California State University, San Bernardino

CSUSB ScholarWorks

Theses Digitization Project John M. Pfau Library

2011

Geodesics of surface of revolution

Wenli Chang

Follow this and additional works at: https://scholarworks.lib.csusb.edu/etd-project

Part of the Geometry and Topology Commons

Recommended Citation
Chang, Wenli, "Geodesics of surface of revolution" (2011). Theses Digitization Project. 3321.
https://scholarworks.lib.csusb.edu/etd-project/3321

This Thesis is brought to you for free and open access by the John M. Pfau Library at CSUSB ScholarWorks. It has
been accepted for inclusion in Theses Digitization Project by an authorized administrator of CSUSB ScholarWorks.
For more information, please contact [email protected].
Geodesics of Surface of revolution

A Thesis

Presented to the

Faculty of

California State University,

San Bernardino

In Partial Fulfillment

of the Requirements for the Degree

Master of Arts

in

Mathematics

by

Wenli Chang

September 2011
 Geodesics of Surface of revolution

A Thesis

Presented to the

Faculty of

California State University,

San Bernardino

by

Wenli Chang

September 2011

Approved by:

Dr. Wenxiang Wang, Committee Chair Date

Dr. Peter Williams , Chair, Dr. Charles Stanton

Department of Mathematics Graduate Coordinator,
Department of Mathematics
 iii

Abstract

In this thesis, I study the differential geometry of curves and surfaces in three-
dimensional Euclidean space. Some important concepts such as, Curvature, Fundamental
Form, Second Fundamental Form, Christoffel symbols, and Geodesic Curvature and equa­
tions are explored.
I then investigate the geodesics on a surface of revolution through solving dif­
ferential equations of geodesic. Main result are stated in Theorem (8.1).
 iv

Acknowledgements

First of all, I would like to extend my sincere gratitude to my supervisor, Dr.

Wenxiang Wang, for his constant encouragement and guidance, for his instructive advice
and useful suggestions and for his patience and thoughtfulness. I am deeply grateful of his
help in the completion of this thesis. Without his consistent and illuminating instruction,
this thesis could not have reached its present form.
I would like to express my gratitude to Dr. Stanton and Dr. Trapp at CSUSB
who gave me suggestions of this thesis. I also owe a special debt of gratitude to all the
professors and staffs in the math department at CSUSB, from whose devoted teaching,
enlightening lectures and kindly help I have benefited a lot.
Finally, I am indebted to my husband and my daughter for their continuous
support and encouragement. Especially, thanks to my husband who have always been
helping me out of difficulties and supporting me taking care of the family.
 V

Table of Contents

Abstract iii

Acknowledgements iv

List of Figures vii

1 Introduction 1

2 Curves 2
2.1 What Is a Curve................................................................................................ 2
2.2 Arc Length.......................................................................................................... 5
2.3 Tangent, Normal and Osculation Plane........................................................... 7
2.4 Curvature............................................................................................................. 8

3 Concepts of a Surface. First Fundamental Form 14

3.1 Concept of a Surface in Differential Geometry............................................... 14
3.1.1 Parametric Representation of Surfaces............................................... 14
3.2 Curve on a Surface , Tangent Plane to a Surface......................................... 18
3.3 First Fundamental Form.................................................................................... 20

4 The Second Fundamental Form. Christoffel Symbols 24

4.1 The Christoffel Symbols.................................................................................... 28

5 Normal and Geodesic Curvature 31

6 Geodesic and Geodesic Equations 36

6.1 Definition and Basic Properties........................................................................ 36
6.2 Derive the Geodesic Equations.................. 36

7 A Surface of Revolution 41
7.1 The Parametric Representation of a Surface of Revolution ...................... 41
7.2 First and Second Fundamental Form of a Surface of Revolution................ 42
7.2.1 The First Fundamental Form of a Surface of Revolution................ 42
7.2.2 The Second Fundamental Form of a Surface of Revolution............. 43
 vi

7.3 The Christoffel Symbols for a Surface of Revolution .................................. 45

8 Geodesic of a Surface of Revolution 47

8.1 Geodesic Equations of a Surface of Revolution............................................... 47
8.2 Examples of Geodesic of a Surface of Revolution......................................... 50

Bibliography 53
 vii

List of Figures

2.1 Helix...................................................................................................................... 3
2.2 Folium Descartes................................................................................................ 4
2.3 LimaQon................................................................................................................ 4
2.4 Tangent, Normal and Binormal Vector........................................................... 7
2.5 Cylinder Circular................................................................................................ 12

5.1 Normal and Geodesic Curvature ..................................................................... 31

7.1 Surface of Revolution .............................................................. 42

8.1 Geodesics on Sphere.......................................................................................... 51

8.2 Geodesic on Cone................................................................................................ 51
 1

Chapter 1

Introduction

The study of geodesics is one of the main subjects in differential geometry. The
shortest path on a surface joining two arbitrary points is represented by a geodesic. In
this project, I focus on the study of geodesics on a surface of revolution. I first introduce
some of the key concepts in differential geometry in the first 6 chapters. In chapter 7, I
derive the differential equations for a curve being a geodesic.
A theorem on geodesics of a surface of revolution is proved in chapter 8.
 2

Chapter 2

Curves

In this Chapter, we discuss the curves in 3-dimentional Euclidean space R3.

2.1 What Is a Curve

Definition 2.1. A curve in R3 is a differentiable map X : I —> R3 of I into R3. I = [a, b]
be an interval on the real line R1. For each t G I we have

X(t) = (x1(t),x2(t),X3(t)), (2.1)

where xi,X2, X3 are the Euclidean coordinate functions of X, and t is called the parameter
of the curve X. X(t) can be considered as the position vector of a moving point on the
image set X(I) of the curve X.

Example 1: Straight Line.

A straight line in space passing through the point A?(0) = a and in b position
can be represented in the following parametric form:

X(t) = a + bt = (ai + bit, a2 + bit, a3 + fet) (2.2)

where
a = (01,02,03), b = (61,62,63).

a, b are two constant vectors.

Example 2: An ellipse with center at origin

 3

An ellipse with center at originis given by

X(t) = (acost, 6sint,0) (2.3)

in particular, if a = b = r we have a circle of radius r as

X(t) = (r cost, r sin t,0). (2.4)

Example 3: Helix

Helix is a curve given by

X(t) = (r cos t,r sin t, ct), a>0, c>0, (2-5)

Helix raise at a constant rate on the cylinder x2 + x2 = r2, and it is said right winding,
if b < 0, then the helix is said to be left winding; and the A?3-axis is called the axis.

Example 4-' Folium of Descartes

The folium of Descartes can be represented in parametric form as follows:

This curve has a double point at (#i,#2) = (0,0).

Example 5: the limagon

 4

Figure 2.2: Folium Descartes

Figure 2.3: Limagon

 5

The limagon is the parametrized curve:

X(t) = ((1 + 2cost) cost, (1 + 2cost) sint)), teR. (2.7)

Note that X has a self-intersection at the origin in the sense that X(t) = 0 for t —
and for t =

2.2 Arc Length

Definition 2.2. Let X(f) (a < t <b) be a curve, then the arc length of the curve X(t)
defined as

s(t) = f* \x'(t)\dt (2.8)

Jto
where,
\X'(t)\dt = a/@1')2 + (4)2 + (®3)2 = (2.9)

is the length of the vector X' (t).

Note that: X' (t) is the tangent vector of the curve and s(t) is a differentiable
function of t.

Theorem 2.3. A curve is parametrized by the arc length if and only if

|XZ(t)| = l (2.10)

Proof. By the Fundamental Theorem of Calculus

ds = ix'ldt,

ds ds
dt ds ’
when t=s is the arc length.

On the other hand, if for the parameter t, we have

|x'(t)| = i =>
ds
dt = ’
 6

Thus,
to-

Let
to = 0,

t is the arc length.

□
I
I
Definition 2.4. We will write

ds2 = dx2 + da?2 + dx2 = dX • dX (2-U)

ds is called the element of arc or linear element of C. ,

Remark: In this paper, some notations are defined

as
t_dPx
X ds2

x’ = 4
d
dt
d2X
X dt2 ’

I
Example 1: For the circular helix as in (2.5):

X (t) = (r cos t, r sin t, ct)

then

X' = (—r sin t,r cost, c)

X' -X' =r2+c2

and therefore
S(t) = ty/(r2 + c2)

Let

I
 7

Then,
X(S) = (rcos(s/w) rsin(r/w)' cs/w).

is the circular helix with arc length as a parameter.

2.3 Tangent, Normal and Osculation Plane

Figure 2.4: Tangent, Normal and Binormal Vector

i
I
i
I
i
i

Definition 2.5. Let C be an arbitrary curve in the space R3, and X(s) be a parametric
representation of C with arc length s as parameter. The vector

(2-12)

is called the unit tangent vector of the curve C at the'point -Y(s). This vector is a unit
 8

vector, because
dX dX
\t\2 = t-t = X-X = (2-13)
ds ds
Definition 2.6. All the vectors pass through a point P of C and orthogonal to the cor­
responding unit tangent vector lie in a plane. This plane is called the normal plane to C
at P. I

Definition 2.7. The plane determined by the unit tarigent t(s) and X(s) is called the
osculating plane of the curve C at P.

Definition 2.8. The intersection of the osculating plane with the corresponding normal
plane is called the principal normal. 1

2.4 Curvature
Curvature is a very important concept of a curve. Curvature measures that how
curved a curve is. ,
Let X(s) represent a curve C with arc length s as parameter.
Since s is the arc length, then by (2.10) : ;

iy(5)i = i

i.e
X(s)-X(s) = 1. I

Differential above equation respect to s, we have

i

X(s) • X(s) = 0

thus
X(s)±X(s) |
I
If X(s) f 0, Y(s) is orthogonal to the unit tangent vector X(s) and lies in the normal
plane to C. ;

Definition 2.9. The unit vector

X(s)
(2-14)

is called the unit principal normal vector to the curve C at the point X(s).
 9

Definition 2.10. :

k(s) = |t(s)| = y/{X(s) ■ X(s)} (k > 0) (2.15)

is called the curvature of the curve C at the point X(s).

Definition 2.11. Set

K(s) = t(s) (2.16)

and it is called curvature vector.

Theorem 2.12. A curve is a straight line if and only if k = 0.

Proof. : Part 1: If C : X = X(s) is a straight line, then

X — Xq T as

where Xo and a are constant.

Therefore,
X=a

k = |X(s)| = 0

Part 2: If k = 0, By k = |X(s) |, we have X is constant, say a. Thus,

a = X(s)

taking integral respect to s of the above equation,

ads = as + Xq

It is a straight line. □

Note that:
i=X

fc(s) = |t(s)|

P(s) =
l^(s)l

It is equivalent to say that

t(s) = kP
 10

Taking the cross product of both sides of this equation by t, we find:

txt = XxX = tx kP

then,
|X x X| = |t x kP\ = k\t x P\

Now, since t and P are unit and orthogonal, |t x P\ = 1 then we obtain that:

k = \Xx X|.

With respect to any parameter t, we find that

ds
dX dt
dt ds
dX
ds
d dX dt.
ds'' dt ds
d dX dt dt
dt dt ds ds
d2X dt dX d2t dt
dt2 ds + dt ds2 ds
dPX (dt\2 ! dX d2t dt

then
dX dt\ (d2X /dt\2 dX d2t dt
dt ds) X y dt2 yds) dt ds2 ds

dt\2 dX (d2X dt dX d2t\

ds J dt X \ dt2 ds + dt ds2 J

dt\2 dX /dXdtV
ds J dt \ dt ds J

That is
k= |XxX| = |X' xX"|

by definition (2.3), we have that

ds3 — dX ■ dX • dX.
 11

We can rewrite above equation as:

k= |X X X| = |X' X x"| (A

i.e
, \X'xX"\ ,dX
(2-17)
k= irp (ir = x)

we can write (2.16) as:

k (y/X' xX")-(X' xX')
(Vx7 • x')3

Since by the the Identity of Lagrange-.

(a x b) • (c x d) = (a • c)(b • d) — (a • d)(b • c)

where a, b, c, d are vectors.

It is equivalent to
, V{(X' • X')(X" • X") - (X7 • X")2}
/v ■ ■ 1 Q (2-18)
(X7 .X')2
In particular t = s, (2.17) reduces to the form (2.15).

Examples: The Curvature of Some Curves

Example 1. When C is a circle represented by

X(t) = (rcost, rsint, 0)

then
X' = (—rsint, rcost, 0), X" = (—rcost, —rsint, 0).

By (2.17),

t y/{(r2 sin21 + r2 cos21) • (r2 cos21 + r2 sin21) — (r2 sin t cos t — r2 sin t cos t)
fV 2
(r2 sin21 + r2 cos21) 2

thus, for a circle

= r.
 12

Figure 2.5: Cylinder Circular

Example 2: The circular helix

X(t) = (r cos t,r sin t,ct)

X = (—r sin t,r cost, c)

X" = (—r cost, —r sin t, 0)

Thus by(2.17), the curvature of the helix is:

, {(r2 sin21 + r2 cos2t + c2) • (r2 cos2t + r2 sin21) — (r2 sin t cos t — r2 sin t cos t)
•V
(r2 sin21 + r2 cos2t + c2) 2

k = yc^ + c2)^2)
(r2 + c2)t
Thus,
k_ r
r2 _|_ c2

So, the curvature of the circular helix is constant.

 13

Example 3: The hyperbolic spiral

X(t) = (a cosh t, a sinht, at), (a > 0);

From above parametric function,

X' = (a sinh t, a cosh t, a),

And,
X" — (a cosh t, a sinh t, 0),

Then,
X' x X" = (—a2 sinht, a2 cosht, —a2),

|-X’Z x X"] = V^acosht,

|A’/| = V2a cosht

Thus, the curvature is :

k ,\X'xX"\
|X'|3
i.e
£■», — \/2a- 2— cosh- t -
2y/2ascosh31

k =------
2a cosh t
 14

Chapter 3

Concepts of a Surface. First

Fundamental Form

In this chapter, we discuss the surfaces in R3.

3.1 Concept of a Surface in Differential Geometry

3.1.1 Parametric Representation of Surfaces

Definition 3.1. Let D C R2 is a open domain, (u,v) represent the points in R2. There
is 1-1 map: r : Z) —> R3, if (xx,X2,x3) represent the Cartesian coordinate of the points
on R3, then the map can be presented as:
f
a?i = xffu, v)
< X2 = X2(u, v) (u,v)&D
k x3 = x3(u,v)

and xffu,v), X2(u,v),x3(u,v) are all differentiable functions respect to u,v, under this
map, the image of D form a surface S in R3. (u, v) is called the parameters of surface S
and the vector parametric function of S is

X(u,v) = (xffu,v),x2(u,v),x3(u,v)) (3.1)

Remark: In Differential Geometry we use calculus to analyst surfaces. The functions

on a surface must be differentiable. In order to be able to apply differential calculus to
 15

geometric problems of X(u, v) with respect to u, v. The following assumptions are made:

The Jacobin matrix

/ 9x1 dxi \

J= dx2,
du dv
dx3
dv J

is of rank 2 in D.
It means the Jacobins JX1, JX2i Jx3 are not all simultaneously zero,i.e.,

Jxi + Jx2 + Jx3 0

and the derivatives of X^, X2 and X3 with respect to u and v are continuous.

The Jacobians are defined as:

_ d(x2,x3)
Jx\
d(u, v)

= d(x3,x1)
Jx2
d(u, v)
= d(x1}X2)
Jx3
d(u, v)
Where,
dx^ dx3
d(x2,x3) _ du du
d(u,v) dx% dx3
dv dv

dxi dx2
3(^1, ^2) du du
5(u, v) d^ dx2
dv dv

dx3 dxi
d(X3,X!) du du
d(u, v) dx3 dxi
dv dv
 16

Remark: Rank J = 2 is equivalent to the condition that and are linearly inde­
pendent.

Examples of surface
Example 1: A sphere of radius r with center at X = (0,0,0)
A sphere is the set of points of R3 that are a fixed distance(the radius) from
a fixed point(its center). So the sphere of radius r with center at X = (0,0,0) can be
represent as:
2,2,2 2
x±+ x% + x$ = r

from this we can obtain a representation of the form:

It represents the two hemispheres when X3 > 0 or X3 < 0.

The parametric representation of the sphere can be given as :

X(u, v) = (r cos v cos u, r cos v sin u, r sin v) (3-2)

that is,

xi = r cos v cos u

X2 = r cos v sin u

X3 = r sin v

As Figure(3.1), u is the longitude and v is the latitude.

Example 2: Let C be a curve in the rri^-plane defined by

X(a;i,a:2) = c, X3 = 0

Then the cylinder lS'(Figure(3.2)) generated by the fine L perpendicule to the aji^-plane
along the curve C is given by
X(X!,X2) = c. (3-3)
 17

If C is not a closed curve, and H(u) = (hi (u), /12(a), 0) is a parametrization of C, then

X (u, v) — (hi (u), h2 (u), v) (3.4)

is a parametrization of the cylinder S.

A circular cylinder is the set of points of R3 that are at a fixed distance( the
radius of the cylinder) from a fixed straight line(its axis).
So C (from above example 2) is a circle with a radius a, the parametrizatric
equation of C is:
H(u) = (acosu, asinu, 0).

Thus,
X(u, v) = (acosu, asinu, v) (3.5)

is a representation of the cylinder of revolution which has radius a and the #3 axis as axis
of revolution.
The corresponding matrix

—asinu 0
J = acosu 0

is of rank 2.

Example 3: A cone of revolution with apex at X = (0,0,0) and with #3-axis as axis of
revolution can be represented as:

X(u,v) = (u cos v,u sin v,au). (3.6)

The curves u = const are circles parallel to the #i#2 — plane while the curves
v = const are the generating straight lines of the cone.

Example 4: Torus is rotating a circle C in a plane II around a straight line Z in II that

does not intersect C. If II to be the xz—plane, Zto be the z—axis, a > 0 the distance of
the center of C from I, and b < a the distance of O',then the torus is a smooth surface
with parametrization.

X(u, v) = ((a + fccosu) cost?) cosu, (a + fccosu) sinu, fesinu)

 18

3.2 Curve on a Surface , Tangent Plane to a Surface

Any curve X(u, v) on a surface S can be determined by a parametric parametric
function:
u = u(t)
a <t <b
v = v(t)

thus, the parametric equation of C can be written as:

X = X(u(t),v(t))

the tangent vector of C on a surface S is:

dX du dXdv
du dt "I” dv dt (3.7)
dt

—X vvu ' ,
vuu' + -X. tvu —
(A vv —
,X ).

This means that the tangent vector of C on a surface S is linear combination of

the vector Xu and Xv. We assume that Xu x Xv 0, Xu, Xv are linearly independent.

Definition 3.2. The vectors Xu, Xv in (3.7) spans a plane E(P) called the tangent plane
at P to the surface S. E(JP) contains the tangent to any curve on S at P passing through
the point P.

Remark: There is a only one tangent direction pass a point at a curve, but there are
infinite tangent directions pass a point on a surface, these tangent vectors form the tangent
plane.

Definition 3.3.
Xu x Xy
(3-8)
X X-u|
is called an unit normal vector to S at D, where Xu = and Xv =
 19

Example 1: Tangent vector and Normal vector of a Circular cylinder

The parametric function of a circular cylinder as (3.5)

X(it, v) = (a cos u, a sin u, i>)

and 0 < u < 2ir, —oo < v < oo, its tangent vector is:

Xi = Xu = (—a sin it, acosu, 0)

X2 = XV = (0,0,1)

its unit normal vector is:

Xu x Xv . .
11 = IY v Yl = (C0S Sm U’ 0)
|-A^ X |

Example 2: Catenary

X(u,v) = (coshucosv, cosh itsinv, it).

Calculate partial derivative:

Xi = (sinh it cos v, sinh u sin v, 1)

X2 = (— cosh u sin v, cosh u cos v, 0)

then
Xi x Xv = (— cosh u cos v, — cosh it sin v, 0)

and,
X^t x Xv 1
n= (— cos v, — sin v, sinh
| Xi x XJ coshiz
 20

3.3 First Fundamental Form

Let S be a surface defined by a parametric equation X(u,v), as we pointed out
in the preceding section, any curve on surface S can be represented in the form.

u = u(t),v = v(t)

the parametric equation of the curve is:

X(u,v) = X(u(t), v(t)).

From (3.7), we have:

/ dX dX du dX dv , /

Multify dt on both sides of the above equation:

dX = (Xu^+Xv^)dt
at at

i.e;
dX = Xudu + Xvdv.

Now, from (2.11)

ds2 = dX • dX

ds2 = dX ■ dX = (Xudu + Xvdv)(Xidu + Xvdv)

thus,
ds2 = XuXu(du)2 + 2XuXvdudv + XvXv(dv)2. (3.9)

We set

Xu • Xu — gn (3.10)

Xu- Xv = gi2 (3-11)

Xv ■ Xu — <721 (3.12)

Xu • Xu = 922 (3.13)

since 912 = <721 > then (3.9) is written as:

ds2 = flli(du)2 + 2gi2dudv + g22(dv)2 (3-14)

 21

3.14 is called the First Fundamental Form of the surface S. It is also expressed as

Edu2 + 2Fdudv + Gdv2.

That is:
Xu ' Xu — gn — E,

Xu • Xu — Xv • Xu — 512 — 521 — F,

Xv ■ Xv — 522 = G

Example 1: Plane
X =(u,v,0)

then
X -dx-1 x _ dX - i
— —- — 1
u ~ du ~ ’ av
Xu • Xu = 1 = E, Xu • Xu = 1 = G

a is the angle between x and y, is right angle.

Thus, the First Fundamental Form for a plane is

ds2 = dx2 + dy2.

Example 2: Circular cylindrical surface

X(u,v) = (a cos u, a sin u, v)

then
Xu = (—a sin u, a cos u, 0)

Thus, the First Fundamental Form of a circular cylinder is

ds2 = a2 du2 + du2

 22

Example 3: Hyperbolic paraboloid:

X = {a(u + v), b(u — v), 2uv}

Xu = (a, b, 2v), Xv = (a, — b, 2u)

Then
E = Xu ■ Xu = a2 + b2 + 4v2

F = Xu • Xv - a2 — b2 + 4uv

G = Xv • Xv — a2 + b2 + 4u2

thus, the First Fundamental Form of Hyperbolic paraboloid is:

ds2 = (a2 + b2 + 4v2)du2 + 2(a2 — b2 + 4uv)dudv + (a2 + b2 + 4u2)dv2

Example j: Sphere

X(u, v) = (a cos u cos v, a cos u sin v, a sin u)

Xu = (—a cost; sin a, — a sinusin v, a cos u)

Xv = (—a cos u sin u, acosucos?;,0).

then

= a2

G = Xv • Xv = (y/a2 cos2 u sin2 v + a2 cos2 u cos2 v + o)

9 9
= a cos u

Xu • Xv — 0.

The first fundamental form of Sphere is:

ds2 = a2 (du2 + cos2 udv2)

Example 5: Catenary

X (u, v) = (cosh u cos v, cosh u sin v, u)

 23

Xu = (sinh u cos v, sinh u sin v, 1)

Xv = ( — cosh u sin v, cosh u cos v, 0)

and,
F = Xu • Xv = 0

thus, The first fundamental form of Catenary is:

ds2 = cosh2 udu2 + cosh2 dv2

 24

Chapter 4

The Second Fundamental Form.

Christoffel Symbols

Consider an arbitrary surface S : X(u, v) of class r >2 and an arbitrary curve

Con S:
u = u(s), V = v(s)

Where s is the arc length of C.

Let 7 be the angle between the unit principal vector p to C and n be the unit
normal vector to S. Since p and n are unit vectors, we have:

p • n = |p|]n| COS7 =>

cos

cos 7 = p • n

since, p = *. = *, we have
A: cos 7 = X • n.

By chain rule
dX du dx dv „ . .
X = ————|- — — = Xuu + Xvv (4.1)
du ds dv ds
and
-X" — Xuuuu “H X^^uv 4” X^^vu “I- X'yyVv X^u d- X^v. (4-2)
 25

Recall the notations:

dX dX
Xu -
du ’
d2X „ d2X
XUV —
dvdu’ vu dvdu
d2X „ d2X
^UU — ■^VV — 11
dudv’ avav

Let n be the unit normal vector of the surface, then n is orthogonal to both Xu
and Xv.

Xu • n = 0, Xv • n = 0

therefore, scalar product of n to (4.2)

X • n = (Xuuuu + Xuvuv + Xvuvu + Xvvvv) ■ n.

Now, we set notation as

&11 — Xuu • n fei2 — Xuv • n

&22 = Xvv ' n &21 = Xvu • n

since,

thus,
fei2 = &21

and
X • n = fen(u)2 + 2b12uv + b22(v)2

i.e
d2X
n = fen (4-3)
ds2
We get :
d2X • n = fen (du)2 + 2b\2dudv + b22(dv)2.

The quadratic form,

fen (du)2 + 2fei2dudu + b22(dv)2 (4-4)
 26

(4.4) is called the the Second fundamental form of the surface S. We are going to rewrite
it as:
L(du)2 + 2Mdudv + N(dv)2 (4.5)

Where

5ii — E — • n,

^12 = &2i = M = Xuv • n = X21 • n,

^22 = N — Xvv ■ n

The second fundamental form of some surfaces are calculated here.

Case l:Plane
From previous chapter we have:

Xu = (—a sin u, a cos u, 0)

Xv = (0,0,1)

xu = 1, xv = 1

so,
Xuu xvv = 0

thus, The second fundamental form of a plane is: 0.

Case 2: Circular Cylinder

X = (u cost, a sin t,ct)

Xu = (—a sin u, a cos u, 0)

Xv = (0,0,1)

E = a2, F = 0, G= 1

Xuu = (—acosu, —asinu, 0)

Xvv = 0
 27

n = (cos u, sin u, 0)

and,
L — luu ' n — ct

M=0 N = 0.

Thus, the second fundamental form of Circular Cylinder is:

—ad2u

Case 3: Catenary
X(u, u) = (cosh u cos v, cosh u sin v, 1)

From above chapter we have:

Xu = (sinh u cos v, sinh u sin v, 1)

Xv = (— cosh u sin v, cosh u sin v, 0)

E = cosh2 u

G = cosh2 u

F=0

Now,
Xuu = (cosh u cos v, cosh u sin v, 0)

Xvv = (— cosh u cos v, cosh u cos v, 0)

And from previous chapter:

n = ——(— cos v, — sin v, sinh u)

coshu
Thus,
E — Xuu' u — 1

M = Xvv • n = 0

N = Xuv • n = 1

The second fundamental form of Catenary is:

—d2u + d2v
 28

4.1 The Christoffel Symbols

Christoffel symbols are shorthand notations for various functions associated with
quadratic differential forms. Each Christoffel symbol is essentially a triplet of three
indices, i, j and k, where each index can assume values from 1 to 2 for the case of two
variables.
Let X(u, v) be a surface with first and second fundamental forms

Edu2 + 2Fdudv + Gdv2

Ldu2 + 2Mdudv + Ndv2

we now consider the partial derivatives

v d2X v d2X
■^UU -- O Q ) -^-uv --- o Q
OU0U OUOV

v d2x v d2x
— q a >
ovou
-A-vv — o o
ovov
of the vectors Xu and Xv.

Theorem 4.1.

(4-6)
(4-7)
(4-8)

Where

GEu-2Fu + FEv 2 2EFU — EEv — FE,

11 2(EG-F2) 11 = 2(EG - F2')
i GEV-FGU 2 EGU - FEv
12 2(EG - F2) 12 2(EG - F2)
2GFV - GGU - FG. 2 2FFV + FG,
22 2(EG - F2) 22 2(EG - F2)

The six T coefficients in these formulas are called Christoffel symbols.

 29

Proof. Since is a basis of R3, the second order derivatives of X should be a

linear combination of them.
We will write

Xuu — CX]XU + ct2Xv + a^n, (4-9)

Xuv — fiiXu. + (32XV + fan, (4-10)
Xuu — r'iiXu + 72 + 73n> (4-11)

Taking the dot product of each above equation with the unit normal vector n,
we have
Xuu • n = • n + (tyXu • n + a3n • n,

L — Xuu - n — O + O + C13 — CC3

similarly,
M'=/?3, N = 73

Recall that L, M and N are the coefficient of Second Fundamental Form, now, taking
dot product of 4.9,4.10,4.11 with Xu, Xv,

XUu * Xu — cx]Xu • Xu T oqXv • Xu T 0311 • Xu

= Eai + Fa% + 0

and
Xuu ’ xv — Faq T Gcx2

Since E = Xu ■ Xu, by differentiationary with respect to u and u,we have

On the other hand,

(4-12)

Therefore, we have
 30

Fai + Gqi2 — Fu — -Ev

solving the equations gives

_ GEU — 2FU + FEv
Q1 ~ 2(EG - F2)

2EFU — EEv - FEU

012 —
2(EG - F2)
similarly,
GEV - FGU
2(EG - F2)

&=r?2 =
EGU - FEv
2(EG - F2)
2GFV - GGU - FGV
71 1722 2(EG - F2)

_ t-,2 _ EGV — 2FFU + FGU

72 ~ 22 “ 2(EG - F2) ‘

The coefficient ai,a2) A,/?2>7i and 72 here are called Christoffel symbols T**,

! GEV—FGU
12 2(EG - F2)

2 _ EGU — FEv
12 _ 2 (EG - F2)
1 2GFV - GGU - FGV
22 2(EG - F2)
2 _ EGv - 2FFV + FGU
22 2(EG - F2)
□
 31

Chapter 5

Normal and Geodesic Curvature

Figure 5.1: Normal and Geodesic Curvature

Let C : X(s) (s is arc length) is a curve on surface S, then X is a unit vector

and a tangent vector to S. Hence, X is perpendicular to the unit normal n of S, so, X,
n and n x X are mutually perpendicular unit vectors. Again since X is perpendicular to
X, and hence is a linear combination of n and n x X:

X — Knn + KgH x X (5-1)

 32

Definition 5.1. The scalars nn and Kg in equation 6.1 are called the normal curvature
and the geodesic curvature of C, respectively.

Proposition 5.2. With above notations (see figure), we have

nn = X • n, (5-2)
, = X ■ (n x X), (5-3)
Kn = K COS </>, (5-4)
ng = ±/s sin (f> (5-5)

Where k is the curvature of C and </> is the angle between n and the unit principal normal
p of C.

Theorem 5.3. The geodesic curvature of Kg of a curve C on a surface S depends on the

first fundamental form of S only.

Proof. : Note that

X = Xuu J- Xvv

Respect to the arc length S of C, we have

X" — Xu-ufa d- J^uvUV T X^uitbu

TXvvbv T XuU T Xvv

Thus,

X x X = [Xuu + Xy v] x [Xuuuu + Xuvuv

+Xvuvu + Xvvvv + Xuv + Xyil]

By previous section, we have

Xu X X-y
n = T—-----—-

Xu X Xy — yfgn,

Xv x Xu yfgxi,
 33

Where, G = EG — F2 since

Xu x Xu = 0 Xv x X v = 0

nx n=0 n•n=1

Xu x Xy = fgn Xv x Xu — fgn

then

(X* x X) * n — [(X^iz T X-yv) x (Xuuuu T XUyUv

+Xvuvu + Xvvvv + Xuu + Xrii)] • n

Note that, in terms of Christoffel symbols,

Xuw = rjjXu + r'jjX.y + tun

xuv = rj2xu + r?2x„ + 6i2n

Xvu = r21xu + r^Xt, + 62111

xvv = r22xu+r^Xv + 62211

then,

(X x X) = (Xuu + Xvv) x (XMttuu + Xuvuv

+X^^V/U + XyyVV + XyU + Xyl})
~ Xu x Xuu(u) + Xu x XUy(u) v

+XU X Xyy(u) V + Xy X XyyU(v)

~t~Xy x Xyuii + Xu x Xyitv + Xy x X^^'w(tt)

+X"V x Xuvu(y) + Xv x Xvuu(v)

+XV x Xw(v)3 + Xv x Xuvu + Xv x Xvvv

-- Xy X XyUV Xy X XyVU + Xy X Xyy(u) + Xy X Xyy(v)

+XW X Xyy(u) V + Xy x XyUv(u)

+Xy X XUVU(V)2 + Xy X Xyy(v)3
 34

Xu x Xy('iw) Xu x Xvvu T Xu x (I\iXy

+r^Xy + &iin)(,u)3 + Xy x (r22Xy + r^2Xy + 62211) (u)3

+xu x (rj2Xu + r22Xu + 61211)(u)2v
+Xy X (I^Xy + I^Xy + 61in)'6('U)2

+Xy x (r}2Xn + r22Xy + 6i2n)u(v)2

+xv X (r£2xu + r22xv + 62211)(v)3

= Xu x Xvuv — Xux Xvi)u + XltXy(u)3r21 + Xu x &n(u)3n

+XV x Xu(i>)3r22 + Xv x 622(v)3n + Xu x Xy(u)2vr22

+Xy x (u)2vb22n + Xv x Xu^u)2?!! + Xv x 6n?;(u)2n

+XU x Xvvu + Xu x XiXu)3?2!

+XU x 6ii(t4)3n + Xy x Xu(v)3Tz2

+X„ x 622 (6) 3n + Xu6i2(u2vn + Xv x Xy^)^)2?^

+XV x 6ni>(ti)2n + Xy x Xuu(v)2r}2 + Xy x 6i2,u(,u)2n

+Xy x Xu(v)3r|2 + X„ x 622('i’)3n

now, since

Xy X Xy = 0, Xy X Xy = 0
Xy x Xy = yfgn, Xv X Xu = ^gn

simplify above, we have

X x X = y/gnuu + ^/t7nur22?w
+A/5n'iir2iV?i + y/gnuT^vv — y/gnvT^uu
—y/gnir^QU^ ~~ i/gnvr21vu — y/gnvT^^

+y/gnuv — y/gnvu + (Xy«6nti/u

+Xu?i62it’it + Xuub22vv + Xvvbiiuu

+Xvvbi2uv + Xvvb2ivu + X2vb22vv) xn

Note-, all ubn(u)(u) ... are constants. Since

nxn=0
 35

fen = Xu • n, fex2 = X12 • n

&21 = -X21 • n, 622 = X22 ■ n

so in above expression, all terms cross multify with n become zero. Thus,

X x X = y/gnuu + y/gnur22uv + y/gnurfvu + y/gnur22vv

—fgnvrfuu — y/gnvrj2uv — y/gnvrfvu

—fgnvT22vv + fgnuv — y/gnvu

Since n • n = 1, then, we can have

(X x X) • n = y/guu + y/gurfy™ + x/guT^feu

+y/guT22vv ~ y/gvTiiUU ~ y/gvl^uv

-y/gvl^vu - y/gvF22vv + y/guv - y/gvu

since
p2 __p2 pl __pl
1 12 — 1 21 > 1 12 ~ 1 21

then,

k9 = (X x X) -n

= V5[r?i(fr)3 + r?2(u)2u + rl1(u)2u

+r22(fe)2u - in(ii)2v - r]2(u)(v)2r^(v)2u

—r22(fe)3 + uv — feu]

so

*9 = v£[r?i(u)3 + (2T22 - rh)(u)2fe - (2T}2 - r|2)u(fe)2 - r*2(fe)3 + uv - ufe]. (5.6)

Thus, Kg is only depend on the first fundamental form since the Christoffel

symbols is only depend on the first fundamental form.

□

Ge
 36

Chapter 6

Geodesic and Geodesic Equations

6.1 Definition and Basic Properties

Definition 6.1. A curve C on a surface S is called a geodesic if its geodesic curvature
is zero everywhere.

Proposition 6.2. Every straight line is a geodesic.

Proof. Assume that a straight fine in S

X(s) =p + sq

Which the arc length as parameter.

Then X = 0,
Kg = X • (n x X) = 0

So it follows that X is a geodesic. □

6.2 Derive the Geodesic Equations

Recall
K = X = Knn + Kg(n x X)

Kg =0

X ±X, X -X = 0.
 37

If C is geodesic, then ng = 0. It implies

X = nnn.

i.e., X is parallel to the normal vector of the surface.

Therefore, C is a geodesic, if and only if X is perpendicular to Xu and Xv. i.e.,

Kg = (X x X) • n = 0.

Let
X — Xuu T Xvv

then
X = -^(Xuu + Xvv)

— Xuuuu “I- Xuvuv “I- Xuii d- Xvuuv 4- XwVV 4~ X^v

= uXu + (u)2Xuu + 2uvXuv + vXv + (v)2Xvv (6-1)

By (4.6), (4.7), (4.8)

xuu = rjx^ + r2jXr + Ln

Xuv = Tj2-^-u + + Mn

Xvv — T^jX^ + T^X^ T Nn

Apply them to equation (6.1)

X = uXu + («)2(rjXu + TfiX + Ln) + 2w(Tl2Xli + rf2X + Mn)

+vXv + (u)2(r^Xu + ri2X + Nn)

Recall that
L — 6ii = Xuu * ri

A7 — 612 — 621 — Xuv * n = Xuu * n

N = 622 = Xvv • n.

Then

X — uXu + ('ii)2(rj:iAlt + T2jXr + Xuun • n)

+2't4'6(rj2Xu + f^Xv + Xuv'n ■ n)

-yvXv + (^(r^ + r^2xr + XvvR •n)

= fiXu + f2Xv + f • n
 38

where

fi = u + (u)2r}1 + 2wrj2 + (*>)2r22 (6-2)

/2 = i) + (u)2?2! + 2wIT22 + (<’)2r22 (6-3)
f = (ii)2Xuli • n + 2uvXuv ■ n + (v)2X22 • n (6-4)

then
X x X = (uXu + vXy) x (JiXu + f2Xv + /n)

= (uXj x faXu + (uXu) x f2Xv + (uXi) x /n

~\~uXv x fi_Xu 4~ vXv x /2-Yv 4“ vXu x fn

now,
Kg = (X x X) -n

and recall

XuxXu = 0, Xv xXv = 0

(Xu x n) • n = 0

(Xv x n) • n = 0

Xu X Xv — y/gri

Xu x Xu — yfgn..

So, combining above, we have

Kg = (X x X) • n

= [(uX„) x 71 Xu] • n 4- [(uXu) x f2Xv] ■ n

4- [(uXu) x f ■ n] • n
4- [vXd x /iXu] • n 4- [i>Xr x f2Xv] ■ n
4- [vX„ x f • n] • n

= uf2[(XU x XJ • n] 4- v/iKX,, x Xu) • n]

= («/2-i/i)[(XuXX,)-n].

Then, k9 = 0 if and only if

uf2 - vfi = 0
 39

thus
Ufa = vfl

Now, considering X • X

XX = ufffXuXf) + uf^) + vf2(XvXv) + vfffXvXu)

= ufffXuXu) + u(XuXv) + vfo(XvXv) + vh(XvXu)

then

u • (X • X) = (u)2fi(XuXu) + u^-(XuXv) + v^(XvXv) + vfffXvXu)

u u
= /1[(«)2(XUXU) + vfffXM + (v)2(X,X„) + v(X,Xu)J
= fi(uXu + vXv)(uXu + iXj

= fi(X-X)

= fl

thus
X-X = l

«(x • x) = A

similarly,
v(X-X) = /2

therefore,

/ts = 0

<=>

ufz = vfi
 40

fi=u(X-X)

h = v(X • X)

Since,
X-X = 0

so
Kg = 0 fi = 0 and f2 = 0

i,e.,

u + (u)2r^ + 2w)rJ2 + (^)2r^ = o (6.5)

v + (u)2r^ + 2wH22 + (u)2r^2 = 0 (6.6)

We have proved the following theorem on geodesics.

Theorem 6.3. Let C : Xs = X(u(s),v(sf) be a curve on the surface with parametric

equation X(u,v). Then C is a geodesic if and only if the equations 6.5 and 6.6 hold.

(Therefore, (6.5) and (6.6) are called the geodesic equations).

 41

Chapter 7

A Surface of Revolution

Definition 7.1. : A surface S generated by a given plane curve C rotating about a fixed
straight line A is called a surface of revolution. A is called the axis of S.C is called the
profile curve.

7.1 The Parametric Representation of a Surface of Revo­

lution
Taking the aixs of rotation to be the Z-axis, the plane to be the xz— plane. By
the definition of the Surface of Revolution, we have that any point, say p of the surface
is obtained by rotation some q of the profile curve through an angle say u around the
z-axis. Now, if
C = (/(v),0,5(v))

is a parametrization of the profile curve containing q, then q is out of the form (see figure)

X(u,v) = (/GO cosu, f(v) sinu, h(vf)

 42

Figure 7.1: Surface of Revolution

7.2 First and Second Fundamental Form of a Surface of

Revolution
7.2.1 The First Fundamental Form of a Surface of Revolution

Recall: The first fundamental form of a surface is,

Edu2 + 2Fdudv + Gdv2

where
E —■ Xu • Xu, F = Xy • Xy,

let
X(u, v) = (/(v) cos u, /(v) sin u, h(v)) (7.1)

be a surface of revolution.

Then
Xu = (—/sin u,f cosu, 0)
 43

dh(y)
Xv = (f' cos u, f' sin u, h')
dv
Where f' denoting

E = Xu- Xu = ((-f sin u)(-f sin u) + (/ cos u)(f cos u) +0)

= (y2 sin2 u+ f2 cos2 u + 0)

=f

and,
F = Xu • Xv = (—f sin u cos u + f sin u cos u + 0) = 0

G = Xv • Xv = (f'2 cos2 u + f 2 sin2 u + h'2) = f 2 + h'2.

So, for a surface of revolution, the first fundamental form has coefficients:

E = f\

F = 0, (7.2)
G = f'2 + h'2

7.2.2 The Second Fundamental Form of a Surface of Revolution

Recall: The second fundamental form of a surface:

Ldu2 + 2Mdudv + Ndv2

where,
 44

Now, for the surface of revolution

XUu = (-/ cos u, - f sin u, 0) (7-3)

X-uv = (—f sin u, f cos u, o) (7-4)
Xvv = (f' cos u, f" sin u, It ) (7-5)

and,
i j k
Xy X Xy — —/sinu f cosu 0
f cos u f sin u f

f cos u 0 —/sinu o —/sinu / cosu

f ,f i- t 1 3+ k
f sinu h f sinu h / cosu / sinu

=(fh' cosh)? + (fh' sinu)/ + (—ff sin2u — ff cos2u)k

=(fh' cosu, fh' sinu, —ff)

We assume that f(v) > 0 and that the profile curves i—> (f(v), 0, h(v)) is unit speed, i.e.
/'2 + = L Then
X —

yj(fh' cosu)2 + (/// sinu)2 + (-ff)2

= y/f2h'2 + (ff)2

= fVf2 + h’2

=f

(fh' cos u, fh' sin u, —ff)

n= = (h cos u, h' sin u, f )
f
So, the second fundamental form of a surface of revolution is:

L = Xuu • n = (—/ cos u, —f sin u, 0) • (h' cos v, if sin v, f)

 45

= ~fh'.

Samilaly,
M = Xuv • n = (—f sin u, f' cos u, 0) • (Ji cos v, li sin v, f')

= (—f'h' sin it cos « + f'h' sinucosu + 0) = 0

N = Xvv • n = (/ cos u, f sin u, h' )(ft cos u, h' sin u, f')

= f'h'1 - f"h'

thus
(/ h — f h )du2 + fh'dv2

is the second fundamental form of surface of revolution.

7.3 The Christoffel Symbols for a Surface of Revolution

Recall the formulas for Christoffel Symbols from the previous chapter:

pi GEU — 2FU + FEv _ 2EFU - EEv - FEU

11 _ 2(EG - F2) 11 _ 2(EG - F2)
! GEV - FGU 2 _ EGU — FEv
12 2(EG - F2) 12 “ 2(EG - F2)
nl _ 2GFV - GGU - FGV p2 _ EGV — 2FFV + FGU
22 2(EG - F2) 22 = 2(EG - F2)

note that the Christoffel symbols only depend on the coefficients of the first fundamental
form. For a surface of revolution, the coefficients of first fundamental form are the
following:
E = f2, F = 0, G = f'2 + h'2 = 1

Replace these first fundamental form of surface of revolution into the previous Christoffel
symbols formula:
 46

! _o-o+o
111- 2/2
ri _ (Z2)/ _ 1
12 2/2 f
ri -Q-o-O
X22- 2/2
-o

0 —/22/— 0
1 11 - 2^2 ”

•n2
p2 /2(0 T 26 h") —0 + 0 = h'h"
J- 22
1 9' —
2/2
r?2 = o

thus

r?2 = o

are the Christoffel symbols of the surface of revolution.

 47

Chapter 8

Geodesic of a Surface of
Revolution

In this chapter, we are going to study the following question: what are the
geodesics on a surface of revolution.
Recall that a surface of revolution obtained by revolving a curve C in XZ—plane
about Zaxis can be parametrized by an equation X(u,v) = (f(v) cosu, f(v) sinu,g(y)),
where (f(v),g(v)) is the parametric equation for the curve C in xz — plane.
In the following, we will assume that C is given by a function of x in xz —plane,
namely Z = h(x).
Then, the parametric equation for the surface can be written as X(u,v) =
(v cos u, v sin u, h(y)).

8.1 Geodesic Equations of a Surface of Revolution

Now, let S be a surface of revolution with the parametric equation

X(u, v) = (v cos u, v sin u, h(v))

We find out
E = v2, F = 0, G = 1 + h"

—vh' -h"
M=0
 48

r]i = o, r]2 = i

^22 = 0, = —v,
p2 k>

22 1 + 7/2

Let a(s) = (v(s) cosu(s),v(s) sinu(s), h(y(sf) be a curve on S.

By the theorem (6.3), a(s) is a geodesic if and only if

v , .,9 „ „ h'h" , ,9.

“-rn/sM +2-°+t7^w)=°

u+0+2• •v+0=0

v • (u)2
1 + h'2 + 1 + 7/2 = 0

i,e., Therefore, the geodesic equations for a surface of revolution curve;

•• 2 ..
u H—uv = 0 (8.1)
v

v • (u)2 tililv)2
=0 (8-2)

if u(s) = constant, then it = 0. The equation (8.1) will obviously holds.

The equation (8.2) will be studied in the following.
First for a curve, on any surface in general:

X(s) = X(u(s), v(s))

we have,
X-X = 1

where
X — Xuu -|- Xvv
 49

therefore,
X • X = (Xuu + Xvv)(Xuu + Xvv)

= (Xu • Xu)(u)2 + 2(XU ■ Xv)uv + (Xv • Xv)(v)2

recall that

are the coefficients of the first fundamental form of a surface of revolution, And, for the
surface of revolution,

E = (u)2, F = 0, G = 1 + h'2 (h' =

dv

then , we have that:

X • X = (u)2 • (u)2 + (1 + h'2)(y)2 = 1. (8.3)

When u(s) is constant,

(u)2 • (u)2 = 0

then (8.4) implies that:

(1 + h'2)(v)2 = 1 (8.4)

that is
(1 + 6'2)(v)2 = (v)2 + (v)2 ■ K2 = 1

differentiating it with respect to the arc length s, we have the following,

2v(y + till' v + h'2v) = 0

Since u(s) here can not be constant for an actual curve,

then the equation (8.5) implies:

v + h h v + h 2v = 0

or
15(1 + h'2) + h'h"v = 0. (8-5)

Now, considering the second geodesic equation (8.2) of a surface of revolution

v ■ (u)2 h’h"(y)2
=0
 50

when u is constant, it becomes,

h'h"(v)2
v—0+ =0
1 + h'2

multiplying both sides by 1 + h,'2, we have that

v(l + h2) + h h (v)2 = 0

therefore, we have just proved the following theorem.

Theorem 8.1. On a surface of revolution S : X(u,v) = (ycosu,vsinu, h(yf) all the

v-curves (i.e., u — constant) are geodesic.

A u-curve on S is also called a meridian which is basically a rotation of the

profile curve C about z-axis. In the mean time, a u-curve is called a parallel and all
parallel are circles.
Therefore, the theorem 8.1 states that all meridians on a surface of revolution
are geodesic.
On the other hand, if v(s) is constant for a curve a(s),then the geodesic equations
(8.1) and (8.2) imply that u — 0, i.e.,u(s) must be constant. Hence, a parallel on a surface
of a revolution is not a geodesic.

8.2 Examples of Geodesic of a Surface of Revolution

In the following, we take a close look at some examples of surface of revolution.
Example 1 Geodesics on a Sphere
We consider the upper half sphere of radius r centered at (0,0,0) as a surface of revolution
by revolving a quarter of circle in xz—plane by z-axis.
Then a meridian is part of a great circle on the sphere. Therefore, by the theorem 8.1,
all the great circle on a sphere are geodesic.
If c = 0 7^ a, then X is a circle around the z — axis.

Example 3 Geodesic on right circular Cone

 51

Figure 8.1: Geodesics on Sphere

Figure 8.2: Geodesic on Cone

 52

A right circular Cone can be realized as a surface of revolution by revolving a

half line z = ax in sz-plane about z-axis, x > 0, therefore, it has a parametric equation
of X(u, v) = (v cos u, v sin u, av).
(That is, h(v) = av in our general notation.)
Meridians on a cone are those straight edges which are the geodesics, by the
theorem (8.1).
Parallels are the circles around the cone which are not geodesics.
We also observe that if one cuts the cone along its edge, the cone unwrap into a
sector of the Euclidean plane. Therefore, the geodesics on the come should yield straight
line segments in the sector.
It is clear that the unwrapping of a parallel (a circle) on the cone is not a straight
line segment in the sector. It shows that parallels on a cone are not geodesics.
 53

Bibliography

[Car76] Manfredo Perdigao Carmo. Differential Geometry of Curves and Surfaces.

Prentice-Hall, New Jersey, 1976.

[E.K91] E.Kreyszig. Differential Geometry. Dover Publications,Inc., New York, 1991.

[Hsu97] Chuan-Chih Hsuing. A First Course in Differential Geometry. International

Press, Boston, 1997.

[Lip69] Seymour Lipschutz. Differential Geometry. McGraw-Hill, New York, 1969.

[O’N66] Barrett O’Neill. Elementary Differential Geometry. Academic Press Inc.,San

Diego, 1966.

[Opr97] John Oprea. Differential Geometry and its Applications. Prentice-Hall Inc, New
Jersey, 1997.

[PrelO] Andrew Pressley. Elementary Differential Geometry. Springer, New York, 2010.