Ecotoxicology (2008) 17:69–81

DOI 10.1007/s10646-007-0168-7

Adverse effects from environmental mercury loads on breeding

common loons
David C. Evers Æ Lucas J. Savoy Æ Christopher R. DeSorbo Æ David E. Yates Æ
William Hanson Æ Kate M. Taylor Æ Lori S. Siegel Æ John H. Cooley Jr Æ
Michael S. Bank Æ Andrew Major Æ Kenneth Munney Æ Barry F. Mower Æ
Harry S. Vogel Æ Nina Schoch Æ Mark Pokras Æ Morgan W. Goodale Æ
Jeff Fair

Accepted: 13 August 2007 / Published online: 2 October 2007

Springer Science+Business Media, LLC 2007

Abstract Anthropogenic inputs of mercury (Hg) into the thresholds for adult loons at 3.0 ug/g (wet weight) in blood
environment have significantly increased in the past century. and 40.0 ug/g (fresh weight) in feathers. Mercury contami-
Concurrently, the availability of methylmercury (MeHg) in nation in parts of Maine and New Hampshire is a driving
aquatic systems has increased to levels posing risks to eco- stressor for creating breeding population sinks. Standardized
logical and human health. We use the common loon (Gavia monitoring programs are needed to determine if population
immer) as an upper trophic level bioindicator of aquatic Hg sinks occur elsewhere and to track aquatic ecosystem
toxicity in freshwater lakes. Multiple endpoints were responses to changes in Hg emissions and deposition.
selected to measure potential negative impacts from MeHg
body burdens on behavior, physiology, survival and repro- Keywords Mercury
Common loon
Population sink

ductive success. A robust spatio-temporal dataset was used Adverse effects
Behavior
that included nearly 5,500 loon Hg measurements over an
18-year period. We measured significant changes related to
elevated MeHg body burdens, including aberrant incubation Introduction
behavior, lethargy, and wing area asymmetry. Mercury body
burdens in adult loons increased an average of 8.4% per Mercury (Hg) toxicity to wildlife is widely documented
year. Increasing Hg body burdens reduced the number of (Thompson 1996; Wolfe et al. 1998; Scheuhammer et al.
fledged chicks per territorial pair, with highest risk loons 2007). The atmospheric deposition of Hg is an important
producing 41% fewer fledged young than our reference contributor to declining local (Frederick et al. 2004) and
group. Our multiple endpoints establish adverse effect global bird populations (Braune et al. 2006). Waterborne

D. C. Evers (&)
L. J. Savoy
C. R. DeSorbo
A. Major
K. Munney

D. E. Yates
K. M. Taylor
L. S. Siegel
M. W. Goodale U.S. Fish and Wildlife Service, Concord, NH, USA
BioDiversity Research Institute, 19 Flaggy Meadow Road,
Gorham, ME 04038, USA B. F. Mower
e-mail: [email protected] Maine Department of Environmental Protection,
Augusta, ME, USA
W. Hanson
FPL Energy Maine Hydro, Waterville, ME, USA N. Schoch
Wildlife Conservation Society’s Adirondack Program,
K. M. Taylor
J. H. Cooley Jr
H. S. Vogel New York, USA
Loon Preservation Committee, Moultonborough, NH, USA
M. Pokras
L. S. Siegel Tufts University, North Grafton,
Siegel Environmental Dynamics, LLC, Hanover, NH, USA MA, USA

M. S. Bank J. Fair
Department of Environmental Health, Harvard University, Fairwinds Wildlife Services, Palmer,
Boston, MA, USA AK, USA

123
70 D. C. Evers et al.

Hg from legacy sources such as mining (Schwarzbach et al. Negative impacts of MeHg exposure on reproductive
2006), chlor-alkali plants (Barr 1986), and manufacturing success and survival of adult loons are of ultimate concern.
facilities (Brasso and Cristol, this issue) have also been There are inherent difficulties in relating Hg exposure to
shown to negatively impact breeding success. However, loon reproductive success that relate to both intrinsic
there are few studies that have directly related adverse (e.g., density dependence, species longevity) and extrinsic
impacts from Hg on reproductive success in free-living bird (e.g., weather, habitat quality) factors as well as anthro-
populations. Long-lived obligate piscivores such as the pogenic stressors (e.g., recreational disturbances and other
common loon (Gavia immer) are considered important bi- contaminants). Still, there are studies that have identified
oindicators of environmental Hg loads in North America Hg as a cause of reduced reproductive success in loons.
(Evers 2006; Wolfe et al. 2007), because of the ability of Barr (1986) found whole body Hg levels in prey fish of
methylmercury (MeHg) to biomagnify through the food- 0.30 ug/g (ww) resulted in reproductive impairments on
chain (Cabana and Rasmussen 1994; Atwell et al. 1998; breeding loons, and no loon reproduction when Hg in fish
Burgess and Hobson 2006) and bioaccumulate over an exceeded 0.40 ug/g (ww). More recently, Burgess and
individual’s lifetime (Evers et al. 1998). Meyer (this issue) found a significant negative correlation
Numerous studies of Hg concentrations in loon blood, between Hg in fish and loon productivity with 50% fewer
feathers (Meyer et al. 1998; Evers et al. 1998; Burgess fledged young in breeding loons foraging on fish averaging
et al. 2005; Champoux et al. 2006) and eggs (Scheuham- 0.21 ug/g (ww) and no reproductive success if prey fish
mer et al. 2001; Evers et al. 2003) provide a robust exceeded 0.41 ug/g (ww).
geographic Hg exposure profile. These and other studies In this study, we examine adverse effects of Hg on a
indicate loons breeding in northeastern North America free-living breeding population of common loons inhabit-
have the greatest risk of adverse effects from Hg due to ing lentic ecosystems in western Maine and New
elevated Hg exposure associated with relatively high rural Hampshire using behavioral, physiological, survivorship,
atmospheric deposition (up to 32 ug/m2 per year of wet and and reproductive endpoints based on observational and Hg
dry deposition) (Miller et al. 2005) and enhanced Hg concentration data over an 18-year period.
methylation in sensitive environments (Driscoll et al.
2007). Biological Hg hotspots have been identified in New
York, New England, and Nova Scotia where concentrations Study area
are high enough to result in population-level impacts to
breeding loons (Evers et al. 2007). BioDiversity Research Institute (BRI) has collected blood
Substantial evidence exists for adverse neurological, and tissue samples to analyze Hg concentrations in North
physiological, and reproductive effects to loons associated America’s breeding loons annually from 1989 to 2006
with environmental Hg levels. Neurotoxic effects of (Evers et al. 1998, 2003, 2005). This broad geographic
MeHg in free-living loons include reduced back-riding by sampling effort has resulted in the testing of 5,477 loon
chicks (Nocera and Taylor 1998) and lowered chick- tissues for blood (n = 2,850), feathers (n = 1,589) and eggs
feeding rates by adults (Counard 2000). Recently, changes (n = 1,038) from 1,086 territories on 701 lakes in 13 states
in neurotransmitter receptor concentration and other and four Canadian provinces (Fig. 1). The extensive con-
neurochemical effects have been correlated with brain Hg tinental dataset was used to examine physiological and
concentrations in wild loons (Scheuhammer et al. this survival endpoints and to develop loon Hg equivalent units
issue). Physiological responses to elevated MeHg levels to support comparisons across tissue type and age and sex
include reduced diving frequency of free-living loons classes.
(because Hg inhibits heme production; Olsen et al. 2000) We conducted a high-resolution study on the impacts
and the production of smaller eggs in the wild (Evers of Hg on loon behavior and reproductive success in the
et al. 2003). Sublethal Hg dosing studies in laboratory upper Androscoggin River watershed in New Hampshire
conditions indicate that histological, immunological, and and Maine, the upper Kennebec River watershed in
biochemical biomarkers are effective for detecting Maine, and selected lakes in southeastern New Hamp-
adverse effects on juvenile piscivorous birds (Spalding shire based on our extensive historical database. This
et al. 2000; Henny et al. 2002; Hoffman et al. 2005), geographic area had some of the highest Hg levels
including loons (Kenow et al. 2003, 2007). However, the recorded in North America. Our focal study area inclu-
limitation of low adult loon survival in captivity forces ded 80 lakes with 178 territories that were regularly
laboratory studies to be limited to juvenile loons, an age surveyed to determine reproductive success from 1996 to
class characterized by its ability to depurate large 2005 (Fig. 2). The behavioral observations were con-
amounts of MeHg, especially into growing feathers centrated on 12 lakes with 43 territories from 1998 to
(Fournier et al. 2002). 2000.

123
Mercury effects on loons 71

Fig. 1 General locations of

sampling effort for feather Hg
levels that correspond with
asymmetry and percent change
biomarkers. All locations also
represent blood and/or egg
sampling areas—tissues used
for developing female loon unit
(FLU) equivalents

Sampling efforts in this same geographic area in 1996 to adult blood Hg levels reflect prey Hg levels in the breeding
1997 by the U.S. Fish and Wildlife Service on fish and loons lake (Evers et al. 2004; Burgess and Hobson 2006; Burgess
to determine levels of organochlorines and polychlorinated and Meyer, this issue). Over 95% of the blood Hg is in the
biphenyls (PCBs) found relatively low levels (USFWS, methyl form (Wolfe et al. 2007). The half-life of MeHg in
unpublished data). This finding further supported the suit- the blood for juvenile loons until 12 weeks of age is
ability of this region for studying the effects of Hg on loons \3 days and post-molt is 116 days (Fournier et al. 2002).
and suggested that the results would not be confounded by The half-life for MeHg in adult loon blood is unknown; in
the potential synergistic effects of organic compounds, such adult Cory’s shearwaters (Calonectris diomedea) it is 40–
as those known with PCBs (Newland 2002). 60 days (Monteiro and Furness 2001). We surmise that
adult loon blood Hg levels reflect uptake of dietary Hg from
the breeding lake because the blood Hg levels of adult loons
Methods returning to their breeding lakes in the spring is likely
\1.2 ug/g (based on 105 loons sampled in wintering areas
Tissue choice along the Pacific and Atlantic coasts; BRI unpublished data)
and adult loons were sampled 2–4 months following their
Three standard biological matrices that could be non-leth- return to lakes.
ally collected (blood, eggs, and feathers) were used to Since female blood Hg levels are highly correlated with
identify adverse effect levels based on pre-existing egg Hg levels (Evers et al. 2003), eggs are also pertinent
knowledge of tissue Hg properties for loons and other birds tissues for predicting Hg risk within a breeding territory.
(Evers et al. 2005). Blood is the primary matrix for evalu- Egg collection was opportunistic and reflects eggs that
ating recent dietary uptake and there is strong evidence that were abandoned because of nest inundation by rising water

123
72 D. C. Evers et al.

Fig. 2 Locations of common

loon territories sampled for Hg
and monitored for reproductive
success. Major lakes are
identified

levels, wave action, human disturbance, storms and other bands (described in Evers 2001). We attempted to recap-
events that force incubating adults from normal incubation ture adults to measure changes in Hg body burdens over
behavior. time in known individuals. High annual territory fidelity
Feather Hg levels are a useful matrix for evaluating rates that average 80% permit effective recapture (Evers
chronic body burdens. Feather Hg can represent 70–93% of 2001).
the total body burden of Hg (Burger 1993); therefore there Tissue collection protocols are well established and
can be a chronic bioaccumulation of MeHg because the described (Evers et al. 1998, 2003, 2005). Blood was
entire Hg body burden is not depurated each year, partic- drawn from the metatarsal vein and placed in a cooler with
ularly for high-risk individuals. Feather Hg reflects blood ice packs. Whole blood samples were then placed in a
Hg levels at the time of molt (Bearhop et al. 2000); how- freezer within 8 h of collection and not removed until
ever, if MeHg is sequestered in the muscle tissue (as is the submitted to an analytical lab. Second secondary feathers
case for individual birds with a high dietary Hg uptake), were symmetrically taken by cutting below the superior
then additional MeHg is available through remobilization umbilicus. They were placed in a polyethylene bag and
at a later time. Individual variation in physiological kept in a cool area. Abandoned eggs were placed in a
response to Hg (Bearhop et al. 2000), as well as the broad polyethylene bag and frozen as soon as possible. Later,
differences of the pharmacokinetics among species, com- eggs were measured for standard metrics and placed into
plicates the interpretation of adult feather Hg levels. sterile I-Chem1 jars (Evers et al. 2003).
Mercury results in tissues are reported as wet weight All sample preparation, handling, labeling, and chain-
(ww) in blood and eggs and fresh weight (fw) in feathers. of-custody efforts followed well-established standard
operating procedures established by BRI.

Capture, marking, and sampling

Loon equivalent development
Capture of adult loons followed night-lighting techniques
described by Evers (2001). Blood and feather samples were To best evaluate and utilize existing data from various
non-lethally taken. All individuals were banded with a U.S. biotic compartments, Hg concentrations require a single
Fish and Wildlife Service numbered aluminum band along common unit. Accordingly, Hg concentrations in the var-
with a unique color combination of custom-formed plastic ious loon tissues and age/sex classes (i.e., adult female

123
Mercury effects on loons 73

blood, adult male blood, juvenile blood, and egg) were individuals have FLUs below 1.0 ug Hg/g and (2) our
converted into a female loon unit (FLU) Hg equivalent. results herein document FLUs [3.0 ug Hg/g as having a
Since adult loon weight varied from 3,900 g to 7,550 g, significant negative adverse effect on reproductive success.
and prey size relates to loon size and Hg levels (Evers et al. We collected behavior data using time-activity budget
2004), we also accounted for weight within our FLU (TAB) methods based on those described by Altmann
model. Using our continental dataset, the Hg per weight (1974), Tacha et al. (1985), and Nocera and Taylor (1998).
values for each loon with blood Hg and weight data were Observation periods were not staggered throughout the
calculated and the mean per territory value determined. photoperiod because of the strong evidence from Evers
Using Statgraphics Centurion XV Version 15.2.00, mean (1994), Mager (1995), Gostomski and Evers (1998), and
blood Hg-weight ratios of females were regressed with Paruk (2000) that minimal or no significant relationships
mean ratios of males and of juveniles. The optimum male exist between time of day and loon behaviors.
to female relationship was linear; however, the optimum Individual loons were observed in one-hour time blocks
juvenile to female ratio was a double reciprocal. Initially, for up to 5 h/day using a 15–45· spotting scope and 10·
the regression for juvenile to female levels was poor due to binoculars. Observers continually monitored behavior
extreme outliers. However, eliminating all values of juve- through a spotting scope and relayed behaviors to a
niles for which the Hg to weight ratio for juveniles, divided recorder, who noted times from a digital stopwatch and
by that for females was outside the 99.99% confidence recorded categorized observations on data sheets. In 1998
interval, improved the regression significantly. Multiplying and 1999, data were collected by six BRI biologists and
the regression by the individual’s weight converts a male trained EarthWatch Institute volunteers. Martin and Bate-
and juvenile blood Hg level, respectively, to a female blood son (1993) addressed potential problems with observer bias
Hg level (i.e., FLU) (Eqs. 1 and 2). Additionally, egg and misinterpretation of behaviors. Therefore, observer
mercury levels (Hgegg) were converted to female levels bias was minimized each year by training all BRI biologists
based on the conversion established in Evers et al. (2003) simultaneously for 3–4 days, and meeting to review
(Eq. 3). methods and results several times throughout the season.
EarthWatch Institute volunteers were designated as
FLU ¼ Hgfemale recorders. Bradley (1985) addressed the importance of
¼ ½ðHgmale =Weightmale Þ
0:948 minimizing visibility and discovery bias when collecting
þ 0:000002 Weightmale ; r2 ¼ 0:68 ð1Þ TABs. We addressed these potential biases by concealing
ourselves and/or through remote observation (up to 300 m
distant).
FLU ¼ Hgfemale
Adult behaviors during post-hatching were designated
¼ ½1=1393:220 þ 0:197=ðHgjuvenile = into two energetically-based categories: (1) high energy
Weightjuvenile Þ Weightjuvenile ; r2 ¼ 0:83 ð2Þ behaviors included foraging for chick, foraging for self,
locomotion (swimming and flying), preening and agonistic
FLU ¼ Hgfemale ¼ Hgegg
1:554 þ 0:224; r2 ¼ 0:79 ð3Þ behaviors and (2) low energy behaviors included brooding
and resting (drifting and sleeping).
Hereafter, we use FLU as our standardized approach for
comparison. FLUs are presented on a wet weight (ww)
basis as ug Hg/g of tissue. Asymmetry development

To determine feather asymmetry we cut the second sec-

Behavioral observations ondary at a standard site on the rachis, where the calamus
meets the feather vane (i.e., at the superior umbilicus), to
Over a 3 year period, we collected a total of 1,540 h of standardize measurements of weight. Secondaries were cut
loon behavioral observations between early May and late on the left and right wing. Measurement precision was
August: 1998 (n = 753 h), 1999 (n = 571 h), and 2000 0.001 g using a high-precision digital scale at the Univer-
(n = 216 h). Loon territories were placed into three risk sity of Southern Maine, Portland, Maine. Differences
categories for interpretive purposes based on Hg concen- between left and right wing secondary weights were used
trations in FLUs: (1) low (\1.0 ug Hg/g), (2) moderate for our determination of symmetry. This method is highly
(1.0–3.0 ug Hg/g), and (3) high ([3.0 ug Hg/g). The risk replicable and without human bias and therefore offers a
categories were defined based on two thresholds for Hg high-confidence measurement. Minimizing error and bias
measurements: (1) in the southern Alaskan loon popula- in morphometrics measurements is critical for accurately
tion, which serves as our low-exposure reference group, all quantifying asymmetry (Helm and Albrecht 2000).

123
74 D. C. Evers et al.

Breeding loon monitoring Statistical analysis

From 1996 to 2005, we regularly surveyed territorial loon Mercury concentrations are expressed as arithmetic means
pairs from early or mid May until late August or early because of generally large datasets that were deemed nor-
September. We used well-established standardized proto- mally distributed based on normal probability plot residuals
cols developed by the Loon Preservation Committee and use of the Kolmogorov–Smirnov test. Homoscedastic-
(LPC) (Taylor and Vogel 2000). Surveys consisted of ity was checked with Bartlett’s test, which is sensitive to the
locating territorial pairs every 6–8 days from a boat or normality assumption. JMP software (SAS Institute 1999)
shore with 10· binoculars. We collected four reproductive was used to test various hypotheses using one-way analysis
parameters from each territory: (1) presence of territorial of variance (ANOVA) followed by Tukey’s honestly sig-
pair, (2) nesting attempts, (3) hatching success and (4) nificant difference tests if our ANOVA showed significant
fledging success. Successful fledging was defined as a differences. Student’s t-test was used when comparing
loon reaching age six weeks or older. This definition is paired data sets. JMP software corrected for inequity of
consistent with protocols developed by the Northeast unbalanced data sets. In cases where normality or homo-
Loon Study Working Group (NELSWG) and used by scedasticity tests failed, we used nonparametric tests. In all
LPC. The unit most representative of overall reproductive cases, means are given with one standard error (SE) unless
success is the number of chicks fledged per territorial pair otherwise noted. The level of statistical significance was
(CF/TP). defined as an alpha value of 0.05. Graphs were generated by
SigmaPlot ver. 10 (Systat 2006).

Laboratory analysis
Results
Laboratory protocols for analyzing total Hg follow Evers
et al. (2003) for eggs and Evers et al. (1998) for blood and Mercury exposure in the focal study area
feathers (except for feathers after 2002). Analyses of loon
tissues were conducted by the Animal Health Diagnostics Blood, egg, and feather Hg levels in the focal Maine–New
Laboratory, University of Pennsylvania, New Bolton, Hampshire study area ranged from levels similar to our
Pennsylvania (for blood and feathers) and the Trace Ele- Alaskan reference group to some of the highest recorded in
ment Research Laboratory, Texas A&M, College Station, North America. Adult blood Hg levels ranged from 0.13 to
Texas (for blood and egg). 11.80 ug/g (ww) with a mean (±SE) of 1.73 ± 0.06 ug/g
Loon feathers from 2003 to 2005 were analyzed by the (n = 644). Egg Hg levels ranged from 0.20 to 9.03 ug/g
Department of Public Health, Harvard University, Cam- (ww) with a mean (±SE) of 1.63 ± 0.82 ug/g (n = 366).
bridge, Massachusetts. Loon feathers were washed three Feather Hg levels ranged from 1.4 to 75.7 ug/g (fw) with a
times in acetone, and rinsed in triplicate with deionized mean (±SE) of 16.7 ± 0.4 ug/g (n = 302). Adverse effect
water followed by a final acetone wash and air dried in a thresholds were exceeded in 19% of the eggs, 16% of the
clean hood (Class 100 Series) for 24 h at Harvard adult blood, and 3% of the adult feathers.
University. After washing, feathers were cut into small
pieces using stainless steel scissors (cleaned between
samples), homogenized in a cryomill, and weighed in Behavioral endpoints
0.5 mg quartz sample boats. Total Hg was measured
in homogenized feather samples (feather sample mass We found a significant negative relationship between adult
range = 0.001–0.033 g) by thermal decomposition, blood Hg levels and the percent time adult male and female
amalgamation, and atomic absorption spectrophotometry loons spent in high energy behaviors while brooding 1–
(EPA method 7473; USEPA 1998) using an automated 40 day old young (F = 11.6, df = 18, p = 0.003; Fig. 3)
system (DMA-80, Milestone Inc., Monroe, Connecticut, that was predictive (r2 = 0.41)
USA). Certified reference materials (CRM; DORM-2, Behavioral data for time on nest did not meet normality
National Research Council Canada, Ottawa, ON, Canada) and homscedasticity assumptions. A ranking of the three
and procedural blanks were analyzed periodically (every Hg risk categories indicate that high Hg loons spent less
10–20 samples) to evaluate accuracy and ensure low time nest sitting than did moderate- and low-risk individ-
blanks and minimal instrument drift. Recovery of Hg in uals (n = 152 h). Males and females at low Hg risk spent
CRM was between 94–103% and precision, as measured an average of 99% of the time incubating eggs, leaving the
by randomly selected duplicate samples, averaged 90% eggs unincubated for only 1% of the time sampled
(n = 12). (n = 45 h). Loons in the moderate risk category left eggs

123
Mercury effects on loons 75

Mean % time spent by adults in

1 Survival endpoint

We determined annual changes of feather Hg levels from

high energy behaviors

known individual loons that were recaptured at least one

breeding season later. On average, the annual change for
adult loons was an increase of 9.2% for males and 7.4% for
females (n = 238, 203 resampled individuals, respectively)
r2=0.41; y = 0.644 - (0.0452 * x)
or 8.4% when sexes were combined (Fig. 5). There was no
significant difference between sexes (t = 0.77, df = 439,
p = 0.44).
0.1
0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0
Adult blood Hg level (ug/g, ww)
Reproductive endpoint
Fig. 3 Average time adult male and female loons spent in high
energy behaviors while brooding 1–40 day-old chicks related to mean
blood Hg levels in Maine, 1998–2000. Semi-log plot includes 95%
The reproductive success of 178 territories on 80 lakes was
confidence intervals monitored for 6 to 10 years—equivalent to 1,529 territory-
years. Summaries of the reproductive success of loons from
our study area provide a high-resolution description of
annual number of territorial pairs monitored and their
unincubated for an average of 10% of the time (n = 33 h),
nesting, hatching and fledging success (Table 1). When
and high risk males and females left the eggs unincubated
comparing the average CF/TP with FLUs we found a sig-
for an average of 14% of the time sampled (n = 74 h).
nificant negative correlation between fledging production
and Hg concentrations (F = 4.04, df = 176, p = 0.04)
(Fig. 6). We did not find significant among-year differ-
Physiological endpoint
ences in CF/TP (F = 0.92, df = 9, p = 0.51).
We used differences of the weights of paired flight feathers
as a measure of fluctuating asymmetry (n = 664 paired
feathers) and compared them with five groups of Hg levels Discussion
that were evenly divided. Feather weights were signifi-
cantly different between the highest Hg group ([40 ug/g) Significant adverse effects from Hg on breeding loons were
and the lowest Hg group (\10 ug/g) with high Hg adults determined using behavioral, physiological, survival, and
having greater flight feather asymmetry (t = 1.97, df = reproductive endpoints. We believe the weight of evidence
230, p = 0.02) (Fig. 4). indicates that population-level effects occur in parts of
Maine and New Hampshire, and potentially in broad areas
of the loon’s range—particularly where there is high
sulfate deposition (Jeremiason et al. 2006) and other
10.0

140
symmetrical secondary feathers

8.0
Mean % weight difference in

120

100
6.0
Annual % change in feather

80
Hg level (ug/g, fw)

60
4.0 40

20

2.0 0

-20

0.0 -40
0 <10 ug/g 10-20 ug/g 20-30 ug/g 30-40 ug/g >40 ug/g
(n=213) (n=314) (n=93) (n=25) (n=16) -60
Range of feather Hg groups (ug/g, ww) -80
Female (n=203) Male (n=238)
Fig. 4 Developmental stability measured through differences in the
mean weight (±SE) of common loon secondaries measured symmet- Fig. 5 Annual percent change of Hg measured in feathers of
rically compared to five feather Hg groups recaptured adult male and female loons

123
76 D. C. Evers et al.

Table 1 Number of territorial pairs and reproductive success of common loons in the Maine and New Hampshire study area
Year # Territories # Territorial # Nesting # Chicks # Chicks # Chicks fledged/
monitored pairs pairs hatched fledged territorial pair

1996 89 81 58 46 31 0.38
1997 115 103 68 65 47 0.46
1998 129 124 96 51 35 0.28
1999 134 128 83 71 50 0.39
2000 183 181 127 116 66 0.36
2001 189 179 124 102 61 0.34
2002 182 163 110 81 52 0.32
2003 173 160 105 101 62 0.39
2004 169 155 104 104 68 0.44
2005 166 148 102 84 61 0.41
Total 1529 1422 977 821 533 0.37

motivation to hunt, and thermoregulation (Heinz 1996;

0.60 Thompson 1996; Bouton et al. 1999).
We found that Hg can negatively impact normal adult
Chicks fledged / Territoril pair

loon behavioral activities and grouped these activities into

0.40 two broad categories based on other studies: activity level
and incubation time. Higher blood Hg concentrations are
correlated with increased lethargy in adults (Fig. 3). High-
0.20 energy behavioral events, such as time adults spent for-
aging for chicks and themselves, declined as blood Hg
y = -0.0294x + 0.4802
r2 = 0.86
levels increased. The reduction in time spent foraging
0.00 could contribute to lowered chick survival.
0.1 - 0.5 0.5 - 1.0 1.0 - 1.5 1.5 - 2.0 2.0 - 2.5 2.5 - 3.0 3.0 - 3.5 >3.5
Female Loon Unit (FLU) equivalent (ug/g, ww) Average time spent on the nest declined from the normal
coverage of 99% to 86% of the time as Hg exposure
Fig. 6 Mean changes (±SE) in overall reproductive success (chicks increased from £1.0 ug/g to ‡ 3.0 ug/g. Incubation is one
fledged per territorial pair, CF/TP) in relation to Hg levels measured
as female loon unit (FLU) equivalents (ug/g, ww) of the few behaviors for which we recorded notable
impacts from Hg during field observations. Generally, a
breeding pair equally shares incubation (Evers 1994;
conditions that favor methylation of Hg, which can lead to Mager 1995; Paruk 2000). In several cases, on loon terri-
biological Hg hotspots (Driscoll et al. 2007, Evers et al. tories where blood Hg levels ranged from 4.1 ug/g to
2007). Below, we describe our findings for each of our 6.7 ug/g in females and 4.4 ug/g to 7.3 ug/g in males,
endpoints, including the combined effects and implications breeding individuals exhibited aberrant incubation
for the sustainability of breeding populations. We also behavior.
identify geographic areas that are most likely to contain The loon territory at Stratton Island on Flagstaff Lake is
population sinks. a representative example. This territory was monitored
dawn to dusk in June of 1999. Observers documented the
routine of the non-incubating male approaching the incu-
bating female using normal, subtle behaviors including soft
Behavioral impacts mew calls, peering, and maintaining a low body profile.
The female responded normally by moving off her eggs
Time spent on the nest is a key behavior for detecting and swimming underwater, and surfacing 15–30 m from
abnormalities related to MeHg body burdens. In our study, the nest. She regularly moved out into the territory to feed
we quantified adult behavior through TABs during three during the next 4–6 h. However, instead of incubating the
distinct breeding periods: pre-nesting, nesting and post- eggs, the male remained on the water and patrolled the nest
nesting. Others have reported relationships between Hg site. After the eggs were exposed to ambient temperatures
and decreased activity levels (or increased lethargy), and the elements for 4–6 h, the female returned, switched

123
Mercury effects on loons 77

nest duties with the male, and moved to an incubation expenditure to fly in the European starling (Sturnus vul-
posture. This nest was abandoned after five days. garis) (Swaddle 1997). Further, lower wing asymmetry in
Unattended eggs have a higher probability of being migrating cliff swallows (Petrochelidon pyrrhonota) cor-
chilled/overheated or predated, which likely results in a responded with significantly greater survivorship during an
higher incidence of nest failures. Such behavior and related extreme climatic disturbance that reduced a local popula-
nest failures were observed on other territories indicating tion by over 50% (Brown and Brown 1998).
that overt abnormal behavior can be quantified at elevated
Hg levels.
Survival impacts

Physiological impacts Although birds have natural defense mechanisms for de-
purating (e.g., feathers), demethylating (e.g., liver and
Several biomarkers were evaluated to determine potential kidney), and sequestering (e.g., egg) Hg (Evers et al.
physiological effects of the blood Hg concentrations in this 2005), high-risk individuals accumulate more Hg than they
study. Tests using these biomarkers varied from no adverse are able to annually regulate. Excess Hg binds to protein in
effects detected in immunosuppression (i.e., white blood the muscle tissue and remobilizes during stressful events.
cell counts) (Haefele et al. 2005) and genetic damage (i.e., Feather molts are energetically demanding, particularly the
DNA fragmentation) (Emery 2007) endpoints to significant full remigial molts that loons experience for two weeks
yet coarse adverse effects in hormonal changes (i.e., cor- during the winter. Because muscle protein reservoirs are
ticosterone levels) (Franceschini 2007). The strongest associated with feather protein (Murphy 1996), the remo-
relationship with Hg was the test of asymmetry between bilization of proteins during feather molt partly reflects the
remige weights. available body burden of MeHg in an individual loon.
Using flight feathers, we measured the relationship In general, feather Hg increased in known individuals by
between lifetime Hg body burden and fluctuating asym- an average of 8.4% per year, which is similar to earlier
metry (FA). Clarke (1995) considered the ability of an findings by Evers et al. (1998). However, feather Hg levels
individual to develop bilateral characters to be one of the exceeding 30.0 ug/g tended to have a greater average
best estimates of developmental stability—an indirect annual increase of 10%. The increase of feather Hg levels
measure of fitness. Because feather growth is linked with with age indicates greater risk of Hg impacts to older
the very protein reserves that are associated with bound- individuals. Scenarios where birds are at lower risk indicate
MeHg in the muscle tissue (Murphy 1996; Scheuhammer no evidence of bioaccumulation in body burdens, such as in
1991), it is likely that remobilization of MeHg coincides known-aged great skuas (Catharacta skua) (Thompson
with the proteins used for feather formation. Clarke (1995) et al. 1991).
and Polak and Trivers (1994) suggested FA to be a sensi- Although Mitro et al. (2007) did not significantly cor-
tive measure of long-term body condition, and Yablokov relate Hg body burdens to apparent adult survivorship, their
(1986) and Moller and Swaddle (1997) considered FA as a elasticity analysis found that an exceedingly large spatio-
sentinel for subtle environmental perturbations prior to temporal sample size is required to detect a significant
visible effects in population viability. Some studies have impact of Hg on small differences in loon survival (likely
linked heavy metal pollution to increased asymmetry in the case because of the loon’s relatively long lifespan that
primaries (Eeva et al. 2003). may exceed 30 years). Because there can be substantial
It appears that the loon’s remige weights are a sensitive annual increases in feather Hg levels, understanding how
indicator of FA and the relationship of FA with high Hg rising Hg body burdens in high risk adults could impact
risk breeding loon populations makes this biomarker their reproductive success will require further detailed
important for monitoring age-confounded impacts from study on known individuals.
Hg. Although other stressors may disrupt developmental
homeostasis, and genetic diversity may predispose some
populations to having greater FA than others (especially in Reproductive impacts
the loon, e.g., Dhar et al. 1997), we found that FA is a valid
endpoint of physiological stress from Hg. Symmetrically- Our emphasis on the analysis of the CF/TP measure is
measured remige weights with average differences of 6.7% based on the high confidence of data (i.e., nests can fail
found on individual loons with feather Hg levels ‡40 ug/g prior to detection and hatched chicks can die before con-
could have substantial impacts on individual fitness. For firmation) and the biological significance of an outcome of
example, flight studies in wind tunnels found a wing area a fledged chick. By standardizing loon Hg levels in blood
with a difference of 5% to result in 20% more energy and eggs, we used a FLU to establish a relationship

123
78 D. C. Evers et al.

between CF/TP in correspondence with Hg concentrations. Based on our endpoints, areas of high Hg exposure could
Such an analysis demonstrates a significant decline in result in age-related increases in Hg concentrations that could
reproductive success from low to higher Hg levels that is reduce an individual’s lifetime reproductive success (LRS)
strongly predictive (r2 = 0.86) (Fig. 6). and alter the population’s age-structure toward younger
Our dataset indicates a break in CF/TP categories at adult individuals. Generally, breeding performance is lowered in
blood Hg levels of 3.0–3.5 ug/g that is significantly lower seabirds when older individuals are replaced by younger ones
than the control group (\1.0 ug/g) (t = 2.38, df = 19, (Sydeman et al. 1991), mate changes occur (Wooller et al.
p = 0.02). In our study population, adult loons with blood 1988), and body condition is compromised (Chastel et al.
Hg levels ‡ 3.0 ug/g produce 41% fewer fledged young than 1995). High quality parents generally account for the
breeding loons with blood Hg levels £ 1.0 ug/g. We predict majority of young recruited in a local population (Newton
total reproductive failure when adult blood Hg levels reach 1989); in red-billed gulls (Larus novaehollandiae scopulinus)
16.5 ug/g. Burgess and Meyer (this issue) had similar find- only 8% of the eggs laid resulted in recruitments that bred
ings based on a linear upper limit of fledging success; they (Mills 1989). A detailed demographic study based on known
documented a loss of approximately 40% of fledged young individual common loons is needed to document the rela-
when adult blood Hg levels reached 3.45 ug/g. tionship between Hg and LRS. Such an effort should examine
if high quality parents tend to choose territories that are in
habitats sensitive to MeHg production and accessibility.
Integrating endpoint measurements

The separate endpoints we analyzed provide compelling Conclusions

evidence of negative impacts on neurological, physiological,
and reproductive abilities of breeding loons due to high Hg In our Maine–New Hampshire study, Hg appears to be a
exposure. They also provide insight into why there is ulti- primary anthropogenic stressor reducing reproductive
mately lowered reproductive success in loons as Hg success to significantly under 0.48 CF/TP. Here, 16% of
concentrations increase. We speculate that flight feather the adult population exceeds tissue thresholds that have
asymmetry could decrease migratory performance for a spe- significant adverse effects (Table 2). Based on a population
cies where wing-loading is restrictive (Mager et al. 2007). matrix model, loon breeding populations producing fewer
Extra energy demands stemming from migratory flights could than 0.48 CF/TP are characterized as population sinks—
impact a loon’s ability to maintain a breeding territory against using model parameters of an annual adult survivorship of
intruding adults that regularly evaluate individual fitness; 92%, subadult survival of 41%, and average first year
prospecting by both pre-breeders and displaced breeders is breeding age of 6 years (Mitro et al. 2007; Evers 2007).
known to use cues of fitness at an established loon’s breeding Influence from extrinsic factors such as weather, predation,
territory (Piper et al. 2006). Earlier-than-normal and more habitat variability, and human disturbance are undoubtedly
frequent loss of a breeding territory would therefore occur for present and need to be factored into a comprehensive risk
individuals with elevated body burdens of Hg. assessment (Nacci et al. 2005); however, because the
The mechanism through which developmental instabil- average CF/TP did not significantly differ among years, the
ities manifest into behavioral aberrations is relatively contribution of intrinsic and extrinsic factors appears to be
unknown; however, the neurotoxic nature of MeHg appears constant and therefore predictive.
to impact adult loons by reducing time spent incubating, as While there are areas where Hg is negatively impacting
well as causing a lethargic response during a high-energy loon population viability in Maine and New Hampshire, less-
time period of feeding chicks. Breeding adults that reside studied areas in the loon’s extensive core breeding range in
in Hg sensitive areas are also prone to annual increases of Ontario and Quebec need further study. These and other
Hg body burdens that result from an inability to sufficiently areas in Canada are generally more sensitive to sulfate
depurate and demethylate dietary uptake of Hg. Prey item deposition and subsequent acidification of lakes than New
Hg levels that are correlated with reductions of 40% and England (Tan 1989) and therefore are even more sensitive to
41% in fledging success are 0.16 ug/g (ww) (Evers et al. the deposition of atmospheric Hg. The lowering of the pH in
2004) and 0.17 ug/g (ww), respectively (Burgess and lake systems is one factor that is associated with enhanced
Meyer, this issue). Prey fish behavior may also succumb to MeHg concentrations in surface water and elevated fish Hg
elevated Hg levels; studies have found cyprinnids with levels (Driscoll et al. 2001, 2007). Sulfate deposition further
0.52 ug Hg/g (whole body, ww) to school erratically enhances methylation by supporting an increase in sulfur-
(Webber and Haines 2003), which could lead to lethargic reducing bacteria that drive the methylation process (Jere-
loons disproportionally selecting fish that exhibit inhibited miason et al. 2006). There are indications that biological Hg
predator-avoidance behaviors. hotspots exist in Canada where loon reproduction may be

123
Mercury effects on loons 79

Table 2 Summary of
Matrix Adverse effect threshold Based on Reference
thresholds for adverse effects in
loon tissues and prey items Adult blood 3.0 ug/g (ww) Reproductive failure This paper
Adult feather 40.0 ug/g (fw) Flight feather asymmetry This paper
Egg 1.3 ug/g (ww) Reduced volume Evers et al. 2003
Prey fish 0.16 ug/g (ww) Relationship with adult Evers et al. 2004
blood Hg levels

negatively affected (Evers et al. 2007), including the La- Atwell L, Hobson KA, Welch H (1998) Biomagnification and
Maurice area in Quebec (Champoux et al. 2006), southern bioaccumulation of mercury in an arctic marine food web:
insights from stable nitrogen isotope analysis. Can J Aquatic Sci
Ontario (Scheuhammer et al. 1998), and southern Nova 55:1114–1121
Scotia (Burgess et al. 2005). Barr JF (1986) Population dynamics of the common loon (Gavia immer)
Spatial gradients and temporal trends in loon Hg burdens associated with mercury-contaminated waters in northwestern
can be used to identify areas at risk for ecological and Ontario. Occ. Paper 56, Can. Wildl. Serv., Ottawa, ON, Canada
Bearhop S, Ruxton GD, Furness RW (2000) Dynamics of mercury in
human health impacts in North America. The creation of an blood and feathers of Great Skuas. Environ Toxicol Chem
international Hg monitoring program that includes avian 19:1638–1643
piscivores, such as the common loon, would provide a Bouton SN, Frederick PC, Spalding MG, McGill H (1999) Effects of
standardized approach to identify biological Hg hotspots chronic, low concentrations of dietary methylmercury on the
behavior of juvenile Great Egrets. Environ Toxicol Chem
and track changes in Hg deposition. The methods for such a 18:1934–1939
project in the United States are now established (Mason Bradley DW (1985) The effects of visibility bias on time-activity
et al. 2005, Harris et al. 2007). Loons are identified as one budget estimates of niche breadth and overlap. Auk 102:493–499
of the higher ranked avian piscivorous bioindicators (Wolfe Brasso RL, Cristol DA (this issue) Effects of mercury exposure on
reproductive success of tree swallows (Tachycineta bicolor).
et al. 2007) and their Hg body burdens are well-established Ecotoxicology doi: 10.1007/s10646-007-0163-z
as providing an approach for measuring both biological Hg Braune BM, Mallory ML, Gilchrist HG (2006) Elevated mercury
hotspots and tracking the impacts of temporal changes in levels in declining population of ivory gulls in the Canadian
Hg deposition (Evers et al. 2007). Expanding this approach Arctic. Marine Poll Bull 52:969–987
Brown RC, Brown MB (1998) Intense natural selection on body size
to other northern hemisphere countries would greatly con- and wing and tail asymmetry in cliff swallows during severe
tribute to international concerns for global action (Selin weather. Evolution 52:1461–1475
2005). Further monitoring will provide the long-term Burger J (1993) Metals in avian feathers: bioindicators of environ-
datasets that are needed for conservation planning, thereby mental pollution. Reviews in Environ Toxicol 5:203–311
Burgess NM, Hobson KA (2006) Bioaccumulation of mercury in
facilitating national and even global decisions for regulating yellow perch (Perca flavescens) and common loons (Gavia
Hg emissions. Such attention will reduce the magnitude and immer) in relation to lake chemistry in Atlantic Canada.
extent of population sinks created by anthropogenic Hg Hydrobiologia 567:275–282
sources—for loons as well as other wildlife. Burgess NM, Meyer MW (this issue) Methylmercury exposure
associated with reduced productivity in common loons. Eco-
toxicology doi: 10.1007/s10646-007-0167-8
Acknowledgments This study was and continues to be an extensive
Burgess NM, Evers DC, Kaplan JD (2005) Mercury and other
effort of BRI’s International Center for Loon Conservation. Numer-
contaminants in common loons breeding in Atlantic Canada.
ous individuals, organizations and agencies assisted with this study.
Ecotoxicology 14:241–252
The Maine Department of Environmental Protection, FPL Energy
Cabana G, Rasmussen JB (1994) Modelling food chain structure and
Maine Hydro, the U.S. Fish and Wildlife Service, and Rawson Wood
contaminant bioaccumulation using stable nitrogen isotopes.
were instrumental for funding major aspects of this study. Robert
Nature 372:255–257
Poppenga of the University of Pennsylvania supervised lab analysis
Champoux L, Masse DC, Evers DC, Lane O, Plante M, Timmerman
for mercury in blood and feathers, and Robert Taylor of Texas
STA (2006) Assessment of mercury exposure and potential
A&M’s Trace Element Research Laboratory analyzed egg mercury
effects on Common Loons (Gavia immer) in Quebec. Hydrobi-
levels. EarthWatch Institute volunteers assisted with the collection of
ologia 567:263–274
behavioral observations. This study was integrated into the workscope
Chastel O, Weimerskirch H, Jouventin P (1995) Influence of body
of the Northeast Loon Study Working Group (NELSWG), a coalition
condition on reproductive decision and reproductive success in
of state and federal agency representatives, universities, non-profit
the blue petrel. Auk 112:964–972
organizations and other interested parties and members. We thank all
Clarke G (1995) Relationships between developmental stability and
members of NELSWG and the many field staff and volunteers who
fitness: application for conservation biology. Conserv Biol 9:18–24
contributed their expertise and enthusiasm to this study.
Counard CJ (2000) Mercury exposure and effects on Common Loon
(Gavia immer) behavior in the Upper Midwestern United States.
Unpublished MS thesis, University of Minnesota, St. Paul, MN
References Dhar AK, Pokras MA, Garcia DK, Evers DC, Gordon ZJ, Alcivar-
Warren A (1997) Analysis of genetic diversity in common loon
Altmann J (1974) Observational study of behavior: sampling method. Gavia immer using RAPD and mitochondrial RFLP techniques.
Behaviour 49:227–267 Mol Ecol 6:581–586

123
80 D. C. Evers et al.

Driscoll CT, Lawrence GB, Bulger AJ, Butler TJ, Cronan CS, Eagar Harris R, Krabbenhoft DP, Mason R, Murray MW, Reash R, Saltman
C, Lambert KF, Likens GE, Stoddard JL, Weathers KC (2001) T (2007) Ecosystem responses to mercury contamination:
The effects of acidic deposition in the northeastern United States Indicators of change. CRC Press, Boca Raton, FL
include the acidification of soil and water, which stressors Heinz GH (1996) Mercury poisoning in wildlife. In: Fairbrother A,
terrestrial and aquatic biota. BioScience 51:180–198 Locke LN, Hoff GL (eds) Noninfectious diseases of wildlife, 2nd
Driscoll CT, Han YJ, Chen CY, Evers DC, Lambert KF, Holsen TM, edn. Iowa State University Press, Ames, IA, pp 118–127
Kamman NC, Munson R (2007) Mercury contamination in Helm B, Albrecht H (2000) Human handedness causes directional
remote forest and aquatic ecosystems in the northeastern U.S.: asymmetry in avian wing length measurements. Anim Behav
Sources, transformations and management options. BioScience 60:899–902
57:17–28 Henny CJ, Hill EF, Hoffman DJ, Spalding MG, Grove RA (2002)
Eeva T, Lehikoinen E, Nikinmaa M (2003) Pollution-induced Nineteenth century mercury: hazard to wading birds and
nutritional stress in birds: an experimental study of direct and cormorants of the Carson River, Nevada. Ecotoxicology
indirect effects. Ecol Appl 13:1242–1249 11:213–231
Emery K (2007) Genotoxicity in the common loon. M.S. thesis, Univ. Hoffman DJ, Spalding MG, Frederick PC (2005) Subchronic effects
Southern Maine, Portland, ME of methylmercury on plasma and organ biochemistries in great
Evers DC (1994) Activity budgets of marked common loon (Gavia egret nestlings. Environ Toxicol Chem 24:3078–3084
immer) nesting population. Hydrobiologia 270–280:415–420 Jeremiason JD, Engstrom DR, Swain EB, Nater EA, Johnson BM,
Evers DC (2001) Common Loon population studies: Continental Almendinger JE, Monson BA, Kolka RK (2006) Sulfate addition
mercury patterns and breeding territory philopatry. Ph.D. increases methylmercury production in an experimental wetland.
dissertation, Univ. Minn., St. Paul, MN Environ Sci Technol 40:3800–3806
Evers DC (2006) Loons as biosentinels of aquatic integrity. Environ Kenow KP, Gutreuter S, Hines RK, Meyer MW, Fournier F, Karasov
Bioindicators 1:18–21 WH (2003) Effects of methyl mercury exposure on the growth of
Evers DC (2007) Status assessment and conservation plan for the juvenile common loons. Ecotoxicology 12:171–182
Common Loon in North America. U.S. Fish and Wildlife Kenow KP, Grasman KA, Hines RK, Meyer MW, Gendron-Fitzpa-
Service, Hadley, MA trick A, Spalding MG, Gray BR (2007) Effects of
Evers DC, Kaplan JD, Meyer MW, Reaman PS, Braselton WE, Major methylmercury exposure on the immune function of juvenile
A, Burgess N, Scheuhammer AM (1998) A geographic trend in common loons (Gavia immer). Environ Toxicol Chem 26:1460–
mercury measured in common loon feather and blood. Environ 1469
Toxicol Chem 17:173–183 Mager JN (1995) A comparison of the time-activity budgets of
Evers DC, Taylor KM, Major A, Taylor RJ, Poppenga RH, breeding male and female Common Loons (Gavia immer). M.S.
Scheuhammer AH (2003) Common loon eggs as indicators of thesis, Miami Univ., Oxford, OH
methylmercury availability in North America. Ecotoxicology Mager JN, Walcott CW, Evers DC (2007) Macrogeographic variation
12:69–81 in the body size and territorial vocalizations of male common
Evers DC, Lane OP, Savoy L, Goodale W (2004) Assessing the loons (Gavia immer). Waterbirds 30:64–72
impacts of methylmercury on piscivorous wildlife using a Martin P, Bateson P (1993) Measuring behavior: an introductory
wildlife criterion value based on the common loon, 1998–2003. guide. Cambridge Univ. Press, Cambridge, MA
Unpubl. report BRI 2004–05 submitted to the Maine Department Mason R, Abbot M, Bodaly D, Bullock R, Driscoll C, Evers D,
of Environmental Protection. BioDiversity Research Institute, Lindberg S, Murray M, Swain E (2005) Monitoring the
Gorham, ME environmental response to changes in mercury contamination
Evers DC, Burgess N, Champoux L, Hoskins B, Major A, Goodale from the atmosphere: a multi-media challenge. Environ Sci
W, Taylor R, Poppenga R, Daigle T (2005) Patterns and Technol 39:15A–22A
interpretation of mercury exposure in freshwater avian com- Meyer MW, Evers DC, Hartigan JJ, Rasmussen PS (1998) Patterns of
munities in northeastern North America. Ecotoxicology 14: common loon (Gavia immer) mercury exposure, reproduction,
193–222 and survival in Wisconsin, USA. Environ Toxicol Chem 17:184–
Evers DC, Han YJ, Driscoll CT, Kamman NC, Goodale MW, 190
Lambert KF, Holsen TM, Chen CY, Clair TA, Butler T (2007) Miller EK, Vanarsdale A, Keeler GJ, Chalmers A, Poissant L,
Identification and evaluation of biological hotspots of mercury in Kamman NC, Brulotte R (2005) Estimation and mapping of wet
the Northeastern U.S. and Eastern Canada. BioScience 57:29–43 and dry mercury deposition across northeastern North America.
Fournier F, Karasov WH, Kenow KP, Meyer MW, Hines RK (2002) Ecotoxicology 14:53–70
The oral bioavailability and toxicokinetics of methylmercury in Mills JA (1989) Red-billed gulls. In: Newton I (ed) Lifetime
Common Loon (Gavia immer) chicks. Comp Biochem Phys Part reproduction in birds. Academic Press, San Diego, CA, pp
A 133:703–714 387–404
Franceschini MD (2007) Mercury and circulating corticosterone in Mitro MG, Evers DC, Meyer MW, Piper WH (2007) Common loon
common loons. PhD. Dissertation, Tufts Univ., Boston, MA survival rates and mercury in New England and Wisconsin. J
Frederick PC, Hylton B, Heath JA, Spalding MG (2004) A historical Wildl Manage, in press
record of mercury contamination in southern Florida (USA) as Moller A, Swaddle J (1997) Asymmetry, developmental stability, and
inferred from avian feather tissue. Environ Toxicol Chem evolution. Oxford University Press, Oxford
23:1474–1478 Monteiro LR, Furness RW (2001) Kinetics, dose-response, excretion,
Gostomski TJ, Evers DC (1998) Time-activity budget for common and toxicity of methylmercury in free-living Cory’s shearwater
loons, Gavia immer, nesting on Lake Superior. Can-Field chicks. Environ Toxicol Chem 20:1816–1824
Natural 112:191–197 Murphy M (1996) Energetics and nutrition of molt. In: Carey C (ed)
Haefele HJ, Sidor J, Evers DC, Hoyt DE, Pokras MA (2005) Avian energetics and nutritional ecology. Chapman and Hall,
Hematologic and physiologic reference ranges for free-living New York, pp 158–198
adult and young common loons (Gavia immer). J Zool Wildl Nacci D, Pelletier M, Lake J, Bennett R, Nichols J, Haebler R, Grear
Manage 36:385–350 J, Kuhn A, Copeland J, Nicholson M, Walters S, Munns WR Jr

123
Mercury effects on loons 81

(2005) An approach to predict risks to wildlife populations from Spalding MG, Frederick PC, McGill HC, Bouton SN, Richey LJ,
mercury and other stressors. Ecotoxicology 14:283–293 Schumacher M, Blackmore CG, Harrison J (2000) Histologic,
Newland MC (2002) Neurobehavioral toxicity of methylmercury and neurologic, and immunologic effects of methylmercury in
PCBs: effects-profiles and sensitive populations. Environ Tox- captive great egrets. J Wildl Dis 36:423–435
icol Pharm 12:119–128 Swaddle JP (1997) Within-individual changes in developmental
Newton I (1989) Lifetime reproduction in birds. Academic Press, San stability affect flight performance. Behav Ecol 8:601–604
Diego, CA Sydeman WJ, Penniman JF, Penniman TM, Pyle P, Ainley DG (1991)
Nocera J, Taylor P (1998) In situ behavioral response of common Breeding performance in the Western Gull: effects of parental
loons associated with elevated mercury exposure. Conserv Ecol age, timing of breeding and year in relation to food availability J
2(2):10 Anim Ecol 60:135–149
Olsen B, Evers DC, DeSorbo C (2000) Effect of methylated mercury on Systat (2006). SigmaPlot 10. Systat software, Inc., Pt. Richmond, CA
the diving frequency of the Common Loon. J Ecol Res 2:67–72 Tacha TC, Vohs PA, Iverson GC (1985) A comparison of interval and
Paruk JD (2000) Incubating roles and patterns in Common Loons, pp continuous sampling methods for behavioral observations. J
50–54. In: McIntyre J, Evers DC (eds) Loons: old history and Field Ornithol 56:258–264
new findings. Proc. of a Symposium from the 1997 meeting, Tan B (1989) Extent and effect of acid precipitation in northeastern
American Ornithologists’ Union. N. Am. Loon Fund, Holder- United States and eastern Canada Arch. Environ Contam Toxicol
ness, NH 18:55–63
Piper WH, Walcott C, Mager JN, Perala M, Tischler KB, Harrington Taylor KM, Vogel H (2000) New Hampshire report. In: McIntyre J,
E, Turcotte AJ, Schwabenlander M, Banfield N (2006) Pros- Evers DC (eds) Loons: Old history and new findings. Proc.
pecting in a solitary breeder: chick production elicits territorial Symp. 1997 meeting, Am. Ornith. Union. N. Am. Loon Fund,
intrusions in common loons. Behav Ecol 17:881–888 Holderness, NH, pp 110–113
Polak M, Trivers R (1994) The science of symmetry in biology. Thompson DR (1996) Mercury in birds and terrestrial animals. In:
Trends Ecol Evol 9:122–124 Beyer WN, Heinz GH, Redmon-Norwood AW (eds) Environ-
SAS Institute (1999) JMP Statistical discovery software. SAS mental contaminants in wildlife: interpreting tissue
Institute, Cary, NC concentrations. Lewis Publisher, Clemson, SC, pp 341–355
Scheuhammer AM (1991) Effects of acidification on the availability Thompson DR, Hamer KC, Furness RW (1991) Mercury accumula-
of toxic metals and calcium to wild birds and mammals. Environ tion in great skuas Catharacta skua of known age and sex, and
Poll 71:329–375 its effects upon breeding and survival. J Applied Ecol 28:672–
Scheuhammer AM, Atchison CM, Wong AHK, Evers DC (1998) 684
Mercury exposure in breeding common loons (Gavia immer) in United States Environmental Protection Agency (USEPA) 1998.
central Ontario, Canada. Environ Toxicol Chem 17:191–196 Mercury in solids and solutions by thermal decomposition,
Scheuhammer AM, Perrault JA, Bond DE (2001) Mercury, methyl- amalgamation, and atomic absorption spectrophotometry. EPA
mercury, and selenium concentrations in eggs of Common Loons Method 7473 Report, January 1998. 15 p
(Gavia immer) from Canada. Environ Monitor Assess 72:79–94 Webber HM, Haines TA (2003) Mercury effects on predator
Scheuhammer AM, Meyer MW, Sandheinrich MB, Murray MW avoidance behavior of a forage fish, golden shiner (Notemigonus
(2007) Effects of environmental methylmercury on the health of crysoleucas). Environ Toxicol Chem 22:556–581
wild birds, mammals, and fish. AMBIO: J Human Environ Wolfe MF, Atkeson T, Bowerman W, Burger K, Evers DC, Murray
36:12–19 MW, Zillioux E (2007) Wildlife indicators. In: Harris R,
Scheuhammer AM, Basu AM, Burgess N, Elliot NM, Campbell JE, Krabbenhoft DP, Mason R, Murray MW, Reash R, Saltman T
Wayland GD, Chapoux M, Rodgrigue J. (this issue) Relation- (eds) Ecosystem response to mercury contamination: indicators
ships among mercury, selenium, and neurochemical parameters of change. CRC Press, SETAC, Webster, NY, pp 123–189
in common loons (Gavia immer) and bald eagles (Haliaeetus Wolfe M, Schwarzbach FS, Sulaiman RA (1998) Effects of mercury
leucocephalus). Ecotoxicology doi: 10.1007/s10646-007-0170-0 on wildlife: a comprehensive review. Environ Toxicol Chem
Schwarzbach SE, Albertson JD, Thomas CM (2006) Effects of 17:146–160
predation, flooding, and contamination on reproductive success Wooller RD, Bradley JS, Serventy DL, Skira IJ (1988) Factors
of California clapper rails (Rallus longirostris obsoletus) in San contributing to reproductive success in short-tailed shearwaters
Francisco Bay. Auk 123:45–60 (Puffinus tenuirostris). Proc Int Ornithol Congr 19:848–856
Selin NE (2005) Mercury rising: is global action needed to protect Yablokov AV (1986) Population biology: progress and problems of
human health and the environment. Environment 47:22–35 studies on natural populations. Moscow: Mir

123