MA4545: Applied Differential Geometry

Lecture 4: Intrinsic Geometry of Surfaces

4.1 Motivation

Recall that the first fundamental form is defined on the tangent plane, i.e., such computation can be made
without “leaving” the surface. We call the concepts can be expressed only on the first fundamental form are
intrinsic to the surface S.
Moreover, as we shall see in this note, many important local properties of a surface can be expressed only
in terms of the first fundamental form. The study of such properties is called the intrinsice geometry of the
surface.

4.2 Isometries

Recall that although the cylinder and the plane are distinct surfaces, their first fundamental forms are
“equal”. In this note, we shall see that many important concepts associated to a regular surface depend only
on the first fundamental form and should be included in the category of intrinsic concepts. It is therefore
convenient that we formulate in a precise way what is meant by two regular surfaces having equal first
fundamental forms.
Notation: S and S̃ will always denote regular surfaces.

Definition 4.1 A diffeomorphism ϕ : S → S̃ is an isometry if for all p ∈ S and all pairs w1 , w2 ∈ Tp (S),
we have
< w1 , w2 >p =< dϕp (w1 ), dϕp (w2 ) >ϕ(p) .
The surfaces S and S̃ are then said to be isometric.

Proposition 4.2 A diffeomorphism ϕ is an isometry if and only if Ip (w) = Iϕ(p) (dϕp (w)) for all w ∈ Tp (S).

Proof: By the definition, a diffeomorphism ϕ is an isometry if the differential dϕp preserves the inner
product. It follows that, dϕ being an isometry,
Ip (w) =< w, w >p =< dϕp (w), dϕp (w) >ϕ(p) = Iϕ(p) (dϕp (w))
for all w ∈ Tp (S). Conversely, if a diffeomorphism ϕ preserves the first fundamental form, that is,
Ip (w) = Iϕ(p) (dϕp (w)) for all w ∈ Tp (S),
then
2 < w1 , w2 >=Ip (w1 + w2 ) − Ip (w1 ) − Ip (w2 )
=Iϕ(p) (dϕp (w1 + w2 )) − Iϕ(p) (dϕp (w1 )) − Iϕ(p) (dϕp (w2 ))
=2 < dϕp (w1 ), dϕp (w2 ) > .
So ϕ is an isometry.

4-1
4-2 Lecture 4: Intrinsic Geometry of Surfaces

Definition 4.3 A map ϕ : V → S̃ of a neighborhood V of p ∈ S is a local isometry at p if there exists a

neighborhood Ṽ of ϕ(p) ∈ S̃ such that ϕ : V → Ṽ is an isometry. If there exists a local isometry into S̃ at
every p ∈ S, the surface S is said to be locally isometric to S̃. S and S̃ are locally isometric if S is locally
isometric to S̃ and S̃ is locally isometric to S.

Remark 4.4 It is clear that if ϕ : S → S̃ of a diffeomorphism and a local isometry for every p ∈ S is said
to be an isometry (globally).

However, it may happen that two surfaces are locally isometric without being (globally) isometric, as shown
in the following example.
Example 1. Let ϕ be a map of the coordinate neighborhood of x̃(U ) of the cylinder over x2 + y 2 = 1 by

x̃(u, v) = (cos u, sin u, v), U = {(u, v) ∈ R2 : 0 < u < 2π, −∞ < v < ∞},

into the plane x(R2 ) by

x(u, v) = p0 + uw1 + vw2 , (u, v) ∈ R2 ,
defined by ϕ = x ◦ x̃−1 .
Then ϕ is a local isometry. In fact, each vector w, tangent to the cylinder at a point p ∈ x̃(U ), is tangent to
a curve x̃(u(t), v(t)), where (u(t), v(t)) is a curve in U ⊂ R2 . Thus, w can be written as

w = x̃u u0 + x̃v v 0 .

On the other hand, dϕ(w) is tangent to the curve

ϕ(x̃(u(t), v(t))) = x(u(t), v(t)).

Thus, dϕ(w) = xu u0 + xv v 0 . Since E = Ẽ, F = F̃ , G = G̃, we obtain

Ip (w) = Ẽ(u0 )2 + 2F̃ u0 v 0 + G̃(v 0 )2 = E(u0 )2 + 2F u0 v 0 + G(v 0 )2 = Iϕ(p) (dϕp (w)).

It follows that the cylinder is locally isometric to a plane.

However, the isometry cannot be extended to the entire cylinder beause the cylinder is not enven homeo-
morphic to a plane. A rigorous proof of this asertion would take us too far afield (see a textbook on the
Topology for it).

Proposition 4.5 Assume the existence of parametrizations x : U → S and x̃ : U → S̃ such that E = Ẽ,
F = F̃ , G = G̃ in U . Then the map ϕ = x̃ ◦ x−1 : x(U ) → S̃ is a local isometry.

Proof: Let p ∈ x(U ) and w ∈ Tp (S). Then w is tangent to a curve x(α(t)) at t = 0, where α(t) = (u(t), v(t))
is a curve in U . Thus w may be written at t = 0 as

w = xu u0 + xv v 0 .

By definition, the vector dϕp (w) is the tangent vector to the curve ϕ(x(α(t))) = x̃ ◦ x−1 ◦ x ◦ α = x̃(α(t)) at
t = 0. Thus,
dϕp (w) = x̃u0 + x̃v v 0 .
Therefore,

Ip (w) = E(u0 )2 + 2F u0 v 0 + G(v 0 )2 = Ẽ(u0 )2 + 2F̃ u0 v 0 + G̃(v 0 )2 = Iϕ(p) (dϕp (w)).

Lecture 4: Intrinsic Geometry of Surfaces 4-3

Figure 4.1: cone

We conclude that Ip (w) = Iϕ(p) (dϕp (w)) for all p ∈ x(U ) and all w ∈ Tp (S). hence, ϕ is a local isometry.
Example 2. We shall prove that the one-sheeted cone (minus the vertex)
p
z = k x2 + y 2 , 6 (0, 0),
(x, y) =

is locally isometric to a plane.

Idea: A cone minus a generator can be “rolled” onto a piece of a plane (see Fig. 4.1).
Solution: Let sector U ⊂ R2 be the open set given in polar coordinates (ρ, θ) by

0 < ρ < ∞, 0 < θ < 2π sin α,

where 2α (0 < 2α < π) is the angle at the vertex of the cone (i.e., where cot α = k), and let f : U → R3 be
the map      
θ θ
f (ρ, θ) = ρ sin α cos , ρ sin α sin , ρ cos α .
sin α sin α
It is clear that f (U ) is contained in the cone because
p q
k x2 + y 2 = cot α ρ2 sin2 α = ρ cos α = z.

Furthermore, when θ describes the interval (0, 2π sin α), θ/ sin α describes the interval (0, 2π). Thus all points
of the cone except the generator θ = 0 are covered by f (U ).
It is easy to checked that f and df are one-to-one in U . Therefore, f is a diffeomorphism of U onto the cone
minus a generator.
We shall now show that f is an isometry. In fact, U may be thought of as a regular surface, parametrized
by
x̃(ρ, θ) = (ρ cos θ, ρ sin θ, 0), 0 < ρ < ∞, 0 < θ < 2π sin α.
The coefficients of the first fundamental form of U in this parametrization are

Ẽ = 1 F̃ = 0 G̃ = ρ2 .

On the other hand, the coefficients of the first fundamental form of the cone in the parametrization f ◦ x̃ are

E=1 F =0 G = ρ2 .

Form Proposition 4.5, we conclude that F is a local isometry, as we wished.

4-4 Lecture 4: Intrinsic Geometry of Surfaces

4.3 Gauss Theorem and Equations of Compatibility

S will denote a regular, orientable and oriented surface. Let x : U ⊂ R2 → S be a parametrization in the
orientation of S. It is possible to assign to each point of x(U ) a natural trihedron given by the vectors xu ,
xv , and N . The study of this trihedron will be the subject of this section.
By expressing the derivatives of the vectors xu , xv , and N in the basis {xu , xv , N }, we obtain

xuu = Γ111 xu + Γ211 xv + L1 N,

xuv = Γ112 xu + Γ212 xv + L2 N,
xvu = Γ121 xu + Γ221 xv + L̃2 N,
(4.1)
xvv = Γ122 xu + Γ222 xv + L3 N,
Nu = a11 xu + a21 xv ,
Nv = a12 xu + a22 xv ,

where aij , i, j = 1, 2 were obtained before as

    −1  
a11 a21 e f E F 1 f F − eG eF − f E
=− = , (4.2)
a12 a22 f g F G EG − F 2 gF − f G f F − gE

and the other coefficients are to be determined below.

Remark 4.6 The coefficients Γkij , i, j, k = 1, 2 are called the Christoffel symbols of S in the parametrization
x.

Since xuv = xvu , we conclude that (by linearly independence)

Γ112 = Γ121 , Γ212 = Γ221 , L2 = L̃2 . (4.3)

That is, the Christoffel symbols are symmetric relative to the lower indices.
By taking the inner product of the first four relations in (4.1) with N , we immediately obtain

L1 = e, L2 = L̃2 = f, L3 = g, (4.4)

where e, f, g are the coefficients of the second fundamental form of S.

To determine the Christoffel symbols, we take the inner product of the first four relations with xu and xv ,
obtaining the system (
Γ111 E + Γ211 F =< xuu , xu >= 21 Eu ,
Γ1 F + Γ211 G =< xuu , xv >= Fu − 21 Ev ,
( 11
Γ112 E + Γ212 F =< xuv , xu >= 21 Ev ,
(4.5)
Γ1 F + Γ212 G =< xuv , xv >= 21 Gu ,
( 12
Γ122 E + Γ222 F =< xvv , xu >= Fv − 12 Gu ,
Γ122 F + Γ222 G =< xvv , xv >= 21 Gv .
Note that the above equations have been grouped into three pairs of equations and that for each pair, the
determinant of the system is EG − F 2 6= 0. Thus

Remark 4.7 It is possible to solve (4.5) and to compute the Christoffel symbols in terms of the coefficients
of the first dundamental form, E, F , G, and their derivatives.
Lecture 4: Intrinsic Geometry of Surfaces 4-5

Moreover, it follows from the remark above that

Proposition 4.8 All geometric concepts and properties expressed in terms of the Christoffel symbols are
invariant under isometries.

Example 3. We shall compute the Christoffel symbols for a surface of revolution parametrized by
x(u, v) = (f (v) cos u, f (v) sin u, g(v)), f (v) 6= 0.
Solution. Since E = (f (v))2 , F = 0, G = (f 0 (v))2 + (g 0 (v))2 , we obtain
Eu = 0, Ev = 2f f 0 , Fu = Fv = 0, Gu = 0, Gv = 2(f 0 f 00 + g 0 g 00 ),
where prime denotes derivative with respect to v. Then it follows from (4.5) that
ff0 ff0 f 0 f 00 + g 0 g 00
Γ111 = 0, Γ211 = − , Γ112 = , Γ212 = 0, Γ122 = 0, Γ222 = .
(f 0 )2+ (g 0 )2 f2 (f 0 )2 + (g 0 )2

Now, we are going to obtain relations between these coefficients. Consider the expressions
(xuu )v − (xuv )u = 0, (xvv )u − (xuv )v = 0, Nuv − Nvu = 0. (4.6)
Plugging (4.1) into (4.6), we may write the above relations in the form
A1 xu + B1 xv + C1 N = 0
A2 xu + B2 xv + C2 N = 0 (4.7)
A3 xu + B3 xv + C3 N = 0
where Ai , Bi , Ci , i = 1, 2, 3 are functions of E, F , G, e, f , g and of their derivatives. Since the vectors xu ,
xv , N are linearly independent, (4.7) implies that there exist nine relations:
Ai = 0, Bi = 0, Ci − 0, i = 1, 2, 3.
They are called the equations of compatibility.
As an example, plugging (4.1) into the first of the relations (4.6), we have
Γ111 xuv + Γ211 xvv + eNv + (Γ111 )v xu + (Γ211 )v xv + ev N = Γ112 xuu + Γ212 xvu + f Nu + (Γ112 )u xu + (Γ212 )u xv + fu N.
(4.8)
By using (4.1) again and equating the coefficients of xv , we obtain
Γ111 Γ212 + Γ211 Γ222 + ea22 + (Γ211 )v = Γ112 Γ211 + Γ212 Γ212 + f a21 + (Γ212 )u .
By (4.2), it follows that
eg − f 2
(Γ212 )u − (Γ211 )v + Γ112 Γ211 + Γ212 Γ212 − Γ211 Γ222 − Γ111 Γ212 = −E = −EK. (4.9)
EG − F 2

Remark 4.9 Expression (4.9), which yields the value of K in terms of the coefficients of the first funda-
mental form and its derivatives, is known as the Gauss formula. It was first proved by Gauss.

Theorem 4.10 (Gauss) The Gaussian curvature K of a surface is invariant by local isometries.

Proof: If x : U ⊂ R2 → S is a parametrization at p ∈ S and if ϕ : V ⊂ S → S̃, where V ⊂ x(U ) is

a neighborhood of p, is a local isometry at p, then y = ϕ ◦ x is a parametrization of S̃ at ϕ(p). Since
ϕ is an isometry, the coefficients of the first fundamental form in the parametrizations x and y agree at
corresponding points q and ϕ(q), q ∈ V . Thus the corresponding Christoffel symbols also agree. By (4.9), K
can be computed at a point as a function of the Christoffel symbols in a given parametrization at the point.
It follows K(q) = K(ϕ(q)) for all q ∈ V .
4-6 Lecture 4: Intrinsic Geometry of Surfaces

Figure 4.2: covariant derivative

4.4 Parallel Transport. Geodesics.

We shall now proceed to a systematic exposition of the intrinsic geometry. We shall start with the definition
of covariant derivative of a vector field, which is the analogue for surfaces of the usual differentiation of
vectors in the plane. We recall that a (tangent) vector field in an open set U ⊂ S of a regular surface S is a
correspondence w that assigns to each p ∈ U a vector w(p) ∈ Tp (S). The vector field w is differentiable at
p if, for some parametrization x(u, v), the components a and b of w = axu + bxv in the basis {xu , xv } are
differentiable functions at p. The vector field w is differentiable in U if it is differentiable for every p ∈ U .

Definition 4.11 (Covariant derivative, see Fig. 4.2) Let w be a differentiable vector field in an open
set U ⊂ S and p ∈ U . Let y ∈ Tp (S). Consider a parametrized curve

α : (−ε, ε) → U,

with α(0) = p and α0 (0) = y, and let w(t), t ∈ (−ε, ε), be the restriction of the vector field w to the curve α.
The vector obtained by the normal projection of dw dt (0) onto the plane Tp (S) is called the covariant derivative
at p of the vector field w relative to the vector y. This covariant derivative is denoted by Dw dt (0) or (Dy w)(p).

The above definition makes use of the normal vector of S and of a particular curve α, tangent to y at p. But
we have the following proposition.

Proposition 4.12 Dw dt , defined in Definition 4.11, depends only on the vector y and not on the curve α.
Furthermore, the covariant differentiation is a concept of the intrinsic geometry.

Proof: Let x(u(t), v(t)) = α(t) be the expression of the curve α and let

w(t) = a(u(t), v(t))xu + b(u(t), v(t))xv = a(t)xu + b(t)xv ,

be the expression of w(t) in the parametrization x(u, v). Then

dw
= a(xuu u0 + xuv v 0 ) + b(xvu u0 + xvv v 0 ) + a0 xu + b0 xv ,
dt
where prime denotes the derivative with respect to t.
Lecture 4: Intrinsic Geometry of Surfaces 4-7

Since Dw/dt is the component of dw/dt in the tangent plane, we use the expressions in (4.1) for xuu , xuv ,
and xvv and, dropping the normal component (i.e., N), we obtain
Dw
= (a0 + Γ111 au0 + Γ112 av 0 + Γ112 bu0 + Γ122 bv 0 )xu + (b0 + Γ211 au0 + Γ212 av 0 + Γ212 bu0 + Γ222 bv 0 )xv . (4.10)
dt
Expression (4.10) shows that Dw/dt depends only on the vector (u0 , v 0 ) = y and not on the curve α.
Furthermore, the covariant differentiation given in (4.10) is through the Christoffel symbols, that is, through
the first fundamental form. Therefore, the proposition is proved.

Remark 4.13 In particular, if S is a plane, then it is possible to find a parametrization in such a way that
E = G = 1 and F = 0. A quick inspection of the equations that give the Christoffel symbols shows that in
0 0
this case, the Γkij become zero. In this case, it follows from (4.10) that Dw
dt = a xu + b xv , i.e., the covariant
derivative agrees with the usual derivative of vectors in the plane. Therefore, the covariant derivative is a
generalization of the usual derivative of vectors in the plane.

Another consequence of equation (4.10) is that the definition of covariant derivative may be extended to a
vector field which is defined only at the points of a parametrized curve.

Definition 4.14 A parametrized curve α : [0, l] → S is the restriction to [0, l] of a differentiable mapping
of (0 − ε, l + ε), ε > 0, into S. If α(0) = p and α(l) = q, we say that α joins p to q.

Notation. In what follows it will be convenient to use the notation [0, l] = I whenever the specification of
the end point l is not necessary.

Definition 4.15 Let α : I → S be a parametrized curve in S. A vector field w along α is a correspondence

that assigns to each t ∈ I a vector
w(t) ∈ Tα(t) (S).
The vector field w is differentiable at t0 ∈ I if for some parametrization x(u, v) in α(t0 ) the components a(t),
b(t) of w(t) = axu + bxv are differentiable functions of t at t0 . w is differentiable in I, if it is differentiable
for every t ∈ I.

Dw
Definition 4.16 Let w be a differentiable vector field along α : I → S. The expression (4.10) of dt (t),
t ∈ I, is well defined and is called the covariant derivative of w at t.

Remark 4.17 If α(t) is a curve on S, we can think of it as the trajectory of a point which is moving on the
surface. α0 (t) is then the speed and α00 the acceleration of α. The covariant derivative Dα0 /dt of the field
α0 (t) is the tangential component of the acceleration α00 (t). Intuitively, Dα0 (t)/dt is the acceleration of the
point α(t) “as seen from the surface S”.

Definition 4.18 A vector field w along a parametrized curve α : I → S is said to be parallel if Dw/dt = 0
for every t ∈ I.

Remark 4.19 In the particular case of the plane, the notion of parallel field along a parametrized curve
reduces to that of a constant field along the curve, because dw/dt = Dw/dt = 0. So in this case, the length
of the vector and its angle with a fixed direction are constant.

These properties are partially reobtained on any surface as the following properties shows.
4-8 Lecture 4: Intrinsic Geometry of Surfaces

Proposition 4.20 Let w and v be parallel vector fields along α : I → S. Then < w(t), v(t) > is constant.
In particular, |w(t)| and |v(t)| are constant, and the angle between v(t) and w(t) is constant.

Proof: Dw/dt = 0 along α means that dw/dt is normal to the plane which is tangent to the surface at α.
That is (because w and v are in the tangent plane Tα(t) (S)),

< v(t), w0 (t) >= 0, t ∈ I.

Similarly, we also have < v 0 (t), w(t) >= 0 for t ∈ I. Thus

< v(t), w(t) >0 =< v 0 (t), w(t) > + < v(t), w0 (t) >= 0.

That is, < v(t), w(t) >= constant.

The following proposition shows that there exist parallel vector fields along a parametrized curve α(t) and
that they are completely determined by their values at a point t0 .

Proposition 4.21 Let α : I → S be a prametrized curve in S and let w0 ∈ Tα(t0 ) (S), t0 ∈ I. Then there
exists a unique parallel vector field w(t) along α(t), with w(t0 ) = w0 .

The proof of Proposition 4.21 will be given later in this section as an application of Proposition 4.40.
Proposition 4.21 allows us to talk about parallel transport of a vector along a parametrized curve.

Definition 4.22 Let α : I → S be a parametrized curve and w0 ∈ Tα(t0 ) (S), t0 ∈ I. Let w be the parallel
vector field along α, with w(t0 ) = w0 . The vector w(t1 ), t1 ∈ I, is called the parallel transport of w0 along
α at the point t1 .

Remark 4.23 If α : I → S, t ∈ I, is regular, then the parallel transport does not depend on the parametriza-
tion of α(I).

Proof: If β : J → S, σ ∈ J is another regular parametrization for α(I), it follows from equation (4.10) that
Dw Dw dt
= , t ∈ I, σ ∈ J.
dσ dt dσ
dt
Since dσ 6= 0 (regular), w(t) is parallel if and only if w(σ) is parallel.

Remark 4.24 Fix two points p, q ∈ S and a parametrized curve α : I → S with α(0) = p, α(l) = q.
Denote by Pα : Tp (S) → Tq (S) the map that assigns to each v ∈ TP (S) its parallel transport along α at q.
Proposition 4.20 says that this map is a linear isometry.

The parametrized curves γ : I → R2 of a plane along which the field of their tangent vectors γ 0 (t) is parallel
are preciesely the straight lines of that plane. The parametrized curves that satisfy an analogous condition
for a surface are called geodesics.

Definition 4.25 A nonconstant, parametrized curve γ : I → S is said to be geodesic at t ∈ I if the field of

its tangent vectors γ 0 (t) is parallel along γ at t. That is,
Dγ 0 (t)
= 0.
dt
γ is a parametrized geodesic if it is geodesic for all t ∈ I.
Lecture 4: Intrinsic Geometry of Surfaces 4-9

Remark 4.26 By Proposition 4.20, we obtain immediately that |γ 0 (t)| = const. = c 6= 0. Therefore, we
may introduce the arc length s = ct as a parameter, and we conclude that the parameter t of a parametrized
geodesic γ is proportional to the arc length of γ.

Remark 4.27 The tangent vector of a parametrized geodesic is never zero, so the parametrization is regular.

Definition 4.28 A regular connected curve C in S is said to be a geodesic if, for every p ∈ C, the
parametrization α(s) of a coordinate neighborhood of p by the arc length s is a parametrized geodesic. That
is, α0 (s) is a parallel vector field along α(s).

Example. Every straight line contained in a surface satisfies Defintion 4.28.

Remark 4.29 From a point of view exterior to the surface S, a regular curve C ⊂ S (k 6= 0) is a geodesic
if and only if its principal normal at each point p ∈ C is parallel to the normal to S at p. The reason is that
Definition 4.28 is equivalent to saying that α00 (s) = kn is normal to the tangent plane.

Example 4. The great circles of a sphere S 2 are geodesics.

Solution. The great circles C are obtained by intersecting the sphere with a plane that passes through the
center O of the sphere. The principal normal at a point p ∈ C lies in the direction of the line that connects
p to O because C is a circle of center O. Since S 2 is a sphere, the normal lies in the same direction, which
verifies our assertion by Remark 4.29.

Definition 4.30 Let w be a differentiable field of unit vectors along a parametrized curve α : I → S on an
oriented surface S. Since w(t), t ∈ I, is a unit vector field, dw
dt (t) is normal to w(t), and therefore

Dw
= λ(N ∧ w(t)).
dt
The real number λ = λ(t), denoted by[Dw/dt], is called the algebraic value of the covariant derivative of w
at t.

Remark 4.31 The sign of [Dw/dt] depends on the orientation of S and that

Dw dw
[ ] =< ,N ∧ w > . (4.11)
dt dt

The proof is left as an exercise.

Definition 4.32 Let C be a regular curve contained in an oriented surface S, and let α(s) be a parametriza-
tion of C, in a neighborhood of p ∈ S, by the arc length s. The algebraic value of the covariant derivative
kg = [Dα0 (s)/ds] of α0 (s) at p is called the geodesic curvature of C at p.

It follows from (4.11) that kg =< α00 , N ∧ α0 >.

Remark 4.33 The geodesics which are regular curves are thus characterized as curves whose geodesic cur-
vature is zero.
4-10 Lecture 4: Intrinsic Geometry of Surfaces

Remark 4.34 From a point of view external to the surface, the absolute value of the geodesic curvature kg
of C at p is the absolute value of the tangential component of the vector α00 (s) = kn, where k is the curvature
of C at p and n is the normal vector of C at p. Recall that the absolute value of the normal component of
the vector kn is the absolute value of the normal curvature kn of C ⊂ S in p. Hence, we have
k 2 = kg2 + kn2 . (4.12)

Remark 4.35 The geodesic curvature of C ⊂ S changes sign when we change the orientation of either C
or S.

We shall now obtain an expression for the algebraic value of the covariant derivative (Proposition 4.40 below).
For that, we need some preliminaries.
Let v and w be two differentiable vector fields along the parametrized curve α : I → S, with |v(t)| = |w(t)| =
1, t ∈ I.
We want to define a differentiable function ϕ : I → R in such a way that ϕ(t), t ∈ I, is a determination of
the angle from v(t) to w(t) in the orientation of S. To do that, we consider the differentiable vector field ṽ(t)
along α, defined by the condition that {v(t), ṽ(t)} is an orthonormal positive basis for every t ∈ I. Thus,
w(t) may be expressed as
w(t) = a(t)v(t) + b(t)ṽ(t),
where a and b are differentiable functions in I and a2 + b2 = 1.
Lemma below shows that by fixing a determination ϕ0 of the angle from v(t0 ) to w(t0 ), it is possible to
“extend it” differentiably in I, and this yields the desired function.

Lemma 4.36 Let a and b be differentiable functions in I with a2 +b2 = 1 and ϕ0 be such that a(t0 ) = cos ϕ0 ,
b(t0 ) = sin ϕ0 . Then the differentiable function
Z t
ϕ = ϕ0 + (ab0 − ba0 )dt
t0

is such that cos ϕ(t) = a(t), sin ϕ(t) = b(t), t ∈ I, and ϕ(t0 ) = ϕ0 .

Proof: It suffices to show that

A := a cos ϕ + b sin ϕ = 1.
Then we know
(a − cos ϕ)2 + (b − sin ϕ)2 = 2 − 2(a cos ϕ + b sin ϕ) = 0.
By using the fact that aa0 + bb0 = 0 and the definition of ϕ (i.e., ϕ0 = ab0 − ba0 ), we easily obtain
A0 = −a(sin ϕ)ϕ0 +b(cos ϕ)ϕ0 +a0 cos ϕ+b0 sin ϕ = −b0 (sin ϕ)(a2 +b2 )−a0 (cos ϕ)(a2 +b2 )+a0 cos ϕ+b0 sin ϕ = 0.
Therefore, A(t) = const. = A(t0 ) = 1. So the lemma is proved.
We may now relate the covariant derivative of two unit vector fields along a curve to the variation of the
angle that they form.

Lemma 4.37 Let v and w be two differentiable vector fields along the curve α : I → S, with |w(t)| =
|v(t)| = 1, t ∈ I. Then    
Dw Dv dϕ
− = ,
dt dt dt
where ϕ is one of the differentiable determinations of the angle from v to w as given in Lemma 4.36.
Lecture 4: Intrinsic Geometry of Surfaces 4-11

Proof: We first prove the lemma for ϕ 6= 0. Since < v, w >= cos ϕ, we obtain

< v 0 , w > + < v, w0 >= − sin ϕϕ0 .

Hence
Dv Dw
< , w > + < v, >= − sin ϕϕ0 .
dt dt
Dv
Note that dt = [ Dv
dt ](N ∧ v), that is, <
Dv
dt , w >= [ Dv
dt ] < N ∧ v, w >. Then
       
Dv Dw Dv Dw Dw Dv
< , w > + < v, >= < N ∧ v, w > + < N ∧ w, v >= ( − ) < N ∧ w, v > .
dt dt dt dt dt dt

Because v ∧ w = sin ϕN , < v, N ∧ w >= − sin ϕ. Then it follows that

   
Dw Dv
( − )(− sin ϕ) = − sin ϕϕ0
dt dt

and this proves the lemma for ϕ 6= 0.

If ϕ = 0 at p, either ϕ ≡ 0 in a neighborhood U of p, or there exists a sequence (pn ) → p with ϕ(pn ) 6= 0.
In the first case, ϕ0 ≡ 0 in U , v = w and the lemma holds trivially. In the second case, the lemma holds by
continuity.

Corollary 4.38 Geodesic curvature is the rate of change of the angle that the tangent to the curve makes
with a parallel direction along the curve.

Proof: Let C be a regular oriented curve on S, α(s) a parametrization by the arc length s of C at p ∈ C,
and v(s) a parallel field along α(s). Then, by taking w(s) = α0 (s), we obtain

Dα0 (s)
 
dϕ
kg (s) = = .
ds ds

Remark 4.39 In the case of the plane, the parallel direction is fixed and the geodesic curvature reduces to
the usual curvature.

We are now able to obtain the promised expression for the algebraic value of the covariant derivative.

Proposition 4.40 Let x(u, v) be an orthogonal parametrization (that is, F = 0) of a neighborhood of an

oriented surface S, and w(t) be a differentiable field of unit vectors along the curve x(u(t), v(t)). Then
   
Dw 1 dv du dϕ
= √ Gu − Ev + ,
dt 2 EG dt dt dt

where ϕ(t) is the angle from xu to w(t) in the given orientation.

√ √
Proof: Let e1 = xu / E, e2 = xv / G be the unit vectors tangent to the coordinate curves. Observe that
e1 ∧ e2 = N , where N is the given orientation of S. By Lemma 4.37, we have
   
Dw De1 dϕ
= + ,
dt dt dt
4-12 Lecture 4: Intrinsic Geometry of Surfaces

where e1 (t) = e1 (u(t), v(t)) is the field e1 restricted to the curve x(u(t), v(t)). Now
 
De1 de1 de1 du dv
=< , N ∧ e1 >=< , e2 >=< (e1 )u , e2 > + < (e1 )v , e2 > .
dt dt dt dt dt
On the other hand, since F = 0, we have
1
< xuu , xv >= − Ev ,
2
and therefore  
x xv 1 Ev
< (e1 )u , e2 >=< √u , √ >= − √ .
E u G 2 EG
Similarly,
1 Gu
< (e1 )v , e2 >= √ .
2 EG
By introducing these relations in the expression of [Dw/dt], we finally complete the proof.
As an application of Proposition 4.40, we shall prove the existence and uniqueness of the parallel transport,
i.e., Proposition 4.21.
Proof:[Proof of Proposition 4.21] Let us assume initially that the parametrized curve α : I → S is con-
tained in a coordinate neighborhood of an orthogonal parametrization x(u, v). Then, with the notations of
Proposition 4.40, the condition of parallelism for the field w becomes
 
dϕ 1 dv du
=− √ Gu − Ev := B(t).
dt 2 EG dt dt
Denoting by ϕ0 a determination of the oriented angle from xu to w0 , the field w is entirely determined by
Z t
ϕ = ϕ0 + B(t)dt,
0

which proves the existence and uniqueness of w in this case.

If α(I) is not contained in a coordinate neighborhood, we shall use the compactness of I to divide α(I) into
a finite number of parts, each contained in a coordinate neighborhood. By using the uniqueness of the first
part of the proof in the nonempty intersections of these pieces, it is easy to extend the result to the present
case.
A further application of Proposition 4.40 is the following expression for the geodesic curvature, known as
Liouville’s formula.

Proposition 4.41 (Liouville) Let α(s) be a parametrization by arc length of a neighborhood of a point
p ∈ S of a regular curve C on an oriented surface S. Let x(u, v) be an orthogonal parametrization of S in p
and ϕ(s) be the angle that xu makes with α0 (s) in the given orientation. Then
dϕ
kg = (kg )1 cos ϕ + (kg )2 sin ϕ + ,
ds
where (kg )1 and (kg )2 are the geodesic curvature of the coordinate curves v = const. and u = const.,
respectively.

Proof: By setting w = α0 (s) in Proposition 4.40, we obtain

 
1 dv du dϕ
kg = √ Gu − Ev + .
2 EG ds ds ds
Lecture 4: Intrinsic Geometry of Surfaces 4-13

√
Along the coordinate curve v = const. u = u(s), we have dv/ds = 0 and du/ds = 1/ E. Therefore,
Ev
(kg )1 = − √ .
2E G
Similarly,
Gu
√ .
(kg )2 =
2G E
By introducing these relations in the above formula for kg , we obtain
√ du √ dv dϕ
kg = (kg )1 E + (kg )2 G + .
ds ds ds
Since
√ du xu √ dv xv
E =< α0 (s), √ >= cos ϕ and G =< α0 (s), √ >= sin ϕ,
ds E ds G
we finally arrive at
dϕ
kg = (kg )1 cos ϕ + (kg )2 sin ϕ + ,
ds
as we wished.

4.5 Gauss-Bonnet Theorem (Local Version)

The Gauss-Bonnet theorem is probably the deepest theorem in the differential geometry of surfaces. We
shall now begin the details of a local version of the Gauss-Bonnet thereom.

Definition 4.42 Let α : [0, l] → S be a continuous map from the closed interval [0, l] into the regular surface
S. We say that α is a simple, closed, piecewise regular, parametrized curve if

1. α(0) = α(l).
2. t1 6= t2 , t1 , t2 ∈ [0, l), implies that α(t1 ) 6= α(t2 ).
3. There exists a subdivision
0 = t0 < t1 < · · · < tk < tk+1 = l,
of (0, l] such that α is differentiable and regular in each (ti , ti+1 ), i = 0, · · · , k.

The points α(ti ), i = 0, · · · , k, are called the vertices of α and the traces α([ti , ti+1 ]) are called the regular
arcs of α. It is usual to call the trace α([0, l]) of α, a closed piecewise regular curve.

Intuitively, this means that α is a closed curve (condition 1) without self-intersections (condition 2), which
fails to have a well-defined tangent line only at a finite number of points (condition 3).
External angle at the vertices (see Fig. 4.3). By the condition of regularity, for each vertex α(ti ), there
exist the limit from tile left, i.e., for t < ti ,

lim α0 (t) = α0 (ti − 0) 6= 0,

t→ti

and the limit from the right, i.e., for t > ti ,

lim α0 (t) = α0 (ti + 0) 6= 0.

t→ti
4-14 Lecture 4: Intrinsic Geometry of Surfaces

Figure 4.3: External angle at the vertices

Figure 4.4: Angle at a cusp

Assume now that S is oriented and let |θi |, 0 ≤ |θi | < π, be the smallest determination of the angle from
α0 (ti − 0) to α0 (ti + 0). If |θi | =
6 π, we give θi the sign of the determinant (α0 (ti − 0), α0 (t0 + 0), N ). This
means that if the vertex α(ti ) is not a ”cusp”, the sign of θi is given by the orientation of S. The signed
angle θi , −π < θi < π is called the external angle at the vertex α(ti ).
Angle at a cusp (see Fig. 4.4). In the case that α(ti ) is a cusp, i.e., |θi | = π, we choose the sign of
θi as follows. Let the closed simple curve α be contained in the image of a parametrization with a given
orientation, and assume that α(ti ) is a cusp. Choose coordinate axis xOy, with α(ti ) = O, in the given
orientation, and assume further that the part of α arrving at α(ti ) is pointing towards the negative part of
the axis Ox (and, of course, the part of α leaving α(ti )) is pointing to the positive part of the axis Ox).
For small ε > 0, the part of α arriving at α(ti ), and near it, is given by a function f (x) = y, 0 < x < ε, and
the part of α leaving ti , and near it, is given by a function f (x) = y, 0 < x < ε, and the part of α leaving
ti , and near it, is given by a function g(x) = y, 0 < x < ε. If f > 0 and g < 0, we set θ(ti ) = π and if f < 0
and g > 0, we set θ(ti ) = −π.

Let α : [0, l] → x(U ) ⊂ S be a simple closed, piecewise regular, parametrized curve, with vertices α(ti ) and
external angles θi , i = 0, · · · , k.
Lecture 4: Intrinsic Geometry of Surfaces 4-15

Let ϕi : [ti , ti+1 ] → R be differentiable functions which measure at each t ∈ [ti , ti+1 ] the positive angle from
xu to α0 (t) (see Lemma 4.36).
The first topology fact that we shall present without proofs is the following.

Theorem 4.43 (Turning Tangents) With the above notation, we have for plane curves

k
X k
X
(ϕi (ti+1 ) − ϕi (ti )) + θi = ±π,
i=0 i=0

where the sign plus or minus depends on the orientation of α.

Before stating the local version of the Gauss-Bonnet theorem, we still need some terminology.
Let S be an oriented surface. Let x : U ⊂ R2 → S be a parametrization of S and let R ⊂ x(U ) be a bounded
region of S.

ˆ A region R ⊂ S (union of a connected open set with its boundary) is called a simple region if R is
homeomorphic to a disk and the boundary ∂R of R is the trace of a simple, closed, piecewise regular,
parametrized curve α : I → S.

ˆ We say then that α is positively oriented if for each α(t), belonging to a regular arc, the positive
orthogonal basis {α0 (t), h(t)} satisfies the condition that h(t) ”points toward” R. More precisely, for
any curve β : I → R with β(0) = α(t) and β 0 (0) = α0 (t), we have that hβ 0 (0), h(t)i > 0.

ˆ If f is a differentiable function on S, then it is easily to see that the integral

ZZ p
f (u, v) EG − F 2 dudv
x−1 (R)

does not depend on the parametrization x, chosen in the class of orientation of x (Exercise). This
integral has, therefore, a geometrical meaning and is called the integral of f over the region R. It is
usual to denote it by
ZZ
f dσ.
R

With these definitions, we now state the

Theorem 4.44 (Gauss-Bonnet Theorem (local)) Let x : U → S be an isothermal parametrization

( i.e., F = 0, E = G = λ2 (u, v)) of an oriented surface S, where U ⊂ R2 is homeomorphic to an open disk
and x is compatible with the orientation of S.
Let R ⊂ x(U ) be a simple region of S and let α : I → S be such that ∂R = α(I). Assume that α is positively
oriented, parametrized by arc length s, and let α(s0 ), · · · , α(sk ) and θ0 , · · · , θk be, respectively, the vertices
and the external angles of α. Then

k Z
X si+1 ZZ k
X
kg (s)ds + Kdσ + θi = 2π,
i=0 si R i=0

where kg (s) is the geodesic curvature of the regular arcs of α and K is the Gaussian curvature of S.
4-16 Lecture 4: Intrinsic Geometry of Surfaces

Remark 4.45 A first version of this theorem was presented by Gauss deals with geodesic triangles on surfaces
(that is, triangles whose sides are arcs of geodesics). That is
3
X ZZ
ϕi − π = Kdσ,
i=1 T

where ϕ1 , ϕ2 , ϕ3 are the interior angles of a geodesic triangle T (because ϕi = π − θi and kg = 0).
P
For instance, if k ≡ 0, we obtain ϕi = π, an extension P of Thales’ theorem of high school geometry to
surfaces of zero curvature. Also, if K ≡ 1, we obtain that ϕi − π = area(T ) > 0. Thus, on a unit sphere,
the sum of the interior angles of any geodesic triangle is greater than π, and the excess over π is exactly the
area of T .

Proof:(Proof of Gauss-Bonnet Theorem (local)). Let u = u(s), v = v(s) be the expression of α in the
parametrization x. By Proposition 4.40, we have
 
1 dv du dϕi
kg (s) = √ Gu − Ev + ,
2 EG ds ds ds
where ϕi = ϕi (s) is a differentiable function which measures the positive angle from xu to α0 (s) in {si , si+1 }.
By integrating the above expression in every interval [si , si+1 ] and adding up the results,
k Z si+1 k Z si+1   k Z si+1
G dv Ev du dϕi
√u
X X X
kg (s)ds = − √ ds + ds.
i=0 si i=0 si 2 EG ds 2 EG ds i=0 si ds

Now we use the Gauss-Green theorem in the uv plane which states the following: If P (u, v) and Q(u, v) are
differentiable functions in a simple region A ⊂ R2 , the boundary of which is given by u = u(s), v = v(s),
then
k Z si+1   ZZ  
X du dv ∂Q ∂P
P +Q ds = − dudv.
i=0 si
ds ds A ∂u ∂v
It follows that
k Z si+1 ZZ      k Z si+1
E G dϕi
√v √u
X X
kg (s)ds = + dudv + ds.
i=0 si x−1 (R) 2 EG v 2 EG u i=0 si ds

Because our parametrization is isothermal, that is, in addition to F = 0, we have that E = G = λ(u, v) > 0,
        
Ev Gu 1 λv λu 1
√ + √ = + = (4 ln λ)λ = −Kλ,
2 EG v 2 EG u 2 λ v λ u 2λ
where we left the last equality as an exercise. Therefore,
k Z si+1 ZZ k Z si+1 ZZ k Z si+1
X X dϕi X dϕi
kg (s)ds = − Kλdudv + ds = − Kdσ + ds.
i=0 si x−1 (R) i=0 si ds R i=0 si ds

On the other hand, by the theorem of turning tangents,

k Z si+1 k k
X dϕi X X
ds = (ϕi (si+1 ) − ϕi (si )) = ±2π − θi .
i=0 si ds i=0 i=0

Since the curve α is positively oriented, the sign should be plus, as can easily be seen in the particular case
of the circle in a plane.
Lecture 4: Intrinsic Geometry of Surfaces 4-17

By putting these facts together, we obtain

k Z
X si+1 ZZ k
X
kg (s)ds + Kdσ + θi = 2π.
i=0 si R i=0

4.6 The Exponential Map. Geodesic Polar Coordinates (Optional)

In this section, we shall introduce some special coordinate systems with an eye toward their geometric
applications. The natural way of introducing such coordinates is by means of the exponential map, which
we shall now describe.
As we learned in Proposition 4.41, given a point p of a regular surface S and a nonzero vector v ∈ Tp (S),
there exists a unique parametrized geodesic γ : (−ε, ε) → S, with γ(0) = p and γ 0 (0) = v. To indicate the
dependence of this geodesic on the vector v, it is convenient to denote it by γ(t, v) = γ.

Lemma 4.46 If the geodesic γ(t, v) is defined for t ∈ (−ε, ε), then the geodesic γ(t, λv), λ ∈ R, λ > 0, is
defined for t ∈ (−ε/λ, ε/λ), and γ(t, λv) = γ(λt, v).

Proof: Let α : (−ε/λ, ε/λ) → S be a parametrized curve defined by α(t) = γ(λt). Then α(0) = γ(0),
α0 (0) = λγ 0 (0), and, by the linearity of D in (4.10),

Dα0 (t) α0 (t) = λ2 Dγ 0 (t) γ 0 (t) = 0.

It follows that α is a geodesic with initial conditions γ, λγ 0 (0), and by uniqueness

α(t) = γ(t, λv) = γ(λt, v).

Intuitively, Lemma 4.46 means that since the speed of a geodesic is constant, we can go over its trace within
a prescribed time by adjusting our speed appropriately.

Definition 4.47 (Expontial Map) If v ∈ Tp (S), v 6= 0, is such that γ(|v|, v/|v|) = γ(1, v) is defined, we
set
expp (v) = γ(1, v) and expp (0) = p. (4.13)

Remark 4.48 Geometrically, the construction corresponds to laying off (if possible) a length equal to |v|
along the geodesic that passes through p in the direction of v. The point of S thus obtained is denoted by
expp (v).

The important point is that expp is always defined and differentiable in some neighborhood of the origin of
Tp (S).

Proposition 4.49 Given p ∈ S, there exists an ε > 0 such that expp is defined and differentiable in the
interior Bε of a disk of radius ε of Tp (S), with center in the origin.
4-18 Lecture 4: Intrinsic Geometry of Surfaces

Proof: It is clear that for every direction of Tp S, by Lemma 4.46, it is posible to take v sufficiently small so
that the interval of definition of γ(t, v) contains 1, and thus γ(1, v) = expp (v) is defined.
Now, we need the following theorem, which is Theorem 1 in page 304 in M. do Carmo’s textbook, that
Given p ∈ S, there exist numbers ε1 > 0, ε2 > 0, and a differentiable map γ : (−ε2 , ε2 ) × Bε1 → S such
that, for v ∈ Bε1 , v 6= 0, t ∈ (−ε2 , ε2 ), the curve γ(t, v) is the geodesic of S with γ(0, v) = p, γ 0 (0, v) = v,
and for v = 0, γ(t, 0) = p.
Set λ = ε2 /2 in Lemma 4.46, we obtain that γ(t, (ε2 /2)v) is defined for |t| < 2 and |v| < ε1 . Therefore, by
taking a disk Bε ⊂ Tp S, with center at the origin and radius ε < ε1 ε2 /2, we have that γ(1, w) = expp (w),
w ∈ Bε , is defined. The differentiability of expp in Bε follows from the differentiability of γ.
An important complement to this result is the following:

Proposition 4.50 expp : Bε ⊂ Tp S → S is a diffeomorphism in a neighborhood U ⊂ B, of the origin 0 of

Tp S.

Proof: We shall show that the differential d(expp ) is nonsingular at 0 ∈ Tp S. To do this, we identify the
space of tangent vectors to Tp S at 0 with Tp S itself. Consider the curve α(t) = tv, v ∈ Tp S. It is obvious
that α(0) = 0 and α0 (0) = v. The curve (expp ◦ α)(t) = expp (tv) has at t = 0 the tangent vector
d d
(expp (tv)) t=0
= (γ(t, v)) t=0
= v.
dt dt
It follows that (dexpp )0 = v, which shows that dexpp is nonsingular at 0. By applying the inverse function
theorem, we complete the proof of the proposition.
It is convenient to call V ⊂ S a normal neighborhood of p ∈ S if V is the image V = expp (U ) of a
neighborhood U of the origin of Tp S restricted to which expp is a diffeomorphism.
Since the exponential map at p ∈ S is a diffeomorphism on U , it may be used to introduce coordinates in
V . Among the coordinate systems thus introduced, the most usual are

1. The normal coordinates which correspond to a system of rectangular coordinates in the tangent plane
Tp S.
2. The geodesic polar coordinates which correspond to polar coordinates in the tangent plane Tp S.

Definition 4.51 (Normal coordinates) Choose in the plane Tp S, p ∈ S, two orthogonal unit vectors e1
and e2 . Since expp : U → V ⊂ S is a diffeomorphism, it satisfies the conditions for a parametrization in p.
If q ∈ V , then q = expp (w), where w = ue1 + ve2 ∈ U , and we say that q has coordinates (u, v). It is clear
that normal coordinates thus obtained depend on the choice of e1 , e2 .

In a system of normal coordinates centered in p, the geodesics that pass through p are the images by expp
of the lines u = at, v = bt which pass through the origin of Tp S. Observe also that at p, the coefficients of
the first fundamental form in such a system are given by E(p) = G(p) = 1, F (p) = 0.

Definition 4.52 (Geodesic polar coordinates) Choose in the plane Tp S, p ∈ S, a system of polar co-
ordinates (ρ, θ), where ρ is the polar radius and θ, 0 < θ < 2π, is the polar angle, the pole of which is the
origin 0 of Tp S. Observe that the polar coordinates in the plane are not defined in the closed half-line l,
which correpsonds to θ = 0. Set expp (l) = L. Since expp : U \l → V \L is still a diffeomorphism, we may
parametrize the point of V \L by the coordinates (ρ, θ), which are called geodesic polar coordinates.
Lecture 4: Intrinsic Geometry of Surfaces 4-19

We shall use the following terminology. The images by expp : U → V of circles in U centered in 0 will be
called geodesic circles of V , and the images of expp of the lines through 0 will be called radial geodesics of
V . In V \L these are the curves ρ = cosnt. and θ = const., respectively.
We shall now determine the coefficients of the first fundamental form in a system of geodesic polar coordi-
nates.

Proposition 4.53 Let x : U \l → V \L be a system of geodesic polar coordinates (ρ, θ). Then the coefficients
E = E(ρ, θ), F = F (ρ, θ), and G = G(ρ, θ) of the first fundamental form satisfy the conditions
√
E = 1, F = 0, lim G = 0, lim ( G)ρ = 1.
ρ→0 ρ→0

Proof: By definition of the exponential map, ρ measures the arc length along the curve θ = const. It follows
immediate that E = 1.
∂x ∂x
To prove that F = 0, we first prove that Fρ = 0. Since F =< ∂ρ , ∂θ >, we obtain

∂ 2 x ∂x ∂x ∂ 2 x
Fρ =< , > + < , >.
∂ρ2 ∂θ ∂ρ ∂θ∂ρ
Since θ = const. is a geodesic, we have
∂x ∂ 2 x 1 ∂ ∂x ∂x
Fρ =< , >= < , >= 0,
∂ρ ∂θ∂ρ 2 ∂θ ∂ρ ∂ρ
as we claimed. Thus F (ρ, θ) does not depend on ρ.
For each q ∈ V , we shall denote by α(σ) the geodesic circle that passes through q, where σ ∈ [0, 2π] if (q = p,
α(σ) is the constant curve α(σ) = p). We shall denote by γ(s), where s is the arc length of γ, the radial
geodesic that passes through q. With this notation, we may write
dα dγ
F (ρ, θ) =< , >.
dσ ds
The coefficient F (ρ, θ) is not defined at p. However, if we fix the radial geodesic θ = const., the second
member (right hand side) of the above equation is defined for every point of this geodesic. Since α(σ) = p
at p, that is, dα
dσ = 0 at ρ = 0, we obtain

dα dγ
lim F (ρ, θ) =< lim , >= 0.
ρ→0 ρ→0 dσ ds
Together with the fact that F does not depend on ρ, this implies that F = 0.
To prove the last assertion of the proposition, we choose a system of normal coordinates (ũ, ṽ) in p in such
a way that the change of coordinates is given by
ũ = ρ cos θ, ṽ = ρ sin θ, ρ 6= 0, 0 < θ < 2π.
√ p
By recalling that EG − F 2 = Ẽ G̃ − F̃ 2 ∂(ũ,ṽ)
∂(ρ,θ) , where
∂(ũ,ṽ)
∂(ρ,θ) is the Jacobian of the change of coordinates
and Ẽ, F̃ , G̃, are the coefficients of the first fundamental form in the normal coordinates (ũ, ṽ), we have
√ p
G = ρ Ẽ G̃ − F̃ 2 , ρ 6= 0.
Since at p, Ẽ = G̃ = 1, F̃ = 0 (the normal coordinates are defined at p), we conclude that
√ √
lim G = 0, lim ( G)ρ = 1,
ρ→0 ρ→0

which coludes the proof of the proposition.