydora rickettsi, n. sp., 78-81 Polymastia fusca, n. sp., 43 granulosa, 44 Porifera from northern New Zealand, 33—48 Pranesus capricornus, n. sp., 540—541 Psilotum nudum, 297 Pterolepis glomerata, 301, 303 Pyrrosia an gust at a, 299 Queen Charlotte Strait, benthonic algae in, 494-539 Queensland, Australia, new species of Pandanus from, 563-575 rabbits, feral, populations on Pacific Islands, 591-593 Rbabderemia stellata, n. sp., 41 Raroia, fungi from, 186-188 Rumohra aristata, 298 Sagitta bierii, n. sp., 67—71 euneritica, n. sp., 12—16 peruviana, n. sp., 105-108 popovicii, n. sp., 108 Sarda chiliensis, 487-493 orientalis, 487—493 Scbizoloma prolongata, n. comb., 65 Scbizophyllum commune, 187 Scypha raphana, 192-193 Seychelles Islands, Pandanus of, 328-346 shark incidents, at Eniwetok Atoll, 605-609 Siboglinum veleronis, n. sp., 542—546 Sigmadocia edapba, 194 soils, Pacific, and amorphous mineral colloids, 477— 482 Solomon Islands, new species of Pandanus from, 576- 590 species commonality index, 307—308 Sphenomeris chusana, 66 Spongia reticulata, 33 strontium-90 in coconut crab, 155-159 Stylissa stipitata, n. sp., 199—200 Suberites cupuloides, n. sp., 42 Syringella amphispicula, n. sp., 200 Tapeinidium flavicans, 65 Tedania fragilis, 197 T ' etraleurodes submarginata, n. sp., 135-136 Tetrazygia bicolor, 301 Thelypteris uliginosa, 298 Tibouchina semidecandra, 301-303 Toxadocia toxophorus, 37 Trichomonas endlicherianum, 297-298 Velella lata, 553-55 6 Vittaria elongata, 300 voles, red-backed, of Japan, 594-604 Xestospongia vanilla, 194 Yanagi Islet, Truk, vegetation of, 561-562 should not be fastened together in any way, and should be mailed flat. Inserts should be either typed on separate sheets or pasted on proper page, and point of insertion should be clearly indicated. Original copy and one carbon copy of manuscript should be submitted. The author should retain a car- bon copy. Although due care will be taken, the editors cannot be responsible for loss of manuscripts. Introduction and Summary. It is desirable to state the purpose and scope of the paper in an introductory paragraph and to give a summary of results at the end of the paper. Dictionary Style. It is recommended that authors fol- low capitalization, spelling, compounding, abbrevia- tions, etc., given in Webster’s New International Dic- tionary (unabridged), second edition; or, if desired, the Oxford Dictionary. Abbreviations of titles of pub- lications should, if possible, follow those given in World List of Scientific Periodicals. Examples of the usual abbreviations are to be found in Pacific Science. Footnotes. Footnotes should be used sparingly and never for citing references (see later). When used, footnotes should be consecutively numbered by supe- rior figures throughout the body of the paper. Foot- notes should be typed in the body of the manuscript on a line immediately below the citation, and sepa- rated from the text by lines running across the page. Citations of Printed Sources. All references cited should be listed alphabetically by author at the end of the paper, typed double-spaced. References to books and to papers in periodicals should conform to the following models : Batzo, Roderick L., and J. K. Ripkin. 1849. A Treatise on Pacific Gastropods. Rice and Shipley, Boston, vii + 326 pp., 8 figs., 1 map. Crawford, David L. 1920^. New or interesting Psyllidae of the Pacific Coast (Homop.). Proc. Hawaii. Ent. Soc. 4(1) : 12-14. 1920A Cerotrioza (Psyllidae, Homoptera). Proc. Hawaii. Ent. Soc. 4(2) : 374-375. Rock, Joseph F. 1916. The sandalwoods of Hawaii; a revison of the Hawaiian species of the genus Santalum. J. Aust. Inst. Agric. Sci. 12(3): 1-43, 13 pis. In the text, sources should be referred to by author, date, and page, as follows: "It was noted (Rock, 1916: 18) that . . .” or "Rock (1916: 21-24) says . . .” Quotations. Quoted matter of fewer than five printed lines (about 200 characters) should be given in the text in the usual form, using double quote marks. Longer quotations should be set flush with left mar- gin. The author is responsible for the accuracy of quoted material. Numbers. Decimals, measurements, money, percent- ages, time; enumerations in which any figure is 10 or over; and isolated enumerations of 10 and over should be given in Arabic figures, rather than spelled out, except when the number begins a sentence. ILLUSTRATIVE MATTER Only the minimum number of illustrations required to supplement the text will be accepted by the editors. Reproduction costs of illustrations in excess of the number allowed by the editors will be paid by the author. Illustrations should accompany manuscript, on sep- arate sheets. Often more valuable than a photograph is a good line drawing, which may be reproduced by the zinc etching process. Figures and Graphs. Copy for figures and graphs should always be drawn large enough to allow for at least one-third reduction by the engraver. Copy should consist of carefully prepared line drawings in one color only, drawn in India ink on plain white draw- ing paper or tracing cloth. Co-ordinate paper with lines in light blue (a color which will not show in a photograph) may be used; but co-ordinates which should appear in the finished graph must be drawn in India ink. If original figures may not be conven- iently submitted with manuscript, duplicate rough sketches or photographic prints may be furnished to aid the editors in their decisions. It is strongly urged that an indication of scale be incorporated as a part of all drawings in which mag- nification and size are critical considerations. Photographs. Photographs should be chosen for clarity in portraying essential information. They should be printed for contrast, on glossy paper, and should be sent unmounted. They should be identified with serial number written in soft pencil on the back to corre- spond with list of captions. Illustrations will be returned to the author. Tables. Tabular matter should be kept to a minimum. Each table, prepared to conform with Pacific Science style, should be typed on a separate page, and its posi- tion indicated on the manuscript. Captions. Readily identifiable captions for figures, graphs, photographs, and other illustrative matter should be supplied on a separate page. PROOF Proof should be corrected immediately and returned at once to Robert Sparks, assistant to the editors. Authors are reminded that the editors will allow only a minimum number of corrections on galley proof. Additions to the printed text and changes in style and content are not permitted. All queries on proof should be answered. If cor- rected proof is not received within four weeks after being sent to the author, author’s changes cannot be accepted. REPRINTS Reprints or separates should be ordered on the form provided and returned with author’s proof. All correspondence concerning separates must be directed to the printer, Star-Bulletin Printing Company, Inc., 420 Ward Avenue, Honolulu 14, Hawaii. / '©• m J* U ■ : V : wtm a, j? 0S$Wk % $ v < c ,> a ll&jtJtif s> 4 jl ^ *• , %-MS w . '$' % W?M M ^ \WI igSSKffi® r^pp^'i §§5fia$S$# BS&'&p JkkS$ © %^ ipsl Itlll flflj tefa SlftiSa tSPv $•. sn Isil v s) Hr jiii M