Differential Geometry of Curves and Surfaces, 1st Edition: Thomas Banchoff, Shiing-Shen Chern, and William Pohl

Thomas Banchoff, Shiing-Shen Chern, and
William Pohl
Differential Geometry of Curves

and Surfaces, 1st Edition
SPIN C346 Differential Geometry Banchoff/Chern/Pohl
Monograph – Mathematics –
July 17, 2002
Springer
Berlin Heidelberg NewYork
Barcelona Hong Kong
London Milan Paris
Tokyo
Table of Contents
Part I Introduction
1 Review of Euclidean Geometry . . . . . . . . . . . . . . . . . . . . . . . . . . . 3

1.1 Motions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.2 Vectors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
2 Curves . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
2.1 Arc Length . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
2.2 Curvature and Fenchel’s Theorem . . . . . . . . . . . . . . . . . . . . . . . . 13
2.3 The Unit Normal Bundle and Total Twist . . . . . . . . . . . . . . . . . 16
2.4 Moving Frames . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
2.5 Curves at a Non-inflexional Point and the Frenet Formulas . . 21
2.6 Local Equations of a Curve . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
2.7 Plane Curves and a Theorem on Turning Tangents . . . . . . . . . 26
2.8 Plane Convex Curves and the Four Vertex Theorem . . . . . . . . 30
2.9 Isoperimetric Inequality in the Plane . . . . . . . . . . . . . . . . . . . . . . 31
3 Fundamental Forms of a Surface . . . . . . . . . . . . . . . . . . . . . . . . . . 35

3.1 The First Fundamental Form . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
3.1.1 Geometry in the Tangent Plane . . . . . . . . . . . . . . . . . . . . 39
3.2 The Second Fundamental Form . . . . . . . . . . . . . . . . . . . . . . . . . . 40
3.2.1 The Shape of a Surface . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
3.2.2 Characterization of the Sphere . . . . . . . . . . . . . . . . . . . . . 44
3.2.3 Principal Curvatures, Principal Directions, and Lines
of Curvature . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
3.3 Gauss Mapping and the Third Fundamental Form . . . . . . . . . . 47
3.4 Triply Orthogonal System; Theorems of Dupin and Liouville . 47
4 Fundamental Equations in Surface Theory . . . . . . . . . . . . . . . . 53

4.1 Weingarten and Gauss Equations . . . . . . . . . . . . . . . . . . . . . . . . . 53
4.2 Levi-Civita Parallelism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56
4.3 Integrability Conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
4.4 The Congruence Theorem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66
VI Table of Contents
5 Global Surface Theory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71

5.1 Moving Frames . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71
5.1.1 INSERT TITLE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71
5.1.2 INSERT TITLE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72
5.1.3 INSERT TITLE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75
5.2 The Gauss-Bonnet Theorem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
5.2.1 INSERT TITLE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
5.2.2 Theorems of Jacobi and Hadamard . . . . . . . . . . . . . . . . . 79
5.2.3 Vector fields on a surface . . . . . . . . . . . . . . . . . . . . . . . . . . 81
5.3 Rigidity Theorems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81
5.3.1 Elliptic W-surfaces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82
5.3.2 Integral formulas; second proof of Liebmann’s theorem. 85
5.3.3 Cohn-Vossen’s rigidity theorem . . . . . . . . . . . . . . . . . . . . . 86
6 INSERT TITLE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89
6.1 Geodesics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89
6.1.1 INSERT TITLE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89
6.1.2 INSERT TITLE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90
6.1.3 INSERT TITLE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91
6.2 Normal Coordinates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94
6.2.1 INSERT TITLE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94
6.2.2 INSERT TITLE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100
6.3 The second variation of arc length . . . . . . . . . . . . . . . . . . . . . . . . 102
6.3.1 INSERT TITLE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102
6.3.2 INSERT TITLE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105
6.3.3 INSERT TITLE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106
6.3.4 INSERT TITLE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 108
6.3.5 INSERT TITLE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 114
Part I
Introduction
1 Review of Euclidean Geometry
1.1 Motions
Three-dimensional Euclidean space consists of points which have as coordi-
nates ordered triples of real numbers x1 , x2 , x3 . In vector notation, we write
x = (x1 , x2 , x3 ). The distance between two points is given by the formula
v
u 3
uX
d(x, y) = t (xi − yi )2 . (1.1)
i=1
Note that d(x, y) ≥ 0 and d(x, y) = 0 if and only if x = y.

An affine transformation T from R3 to R3 is defined by T (x) = Ax + b,
where A is a 3 × 3 matrix and b is a vector in R3 . An affine transformation
that preserves distance between points, so that d(T (x), T (y)) = d(x, y) for
all x and y is called a motion of R3 .
Proposition 1. An affine transformation is a motion if and only if A is an
orthogonal matrix, i.e. a matrix with columns that are mutually perpendicular
unit vectors.
Proof. Let the points (x1 , x2 , x3 ) and (y1 , y2 , y3 ) be mapped by an affine
transformation T into the points (x01 , x02 , x03 ) and (y10 , y20 , y30 ) respectively, so
that
3
X
x0i = aij xj + bi
j=1
3
X
yi0 = aij yj + bi ,
j=1
where aij denotes the entries of the matrix A and the bi denotes the
components of b. If we subtract these two equations, we get
3
X
x0i − yi0 = aij (xj − yj ) .
j=1
4 1 Review of Euclidean Geometry
Taking the sum of the squares gives us

2
X X
(x0i − yi0 ) = aij aik (xj − yj ) (xk − yk ) ,
i i,j,k
where all the indices run from 1 to 3. This equality will only hold true if
3
X
aij aik = δjk , (1.2)
i=1
where δjk = 1 if j = k and 0 otherwise.

P3
Given an affine transformation x0i = j=1 aij xj + bi , we can solve ex-
P3
plicitly for the xi in terms of the x0i . We first set x0i − bi = j=1 aij xj then
multiply by aji and sum over i and j to get
3
X
aji (x0i − bi ) = xj .
i=1
Remark 1. The quantities δjk defined in (1.2) are called Kronecker deltas.
We have illustrated their usefulness in the above proof, and they will be used
consistently.
It will be convenient to introduce the matrices
   
a11 a12 a13 a11 a21 a31
A =  a21 a22 a23  AT =  a12 a22 a32  .
a31 a32 a33 a13 a23 a33
The second is obtained from the first by interchanging the rows and columns
and is called the transpose of A. Using this notation, (1.2) can be re-written
as
AT A = I (1.3)
where I denotes the unit matrix (δij ). A matrix A with this property is called
orthogonal.
We may rewrite the definition of a motion in terms of matrices as
x0 = Ax + b where x is the column matrix with entries xi . We may then
solve explicitly for x in terms of x0 by writing x0 − b = Ax , so
AT (x0 − b) = AT (Ax)
= (AT A)x
= Ix = x.
1.1 Motions 5
A basic result in linear algebra states that, for square matrices,
(AC)T = CT AT ,
where the order of the multiplication is important. From this it follows that
if A and C are orthogonal matrices, then
(AC)T AC = CT AT AC = CT C = I
so the product AC is also an orthogonal matrix. For an orthogonal matrix A,

we have A−1 = AT . Moreover, AT (A−1 )T = (A−1 A)T = I T = I so (A−1 )T =
(AT )−1 . It follows that if A is orthogonal, then so is A−1 . A collection of
matrices that is closed under multiplication and such that the inverse of
every element of the collection is also in the collection is called a group.
The succesive application of two motions, as in (1.3) above, is called their
product. This multiplication is in general not commutative. It is easily seen
that all of the motions in E form a group under this multiplication, called
the group of motions. Euclidean geometry studies the properties of E that
are invariant under the group of motions.
From (1.3) we find (det A)2 = 1 so that det A = ±1. The motion is called
proper if the determinant is +1, and improper if it is −1. It is easily verified
that the product of two proper motions is a proper motion, and it is a simple
result that all proper motions form a subgroup of the group of motions.
Example 1. The mirror reflection, (x1 , x2 , x3 ) → (−x1 , x2 , x3 ), is an improper
motion.
A motion of the form x0i = xi + bi , for i = 1, 2, 3 is called a translation. A
motion of the form
3
X
x0i = aij xj
i=1
where j = 1, 2, 3 is called an orthogonal transformation. In matrix form, a

translation can be written x0 = x + b and an orthogonal transformation can
be written x0 = Ax. An orthogonal tranformation is called proper or improper
according to the sign of det A. A proper orthogonal transformation can also
be called a reflection. All of the translations form a group, as do all of the
othogonal transformations and all the rotations. The group of all rotations
can be characterized as the subgroup of all proper motions leaving the origin
fixed.
Exercise 1. Prove that the quadratic polynomial

X
αij ξi ξj
i,j
where αij = αji is zero for all ξi if and only if αij = 0. Show that this is not
true without the symmetry condition on the coefficients αij . This result is
used in the proof of (1).
Exercise 2. Show that the inverse motion of x0 = Ax + b is x0 = A−1 x −

A−1 b . Let Ti be the motions x0 = Ai x + bi where i = 1, 2 and show that
T1 T2 is the motion x0 = A2 A1 x + A2 b1 + b2 .
Exercise 3. Let G be a group with the elements {e, a, b, . . . }, where e is the

unit element. The left and right inverses of an element a are defined by
a−1 −1
l a = e, and aar = e,
respectively. Prove that a−1

l = a−1r . Observe that the equivalence of the
T T
conditions AA = I and A A = I means group-theoretically that the matrix
A has the same right and left inverse, which is AT .
Exercise 4. Prove that the translations form a normal subgroup of the group
of motions, while the rotations do not.
Exercise 5. Show that the helicoidal motions
x01 = cos(t)x1 + sin(t)x2

x02 = − sin(t)x1 + cos(t)x2
x03 = x3 + bt,
where b is a constant and t is a parameter, form a group. Draw the orbit of

the point (a, 0, 0), and distinguish the cases when b < 0 and b > 0.
Exercise 6. Prove that the rotation

X
x0i = aij xj
j
where i, j = 1, 2, 3 and det(aij ) = 1 has a line of fixed points through the ori-
gin, the axis of rotation. Hence prove that the group of rotations is connected.
Prove also that the group of orthogonal transformations is not connected.
(Note: A subgroup of motions is connected if any two points can be joined
by a continuous arc.)
1.2 Vectors
Two ordered pairs of points, p(x1 , x2 , x3 ), q(y1 , y2 , y3 ) and p0 (x01 , x02 , x03 ),
q 0 (y10 , y20 , y30 ) are called equivalent if there is a translation T which maps p
to p0 and q to q 0 . The last property can be expressed by the conditions
x0i = xi + bi and yi0 = yi + bi , where i = 1, 2, 3. It follows that a necessary and
1.2 Vectors 7
sufficient condition for the equivalence of the two ordered pairs of points is
yi0 − x0i = yi − xi . Such an equivalence class is called a vector. We denote the
vector by V = − → and call
pq
vi = yi − xi , (1.4)
where i = 1, 2, 3 denote its components. A vector is therefore completely

−
→ −→
determined by its components. Geometrically pq = p0 q 0 if and only if the
0 0
segments pq and p q are of the same length and parallel in the same sense.
Using the origin O of our coordinate system, we can set up a one-to-one
−→
correspondence between the points p of R3 and the vectors Op. The latter
will be called the position vector of p. Notice that it is defined with reference
to the origin O.
Given two vectors v = (v1 , v2 , v3 ) and w = (w1 , w2 , w3 ) their sum is
v+w = (v1 +w1 , v2 +w2 , v3 +w3 ), and multiplication of a vector by a number
is defined by λv = (λv1 , λv2 , λv3 ). Throughout the book real numbers are
sometimes called scalars, in order to emphasize their difference from vectors.
By (1.4) we see that under a motion the vectors are transformed according
to the equations
X
vi = Aij vj (1.5)
j
where i, j = 1, 2, 3, and v0 = (v10 , v20 , v30 ) is the image vector. Using (1.5) and (1)
we get
0 0 0
v12 + v22 v32 = v12 + v22 + v32 .
This leads us to define
v2 = v12 + v22 + v32 . (1.6)

√
We call + v2 the length of v. Thus, the length of a vector is invariant under
motions.
More generally, we find
1 2 2 2
P
2 {(v + w) − v − w } = vi wi ,
i
where i, j = 1, 2, 3. Since the left-hand side involves only lengths of vectors,
which are invariant under motions, the same property holds for the expres-
sions at the right-hand side. Generalizing (1.6), we introduce the notation
X
v·w =w·v = vi wi ,
i
which is to be called the scalar or dot product of v and w.

Relative to addition and scalar multiplication of vectors, the scalar prod-

uct has the following properties: (v1 + v2 ) · w = v1 · w + v2 · w and
(λv) · w = v · (λw) = λ(v · w) where λ = scalar. The relation
(v + λw)2 = v2 + 2λv · w + λ2 w2 ≥ 0
is true for all λ. So by elementary algebra we get
v2 w2 − (v · w)2 ≥ 0, (1.7)
which is called the Cauchy-Schwartz inequality. The angle θ between v and w

is defined by
v·w
cos θ = √ .
v2 w2
This is meaningful because by the Cauchy-Schwartz inequality the right-
hand side has absolute value ≤ 1. The vectors v and w are perpendicular or
orthogonal if v · w = 0.
The determinant of three vectors u, v, w with components ui , vi , wi for
i = 1, 2, 3 respectively, is defined by

u1 u2 u3

det(u, v, w) = v1 v2 v3 .
w1 w2 w3
Under the transformation in (1.5) the determinant (u, v, w) is multiplied by

det A = det (aij ). Hence it is invariant under proper motions and changes its
sign under improper motions. The following properties follow immediately
from the definition:
(u + u1 , v, w) = (u, v, w) + (u1 , v, w),

(λu, v, w) = λ(u, v, w) where λ = scalar,
(u, v, w) = −(v, u, w) etc.
The vector product of two vectors v, w is the vector z such that the
relation
(v, w, x) = z · x (1.8)
holds for all vectors x. It follows that z has the components
z1 = v2 w3 − v3 w2 , z2 = v3 w1 − v1 w3 , z3 = v1 w2 − v2 w1 . (1.9)
We write
1.2 Vectors 9
z=v×w .
The vector product has the following properties:
v × w + w × v = 0,
(v1 + v2 ) × w = v1 × w + v2 × w,
(λv) × w = λ(v × w) where λ = scalar.
The vector z can be described geometrically as follows: We see from (1.9)

that v × w = 0 if and only if one of the vectors v, w is a multiple of the
other. In this case we say that v and w are linearly dependent. Suppose next
that v × w 6= 0, i.e., v and w are linearly independent. Putting x = v, w
successfully in (1.8), we get z · v = z · w = 0, so that z is orthogonal to both
v and w. We write z = λu, where u is a unit vector orthogonal to v and w.
Thus from (1.8) we get
(v, w, u) = z · u = λu2 = λ .
Hence we have
z = v × w = (v, w, u) u 6= 0 .
The unit vector u is defined up to its sign and we can choose u such that
(v, w, u) > 0. z is therefore a multiple of u and of length (v, w, u). This
completely determines z.
Three vectors u, v, w are called linearly dependent if (u, v, w) = 0; other-
wise they are linearly independent. An ordered set of three linearly indepen-
dent vectors is called a right-handed or left-handed frame according to the sign
of its determinant. The property of right-handedness or left-handedness of a
frame remains unchanged under proper motion, while they interchange un-
der an improper motion. Also a right-handed (or left-handed) frame becomes
left-handed (or right-handed) when any two of its vectors are interchanged.
Remark 2. The importance of vectors in analytic geometry is due to the alge-

braic structure. Two vectors can be added and a vector can be multiplied by
a scalar. It is important to observe that corresponding operations are mean-
ingless on the points of R3 , because they are not invariant under motions.
Exercise 7. a) The vector equation of a line is x(t) = at + b, where a, b =

constant and a 6= 0. Find its angles with the coordinate axes.
b) The vector equation of a plane is a · x = b where a 6= 0. Give the
geometrical meaning of b when a is a unit vector.
c) The vector equation of a sphere is (x − a)2 = r2 . What are a and r?
In each case draw the relevant figure.
Exercise 8. Prove that x(t) = A cos t + b sin t for A, b 6= 0, represents an

ellipse.
Exercise 9. Let yi , for i = 1, 2, 3, be three linearly independent vectors.

Prove that any vector x can be written
X
x= λi yi .
i
Hence prove that x = 0 if and only if x · yi = 0.
Exercise 10. Let (u, v, w) = 0, u × v 6= 0. Prove that w is a linear com-

bination of u and v, i.e., w can be written w = λu + µv, where λ, µ are
scalars.
Exercise 11. Prove Lagrange’s identity:
(v × w) · (x × y) = (v · x)(w · y) − (v · y)(w · x) .
Hence prove that
(v × w) × x = (vx)w − (wx)v .
Hint. To prove the first equation, write out both sides in components.
2 Curves
2.1 Arc Length

A parametrized curve in Euclidean three-space e3 is given by a vector function
x(t) = (x1 (t), x2 (t), x3 (t))
that assigns a vector to every value of a parameter t in a domain interval

[a, b]. The coordinate functions of the curve are the functions xi (t). In order
to apply the methods of calculus, we suppose the functions xi (t) to have as
many continuous derivatives as needed in the following treatment.
For a curve x(t), we define the first derivative x0 (t) to be the limit of the
secant vector from x(t) to x(t + h) divided by h as h approaches 0, assuming
that this limit exists. Thus,

x(t + h) − x(t)
x0 (t) = lim .
h→0 h
The first derivative vector x0 (t) is tangent to the curve at x(t). If we think
of the parameter t as representing time and we think of x(t) as representing
the position of a moving particle at time t, then x0 (t) represents the velocity
of the particle at time t. It is straightforward to show that the coordinates of
the first derivative vector are the derivatives of the coordinate functions, i.e.
x0 (t) = (x1 0 (t), x2 0 (t), x3 0 (t)) .
For most of the curves we will be concerning ourselves with, we will make
the “genericity assumption” that x0 (t) is non-zero for all t. (MISSING SEC-
TIONS) lengths of polygons inscribed in x as the lengths of the sides of these
polygons tend to zero. By a theorem of calculus, this limit can be expressed
as the integral of the speed s0 (t) = |x0 (t)| between the parameters of the
end-points of the curve, a and b. That is,
v
Z b Z bu
uX 3
s(b) − s(a) = |x0 (t)| dt = t x0i (t)2 dt .
a a i=1
For an arbitrary value t ∈ (a, b), we may define the distance function
12 2 Curves
Z t
s(t) − s(a) = |x0 (u)| du ,
a
which gives us the distance from a to t along the curve.

Notice that this definition of arc length is independent of the parametriz-
tion of the curve. If we define a function v(t) from the interval [a, b] to itself
such that v(a) = a, v(b) = b and v 0 (t) > 0, then we may use the change of
variables formula to express the arc length in terms of the new parameter v:
Z b Z b Z b
|x0 (t)| dt = |x0 (v(t))| v 0 (t) dt = |x0 (v)| dv .
a a a
We can also write this expression in the form of differentials:

ds = |x0 (t)| dt = |x0 (v)| dv.
This differential formalism becomes very significant, especially when we use
it to study surfaces and higher dimensional objects, so we will reinterpret
results that use integration or differentiation
qP in differential notation as we go
0 3 0 2
along. For example, the statement s (t) = i=1 xi (t) can be rewritten as
2 3 2
ds X dxi
= ,
dt i=1
dt
and this may be expressed in the form

3
X
2
ds = dx2i ,
i=1
which has the advantage that it is independent of the parameter used to

describe the curve. ds is called the element of arc. It can be visualized as the
distance between two neighboring points.
One of the most useful ways to parametrize a curve is by the arc length
s itself. If we let s = s(t), then we have
s0 (t) = |x0 (t)| = |x0 (s)| s0 (t) ,
from which it follows that |x0 (s)| = 1 for all s. So the derivative of x with
respect to arc length is always a unit vector.
This parameter s is defined up to the transformation s → ±s + c, where
c is a constant. Geometrically, this means the freedom in the choice of initial
point and direction in which to traverse the curve in measuring the arc length.
Exercise 12. One of the most important space curves is the circular helix
x(t) = (a cos t, a sin t, bt) ,
where a 6= 0 and b are constants. Find the length of this curve over the
interval [0, 2π].
2.2 Curvature and Fenchel’s Theorem 13
Exercise 13. Find a constant c such that the helix

x(t) = (a cos(ct), a sin(ct), bt)
is parametrized by arclength, so that |x0 (t)| = 1 for all t.
Exercise 14. The astroid is the curve defined by
x(t) = a cos3 t, a sin3 t, 0 ,

on the domain [0, 2π]. Find the points at which x(t) does not define an
immersion, i.e., the points for which x0 (t) = 0.
Exercise 15. The trefoil curve is defined by
x(t) = ((a + b cos(3t)) cos(2t), (a + b cos(3t)) sin(2t), b sin(3t)) ,
where a and b are constants with a > b > 0 and 0 ≤ t ≤ 2π. Sketch this
curve, and give an argument to show why it is knotted, i.e. why it cannot be
deformed into a circle without intersecting itself in the process.
Exercise 16. (For the serious mathematician) Two parametrized curves x(t)
and y(u) are said to be equivalent if there is a function u(t) such that u0 (t) > 0
for all a < t < b and such that y(u(t)) = x(t). Show that relation satisfies
the following three properties:
1. Every curve x is equivalent to itself
2. If x is equivalent to y, then y is equivalent to x
3. If x is equivalent to y and if y is equivalent to z, then x is equivalent to
z
A relation that satisfies these properties is called an equivalence rela-
tion. Precisely speaking, a curve is considered be an equivalence class of
parametrized curves.
2.2 Curvature and Fenchel’s Theorem

If x is an immersed curve, with x0 (t) 6= 0 for all t in the domain, then we may
0
define the unit tangent vector T(t) to be |xx0 (t) (t)| . If the parameter is arclength,
then the unit tangent vector T(s) is given simply by x0 (s). The line through
x(t0 ) in the direction of T(t0 ) is called the tangent line at x(t0 ). We can write
this line as y(u) = x(t0 ) + uT(t0 ), where u is a parameter that can take on
all real values.
Since T(t) · T(t) = 1 for all t, we can differentiate both sides of this
expression, and we obtain 2T0 (t) · T(t) = 0. Therefore T0 (t) is orthogonal
to T(t). The curvature of the space curve x(t) is defined by the condition
0
κ(t) = |T (t)| 0 0
|x0 (t)| , so = κ(t)s (t) = |T (t)|. If the parameter is arclength, then
x0 (s) = T(s) and κ(s) = |T0 (s)| = |x”(s)|.
14 2 Curves
Proposition 2. If κ(t) = 0 for all t, then the curve lies along a straight line.
Proof. Since κ(t) = 0, we have T0 (t) = 0 and T(t) = a, a constant unit
vector. Then x0 (t) = s0 (t)T(t) = s0 (t)a, so by integrating both sides of the
equation, we obtain x(t) = s(t)a + b for some constant b. Thus x(t) lies on
the line through b in the direction of a.
Curvature is one of the simplest and at the same time one of the most
important properties of a curve. We may obtain insight into curvature by
considering the second derivative vector x”(t), often called the acceleration
vector when we think of x(t) as representing the path of a particle at time t. If
the curve is parametrized by arclength, then x0 (s)·x0 (s) = 1 so x”(s)·x0 (t) = 0
and κ(s) = |x”(s)|. For a general parameter t, we have x0 (t) = s0 (t)T(t) so
x”(t) = s”(t)T(t) + s0 (t)T0 (t). If we take the cross product of both sides with
x0 (t) then the first term on the right is zero since x0 (t) is parallel to T(t).
Moreover x0 (t) is perpendicular to T0 (t) so
|T0 (t) × x0 (t)| = |T0 (t)||x0 (t)| = s0 (t)2 κ(t) .
Thus
x”(t) × x0 (t) = s0 (t)T0 (t) × x0 (t)
and
|x”(t) × x0 (t)| = s0 (t)3 κ(t) .
This gives a convenient way of finding the curvature when the curve is defined
with respect to an arbitrary parameter. We can write this simply as
|x”(t) × x0 (t)|
κ(t) = .
|x0 (t)x0 (t)|3/2
Note that the curvature κ(t) of a space curve is non-negative for all t. The
curvature can be zero, for example at every point of a curve lying along a
straight line, or at an isolated point like t = 0 for the curve x(t) = (t, t3 , 0).
A curve for which κ(t) > 0 for all t is called non-inflectional.
The unit tangent vectors emanating from the origin form a curve T(t) on
the unit sphere called the tangential indicatrix of the curve x. To calculate
the length of the tangent indicatrix, we form the integral of |T0 (t)| = κ(t)s0 (t)
with respect to t, so the length is κ(t)s0 (t)dt = κ(s)ds. This significant integral
is called the total curvature of the curve x.
Up to this time, we have concentrated primarily on local properties of
curves, determined at each point by the nature of the curve in an arbitrarily
small neighborhood of the point. We are now in a position to prove our first
result in global differential geometry or differential geometry in the large.
By a closed curve x(t), a ≤ t ≤ b, we mean a curve such that x(b) = x(a).
We will assume moreover that the derivative vectors match at the endpoints
of the interval, so x0 (b) = x0 (a).
2.2 Curvature and Fenchel’s Theorem 15
Theorem 1 (Fenchel’s Theorem). The total curvature of a closed space

curve x is greater than or equal to 2π.
κ(s)ds ≥ 2π
The first proof of this result was found independently by B. Segre in 1934
and later independently by H. Rutishauser and H. Samelson in 1948. The
following proof depends on a lemma by R. Horn in 1971:
Lemma 1. Let g be a closed curve on the unit sphere with length L < 2.
Then there is a point m on the sphere that is the north pole of a hemisphere
containing g.
To see this, consider two points p and q on the curve that break g up
into two pieces g1 and g2 of equal length, therefore both less than π. Then
the distance from p to q along the sphere is less than π so there is a unique
minor arc from p to q. Let m be the midpoint of this arc. We wish to show
that no point of g hits the equatorial great circle with m as north pole. If
a point on one of the curves, say g1 , hits the equator at a point r, then we
may construct another curve g10 by rotating g1 one-half turn about the axis
through m, so that p goes to q and q to p while r goes to the antipodal
point r0 . The curve formed by g1 and g10 has the same length as the original
curve g, but it contains a pair of antipodal points so it must have length at
least 2π, contradicting the hypothesis that the length of g was less than 2π.
From this lemma, it follows that any curve on the sphere with length less
than 2π is contained in a hemisphere centered at a point m. However if x(t)
is a closed curve, we may consider the differentiable function f (t) = x(t) · m.
At the maximum and minimum values of f on the closed curve x, we have
0 = f 0 (t) = x0 (t) · m = s0 (t)T(t) · m
so there are at least two points on the curve such that the tangential image
is perpendicular to m. Therefore the tangential indicatrix of the closed curve
x is not contained in a hemisphere, so by the lemma, the length of any such
indicatrix is greater than 2π. Therefore the total curvature of the closed curve
x is also greater than 2π.
1
Corollary 1. If, for a closed curve x, we have κ(t) ≤ R for all t, then the
curve has length L ≥ 2πR.
Proof.
Z Z Z
L= ds ≥ Rκ(s)ds = R κ(s)ds ≥ 2πR
16 2 Curves
Fenchel also proved the stronger result that the total curvature of a closed
curve equals 2π if and only if the curve is a convex plane curve.
I. Fáry and J. Milnor proved independently that the total curvature must
be greater than 4π for any non-self-intersecting space curve that is knotted
(not deformable to a circle without self-intersecting during the process.)
Exercise 17. Let x be a curve with x0 (t0 ) 6= 0. Show that the tangent line
at x(t0 ) can be written as y(u) = x(t0 ) + ux0 (t0 ) where u is a parameter that
can take on all real values.
Exercise 18. The plane through a point x(t0 ) perpendicular to the tangent
line is called the normal plane at the point. Show that a point y is on the
normal plane at x(t0 ) if and only if
x0 (t0 ) · y = x0 (t0 ) · x(t0 )
Exercise 19. Show that the curvature κ of a circular helix
x(t) = (r cos(t), r sin(t), pt)
is equal to the constant value κ = r2|r|

+p2 . Are there any other curves with
constant curvature? Give a plausible argument for your answer.
Exercise 20. Assuming that the level surfaces of two functions f (x1 , x2 , x3 ) =
0 and g(x1 , x2 , x3 ) = 0 meet in a curve, find an expression for the tangent
vector to the curve at a point in terms of the gradient vectors of f and g
(where we assume that these two gradient vectors are linearly independent
at any intersection point.) Show that the two level surfaces x2 − x21 = 0 and
x3 x1 − x22 = 0 consists of a line and a “twisted cubic” x1 (t) = t, x2 (t) = t2 ,
x3 (t) = t3 . What is the line?
Exercise 21. What is the geometric meaning of the function f (t) = x(t) · m
used in the proof of Fenchel’s theorem?
Exercise 22. Let m be a unit vector and let x be a space curve. Show that
the projection of this curve into the plane perpendicular to m is given by
y(t) = x(t) − (x(t) · m)m .
Under what conditions will there be a t0 with y0 (t0 ) = 0?
2.3 The Unit Normal Bundle and Total Twist

Consider a curve x(t) with x0 (t) 6= 0 for all t. A vector z perpendicular to
the tangent vector x0 (t0 ) at x(t0 ) is called a normal vector at x(t0 ). Such a
vector is characterized by the condition z · x(t0 ) = 0, and if |z| = 1, then z is
2.3 The Unit Normal Bundle and Total Twist 17
said to be a unit normal vector at x(t0 ). The set of unit normal vectors at a
point x(t0 ) forms a great circle on the unit sphere. The unit normal bundle
is the collection of all unit normal vectors at x(t) for all the points on a curve
x.
At every point of a parametrized curve x(t) at which x0 (t) 6= 0, we may
consider a frame E2 (t), E3 (t), where E2 (t) and E3 (t) are mutually orthogonal
unit normal vectors at x(t). If E 2 (t), E 3 (t) is another such frame, then there
is an angular function φ(t) such that
E2 (t) = cos(φ(t))E 2 (t) − sin(φ(t))E 3 (t)
E3 (t) = sin(φ(t))E 2 (t) + cos(φ(t))E 3 (t)
or, equivalently,
E 2 (t) = cos(φ(t))E2 (t) + sin(φ(t))E3 (t)
E 3 (t) = sin(φ(t))E2 (t) + cos(φ(t))E3 (t) .
From these two representations, we may derive an important formula:
E20 (t) · E3 (t) = E 02 (t) · E 3 (t) − φ0 (t)
Expressed in the form of differentials, without specifying parameters, this

formula becomes:
dE2 E3 = dE 2 E 3 − dφ .
Since E3 (t) = T(t) × E2 (t), we have:
E20 (t) · E3 (t) = −[E20 (t), E2 (t), T(t)]
or, in differentials:
dE2 E3 = −[dE2 , E2 , T] .
More generally, if z(t) is a unit vector in the normal space at x(t), then we
may define a function w(t) = −[z0 (t), z(t), T(t)]. This is called the connection
function of the unit normal bundle. The corresponding differential form w =
−[dz, z, T] is called the connection form of the unit normal bundle.
A vector function z(t) such that |z(t)| = 1 for all t and z(t) · x0 (t) =
0 for all t is called a unit normal vector field along the curve x. Such a
vector field is said to be parallel along x if the connection function w(t) =
−[z0 (t), z(t), T(t)] = 0 for all t. In the next section, we will encounter several
unit normal vector fields naturally associated with a given space curve. For
now, we prove some general theorems about such objects.
18 2 Curves
Proposition 3. If E2 (t) and E 2 (t) are two unit normal vector fields that are
both parallel along the curve x, then the angle between E2 (t) and E 2 (t) is
constant.
Proof. From the computation above, then:
E20 (t) · (−E2 (t) × T(t)) = E 02 (t) · (−E 2 (t) × T(t)) − φ0 (t) .
But, by hypothesis,
E20 (t) · (−E2 (t)xT(t)) = 0 = E 02 (t)(−E 2 (t) × T(t))
so it follows that φ0 (t) = 0 for all t, i.e., the angle φ(t) between E2 (t) and
E 2 (t) is constant.
Given a closed curve x and a unit normal vector field z with z(b) = z(a),
we define
Z
1 1
µ(x, z) = − [z0 (t), z(t), T(t)]dt = − [dz, z, T] .
2π 2π
If z is another such field, then
Z
1
µ(x, z) − µ(x, z) = − [z0 (t), z(t), T(t)] − [z0 (t), z(t), T(t)]dt
2π
Z
1 1
=− φ0 (t)dt = − [φ(b) − φ(a)] .
2π 2π
Since the angle φ(b) at the end of the closed curve must coincide with
the angle φ(a) at the beginning, up to an integer multiple of 2π, it follows
that the real numbers µ(x, z) and µ(x, z) differ by an integer. Therefore the
fractional part of µ(x, z) depends only on the curve x and not on the unit
normal vector field used to define it. This common value µ(x) is called the
total twist of the curve x. It is a global invariant of the curve.
Proposition 4. If a closed curve lies on a sphere, then its total twist is zero.
Proof. If x lies on the surface of a sphere of radius r centered at the origin,
then |x(t)|2 = x(t) · x(t) = r2 for all t. Thus x0 (t) · x(t) = 0 for all t, so x(t)
is a normal vector at x(t). Therefore z(t) = x(t) r is a unit normal vector field
defined along x, and we may compute the total twist by evaluating
Z
1
µ(x, z) = − [z0 (t), z(t), T(t)]dt .
2π
But
x0 (t) x(t)
[z0 (t), z(t), T(t)] = [ , , T(t)] = 0
r r
for all t since x0 (t) is a multiple of T(t). In differential form notation, we get
the same result: [dz, z, T] = r12 [x0 (t), x(t), T(t)]dt = 0. Therefore µ(x, z) = 0,
so the total twist of the curve x is zero.
2.4 Moving Frames 19
Remark 3. W. Scherrer proved that this property characterized a sphere, i.e.

if the total twist of every curve on a closed surface is zero, then the surface
is a sphere.
Remark 4. T. Banchoff and J. White proved that the total twist of a closed
curve is invariant under inversion with respect to a sphere with center not
lying on the curve.
Remark 5. The total twist plays an important role in modern molecular bi-
ology, especially with respect to the structure of DNA.
Exercise 23. Let x be the circle x(t) = (r cos(t), r sin(t), 0), where r is a
constant > 1. Describe the collection of points x(t) + z(t) where z(t) is a unit
normal vector at x(t).
Exercise 24. Let Σ be the sphere of radius r > 0 about the origin. The
x
inversion through the sphere S maps a point x to the point x = r2 |x| 2 . Note
that this mapping is not defined if x = 0, the center of the sphere. Prove
that the coordinates of the inversion of x = (x1 , x2 , x3 ) through S are given
r 2 xi
by xi = x2 +x 2 +x2 . Prove also that inversion preserves point that lie on the
1 2 3
sphere S itself, and that the image of a plane is a sphere through the origin,
except for the origin itself.
Exercise 25. Prove that the total twist of a closed curve not passing through
the origin is the same as the total twist of its image by inversion through the
sphere S of radius r centered at the origin.
2.4 Moving Frames

In the previous section, we introduced the notion of a frame in the unit normal
bundle of a space curve. We now consider a slightly more general notion. By
a frame, or more precisely a right-handed rectangular frame with origin, we
mean a point x and a triple of mutually orthogonal unit vectors E1 , E2 , E3
forming a right-handed system. The point x is called the origin of the frame.
Note that Ei · Ej = 1 if i = j and 0 if i 6= j.
Moreover,
E1 × E2 = E3 , E2 = E3 × E1 , andE3 = E1 × E2 .
In the remainder of this section, we will always assume that small Latin
letters run from 1 to 3.
Note that given two different frames, x, E1 , E2 , E3 and x, E1 , E2 , E3 , there
is exactly one affine motion of Euclidean space taking x to x and taking Ei
to E i . When x(t), E1 (t), E2 (t), E3 (t) is a family of frames depending on a
parameter t, we say we have a moving frame along the curve.
20 2 Curves
Proposition 5. A family of frames x(t), E1 (t), E2 (t), E3 (t) satisfies a system

of differential equations:
x0 (t) = Σpi (t)Ei (t)

Ei0 (t) = Σqij (t)Ej (t)
where pi (t) = x0 (t) · Ei (t) and qi j(t) = Ei0 (t) · Ej (t).

Since Ei (t) · Ej (t) = 0 for i 6= j, it follows that
qij (t) + qji (t) = Ei0 (t) · Ej (t) + Ei (t) · Ej0 (t) = 0
i.e. the coefficients qij (t) are anti-symmetric in i and j. This can be expressed
by saying that the matrix ((qi j(t))) is an anti-symmetric matrix, with 0 on
the diagonal.
In a very real sense, the function pi (t) and qij (t) completely determine
the family of moving frames.
Specifically we have:
Proposition 6. If x(t), E1 (t), E2 (t), E3 (t) and x(t), E 1 (t), E 2 (t), E 3 (t) are
two families of moving frames such that pi (t) = pi (t) and qij (t) = q ij (t) for
all t, then there is a single affine motion that takes x(t), E1 (t), E2 (t), E3 (t)
to x(t), E 1 (t), E 2 (t), E 3 (t)) for all t.
Proof. Recall that for a specific value t0 , there is an affine motion tak-
ing x(t0 ), E1 (t0 ), E2 (t0 ), E3 (t0 ) to x(t0 ), E 1 (t0), E 2 (t0 ), E 3 (t0 ). We will show
that this same motion takes x(t), E1 (t), E2 (t), E3 (t) to x(t), E 1 (t), E 2 (t), E 3 (t)
for all t. Assume that the motion has been carried out so that the frames
x(t0 ), E1 (t0 ), E2 (t0 ), E3 (t0 ) and x(t0 ), E 1 (t0), E 2 (t0 ), E 3 (t0 ) coincide.
Now consider
(ΣEi (t) · E i (t))0 = ΣEi0 (t) · E i (t) + ΣEi (t) · E 0i (t)

= ΣΣqij (t)Ej (t) · E i (t) + ΣEi (t) · Σqij (t)E j (t)
= ΣΣqij (t)Ej (t) · E i (t) + ΣΣqij (t)Ei (t) · E j (t)
= ΣΣqij (t)Ej (t) · E i (t) + ΣΣqji (t)Ej (t) · E i (t)
=0.
It follows that
ΣEi (t) · E i (t) = ΣEi (t0 ) · E i (t0 ) = ΣEi (t0 ) · Ei (t0 ) = 3
for all t. But since |Ei (t) · E i (t)| ≤ 1 for any pair of unit vectors, we must
have Ei (t) · E i (t) = 1 for all t. Therefore Ei (t) = E i (t) for all t.
Next consider
(x(t) − x(t))0 = Σpi (t)Ei (t) − Σpi (t)E i (t) = Σpi (t)Ei (t) − Σpi (t)Ei (t) = 0 .
2.5 Curves at a Non-inflexional Point and the Frenet Formulas 21
Since the origins of the two frames coincide at the value t0 , we have
x(t) − x(t) = x(t0 ) − x(t0 ) = 0
for all t.
This completes the proof that two families of frames satisfying the same
set of differential equations differ at most by a single affine motion.
Exercise 26. Prove that the equations Ei0 (t) = Σqij (t)Ej (t) can be written
Ei0 (t) = d(t) × Ei (t), where d(t) = q23 (t)E1 (t) + q31 (t)E2 (t) + q12 (t)E3 (t).
This vector is called the instantaneous axis of rotation.
Exercise 27. Under a rotation about the x3 -axis, a point describes a circle
x(t) = (a cos(t), a sin(t), b). Show that its velocity vector satisfies x0 (t) =
d × x(t) where d = (0, 0, 1). (Compare with the previous exercise.).
Exercise 28. Prove that (v ·v)(w ·w)−(v ·w)2 = 0 if and only if the vectors
v and w are linearly dependent.
2.5 Curves at a Non-inflexional Point and the Frenet

Formulas
A curve x is called non-inflectional if the curvature k(t) is never zero. By our
earlier calculations, this condition is equivalent to the requirement that x0 (t)
and x00 (t) are linearly independent at every point x(t), i.e. x0 (t) × x00 (t) 6= 0
for all t. For such a non-inflectional curve x, we may define a pair of natural
unit normal vector fields along x.
0 00
Let b(t) = |xx0 (t)×x (t)
(t)×x00 (t)| , called the binormal vector to the curve x(t).
Since b(t) is always perpendicular to T(t), this gives a unit normal vector
field along x.
We may then take the cross product of the vector fields b(t) and T(t) to
obtain another unit normal vector field N(t) = b(t)×T(t), called the principal
normal vector. The vector N(t) is a unit vector perpendicular to T(t) and
lying in the plane determined by x0 (t) and x00 (t). Moreover, x00 (t) · N(t) =
k(t)s0 (t)2 , a positive quantity.
Note that if the parameter is arclength, then x0 (s) = T(s) and x00 (s)
is already perpendicular to T(s). It follows that x00 (s) = k(s)N(s) so we
00
may define N(s) = xk(s) (s)
and then define b(s) = T(s) × N(s). This is the
standard procedure when it happens that the parametrization is by arclength.
The method above works for an arbitrary parametrization.
We then have defined an orthonormal frame x(t)T(t)N(t)b(t) called the
Frenet frame of the non-inflectional curve x.
By the previous section, the derivatives of the vectors in the frame can
be expressed in terms of the frame itself, with coefficients that form an anti-
symmetric matrix. We already have x0 (t) = s0 (t)T(t), so
22 2 Curves
p1 (t) = s0 (t) , p2 (t) = 0 = p3 (t) .
Also T0 (t) = k(t)s0 (t)N(t), so
q12 (t) = k(t)s0 (t) and q13 (t) = 0 .
We know that
b0 (t) = q31 (t)T(t) + q32 (t)N(t) , and q31 (t) = −q13 (t) = 0 .
Thus b0 (t) is a multiple of N(t), and we define the torsion w(t) of the curve
by the condition
b0 (t) = −w(t)s0 (t)N(t) ,
so q32 (t) = −w(t)s0 (t) for the Frenet frame. From the general computations
about moving frames, it then follows that
N0 (t) = q21 (t)T(t) + q23 (t)b(t) = −k(t)s0 (t)T(t) + w(t)s0 (t)b(t) .
The formulas for T0 (t), N0 (t), and b0 (t) are called the Frenet formulas for
the curve x.
If the curve x is parametrized with respect to arclength, then the Frenet
formulas take on a particularly simple form:
x0 (s) = T(s)
T0 (s) = k(s)N(s)
N0 (s) = −k(s)T(s) + w(s)b(s)
b0 (s) = −w(s)b(s) .
The torsion function w(t) that appears in the derivative of the binormal
vector determines important properties of the curve. Just as the curvature
measures deviation of the curve from lying along a straight line, the torsion
measures deviation of the curve from lying in a plane. Analogous to the result
for curvature, we have:
Proposition 7. If w(t) = 0 for all points of a non-inflectional curve x, then

the curve is contained in a plane.
Proof. We have b0 (t) = −w(t)s0 (t)N(t) = 0 for all t so b(t) = a, a constant

unit vector. Then T(t)a = 0 for all t so (x(t) · a)0 = x0 (t) · a = 0 and
x(t) · a = x(a) · a, a constant. Therefore (x(t) − x(a)) · a = 0 and x lies in
the plane through x(a) perpendicular to a.
If x is a non-inflectional curve parametrized by arclength, then
w(s) = b(s) · N0 (s) = [T(s), N(s), N0 (s)] .

2.5 Curves at a Non-inflexional Point and the Frenet Formulas 23
x00 (s)
Since N(s) = k(s) , we have
x000 (s) −k 0 (s)

N0 (s) = + x00 (s) ,
k(s) k(s)2
so
x00 (s) x000 (s) −k 0 (s) [x0 (s), x00 (s), x000 (s)]

w(s) = x0 (s), , + x00 (s) 2
= .
k(s) k(s) k(s) k(s)2
We can obtain a very similar formula for the torsion in terms of an arbi-
trary parametrization of the curve x. Recall that
x00 (t) = s”(t)T(t) + k(t)s0 (t)T0 (t) = s00 (t)T(t) + k(t)s0 (t)2 N(t) ,
so
0
x000 (t) = s000 (t)T(t) + s00 (t)s0 (t)k(t)N(t) + k(t)s0 (t)2 N(t) + k(t)s0 (t)2 N0 (t) .

Therefore
x000 (t)b(t) = k(t)s0 (t)2 N0 (t)b(t) = k(t)s0 (t)2 w(t)s0 (t) ,
and
x000 (t) · x0 (t) × x00 (t) = k 2 (t)s0 (t)6 w(t) .
Thus we obtain the formula

x000 (t) · x0 (t)xx00 (t)
w(t) = ,
|x0 (t) × x00 (t)|2
valid for any parametrization of x.
Notice that although the curvature k(t) is never negative, the torsion w(t)
can have either algebraic sign. For the circular helix x(t) = (r cos(t), r sin(t), pt)
p
for example, we find w(t) = r2 +p 2 , so the torsion has the same algebraic sign
as p. In this way, the torsion can distinguish between a right-handed and a

left-handed screw.
Changing the orientation of the curve from s to −s changes T to −T, and
choosing the opposite sign for k(s) changes N to −N. With different choices,
then, we can obtain four different right-handed orthonormal frames, xTNb,
x(−T)N(−b), xT(−N)(−b), and x(−T)(−N)b. Under all these changes of
the Frenet frame, the value of the torsion w(t) remains unchanged.
A circular helix has the property that its curvature and its torsion are
both constant. Furthermore the unit tangent vector T(t) makes a constant
angle with the vertical axis. Although the circular helices are the only curves
with constant curvature and torsion, there are other curves that have the
second property. We characterize such curves, as an application of the Frenet
frame.
24 2 Curves
Proposition 8. The unit tangent vector T(t) of a non-inflectional space

curve x makes a constant angle with a fixed unit vector a if and only if
the ratio w(t)
k(t) is constant.
Proof. If T(t) · a = constant for all t, then differentiating both sides, we

obtain
T0 (t) · a = 0 = k(t)s0 (t)N(t) · a ,
so a lies in the plane of T(t) and b(t). Thus we may write a = cos(φ)T(t) +
sin(φ)b(t) for some angle φ. Differentiating this equation, we obtain
0 = cos(φ)T0 (t) + sin(φ)b0 (t) = cos(φ)k(t)s0 (t)N(t) − sin(φ)w(t)s0 (t)N(t) ,
so w(t) sin(φ)
k(t) = cos(φ) = tan(φ). This proves the first part of the proposition and
identifies the constant ratio of the torsion and the curvature.
Conversely, if w(t)
k(t) = constant = tan(φ) for some φ, then, by the same
calculations, the expression cos(φ)T(t) + sin(φ)b(t) has derivative 0 so it
equals a constant unit vector. The angle between T(t) and this unit vector
is the constant angle φ.
Curves with the property that the unit tangent vector makes a fixed
angle with a particular unit vector are called generalized helices. Just as a
circular helix lies on a circular cylinder, a generalized helix will lie on a
general cylinder, consisting of a collection of lines through the curve parallel
to a fixed unit vector. On this generalized cylinder, the unit tangent vectors
make a fixed angle with these lines, and if we roll the cylinder out onto a
plane, then the generalized helix is rolled out into a straight line on the plane.
We have shown in the previous section that a moving frame is completely
determined up to an affine motion by the functions pi (t) and qij (t). In the
case of the Frenet frame, this means that if two curves x and x have the same
arclength s(t), the same curvature k(t), and the same torsion w(t), then the
curves are congruent, i.e. there is an affine motion of Euclidean three-space
taking x(t) to x(t) for all t. Another way of stating this result is:
Theorem 2. The Fundamental Theorem of Space Curves. Two curves parametrized

by arclength having the same curvature and torsion at corresponding points
are congruent by an affine motion.
Exercise 29. Compute the torsion of the circular helix. Show directly that
the principal normals of the helix are perpendicular to the vertical axis, and
show that the binormal vectors make a constant angle with this axis.
Exercise 30. Prove that if the curvature and torsion of a curve are both
constant functions, then the curve is a circular helix (i.e. a helix on a circular
cylinder).
2.6 Local Equations of a Curve 25
Exercise 31. Prove that a necessary and sufficient condition for a curve x
to be a generalized helix is that
x00 (t) × x000 (t) · xiv (t) = 0 .
Exercise 32. Let y(t) be a curve on the unit sphere, so that R |y(t)| = 1 and
y(t)·y0 (t)×y00 (t) 6= 0 for all t. Show that the curve x(t) = c y(u)×y00 (u)du
with c 6= 0 has constant torsion 1c .
Exercise 33. (For students familiar with complex variables) If the coordi-
nate functions of the vectors in the Frenet frame are given by
T = (e11 , e12 , e13 ) ,
N = (e21 , e22 , e23 ) ,
b = (e31 , e32 , e33 ) ,
then we may form the three complex numbers
e1j + ie2j 1 + e3j
zj = = .
1 − e3j e1j − ie2j
Then the functions zj satisfy the Riccati equation
i
zj0 = −ik(s)zj + w(s)(−1 + zj2 ) .
2
This result is due to S. Lie and G. Darboux.
2.6 Local Equations of a Curve

We can “see” the shape of a curve more clearly in the neighborhood of a
point x(t0 ) when we consider its parametric equations with respect to the
Frenet frame at the point. For simplicity, we will assume that t0 = 0, and we
may then write the curve as
x(t) = x(0) + x1 (t)T(0) + x2 (t)N(0) + x3 (t)b(0) .
On the other hand, using the Taylor series expansion of x(t) about the point
t = 0, we obtain
t2 00 t3
x(t) = x(0) + tx0 (0) + x (0) + x000 (0) + higher order terms .
2 6
From our earlier formulas, we have
x0 (0) = s0 (0)T(0) ,
x00 (0) = s00 (0)T(0) + k(0)s0 (0)2 N(0) ,
x000 (0) = s000 (0)T(0) + s00 (0)s0 (0)k(0)N(0) + (k(0)s0 (0)2 )0 N(0)
+ k(0)s0 (0)2 (−k(0)s0 (0)T(0) + w(0)s0 (0)b(0)) .
26 2 Curves
Substituting these equations in the Taylor series expression, we find:

t2 00 t3 000

0 2 0 3

x(t) = x(0) + ts (0) + s (0) + s (0) − k(0) s (0) + ... T(0)
2 6
2
t3 00

t 0 2 0 0 2 0

+ k(0)s (0) + s (0)s (0)k(0) + (k(0)s (0) ) + ... N(0)
2 6
3
t 0 3
+ k(0)w(0)s (0) + ... b(0) .
6
If the curve is parametrized by arclength, this representation is much
simpler:
k(0)2 3

x(s) = x(0) + s − s + ... T(0)
6
k(0) 2 k 0 (0) 3

+ s + s + ... N(0)
2 6

k(0)w(0) 3
+ s + ... b(0) .
6
Relative to the Frenet frame, the plane with equation x1 = 0 is the normal
plane; the plane with x2 = 0 is the rectifying plane, and the plane with x3 = 0
is the osculating plane. These planes are orthogonal respectively to the unit
tangent vector, the principal normal vector, and the binormal vector of the
curve.
2.7 Plane Curves and a Theorem on Turning Tangents

The general theory of curves developed above applies to plane curves. In the
latter case there are, however, special features which will be important to
bring out. We suppose our plane to be oriented. In the plane a vector has
two components and a frame consists of an origin and an ordered set of two
mutually perpendicular unit vectors forming a right-handed system. To an
oriented curve C defined by x(s) the Frenet frame at s consists of the origin
x(s), the unit tangent vector T(s) and the unit normal vector N(s). Unlike
the case of space curves this Frenet frame is uniquely determined, under the
assumption that both the plane and the curve are oriented.
The Frenet formulas are
x0 = T,
T0 = kN , (2.1)
N0 = −kT .
The curvature k(s) is defined with sign. It changes its sign when the
orientation of the plane or the curve is reversed.
2.7 Plane Curves and a Theorem on Turning Tangents 27
The Frenet formulas in (2.1) can be written more explicitly. Let
x(s) = (x1 (s) , x2 (s)) (2.2)
Then
T(s) = ( x01 (s) , x02 (s)) ,
(2.3)
N(s) = (−x02 (s) , x01 (s)) .
Expressing the last two equations of (2.1) in components, we have
x001 = −kx02 (2.4)

x002 = kx01 . (2.5)
These equations are equivalent to (2.1).

Since T is a unit vector, we can put
T(s) = (cos τ(s) , sin τ(s)) , (2.6)
so that τ(s) is the angle of inclination of T with the x1 -axis. Then
N(s) = (− sin τ(s) , cos τ(s)) , (2.7)
and (2.1) gives

dτ
= k(s) (2.8)
ds
This gives a geometrical interpretation of k(s).
A curve C is called simple if it does not intersect itself. One of the most
important theorems in global differential geometry is the theorem on turning
tangents:
Theorem 3. For a simple closed plane curve we have

I
1
k ds = ±1 .
2π
To prove this theorem we give a geometrical interpretation of the integral
at the left-hand side of (3). By (2.8)
I I
1 1
k ds = dτ .
2π 2π
But τ , as the angle of inclination of τ(s), is only defined up to an integral
multiple of 2π, and this integral has to be studied with care.
Let O be a fixed point in the plane. Denote by Γ the unit circle about
O; it is oriented by the orientation of the plane. The tangential mapping or
Gauss mapping
28 2 Curves
g : C 7→ Γ (2.9)
is defined by sending the point x(s) of C to the point T(s) of Γ. In other

words, g(P ), P ∈ C, is the end-point of the unit vector through O parallel
to the unit tangent vector to C at P . Clearly g is a continuous mapping. If
C is closed, it is intuitively clear that when a point goes along C once its
image point under g goes along Γ a number of times. This integer is called
the rotation index of C. It is to be defined rigorously as follows:
We consider O to be the origin of our coordinate system. As above we
denote by τ(s) the angle of inclination of T(s) with the x1 -axis. In order to
make the angle uniquely determined we suppose O ≤ τ(s) < 2π. But τ(s) is
not necessarily continuous. For in every neighborhood of s0 at which τ(sc ) = 0
there may be values of τ(s) differing from 2π by arbitrarily small quantities.
We have, however, the following lemma:
Lemma 2. There exists a continuous function τ̃(s) such that τ̃(s) ≡ τ(s)
mod 2π .
Proof. We suppose C to be a closed curve of total length L. The continuous

mapping g is uniformly continuous. There exists therefore a number δ > 0
such that for |s1 − s2 | < δ, T(s1 ) and T(s2 ) lie in the same open half-plane.
Let s0 (= O) < s1 < · · · < si (= L) satisfy |si − si−1 | < δ for i = 1, . . . , m. We
put τ̃(s0 ) = τ(s0 ). For s0 ≤ s ≤ s1 , we define τ̃(s) to be τ̃(s0 ) plus the angle
of rotation from g(s0 ) to g(s) remaining in the same half-plane. Carrying
out this process in successive intervals, we define a continuous function τ̃(s)
satisfying the condition in the lemma. The difference τ̃(L)− τ̃(O) is an integral
multiple of 2π. Thus, τ̃(L) − τ̃(O) = γ2π. We assert that the integer γ is
independent of the choice of the function τ̃. In fact let τ̃0 (s) be a function
satisfying the same conditions. Then we have τ̃0 (s) − τ̃(s) = n(s)2π where
n(s) is an integer. Since n(s) is continuous in s, it must be constant. It
follows that τ̃0 (L) − τ̃0 (O) = τ̃(L) − τ̃(O), which proves the independence of
γ from the choice of τ̃. We call γ the rotation index of C. In performing
integration over C we should replace τ(s) by τ̃ in (2.8). Then we have
I I
1 1
k ds = d τ̃ = γ . (2.10)
2π 2π
We consider the mapping h which sends an ordered pair of points x(s1 ), x(s2 ),
O ≤ s1 ≤ s2 ≤ L, of C into the end-point of the unit vector through O parallel
to the secant joining x(s1 ) to x(s2 ). These ordered pairs of points can be
represented as a triangle 4 in the (s1 , s2 )-plane defined by O ≤ s1 ≤ s2 ≤ L.
The mapping h of 4 into Γ is continuous. Moreover, its restriction to the
side s1 = s2 is the tangential mapping g in (2.9).
To a point p ∈ 4 let τ(p) be the angle of inclination of Oh(p) to the x1 -
axis, satisfying O ≤ τ(p) < 2π. Again this function need not be continuous.
˜ p ∈ 4,
We shall, however, prove that there exists a continuous function τ(p),
2.7 Plane Curves and a Theorem on Turning Tangents 29
such that τ̃(p) ≡ τ(p) mod 2π. In fact, let m be an interior point of 4.
We cover 4 by the radii through m. By the argument used in the proof of
the above lemma we can define a function τ̃(p), p ∈ 4, such that τ̃(p) ≡ τ(p)
mod 2π, and such that it is continuous on every radius through m. It remains
to prove that it is continuous in 4.
For this purpose let p0 ∈ 4. Since h is continuous, it follows from the com-
pactness of the segment mp0 that there exists a number η = η(p0 ) > 0, such
that for q0 ∈ mp0 and for any point q ∈ 4 for which the distance d(q, q0 ) < η
the points h(q) and h(q0 ) are never antipodal. The latter condition can be
analytically expressed by
τ̃(q) 6≡ τ̃(q0 ) mod π . (2.11)
Now let > 0, < π2 be given. We choose a neighborhood U of p0 such

that U is contained in the η-neighborhood of p0 and such that, for p ∈ U ,
the angle between Oh(p0 ) and Oh(p) is < . This is possible, because the
mapping h is continuous. The last condition can be expressed in the form
τ̃(p) − τ̃(p0 ) = 0 + 2k(p)π , (2.12)
where k(p) is an integer. Let q0 be any point on the segment mp0 . Draw
the segment qq0 parallel to pp0 , with q on mp. The function τ̃(q) − τ̃(q0 )
is continuous in q along mp and equals O when q coincides with m. Since
d(q, q0 ) < η, it follows from (2.11) that |τ̃(q) − τ̃(q0 )| < π. In particular, for
qo = p0 this gives |τ̃(p) − τ̃(p0 )| < π. Combining this with (2.12), we get
k(p) = 0. Thus we have proved that τ̃(p) is continuous in 4, as asserted
above. Since τ̃(p) ≡ τ(p) mod 2π, it is clear that τ̃(p) is differentiable.
Now let A(O, O), B(O, L), D(L, L) be the vertices of 4. The rotation
index γ of C is, by (2.10), defined by the line integral
I
2πγ = dτ̃ .
AD
Since τ̃(p) is defined in 4, we have

I I I
dτ̃ = dτ̃ + dτ̃ .
AD AB BD
To evaluate the line integrals at the right-hand side, we suppose the origin
O to be the point x(O) and C to lie in the upper half-plane and to be tangent
to the x1 -axis at O. This is always possible for we only have to take x(O)
to be the point on C at which the x2 -coordinate is a minimum. Then the
x1 -axis is either in the direction of the tangent vector to C at O or opposite
to it. We can assume the former case, by reversing the orientation of C if
necessary. The line integral along AB is then equal to the angle rotated by
OP as P goes once along C. Since C lies in the upper half-plane, the vector
OP never points downward. It follows that the integral along AB is equal
30 2 Curves
to π. On the other hand, the line integral along BD is the angle rotated by
P O as P goes once along C. Since the vector P O never points upward, this
integral is also equal to π. Hence their sum is 2π and the rotation index γ is
+1. Since we may have reversed the orientation of C, the rotation index is
±1 in general.
Exercise 34. Consider the plane curve x(t) = (t, f (t)). Use the Frenet for-
mulas in (2.1) to prove that its curvature is given by
f¨
k(t) = . (2.13)
(1 + f˙2 )3/2
Exercise 35. Draw closed plane curves with rotation indices 0,−2, +3 re-
spectively.
Exercise 36. The theorem on turning tangents is also valid when the simple
closed curve C has ”corners.” Give the theorem when C is a triangle con-
sisting of three arcs. Observe that the theorem contains as a special case the
theorem on the sum of angles of a rectilinear triangle.
Exercise 37. Give in detail the proof of the existence of η = η(p0 ) used in
the proof of the theorem on turning tangents. η = η(p0 ) .
2.8 Plane Convex Curves and the Four Vertex Theorem

A closed curve in the plane is called convex, if it lies at one side of every
tangent line.
Proposition 9. A simple closed curve is convex, if and only if it can be so
oriented that its curvature k is ≥ 0.
The definition of a convex curve makes use of the whole curve, while the
curvature is a local property. The proposition therefore gives a relationship
between a local property and a global property. The theorem is not true if
the closed curve is not simple. Counter examples can be easily constructed.
˜ be the function constructed above, so that we have k = dτ .
Let τ(s) ds
The condition k ≥ O is therefore equivalent to the assertion that τ̃(s)) is a
monotone non-decreasing function. We can assume that τ(O) ˜ = O. By the
theorem on turning tangents, we can suppose C so oriented that τ(L) ˜ = 2π.
Suppose τ̃(s) , O ≤ s ≤ L, be monotone non-decreasing and that C is
not convex. There is a point A = x(s0 ) on C such that there are points of
C at both sides of the tangent λ to C at A. Choose a positive side of k and
consider the oriented perpendicular distance from a point x(s) of C to λ. This
is a continuous function in s and attains a maximum and a minimum at the
points M and N respectively. Clearly M and N are not on λ and the tangents
2.9 Isoperimetric Inequality in the Plane 31
to C at M and N are parallel to x. Among these two tangents and k itself

there are two tangents parallel in the same sense. Call s1 < s2 the values of
the parameters at the corresponding points of contact. Since τ̃(s) is monotone
non-decreasing and O ≤ τ̃(s) ≤ 2π, this happems only when τ̃(s) = τ(s˜1 ) for
all s satisfying s1 ≤ s ≤ s2 . It follows that the arc s1 ≤ s ≤ s2 is a line
segment parallel to λ. But this is obviously impossible.
Next let C be convex. To prove that τ̃(s) is monotone non-decreasing,
suppose τ(s˜1 ) = τ(s˜2 ), s1 < s2 . Then the tangents at x(s1 ) and x(s2 ) are
parallel in the same sense. But there exists a tangent parallel to them in the
opposite sense. From the convexity of C it follows that two of them coincide.
We are thus in the situation of a line λ tangent to C at two distinct
points A and B. We claim that the segment AB must be a part of C. In fact,
suppose this is not the case and let D be a point on AB not on C. Draw
through D a perpendicular µ to λ in the half-plane which contains C. Then
µ intersects C in at least two points. Among these points of intersection let
F be the farthest from λ and G the nearest one, so that F 6= G. Then G
is an interior point of the triangle ABF . The tangent to C at G must have
points of C in both sides which contradicts the convexity of C.
It follows that under our assumption, the segment AB is a part of C , so
that the tangents at A and B are parallel in the same sense. This proves that
the segment joining x(s1 ) to x(s2 ) belongs to C. Hence τ̃(s) remains constant
in the interval s1 ≤ s ≤ s2 . We have therefore proved that τ̃(s) is monotone
and K ≥ O.
A point on C at which k 0 = 0 is called a vertex. A closed curve has at
least two vertices, e.g., the maximum and the minimum of k. Clearly a circle
consists entirely of vertices. An ellipse with unequal axes has four vertices,
which are its intersection with the axes.
Theorem 4 (Four-vertex Theorem.). A simple closed convex cuxve has
at least four vertices.
Remark 6. This theorem was first given by Mukhopadhyaya (1909). The fol-
lowing proof was due to G.Herglotz. It is also true for non-convex curves, but
the proof will be more difficult.
2.9 Isoperimetric Inequality in the Plane

Among all simple closed curves having a given length the circle bounds the
largest area, and is the only curve with this property. We shall state the
theorem as follows:
Theorem 5. Let L be the length of a simple closed curve C and A be the
area it bounds. Then
L2 − 4πA ≥ 0 . (2.14)
32 2 Curves
Moreover, the equality sign holds only when C is a circle.
The proof given below is due to E. Schmidt (1939).

We enclose C between two parallel lines g, g 0 , such that C lies between
g, g 0 and is tangent to them at the points P , Q respectively. Let s = 0, s0
be the parameters of P , Q. Construct a circle C tangent to g, g 0 at P , Q
respectively. Denote its radius by r and take its center to be the origin of a
coordinate system. Let x(s) = (x1 (s), x2 (s)) be the position vector of C, so
that
(x1 (0), x2 (0)) = (x1 (L), x2 (L)) .
As the position vector of C we take (x1 (s), x2 ), such that
x1 (s) = x1 (s) ,
q
x2 (s) = − r2 − x21 (s) , 0 ≤ s ≤ s0
q
= + r2 − x21 (s) , s0 ≤ s ≤ L .
Denote by A the area bounded by C. Now the area bounded by a closed

curve can be expressed by the line integral
Z L Z L
1 L
Z
A= x1 x02 ds = − x2 x01 ds = (x1 x02 − x2 x01 )ds .
0 0 2 0
Applying this to our two curves C and C, we get

Z L
A= x1 x02 ds ,
0
Z L Z L
A = πr2 = − x2 x01 ds = − x2 x01 ds .
0 0
Adding these two equations, we have

Z L Z L q
A + πr2 = (x1 x02 − x2 x01 )ds ≤ (x1 x02 − x2 x01 )2 ds
0 0
Z L q
0 0
≤ (x21 + x22 )(x12 + x22 )ds (2.15)
0
Z L q
= x21 + x22 ds = Lr .
0
Since the geometric mean of two numbers is ≤ their arithmetic mean, it

follows that
√ √ 1 1
A πr2 ≤ (A + πr2 ) ≤ Lr .
2 2
2.9 Isoperimetric Inequality in the Plane 33
This gives, after squaring and cancellation of r, the inequality (2.14).

Suppose now that the equality sign in (2.14) holds. A and πr2 have then
the same geometric and arithmetic mean, so that A = πr2 and L = 2πr.
The direction of the lines g, g 0 being arbitrary, this means that C has the
same ”width” in all directions. Moreover, we must have the equality sign
everywhere in (2.15). It follows in particular that
0 0
(x1 x02 − x2 x01 )2 = (x21 + x22 )(x12 + x22 ) ,
which gives
p
x1 −x2 x21 + x22
0 = 0 = p 0 0
= ±r .
x2 x1 x12 + x22
From the first equality in (2.15) the factor of proportionality is seen to be r,

i.e.,
x1 = rx02 , x2 = −rx01 .
This remains true when we interchange x1 and x2 , so that
x2 = rx01 .
Therefore we have
x21 + x22 = r2 ,
which means that C is a circle.

3 Fundamental Forms of a Surface
3.1 The First Fundamental Form

Let D be an open domain in the (u1 , u2 ) plane. A parametrized surface is a
smooth mapping
x : D 7→ E 3 .
of D into three-dimensional Euclidean space E 3 We will denote the value of

x at (u1 , u2 ) by x(u1 , u2 ), the position vector from the origin considered as
a function of u1 and u2 .
For the rest of our study of surfaces, we will be considering sums over
the indices i = 1, 2. We will therefore make adopt a convention, standard
in tensor analysis, that all small Latin indices have the range 1 − −2, and
that they will be summed whenever repeated. We will also make the conven-
tion that subscripts to a vector mean pertial derivatives with respect to the
corresponding parameter. Thus,
∂x ∂2x
xi = and xik = etc.
∂ui ∂ui ∂uk
The parametrized surface defined by x is said to be regular or immersed
if x1 (u1 , u2 ) × x2 (u1 , u2 ) 6= 0 for every (u1 , u2 ) in the domain D. A point
where x1 (u1 , u2 ) × x2 (u1 , u2 ) = 0 is said to be a singular point. In most cases
we will consider regular surfaces without singular points.
A point x of E 3 can be given in terms of its coordinates (x1 , x2 , x3 ) and
the vector function x(u1 , u2 ) can be expressed in terms of its component func-
tions x(u1 , u2 ) = (x1 (u1 , u2 ), x2 (u1 , u2 ), x3 (u1 , u2 )). In terms of component
functions, the regularity condition from above means that the matrix
∂x1 (u1 , u2 ) ∂x2 (u1 , u2 ) ∂x3 (u1 , u2 )

 
 ∂u1 ∂u1 ∂u1 
(3.1)
 
 
 ∂x (u1 , u2 ) ∂x (u1 , u2 ) ∂x (u1 , u2 ) 
1 2 3
∂u2 ∂u2 ∂u2
has rank two everywhere.
36 3 Fundamental Forms of a Surface
Example 2. The graph of the function f, x3 = f (x1 , x2 ) can be expressed as a

1
,u2 )
parametrized surface, x(u1 , u2 , f (u1 , u2 )). Since x1 (u1 , u2 ) = (1, 0, ∂f (u
∂u1 )
1 2
and x2 (u1 , u2 ) = (0, 1, ∂f (u ,u )
∂u2 ), the rank of the matrix in (3.1) is always
two, so the regularity condition is satisfied at all points of the domain of f .
Example 3. The sphere of radius b is defined by the equation
x21 + x22 + x23 = b2
may be parametrized as
x(u1 , u2 ) = (b sin(u1 ) cos(u2 ), b sin(u1 ) sin(u2 ), b cos(u1 )) .
Then,
x1 (u1 , u2 ) = (b cos(u1 ) cos(u2 ), b cos(u1 ) sin(u2 ), −b sin(u1 ))

x2 (u1 , u2 ) = (−b sin(u1 ) sin(u2 ), b sin(u1 ) cos(u2 ), 0) .
So,
x1 (u1 , u2 ) × x2 (u1 , u2 ) = b sin(u1 )x(u1 , u2 ) .
At the points where sin(u1 ) = 0, i.e. at the north pole (0, 0, 1) and the
south pole (0, 0, −1), the parametrized surface is not regular, although from
a geometric point of view, all points of the sphere are like all others. By
rotating the sphere one quarter turn around the x1 -axis, we may obtain
a parametrization which is regular at the north and south poles, although
singular at two other points. This example shows the importance of defining a
surface by using enough parametrizations so that each point is a regular point
of at least one of them. The process of defining a surface as a combination of
parametrized surfaces is at the foundation of differential geometry, and will
be discussed in detail in later chapters.
Example 4. The torus of revolution with the parametrization
x(u1 , u2 ) = ((a + b sin(u1 )) cos(u2 ), ((a + b sin(u1 )) sin(u2 ), b cos(u1 ))
for constants a > b > 0 is a regular surface for all (u1 , u2 ).
Let P = x(u10 , u20 ) be the point having the parameters (u10 , u20 ). The equa-
tions ui = ui (t) for i = 1, 2 satisfying the conditions ui0 = ui (t0 ) for i = 1, 2
define a curve through P on the parametrized surface. By the chain rule, the
tangent vector to the curve at P is
2 2
dx X ∂x dui X dui
= i
= xi . (3.2)
dt i=1
∂u dt i=1
dt
3.1 The First Fundamental Form 37
When two indices appear in a formula, it will be assumed that they will be
summed unless otherwise mentioned. Thus we may rewrite (3.2) as
dx dui
= xi .
dt dt
Thus, the tangent vector to the curve is a linear combination of x1 (t0 ) and
x2 (t0 ). All tangent vectors at P to curves passing through P lie in the plane
spanned by the linearly independent vectors x1 (u10 , u20 ) and x2 (u10 , u20 ). This is
called the tangent plane to the surface at P . The line through P perpendicular
to this plane is called the normal line at P . It consists of all vectors of the
form x(u10 , u20 ) + r(x1 (u10 , u20 ) × x2 (u10 , u20 )).
The unit normal for the parametrized surface x is defined by the condition
(x1 × x2 )
N= p .
(x1 × x2 ) · (x1 × x2 )
It is the unique unit vector perpendicular to the tangent plane such that the
triple product (
vx1 , x2 , N) > 0.
By the chain rule, we observed above that a tangent vector v at P can
be written as a linear combination of the partial derivatives of x,
v = ai xi = a1 x1 + a2 x2 ,
where a1 and a2 are the components of v (relative to the basis, x1 , x2 ).

The dot product of v with itself is then
v · v = a1 a1 x1 · x1 + a1 a2 x1 · x2 + a2 a1 x2 · x1 + a2 a2 x2 · x2 ,
which we abbreviate, using the summation convention, as
v · v = ai aj xi · xj .
We introduce the metric coefficients
xi · xj = gij
so we have a final abbreviation
v · v = gij ai aj .
More generally, if w is another tangent vector at P with components bi we

many then write
w = ai xi = a1 x1 + a2 x2 ,
and the dot product of the two vectors is

v · w = ai bj xi · xj = gij ai bj .
We will write these expression using the notation
I(v, w) = v · w and I(v) = I(v, v) = v · v.
This expression is a symmetric bilinear form on the tangent space at P , which

we call the first fundamental form. Note that this form has the property that
I(v) ≥ 0, with I(v) = 0 if and only if v is the zero vector. A form with this
property is called positive definite.
The first fundamental form then gives a bilinear form on each of the
tangent planes of a parametrized surface, and we can use this form to compute
the length of a curve x(t) = x(u1 (t), u2 (t)) between the limits t1 and t2 :
r
t2 t2
dui duj
Z p Z
s= I(x0 (t)) dt = gij dt (3.3)
t1 t1 dt dt
or
2
dui duj

ds
= gij . (3.4)
dt dt dt
In differential form notation, this formula for length can be written
ds2 = gij dui duj .
Exercise 38. Find the metric coefficients, gij , for the ellipsoid with equation
x(u1 , u2 ) = (a sin(u1 ) cos(u2 ), b sin(u1 ) sin(u2 ), c cos(u1 )) ,
where a, b, and c are constants. At which points does the surface fail to be
regular? Find the unit normal vector at regular points.
Exercise 39. Find the metric coefficients, gij , for the surface of revolution
with equation
x(u1 , u2 ) = (u1 cos(u2 ), u1 sin(u2 ), f (u1 )) .
At which points does the surface fail to be regular? What is the length of the
curve u1 (t) = c, u2 (t) = t? Find an expression for the unit normal vector.
Exercise 40. Find the metric coefficients, gij , for the right helicoid
x(u1 , u2 ) = (u1 cos(u2 ), u1 sin(u2 ), g(u2 )) .
What is the length of the curve u1 (t) = t, u2 (t) = c? Find an expression for
the unit normal vector.
3.1 The First Fundamental Form 39
Exercise 41. Find the metric coefficients, gij , for the catenoid with equation
x(u1 , u2 ) = (cosh(u1 ) cos(u2 ), cosh(u1 ) sin(u2 ), u1 ) .
What is the length of the curve u1 (t) = t, u2 (t) = c? What is the length of
the curve u1 (t) = c, u2 (t) = t? Find an expression for the unit normal vector.
Exercise 42. Find the metric coefficients, gij , for the helicoid with equation
x(u1 , u2 ) = (sinh(u1 ) sin(u2 ), − sinh(u1 ) cos(u2 ), u2 ) .
What is the length of the curve u1 (t) = t, u2 (t) = c? What is the length of
the curve u1 (t) = c, u2 (t) = t? Find an expression for the unit normal vector.
Exercise 43. Find the metric coefficients, gij , and the unit normal vector
for the function graph
x(u1 , u2 ) = (u1 , u2 , f (u1 , u2 )) .
Exercise 44. Find the metric coefficients, gij , and the unit normal vector of
the torus
x(u1 , u2 ) = ((a + b sin(u1 )) cos(u2 ), ((a + b sin(u1 )) sin(u2 ), b cos(u1 )) .
3.1.1 Geometry in the Tangent Plane
In order to study the geometry of the tangent plane at a point P on a surface,

we need a notion of the angle φ between two non-zero vectors v and w. This
is obtained by using the first fundamental form on the surface to define
I(v, w)
cos(φ) = p .
I(v)I(w)
In particular, the angle φ1 between a vector v and the vector xi is given by
v · xi aj gij
cos(φi ) = p =
I(v)I(xi ) gii I(v)
with no summation on the index i.

Two vectors v and w in the tangent plane at P are orthogonal if I(v, w) =
0. If v and w are orthogonal, so are the scalar multiples, av and bw. The
totality of all non-zero scalar multiples of a vector v is called the direction
determined by v. Thus, a direction is determined by the ratio a1 : a2 of its
components. Orthogonality is therefore a property of a pair of directions.
Suppose that two directions are determined by the quadratic condition
cij ai aj = 0 where cij = cji. Then these directions are orthogonal if and only
if g22 c11 − 2g12 c12 + g11 c22 = 0. To see this, let p and q denote the solutions of
the quadratic condition, c11 + 2c12 x + c22 x2 = 0, so p and q satisfy pq = cc11

22
and p + q = −2c i j
c22 . Let v = a xi and w = b xj be two non-zero vectors with
12
a2 b2
directions p and q respectively. Set p = a1 and q = b1 . Then
I(v, w) b2 a2 a2 b2
= g 11 + g12 + g 21 + g22
a1 b1 b1 a1 a1 b1
= g11 + (p + q)g12 + (pq)g22
c12 c11
= g11 − 2 g12 + g22
c22 c22
1
= (g22 c11 − 2g12 c12 + g11 c22 ) .
c22
So, I(v, w) = 0 if and only if
g22 c11 − 2g12 c12 + g11 c22 = 0 .
Exercise 45. The curves defined by u1 = constant for a fixed index i are
called parametric curves of the surface. There are therefore two families of
parametric curves on a surface, and through each point of the surface there
passes exactly one parametric curve from each family. Find the condition
that the parametric curves are orthogonal. Under which conditions will the
parametric curves meet at a 45 ◦ angle?
Exercise 46. For each of the surfaces in the problem set for (3.1), find the
angle between parametric curves.
Exercise 47. Consider a family of curves given by the conditions ci ai = 0,

for a fixed set of constants ci . Show that the orthogonal trajectories of this
family satisfy the condition
(g11 c1 − g12 c2 )a1 + (g12 c1 − g22 c2 )a2 = 0 .
Use this formula to find the orthogonal trajectories of the family of circles
given in polar coordinates by r = c cos(φ). Sketch the curves in this family
and sketch their orthogonal trajectories.
3.2 The Second Fundamental Form

The first fundamental form arises from studying the length of curves on a
surface and the angle between vectors in the tangent planes. The study of
curvature for curves on a surface leads to another bilinear form, the second
fundamental form. Let C be a curve on a parametrized surface x(u1 , u2 )
determined by a curve (u1 (t), u2 (t)) in the parameter domain D. We write
x(t) = x(u1 (t), u2 (t)), and we let T(t) and N(t) denote the unit tangent and
3.2 The Second Fundamental Form 41
(unit) principal normal vectors at x(t). As a curve in space, the curvature

k(t) of x(t) is given by the Frenet formula
k(t)s0 (t)N(t) = T0 (t) . (3.5)
We now wish to express this formula in terms of quantities associated with

the parametrized surface.
If φ(t) denotes the angle between the principal normal N(t) and the unit
normal ν(t) to the surface at x(t), then, taking the dot product of both sides
of (3.5) with ν(t) and using the fact that T(t) · ν(t) = 0, we obtain
k(t) cos(φ(t))s0 (t) = T0 (t) · ν(t) = −T(t) · ν 0 (t) . (3.6)
This can also be written
k(t) cos(φ(t))(s0 (t))2 = −T(t)s0 (t)ν 0 (t) = −x0 (t) · n0 (t) . (3.7)
i j
By the chain rule, x0 (t) = xi (t) du 0 du
dt and ν (t) = nj (t) dt . Therefore we may
write
dui duj dui duj
−x0 (t) · ν 0 (t) = −xi (t) · νj (t) = hij (t) (3.8)
dt dt dt dt
where hij = −xi ·νj are the coefficients of a differential form called the second
fundamental form. Notice that the condition xi · ν = 0 leads to the relation
xij · ν = −xi · νj = hij , (3.9)
from which it follows that hij ≤ hji , so the second fundamental form is
symmetric. Recall that
(x1 × x2 ) · (x1 × x2 ) = (x1 · x1 )(x2 · x2 ) − (x1 · x2 )(x2 · x1 ) = g11 g22 − g12 g21
by Lagrange’s identity (1.15). We set g = g11 g22 − g12 g21 , so we may write
I(x1 × x2 ) = g and ν = (x1√×x
g
2)
. Therefore (3.9) can be rewritten
(x1 , x2 , ν)
hij = √ . (3.10)
g
For two tangent vectors v = ai xi and w = bj xj , we define the bilinear form

II(v, w) = hij ai bj to be the second fundamental form on each tangent space
of the parametrized surface. For a single tangent vector v, we set II(v, v) =
II(v) = hij ai aj. Note that II(v, w) = II(w, v). We can now rewrite (3.5)
as
II(x0 (t))
k(t) cos(φ(t)) = . (3.11)
I(x0 (t))
The right hand side of this expression depends only on the direction of x0 (t).
This indicates that k(t) cos φ(t) will be the same for any two curves through
a point that have the same tangent direction there. Given a vector v = ai xi
in the tangent plane at a point, there are many curves C = x(t) through the
point on the surface so that the tangent vector x0 (t) has the same direction
as v. One such curve is obtained by cutting the surface by the plane deter-
mined by v and the surface normal n at P , the so-called normal section at
P in the direction of v. The curvature of the normal section in the direction
of v is called the normal curvature and is denoted kN (v). Hence we may
rewrite (3.11) as
II(x0 (t))
k(t) cos(φ(t)) = = kN (x0 (t)) . (3.12)
I(x0 (t))
This relationship is known as Meusnier’s Theorem. It determines the curva-
ture k(t) of a curve in terms of the normal curvature and the angle between
the principal normal and the surface normal, provided that the cos φ(t) is
not zero, i.e. provided that the principal normal does not lie in the tangent
plane at the point. If N(t) does lie in the tangent plane at x(t), then the
osculating plane of the curve coincides with the tangent plane of the surface
at that point. Otherwise the tangent line of the curve is the intersection of
the osculating plane of the curve and the tangent plane of the surface, and
it follows that two curves through P on the surface with the same osculating
plane at P must have the same curvature there.
Exercise 48. Find the normal curvature at a point of the circle of latitude
x(u1 , u2 ) = (cos(u1 ) cos(u2 ), cos(u1 ) sin(u2 ), sin(u1 ))
where u1 = c, a constant.
Exercise 49. Calculate the normal curvature at the point (0, 0, 0) for the
curve determined by u1 (t) = t cos(b), u2 (t) = tsin(b) on the surface
x(u1 , u2 ) = (u1 , u2 , p(u1 )2 + q(u2 )2 ) for constants p and q.
Exercise 50. Find the normal curvature of the parametric curves in the
exercises 1-6 in section 1.1. Find the coefficients of the second fundamental
form for each of these surfaces.
3.2.1 The Shape of a Surface
Consider a parametrized surface x(u1 , u2 ) such that x(0, 0) = P. By Taylor’s

Theorem, we have
x(u1 , u2 ) = x(0, 0) + xi (0, 0)ai + 12 xij (0, 0)ai aj + O(r3 )
where r2 = (u1 )2 + (u2 )2 . Then the distance from x(u1 , u2 ) to the tangent
plane at x(0, 0) is given by
(x(u1 , u2 ) − x(0, 0)) · ν(0, 0) = 12 hij (0, 0)ai aj + O(r3 ) ,

where we set xij (0, 0) · ν(0, 0) = hij (0, 0). This expression is dominated by
the quadratic term, so the shape of the surface will be similar to the shape
of the graph of the quadratic function f (x1 , x2 ) = h11 x21 + 2h12 x1 x2 + h22 x22 .
This shape is determined by the expression h = h11 h22 −(h12 )2 . When h > 0,
the graph of the quadratic function intersects the horizontal plane only at
the origin, while if h < 0, the intersection is a pair of intersecting lines.
The point P of a parametrized surface is called elliptic, hyperbolic, or
parabolic according to if h is positive, negative, or zero at P. At an elliptic
point P , the tangent plane meets the surface only at P locally, i.e. there is
a neighborhood of P that meets the tangent plane only at P. On the other
hand, if P is hyperbolic, the intersection of the tangent plane and a small
neighborhood of P will consist of two curves with distinct tangent lines, so
arbitrarily close to P , there are points of the surface on both sides of the
tangent plane. At a parabolic point, the behavior of the surface can be quite
a bit more complicated (see the exercises at the end of this section for some
examples.)
The behavior of a parametrized surface in a neighborhood of P can also
be described in terms of the normal sections at P , i.e. the intersections of
the surface with planes through the normal line at P . If the curvature κN of
such a normal section is non-zero, then the center of the osculating circle to
this curve is P + κ1N ν. At an elliptic point, all normal curvatures will have
the same sign and the centers of osculating circles for all normal sections
lie on the same side of the tangent plane. At a hyperbolic point, the normal
curvature takes on both positive and negative values, and the centers of some
osculating circles lie on one side of the tangent plane and some will lie on the
other. At a parabolic point, at least one normal section has normal curvature
0.
Exercise 51. For all surfaces in the exercises 1-6 of section 1.1, which points
are elliptic, which are hyperbolic and which are parabolic?
Exercise 52. Let x(u1 , u2 ) = (u1 , u2 , (u1 )3 −3u1 (u2 )2 ). Show that the origin
is the only parabolic point and that all of the metric coefficients hij are zero
at this point. (Such a point is called planar.) What is the intersection of the
surface and the tangent plane at this point? (This point is called a monkey
saddle. A saddle for a person riding a bicycle has two dips, but a saddle for
a monkey requires a third depression for the tail.)
Exercise 53. Let x(u1 , u2 ) = (u1 , u2 , (u1 )4 + (u2 )4 ). Show that the origin
is the only parabolic point which is planar (i.e. all of the metric coefficients
hij are zero at this point.) What is the intersection of the surface and the
tangent plane at this point?
Exercise 54. Let x(u1 , u2 ) = (u1 , u2 , (u1 )4 + c(u2 )2 ) for a constant c. Show
that the origin is the only parabolic point. What is the intersection of the
surface and the tangent plane at this point?
Exercise 55. Let x(u1 , u2 ) = (u1 , u2 , ((u1 )2 −u2 )((u1 )2 −cu2 )) for a constant
c. Show that the origin is a parabolic point. What is the intersection of the
surface and the tangent plane at this point? Which points are parabolic,
which are ellipic, and which are hyperbolic?
Exercise 56. Show that all points on a hyperboloid of one sheet are hyper-
bolic:
x21 x2 x2
2 + 22 + 23 = 1 .
a1 a2 a3
Exercise 57. For the torus of revolution of problem 7 in section 1.1, which
points are parabolic, which are elliptic, and which are hyperbolic?
3.2.2 Characterization of the Sphere
A point P is called an umbilic if the curvatures of all normal sections are

equal, so the normal curvature is independent of the direction in the tangent
space. Since κN = II(v)
I(v) , the condition for P to be an umbilic is equivalent
to hij = cgij for all indices i, j. The constant c = c(u1 , u2 ) depends on the
point P .
For a plane, all points are umbilic with normal curvature 0. For a sphere
of radius r, all points are umbilic with normal curvature 1r . We shall now
prove the converse.
Proposition 10. A surface consisting entirely of umbilics is either a plane

or a sphere.
We have c(u1 , u2 )xi · xj = −xij so (cxj + νj ) · xi = 0. Furthermore

ν · ν = 1 implies that ν · νj = 0, so (cxj + νj ) · ν = 0 as well. Therefore
cxj + νj , being perpendicular to three independent vectors, must be the zero
vector, so cxj + xj = 0. By differentiating this expression, we have
cxij + ci xj + νij = 0 (3.13)
and similarly,
cxji + cj xj + νij = 0 . (3.14)
Since we are assuming that the second partial derivatives of x and ν are
continuous, we have xij = xji and νij = νji . Subtracting (3.13) from (3.14),
we obtain ci xj − cj xi = 0. Since x1 and x2 are linearly independent by
hypothesis, it follows that ci = 0 = cj identically, so c is constant.
If c = 0, then νi = 0 = νj so ν is constant, equal to a fixed unit vector a.
Then (x · a)i = xi · a = 0 for i = 1, 2 so x · a = d, a constant, and x therefore
lies in the plane perpendicular to a at distance d from the origin.
If c 6= 0, then (x − ( 1c )n)i = 0f ori = 1, 2, so x − (1/c) = c, a con-

stant vector, and x(u1 , u2 ) − C = 1c (u1 , u2 ) for all u1 , u2 . Thus the length
of x(u1 , u2 ) − C is a constant, 1c , and the parametrized surface lies on the
sphere of radius 1c about the center C.
Corollary 2. A surface consisting entirely of planar points is a plane.
3.2.3 Principal Curvatures, Principal Directions, and Lines of

Curvature
We now wish to study the normal curvature κN as a function of the tangent

vector v = ai xi 6= 0. As remarked earlier, κN depends only on the direction
of the vector v and not on its length, so we consider only unit vectors v, for
which gij ai aj = 1. Then κN is a continuous function of the variables (a1 , a2 )
on the unit circle in the tangent plane so it must have a maximum and a
minimum. We set I = I(a1 , a2 ) and II = II(a1 , a2 ). Then ∂I j
ai = 2gij a and
∂II j
ai = 2hij a so
∂κN [(2gij aj )II − (2hij aj )I] 2(κN gij aj − hij aj )

i
= 2
= .
a I I
At a maximum or minimum of κN , we therefore have
hij aj − κN gij aj = 0 .
Thus,
(hi1 a1 + hi2 a2 ) = κN (gi1 a1 + gi2 a2 )
so, by eliminating κN , we obtain
(h11 a1 + h12 a2 )(g21 a1 + g22 a2 ) = (h21 a1 + h22 a2 )(g11 a1 + g12 a2 ) ,
Expanding this out, we obtain the condition
(g11 h21 − g21 h11 )(a1 )2 + (g11 h22 − g22 h11 )(a1 a2 ) + (g12 h22 − g22 h12 )(a2 )2 = 0 .
At an umbilic this expression is satisfied identically, and at a non-umbilic

point, there must be two solutions since the function must have at least one
maximum and one minimum. The solutions of this quadratic equation give
the directions for which κN attains its maximum and minimum, and these
are called principal directions. By our previous computation in section 1.2,,
it follows that the principal directions at a non-umbilic point are orthogonal.
The maximum and minimum values of κN are called principal curvatures.
To determine them, we rewrite
(hi1 a1 + hi2 a2 ) = κN (gi1 a1 + gi2 a2 )

as
(hi1 − κN gi1 )a1 + (hi2 − κN gi2 )a2 = 0 .
Since this system of equations has non-trivial solutions, we must have
(h11 − κN g11 )(h22 − κN g22 ) − (h21 − κN g21 )(h12 − κN g12 )
so κN satisfies the quadratic equation
g(κN )2 − mκN + h = 0
where m = g11 h22 − 2g12 h12 + g22 h11 . The roots are the principal curvatures,
denoted κ1 and κ2 .
The average of the roots of this equation is called the mean curvature of
the surface at the point, denoted by H, so
κ1 + κ2 m (g11 h22 − 2g12 h12 + g22 h11 )
H= = = .
2 2g 2g
The product of the roots of this equation is called the total curvature or
Gaussian curvature of the surface at the point, so
h
K = κ1 κ2 = .
g
The principal curvatures, mean curvature, and Gaussian curvatures are
scalar functions defined on the parametrized surface. They play an important
role in the description of the geometric properties of the surface. For example
a point is elliptic, parabolic, or hyperbolic according to if K is positive, zero,
or negative. Furthermore
(κ1 − κ2 )2
H2 − K = ≥0,
4
with equality only at an umbilic.
A curve for which all unit tangent vectors are principal directions is called
i
a line of curvature on the surface. For such a curve, x0 (t) = xi du dt and the
condition that each such vector determine a principal direction is
1 2 1 2
du du1 dtwo du
g11 h21 − g21 h11 ) + (g11 h22 − g22 h11 ) + (g12 h22 − g2 h12 ) =0.
dt dt dt dt
In the language of differential forms, this can be written
(g11 h21 − g21 h11 )(du1 )2 + (g11 h22 − g22 h11 )du1 du2 + (g12 h22 g22 h12 )(du2 )2 = 0 .
This is a quadratic equation in the first derivatives of the component
functions u1 (t) and u2 (t), and by a standard existence theorem in the theory
of ordinary differential equations, it follows that in any portion of the domain
where there are no umbilics, it is possible to find two families of solution
curves, the lines of curvature. These form an orthogonal net, so that at each
point there are two lines of curvature with perpendicular tangent vectors.
3.4 Triply Orthogonal System; Theorems of Dupin and Liouville 47
Exercise 58. Find conditions in terms of gij and hij for the parametric
curves to be lines of curvature.
Exercise 59. For which of the six surfaces in the exercises 1-6 of section 1.1
are the parametric curves lines of curvature?
Exercise 60. Find the lines of curvature for the right helicoid.
3.3 Gauss Mapping and the Third Fundamental Form

The unit normal vector ν = (u1 , u2 ) can be considered as a point on the unit
sphere S0 centered at the origin of our space E. The resulting mapping
ν : x 7→ S0
which assigns to the point x(u1 , u2 ) the point ν(u1 , u2 ), is called the Gauss
map. Using the first fundamental form on S, we define the third fundamental
form
III(v, w) = mij ai bj
where v = ai xi and w = bj xj and mij = νi · νj .
3.4 Triply Orthogonal System; Theorems of Dupin and

Liouville
The notion of a parametrized surface can be generalized to that of a triple
system, as follows: Let ∆ be an open domain in the (ul , u2 , u3 )-space. Consider
a mapping
x : ∆ 7→ E (3.15)
of ∆ into our Euclidean space E such that x(u1 , u2 , u3 ) has continuous first
and second partial derivatives and that the Jacobian determinant
∂(x1 , x2 , x2 )
6= 0 . (3.16)
∂(u1 , u2 , u3 )
The last condition can also be written
det(x1 , x2 , x3 ) 6= 0 (3.17)
where
∂x
xi = (3.18)
ui
for i = 1, 2, 3.
The name triple system derives from the fact that each of the equations
defines a family of parametric surfaces in E. The system is called a triply
orthogonal system, if
xi · xj = 0 (3.19)
for i, j = 1, 2, 3, i 6= j, i.e., two surfaces of different families are everywhere

orthogonal.
Example 5. Consider the spherical coordinates defined by the equations
x = r cos u1 cos u2
x = r cos u1 sin u2 (3.20)
1
x = r sin u .
The three families of surfaces are:

1. The spheres r = constant.
2. The cones of revolution u = constant.
3. The half-planes u2 = constant.
We leave it as an exercise to verify that (3.20) defines a triply orthogonal
system.
Theorem 6 (Dupin’s Theorem). The surfaces of a triply orthogonal sys-

tem intersect each other in lines of curvature.
Proof. We take a surface u3 = constant. It suffices to show that on it the

parametric curves u1 = constant and u2 =constant are lines of curvature. We
differentiate (3.19) with respect to uk , for k 6= i, j, getting xik ·xj +xjk ·xi = 0.
These equations are, when written in detail,
x13 · x2 + x23 · x1 = 0
x21 · x3 + x31 · x2 = 0
x32 · x1 + x12 · x3 = 0 .
It follows that
x12 · x3 = x23 · x1 = x31 · x2 = 0 .
The first equation shows that x12 is orthogonal to x3 . By (3.19), so are x1

and x2 . Hence we have
det(x1 , x2 , x3 ) = 0 .
On the surface u3 = constant we have therefore g12 = h12 = 0.

A non-trivial example of a triply orthogonal system is given by the “con-

focal quadrics.” For central quadrics this is the family defined by the equation
x21 x22 x23
+ + =1 (3.21)
t − a1 t − a2 t − a3
where t is the parameter and where we assume 0 < a3 < a2 < a1 . If (xl , x2 , x3 )
are the coordinates of a given point distinct from the origin, then (3.21) is a
cubic equation in t, which is easily seen to have three real roots t1 , t2 , and t3
satisfying the inequalities
t 1 > a1 > t 2 > a2 > t 3 > a3 . (3.22)
Through this point there pass therefore three quadrics of the family: an el-
lipsoid t1 , a hyperboloid of one sheet t2 , and a hyperboloid of two sheets t3 .
It can be verified that they are mutually orthogonal. (t1 , t2 , t3 ) can be taken
as coordinates in a domain not containing the origin. They are called ellip-
tic coordinates. These coordinates are useful in studying the properties of
ellipsoids, which are among the most important surfaces.
Since every central quadric can be considered a member of a confocal
family, the following property of its lines of curvature follows from Dupin’s
theorem:
Proposition 11. The lines of curvature of a central quadric are its curves
of intersection by the confocal quadrics and are generally quartic curves.
An application of Dupin’s theorem gives Liouville’s theorem on conformal
mappings in the three-dimensional Euclidean space E. A mapping of a do-
main of E into another domain is called conformal, if it preserves the angle
between two curves. A motion in E is conformal; so is an inversion (INSERT
REFERENCE HERE). We say that a surface is totally umbilical if it consists
entirely of umbilics. By (10), a totally umbilical surface is either a plane or
a sphere. In fact, we proved that such a surface satisfies the condition
νi + ρ xi = 0 (3.23)
for i = 1, 2 where ρ is a constant.

Theorem 7 (Liouville’s Theorem). A conformal mapping E maps a to-
tally umbilical surface into a totally umbilical surface.
Proof. Let S be a totally umbilical surface, given analytically by x(u1 , , u2 ).
We suppose that the parametric curves are orthogonal, so that
x1 · x2 = 0 . (3.24)
Let
x(u1 , u2 , u3 ) = x(u1 , u2 ) + u3 ν(u1 , u2 ) . (3.25)

Then
xi = xi + u3 νi = (1 − ρu3 )xi (3.26)
for i = 1, 2. So,
∂x
x3 = =ν. (3.27)
∂u3
It follows that x defines a triply orthogonal system where 1 − ρ u3 6= 0.
Let S 0 be the image of S under the conformal mapping T . Under a triply
orthogonal system goes into a triply orthogonal system. It follows by Dupin’s
theorem that the parametric curves on S 0 are lines of curvature. Let C 0 be
any curve on S 0 . We can choose parameters u1 and u2 on S 0 so that the
parametric curves are orthogonal and C 0 is a parametric curve. Under the
mapping T −1 these can be used as parameters on S. T being conformal, S
has orthogonal parametric curves. By what was proved above, C 0 is a line
of curvature on S 0 . Since any curve on S 0 is a line of curvature, S 0 is totally
umbilic.
Remark 7. Liouville’s theorem is a local theorem. It can be proved that the
locally conformal transformations in E depends on ten parameters and are
generated by the similarity transformations and inversions. In contrast the
locally conformal transformations in the plane depend on arbitrary functions.
Exercise 61. The cylindrical coordinates r, θ, z in E are defined by
x = r cos θ
x = r sin θ
x=z.
Show that they form a triply orthogonal system and describe the three fam-
ilies of surfaces.
Exercise 62. Prove the following properties on confocal quadrics stated
above:
a) Through a point distinct from the origin there pass three confocal quadrics:
an ellipsoid, a hyperboid of one sheet, and a hyperboloid of two sheets.
b) Two confocal quadrics of different families meet orthogonally.
Exercise 63. Let x∗i = fi (x1 , x2 , x3 ) for i = 1, 2, 3 define a mapping in E
with
∂(x∗1 , x∗2 x∗3 )
6= 0 .
∂(x1 , x2 , x3 )
Prove that it is conformal if
dx∗2 ∗2 ∗2 2 2 2
1 + dx2 + dx3 = λ(dx1 + dx2 + dx3 )
for λ > 0. Hence prove that an inversion is conformal.

Exercise 64. Let z = x + ıy and w = u + ıv be complex variables and

w = f (z) be a holomorphic function. The function defines a mapping
(x, y) 7→ (u, v). Show that it is conformal at a point z0 . where f 0 (z0 ) 6= 0.
Hence show that a conformal mapping in the plane generally does not map
a circle into a circle or a line.
4 Fundamental Equations in Surface Theory
In curve theory, the fundamental invariants are the curvature and torsion,
and the fundamental equations are the Frenet formulas. These express the
derivatives of the vectors T, N, and b of the Frenet frames of a curve in terms
of these vectors T, N, and b themselves, with the coefficients essentially being
the curvature and torsion.
The situation in surface theory is more complicated, but conceptually no
different. The fundamental invariants are now the coefficients of the first and
second fundamental forms, namely the six functions gij and hij of the pa-
rameters (u1 , u2 ). These functions are not independent, and the relationship
between them is contained in the theorems of Gauss and Codazzi, to be es-
tablished in (INSERT REF) of this chapter. Taking the place of the Frenet
frame is the frame x1 , x2 , ν, where x1 and x2 are not orthogonal unit vectors
in general. The counterpart of the Frenet equations are the Weingarten and
Gauss equations, expressing the partial derivatives of x1 , x2 , ν in terms of
the vectors x1 , x2 , ν themselves.
4.1 Weingarten and Gauss Equations

Since any vector in E can be written as a linear combination of the three
linearly independent vectors x1 , x2 , ν, we may write
νi = Aji xj + Bi ν .
To determine the coefficients Aji and Bi , we take the dot product with ν, and
the fact that ν · ν = 1 immediately leads to ν · νi = 0 so Bi = 0. Taking the
dot product of νi with xj leads to the expression (MISSING EXPRESSION)
and we may then solve for the coefficients Aji . To do so, we introduce the
inverse matrix (g ij ) to the matrix (gij ), so
gij g jk = dki .
We then get
−hik g km = Aji gjk g km = Aji dm m

j = Ai .
54 4 Fundamental Equations in Surface Theory
Thus in the original expression, we have Aji = −hik g kj , which we now denote
by −hji . The Weingarten equations for the surface are then
νi = −hji xj .
From the Weingarten equations, we immediately derive the theorem,

proved earlier:
Theorem 8. A surface consisting entirely of planar points is a plane.
This follows since the hypothesis implies that hij = 0 for all i, j, so νi = 0 =
νj and is constant. This gives xi · ν = (x · ν)i = 0 and (x · ν)j = 0 so x · ν = c,
a constant, the condition for the surface x to lie in a plane.
More subtle is another consequence of the Weingarten equations:
Proposition 12. A surface with Gaussian curvature K = 0 is, in the neigh-

borhood of a non-planar point, generated by a family of lines such that the
tangent plane remains the same along any generator.
Proof. By hypothesis, we have h11 h22 − (h12 )2 = 0 where h11 and h22 are
not both zero. Suppose h 6= 0. Then the equation of the asymptotic curves is
p p 2
h11 (du1 )2 + 2h12 du1 du2 + h22 (du2 )2 = 0 = h11 du1 + h22 du2
r
1 h22 2
du = du .
h11
We can now make a change of parameters from (u1 , u2 ) to (u1 , u2 ) so that the
parametric curve u1 = constant are asymptotic. Then h22 = 0 and h12 = 0
as well so ν2 = 0 and ν = ν(u1 ) is a function of u1 alone. This implies that
the normal vectors remain parallel to one another along an asymptotic curve.
Also, we have
(x · ν)2 = x2 · ν + x · ν2 = 0 ,
so x · ν = p(u1 ) is a function depending only on u1 . Differentiating with

respect to u1 gives x · ν1 = p0 (u1 ). Note that
ν1 = −hj1 xj = −h1k g kj xj 6= 0
since h11 6= 0 6= g11 . Therefore the asymptotic curve lies in this plane x · ν1 =
p0 (u1 ). But it already lies in the plane x · ν = c and the vectors ν and ν1 are
distinct, so it follows that the asymptotic curve lies in the intersection of two
distinct plane, and therefore that it is a straight line. As remarked earlier,
the normal vector is constant along this line, so the theorem is proved.
A surface with K = 0 and no planar points is called developable. The rea-

son for this is that any such surface is isometric to a portion of the plane. We
4.1 Weingarten and Gauss Equations 55
may approach this fact in a different way by concentrating on ruled surfaces

and singling out those for which the Gaussian curvature is everywhere 0. A
ruled surface is of the form x(u1 , u2 ) = y(u1 ) + u2 z(u1 ) for a regular curve
y(u1 ) and a unit vector field z(u1 ). Then x1 (u1 , u2 ) = y0 (u1 ) + u2 Z 0 (u1 )
while X2 (u1 , u2 ) = z(u1 ). Thus
x1 (u1 , u2 ) × x2 (u1 , u2 ) = y0 (u1 ) × z(u1 ) + u2 Z 0 (u1 ) × z(u1 ) .
When we normalize this vector, the result will be independent of u2 only

when y0 (u1 ) × z(u1 ) = 0 identically, i.e. the vector z(u1 ) is a multiple of the
tangent vector y0 (u1 ) so we may assume z(u1 ) = T(u1 ), the unit tangent.
Thus the ruled surface is the tangential surface of the curve y(u1 ) and we
have already seen that the metric coefficients of that surface depend only on
the curvature of the curve, not on its torsion. There is therefore a curve in the
plane with the same curvature and torsion zero, and the tangential surface
of this curve, a portion of the plane, will be isometric to the original surface.
The Gauss equations express the second partial derivatives Xik in terms
of the vectors x1 , x2 , and ν. Let
j
xik = Γik xj + cik ν
j j
where Γik = Γki , cik = cki . Taking the dot product with ν gives cik =
xik · ν = hik by the definition of the second fundamental form. To determine
j
the coefficients Γik , we take the dot product with xm to obtain
j j
xik · xm = Γik xj · xm = Γik gjm .
j
We set Γimk = Γik gjm (being careful about the order of the subscripts). Note
that Γimk = Γkmi but in general there is no symmetry in the other subscripts.
j
Using the inverse of the first fundamental form, we may write Γik =
jm
Γimk g . This process is called “raising or lowering the indices” and the sets
j
of coefficients Γik and Γimk (or hji and hij ) are called “associated”; one set
completely determines the other.
We then have xik · xm = Γimk . It follows that
∂gim ∂
k
= xi · xm = xik · xm + xi · xmk = Γimk + Γmik ,
∂u ∂uk
and similarly
∂gmk
= Γmki + Γkmi ,
∂ui
∂gki
= Γkim + Γikm .
∂um
Adding these last two equations and subtracting the first, we obtain:

1 ∂gmk ∂gki ∂gim
Γikm = + − .
2 ∂ui ∂um ∂uk
j
It follows that the coefficients Γikm , and hence also the Γik , can be expressed
completely in terms of the coefficients gik of the first fundamental form and
j
their partial derivatives. The coefficients Γikm and Γik are called the Christof-
fel symbols of the first and second type. It is remarkable that they are intrinsic,
depending only on the first fundamental form of the surface.
In summary, the Gauss equations of the surface are
j
xik = Γik xj + hik ν .
Exercise 65. When Gauss originally introduced the first and second funda-
mental forms, he did not use multi-indexed quantities, rather writing
I = Edu2 + 2F dudv + Gdv 2

II = Ldu2 + 2M dudv + N dv 2 .
1 EN −2F M +GL LN −M 2
Exercise 66. Show that H = 2 EG−F 2 and K = EG−F 2 .
Exercise 67. Suppose that the lines of curvature are parametric curves, so
F = 0 = M . Show that the Weingarten equations become νu = − LX E
u
and
N Xv
νv = − G , and use these to prove the identity KI −2HII +III = 0 relating
the three fundamental forms.
Exercise 68. Use the result of the previous problem to prove that for an
asymptotic curve, we have w2 = −K, where w denotes the torsion of the
curve (a theorem of Beltrami and Enneper).
Exercise 69. If the parametric curves are orthogonal, so F = 0, show that
1 1 Eu 2 1 Ev
Γ11 = , Γ11 =− ,
2 E 2 G
1 1 Ev 2 1 Gu
Γ12 = , Γ12 = ,
2 E 2 G
1 1 Gu 2 1 Gv
Γ22 =− , Γ22 = .
2 E 2 G
Exercise 70. Determine the Christoffel symbols corresponding to polar co-
ordinates in the plane.
i ∂
Exercise 71. Prove that Γik = ∂uk
log g where g = g11 g22 − (g12 )2 .
4.2 Levi-Civita Parallelism

Levi-Civita parallelism follows as an application of the Gauss equations.
Consider a curve C on a surace x defined by ui = ui (t). At each point of
C, let y(t) = y i (t)xi (t) be a tangent vector to x, but not necessarily to C.
Then
4.2 Levi-Civita Parallelism 57
dy i
y0 (t) = xi (t) + y i (t)x0i (t)
dt
dy i duk

j
= xi (t) + y (t) xjk (t)
dt dt
i k
dy du
xi (t) + y j (t) i

= Γjk xi (t) + hjk ν
dt dt
i k
duk

dy i du
= + Γjk y (t) xi (t) + y j (t)
j
hjk ν .
dt dt dt
The vector
i k

Dy dy i du j
= + Γjk y (t) xi (t)
dt dt dt
is the projection of the derivative of y(t) into the tangent space at y(t). From
the vector field y(t), we thus obtain another tangent vector field Dy dt called
the absolute or covariant derivative of yY.
In differential formalism, we may write the absolute or covariant differ-
ential as
Dy = dy i + Γjk i
duk y j (t) xi (t) .

This differential form depends only on the first fundamental form of x, on

the equations ui (t) determining the curve C, and the vector field y(t).
Let z(t) be a second tangential vector field along C. Then
d(y · z) = dy · z + y · dz = Dy · z + y · Dz ,
since the differences Dy − dy and Dz − dz are multiples of the normal vector

so their dot product with the tangent vectors z and y will be zero.
The vector field y(t) is said to be parallel along C if Dy = 0 for all t.
It follows that if two vector fields y and z are parallel along C, then their
dot product is constant. In particular, if y is a parallel vector field, then the
length of y(t) is constant.
The condition that y(t) be parallel along C is then
dy i i du
k
+ Γjk y j (t) = 0 .
dt dt
This is a system of linear homogeneous ordinary differential equations of
first order, with the coefficients y i (t) as dependent variables and t as the
k
independent variable. Note that the coefficients Γjk i
and du
dt are all functions
of t. By a basic existence theorem in ordinary differential equations, there
exists a uniquely determined solution y i (t), which takes given initial values
y i (t0 ) at t = t0 . Geometrically this can be described by saying that a vector
y(t0 ) at t0 can be displaced parallelly along C.
As an example, consider the two-dimensional Euclidean plane with first

fundamental form ds2 = (du1 )2 + (du2 )2 . Then Γjk
i
= 0 and the above equa-
i
tions show that a vector field y(t) is parallel if and only if dy
dt = 0 for all t, so
the coefficients y i (t) are constant. Thus in this case, Levi-Civita parallelism
coincides with ordinary parallelism.
Consider two surfaces x and x intersecting along a curve C. The surfaces
are said to be tangent along C if they have the same normal vectors at each
point of C. Since the absolute differential is defined by the use of the normal
vector, it follows that when two suraces are tangent to each other along C,
the absolute differential is the same whether C is considered as a curve on
either of the surfaces.
This property leads to the following geometrical construction of Levi-
Civita parallelism: The tangent planes to x along C envelop a developable
surface S.
We can make this more explicit by observing that the tangent planes at
x(t0 ) and x(t0 + h) will intersect in a line perpendicular to both ν(t0 ) and
ν(t0 + h). This line has direction
ν(t0 ) × ν(t0 + h) = ν(t0 ) × [ν(t0 + h) − (t0 )] ,
and the limit of this as h tends to zero will have the same direction as ν(t0 ) ×
ν 0 (t0 ) = v(t0 ). Then the surface z(t, v) = x(t) + vv(t) will have
z1 = x0 (t) + vv0 (t) , z2 = v(t) ,
so
z1 × z2 = x0 (t) × v(t) + vv0 (t) × v(t) .
But v(t) = ν(t) × ν 0 (t) so v0 (t) = ν(t) × ν 00 (t) so v0 (t) lies in the tangent
plane. It follows that the normal to z is the same as the normal to x. We
would like to show that the surface z has Gaussian curvature everywhere 0.
But z22 = 0 so −h22 = 0 and z12 = v(t) so z12 · ν(t) = −h12 = 0 also.
Therefore h11 h22 − h212 = 0 and K = 0.
To determine parallelism of a vector field y along C it suffices to consider
parallelism along C considered as a curve on S. But S is isometric to a region
in the plane, where the parallelism coincides with ordinary parallelism. We
may then construct parallelism on the original surface x by reversing the
process.
The curve C defined by the coordinate functions ui (t) is called a geodesic
if it is auto-parallel, i.e. if the tangent vector field is parallel to itself along
C. If x(t) gives the equation for the curve C, then the tangent vector is
i
dui
x0 (t) = du i
dt xi (t). Thus in the above calculations, we use y (t) = dt and the
Dx0 (t)
condition for dt = 0 is
d2 (ui ) k
i du du
j
+ Γ jk =0.
dt2 dt dt
This is the second-order system of ordinary differential equations satisfied

by the geodesics, with dependent variables ui (t) and independent variable
t. By a basic existence theorem in ordinary differential equations, there is a
uniquely determined solution ui (t) that takes on given initial values ui (t0 )
i
and dudt at t = t0 . Geometrically, this means that for each point and each
tangent vector at that point, there is a uniquely determined geodesic through
that point with the given tangent vector as its derivative at the point.
Since the tangent vectors to the geodesic are parallel, in particular they
have the same length. There is no loss of generality in assuming that these
tangent vectors are all of unit length, so t becomes the arc-length parameter.
We now show that the geodesics on a surface can be characterized in
terms of the previously defined geodesic curvature.
Proposition 13. The geodesics on a surface are the curves of zero geodesic
curvature.
00
Proof. The geodesic curvature of a curve x(t) is given by kg (t) = xx0 (t)·x
(t)·u(t)
0 (t) ,
where u(t) = ν(t) × T(t). It can also be determined by the equation

dT DT
kg (t)s0 (t) = · u(t) = ·u.
dt dt
Thus the condition that DT
dt = 0 implies that the geodesic curvature is zero.
Moreover, since T · T = 1, we have
dT DT
·T=0= ·T
dt dt
DT
so dt is a multiple of u, specifically
DT
= kg (t)s0 (t)u(t) .
dt
DT
Thus if the geodesic curvature is zero along the curve, it follows that dt =0
so the curve is a geodesic.
Corollary 3. A curve C with curvature k 6= 0 on a surface is a geodesic if
and only if the principal normal is a multiple of the surface normal at every
point.
Proof. By a previous exercise, kg (t) = k(t) sin(φ(t)) where φ(t) is the angle
between the principal normal to the curve and the normal vector to the
surface.
Exercise 72. On the sphere
x(u1 , u2 ) = (a sin(u1 ) cos(u2 ), a sin(u1 ) sin(u2 ), a cos(u1 ))
consider the latitude circle u1 = a, constant. Show that by displacing a vector
parallelly once along the circle, the vector turns an angle of 2π(1 − cos a).
Exercise 73. Two directions v = ai xi and w = bj xj are said to be conjugate

if hij ai bj = 0. Consider the tangent planes to a surface x along C. Show that
the limit of the intersection of two neighboring tangent planes is conjugate
to the tangent vector to C.
Exercise 74. In the x − y-plane, consider the conic ax2 + 2bxy + cy 2 = 1,

ac − b2 6= 0, with center at the origin (0, 0). Two lines through the origin with
slopes m and m are called conjugate if a + b(m + m) + cmm = 0. Show that
the midpoints of a family of parallel chords lie on a line through the origin
which is conjugate to the chord of the family through the origin.
Exercise 75. Show that the geodesic curvature of the latitude circle in (72)
is cota a . Hence a latitude circle is a geodesic if and only if it is the equator.
The Levi-Civita parallelism follows as an application of the Gauss equa-

tions (??).
Consider on the surface S a curve C defined by
ui = ui (t) , i = 1, 2 . (4.1)
At each point of C let
y(t) = y i (t)xi (4.2)
be a tangent vector to S, but not necessarily to C. By (??) its derivative is

i k

dy dy i du j
= + Γjk y xi + (...)ν .
dt dt dt
The vector
dy i k

Dy i du j
= + Γjk y xi (4.3)
dt dt dt
is its orthogonal projection into the tangent plane. y(t) defines a tangent
vector field to S along C and DY
dt is called its absolute or covariant derivative.
It will be convenient to omit the factor dt; then
Dy = (dy i + Γijk duk y j )xi
is called the absolute or covariant differential. The covariant differential de-

pends only on the first fundamental form of S, and on C and y(t).
Let z(t) be a second tangent vector field to S along C. Then
d(yz) = dyz + yDz . (4.4)
This follows because Dy and Dz differ respectively from dy and dz by terms

involving only the normal vector v, whose dot product with a tangent vector
is zero.
The vector field y(t) is said to be parallel along C in the sense of Levi-
Civita if Dy = 0. It follows from (4.4) that if two vector fields are both
parallel along C, their scalar product is constant. In particular, the length of
a parallel vector field is constant.
It follows from (4.3) that the condition for the vector field (4.2) to be
parallel along C is
dy i duk j
+ Γijk y =0. (4.5)
dt dt
This is a system of linear homogeneous differential equations of the first order
with y i as dependent variables and t as the independent variable. In fact,
k
both Γijk and dudt are functions of t. By an existence theorem on differential
equations there exists a uniquely determined solution y i (t), which takes given
initial values at t = t0 . Geometrically this can be described by saying that a
vector y(t0 ) at t0 can be displaced parallelly along C.
Example 6. S is the plane with the first fundamental form
ds2 = (du1 )2 + (du2 )2 .
Then Γjik = 0 and equations (4.5) show that y(t) is parallel if y i = const.
Thus Levi-Civita parallelism coincides with ordinary parallelism.
Consider two surfaces S and S ∗ intersecting in a curve C. They are said

to be tangent to each other along C if they have the same normal along C.
Since the absolute differential is defined by the use of the normal vector, we
have immediately: When two surfaces S and S ∗ are tangent to each other
along C, the absolute differential remains the same whether C is considered
as a curve on S or on S ∗ .
This property gives the following geometrical construction of the Levi-
Civita parallelism: The tangent planes to S along C envelop a developable
Σ. The parallelism is the same by considering C as a curve on Σ. Roll Σ
onto a plane and the parallelism becomes the ordinary one. The parallelism
on S can be constructed by reversing this process. (INSERT REFERENCE
HERE).
The curve C defined by (4.1) is called a geodesic if it is auto-parallel, i.e.,if
its tangent vector is parallel to itself along C. By putting y i = du
dt in (4.5) we
get the differential system satisfied by the geodesics:
d2 ui j
i du du
k
+ Γjk =0.
dt2 dt dt
It is a second-order system with the dependent variables ui and the inde-
pendent variable t. By a known existence theorem on ordinary differential
equations there is a uniquely determined solution ui (t) which takes given ini-
tial values ui (t0 ), u̇i (t0 ) at t = t0 . Geometrically this means that there is a
uniquely determined geodesic through a given point and tangent to a given

vector. All the tangent vectors to the geodesic, being parallel, are of the same
length. There is thus no loss of generality in supposing them to be of unit
length, which means that t is the arc length s.
Proposition 14. The geodesics are the curves of zero geodesic curvature.
Proof. By (INSERT REFERENCE HERE), the geodesic curvature of a curve

is given by
c = T0 Ni = (T, T0 , ν) ,
DT dT
where Ni = ν × T. Since ds and ds differ by a multiple of ν this can be
written
DT
c= Ni .
ds
From T2 = 1 we have
DT dT
T= T=0.
ds ds
DT
Being a tangent vector, ds is also orthogonal to ν. Hence
DT
= cNi . (4.6)
ds
The vanishing of the left-hand side of (4.6) defines the geodesics, and the
Proposition follows.
Corollary 4. A curve C on S with curvature k 6= 0 is a geodesic if and only

if it has the surface normal as the principal normal at every point.
Proof. By the same Exercise referred to above we have
c = k sin Θ , (4.7)
where Θ is the angle between the principal normal of C and the surface
normal. The corollary is an immediate consequence of (4.7).
Exercise 76. On the sphere
x(ul , u2 ) = (a sin u1 cos u2 , a sin u1 sin u2 , a cos u1 )
consider the latitude circle u1 = α = const. Show that by displacing a vector

parallelly once along the circle, the vector turns an angle 2π(1 − cos α).
4.3 Integrability Conditions 63
Exercise 77. Two directions defined by the vectors
v = ξ i xi , w = η j xj
are called conjugate if
hij ξ i η j = 0 .
Consider the tangent planes to a surface S along a curve C on S. Show that

the limit of the intersection of two neighboring tangent planes is a line in the
conjugate direction of the tangent direction of C.
Exercise 78. In the xy-plane consider the conic
ax2 + 2bxy + cy 2 = 1 , ac − b2 6= 0
with center at 0 = (0, 0). Two lines through 0 with the slopes m, m0 are
called conjugate if
a + b(m + m0 ) + cmm0 = 0 .
Show that the mid-points of a family of parallel chords lie on a line through
0, which is conjugate to the chord of the family through 0.
Exercise 79. Show that the geodesic curvature of the latitude circle in (IN-
SERT REFERENCE HERE) is cot αa . Hence the latitude circle is a geodesic
if and only if it is an equator.
4.3 Integrability Conditions

By successive applications of the Weingarten equations (??) and the Gauss
equations (??) it is possible to express all partial derivatives of x and ν as
linear combinations of x1 , x2 , ν. The integrability conditions express the
fact that they are independent of the order of differentiation. The resulting
relations are among the most fundamental in surface theory.
By differentiating (??) and using (??) we find
!
∂hji
νik = − + hi Γlk xj − hji hjk ν .
l j
∂uk
Since the left-hand side is symmetric in i, k, the same is true of the right-hand
side, and of the coefficients of xj , ν at the right-hand side. The coefficient of
ν is
hji hjk = g jl hij hlk ,

which is symmetric in i, k. Expressing that the coefficient of xj is symmetric

in i, k, we get
∂hji ∂hjk
− + hli Γjlk − hlk Γjli = 0 . (4.8)
∂uk ∂ui
These are called the Codazzi equations.
These equations have an equivalent form expressed in terms of hij . For
we have, using (??),
∂hij ∂(hpi gpj )

k
=
∂u ∂uk
∂hpi
= g + hpi (Γpjk + Γjpk )
k pj
∂u
p
∂hi q p p
= + h i qk gpj + hi Γjpk
Γ
∂uk
which can be written

∂hpi

∂hij
− hip Γpjk = + h q p
i qk gpj .
Γ
∂uk ∂uk
Hence we have
∂hij ∂hkj
− − hil Γljk + hkl Γlji = 0 , (4.9)
∂uk ∂ui
which are equivalent to (4.8).
By differentiating (??) and using (??) and (??), we have
l
∂Γik p l l ∂hik l
xipj = + Γik Γpj − hik hj xl + + Γik hlj ν .
∂uj ∂uj
We introduce the Riemann-Christoffel tensor
∂Γlik ∂Γlij
l
−Rikj l
= +Rijk = − + Γpik Γlpj − Γpij Γlpk . (4.10)
∂uj ∂uk
The symmetry in k, j of the coefficient of ν in the above equation for xikj
gives the Codazzi equation (4.9), while the same symmetry of the coefficient
of xl gives
l
−Rikj − hik hlj − hij hlk = 0 . (4.11)
These are called the Gauss equations.

Let
p
Rilkj = Rkj gpl ,
4.3 Integrability Conditions 65
or, equivalently,
p
Rikj = Rilkj g pl .
Then the Gauss equations (4.11) can be written
Rilkj + hik hjl − hij hkl = 0 .(CHECK THIS) (4.12)
From (4.12) we derive the following symmetry properties of Rilkj ;
Rilkj = −Rlikj = −Riljk ,

Rilkj = Rkjil .
In particular, (4.12) gives
R1212 + h11 h22 − h212 = 0 .(CHECK THIS)
By (INSERT REFERENCE HERE) we have
R1212
K= ,
g
where the right-hand side depends only on the first fundamental form. This
fact is called:
Theorem 9. The Remarkable Theorem of Gauss. The Gaussian curvature
of a surface depends only on the first fundamental form.
Exercise 80. Show that in the notation of the Exercises in (INSERT REF-
ERENCE HERE) the Codazzi equations can be written
∂L ∂M
− = LΓ112 + M (Γ212 − Γ111 ) − N Γ211 ,
∂v ∂u
∂M ∂N
− = LΓ222 + M (Γ222 − Γ112 ) − N Γ212 .
∂v ∂u
Exercise 81. With the same notation show that the Gauss equation can be
written
EGK = − 12 (Guu + Evv ) + 14 EuEGu + EvGGv ,

provided F = 0.
There are therefore two Codazzi equations and one Gauss equation be-
tween the six functions which enter into the first and second fundamental
forms. This is in agreement with the fact that a parametric surface depends
on three arbitrary functions in two variables: x(u, v).
4.4 The Congruence Theorem

The congruence theorem addresses to the question as to when two parametric
surfaces are congruent. More precisely, let ∆ be a connected domain in the
(u1 , u2 )-plane and let
INSERT GRAPHIC HERE
be two parametric surfaces, E being the euclidean space. We wish to know
the conditions that there exists a motion T so that the above diagram is
commutative.
Let I, II (resp I ∗ , II ∗ ) be the.first and second fundamental forms of the
surface x (resp. x∗ ). If x and x∗ are congruent, we have I ∗ = I. The form II
depends on the orientation of the space E, because it depends on the sign of
the unit normal vector ν(u1 , u2 ). As usual suppose E be oriented. Then the
sign of ν is defined by the condition
(x1 , x2 , ν) > 0 .
Under an improper motion this determinant changes sign. As a result II
changes sign under an improper motion.
Theorem 10. Let
x, x∗ : ∆ → E
be two surfaces whose fundamental forms are to be denoted by I, II and I ∗ ,
II ∗ respectively. If
I ∗ = I , ±II ∗ = II , (4.13)
then x∗ differs from x by a motion.
In terms of the coefficients of the fundamental forms conditions (4.13) can
be written
∗
gij = gij , h∗ij = ±hij , i, j = 1, 2 .
To prove the theorem we consider frames xx1 x2 ν(u1 , u2 ), which satisfy
the conditions
xi xk = gik , xi ν = 0 , ν 2 = 1 , i, k = 1, 2 , (4.14)
and the partial differential equations
∂x
= xi ,
∂ui
∂xk
= xki = Γjki xj + hki ν , i, j, k = 1, 2 ,
∂ui
∂ν
= −hji xj .
∂ui
4.4 The Congruence Theorem 67
The first equation is the definition of xi , the second is identical to the Gauss
equations, and the third to the Weingarten equations. Their coefficients are
completely determined by I and II.
We reduce the congruence theorem to a theorem on a one-parameter
family of frames, whose special case on rectangular frames was proved earlier
(INSERT REFERENCE HERE):
Proposition 15. Let xE1 E2 E3 (t) and x∗ E1∗ E2∗ E3∗ (t) be two smooth families
of frames in the parameter t which satisfy the conditions
Eα Eβ = Eα∗ Eβ∗ = gαβ , α, β = 1, 2, 3, det(gαβ ) 6= 0 , (4.15)
and the differential equations
ẋ = pα Eα , Ėα = qαβ Eβ , (4.16)
ẋ∗ = pα Eα∗ , Ėα∗ = qαβ Eβ∗ , (4.17)
with the functions pα , qαβ , with dot denoting differentiation with respect to t.
If the frames are identical at one value t0 :
x∗ (t0 ) = x(t0 ) , Eα∗ (t0 ) = Eα (t0 ) , (4.18)
they are identical for all values t:
x∗ (t) = x(t) , Eα∗ (t0 ) = Eα (t) . (4.19)
Proof. This theorem is a special case of a uniqueness theorem on the solution

of a system of differential equations of the first order. The following direct
proof involves techniques which are useful in differential geometry.
We will write the equations in matrix notation. Let
Eα = (e1α , e2α , e3α )
and introduce the matrix

e11 e21 e31
 
M = e12 e22 e32  ,

e13 e23 e33
whose rows are therefore the components of Eα . Similary we define M ∗ . Then

the second equations in (4.16) and (4.17) can be written
Ṁ = QM , Ṁ ∗ = QM ∗ , (4.20)
where
q11 q12 q13

 
Q = q21 q22 q23  .

q31 q32 q33
On the other hand, putting

 1 2 3
g1 g1 g1
G = g21 g22 g23  ,
g31 g32 g33
equation (4.15) can be written
M tM = G . (4.21)
G being non-singular, we have, on taking the inverse of the above equation,
G−1 = t M −1 M −1 ,
or
t
M G−1 M = I = identity matrix . (4.22)
Differentiating (4.21), we get, on using (4.20),
Ġ = Ṁ t M + M t Ṁ = QG + Gt Q .
Since GG−1 = I, we have
ĠG−1 + GĠ−1 = 0 ,
so that
Ġ−1 = −G−1 ĠG−1 = −G−1 Q − t QG−1 . (4.23)
By using (4.20) and (4.23), we get
d(t M ∗ G−1 M ) t ∗t
= M QG−1 M + t M ∗ (−G−1 Q − t QG−1 )M + t M ∗ G−1 QM = 0 .
dt
Hence the matrix in the parentheses at the left-hand side is constant; it is
equal to the matrix at t0 . which, by (4.18), is equal to t M G−1 M . The latter
is I by (4.22). It follows that
t
M ∗ G−1 M = I .
Comparing this equation with (4.22), we get

t
M ∗ = t M , orM ∗ = M .
This gives the second equation of (4.19).

4.4 The Congruence Theorem 69
From the first equations of (4.16) and (4.17) we get
d(x∗ − x)
=0.
dt
Hence the vector in the parentheses is constant and is, by (4.18), zero.
Proof of congruence theorem. By an improper motion if necessary we can
suppose II ∗ = II. Let 0 be a fixed point of ∆. The frames xx1 x2 ν and
x∗ x∗1 x∗2 ν ∗ at 0, satisfying the conditions (4.14), differ by a motion, proper or
improper. By applying this motion, we can suppose them to be identical. It
remains to prove that xx1 x2 ν and x∗ x∗1 x∗2 ν ∗ are identical at any other point
P of ∆.
We join 0 to P by an arc, whose equations we suppose to be
ui = ui (t) , i = 1, 2 .
Along the arc, xx1 x2 ν satisfies the differential system
ẋ = u̇i xi ,
ẋk = Γjki u̇i xj + hki u̇i ν ,
ν̇ = −hji u̇i xj ,
and x∗ x∗1 x∗2 ν ∗ satisfies the same differential system. Since they are identical
at 0, it follows from the above Proposition that they are identical at P . This
proves the congruence theorem.
5 Global Surface Theory
We shall treat some selected problems in the theory of surfaces in the large,
i.e., results which are valid only when the whole surface is taken into ac-
count. For such global problems it will be most convenient to use the method
of “moving frames,” which will be developed in the first section. This ap-
proach is consistent with our study of curves, where the Frenet formulas
were the culmination. In surface theory it is necessary to consider families of
frames depending on two variables. The effectiveness of the moving frames
can be theoretically justified by the observation that euclidean geometry is
dominated by the group of its motions and the space of all frames can be
identified in an obvious way with the space of all motions.
5.1 Moving Frames

5.1.1 INSERT TITLE
We will consider differential forms in two variables, such as
ω = P (u, v)du + Q(u, v)dv ,
where the coefficients P , Q are smooth functions. When
ω 0 = P 0 du + Q0 dv
is another form, we introduce a multiplication, called the exterior multipli-

cation,which is anti-symmetric in the differentials, thus:
ω ∧ ω 0 = −ω 0 ∧ ω = (P Q0 − QP 0 )du ∧ dv .
It follows that ω and ω 0 are linearly dependent if and only if
ω ∧ ω0 = 0 .
The exterior derivative of ω is defined to be the form of degree two:
dω = (Qu − Pv )du ∧ dv .
72 5 Global Surface Theory
When it is applied to the differential df of a smooth function f , we have

clearly
d(df ) = 0 . (5.1)
Also we have
d(f ω) = df ∧ ω + f dω .
The exterior derivative allows the Green-Stokes formula to be written in

the form
ZZ Z
dω = ω,
D ∂D
where D is a domain in the (u, v)-plane and ∂D is its boundary. The same
formula is true for a domain on a surface. This is proved by the “additivity”
of the formula: Cut D into subdomains so small that each lies in a coordi-
nate neighborhood; apply the formula to each subdomain; and then add the
results.
5.1.2 INSERT TITLE
As in Chapter 1 by a frame we mean the vectors xe1 e2 e3 , satisfying the

relations
eα · eβ = δαβ , α, β = 1, 2, 3 , (5.2)
(e1 , e2 , e3 ) = 1 . (5.3)
Condition (5.3) means that we restrict our frames to be right-handed. The

point with the position vector x is called the origin of the frame.
The local theory of surfaces can be developed as follows: Let, a surface S
be described by x(u, v). We suppose S to be oriented and e3 (u, v) to be the
unit normal vector at (u, v). Choose a smooth field of unit tangent vectors
e1 (u, v), and let e2 = e3 × e1 . Since eα are linearly independent, we can write
X
dx = ωi ei , i = 1, 2 , (5.4)
i
X
deα = ωαβ eβ , α, β = 1, 2, 3 . (5.5)
β
The sum in (5.4) does not involve e3 , because dx is a tangent vector.

From (5.4) and (5.5) we find
ωi = dx · ei , ωαβ = deα · eβ . (5.6)
By differentiating (5.2) and using the second equation of (5.6), we get
ωαβ + ωβα = 0 .
Thus the matrix (ωαβ ) is anti-symmetric; there are only three essential en-
tries. observe that the equation de3 in (5.5) is the same as the Weingarten
formulas and the equations for de1 , de2 are equivalent to the Gauss equations.
The first fundamental form is
I = ds2 = dx · dx = ω12 + ω22 ,
and the element of area is
dA = ω1 ∧ ω2 6= 0 .
Thelatter implies that ω1 and ω2 are linearly independent. As a consequence

ωαβ are their linear combinations. These coefficients, or their combinations,
are to have geometrical meaning.
The equation (5.1) is true for each of the components of the vectors x, eα
so that we have
d(dx) = d(deα ) = 0 . (5.7)
These equations contain all the important information on local surface theory
and are to be studied in detail.
First we have
X X
d(dx) = dωi ei − ωi ∧ ωαβ eβ , i = 1, 2 ; β = 1, 2, 3 .
i i,β
Its vanishing implies the vanishing of the coefficients of eβ , which gives

X
dωi = ωj ∧ ωji , i, j = 1, 2 , (5.8)
j
ω1 ∧ ω13 + ω2 ∧ ω23 = 0 . (5.9)
If we write
ω13 = aω1 + bω2 , ω23 = b0 ω1 + cω2 ,
equation (5.9) gives b0 = b.

The second fundamental form is
II = −dx · de3 = ω1 ω13 + ω2 ω23 = aω12 + 2bω1 ω2 + cω22 .

The two principal curvatures k1 and k2 are the roots of the equation
(a − k)(c − k) − b2 = 0 .
Their two elementary symmetric functions are respectively the mean curva-
ture and the gaussian curvature and are given by
1 1
H= (k1 + k2 ) = (a + c) , K = k1 k2 = ac − b2 .
2 2
The equations (5.8) completely determine ω12 . For, if
X
0
dωi = ωj ∧ ωji , i, j = 1, 2 ,
j
we get by subtraction
X
0
ωj ∧ (ωji − ωji ) = 0 ,
j
or explicitly,
0 0
ω1 ∧ (ω12 − ω12 ) = ω2 ∧ (ω12 − ω12 = 0 ,
0
which implies ω12 = ω12 .
Next we have
!
X X
d(deα ) = dωαβ − ωαγ ∧ ωγβ eβ , α, β, γ = 1, 2, 3 .
β γ
The vanishing of this expression implies that the coefficient of eβ at the

right-hand side is zero, i.e.,
X
dωαβ = ωαγ ∧ ωγβ , α, β, γ = 1, 2, 3 . (5.10)
γ
Because of the anti-symmetry of ωαβ this equation is identically satisfied

if α = β and remains the same equation when α and β are interchanged.
Hence (5.10) involves only three independent equations. Written explicitly,
the equation
dω12 = −ω13 ∧ ω23 = −Kω1 ∧ ω2 (5.11)
is commonly known as the Gauss equation and the equations
dω13 = ω12 ∧ ω23 , dω23 = ω21 ∧ ω13 (5.12)
are the Codazzi equations.

The absolute differential of a vector field is by definition the orthogonal

projection of the ordinary differential into the tangent plane. Hence from (5.5)
we have
X
Dei = ωij ej , i, j = 1, 2 .
j
It follows that
ω12 = e2 · De1 = −e1 · De2 ,
which says that ω12 depends only on the first fundamental form I (and on
the vector field e1 ). Since ω1 ∧ω2 is the element of area and depends also only
on I, equation (5.11) contains the Gauss remarkable theorem: The Gaussian
curvature K depends only on the first fundamental form.
The above discussion contains everything that is needed to know on the
formal aspect of local surface theory. Since it depends on the choice of a
vector field, we will have a complete control if we know the behavior of the
differential forms under a change of the vector field. Such a change is given
by
e∗1 = cos Θe1 + sin Θe2 ,

e = sin Θe1 + cos Θe2 ,
where Θ(u, v) is a function of u, v. Then
ω1∗ = dxe∗1 = cos Θω1 + sin Θω2 ,

ω2∗ = dxe∗2 = − sin Θω1 + cos Θω2 .
It follows that
dω1∗ = (dΘ + ω12 ) ∧ ω2∗ , dω2∗ = −(dΘ + ω12 ) ∧ ω1∗ .

∗
But ω12 is completely determined by the equations (5.8) when the ω’s are
provided with asterisks. Comparing with the above equations, we have
∗
ω12 = dΘ + ω12 .
Remark 8. When applied to the six-dimensional space of all frames in Eu-

clidean space, the equations (5.7) lead to the structure equations of the group
of motions in the sense of the theory of Lie groups. Herein lies the basic reason
for the importance of the Gauss and Codazzi equations.
5.1.3 INSERT TITLE
In this section we will relate the moving frames with our earlier treatment.
Suppose the parametric curves be orthogonal, so that
ds2 = Edu2 + Gdv 2 .
We choose
xu xv
e1 = √ , e2 = √ ,
E G
which implies
1 1
ω1 = E 2 du , ω2 = G 2 dv . (5.13)
Then
1 1
dω1 = √ Ev dv ∧ du = √ Ev ω2 ∧ du ,
2 E 2 EG
1 1
dω2 = √ Gu du ∧ dv = √ Gu ω1 ∧ dv .
2 G 2 EG
From the fact that ω12 is completely determined by (5.8) we find
1
ω12 = √ (−Ev du + Gu dv) . (5.14)
2 EG
Therefore

1 G E
dω12 = √ u + √ v du ∧ dv ,
2 EG u EG v
and we get the classical Gauss equation

1 Gu Ev
K=− √ √ + √ .
2 EG EG u EG v
To compare the Codazzi equations observe that the second fundamental
form is
II = Ldu2 + 2M dudv + N dv 2 = aω12 + 2bω1 ω2 + cω22 .
Comparing with (5.13), we have

√
aE = L , b EG = M , cG = N .
Hence
1 1
ω13 = aω1 + bω2 = √ (Ldu + M dv) , ω23 = bω1 + cω2 = √ (M du + N dv) .
E G
Substituting these and (5.14) into (5.12), we get
2EG(Lv − Mu ) − (EN + GL)Ev − M (EGu − GEu ) = 0 ,

2EG(Mv − Nu ) + (EN + GL)Gu − M (EGv − GEv ) = 0 ,
which are the classical Codazzi equations.

5.2 The Gauss-Bonnet Theorem 77
5.2 The Gauss-Bonnet Theorem

The Gauss-Bonnet theorem has its analytical basis in the Gauss equa-
tion (5.11). It is nothing more than a geometrical interpretation of that
equation. But it has profound consequences and is perhaps one of the most
important theorems in mathematics.
5.2.1 INSERT TITLE
Let our surface be described by the position vector x(u, v) as a function of

the parameters u, v. As in (INSERT REFERENCE HERE) we let e3 (u, v)
be the unit normal vector. We choose e1 to be the unit vector tangent to the
u-curve, i.e., e1 = √xuE ; then e2 = e3 × e1 . This defines a field of frames and
the formulas in (INSERT REFERENCE HERE) are valid.
Consider an oriented curve C on the surface S. Let s be its arc length
and T = dxds be its unit tangent vector. We can write
T = cos Θe1 + sin Θe2 , (5.15)
so that Θ(s) is the angle that T makes with e1 . Then
DT = (dΘ + ω12 )N ,
where
N = − sin Θe1 + cos Θe2 (5.16)
is the unit normal vector to C. The geodesic curvature kg of C is defined by

the equation
dΘ + ω12 = kg ds .
Consider now a domain D whose boundary is a sectionally smooth simple

closed curve C. Then we have
Z Z Z
kg ds = dΘ + ω12 .
C C C
By the theorem of turning tangents, we have

Z X
dΘ = 2π − (π − αi ) ,
C i
where π − αi are the exterior angles at the corners of C. On the other hand,
by Stokes theorem and (5.11), we have
Z ZZ
ω12 = − KdA .
C D
We have thus proved a special case of the Gauss-Bonnet theorem.

Theorem 11. Let D be a compact oriented domain on S bounded by a sec-

tionally smooth curve C. Then
Z ZZ X
kg ds + KdA + (π − αi ) = 2πχ , (5.17)
C D i
where kg is the geodesic curvature of C, π − αi are the exterior angles at the

corners of C, and χ is the Euler characteristic of D.
Proof. We have proved the theorem for the case that D lies in a neighborhood
where the same set of parameters u, v is valid, with χ = 1. In the general
case suppose D be subdivided into a union of polygons Dλ , λ = 1, ..., f , such
that: 1) each Dλ lies in one coordinate neighborhood; 2) two Dλ ’s have either
no point, or one vertex, or a whole side in common. Such a subdivision is
clearly possible. We orient Dλ coherently with D and apply (5.17) to each
Dλ , for which the theorem has been proved. Adding the resulting equations,
we have, because the integrals of geodesic curvature along the interior sides
cancel,
Z ZZ XX
kg ds + KdA + (π − αλ,i ) = 2πf . (5.18)
C D λ i
We will evaluate the double sum in the formula, by summing it with

respect to the vertices of the subdivision. Let vI , vE be respectively the
numbers of interior and exterior vertices and let s1 (respectively s2 be the
number of sides which join two interior vertices (respectively one interior
vertex
P P and one exterior vertex). Then at the interior vertices
P P the contribution
of λ i π is 2πs1 + πs2 while the contribution
P P of − λ is −πvI . At
i αλ,i P
the exterior vertices the contribution of λ i (π − αλ,i ) is πs2 + i (π − αi ),
where the last sum is the sum of the exterior angles of D. Since C has the
same number of sides as vertices, the total number of sides of the subdivision
is s = s1 + s2 + vE . Let v = vI + vE be the total number of vertices. The
number
χ = v − s + f = vI − s1 − s2 + f
is called the Euler characteristic of the subdivision. Substituting the above

values into (5.18), we have proved the Gauss-Bonnet theorem.
It follows from (5.17) that the Euler characteristic, which is defined in
terms of a subdivision, depends only on D.
Remark 9. The theorem is true on an abstract riemannian manifold of two
dimensions, and essentially the same proof applies.
Special cases: 1) S is a surface of constant curvature K and D is bounded
by a triangle formed by three geodesic arcs. Then (5.17) becomes
KA = α1 + α2 + α3 − π , (5.19)
5.2 The Gauss-Bonnet Theorem 79
where A is the area of D and αi , i = 1, 2, 3, are the angles of the triangle. It

follows that α1 + α2 + α3 is greater than, equal to, or less than π, according
as K > 0, = 0, or < 0. Moreover, in the case K 6= 0, A is proportional to the
expression at the right-hand side of (5.19).
2) S is a compact orientable surface without boundary. Then (5.17) gives
ZZ
KdA = 2πχ .
D
It follows that if K = 0, the Euler characteristic of S is zero and S is home-

omorphic to a torus. Also if K > 0, then χ > 0 and S is homeomorphic to
the sphere.
5.2.2 Theorems of Jacobi and Hadamard
Consider a closed curve x(s) in space, referred to the arc length s as param-
eter. We suppose that the curvature k(s) never vanishes, and there is no loss
of generality in supposing k(s) > 0. Then the unit principal normal vector
00
N(s) = x k(s) and the unit binormal vector b(s) = x00 (s) × N(s) are both well
defined over the whole curve. Their end-points describe on the unit sphere the
principal normal indicatrix and the binormal indicatrix respectively. Jacobi’s
theorem is the following:
Theorem 12. Let x0 (s) be a closed space curve whose curvature is nowhere
zero. If its principal normal indicatrix does not intersect itself, it divides the
unit sphere in two domains with the same area.
Proof. To prove the theorem we define τ by the equations

1 1
k = (k 2 + w2 ) 2 cos τ , w = (k 2 + w2 ) 2 sin τ ,
where w is the torsion of the curve. Then we have

1
d(− cos τ T + sin τ b) = (sin τ T + cos τ b)dτ − (k 2 + w2 ) 2 Nds ,
where T = x0 is the unit tangent vector. By Frenet’s formula the vector

between parentheses at the left-hand side is the unit tangent vector to the
principal normal indicatrix. It follows that, if σ denotes the arc length of
dτ
N(s), dσ is the geodesic curvature of N(s) on the unit sphere. Let D be
one of the domains bounded by N(s), and A its area. By the Gauss-Bonnet
formula we have, since K = 1,
Z ZZ
dτ + dA = 2π .
N(s) D
The first integral at the left-hand side is zero because it is integrated over a
closed curve. Hence we have A = 2π and the theorem is proved.
Another application of the Gauss-Bonnet theorem gives Hadamard’s the-

orem on convex surfaces:
Theorem 13. If the Gaussian curvature of a closed orientable surface in
space is everywhere positive, the surface is convex. (That is, it lies at one
side of every tangent plane.)
Observe that for curves in the plane a similar theorem is true only un-
der the additional assumption that the curve is simple. (INSERT REFER-
ENCE HERE) For a surface it is not necessary to suppose that it has no
self-intersection.
Let S denote the surface. It follows from the Gauss-Bonnet formula that
its Euler characteristic χ(S) is positive, so that χ(S) = 2 and we have
ZZ
KdA = 4π .
S
Suppose S be oriented. If SO is the unit sphere about a fixed point O, the

Gauss mapping
g : S → SO
is defined by assigning to every point p ∈ S the end-point of the unit vector

through O parallel to the unit normal vector to S at p. The condition K > 0
implies that g has everywhere a nonzero functional determinant and is locally
one-to-one. It follows that g(S) is an open subset of SO . Since S is compact,
g(S) is a compact subset of SO , and hence is also closed. Therefore g maps
onto SO .
Suppose that g is not one-to-one, that is, there exist points p and q of
S, p 6= q, such that g(p) =RRg(q). Then there is a neighborhood U of q such
that g(S − U ) = SO . Since S−U KdA is the area of g(S − U ), counted with
multiplicities, we will have
ZZ
KdA ≥ 4π .
S−U
But
ZZ
KdA > 0 ,
U
so that
ZZ ZZ ZZ
KdA = KdA + KdA > 4π ,
S U S−U
which is a contradiction. Hence g is one-to-one and we see easily that S is

convex.
5.3 Rigidity Theorems 81
5.2.3 Vector fields on a surface

The tools developed above also provide a convenient means to study the
behavior at the singularities of a vector field on the surface. We use the
notations of (INSERT REFERENCE), where Θ(u, v) is now a function of u,
v, defines a vector field on S; then equation (INSERT REFERENCE) defines
the vector field perpendicular to it. We suppose that T has a singularity at
the point p0 with the parameters u = v = 0. Let U be a neighborhood of p0 .
In U − p0 we have
DT · N = dΘ + ω12 .
Let γ be a circle of radius about p0 . The limit
Z
1
lim dΘ + ω12
2π →0 γ
is an integer, called the index of the vector field at p0 . Clearly the index is
zero if p0 is no singularity.
We will prove the theorem:
Theorem 14. Let S be a closed orientable surface and ξ be a smooth vector
field with a finite number of singularities. Then the sum of the indices at the
singularities is equal to the Euler characteristic χ(S).
It follows that no continuous vector field ξ exists on a closed orientable
surface other than the torus, such that ξ 6= 0. For if such a ξ exists, then
ξ
1 defines a unit vector field without singularity on S. By our theorem
(ξ·ξ) 2
this is only possible when χ(s) = O.
Proof. To prove the theorem let pi , 1 ≤ i ≤ n, be the singularities of ξ. Let
γi () be a circle of radius about pi and ∆i () the disk bounded by γi ().
By (INSERT REFERENCE) we have
d(dΘ + ω12 ) = −KdA .
S
Applying Stokes Theorem to the domain S − i ∆i (), we get
ZZ XZ
S
S−
KdA =
∆i ()
dΘ + ω12 ,
γi ()
i i
where γi () is so oriented that it is the boundary of ∆i (). The theorem

follows by letting → 0.
5.3 Rigidity Theorems

We shall establish some global rigidity (or uniqueness) theorems on closed
surfaces. They characterize surfaces by the validity of certain local properties
throughout. The two most elementary methods are maximum principle and
integral formulas. We shall show how they are applied.
5.3.1 Elliptic W-surfaces
For definiteness and for historical reason we consider the problem of deter-
mining the closed surfaces of constant Gaussian curvature K. We shall begin
by some general analytical considerations.
We take a frame field over a neighborhood of S and use the notations and
results of (INSERT REFERENCE). Let
yα = x · eα , α = 1, 2, 3 .
Geometrically y3 is the oriented distance from the origin to the tangent plane
at (u, v). By (INSERT REFERENCE) and (INSERT REFERENCE) we have
dy1 = ω1 + ω12 y2 + ω13 y3 , (5.20)

dy2 = ω2 − ω12 y1 + ω23 y3 , (5.21)
dy3 = −ω13 y1 − ω23 y2 . (5.22)
The following lemma results from the elementary study of the behavior
of a geometrically defined function at a local maximum:
Lemma 3. Let pO ∈ S be a point where the function w = x2 attains a local
maximum. Then the Gaussian curvature of S at pO is positive.
The function w on S is the square of distance from O to a point of S. We
have
1
dw = x · dx = y1 ω1 + y2 ω2 .
2
If the subscript O denotes the value of a function at pO , we have
(y1 )O = (y2 )O = O .
By (5.20) the second differential of w at pO is given by

1 2
(d w)O = IO + (y3 )O IIO ,
2
which is negative semi-definite. Using the expressions (INSERT REF) and
(INSERT REF), this implies
1 + (y3 )O aO ≤ O , 1 + (y3 )O cO ≤ O ,
(1 + (y3 )O aO )(1 + (y3 )O cO ) − (y3 )2O b2O ≥ O .
From these we derive KO = aO cO − b2O > O by a simple manipulation.

It follows that a closed surface S must have a point with positive Gaussian
curvature, i.e., the point at a maximum distance from a fixed point in space.
Hence if its Gaussian curvature is constant, the constant is positive and, by
Hadamard’s theorem, the surface is convex.
The answer to our problem is given by Liebmann’s theorem:
Theorem 15. A closed surface of constant Gaussian curvature K is a sphere

of radius √1K .
The proof that we will give was due to Hilbert. It relies on a lemma
which is similar to the above but is more sophisticated. Hilbert’s lemma is
the following:
Lemma 4. Let pO ∈ S be a non-umbilical point and let k1 (p) > k2 (p) be

the two principal curvatures, where p is in a neighborhood of pO . Suppose
k1 has a local maximum and k2 a local minimum at pO . Then the Gaussian
curvature of S at pO is ≤ 0.
Proof. By hypothesis the principal directions are well defined in a neighbor-

hood of pO . We choose our frame field such that e1 , e2 are in these directions.
Observe that the principal directions are given by the equation

ω1 aω1 + bω2
ω2 bω1 + cω2 = 0

or
(a − c)ω1 ω2 − b(ω12 − ω22 ) = 0 .
Our choice implies b = 0. Then a and c are clearly the normal curvatures in
the principal directions and we can write
ω13 = k1 ω1 , ω23 = k2 ω2 .
Taking the exterior derivatives of these equations and using (INSERT REF)
and (INSERT REF) (Codazzi equations), we get
dk1 ∧ ω1 + (k1 − k2 )ω12 ∧ ω2 = 0 , (5.23)

dk2 ∧ ω2 + (k1 − k2 )ω12 ∧ ω1 = 0 . (5.24)
We will use ω1 , ω2 (instead of du, dv) to express the differentials of functions

and we put
X
dki = kij ωj ,
j
X
dkij = kijk ωk , i, j, k = 1, 2 .
k
Substituting into (5.23), we get
(k1 − k2 )ω12 = k12 ω1 + k21 ω2 . (5.25)
By the extremal properties of k1 , k2 at pO we have, at pO ,

(kij )O = 0 .
The second differential of ki at pO is given by

X
(d2 ki )O = (kijk )O ωj ωk ,
j,k
which is a negative (respectively positive) semi-definite form when i = l

(respectively i = 2). Hence
(k1jj )O ≤ 0 , (k2jj )O ≥ 0 , j = 1, 2 .
Taking the exterior derivative of (5.25) and using the Gauss equation
(INSERT REF), we get
−(k1 − k2 )Kω1 ∧ ω2 = (k211 − k211 )ω1 ∧ ω2 + ... ,
where the dots denote terms which vanish at pO . Equating the coefficients of
ω1 ∧ ω2 at both sides and taking their values at pO , we get

k122 − k211
KO = ≤0.
k1 − k2 O
This proves the lemma.
From the lemma follows a theorem more general than Liebmann’s. A

surface S is called a Weingarten surface or a W-surface if there is a functional
relation between the principal curvatures:
W (k1 , k2 ) = 0 .
If, moreover, the relation

∂W ∂W
>0 (5.26)
∂k1 ∂k2
holds, S is called an elliptic W-surface. Examples of elliptic W-surfaces in-
clude the surfaces of constant mean curvature and surfaces of constant posi-
tive Gaussian curvature. The condition (5.26) means that k2 is a monotone
decreasing function of k1 .
We will prove the theorem:
Theorem 16. A closed convex (i.e. K > 0) elliptic W-surface is a sphere.
We choose e3 to be the unit inward normal vector, so that the second

fundamental form is positive definite everywhere. Let k1 (p) ≥ k2 (p) > 0,
p ∈ S, be the two principal curvatures. Then k1 (p) attains a maximum at
a point pO , say. Since k2 is a decreasing function of k1 , pO is at the same
time a minimum of k2 . If pO is non-umbilical, Hilbert’s lemma asserts that
the Gaussian curvature K(pO ) ≤ 0. This contradicts our hypothesis K > 0.

Hence pO is an umbilic, i.e., k1 (pO ) = k2 (pO ). But, we have, for any point
p ∈ S,
k1 (pO ) ≥ k1 (p) ≥ k2 (p) ≥ k2 (pO ) .
The equality of the two extreme sides implies that p is an umbilic. Therefore
S consists entirely of umbilics and is a sphere.
Clearly our theorem implies Liebmann’s theorem. We will state also the
following corollary:
Corollary 5. A closed convex surface of constant mean curvature is a sphere.
5.3.2 Integral formulas; second proof of Liebmann’s theorem.
By (5.20) and the equations (INSERT REF), (INSERT REF) we get
d(y2 ω1 − y1 ω2 ) = −2(1 + y3 H)ω1 ∧ ω2 ,

d(y2 ω13 − y1 ω23 ) = −2(H + y3 K)ω1 ∧ ω2 .
The expressions under d at the left-hand sides are independent of the choice
of the frame field, for we have
(x, e3 , dx) = y2 ω1 − y1 ω2 ,
−(x, e3 , de3 ) = y2 ω13 − y1 ω23 .
These are therefore linear differential forms defined over the whole surface S.
Applying Stokes theorem, we have the integral formulas:
Let S be a closed orientable surface without boundary. Then
ZZ
(1 + y3 H)dA = 0 , (5.27)
ZZ S
(H + y3 K)dA = 0 . (5.28)
S
Let A be the total area of S and let

ZZ
M= HdA
(the integral of mean curvature). Then (5.27) can be written

ZZ
A=− y3 HdA , (5.29)
ZZ
M =− y3 KdA . (5.30)
To illustrate the way that integral formulas can be used to derive global
geometrical theorems we shall use (5.29) to give a second proof of Liebmann’s
theorem. In fact, let
K = κ2 > 0 , κ > 0 .
By the Gauss-Bonnet theorem we have

ZZ
KdA = κ 2 A = 4π ,
from which follows that

ZZ
(k1 − κ)(k2 − κ)dA = 8π − 2κM .
On the other hand, using (5.29), we get

ZZ ZZ ZZ
8π
(k1 − κ)(k2 − κ)y3 dA = 2 y3 KdA − 2κ y3 HdA = −2M + 2κA = −2M + .
κ
If the integrals at the left-hand sides of the last two equations are denoted
by J1 , J2 respectively, we have
J1 = κJ2 .
As above, we choose e3 to be the unit inward normal vector, so that the

second fundamental form is positive definite, i.e., k1 > 0, k2 > 0, H > 0, and
y3 = x · e3 < 0. Then we have
(k1 − κ)(k2 − κ) = 2(K − κH) = 2κ(κ − H) ≤ 0 ,
since κ is the geometric mean of k1 , k2 and H is their arithmetic mean. It

follows that
J 1 ≤ 0 , J2 ≥ 0 .
This is possible only if J1 = J2 = 0. Since the integrands keep constant signs,

this implies κ − H = 0. Hence every point is an umbilic and S is a sphere.
5.3.3 Cohn-Vossen’s rigidity theorem
Cohn-Vossen’s theorem is the following:
Theorem 17. An isometry between two closed convex surfaces is established

either by a motion or by a motion and a reflection.
In other words, such an isometry is always trivial. The theorem is clearly

not true locally. For example, the plane and the circular cylinder are isometric
but are not congruent. Also the convexity assumption cannot be removed
(See figure - INSERT FIG). The question answered by the theorem was long-
outstanding, till it was solved by Cohn-Vossen in 1927. The following proof,
based on an integral formula, was due to G. Herglotz.
We shall begin by an algebraic lemma:
Lemma 5. Let ax2 +2bxy+cy 2 and a0 x2 +2b0 xy+c0 y 2 be two positive definite
quadratic forms, with
ac − b2 = a0 c0 − b02 .
Then
0
a − a b0 − b

b − b c0 − c ≤ 0 , (5.31)
0
and the equality sign holds in (5.31) only when the two forms are identical.
To prove the lemma we observe that the statement of the lemma remains
unchanged under a linear transformation of the variables. Applying such a
linear transformation when necessary, we can assume b0 = b. Then the left-
hand side of (5.31) becomes
c 0
(a0 − a)(c0 − c) = − (a − a)2 ≤ 0 ,
a0
as to be proved. Moreover, the quantity equals zero only when we also have
a0 = a and c0 = c.
Remark 10. Geometrically the lemma means that two concentric ellipses with
the same area meet in four distinct points or are congruent.
To prove the theorem let S and S ∗ be the two surfaces which are related
by an isometry T : S → S ∗ . Suppose both be oriented by inward normals.
The isometry T either preserves or reverses the orientation. In the latter case
we replace S ∗ by its mirror image so that we can assume without loss of
generality that T preserves orientation.
We now study local properties of the isometry. Let u, v be parameters on
both surfaces such that corresponding points have the same parameters. T
maps tangent vectors of S to tangent vectors of S ∗ and preserves their scalar
products. Since it also preserves orientation, it maps a (local) frame field on S
into a frame field on S ∗ . We will use the results developed in (INSERT REF)
on the differential geometry of S relative to a frame field. This also applies
to S ∗ relative to the image frame field and we will denote the corresponding
quantities by the same notations with asterisks. By (INSERT REF) we have
ωi = dxei = (xu ei )du + (xv ei )dv , i = 1, 2 .

A similar equation holds for ωi∗ . Since T preserves scalar products of tangent
vectors, we have
ωi∗ = ωi , i = 1, 2 .
By exterior differentiation and use of (INSERT REF) we find that the con-
nection forms are equal, i.e.,
∗
ω12 = ω12 .
Exterior differentiation of this equation and use of the Gauss equation (IN-
SERT REF) give the equality K ∗ = K of Gaussian curvatures at correspond-
ing points. This is to be expected, because we know that the Gaussian curva-
ture depends only on the first fundamental form and thus remains invariant
under an isometry.
To prove the theorem it suffices to show that the second fundamental
forms are equal. By (5.20) and the Codazzi equations (INSERT REF) applied
to S ∗ , we find
∗ ∗
d(y1 ω23 − y2 ω13 ) = 2H ∗ ω1 ∧ ω2 + (ac∗ + a∗ c − 2bb∗ )y3 ω1 ∧ ω2 .
The expression under d at the left-hand side can be written
∗ ∗ ∗ ∗
y1 ω23 − y2 ω13 = −(x, e3 , ω13 e1 + ω23 e2 ) .
It remains invariant under a rotation of the frame field on S; it is therefore a
linear differential form defined everywhere on S. By applying Stokes theorem
we get the integral formula
ZZ ZZ
y3 JdA = −2 H ∗ dA ,
where we define
J = ac∗ + a∗ c − 2bb∗ .
Combining with the second equation of (5.29), we get
ZZ ∗
a − a b∗ − b

∗
b − b c∗ − c y3 dA = 2(M − M ) ,
∗
S
where M and M are the integrals of mean curvature of S and S ∗ respec-

∗
tively. The integrand at the left-hand side is ≥ 0. Hence the same is true of
the integral and we have M ∗ ≥ M . But the relation of the two surfaces is
symmetric, so that the above inequality remains true when the two sides are
interchanged. This gives M ∗ = M . Hence the integrand at the left-hand side
must be zero. By our algebraic lemma this is possible only when
a∗ = a , b ∗ = b , c ∗ = c .
This completes our proof of Cohn-Vossen’s theorem.
6 INSERT TITLE
6.1 Geodesics
6.1.1 INSERT TITLE
Consider a curve C on a Riemannian surface. In terms of the local coordinates
u, v it is defined by the equations
u = u(t) , v = v(t) . (6.1)
The vector with components du dv
dt , dt (relative to the coordinates u,v) is called
its tangent vector. From (INSERT REF) and (INSERT REF) it follows that
it is a unit vector when t is the arc length of C. When we use arc length as
the parameter we will write s for t.
We use the notation of the last section and, at every point of C, we let
e1 be the unit tangent vector. Then u, v and τ are all functions of s and we
can write ω12 = kg ds on C. The number kg clearly generalizes the curvature
of a curve in the Euclidean plane. We call it the geodesic curvature of C (at
the point with parameter s). It is to be observed that kg is a well-defined
quantity only when both M and C are oriented (the latter means that we
have made a choice of the sign of the arc length). If M is non-orientable or
unoriented, then only the absolute value of kg is defined.
We will give a geometrical interpretation of the geodesic curvature, which
illustrates even more clearly its analogy with the case of the Euclidean plane.
In fact. let v(s) be a family of unit vectors parallel along C. Denote by τ 0 (s)
∂
the angle from ∂u to v(s), while, as in section 7.1 (CHANGE THIS?), τ (s)
∂
is the angle from ∂u to the tangent vector e1 (s) of C. Then we have
dτ (s) + αθ1 + βθ2 = kg ds
dτ 0 (s) + αθ1 + βθ2 = 0 .
where in α, β, θ1 , θ2 the coordinates u, v are to be considered as functions
of s, as given by (6.1). From these two equations we get
d(τ − τ 0 ) = kg ds .
Since τ − τ 0 is the angle from v(s) to e1 (s), the geodesic curvature is the rate
of change of this angle with respect to arc length. Clearly it is independent
of the choice of the family of parallel vectors v(s) along C.
90 6 INSERT TITLE
6.1.2 INSERT TITLE
A geodesic is a curve whose geodesic curvature vanishes at every point. An

alternative way of saying this is that the family of unit tangent vectors is
parallel along the curve. This property is preserved if we reverse the orienta-
tion of the curve or the orientation of the surface, so that being a geodesic is
a property of an unoriented curve on an unoriented Riemannian surface.
We next derive the “differential equation” of geodesics. For simplicity we
consider only the geodesics which can be defined by an equation of the form
v = v(u). This restriction is not essential, for, by interchanging u and v,
we can get hold of the geodesics which escape our consideration. Where a
geodesic is defined by such an equation, its tangent vector has components
(l, v 0 ), where v 0 = du
dv
. We assume now that the curves u = const., v = const.
form an orthogonal net, and use the equations of (INSERT REF). We observe
that θ1 , θ2 has as dual basis
1 ∂ 1 ∂
a1 = , a2 = ,
p ∂u q ∂v
(which is orthonormal). Hence the tangent to the curve can be written as
pa1 + v 0 qa2 .
From this we find that

q
τ = tan−1 v 0 .
p
By (INSERT REF) the geodesics are characterized by the condition
dτ
+ αp + βqv 0 = 0 .
du
Substituting into this the expression for τ we get a differential equation of
the form
v 00 = A(u, v) + B(u, v)v 0 + C(u, v)v 02 + D(u, v)v 03 , (6.2)
where A, B, C, D are functions of u, v. If (u0 , v0 ) is any point of U , v00 an

arbitrary constant, the differential equation (6.2) has a uniquely determined
solution v(u) satisfying the initial conditions
v(u0 ) = v0 , v 0 (u0 ) = v00 .
Geometrically this means that through any point of M and tangent to any
vector at the point, there passes exactly one geodesic.
We do not care about the explicit expressions for the coefficients A, B, C,
D in (6.2). We only remark that in order that (6.2) defines the geodesics these
coefficients are not arbitrary. The question of determining what conditions
they must satisfy is unsolved. It involves very complicated computations.
6.1 Geodesics 91
Exercise 82. Determine the geodesics on a plane, cylinder, and sphere, both
analytically and by direct geometric arguments. Note: for a curve on a surface
in ordinary space, if we take frames along the curve such that e1 is tangent
to the curve and e2 is normal to the curve but tangent to the surface, then
ω12 = de1 · e2 .
6.1.3 INSERT TITLE
One of the standard methods in geometry is to establish. for each problem, a

suitable coordinate system. In what follows we are going to give some different
coordinate systems, each with certain advantages. Let C0 be a curve. It is
given locally by (u(t), v(t)), where u02 + v 02 6= 0. By interchanging u and
v, if necessary, we can arrange that v 0 6= 0. Then by the Inverse Function
Theorem t is locally a differentiable function of v, so that the curve is locally
represented by u = u(t(v)) = f (v). Let
ũ = u − f (v) + u0
ũ = v0 ,
where u0 is some number. The Jacobian is

∂(ũ, ṽ) 1 −f 0 (v)

= =1,
∂(u, v) 0 1
so that ũ, ṽ are valid local coordinates. Writing u, v in place of ũ, ṽ, we have
coordinates such that C0 is locally defined by u = u0 . Construct through
every point (u0 , v) of C0 the geodesic perpendicular to C0 . Such a geodesic
can be defined by an equation of the form v = v(u, v0 ). By the theorem on
the dependence of the solution of (6.2) on the initial conditions, the function
v(u, v0 ) is differentiable with respect to both u and v0 . Since v(u, v0 ) = v0 ,
we have

∂v
=1.
∂v0 u=u0
Therefore at (u0 , v0 ) the Jacobian of u, v(u, v0 )

∂(u, v) 1 0
= ∂v = 1 .

∂(u, v0 ) ∂u 1
Hence we can take u and v0 as local coordinates in a sufficiently small neigh-

borhood of (u0 , v0 ). We now write v for v0 , so that the geodesics perpendicular
to C0 are u-curves, that is curves with v = const.
We wish to have coordinates such that the u-curves are orthogonal to the
v-curves. To do this, we observe that the condition that a curve defined by
u = u(v), whose tangent vector has components (u0 , 1), be orthogonal to the
u-curves, whose tangent vectors have components (1, 0), is that
92 6 INSERT TITLE
du
E(u, v) + F (u, v) = 0 . (6.3)
dv
by (INSERT REF) using the notation of (INSERT REF). Considered as
a differential equation for u, this has a unique solution u(v, u) such that
u(v0 , u0 ) = u0 , provided that (v0 , u0 ) lies in the domain of validity of u, v. It
is readily checked that the Jacobian of u, v with respect to u, v is 1 at (u, v0 ),
so that u and v can be taken as local coordinates in a neighborhood of (u0 , v0 ).
The v-curves are orthogonal to the u-curves, since (6.3) is satisfied. Writing
u for u, we can say, in summary, that the u-curves are geodesics orthogonal
to the curve u = u0 , and the v-curves are their orthogonal trajectories. A
neighborhood in which such a system of local coordinates u, v is valid is
called a geodesic field. The v-curves are called geodesic parallels.
Suppose that u, v form such a local coordinate system. Then since the
parameter curves form an orthogonal net we may apply the formulas of (IN-
SERT REF). We take a field of frames such that e1 is in the u direction which
makes τ = 0. Since the u-curves are geodesics, we have on each u-curve by
(INSERT REF)
pv qu pv
0 = ω12 = dτ − du + dv = − du .
q p q
Hence pv = 0, which says that p depends on u only, p = p(u). We change
coordinate u to ũ defined by
Z
ũ = p(u)du .
Writing again u for ũ we have
ω1 = θ1 = du , ω2 = θ2 = qdv ,
which implies by (INSERT REF) that
ds2 = du2 + Gdv 2 , G = q 2 .
Thus ds2 assumes a quite simple form in our local coordinate system u, v.
The arc length on a u-curve is given by
Z u2
s= du = u2 − u1 , if , u2 ≥ u1 .
u1
Hence we have the theorem:

Theorem 18. The geodesics of a field are cut by any two geodesic parallels
in arcs of the same length.
Conversely, suppose that we have a coordinate system u, v such that the
parameter curves form an orthogonal net. Suppose that the v-curves cut equal
lengths on the u-curves. Then
6.1 Geodesics 93
Z u2
pdu
u1
is independent of v, so that
Z u2 Z u2
d
0= pdu = pv du .
dv u1 u1
Since this is true for any values of u1 , u2 , v, we must have pv = 0. Taking

frames such that e1 is in the u direction so that τ = 0 we have on any u-curve,
where v = const.,
pv qu
ω12 = dτ − du + dv = 0 .
q p
Hence the u-curves are geodesics. We have proven the converse of the last
theorem: if the v-curves cut equal lengths on their orthogonal trajectories,
the latter are geodesics.
Using coordinates in a geodesics field as above we can see that the
geodesics are the curves of shortest length in a certain sense. Let (u1 , v1 )
and (u2 , v1 ) two points, u2 > u1 and v = v(u) a curve joining them which
lies in the geodesic field. Its length between these two points is
Z u2 p
1 + Gv 02 du ≥ u2 − u1 .CHECK THIS
u1
However u2 − u1 is the length of the geodesic v = v1 joining them. Moreover

the inequality is an equality only if v 0 = 0 which says that the v = const. and
the curve is a u-curve. This gives the theorem:
Theorem 19. A geodesic of a field is the unique shortest path in two of its
points and lying in the geodesic field.
Example 7. Let M be the unit sphere with its first fundamental form as
Riemannian metric. Let C0 be the equator. The orthogonal geodesics are the
lines of longitude and their orthogonal trajectories are the usual parallels, or
lines of latitude. The coordinates in this geodesic field are u = ϕ and v = θ,
the usual angular spherical coordinates.
Exercise 83. Show that the meridians of a surface of revolution in ordinary

space are geodesics.
Exercise 84. a) Show that a Riemannian surface M is locally isometric to

a surface of revolution in ordinary space if and only if there exist local coor-
dinates u, v on M such that
ds2 = du2 + p(u)dv 2
with |p0 (u)| ≤ 1.

94 6 INSERT TITLE
b) Show that a Riemannian manifold M is locally isometric to a surface

of revolution in ordinary space if and only if the orthogonal trajectories of
the curves K = const. are geodesics and at least one of the curves K = const.
has constant geodesic curvature.
c) Which Riemannian surfaces are locally isometric to a surface of revo-
lution in more than one way?
Exercise 85. Let P ∈ M and C an arc through P . Show that there exist
local coordinates u, v in a neighborhood of P such that on the arc C we have
p = q = 1, pv = qu = 0. Such coordinates are called Fermi coordinates relative
to C. Their significance might be described as follows. A two-dimensional
creature living on a Riemannian surface cannot determine experimentally
by measurements along an arc whether the Gauss curvature of his world
vanishes. To do so he must make measurements around a closed path.
Hint: Let a1 a2 be a family of frames on C which is parallel on C and such
that a2 is tangent to C at P . Show that coordinates in a geodesic field can
be chosen in such a way that the geodesics of the field are tangent to a1 . In
these coordinates p = 1; show that qu = 0 on C. Now find a function ṽ(v)
such that dṽ = q(u, v)dv on C.
6.2 Normal Coordinates

6.2.1 INSERT TITLE
Similar to the above coordinate system is the system of so-called geodesic po-
lar coordinates. We take a point O of M and construct the geodesics through
O. If ϕ is the angle which such a geodesic makes with some fixed direction at
O and we go along the geodesic a distance r, then r, ϕ are the geodesic polar
coordinates with center O of the resulting point. We will show that r, ϕ are
valid local coordinates in a neighborhood of each point of some neighborhood
of O, excluding O itself. In order to show this we first construct the so called
normal coordinates centered at O.
To construct these coordinates we start with some system of coordinates
in a geodesic field u, v. By a linear transformation, if necessary, we arrange
that at O,
u=v=O, p=q=1.
We now consider the geodesics (u(t), v(t)) with (u(0), v(0)) = (0, 0), where t
is a constant non-zero multiple c of arc length from 0. By (INSERT REF)
the components of the tangent vector of (u(t), v(t)) with respect to a1 , a2
are (u0 , qv 0 ), where the prime denotes differentiation with respect to t. Now
t = cs. Hence
1 ds p 02
= = u + q 2 v 02 . (6.4)
c dt
6.2 Normal Coordinates 95
By (INSERT REF), (INSERT REF) and taking into account that p = 1 we

must have
qv 0
0 = ω12 = dτ + qu dv , τ = tan−1 .
u0
Combining the last two equations, and differentiating (6.4) we obtain
−qv 0 u00 + u0 (qv 0 )0 = −qu v 0 (u02 + q 2 v 02 ) (6.5)

u0 u00 + qv 0 (qv 0 )0 = 0 . (6.6)
Solving this system for u00 and (qv 0 )0 , using Cramer’s rule, the denominator
being −(u02 + q 2 v 02 ), we obtain
u00 = qqu v 02 , (qv)0 = −qu u0 v 0
or equivalently,
u00 = qqu v 02 (6.7)
1
v 00 = − (2qu u0 v 0 + qv0 2 ) . (6.8)
q
Thus our geodesics through (0, 0) parametrized by a constant non-zero mul-
tiple of arc length are solutions of (6.7). Conversely any solution of (6.7) with
u02 + q 2 v 02 6= 0 satisfies (6.5) and is therefore a geodesic parametrized by a
constant multiple of arc length. On the other hand if u02 + q 2 v 02 = 0 then
u0 = v 0 = 0 so u(t), v(t) are constant.
By the Fundamental Theorem of Differential Equations the system (6.7)
has a unique solution u(t), v(t) characterized by the properties that u(0) =
v(0) = 0 and u0 (0) = u00 , v 0 (0) = v00 , where u00 , v00 are arbitrary constants.
This solution is defined in an interval containing 0, and we take this interval
to be as large as possible. Furthermore, if we express u and v as functions of
the initial conditions u00 , v00 as well as of t:
u(t, u00 , v00 ), v(t, u00 , v00 ) ;
then these functions are differentiable and defined at least provided

p
|t| ≤ , u0o 2 + v00 2 ≤ η ,
for some positive numbers and η

Let us now suppose that t, u00 , v00 are numbers satisfying
p
|t| u0o 2 + v00 2 ≤ η . (6.9)
Define functions u, v by
u(t, u00 , v00 ) = u(Ct, C −1 u00 , C −1 v00 )

v(t, u00 , v00 ) = v(Ct, C −1 u00 , C −1 v00 ) ,
96 6 INSERT TITLE
where
1
C = η −1 (u00 2 + v00 2 ) 2 .
The right-hand sides are defined and differentiable in all three variables since
1
|Ct| ≤ |t|(u00 2 + v00 2 ) 2 η −1 ≤ , and
1
((C −1 u00 )2 + (C −1 v00 )2 ) 2 = η .
Considered as functions of t, u, and v give a parametrization of a geodesic

through 0 by a constant multiple C −1 c of arc length where c = ((C −1 u00 )2 +
1
(C −1 v00 )2 )− 2 by (6.4). It follows that u, v satisfy the system (6.7). They
satisfy the initial conditions
u(0, u00 , v00 ) = v(0, u00 , v00 ) = 0 ,
and

d
u(t, u00 , v00 ) = Cu1 (0, C −1 u00 , C −1 v00 ) = u00
dt t=0

d
v(t, u00 , v00 ) = Cv1 (0, C −1 u00 , C −1 v00 ) = v00 .
dt t=0
It follows from the uniqueness part of the Fundamental Theorem of Ordinary
Differential Equations that
u(t, u00 , v00 ) = u(t, u00 , v00 )

v(t, u00 , v00 ) = v(t, u00 , v00 ) ,
which tells us in effect that u, v are defined and differentiable in t, u00 , v00
provided (6.9) is satisfied.
Suppose that t, u00 , v00 satisfy (6.9) and let C be any number different
from zero. Then
1
|Ct|((C −1 u00 )2 + (C −1 v00 )2 ) 2 ≤ η ,
so that u(Ct, C −1 u00 , C −1 v00 ), v(Ct, C −1 u00 , C −1 v00 ) are defined and differen-
tiable. As before, u(0, C −1 u00 , C −1 v00 ) = v(0, C −1 u00 , C −1 v00 ) = 0,

d −1 0 −1 0 0 d
u(Ct, C u0 , C v0 ) = u0 , v(Ct, C −1 u00 , C −1 v00 ) = v00 ,
dt t=0 dt t=0
and these functions as functions of t represent a parametrization of a geodesic
by a constant multiple of arc length and therefore satisfy (6.7). It follows as
above that
u(Ct, C −1 u00 , C −1 v00 ) = u(t, u00 , v00 )

v(Ct, C −1 u00 , C −1 v00 ) = v(t, u00 , v00 ) .
1
Letting C = t gives
u(1, tu00 , tv00 ) = u(t, u00 , v00 ) , (6.10)

v(1, tu00 , tv00 ) = v(t, u00 , v00 ) . (6.11)
The functions
u(x, y) = u(1, x, y)
v(x, y) = v(1, x, y)
p
are defined and differentiable provided that x2 + y 2 < η, as we have seen.
Using (6.10) we find that

∂u d
= u(x, 1, 0) = 1 .
∂x (0,0) dx 0
The other derivatives of u, v with respect to x, y at (0, 0) may be found in a

similar fashion, with the result that the Jacobian matrix of u, v with respect
to x, y at (0, 0) is the identity matrix. Hence x, y can be taken as local
coordinates on M in some neighborhood of P called a normal neighborhood ;
x, y are called normal coordinates. Since by choice of u, v, ds2 = du2 + dv 2
at P , and the Jacobian matrix is the identity matrix, we have
ds2 = dx2 + dy 2 at P . (6.12)
We now introduce polar coordinates
x = r cos ϕ , y = r sin ϕ (6.13)
in the x, y plane in the usual way. These form a local coordinate system
in a neighborhood of any point in the plane except (0, 0), and consequently
they can be taken as local coordinates in some neighborhood of any point of
our normal neighborhood except for P itself. They are called geodesic polar
coordinates. If we hold ϕ fixed we note that u(x, y) = u(l, r cos ϕ, r sin ϕ) =
u(r, cos ϕ, sin ϕ) and similarly for v. Since cos ϕa1 + sin ϕa2 is a unit vector
the resulting curve u(r), v(r) is a geodesic through P parametrized by arc
length from P . We wish next to express our metric in these coordinates. Let
U ⊂ R2 be the domain of validity of the normal coordinates centered at P
and consider the map (6.13), which we call Π, from the r, ϕ-plane to the
x, y-plane. Here it will be convenient to allow r to be negative. Ordinarily in
using geodesic polar coordinates we take r ≥ 0 just as we do in the familiar
case of the plane. Let Π −1 (U ) = V . Then we have a map T from V to M
∂
defined by (6.13), taking x, y as normal coordinates. Let e1 = T∗ ( ∂r ). Then
e1 is a unit tangent vector, since for ϕ fixed, as we just saw, u(r cos ϕ, r sin ϕ),
v(r cos ϕ, r sin ϕ) represents a geodesic through P with arc length as param-
eter. Let e2 be the unit normal in M to e1 , chosen so that e1 e2 agrees with
98 6 INSERT TITLE
the orientation of M . Thus we obtain a map to =, the bundle of frames of

M . Pull back ω1 , ω2 , ω12 . We write these forms in terms of dr and dϕ. On
the geodesic dϕ = 0, we have ω1 = dr, ω2 = 0, and ω12 = 0. Consequently
these forms can be written as
ω1 = dr + mdϕ
ω2 = qdϕ
ω12 = Qdϕ ,
for some functions m, q, Q. The structure equation gives
dω1 = mr dr ∧ dϕ
= ω12 ∧ ω2 = 0 .
Hence mr = 0. But for r = 0 the map T is constant, that is to say T (0, ϕ) = P ,

so that ω1 = ω2 = 0 on the line r = 0. Consequently m(0, ϕ) = 0. Since
mr = 0 everywhere we have that m vanishes identically. Our metric therefore
takes the form ds2 = ω12 + ω22 = dr2 + Gdϕ2 , where G = q 2 .
We note that C is well-defined and differentiable at every point where
the geodesic polar coordinates are valid, that is to say at every point in
the normal neighborhood except at P . We now show that G extends to a
function differentiable at P and we prove certain expansion formulas. Invert
formulas (6.13) in the usual way:
y
r2 = x2 + y 2 , ϕ = tan−1 .
x
Differentiating these and substituting in our expression for ds2 we obtain
x2 dx2 + 2xydxdy + y 2 dy 2 y 2 dx2 − 2xydxdy + x2 dy 2

ds2 = 2
+G (6.14)
r r4
2 2 2
r y + Gx2 2
2 2

r x + Gy 2 1 G
= dx + − 2xydxdy + dy . (6.15)
r4 r2 r4 r4
Now the coefficients of dx2 , dxdy, and dy 2 must extend differentiably to P ,

since x, y are valid local coordinates. By (6.12) their values must be 1, 0, 1
respectively, at P . Adding the coefficients of dx2 and dy 2 and simplifying, we
obtain
G
1+ ,
r2
which shows that rG2 extends differentiably to P with limiting value 1. (It
follows that C itself extends differentiably to P .) Consequently we may write
G
= H , where H(0, 0) = 1 ,
r2
with H everywhere differentiable. The coefficient of dxdy in (6.14) may now

be written as
1−H
J(x, y) = 2xy , (6.16)
r2
which, as we said, must extend differentiably to P and have 0 as value at P .
Now J(0, y) = 0. Hence
Z 1
d
J(x, y) = J(x, y) − J(0, y) = J(tx, y)dt
0 dt
Z 1
=x J1 (tx, y)dt .
0
The last integral is differentiable in x and y since it can be differentiated under

the integral sign. We write J(x, y) = L(x, y)x. Now J(x, 0) = 0, by (6.16)
and continuity of J at P . Hence L(x, 0) = 0 for all x. By an argument similar
to what we have just given, we find that
L(x, y) = yM (x, y) ,
for some differentiable function M (x, y). This gives
1 − H(x, y)
xyM (x, y) = 2xy ,
r2
and cancelling xy, which is permitted by continuity, we have
1 − H(x, y) 1
= M (x, y) ,
r2 2
so that

2 1 2
2
G(x, y) = r H(x, y) = r 1 − r M (x, y) ,
2
which we write as
G(x, y) = r2 1 + r2 Q(x, y) ,

(6.17)
where Q = − 21 M is differentiable in x and y for all x, y in the normal

neighborhood, P included. Equation (6.17) can also be written as
G = r2 + αr 4 + r5 (β cos ϕ + γ sin ϕ) + O(r6 ) , (6.18)
where α, β, γ are constants and O(r6 ) denotes a function of x, y which is a

multiple of r6 . Substituting into (6.14), we get
ds2 = dx2 + dy 2 + α(xdy − ydx)2 + ... (6.19)

100 6 INSERT TITLE
where the dots represent a quadratic differential form in x, y whose coeffi-

cients are O(r3 ). This equation expresses the ds2 in terms of normal coordi-
nates. Now the normal coordinates are determined up to orthogonal transfor-
mation about (0, 0) depending on the choice of the vector a1 from which the
angle ϕ is measured and the sense of the angle; it follows that the coefficients
of (6.19), or their combinations, which are invariant under an orthogonal
transformation on x, y, are invariants of the Riemannian metric at P . This
is in particular the case with α. Its relation with known invariants is easily
found. In fact, from (INSERT REF) we have
√
1 ∂2 G
K = −√ .
G ∂r2
By (6.18) we find
K(P ) = −3α ,
where K(P ) is the Gaussian curvature at P .

Using geodesic polar coordinates, we can obtain some simple geometric
interpretations of the Gaussian curvature. Call the curves r = const. geodesic
circles, with center P and radius r. The perimeter of a geodesic circle of
radius r is
Z π √ Z π
K(P ) 3
L= Gdϕ = r− r + ... dϕ
−π −π 6
π 3
= 2πr − K(P )r + ... .
3
Hence we get
3 2πr − L
K(P ) = lim .
r→0 π r3
Similarly, the area of a geodesic circle of radius r is
ZZ Z rZ π √ Z r
π
A= θ1 ∧ θ2 = Gdϕdr = Ldr = πr2 − K(P )r4 + ... .
0 −π 0 12
From this it follows that
12 πr2 − A
K(P ) = lim .
r→0 π r4
6.2.2 INSERT TITLE
From (6.2.1) we can derive the following useful result.

Theorem 20. Every point P of N has a neighborhood U such that the
geodesic lying in U and joining P to any point Q ∈ U is the only curve
of shortest length in M joining these two points.
Proof. To prove this it suffices to take the geodesic polar coordinates r, ϕ

about P and to define U by r < c, where c is a positive number so small that
the geodesic polar coordinates are valid for all r, ϕ such that r < c. If C is a
curve joining P to Q and lying in U then its length is
Z Z p Z
ds = dr2 + Gdϕ2 ≥ dr = r(Q) .
C C C
The inequality is an equality if and only if dϕ = 0, which is to say if and only
if C is the unique geodesic lying in U joining P to Q. On the other hand if
C goes outside of U then it must meet the geodesic circle r = c and so its
length is greater than c and so is longer than the unique geodesic joining P
and Q and lying in U .
The notions we have introduced above, the parallelism of Levi-Civita, the
geodesic curvature of a curve, the Gaussian curvature, etc., all pertain to a
neighborhood. Some of the most interesting results in Riemannian geometry
are concerned with the relations of such properties with the properties of the
manifold as a whole, or, more specifically, with its topological properties. To
study these an effective tool is the distance function in M . Such a function
assigns to any two points P , Q of M a number d(P, Q) > 0, such that the
following conditions are satisfied:
1. d(P, Q) = 0 if and only if P = Q.
2. d(P, Q) = d(Q, P ).
3. d(P, Q) + d(Q, R) ≥ d(P, R).
In topology we say that the distance function defines a metric in M and that
M is a metric space. Notice that such a metric is entirely different from what
we call a Riemannian metric.
We wish to define a distance function in M which bears a close relation
with its Riemannian metric. For any two points P, Q ∈ M we define d(P, Q)
to be inf P Q, where P Q is the arc length of a sectionally smooth curve, called
a path, joining P , Q. To justify this definition we have to show that such a
curve exists. In fact, from the connectedness of M , which we assume here, it
follows that there is a sequence of coordinate neighborhoods U1 , ..., Um , such
that P ∈ U1 , Q ∈ Um , and Ui ∩ Ui+1 6= 0, i = 1, ..., m − 1. Let Pi ∈ Ui ∩ Ui+l .
Then any two consecutive points of the sequence P, P1 , ..., Pm−l , Q lie in the
same coordinate neighborhood and can be joined by a smooth arc. These arcs
make up a sectionally smooth arc joining P , Q.
This distance function clearly has the properties:
d(P, P ) = 0 , d(P, Q) = d(Q, P ) ,
d(P, Q) + d(Q, R) ≥ d(P, R) .
Moreover, if P 6= Q, it follows from the statement in the beginning of this
section that d(P, Q) > 0. Hence our distance function has the properties 1)-3)
and makes M into a metric space.
It is important to observe that the topology defined in (MISSING TEXT)
102 6 INSERT TITLE
6.3 The second variation of arc length

Earlier in this chapter we have seen that geodesics have the following two
minimizing properties: firstly, if C is a geodesic of a field then it is the shortest
path joining two of its points and lying in the geodesic field; secondly, every
point O has a “normal neighborhood” Dr such that for any P ∈ Dr there
is a unique geodesic lying in Dr and joining O to P , and this is the unique
shortest path in M joining O and P . Our approach to these results has been
local in nature; we now want to consider how they can be extended globally.
In particular, if g is a geodesic which starts from a point O of M , and if
P is a variable point on C, then C furnishes the unique shortest path in
M joining O and P at least as long as P lies in a normal neighborhood
centered at O. Beyond that, if G between O and P belongs to a geodesic
field, then G furnishes the unique shortest path joining O and P lying in the
geodesic field. It is natural to ask when, if ever, G fails to provide the shortest
path between O and P as P moves farther and farther away from O. The
study of this question leads to relations between the topology of M and its
differential geometry, and to information on the topology of M , particularly
in higher dimensions. We are not in a position here to say much more about
this development, but we trust that the results we present in two dimensions
will already be found sufficiently attractive.
6.3.1 INSERT TITLE
We begin by examining how the arc length of a curve is changed if the curve
is smoothly deformed. Let M be a Riemannian manifold of dimension 2 and
R the rectangle in the (x, t)-plane
R = {(x, t) | 0 ≤ x ≤ V , −ζ < t < ζ} ,
where V , ζ are two positive numbers. By a 1-parameter family of curves in

M we mean a differentiable map F from R to M which has rank 1 if t is
fixed, i.e. in local coordinates u, v on M .

∂u ∂v
, 6= (0, 0) .
∂x ∂x
To each (x, t) we assign the frame P e1 e2 of M such that

∂u ∂ ∂v ∂
P = F (x, t) , e1 along + ;
∂x ∂u ∂x ∂v
we assume that M is oriented, so that e2 is then uniquely determined. Under
this map the various differential forms on the frame bundle of M are pulled
back to R.
The exterior derivative on R may be decomposed into the sum of differ-
entiation with respect to x and differentiation with respect to t:
6.3 The second variation of arc length 103
∂ ∂
d = dC + dt , dC = dx ∧ , dt = dt ∧ .
∂x ∂t
Likewise, the forms ω1 , ω12 , pulled back from the frame bundle of M , are
decomposed as
ωi = θi + ai dt , ω12 = θ12 + a12 dt .
where θ1 , θ12 are characterized by
θi ∧ dx = θ12 ∧ dx = 0 .
For t fixed, let Ct denote the curve F (t, x) in M . (It is parametrized by

x.) The length of Ct , L(t), is given by
Z
L(t) = θ1
t×[0,V ]
and its derivative by

Z
dL d ∂θ1
(t) ≡ . (6.20)
dt dt t×[0,V ] ∂t
To find the differential form in the second integral we calculate dω, in two
different ways (differentiate and then decompose, decompose and then differ-
entiate):
dω1 = ω12 ∧ ω2 = (a2 θ12 − a12 θ2 ) ∧ dt

∂θ1
dω1 = d(θ1 + a1 dt) = − + dC a1 ∧ dt .
∂t
Cancelling dt gives
dθ1
= dC a1 + a12 θ2 − a2 θ12 . (6.21)
dt
Now θ2 = 0, since e1 is tangent to Ct ; θ12 is the connection form, which
vanishes on Ct if and only if Ct is a geodesic. By combining (6.20) and (6.21)
we obtain
Z Z
dL
(t) = (−a2 θ12 + dC a1 ) = − a2 θ12 + a1 (V, t) − a1 (0, t) , (6.22)
dt Ct Ct
where the second equality is gotten by using the Fundamental Theorem of

Calculus, and where we have indicated the domain of integration t × [O, V ]
by the somewhat imprecise but simpler expression Ct . Any family Ct is called
a variation of C0 and a1 (0, t), a2 (0, t) are called, respectively, the tangential
and normal components of the variation vector. On Ct , θ12 = kg ds, where kg
is the geodesic curvature and ds is the element of arc.
104 6 INSERT TITLE
We next show that given any curve C(x), 0 ≤ x ≤ V , in a Riemannian

surface and any differentiable real-valued function f (x), 0 ≤ x ≤ V , we
can find a 1-parameter family of curves Ct in M such that C = C0 and
a2 (x, 0) = f (x). In fact, taking frames e1 e2 along C0 , with e1 tangent to C
we let F (x, t) be the point obtained by going along the geodesic perpendicular
to C at C(x) a distance f (x)t from C(x) in the e2 -direction. We must find
a positive ζ such that F (x, t) is defined for all (x, t) with 0 ≤ x ≤ V and
−ζ < t < ζ, ∂u ∂v
∂x ∂x 6= 0. Note that the last condition is assumed to hold
,
when t = 0. The Fundamental Theorem of Ordinary Differential Equations
guarantees that for each y, 0 ≤ y ≤ V , there are positive numbers y , ζy such
that F (x, t) is defined for all (x, t) with |x−y| < y , |t| < ζy , and the required
ζ for all x is then gotten by a compactness argument. It remains to show that
a2 (x, 0) = f (x). But along the geodesic through C(x) perpendicular to C the
element of arc is ds = f (x)dt, while ω2 = a2 dt is the element of arc at t = 0,
i.e. ω2 = ds at t = 0. Hence a2 (x, 0) = f (x), as was to be shown. This one-
parameter family of curves is called the one-parameter family corresponding
to a2 . Note that if f (O) = f (V ) = 0 then the endpoints of the corresponding
family are fixed at C0 (0) and C0 (V ), respectively. If f is continuous and
piecewise differentiable this construction can still be carried out; the resulting
curves are continuous but may have finitely many corners.
Formula (6.22) has various geometric consequences. First of all, if C is a
geodesic and we extend it to a 1-parameter family of curves, as above, we
have θ12 = 0 on C = C0 . If the curves of the family have the same fixed
endpoints P , Q, then a1 (V, t) = a1 (0, t) = 0, and hence
dL
δL = (0) = 0 .
dt
Conversely if C is a given curve while is not a geodesic, we can find a 1-
parameter family of curves Ct with fixed endpoints such that C0 = C, and
such that a2 (x, 0) is non-zero only on an interval where the geodesic curvature
of C is bounded away from zero. Taking a2 to have the same sign as the
geodesic curvature, we find that δL < 0. (This implies that there are curves
of the family Ct of strictly shorter length than C0 = C.) What we have shown
can be summed up in the following way: for curves joining fixed points P ,
Q of M , the geodesic or geodesics, if any, joining P and Q are precisely the
curves for which the first variation of arc length, δL, vanishes.
We mention another consequence of formula (6.22). Suppose that D is
a curve in M and consider the family of geodesics perpendicular to D; al-
ternatively, let D be a point and consider a 1-parameter family of geodesics
through D. Suppose that the geodesics in question envelop a curve E (that
is to say are tangent to a curve E). We parametrize the geodesics so that x0
lies on D and x = V is the point of tangency with E. Then in formula (6.22)
θ12 = 0 because the curves are geodesics, and a1 (0, t) = 0 because the end-
points at 0 are fixed or move perpendicularly to the geodesics as t varies. On
the curve endpoints at V , i.e. E, the geodesics, and hence the vectors e1 , are
tangent to E, and therefore ω1 = a1 (V, t)dt = ds is the element of arc on E.

Hence formula (6.22) reduces to
dL = ds .
Integrating this gives the following result: the difference in length of two
geodesics of the family perpendicular to (or passing through) D and enveloping
E is equal to the length of the arc of E lying between the points of tangency
of the two geodesics.
(INSERT FIGURE)
Exercise 86. Let D be a plane curve and let E be the locus centers of
curvature of D. Let P, Q ∈ D and let k denote the curvature of D. Find the
length of the arc of E between the centers of curvature of D at P and at Q
by a) direct computation using the Frenet formulas, and b) as a special case
of the result just proved.
6.3.2 INSERT TITLE

2
We next calculate ddtL
2 (0) for a 1-parameter family of curves Ct on M . For
simplicity, and because it suffices for our needs, we will assume that C0 is a
geodesic. This condition can be expressed analytically as
θ12 = 0 when t = 0 . (6.23)

∂θ12
The first task will be to find ∂t . To do this we calculate dω12 in two different
ways:
dω12 = −Kω1 ∧ ω2 = K(a1 θ2 − a2 θ1 ) ∧ dt

∂
dω12 = (dC + dt ∧ )(θ12 + a12 dt)
∂t
∂θ12
= − + dC a12 ∧ dt .
∂t
Remembering that θ2 = 0, and cancelling dt, we obtain
∂θ12
= dC a12 + Ka2 θ1 . (6.24)
∂t
In order to determine dC (a12 ) along t = 0, we differentiate ω2 in two different
ways:
dω2 = ω1 ∧ ω12 = (θ1 + a1 dt) ∧ (θ12 + a12 dt)

= (a12 θ1 − a1 θ12 ) ∧ dt
dω2 = d(a2 dt) = dC a2 ∧ dt .
At t = 0, where θ12 = 0, we obtain

106 6 INSERT TITLE
dC a2 = a12 θ1 . (6.25)
Now θ1 is the element of arc along the curve C0 , and we will henceforth write
it as dr. Formula (6.25) then becomes
da2
a12 =
dr
and therefore
d2 a2
dC a12 = dr . (6.26)
dr2
From (6.22), taking into account restriction (6.23) and formulas (6.24)
and (6.26), we obtain the formula for the second variation
d2 L
Z
2 ∂θ12 ∂a1 ∂a1
δ L = 2 (0) = − a2 + (V, 0) − (0, 0)
dt C ∂t ∂t ∂t
Z 2
d a2 ∂a1 ∂a1
=− a2 2
+ Ka2 dr + (V, 0) − (0, 0) .
C dr ∂t ∂t
where we write C for C0 and recall that this is proved under the assumption
that C is a geodesic. Integration by parts yields another formula for the
second variation:
Z 2 !
2 da2 2 da2 da2
δ L= − Ka2 dr + a2 (0) (0) − a2 (L) (L)
C dr dr dr
∂a1 ∂a1
+ (V, 0) − (0, 0) .
∂t ∂t
where, to simplify the notation, we have written
a2 (0) = a2 (0, 0) , a2 (L) = a2 (V, 0) .
where L is the length of C. We now regard a2 as a function of r, 0 < r < L,

with r = 0 corresponding to x = 0.
6.3.3 INSERT TITLE
In order to draw out the geometric consequences of the second variation

formula, we must first re-examine geodesic polar coordinates, which we con-
structed in (INSERT REF), from a global point of view. Let O ∈ M be a
point of a Riemannian surface and let v be a tangent vector of M at O. Then
expO (v) is defined to be the point of M obtained by going along the geodesic
through O tangent to v a distance |v| in the direction of v, provided that this
point exists. The mapping expO goes by the name of the exponential mapping
at O. We note that expO is defined on some neighborhood of O in the tangent
space V (O) of M at O, where it is essentially nothing more than the normal,

or geodesic polar coordinates, which we have previously introduced, but it
may not be defined on all of V (O). (If the metric on M defined in (INSERT
REF) has the so-called completeness property, then it can be proved that
expO is defined on all of V (O).)
We claim that expO , which is defined on an open subset of V (O), is dif-
ferentiable. To prove the claim we let vO ∈ V (O) be such that expO (vO )
is defined. Note that if |vO | is sufficiently small then the claimed differen-
tiability is part of the conclusion of the Fundamental Theorem of Ordinary
Differential Equations. In general, assuming that vO is not the 0-vector, we
let vO = |vO |eO , where eO is a vector of unit length. Let C(r), 0 ≤ r ≤ |vO |,
be the geodesic expO (reO ), which is parametrized by arc length. Using the
Fundamental Theorem and an argument from elementary topology, we find
that there is a sequence of numbers rk , 0 = r0 < r1 < ... < rn = |vO |, such
that if ek denotes the tangent vector of C(r) at rk (which is a unit vector
since C(r) is parametrized by arc length), then there is a neighborhood Uk
of ek in the unit tangent bundle F of M for each k, 0 ≤ k ≤ n − 1, hav-
ing the property that if e ∈ Uk then expO (λe) is defined and differentiable
in λe provided |λ| < 2(rk+1 − rk ). If e ∈ Uk ∩ V (P ) let e0 be the tangent
vector of D(λ) = expP (λe) at λ = rk+1 − r. Note that e0k = ek+l and that
the correspondence e → e0 is differentiable. We call e0 the k-translate of e.
0
Now by continuity there is a neighborhood Un−2 ⊂ Un−2 of en−2 such that
0
if e ∈ Un−2 then the (n − 2)-translate of e lies in Un−1 ; likewise there ex-
0 0
ists a neighborhood Un−3 ⊂ Un−3 of en−3 such that if e ∈ Un−3 then the
0
(n − 3)-translate of e lies in Un−2 . Continuing in this fashion we arrive at a
neighborhood UO0 of eO such that if e ∈ UO0 ∩ V (P ) then the composition of
the k-translates of eO ≤ k ≤ n − 2, that is to say the tangent vector e00 of
D(r) = expP (re) at r = rn−1 , lies in Un−1 and depends differentiably on re.
From this it follows that D(r) is defined for all r such that 0 < r < 2rn −rn−1
and depends differentiably on re. The same is true if e is restriced to lie in
V (O), which is a submanifold of the unit tangent bundle. Hence expO is
differentiable at VO , which proves the claim.
For each v in the tangent space at O for which expO is defined, we associate
the right-handed frame e1 e2 of M at expO (v) such that e1 is tangent to
C(r) = expO (rv) in the direction of increasing v. Pulling back ω1 , ω2 from
the frame bundle of M , and taking polar coordinates r, ϕ in the tangent
space to M at O with respect to some right-handed orthonormal frame, we
have, by the argument given in (INSERT REF) in connection with geodesic
polar coordinates,
θ1 = dr (6.27)
θ2 = qdϕ , (6.28)
for some function q(r, ϕ). Now expO may fail to be one-to-one, and its rank
may fall below 2. By (6.27) and (6.28), and the fact that expO is non-singular
108 6 INSERT TITLE
at O, we find that the rank falls exactly where r > O and q = O. (A point
where the rank of a mapping falls is called a singular point.) At such a point
the rank is 1. Note that the metric pulled back to V (O) is given by
ds2 = θ12 + θ22 = dr2 + Gdθ2 , G = q 2 ,
and, by (INSERT REF), q satisfies the differential equation
∂2q
+ Kq = 0 , (6.29)
∂r2
where K is the Gaussian curvature.
Exercise 87. On the sphere of radius ρ, using spherical coordinates, and on

the circular cylinder of radius ρ, using cylindrical coordinates, determine the
exponential map expO , where O is the north pole, in the case of the sphere,
and O is any convenient point in the case of the cylinder. Determine the
domain of definition and the singular points of expO in each case.
6.3.4 INSERT TITLE
Our preparations are now completed and we have in hand the tools for car-
rying out a rather powerful and global analysis of the minimizing properties
of geodesics. Let C(r), A < r < B be a geodesic in M parametrized by arc
length r. By a Jacobi field along C we mean a differentiable function a(r),
A < r < B, which satisfies the differential equation
d2 a
+ K(C(r))a = 0 , (6.30)
dr2
where K is the Gaussian curvature. If r1 6= r2 , and there is a non-zero Jacobi
field along C which vanishes at r1 and r2 , we say that r1 and r2 are conjugate
along C. We specify this relation in terms of the parameter values, instead
of the image points P = C(r1 ), Q = C(r2 ), since C may fail to be one-
to-one. However, if r1 and r2 are conjugate along C and when there is no
danger of confusion, we say, simply that P and Q are conjugate along C. The
differential equation (6.30) is linear in a. Consequently, if a is a Jacobi field
then λa is also a Jacobi field, for any constant λ. Moreover, by the uniqueness
part of the Fundamental Theorem of Ordinary Differential Equations, if a, a
are two non-zero Jacobi fields which vanish at r1 , then
a0 (r1 )
a= a,
a0 (r1 )
since both sides are Jacobi fields and have the same first derivative at r1 .
Let rO be a point in the domain of C, O = C(rO ). Let us change the
parameter so that arc length is measured from rO and O = C(O). Let e be a
unit tangent vector to C at O in the direction of increasing arc length. Then

C(r) = expO (re). Let us introduce polar coordinates r, ϕ in V (O) so√that
e is in the direction ϕ = ϕO . From equation (6.29) we see that q = G is
a Jacobi field on C. Moreover, q vanishes at r = 0, as we saw in (INSERT
REF). Let r1 be such that C(r1 ) is defined. Reversing the parametrization
of C if necessary, we can assume that r1 > 0. As we showed in (INSERT
REF), q(r1 , ϕO ) = 0 if and only if expO is singular at r1 e. Now any Jacobi
field which vanishes at 0 is a constant multiple of q, as we just saw. Hence
r1 is conjugate to 0 along C if and only if q(r, ϕfO ) = 0. It follows that r1
is conjugate to O along C(r) if and only if expO is singular at r1 e1 . Now
let e be a unit vector tangent to C(r) at r1 and directed in the direction
of decreasing r. Let P = C(r1 ). Then since r1 and O are conjugate expP is
singular at r1 e1 . This remarkable fact can be expressed simply but crudely as
follows: expO is singular at P if and only if expP is singular at O. We note
also that the conjugate points of O along C are isolated. For otherwise the
zeros of q would have an accumulation point on C and therefore, by the Mean
∂q ∂q
Value Theorem, ∂r would vanish at the accumulation point. But q and ∂r
cannot vanish simultaneously because q satisfies (6.30) and the Fundamental
Theorem of Ordinary Differential Equations tells us that a solution of (6.30)
which vanishes simultaneously with its derivative must be identically zero,
whereas q 6= 0 in the domain of validity of the geodesic polar coordinates.
Let O ∈ M and let C be a geodesic parametrized by arc length through
O with unit tangent vector e. Suppose that expO (re) = P is defined and that
there are no conjugate points to O along D between O and r, r included,
which is to say that expO (re) is non-singular for all r such that 0 ≤ r ≤ r.
By continuity of the derivatives the set of points of V (O) where expO is
non-singular forms an open neighborhood N of re|0 ≤ r ≤ r. Now let C(t),
0 ≤ t ≤ V , be a path in M joining O to P . We say that C(t) is nearby to
C(r) if it can be “lifted” under the exponential map to N , i.e. there is a curve
Ĉ(t) in N such that Ĉ(0) = 0, Ĉ(V ) = re, and expO (Ĉ(t)) = C(t). If C(t)
is nearby to C(s), then since its “lift” Ĉ(t) lies in N where the pull-back of
ds2 is given by
ds2 = dr2 + Gdϕ2 , G > 0 if r > 0 ,
its length is given by

Z p Z
L(C) = dr2 + Gdϕ2 ≥ dr = r ,
Ĉ(t) Ĉ(t)
with equality holding if and only if ϕ = const., that is to say if and only if
Ĉ(t) lies on the line re. Thus we have shown the following
Theorem 21. If, along the geodesic C(r), 0 ≤ r < r, there are no conjugate-
points of O, then C(r) is shorter than any other path nearby to C(r) and
joining C(O) and C(r).
110 6 INSERT TITLE
On the other hand we will next show the following
Theorem 22. If O and r̂, r̂ > 0, are conjugate along the geodesic C(r), and
r > r̂, then C(r) is not the shortest path joining C(O) to C(r). There exists,
in fact, a one-parameter family of paths Ct with CO = C such that for all
t 6= 0 Ct is strictly shorter than C.
We may express this informally by saying that a geodesic starting from 0

fails to be minimizing beyond the first conjugate point.
Remark 11. The curves to be constructed are not smooth but have each one
a corner. It can be shown that the corner of any one of them can be rounded
off to give a smooth curve of length shorter than that of D(r).
Proof. Recall the second variation formula

Z 2
2 d a2 ∂a1 ∂a1
δ L=− a2 2
+ Ka2 dr + (V, 0) − (0, 0) , (6.31)
C dr ∂t ∂t
valid where C is a geodesic.

Suppose that O and r̂, r̂ > O, are conjugate along the geodesic C(r) and
that r > r̂ with C defined at r. Take a Jacobi field which vanishes at O and
r̂, we can suppose that it is q, referring to the exponential map at C(O) and
let f (r) be the function defined by

q(r) , O ≤ r ≤ r̂
f (r) =
O , r̂ ≤ r ≤ r .
Then f (r) is continuous, but is not differentiable at r̂, although right and left
derivatives exist at r̂, with
f 0 (r̂−) 6= 0 , f 0 (r̂+) = 0 .
Let
a2 (r) = f (r) + η(r) ,
where η is a function and a number yet to be determined. Then

Z r 2 Z r 2
d a2 d f
a2 + Ka2 dr = f + Kf dr
0 dr2 0 dr2
Z r 2 Z r 2 !
d η d f
+ f + Kη dr + η + Kf dr
0 dr2 0 dr2
Z r 2
d η
+ 2 η + Kη dr .
0 dr2
Now the first integral on the right, and the second integral in the term,
vanish because the integrand vanishes. Integrating by parts twice on [O, r̂]
and [r̂, r] gives
Z r 2 Z r̂ 2 Z r 2 Z r
d η d η d η
f + Kη dr = f dr + f dr + f Kηdr
0 dr2 0 dr2 r̂ dr2 0
dη dη dη dη
= f (r̂) (r̂) − f (0) (0) + f (r) (r) − f (r̂) (r̂)
dr dr dr dr
Z r̂ Z r Z r
df dη df dη
− dr − dr + f Kηdr
0 dr dr r̂ dr dr 0
df df df df
= − (r̂−)η(r̂) + (0)η(0) − (r)η(r) + (r̂+)η(r̂)
dr dr dr dr
Z r 2
d f
+ η + Kf dr
0 dr2
df df
= − (r̂−)η(r̂) + (0)η(0) .
dr dr
Hence
Z r 2 r !
d2 η
Z
d a2 df df
− + Ka2 dr = (r̂−)η(r̂) − (0)η(0) − η + Kη dr .
0 dr2 dr dr 0 dr2
We now let η be a continuous function on [0, r] such that η(0) = η(r) = 0,

and η(r̂) is non-zero and has the opposite sign to
df dq
(r̂−) = (r̂) .
dr dr
(Note that this quantity is non-zero, since q and q 0 cannot vanish simultane-
ously. For example, let η(r) = ± sin(πr−1 r).
Then
Z r 2 Z r 2
d a2 dq 2 d η
− a2 + Ka 2 dr = (r̂)η(r̂) − η + Kη dr .
0 dr2 dr 0 dr2
(6.32)
Our aim is to calculate δ 2 L for the variation corresponding to a2 . We have

just calculated the integral in the formula; we next calculate the other terms.
Since a2 (0) = a2 (r) = 0, the endpoints are fixed and therefore
a1 (0, t) = a1 (r, t) = 0 .
The only remaining non-zero terms are

∂a1 ∂a1
(r̂−, 0) − (r̂+, 0) .
∂t ∂t
112 6 INSERT TITLE
We take coordinates u, v in a geodesic field in a neighborhood of C(r̂) so that

C is defined by u = 0 and the v-curves are the geodesics perpendicular to C
(see (INSERT REF)). v can be taken to be arc length on C, in fact, so that
v = r on C. The Riemannian metric can be written as
ds2 = ρ21 + ρ22
where
ρ1 = du
ρ2 = Q(u, v)dv ,
for some function Q. Change the sign of ρ2 if necessary to make Q positive.

Let g(r) be a function and consider the variation of C corresponding to g(r).
For each curve t = const. we take frames e1 e2 such that e1 is tangent to
Ct , as before. Let ω1 ω2 be the dual basis. Then, after changing the sign of
ρ1 = du, if necessary, we can write
ω1 = cos ψρ1 + sin ψρ2

ω2 = − sin ψρ1 + cos ψρ2 ,
for some function ψ, since ρ1 , ρ2 is also dual to an orthonormal basis. On

any curve t = const., ω2 = 0, so that
0 = ω2 = − sin ψρ1 + cos ψρ2 .
On u = 0, that is to say, on the geodesic C, we have ρ1 = 0, and since

dr2 = ds2 = ρ22 , it follows that Q = 1. Hence, on C we have
dr = ω1 = sin ψρ2
0 = ω2 = cos ψρ2 .
It follows that on u = 0 sin ψ = 1, cos ψ = 0, so that ψ = π2 . At any point of

C we therefore have ω2 = −ρ1 = −du, so that u decreases in the a1 -direction.
Using this we find that the variation corresponding to g is given by
u = −g(r)t
v=r.
Hence
ρ1 = −g 0 tdr − gdt
ρ2 = Qdr
ω1 = (− cos ψg 0 t + sin ψQ)dr − cos ψgdt
ω2 = (sin ψg 0 t + cos ψQ)dr + sin ψgdt .
Since ω2 = 0 on t = const., we have
sin ψg 0 t + cos ψQ = 0 . (6.33)
From the expression for ω1 we see that
a1 = − cos ψg .
∂a1
Hence ∂t (r, 0) = ψ 0 g. Differentiating (6.33) with respect to t gives, at t = 0,
g0 − ψ0 = 0 .
Hence
∂a1
= gg 0 .
∂t
We apply this last formula to calculate δ 2 L. If
g(r) = f (r) + η(r) ,
then

∂a1 ∂a1 ∂q ∂η ∂η
(r̂−, 0) − (r̂+, 0) = η(r̂) (r̂) + (r̂) − (r̂)
∂t ∂t ∂r ∂r ∂r
∂q
= η(r̂) (r̂) .
∂r
From (6.31), (6.32), and the last formula, we obtain
r !
d2 η
Z
∂q
δ 2 L = 2 (r̂)η(r̂) − η + Kη dr .
∂r 0 dr2
Now η has been so chosen that

∂q
(r̂)η(r̂) < 0 .
∂r
If we choose so that

dq
dr (r̂)η(r̂)
0 < < R r 2
.
d η
0 η dr2 + Kη dr
we have δ 2 L < 0. Hence the length of Ct , t = 6 0, is smaller than that of C0

for all t sufficiently small, that is to say that C0 is a strict relative maximum
in this family. This proves the theorem.
114 6 INSERT TITLE
6.3.5 INSERT TITLE
We next ask whether conjugate points exist.
Theorem 23. a) If the Gaussian curvature K ≤ 0 on a surface M , then M

has no pairs of conjugate points.
b) If K > b > 0, for some constant b, then for every point 0 and every
geodesic C starting at 0, there is a conjugate point r on C to 0 at distance
π
r<√ ,
b
π
provided that C extends to distance √
b
from 0.
Proof. Part a) is a calculus exercise. Suppose that K ≤ 0 and that there

is a geodesic C(r) having two conjugate points along it, 0 and r̂. Since the
conjugate points of 0 along C(r) are isolated, proceeding from 0 along C
toward r̂ there is a first conjugate point of 0 which we take to be r̂. Let a(r)
be a non-zero Jacobi field which vanishes at 0 and r̂. Then a0 (0) 6= 0, since
otherwise a is identically zero. If a0 (0) < 0, we replace a by −a, so we may
assume that a0 (0) > 0. Then, we have a(r) > 0 for r sufficiently close to 0
and therefore a(r) > 0 on (0, r̂). Since a(r̂) = 0 this implies that a0 (r̂) ≤ 0
and therefore that a0 (r̂) < 0. But
a00 = −Ka ≥ 0 on (0, r̂) ,
which says that a0 is non-decreasing on [O, r̂], contradicting the condition

a0 (O) > 0, a0 (r̂) < 0. This proves part a).
Part b) follows from the following, which is quite remarkable in itself.
Theorem 24. The Sturm Comparison Theorem. If K(r), L(r) are continu-
ous functions on [r1 , r2 ] with
K(r) > L(r)
for all r ∈ [r1 , r2 ], and if a, a are functions on [r1 , r2 ] satisfying
a00 + Ka = 0 ,
a00 + La = 0 ,
respectively, and if a(r1 ) = a(r2 ) = 0, a(r) 6= 0 for r ∈ (r1 , r2 ) then a(r) = 0

for some r ∈ (r1 , r2 ).
Proof. Assume that a(r) 6= 0 for r ∈ (r1 , r2 ). Replacing a by −a, a by −a,

as needed, we may assume a, a > 0 on (r1 , r2 ), and the differential equations,
being linear, are still satisfied. Multiplying the left-hand side of the first
differential equation by a, multiplying the left-hand side of the second by a,
subtracting and integrating gives
Z r2
0= (a(a00 + Ka) − a(a00 + La)) dr .
r1
Integrating by parts then gives
0 = a(r2 )a0 (r2 ) − a(r1 )a0 (r1 ) − a(r2 )a0 (r2 )

Z r2
+ a(r1 )a0 (r1 ) + aa(K − L)dr ,
r1
or
Z r2
aa(K − L)dr = a(r2 )a0 (r2 ) − a(r1 )a0 (r1 ) .
r1
Since a(r) > 0 for r ∈ (r1 , r2 ), a(r1 ) = a(r2 ) = 0, we have a0 (r1 ) ≥ 0,

a0 (r2 ) ≤ 0, and hence a0 (r1 ) > 0, a0 (r2 ) < 0, since otherwise a(r) is identically
zero, by the Fundamental Theorem, since it satisfies the linear differential
equation a00 + La = 0. Also, a(r1 ), a(r2 ) ≥ 0. Hence
Z r2
0< aa(K − L)dr = a(r2 )a0 (r2 ) − a(r1 )a0 (r1 ) ≤ 0 ,
r1
which is a contradiction. Hence a(r) = 0 for some r ∈ (r1 , r2 ), which proves

the Sturm Comparison Theorem.
Returning now to the proof of the previous theorem, Part b), we suppose
that M is a surface whose Gaussian curvature satisfies K > b > 0, for b a
constant. Let C(r) be a geodesic parametrized by arc
√ length which starts at
C(0) = 0 and extends at least for r ≤ √πb . Let q = G, where the pull-back
of the Riemannian metric on M under expO is given by
ds2 = dr2 + Gdϕ2 .
Then q is a Jacobi field, which is to say that q satisfies
q 00 + Kq = 0
along C(r). Now let

√
q = sin br .
Then q satisfies
q 00 + bq = 0 .

π
But K > b and q(0) = q = 0, with q(r) 6= 0 for r ∈ 0, √πb . The Sturm
√
b
Comparison Theorem now implies that q(r) = 0, for some r ∈ 0, √πb . Since
q(0) = 0, by (INSERT REF), we have r conjugate to 0, and the proof of Part
b) is complete.
116 6 INSERT TITLE
The theorem we have just proved has interesting topological consequences,

one of which we now wish to describe. It can be shown that if the metric on
M defined in (INSERT REF) has the so-called “completeness” property, then
any geodesic C(r) parametrized by arc length r is defined for all values of
r, which is to say that for any O ∈ M expO (v) is defined for all v ∈ V (O);
secondly, any two points of M are joined by a minimizing geodesic, that is
a geodesic arc whose length is less than or equal to the length of any path
joining the two points. Let us suppose that M is a complete surface whose
Gaussian curvature satisfies K(s) > b > 0, b constant. Then any minimizing
geodesic has length < √πb , since otherwise, by previous theorems, it would
contain a conjugate point of one of the endpoints and would therefore not
minimize the distance between its endpoints. Thus the distance of any point
P to a fixed point O is less than √πb , and from this it can be shown that M
is compact, that is to say that M is a closed surface without boundary.
As always in mathematics, one must consider the theory we have pre-
sented as it is realized in some examples. First of all, consider a unit sphere
and O a point on it. If C(r) is a geodesic through O parametrized by arc
length from O then π is conjugate to O along C(r); in fact C(π) = O0 is
the antipodal point, and if we express the exponential map expO in terms of
polar coordinates (r, ϕ) in V (O), we have expO (π, ϕ) = O0 , for all ϕ. Hence
expO fails to be locally one-to-one at (π, ϕ), so that (π, ϕ) is a singular point
of expO and π is conjugate to O along C(r). By the same reasoning 2π is
conjugate to O along C(r). But C(2π) = 0, so that one might say that O is
conjugate to itself. For rO < π, C is the unique minimizing geodesic joining
O to C(rO ) since M − O0 is a normal neighborhood of O. If π < rO ≤ 2π,
then C(r), 0 ≤ r ≤ r0 , is no longer a minimizing geodesic; the minimizing
geodesic joining O to C(rO ) is now C(−(2π − rO )).
A right circular cylinder of radius 1 provides a second interesting example.
In this case K = 0, so there are no conjugate points. A base circle C(r) is a
geodesic since its geodesic curvature vanishes, as may be seen by drawing the
right picture. If we start at r = 0, then C(r) is a unique minimizing geodesic
until r = π; for r > π it fails to be minimizing, even though there are no
conjugate points. It remains, however, the shortest path among all nearby
paths joining its endpoints, even for r = 2π, when it joins C(O) to itself.
In this section we have covered only some of the most basic aspects of the
geometry of geodesics on a surface. More on the subject can be found in W.
Blaschke, Vorlesungen uber Differentialgeometrie, vol. I, Chapters 6 and 7.
New York: Chelsea, 1967. Blaschke’s approach, which is in certain respects
very attractive, sets up the variation in a way different from ours.

Differential Geometry of Curves and Surfaces, 1st Edition: Thomas Banchoff, Shiing-Shen Chern, and William Pohl

Transféré par

Informations du document

Description originale:

Titre original

Copyright

Formats disponibles

Partager ce document

Partager ou intégrer le document

Options de partage

Avez-vous trouvé ce document utile ?

Ce contenu est-il inapproprié ?

Droits d'auteur :

Formats disponibles

Differential Geometry of Curves and Surfaces, 1st Edition: Thomas Banchoff, Shiing-Shen Chern, and William Pohl

Transféré par

Droits d'auteur :

Formats disponibles

Thomas Banchoff, Shiing-Shen Chern, and

Differential Geometry of Curves

July 17, 2002

1 Review of Euclidean Geometry . . . . . . . . . . . . . . . . . . . . . . . . . . . 3

3 Fundamental Forms of a Surface . . . . . . . . . . . . . . . . . . . . . . . . . . 35

4 Fundamental Equations in Surface Theory . . . . . . . . . . . . . . . . 53

5 Global Surface Theory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71

Note that d(x, y) ≥ 0 and d(x, y) = 0 if and only if x = y.

Taking the sum of the squares gives us

where δjk = 1 if j = k and 0 otherwise.

A basic result in linear algebra states that, for square matrices,

so the product AC is also an orthogonal matrix. For an orthogonal matrix A,

where j = 1, 2, 3 is called an orthogonal transformation. In matrix form, a

Exercise 1. Prove that the quadratic polynomial

Exercise 2. Show that the inverse motion of x0 = Ax + b is x0 = A−1 x −

Exercise 3. Let G be a group with the elements {e, a, b, . . . }, where e is the

respectively. Prove that a−1

Exercise 5. Show that the helicoidal motions

x01 = cos(t)x1 + sin(t)x2

where b is a constant and t is a parameter, form a group. Draw the orbit of

Exercise 6. Prove that the rotation

where i = 1, 2, 3 denote its components. A vector is therefore completely

This leads us to define

v2 = v12 + v22 + v32 . (1.6)

which is to be called the scalar or dot product of v and w.

Relative to addition and scalar multiplication of vectors, the scalar prod-

is true for all λ. So by elementary algebra we get

which is called the Cauchy-Schwartz inequality. The angle θ between v and w

Under the transformation in (1.5) the determinant (u, v, w) is multiplied by

(u + u1 , v, w) = (u, v, w) + (u1 , v, w),

holds for all vectors x. It follows that z has the components

The vector product has the following properties:

The vector z can be described geometrically as follows: We see from (1.9)

Remark 2. The importance of vectors in analytic geometry is due to the alge-

Exercise 7. a) The vector equation of a line is x(t) = at + b, where a, b =

Exercise 8. Prove that x(t) = A cos t + b sin t for A, b 6= 0, represents an

Exercise 9. Let yi , for i = 1, 2, 3, be three linearly independent vectors.

Hence prove that x = 0 if and only if x · yi = 0.

Exercise 10. Let (u, v, w) = 0, u × v 6= 0. Prove that w is a linear com-

Exercise 11. Prove Lagrange’s identity:

Hence prove that

2.1 Arc Length

x(t) = (x1 (t), x2 (t), x3 (t))

that assigns a vector to every value of a parameter t in a domain interval

x0 (t) = (x1 0 (t), x2 0 (t), x3 0 (t)) .

which gives us the distance from a to t along the curve.

We can also write this expression in the form of differentials:

and this may be expressed in the form

which has the advantage that it is independent of the parameter used to

Exercise 13. Find a constant c such that the helix

2.2 Curvature and Fenchel’s Theorem

Theorem 1 (Fenchel’s Theorem). The total curvature of a closed space

0 = f 0 (t) = x0 (t) · m = s0 (t)T(t) · m

x0 (t0 ) · y = x0 (t0 ) · x(t0 )

Exercise 19. Show that the curvature κ of a circular helix

x(t) = (r cos(t), r sin(t), pt)

is equal to the constant value κ = r2|r|

y(t) = x(t) − (x(t) · m)m .

Under what conditions will there be a t0 with y0 (t0 ) = 0?

2.3 The Unit Normal Bundle and Total Twist

E2 (t) = cos(φ(t))E 2 (t) − sin(φ(t))E 3 (t)

E3 (t) = sin(φ(t))E 2 (t) + cos(φ(t))E 3 (t)

Now let > 0, < π2 be given. We choose a neighborhood U of p0 such

τ̃(p) − τ̃(p0 ) = 0 + 2k(p)π , (2.12)