DiffGeomI - Notes Marc Burger
DiffGeomI - Notes Marc Burger
MARC BURGER
STEPHAN TORNIER
Abstract. These are notes of the course Differential Geometry I held at ETH
Zurich in 2015.
Contents
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1. Differential Manifolds and Differentiable Maps . . . . . . . . . . . . . . 2
2. Tangent spaces, Differential and Whitney’s Embedding Theorem . . . 21
3. Differential Forms and Integration on Manifolds . . . . . . . . . . . . . 31
4. De Rham Cohomology . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
5. De Rham’s Theorem . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71
Introduction
Differential geometry is a synthesis of three different subjects: Analysis in Rn ,
topology and multilinear algebra. It precisely defines a class of “spaces” on which one
can do analysis, termed differential manifolds. These spaces as well as the associated
notion of differentiable functions are the central concept of this course. Differential
manifolds look locally like Rn but are globally much less boring. Examples are the
sphere as well as surfaces with holes:
p
DIFFERENTIAL GEOMETRY 3
(iv) In dimension two, the situation is already so rich that it defies any reason-
able classification. Examples are, as before, the family of surfaces with g
holes (g ∈ N0 ): These surfaces are all compact and connected and in fact
a classification of compact, connected, two-dimensional topological mani-
folds is managable. To this end we will later on introduce the notion of
orientability to distinguish between orientable examples as above and non-
orientable examples like projective space. To get an idea of the wealth of
general (connected) two-dimensional topological manifolds, note that the
complement in R2 of a Cantor set is an example.
Non-Example 1.3.
(i) Consider M := [0, 1] ⊂ R with the induced topology. Every interior point
of M satisfies the third requirement of the definition of a topological man-
ifold for n = 1 but the points 0, 1 ∈ M do not. For instance, a typical
neighbourhood of 0 ∈ M is given by U = [0, ε) for some 0 < ε < 1. Sup-
pose ϕ : U → V ⊆ R is a homeomorphism onto an open subset V of R.
Note U \{0} is connected, but V \{ϕ(0)} is not. This contradicts ϕ being
continuous.
(ii) The set M := [0, 1]2 is not a (two-dimensional) topological manifold either.
Here one may argue using the fundamental group instead of connectedness.
Both non-examples above are manifolds with boundary though, as defined later.
(iii) Let M = Q ⊂ R with the induced topology. Then every open set of Q is
countable and hence cannot be in bijection with an open subset of any Rn .
The next definition constitutes the next important step towards the definition
of differential manifolds.
We are now going to examine the case in which two charts intersect, see below.
4 MARC BURGER STEPHAN TORNIER
Uα Uβ
ϕα ϕβ
θαβ
The maps
θβα := ϕβ ◦ (ϕα |Uα ∩Uβ )−1 : ϕα (Uα ∩ Uβ ) → ϕβ (Uα ∩ Uβ ),
Note that our definition perfectly resembles real-life atlases, introduced by Mer-
cator in 1585. By Definition 1.1, every topological manifold admits a C 0 -atlas. As a
next step towards the definition of differential manifolds we now introduce smooth
atlases.
Example 1.8. We now give examples of topological manifolds with smooth atlases.
(i) Let M := U be an open subset of Rn and A = {(U, id)}. This example may
seem rather trivial but nevertheless includes an important example of a Lie 2
group, namely GL(n, R) := {A ∈ Mn,n (R) | det A 6= 0} ⊆ Mn,n (R) ∼ = Rn .
n n+1
(ii) Let M := S ⊆ R with the induced topology. If n ≥ 1 then an atlas of
M requires at least two charts since M is compact whereas an open subset
of Rn is not. Now, there are in fact many choices for an atlas with two
elements one of which may be constructed using stereographic Pn+1 2 projection
n+1
as follows: Let S = {x = (x1 , . . . , xn+1 ) ∈ R
n
| i=1 xi = 1}, N :=
(0, . . . , 0, 1) ∈ S n and S := (0, . . . , 0, −1). Further, set UN := S n \{N } and
US := S n \{S}. Then the pairs (UN , ϕN ) and (US , ϕS ), where
(x1 , . . . , xn )
ϕN,S : S n → Rn , x 7→
1 ∓ xn+1
DIFFERENTIAL GEOMETRY 5
x
ϕN (x)
b
We shall only concern ourselves with smooth manifolds in this course and not
discuss subtleties arising from the exact value of k. One may wonder whether any
topological manifold admits a smooth atlas. The following remark answers this
question in the negative.
Remark 1.12. There is a compact topological manifold that does not admit a
smooth atlas. However it does admit a system of pairwise compatible charts that
cover all but one point. See [Ker60].
Even more remarkable is the fact that there are topological manifolds which
admit incompatible maximal atlases that do not even arise from each other in the
above fashion.
Remark 1.14. The topological manifold S 7 admits a maximal atlas which is not
compatible with the one defined in Example 1.8 and can even be given by polyno-
mial equations. See [Mil56].
Example 1.15. We now describe three standard ways of producing new manifolds
out of old ones.
(i) (Regular value) Let f : Rn → Rm be a smooth map and let a ∈ Rm be
a regular value, i.e. for every x ∈ f −1 (a) the derivative Dx f : Rn → Rm
has maximal rank m. Then f −1 (a) = {x ∈ Rn | f (x) = a} is a smooth
manifold in a natural way. Note that all manifolds produced in this way are
naturally subsets of Rn . Later on we give the details of this construction in
much greater generality.
(ii) (Open subset) Let M be a smooth manifold with atlas A and let U ⊆ M be
open. Then A |U := {(U ∩ V, ϕU∩V ) | (V, ϕ) ∈ A} is a smooth atlas on U .
Note that A |U need not be maximal. However, it is contained in a unique
maximal smooth atlas by Lemma 1.10.
(iii) (Products) Let M, N be smooth manifolds with atlases A = {(Uα , ϕα | α ∈
A} and B = {(Vβ , ψβ ) | β ∈ B}. Then P := {(Uα × Vβ , ϕα × ψβ | (α, β) ∈
A × B} is a smooth atlas on M × N . We recall the definition of ϕα × ψβ
((α, β) ∈ A × B): Let ϕα : Uα → Rm and ψβ : Vβ → Rn where m = dim M
and n = dim N . Then
Next, we describe a construction which does not yield smooth manifolds that are
naturally subsets of some euclidean space, namely quotients. As a matter of fact,
the construction need not even begin with a smooth manifold. For instance consider
a square and glue two opposite sides together. The result is a cylinder. Continuing
by gluing together its ends yields a torus which is a smooth manifold.
DIFFERENTIAL GEOMETRY 7
The reader is invited to think about what the following identification produces.
For i ∈ {1, 2}, the edge labelled ai (bi ) is identified with the edge labelled a−1
i (b−1
i )
in the direction shown.
a2 b−1
1
b2 a−1
1
a−1
2
b1
b−1 a1
2
The above construction can be generalized to every regular polygon whose num-
ber of sides is divisible by four. Anyway, some work is needed to make the above
constructions precise. We begin by reviewing the quotient topology: Let X be a
topological space and let ∼ be an equivalence relation on X. Further, let X/ ∼
denote the set of equivalence classes, termed the quotient of X by ∼, and let
π : X → X/ ∼ be the map which to every x ∈ X associates its equivalence
class. We turn X/ ∼ into a topological space as follows: A set U ⊆ X/ ∼ is open if
π −1 (U ) is open. Using the fact that π respects all Boolean operations one verifies
that this definition does indeed turn X/ ∼ into a topological space. Furthermore,
π is continuous.
Example 1.16. (Torus). Let X := R2 . For all x, y ∈ X, set x ∼ y if and only
if x − y ∈ Z2 . Since (Z2 , +) is a group, this is indeed an equivalence relation.
We are going to argue that X/ ∼ is homeomorphic to the torus. First of all we
determine a suitable piece of X which meets every equivalence class. The unit square
S := [0, 1]2 ⊂ R2 will serve. Indeed, let π : X → X/ ∼ denote the quotient map.
Then π([0, 1]2 ) = X/ ∼. To see this, just floor the components of a vector x ∈ X
and subtract the resulting vector from x. However, it remains to be understood
which points of S are identified under the equivalence relation. Clearly, if x ∈ S is
in the interior of S, no other point in S is equivalent to x. However, (x1 , 0)T ∈ S is
identified with (x1 , 1)T ∈ S for all x1 ∈ [0, 1] and, similarly, (0, x2 )T ∈ S is identified
with (1, x2 )T ∈ S for all x2 ∈ [0, 1]. As observed above, these identifications yield a
torus. The details of a homeomorphism are to be worked out.
The following example observes that not all equivalence relations yield reasonable
quotient spaces.
Example 1.17. (A pathology). Let X := R. For all x, y ∈ X, set x ∼ y if and only
if x − y ∈ Q. We show that X/ ∼ is an uncountable space whose quotient topology
has only the trivial open sets. Hence the topology is not suited at all to study
the quotient in this case, it does not give any shape. First of all, X/ ∼ is in fact
uncountable since otherwise R would be a countable union of countable equivalence
classes and thus countable. Now, let U ⊆ X/ ∼ be open, i.e. π −1 (U ) is open. Hence
π −1 (U ) contains a non-empty open interval (a, b). As a result, (a, b) + Q = R ⊆
π −1 (U ) since any real number can be approximated arbitrarily well by rationals.
Hence the assertion.
The last example shows that in order to obtain quotient topologies with good
properties one has to impose some conditions on the equivalence relation.
8 MARC BURGER STEPHAN TORNIER
In any case, open equivalence relations yield quotients which potentially sat-
isfy the first two requirements on topological manifolds as follows. Recall that an
equivalence relation ∼ on a set X is the subset R := {(x, y) ∈ X × X | x ∼ y}.
Proof. As to (i), let B be a basis of open subsets of X and set B ′ := {π(U ) | U ∈ B}.
Since ∼ is open, B ′ consists of open sets. S
In addition, if V ⊆ X/ ∼ is open then by
definition so is π −1 (V ), hence π −1 (V ) = U∈B, U⊆π−1 (V ) U . Therefore
[
V = π(π −1 (V )) = π(U )
U ∈B
U ⊆ π −1 (V )
The second part of Proposition 1.19 can be applied to Example 1.16. Indeed, the
equivalence relation
R = {(x, y) ∈ R2 × R2 | x − y ∈ Z2 }
= {(x, y) ∈ R2 × R2 | (e2πi(x1 −x2 ) , e2πi(y1 −y2 ) ) = (1, 1)}
is a level set of a continuous function and as such closed.
Example 1.20. (Real projective space). In linear algebra, Pn R is defined as the
quotient of Rn+1 \{0} by the equivalence relation x ∼ y if and only if x = λy for
some λ ∈ R∗ . In fact, we could replace R with any other field. In this case, the
equivalence relation is open and its graph R is closed. Hence Pn R = Rn+1 \{0}/ ∼
is second-countable and Hausdorff. S
To see that ∼ is open, let A ⊆ Rn+1 \{0} be open, then π −1 (π(A)) = λ∈R∗ λA
is open as well since multiplication by λ is a homeomorphism of Rn+1 \{0}.
Now we show that R is closed. As before, we identify it as the level set of a
continuous function. Define
n+1
X
n+1 n+1
f :R ×R → R, (x, y) 7→ (xi yj − xj yi )2 .
i,j=1
e3 ⊆ R3
U
b
is surjective. (It is not injective, however; in fact every point in Pn R has two pre-
images.
Example 1.21. (Grassmann manifolds). Grassmann manifolds generalize projective
space and play a major role in the theory of vector bundles which we study later
on. Let 1 ≤ k ≤ n and let G(k, n) be the set of all subspaces of Rn of dimension k.
First of all, we describe a topology on G(k, n) utilizing linear algebra: Let
F (k, n) := {(v1 , . . . , vk ) ∈ (Rn )k | v1 , . . . , vk linearly independent}. Since every
vector space admits a basis, we have a surjective map π : F (k, n) → G(k, n) given
by (v1 , . . . , vk ) 7→ R v1 + · · ·+ R vk . This already puts us in a good position to define
a topology: To k vectors w1 , . . . , wk of Rn we associate the k × n-matrix
w1
..
M (w1 , . . . , wk ) := . .
wk
Then w1 , . . . , wk are linearly independent if and only if rankM (w1 , . . . , wk ) = k. In
this way, F (k, n) is identified with {A ∈ Mk,n (R) | rankA = k}, that is
A1j1 · · · A1jk
.. .
.. 6= 0
A ∈ Mk,n (R) ∃1 ≤ j1 < · · · < jk ≤ n : det .
Akj1 · · · Akjk
which is union over multi-indices of open subsets of Mk,n (R) and hence open. Fur-
thermore, given w1 , . . . , wk and w1′ , . . . , wk′ one verifies that R w1 + · · · + R wk =
R w1′ + · · · R wk′ if and only if there is B ∈ GL(k, R) such that M (w1 , . . . , wk ) =
BM (w1′ , . . . , wk′ ). Using this, one can show that (w1 , . . . , wk ) ∼ (w1′ , . . . , wk′ ) if
and only if R w1 + · · · R wk = R w1′ + · · · R wk′ is an open equivalence relation:
Indeed, given B ∈ GL(k, R), the map LB : Mk,n (R) → Mk,n (R), A 7→ BA is
an invertible linear map with inverse LB −1 . In particular, LB is a homeomor-
phism. It preserves
S the open subset F (k, n) and if Ω ⊆ Fk,n is open then so is
π −1 (π(Ω)) = B∈GL(k,R) LB (Ω). One may also show that the graph of ∼ is a closed
subset of F (k, n)× F (k, n). We have thus equipped G(k, n) with a second-countable
Hausdorff topology.
We now define charts on G(k, n) which turn it into a smooth manifold. Notice the
analogy to the construction of Example 1.20. For any multi-index J = (j1 , . . . , jk )
where 1 ≤ j1 ≤ · · · ≤ jk ≤ n, set
w1
w1j1 · · · w1jk
eJ = . . .. = Id ∈ M (R) .
U .. ∈ Mk,n (R) .. . k,k
wk wkj1 · · · wkjk
eJ → G(k, n) is injective and
Then π|UeJ : U
v1
v1j1 ··· v1jk
−1 e .. .. .. 6= 0 .
π (π(UJ )) = . det . .
vk vkj1 ··· vkjk
S
Therefore, π(UeJ ) ⊆ G(k, n) is open and we have e
J π(UJ ) = G(k, n). As a con-
sequence π|UeJ : UeJ → π(U eJ ) is bijective, continuous and open, that is, a home-
∼
=
omorphism. In addition, we have the projection maps pJ : U eJ −
→ Mk,n−k (R) and
one verifies that (π(UeJ ), pJ ◦ (π| e ) )J is a smooth atlas on G(k, n). We have
−1
UJ
dim G(k, n) = k(n − k).
DIFFERENTIAL GEOMETRY 11
Recall that in fact the existence of partial derivatives alone does not imply dif-
ferentiability. For instance, the function
(
xy
2 2 2 (x, y) 6= (0, 0)
f : R → R, (x, y) 7→ x +y
0 (x, y) = (0, 0)
has partial derivatives everywhere but is not differentiable at (x, y) = (0, 0). In fact,
it is not even continuous at this point.
We now define smooth functions on manifolds.
Definition 1.24. Let M be a smooth manifold and p ∈ M . A function f : M → R
is differentiable at p if for some chart (U, ϕ) at p, the function ϕ(U ) → R, x →
7
f ◦ ϕ−1 (x) is differentiable at ϕ(p).
Definition 1.24 is in fact independent of the chart chosen. The argument which
shows this is an important and much used one: Let (U, ϕ) and (V, ψ) be charts
including p ∈ M . For x ∈ ψ(U ∩ V ) we compute f ◦ ψ −1 (x) = f ◦ ϕ−1 ◦ (ϕ ◦ ψ −1 )(x)
which is a composition of the smooth map ϕ ◦ ψ −1 defined on ψ(U ∩ V ) and f ◦
ϕ−1 which is differentiable at ϕ(p). The same argument shows that the following
definitions make sense.
Definition 1.25. Let M and N be smooth manifolds. Then
(i) f ∈ C k (M, R) if for every chart (U, ϕ) of M the function f ◦ϕ−1 : ϕ(U ) → R
is C k ,
(ii) f : M → N is differentiable at p ∈ M if there are charts (U, ϕ) at p and
(W, ψ) at f (p) such that (1) f (U ) ⊆ W and (2) ψ ◦ f ◦ ϕ−1 : ϕ(U ) → ψ(W )
is differentiable at ϕ(p),
(iii) f ∈ C k (M, N ) if f : M → N is continuous and for every pair (U, ϕ),
(W, ψ) of charts respectively on M and N such that f (U ) ⊆ W the map
ϕ(U ) → ψ(W ), x 7→ ψ ◦ f ◦ ϕ−1 (x) is C k , and
(iv) f : M → N is a C k -diffeomorphism if f is a homeomorphism and both f
and f −1 are C k -maps.
12 MARC BURGER STEPHAN TORNIER
Retain Definition 1.25. Note that in (ii) we do not simply require (2) because
without (1) it might be satisfied automatically for rather bad choices of f . Also,
the reader is invited to check that Definition 1.25 is consistent, i.e. for instance that
part (iii) for N = R yields the same notion as part (i). The following is an example
of how results of calculus, namely Theorem 1.23 generalize to manifolds.
Theorem 1.26. Let M and N be manifolds and let f : M → N be a C 1 -map. Then
f is differentiable at every point p ∈ M .
Remark 1.27. We now collect several examples to illustrate Definition 1.25.
(i) Given a manifold M , the set C k (M, R) of C k -functions on M is an R-
algebra for pointwise addition and multiplication of functions. This follows
from the according statement for M = R.
(ii) Let M, N and R be manifolds and let f ∈ C k (M, N ) as well as g ∈ C k (N, R)
be given. Then g ◦ f ∈ C k (M, R) by the chainS rule.
(iii) Let M and N be a manifolds and let M = α∈A Uα be an open covering
of M . Then f ∈ C k (M, N ) if and only if f |Uα ∈ C k (Uα , N ) for all α ∈ A,
i.e. being C k is a local condition.
(iv) Let M = R and N = S 1 = {z ∈ C | |z| = 1}. Then exp : R → S 1 , t 7→ e2πit
is smooth.
(v) Consider GL(n, R) with its smooth structure defined in Example 1.8. Then
the map GL(n, R) × GL(n, R) → GL(n, R), (A, B) 7→ AB is smooth. As
a consequence, given g ∈ GL(n, R) the map Lg : GL(n, R) → GL(n, R),
a 7→ ga is a diffeomorphism. In fact, a two-sided inverse of Lg is Lg−1 .
Remark 1.28. We remark further that any homeomorphism of a manifold can be
turned into a smooth map: Let M be a smooth manifold with maximal atlas A and
let F : M → M be a homeomorphism. Set A′ = {(F (U ), ϕ ◦ F −1 ) | (U, ϕ) ∈ A}.
Then A′ is a (maximal) smooth atlas on M and F : (M, A) → (M, A′ ) is a smooth
diffeomorphism.
We are now working towards the rank theorem for which we recall that the rank
of a linear map T : Rn → Rm is given by the dimension of its image which equals
the row and column rank of any coordinate matrix of T .
Definition 1.29. Let M and N be manifolds, p ∈ M and let f : M → N be
differentiable at p. Further, let (U, ϕ) be a chart of M at p and (W, ψ) a chart of N
at f (p). The rank of f at p is the rank at ϕ(p) of the linear map Dϕ(p) (ψ ◦ f ◦ ϕ−1 ) :
Rn → Rm .
The notion of rank of a map will be exploited a lot in the sequel. It leads to
the concept of immersion, submersion, immersed manifolds etc. The key ingredient
comes from the rank theorem of calculus to which we now turn.
1.4. The Rank Theorem. First of all, we recall the inverse function theorem.
Theorem 1.30. Let W ⊆ Rn be open and let F : W → Rn be a C r -map (r ≥ 1) such
that at a ∈ W , the derivative Da F : Rn → Rn is invertible. Then there are open
subsets U and V of Rn containing a and F (a) respectively such that F |U : U → V
is a C r -diffeomorphism.
This theorem is adaptable to the setting of smooth manifolds. However, for
now we apply it to the following calculus result which says that up to smooth
change of coordinates a map with constant rank k is the projection onto the first
k components. To this end, recall the following fact from linear algebra: Let L :
Rn → Rm be a linear map of rank k. Note that necessarily k ≤ min(m, n). Then
there are invertible linear maps G : Rn → Rn and H : Rm → Rm such that
(H ◦ L ◦ G−1 )(x1 , . . . , xn ) = (x1 , . . . , xk , 0, . . . , 0).
DIFFERENTIAL GEOMETRY 13
G H
U
V
HF G−1
0 b b
0
We state that the change of variable maps G and H of Theorem 1.31 are not
generally defined on the whole of A0 and B0 .
Define G(x) := (F1 (x), . . . , Fk (x), xk+1 , . . . , xn ) where F (x) = (F1 (x), . . . , Fm (x)).
Observe that G(0) = 0 Furthermore, we have x ∈ A0 :
∂F ∂F1 ∂F1 ∂F1
∂x1
1
· · · ∂x k ∂x k+1
· · · ∂x n
. .. .. ..
.
. . . . !
∂Fk ∂Fk ∂Fi
∂x1 · · · ∂F k ∂Fk
· · · ∂xj ∗
Dx G = ∂xk ∂xk+1 ∂xn = .
0 ··· 0 1 0 Idn−k
.. .. ..
. . .
0 ··· 0 1
That is, det Dx G = det(∂Fi /∂xj ) 6= 0. Hence we may apply the Inverse Function
Theorem to G. There are open subsets A1 ⊂ A and U1 ⊆ Rn , containing 0 respec-
tively such that G : A1 → U1 is a C r -diffeomorphism. Now, let us compute F ◦ G−1
on U1 : Let y ∈ U1 and write y = G(x) with x ∈ A1 . Then
F ◦ G−1 (y) = F ◦ G−1 (F1 (x), . . . , Fk (x), xk+1 , . . . , xn ) = F (x1 , . . . , xn )
= (F1 (x), . . . , Fk (x), Fk+1 (x), . . . , Fm (x))
= (y1 , . . . , yk , fk+1 (y), . . . , fm (y))
for some fl (y) = Fl (x) with l ≥ k + 1. Now observe that F ◦ G−1 has constant rank
k on U1 and that for y ∈ U1 we have Dy (F ◦ G−1 ) = DG−1 (y) F ◦ Dy G. Hence the
14 MARC BURGER STEPHAN TORNIER
vanishes. Modulo replacing U1 with a smaller connected ε-ball we may thus assume
that the functions fk+1 , . . . , fm only depend on y1 , . . . , yk . Thus
F ◦ G−1 (y1 , . . . , yn ) = (y1 , . . . , yk , fk+1 (y1 , . . . , yk ), . . . , fm (y1 , . . . , yk ))
Now define T (z) := (z1 , . . . , zk , zk+1 + fk+1 (z1 , . . . , zk ), . . . , zm + fm (z1 , . . . , zk )) for
z ∈ Rm with (z1 , . . . , zk ) ∈ prk1 (U1 ) where prk1 projects onto the first k coordinates.
Then T (0) = 0 and
Idk 0
∗ Idn−k
is invertible. By the Inverse Function Theorem there is an open set B1 ⊆ Rm
containing 0 such that T : B1 → T (B1 ) ⊆ Rm is a C r -diffeomorphism onto its open
image. Furthermore,
T (z1 , . . . , zk , 0, . . . , 0) = (z1 , . . . , zk , fk+1 (z1 , . . . , zk ), . . . , fm (z1 , . . . , zk ))
which implies the assertion: T −1 F G−1 (y1 , . . . , yn ) = (y1 , . . . , yk , 0, . . . , 0). Now set
H = T −1 .
We remark that in Mm,n (R) the condition of being of rank k is neither open
nor closed and hence constitutes a strong assumption in the above theorem which
utterly fails without it. As an immediate corollary of Theorem 1.31, we record
that for a smooth map between manifolds there are always charts “adapted” to
it. To this end we introduce the following notation: Given a ∈ Rn and ε > 0 set
Cεn (a) := {x ∈ Rn | ∀i ∈ {1, . . . , n} : |xi − ai | < ε}. The sets Cεn (a) are hypercubes,
for instance Cεn (0) = (−ε, ε)n .
Corollary 1.32. Let N and M be manifolds of dimension n and m respectively and
p ∈ M . Further, let F : M → N be a smooth map of constant rank k. Then there
are charts (U, ϕ) at p and (W, ψ) at F (p) such that ϕ(p) = 0, ϕ(U ) = Cεn (0) ⊆ Rn
as well as ψ(F (p)) = 0, ψ(W ) = Cεm (0) and
ψF ϕ−1 (x1 , . . . , xn ) = (x1 , . . . , xk , 0, . . . , 0)
for some ε > 0.
1.5. Submanifolds, Immersions, Embeddings etc. We have defined the cat-
egory of smooth manifolds and smooth maps. In some categories the image and
kernel of a morphism are naturally identified as subobjects of the codomain and
domain of the morphism. For instance, if L : V → W is a linear map between
vector spaces then L(V ) is a subspace of W and L−1 (0) ⊆ V is subspace of V . An
analogous statement holds true in the category of groups. For smooth manifolds,
however, images and level sets of smooth maps can be pretty bad. In the sequel
we address both of these problems. First we establish conditions under which level
sets of smooth maps are “nice”. This leads us to introduce the notion of a subman-
ifold. For instance, S 1 is a submanifold of the torus in e.g. the way depicted below.
However, a real line wrapped around the torus at an irrational angle α ∈ R \ Q as
in ι : R → S 1 × S 1 , t 7→ (e2πit , e2πiαt ) is not going to be a submanifold because a
neighbourhood of a point in the image contains a dense set of image points. In fact,
the image of ι is dense in S 1 × S 1 .
DIFFERENTIAL GEOMETRY 15
ϕ ψ
Cεn (0) / C m (0)
ε
prk
1
In particular, (F |U )−1 (q) = U ∩ F −1 (q) and ϕ(U ∩ F −1 (q)) = (prk1 )−1 (0) = {x ∈
Cεn (0) | x1 = · · · = xk = 0} which is a hypercube. Up to swapping coordinates this
shows that F −1 (q) is a submanifold in the sense of Definition 1.33.
Note that in Theorem 1.36, the set f −1 (y) on which f has constant rank is a
closed subset of N . Therefore, a priori, Theorem 1.34 is useless. However, we will
exploit the fact that the rank at any point in f −1 (y) is the maximal possible one
using the following lemma.
Lemma 1.37. Let n ≥ m be integers. Then the set Rm := {A ∈ Mm,n (R) | rank A =
m} is open in Mm,n (R).
Proof. For every A ∈ Rm we construct an open neighbourhood of A contained in
Rm . Let 1 ≤ j1 < · · · < jm ≤ n be a multi-index such that the associated minor of
A has non-zero determinant:
A1j1 · · · A1jm
det ... .. 6= 0.
.
Amj1 ··· Amjm
Then ( )
VA := B ∈ Mm,n (R) det(Bijl ) 1≤l≤m 6= 0
1≤i≤m
Having discussed level sets of smooth functions we now turn to the question
under which circumstances the image of a smooth map between manifolds is a
manifold.
Notice that f (π/2) = (0, 0)T = f (3π/2). In particular, immersions need not
be injective. However, the tangent vectors of f at these two points differ.
(ii) As before, consider the irrationally imbedded real line into the torus given
by f : R → S 1 × S 1 , t 7→ (e2πit , e2πiαt ) where α ∈ R \ Q. Whereas f is
injective, its image is dense which is considered bad for reasons that will
become clear later.
1.6. Sard’s Theorem. We now turn to the arguably most important theorem in
Differential Topology. While individual level sets of smooth maps can be very bad,
there are quite general situations in which a smooth map has lots of regular values;
this follows from Sard’s Theorem which says that in any case the image of the set
of critical points is always small.
Theorem 1.43 (Sard, 1942). Let U ⊆ Rn be open and let f : U → Rm be smooth.
Let C := {x ∈ U | rank Dx f < m} be the set of critical points of f . Then f (C) ⊆
Rm has Lebesgue measure zero.
A few remarks about this theorem are in order. First of all, recall that S∞a subset
E ⊆ Rm has measure zero if for everyP ε > 0 there is a covering E ⊆ i=1 Ci by
∞
countably many hypercubes such that i=1 vol(Ci ) < ε. We will use several times
the fact that a countable union of sets of measure zero has measure zero. This
follows from the above definition by choosing geometrically decaying values for the
respective ε.
For instance, this implies the disturbing fact that Q ⊆ R has measure zero:
Indeed, let f : N → Q be a bijection. Then f (n) ∈ (f (n) − ε/2n , f (n) + ε/2n ) and
the sum of the volumes of these intervals is 2ε.
Also, note the following: Let f : [0, 1] → R be monotonically increasing, i.e.
f (t1 ) ≤ f (t2 ) whenever t1 ≤ t2 . Then the set of points at which f is not differen-
tiable is of measure zero. It therefore seems like one could deduce a lot of information
about f using its derivative; however, there is a monotonically increasing f whose
derivative is zero whenever it exists. See [RSN72] for a nice introduction to examples
of this sort.
Remark 1.44. We now discuss Theorem 1.43 in the two cases n < m and n ≥ m
because the flavour is somewhat different.
(i) (n < m). In this case, C = U and hence the statement is that f (U ) is of
measure zero in Rm . As a warm-up to this, we prove that if U ⊆ R is open
and f : U → Rm for m ≥ 2 is C 1 then f (U ) has measure zero. First of all,
note that the C 1 assumption is indeed necessary as there is a continuous
surjective map f : [0, 1] → [0, 1]2 . By reparameterization, assume that
[0, 1] ⊆ U . Then there is a constant C > 0 such that kf (x) − f (y)k ≤
C|x − y| for all x, y ∈ [0, 1]. Given k ∈ N, subdivide the interval [0, 1]
into the subintervals [j/k, (j + 1)/k]. Then f ([j/k, (j + 1)/k]) is contained
in a hypercube of side length at most C/k and f (C) can be covered by
k hypercubes of this side length. The volume of each such hypercube is
(C/k)m and hence the sum of their volumes is at most k(C/k)m = C m k 1−m .
Since m ≥ 2, this tends to zero as k tends to infinity.
(ii) (n ≥ m). Assume that f : U → Rm is defined on an open subset U ⊆ Rn .
Also, let us assume that C 6= U , that is, there is a regular point x ∈ U .
Then there is an open subset V in U containing x on which f has maximal
rank. Hence f (V ) ⊆ Rm is open and as a consequence cannot have measure
zero. Hence f (U )\f (C) is non-empty. In particular, there are tons of regular
values! In fact, Sard’s Theorem 1.43 is often applied by stating the existence
of a regular value.
(iii) We remark that Sard’s Theorem remains true under the weaker assumption
that f be C k for some k ≥ max{1, n − m + 1}. Thus for n = 1 it suffices in
fact to ask for f to be C 1 whereas when m = 1 the function f needs to be
C n . For instance, there is an example of a C 1 -function f : R2 → R whose
image contains an interval of singular values.
We now turn to the proof of Theorem 1.43. It contains two major ideas.
DIFFERENTIAL GEOMETRY 19
Proof. The proof proceeds by induction on n. Note that the statement makes sense
for n ≥ 0 and m ≥ 1. For n = 0 and m ≥ 1 the theorem is true since the image
of the one-point space R0 has measure zero in Rm for m ≥ 1. Now, let n ≥ 1 and
write f = (f1 , . . . , fm ). Recall that C = {x ∈ U | rank Dx f < n}. Set
∂fr
C1 := {x ∈ U | Dx f ≡ 0} = x ∈ U (x) = 0 ∀r ∈ {1, . . . , m} ∀i ∈ {1, . . . , n} .
∂xi
More generally, for i ≥ 1, let Ci denote the set of all x ∈ U for which all partial
derivatives of f up to order i vanish at x. Then
· · · ⊆ Ci+1 ⊆ Ci ⊆ · · · ⊆ C1 ⊆ C.
The proof is now divided into the follwing three steps.
(i) Show that f (C\C1 ) has measure zero.
(ii) Show that f (Ci \Ci+1 ) has measure zero for all i ≥ 1.
These two steps are not
T yet enough to conclude the assertion as we do not know
about the measure of i≥1 Ci . We therefore end with the following step.
(iii) Show that f (Ck ) has measure zero for large enough k. In fact, k > n/m − 1.
The first two steps are based on the same idea whereas the third one introduces a
new one.
Step (i). We consider C\C1 . Observe that for m = 1 we have C = C1 . Indeed, for a
function f : U → R the condition that rank Dx f < 1 is equivalent to the vanishing
of all first partial derivative at x. Hence f (C\C1 ) is empty. We may thus assume
m ≥ 2. In this situation, let x ∈ C\C1 . By definition and without loss of generality
we may thus assume (∂f1 /∂x1 )(x) 6= 0. Define a change of variables h : U → Rn
by x 7→ (f1 (x), x2 , . . . , xn ) and observe that
∂f1 ∂f1
∂x1 · · · · · · ∂x m
1
Dx h = ..
.
1
which has non-zero determinant. Hence, by the Inverse Function Theorem, there is
an open subset V of U containing x and an open subset V ′ ⊆ Rn containing h(x)
such that h|V : V → V ′ is a diffeomorphism. Now define g := f ◦(h|V )−1 : V ′ → Rm .
Let C ′ ⊆ V ′ be the set of critical points of g. Then the chain rule implies that
h(V ∩ C) = C ′ . Hence f (V ∩ C) = g(C ′ ). It now suffices to show that g(C ′ ) has
measure zero since we can cover U \C1 with countably many such V ’s and deduce
that f (C\C1 ) has measure zero.
As in Theorem 1.34, we show that g has a special form. Let (t, x2 , . . . , xn ) =
(f1 (x), x2 , . . . , xn ) = h(x) ∈ V ′ . Then
g(t, x2 , . . . , xn ) = gh(x) = f (x)
= (f1 (x), . . . , fm (x))
= (t, f2 h−1 (t, x2 , . . . , xn ), . . . , fm h−1 (t, x2 , . . . , xn ))
= (t, g t (x2 , . . . , xn )
where g t : Rn−1 → Rm−1 is given by
g t (x2 , . . . , xn ) := (f2 h−1 (t, x2 , . . . , xn ), . . . , fm h−1 (t, x2 , . . . , xn ).
Observe that n − 1 ≥ 0 and m − 1 ≥ 1 which is covered by the induction hypothesis.
Compute
1 0
D(t,x2 ,...,xn ) g = .
∗ D(x2 ,...,xn ) g t
20 MARC BURGER STEPHAN TORNIER
∂ i fr (x)
w(x) := =0
∂xj2 · · · ∂xji+1
but ∂x
∂w
j1
6= 0. Without loss of generality, assume j1 = 1, that is ∂x ∂w
1
6= 0. Again,
n
consider the map h : U → R given by x 7→ (w(x), x2 , . . . , xn ) which is a diffeomor-
phism from some open set V ⊆ U containing x onto some open set V ′ ⊆ Rn . By
definition, h(Ci ∩ V ) ⊆ ({0} × Rn−1 ). Now, consider again the map g := f ◦ h−1 :
V ′ → Rm . Then gh(Ci ∩ V ) = f (Ci ∩ V ) but since h(Ci ∩ V ) ⊆ {0} × Rn−1 we may
consider g := g|{0}×Rn−1 . Observe that any point in h(Ci ∩ V ) is certainly a critical
point of g. Hence we are done by recurrence.
Step (iii). We now show for large enough k, the set f (Ck ) has measure zero. To this
end, let I δ ⊆ U be a closed hypercube whose edges have length δ. We show that
f (Ck ∩ I δ ) has measure zero for k > n/m − 1. By Taylor, we have for all x ∈ Ck ∩ I δ
and h with x + h ∈ I δ that
f (x + h) = f (x) + R(x, h)
where kR(x, h)k ≤ Ckhkk+1 and C only depends on f and I δ . More precisely, this
is a consequence of the integral form of the remainder. Now, pick N to subdivide
I δ into cubes of side length δ/N . Let I be one of these small cubes containing a
point x ∈ Ck ∩ I.
Ck
b
Iδ
Hence f (I) is contained in a hypercube with edges of the above length. There are
at most N n such little cubes and hence f (Ck ∩ I δ ) is contained in a union of hyper-
√ √
cubes whose sum of volumes is less than N n (C nδ/N )(k+1)m = (C nδ)(k+1)m ·
N n−(k+1)m . For k > n/m − 1 we can make this sum arbitrarily small by choosing
N large enough. Hence the assertion.
DIFFERENTIAL GEOMETRY 21
Rn
b
⇀ p b
v
b
⇀
The mental picture of course is that a tangent vector v at a point p is attached
to p as it represents a small increment of the position vector of p. Now, let A be an
atlas on M and define
Ap := {(U, ϕ, ξ) | (U, ϕ) is a chart at p, ξ ∈ Rm }
On Ap we define a relation ∼p as follows. Set (U, ϕ, ξ) ∼p (V, ψ, η) if and only if
Dϕ(p) (ψ ◦ ϕ−1 )ξ = η.
U V
ϕ ψ
Lemma 2.1. Retain the above notation. The relation ∼p is an equivalence relation.
Proof. Clearly, ∼p is reflexive. As to symmetry, suppose (U, ϕ, ξ) ∼ = (V, ψ, η), that
is Dϕ(p) (ψϕ−1 )ξ = η. Hence ξ = (Dϕ(p) (ψϕ−1 ))−1 (η). However, by the chain
rule, (Dϕ(p) (ψϕ−1 ))−1 = Dψ(p) (ϕψ −1 ) and hence ξ = Dψ(p) (ϕψ −1 )η which is the
= (U2 , ϕ2 , ξ2 ) and (U2 , ϕ2 , ξ2 ) ∼
assertion. For transitivity, suppose (U1 , ϕ1 , ξ1 ) ∼ =
−1
(U3 , ϕ3 , ξ3 ). Then Dϕ1 (p) (ϕ2 ϕ1 )ξ1 = ξ2 and Dϕ2 (p) (ϕ3 ϕ−1
2 )ξ2 = ξ3 . Now observe
that on ϕ1 (U1 ∩ U2 ∩ U3 ), which contains ϕ1 (p) we have (ϕ3 ϕ−1 −1
2 )(ϕ2 ϕ1 ) = ϕ3 ϕ1 .
−1
Lemma 2.2. Let M be a manifold and let (U, ϕ) be a chart at p ∈ M . Then the
map Rm → Tp M, ξ 7→ [(U, ϕ, ξ)] is a bijection. The resulting vector space structure
on Tp M is independent of the choice of the chart (U, ϕ).
Remark 2.3. With this definition of tangent space we indeed recover the following:
Let U ⊆ Rm be an open subset, considered as a smooth m-manifold. Then using
the chart (U, Id) at any point p ∈ U leads to the identification Tp U = Rm .
The following proposition underlines once more that our definition captures the
intuition: Let N, M be manifolds and f : N → M a smooth map of constant rank
k. Then for y ∈ f (N ), the set f −1 (y) ⊆ N is a regular (n − k)-submanifold. Let us
consider f −1 (y) as an (n− k)-manifold and the injection i : f −1 (y) ֒→ N which is of
rank n − k. Then for all x ∈ f −1 (y) we have the map Dx i : Tx (f −1 (y)) → Ti(x) N .
Proposition 2.6. Retain the above notation. For every x ∈ f −1 (y) we have
The reader is encouraged to verify that this captures the intuition from above in
the case N = R3 \{0}, M = (0, ∞), f (x) = kxk2 , y = 1 and f −1 (y) = S 2 :
DIFFERENTIAL GEOMETRY 23
Proof. This can be proven directly by unravelling the definitions but there is also
a way to avoid that. Consider the constant map f ◦ i : f −1 (y) → M, x 7→ y. Then
Dx (f ◦ i) = 0. Using the chain rule we deduce Di(x) f ◦ Dx i = 0. In particular,
Dx i(Tx f −1 (y)) ⊆ ker(Di(x) f ). For the other inclusion, observe that rank of i at
any point x ∈ f −1 (y) is n − k. Hence dim Dx i(Tx f −1 (y)) = n − k. Since the
rank of f at any point is k we conclude that dim ker Di(x) f = n − k because
dim Ti(x) N − dim ker Di(x) f = dim Di(x) f = k and dim Ti(x) N = n. This implies
the converse inclusion.
2.2. Tangent Vectors and Derivations. In this section, we discuss an equiva-
lent, more algebraic definition of the tangent space which is particularly useful e.g.
in the setting of Lie groups. Let N be an n-manifold and let p ∈ N . Recall that
C ∞ (N ) := C ∞ (N, R) is an R-algebra.
Definition 2.7. Let N be a manifold and p ∈ N . A derivation of C ∞ (N ) at p is a
map δ : C ∞ (N ) → R such that
(i) δ is an R-linear map, and
(ii) (Leibniz rule) for all f, g ∈ C ∞ (N ) we have δ(f g) = δ(f )g(p) + f (p)δ(g).
Let Derp C ∞ (N ) be the vector space of derivations at p
Theorem 2.8. Let N be a manifold and p ∈ N . The map
Tp N → Derp C ∞ (N ), v 7→ (δv : f 7→ Dp f (v))
is an isomorphism of vector spaces.
Note that Definition 2.7 is much easier to write down as a definition of the
tangent space but is also a lot less transparent. Nevertheless, as noted above, it is
an important description of the tangent space. We remark that the C ∞ -assumption
in Definition 2.7 is crucial. There are much more derivations of C k (N ) than those
which come from tangent vectors.
The strategy of proof of Theorem 2.8 is to translate the problem to Rn using
charts, solve it there and translate back the solution. The following lemmas are
concerned with the Euclidean setting.
In particular, the following function is needed in
the proof: Let σ : R → R be defined by
( 1
e − x2 x > 0
σ(x) := .
0 x≤0 b
Then σ ∈ C (R).
∞ 0
Proof. We utilize the following more precise statement which will also be used in
the proof of Whitney’s embedding theorem: Let a ∈ Rn and ε > 0. Then there is
f ∈ C ∞ (Rn ) such that 0 ≤ f (x) ≤ 1 for all x ∈ Rn as well as f −1 (1) = Cε/2
n (a)
n
and f (0) = R \Cε (a).
−1 n
First, recall the function σ from above and consider the map R → R which sends
t to σ(t)σ(1/2 − t) and whose graph looks as follows:
b b
0 1
2
Rx
Next, consider the map R → R, x 7→ −∞ σ(t)σ(1/2 − t) dt. One easily determines
that its graph looks as follows:
b b
0 1
2
Finally, normalize the above function so that the constant value which it assumes
on [1/2, ∞) is one, i.e. define g : R → R by
Rx
σ(t)σ( 12 − t) dt
g(x) := R−∞
∞ 1
−∞ σ(t)σ( 2 − t) dt
b b b b b
−1 − 21 0 1 1
2
We are now in a position to prove the actual statement of the Lemma: Let
Sl Sl
ε > 0 and a1 , . . . , al ∈ K such that K ⊆ i=1 Cε/2 n
(ai ) ⊆ i=1 Cεn (ai ) ⊆ Rn \F .
For each ai (i ∈ {1, . . . , l}) pick βi as above for the hypercube Cε/2 n
(ai ) and set
Ql
f (x) = 1 − i=1 (1 − βi (x)). This function has all the right properties: Firstly, it
ranges betweeen zero and one because the βi do. Secondly, for x ∈ K there is some
i0 ∈ {1, . . . , l} such that x ∈ Cε/2
n
(ai0 ); hence βi0 (x) = 1 whence f (x) = 1. Thirdly,
if x ∈ F then x ∈ / Cε (ai ) and hence βi (x) = 0 for all i ∈ {1, . . . , l} and therefore
n
A good account for analytical aspects of the theory as the one above is [dR56].
The next lemma states that the image of a derivation on a function is determined
by the local behaviour of that function.
Lemma 2.10. Let N be a manifold and p ∈ N . Further, let δ ∈ Derp (C ∞ (N )) and
g ∈ C ∞ (N ) such that g ≡ 1 in a neighbourhood of p. Then δ(f ) = δ(f g) for all
f ∈ C ∞ (N ).
DIFFERENTIAL GEOMETRY 25
We now get to the C ∞ -part of the argument without which Theorem 2.8 fails.
Lemma 2.11. Let a ∈ Rn and U ⊆ Rn a star-shaped neighbourhood of a. Further,
let f ∈ C ∞ (U ). Then there are g1 , . . . , gn ∈ C ∞ (U ) with
∂f ∂f
(i) (g1 (a), . . . , gn (a)) = ∂x (a), . . . , ∂x (a) , and
P 1 n
Proof. (Theorem 2.8). Let (U, ϕ) be a chart at p. Pick ε > 0 such that C2ε
n
(ϕ(p)) ⊆
n
ϕ(U ). Now let g ∈ C (R ) be as f in the proof of Lemma 2.9, i.e. g equals one on
∞
n
Cε/2 (ϕ(p)) and g equals zero outside Cεn (ϕ(p)).
For f ∈ C ∞ (N ), we have δ(f ) = δ(f · (g ◦ ϕ)) by Lemma 2.10 where we agree
that g ◦ ϕ is extended to N \U by zero. This in turn equals δ(((f ◦ ϕ−1 ) · g) ◦ ϕ)
where now (f ◦ ϕ−1 ) · g is a smooth function on Rn with support in Cεn (ϕ(p)).
Now, for every F ∈ C ∞ (C2ε n
(ϕ(p))) define α(F ) := δ((F · g) ◦ ϕ) which is a
derivation of C (C2ε (ϕ(p))) at ϕ(p). For ease of notation, set a := ϕ(p). Apply
∞ n
Applying α yields
n
X n
X
α(F ) = α(F (a) · 1) + α(x 7→ (xi − ai ))Gi (a) + 0 · α(Gi )
i=1 i=1
n
X ∂F
= α(x 7→ (xi − ai )) (a).
| {z } ∂xi
i=1
ci
We therefore have
n
X ∂(f ◦ ϕ−1 )
δ(f ) = α(f ◦ ϕ−1 ) = ci ϕ(p)
i=1
∂xi
= Dϕ(p) (f ◦ ϕ−1 )(v) = (Dp f )(Dϕ(p) ϕ−1 )(v)
which is the assertion.
Now, let N and M be manifolds and let f : N → M be a smooth map. Define
f ∗ : C ∞ (M ) → C ∞ (N ) by g 7→ g ◦ f . This is an R-algebra homomorphism. Given
p ∈ N we further define
f∗ : Derp (C ∞ (N )) → Derf (p) (C ∞ (M )), δ 7→ δ ◦ f ∗ .
The map f∗ does indeed range in Derp (C ∞ (M )): Given δ ∈ Derp (C ∞ (N )) and
g1 , g2 ∈ C ∞ (M ), we have
f∗ δ(g1 g2 ) = δ(f ∗ (g1 g2 )) = δ((f ∗ g1 )(f ∗ g2 )) = δ((g1 ◦ f )(g2 ◦ f ))
= (g1 ◦ f )(p)δ(g2 ◦ f ) + (g2 ◦ f )(p)δ(g1 ◦ f )
= g1 (f (p))f∗ (δ)(g2 ) + g2 (f (p))f∗ (δ)(g1 ).
Also, the following diagram commutes:
Dp f
Tp N / Tf (p) M
∼
= ∼
=
Derp C ∞ (N ) / Derf (p) C ∞ (M )
f∗
In particular, the derivative of f can be defined without ever writing down an actual
derivative in Rn . This algebraization of the derivative is used a lot in algebraic
geometry. It is also very useful when working with vector fields as we shall see later.
However, both versions of the tangent space and the derivative are important.
2.3. The Tangent Bundle. In this section, we want to show how one can organize
the set of tangent vectors of a manifold M into a manifold TM on its own. For
instance, this is essential to define smooth vector fields and plays a major rule in
Whitney’s embedding theorem which we will see later. As a set, we have
[
TM = {x} × Tx M = {(x, v) | x ∈ M, v ∈ Tx M }.
x∈M
The tangent bundle TM of a manifold M comes with the natural projection map
π : TM → M, (x, v) 7→ x. For any subset U ⊆ M we define
TU := π −1 (U ) = {(x, v) | x ∈ U, v ∈ Tx M }
and use these subsets of TM to define a topology as follows: Let (U, ϕ) be any chart
of M . Then Dϕ : TU → ϕ(U ) × Rm given by (x, v) 7→ (ϕ(x), Dx ϕ(v)) is a bijection.
We now declare a subset E ⊆ TM to be open if and only if for every chart (U, ϕ)
of M the set Dϕ(E ∩ TU ) ⊆ ϕ(U ) × Rm is open.
DIFFERENTIAL GEOMETRY 27
Lemma 2.12. Retain the above notation. Then TM is a Hausdorff and second
countable topological space. Also, given a chart (U, ϕ) of M the map Dϕ from TU
to ϕ(U ) × Rm is a homeomorphism, and π : TM → M is continuous and open.
Proof. First of all, we show that the set
{E ⊆ TM | ∀(U, ϕ) chart of M : Dϕ(E ∩ TU ) ⊆ ϕ(U ) × Rm is open}
defines a topology on TM . Note that for a chart (U, ϕ) of M the subset TU ⊆ TM
is open. Hence so is TM since TM ∩ TU = TU . Also, the empty set is open. As to
finite intersections, suppose that E1 , . . . , En are open subsets of TM . Then
n
\ n
\
Ei ∩ TU = (Ei ∩ TU )
i=1 i=1
2.4. Whitney’s Embedding Theorem. In this section we prove that every (com-
pact) manifold can be embedded into some Euclidean space. The strong version of
this Theorem due to Whitney reads as follows.
Theorem. Let M be an m-manifold. Then M embeds into R2m .
In full generality, the dimension of the Euclidean target space cannot be reduced
any further: For instance, the two-dimensional manifold P2 R cannot be embedded
into R3 for orientability reasons. We shall prove the following version of the above
theorem.
Theorem 2.15. Let M be a compact m-manifold. Then M embeds into R2m+1 .
A byproduct of the proof of Theorem 2.15 is that in fact M can be immersed
into R2m . However, it requires a new idea to get rid of potential double points.
Proof. The first step of the proof constitutes in constructing an embedding of M
into a Euclidean space of some large dimension. To this end, recall Proposition 1.42
by which it suffices to construct an immersion of M into Euclidean space which is
a homeomorphism onto its image. We then work on reducing the dimension using
Sard’s Theorem 1.43.
Given p ∈ M , let (Up , ϕp ) be a chart at p with ϕp (p) = 0 and pick εp > 0 such
that C3ε
m
p
(0) ⊆ ϕp (Up ). Further, given any ε > 0, let ψε : Rm → [0, 1] be a smooth
function with ψε−1 (1) = Cεm (0) and ψε = 0 outside of C2εm
(0). Now define
(
ψεp ◦ ϕp (x) x ∈ Up
fp : M → R, x 7→
0 x 6∈ Up
which is smooth for every p ∈ M . Using fp we further define
(
m fp (x) · ϕp (x) x ∈ Up
Fp : M → R , x 7→ .
0 x 6∈ Up
well and assume that Ψ(x) = Ψ(y). Since Fpi and ϕpi agree on V pi we conclude
that ϕpi (x) = ϕpi (y) and hence x = y. Now suppose that y 6∈ V pi . Then fpi (y) < 1.
But since fpi (x) = 1 this implies that Ψ(x) 6= Ψ(y).
The strong version of Whitney’s embedding theorem requires an additional idea
in this step to deal with non-compactness.
As a second step, we now reduce the dimension of the target Euclidean space
utilizing Sard’s Theorem 1.43. We have an embedding of M into some large Rn and
consider TM as a regular submanifold of T Rn = Rn × Rn , and in fact
σ(M ) = {(p, v) ∈ TM | kvk = 1}
as a (2m−1)-dimensional regular submanifold of Rn × Rn . We now identify Rn−1 as
{x ∈ Rn | xn = 0} and let v ∈ Rn \ Rn−1 with kvk = 1. Further, let pv : Rn → Rn−1
be the projection parallel to v given by the decomposition Rn = Rn−1 ⊕ R v. Now
consider the map f : σ(M ) → Sn−1 given by (p, w) 7→ w and observe that pv |M is
an immersion if and only if v 6∈ f (σM ). If dim(σ(M )) = 2m−1 < n−1 = dim S n−1 ,
then by Sard’s Theorem f (σM ) has measure zero in S n−1 . Hence there is v ∈ S n−1
with v 6∈ f (σM ) for which consequently pv : M → Rn−1 is an immersion. Repeating
the argument, one eventually gets an immersion of M to R2m .
As a third step, we adress the injectivity issue. In doing so, we loose one di-
mension. The strong version of Whitney’s Theorem avoids with using a deeper
understanding about how to get rid of potentially introduced double points. Con-
sider
x−y
g : (M × M )\∆(M ) → S n−1 , (x, y) 7→
kx − yk
which, using the immersion above, is smooth as the restriction of a smooth map
from Rn × Rn . Observe that pv |M is injective if and only if v 6∈ im(g): We are
only intersted in the following direction: Suppose that pv |M is not injective. Then
there are x, y ∈ M with pv (x) = pv (y) and hence pv (x − y) = 0, i.e. x − y ∈ R v.
Therefore g(x, y) ∈ {±v} and hence v ∈ im(g). By Sard’s Theorem, the image of
g has measure zero in S n−1 if 2m < n − 1. Putting everything together, we have
that if 2m < n − 1, the set im(f ) ∪ im(g) has measure zero in S n−1 . Therefore we
can find v 6∈ im(f ) ∪ im(g) for which pv is an injective immersion and hence an
embedding by Proposition 1.42 since M is compact.
Whitney’s Theorem and Sard’s Theorem can be applied to obtain considerable
information about a given manifold. For instance, suppose that M is a compact
manifold and that f : M → R is a smooth map all of whose values are regular.
Then f −1 (y) for y ∈ im(f ) is a regular submanifold of M of lower dimension and
30 MARC BURGER STEPHAN TORNIER
ψ∗ ψ∗
C ∞ (N ) / Ω1 (N ).
dN
Also, note that Ω (M ) is not only a vector space but in fact a C ∞ (M )-module with
1
have ωx = Pi=1 ai (x)(dxi )x and the functions ai : U → R are smooth. We will also
n
write ω = i=1 ai dxi in the C ∞ (U )-module Ω1 (U ). Furthermore, if V ⊆ Rm is
open we will also consider V as a smooth manifold with the single chart (V, id) and
use the notation dxi ∈ Ω1 (V ) for the 1-form obtained by using the chart (V, id).
Lemma 3.2. Retain the above notation and let µ be a multilinear k-form on V .
Then the following are equivalent.
(i) The form µ is zero whenever two arguments coincide.
(ii) The form µ changes its sign whenever two arguments are interchanged.
(iii) For all v1 , . . . , vk ∈ V and σ ∈ Sk : µ(vσ(1) , . . . , vσ(k) ) = sign(σ)µ(v1 , . . . , vk ).
Based on Lemma 3.2 we make the following definition.
Definition 3.3. Retain the above notation. A multilinear, alternating k-form on V
is a multilinear form on V satisfying one of the equivalent properties of Lemma 3.2.
Proof. (Lemma 3.2). We show that (i) implies (ii): Let v1 , . . . , vk ∈ V . Then we
have for all 1 ≤ i < j ≤ n:
0 = µ(v1 , . . . , vi−1 , vi + vj , vi+1 , . . . , vj−1 , vi + vj , vj+1 , . . . , vn )
= µ(. . . , vi , . . . , vi , . . .) + µ(. . . , vi , . . . , vj , . . .)+
+ µ(. . . , vj , . . . , vi , . . .) + µ(. . . , vj , . . . , vj , . . .)
= µ(. . . , vi , . . . , vj , . . .) + µ(. . . , vj , . . . , vi , . . .).
Hence the assertion. Also, (ii) implies (iii): Rephrasing (ii), we have
µ(vτ (1) , . . . , vτ (k) ) = (−1) · µ(v1 , . . . , vk )
for every transposition τ ∈ Sk . Now use the fact that every permutation σ ∈ Sk
can be written as a product of transpositions and that sign : Sk → {±1} is a group
homomorphism. The implications (iii)⇒(ii) and (ii)⇒(i) are immediate.
We now organize the alternating forms on a vector space into a graded algebra:
First of all, given k ∈ N0 , let Λk (V ∗ ) denote the vector space of alternating k-forms
on V where we define Λ0 (V ∗ ) := R. In particular, we have Λ1 (V ∗ ) = V ∗ . Now
consider the graded vector space
M
Λ∗ (V ∗ ) = Λk (V ∗ ),
k≥0
where equality is left as an exercise. The special kind of permutation of {1, . . . , p+q}
that occurs in the second expression is called (p, q)-shuffle for apparent reasons.
Proposition 3.4. Retain the above notation. Then Λ∗ (V ∗ ) is an associative, graded
commutative R-algebra.
Before we proceed to the proof of Proposition 3.4 several remarks are in order.
Remark 3.5. Let α ∈ Λp (V ∗ ) and β ∈ Λq (V ∗ ).
(i) If p = 0 then α ∧ β = α · β.
(ii) The term graded-commutative means that β∧α = (−1)pq α∧β. In particular,
even order forms commute with any other form.
DIFFERENTIAL GEOMETRY 33
(iii) The definition of the wedge product may be motivated as follows: Given an
multilinear k-form µ on V there is a natural alternating k-form Aµ on V
associated to µ, namely
1 X
Aµ (v1 , . . . , vk ) := sign(σ)µ(vσ(1) , . . . , vσ(k) ).
k!
σ∈Sk
which equals
1 X
sign(σ)α(vσ(1) , . . . , vσ(p) ) β(vσ(p+1) , . . . , vσ(p+q) )·
p!q!r!
σ∈Sp+q+r ·γ(vσ(p+q+1) , . . . , vσ(p+q+r) ).
Computing α ∧ (β ∧ γ) in a similar way yields the same. This proves associativity.
We now turn to proving that Λ∗ (V ∗ ) is graded commutative: Let α ∈ Λp (V ∗ ),
β ∈ Λq (V ∗ ) and v1 , . . . , vp+q ∈ V . Then
1 X
(α ∧ β)(v1 , . . . , vp+q ) = sign(σ)α(vσ(1) , . . . , vσ(p) )β(vσ(p+1) , . . . , vσ(p+q) )
p!q!
σ∈Sp+q
Again, we use that the map Sp+q → Sp+q , σ 7→ στ is a bijection for any τ ∈ Sp+q .
Applying this to
1 ··· p p + 1 ··· p + q
τ := ∈ Sp+q
q + 1 ··· q + p 1 ··· q
which has sign (−1)pq yields
1 X
sign(στ )α(vστ (1) , . . . , vστ (p) )β(vστ (p+1) , . . . , vστ (p+q) )
p!q!
σ∈Sp+q
1 X
= (−1)pq sign(σ)β(vσ(1) , . . . , vσ(q) )α(vσ(q+1) , . . . , vσ(q+p) )
p!q!
σ∈Sp+q
Corollary 3.7. Let V be an n-dimensional real vector space with basis (e1 , . . . , en )
and let (e∗1 , . . . , e∗n ) denote the dual basis. Then (e∗j1 ∧ · · · ∧ e∗jk )1≤j1 <···<jk ≤n is a
basis of Λk (V ∗ ). In particular
k ∗ n
dim Λ (V ) = .
k
Proof. We show that the vectors of said tuple are linearly independent and that
Λk (V ∗ ) has at most the asserted dimension. To this end, let 1 ≤ j1 < · · · < jk ≤ n
and 1 ≤ l1 < · · · < lk ≤ n. Then by Proposition 3.6,
∗
ej1 (el1 ) · · · e∗j1 (elk )
(e∗j1 ∧ · · · ∧ e∗jk )(el1 , . . . , elk ) = det ... .. .
.
∗ ∗
ejk (el1 ) · · · ejk (elk )
Assume that the above expression is non-zero. Then in particular the first row
is non-zero which implies j1 ∈ {l1 , . . . , lk } by definition of the dual basis. Sim-
ilarly, ji ∈ {l1 , . . . , lk } for all i ∈ {1, . . . , k} and therefore ji = li for all i ∈
{1, . . . , k} by the ordering of the multi-indices. As a consequence, we have (e∗j1 ∧
· · · ∧ e∗jk )(el1 , . . . , elk ) = δj1 l1 · · · δjk lk . Suppose now that there is a linear relation
X
aj1 ···jk e∗j1 ∧ · · · ∧ e∗jk = 0 (aj1 ···jk ∈ R)
j1 <···<jk
among the asserted basis vectors. Then evaluation on (el1 , . . . , elk ) shows al1 ···lk = 0
by the above.
Now we show that Λk (V ∗ ) has at most the asserted dimension. For this, it suffices
to show that the linear map
n
e : Λk (V ∗ ) → R(k ) , ω 7→ (ω(ej1 , . . . , ejk ))1≤j1 <···<jk ≤n
is injective: Suppose e(ω) = 0 and let v1 , . . . , vk ∈ V . It suffices to show that
ω(v1 , . . . , vk ) = 0. Indeed, we compute
X X
ω(v1 , . . . , vk ) = ω v1j ej , . . . , vkj ej
j j
X X
= v1j1 · · · vkjk ω(ej1 · · · ejk ) = v1j1 · · · vkjk ω(ej1 · · · ejk )
j1 ,...,jk j1 ,...,jk
distinct
The last sum vanishes since for every k-tuple j1 , . . . , jk ∈ {1, . . . , n} of distinct
numbers there is a permutation σ ∈ Sk such that jσ(1) < · · · < jσ(k) and therefore
ω(ej1 , . . . , ejk ) = sign(σ)ω(ejσ(1) , . . . , ejσ(k) ) which vanishes by assumption.
Recall that binomial coefficients are the entries of Pascal’s triangle. In particular,
Λk (V ∗ ) = 0 for k > dim V and Λk (V ∗ ) = Λdim V −k (V ∗ ) for k ∈ {0, . . . , dim V }.
1
1 2 1
1 3 3 1
..
.
Remark 3.8. Before organizing k-forms on a manifold into a bundle we remark on
contravariance and coordinate-free definitions of the determinant of a matrix and
the cross-ratio of four lines.
36 MARC BURGER STEPHAN TORNIER
(i) Let V, W be vector spaces and let T : V → W be a linear map. Further, let
T ∗ : W ∗ → V ∗ denote the dual map of T . Then there is an induced map
Λk T ∗ : Λk (W ∗ ) → Λk (V ∗ ) defined by
Λk T ∗ (α)(v1 , . . . , vk ) := α(T v1 , . . . , T vk )
where α ∈ Λk (W ∗ ) and v1 , . . . , vk ∈ V .
(ii) Let V be an n-dimensional real vector space and let T ∈ End(V ). Then
Λn T ∗ ∈ End(Λn (V ∗ )). Since Λn (V ∗ ) is one-dimensional, so is its endomor-
phism algebra. Furthermore, the endomorphism algebra of a one-dimensional
real vector space E is, in contrast to the space itself, canonically isomorphic
to R via R → End(E), λ 7→ λ Id. Therefore Λn T ∗ is canonically associated
to a real number. It is an exercise to show that this number is det T .
(iii) To further illustrate the importance of the fact that the endomorphism
algebra of a one-dimensional real vector space is canonically isomorphic to
R consider the following: In geometry, one defines the cross-ratio of four
non-zero complex numbers z1 , z2 , z3 , z4 by
(z1 − z3 )(z2 − z4 )
(z1 , z2 ; z3 , z4 ) :=
(z2 − z3 )(z2 − z4 )
Now, identify C as a two-dimensional real vector space and let l1 , l2 , l3 and
l4 be the lines through the origin associated to z1 , z2 , z3 and z4 respectively.
It is a fact that for a two-dimensional vector space V its general linear group
GL(V ) acts transitively on the set of lines through the origin. It even acts
transitively on the set of triples of lines through the origin. Hence there
is no non-trivial linear invariant of such lines. However, it does not act
transitively on quadruples of lines through the origin and the cross-ratio
provides a linear invariant for such quadruples. Now, consider the map
ϕ3 : l1 → l2 given by moving parallel to l2 towards l3 and then parallel to
l1 towards l2 . Similarly, define ϕ4 : l1 → l2 . Then ϕ−1 ∼ ∗
4 ϕ3 ∈ GL(l1 ) = R is
the cross-ratio.
l2 l3
l4
l1
Λk (M ). As in the case of the tangent and cotangent bundle we equip Λk (M ) with the
structure of a smooth manifold such that π becomes a smooth map which we outline
now: Given a coordinate chart (U, ϕ) on M and x ∈ M , we have an isomorphism
DIFFERENTIAL GEOMETRY 37
Define the topology and smooth structure on Λk as before using the bijections
The pullback operation behaves well with respect to the algebra structures.
The last part of Proposition 3.11 implies that the functor which to a manifold
associates its space of k-forms is contravariant.
38 MARC BURGER STEPHAN TORNIER
Proof. (Proposition 3.11). We only prove (ii) and (iv). For (ii), let x ∈ N and
v1 , . . . , vp+q ∈ Tx N . Then
f ∗ (ω1 ∧ ω2 )x (v1 , . . . , vp+q ) = (ω1 ∧ ω2 )f (x) (Dx f v1 , . . . , Dx f vp+q )
= (ω1,f (x) ∧ ω2,f (x) )(Dx f v1 , . . . , Dx f vp+q )
1 X
= sign(σ)ω1,f (x) (Dx f vσ(1) , . . . , Dx f vσ(p) )
p!q!
σ∈Sp+q ·ω2,f (x) (Dx f vσ(p+1) , . . . , Dx f vσ(p+q) )
= (f ∗ ω1 )x ∧ (f ∗ ω2 )x (v1 , . . . , vp+q )
= (f ∗ ω1 ∧ f ∗ ω2 )x (v1 , . . . , vp+q )
which is the assertion of part (ii). Part (iv) is yet another incarnation of the chain
rule. Given ω ∈ Γ(Λk (P )) and v1 , . . . , vk ∈ Tx N we compute
(g ◦ f )∗ (ω)x (v1 , . . . , vk ) = ωgf (x) (Dx (gf )v1 , . . . , Dx (gf )vk )
= ωgf (x) (Df (x) gDx f (v1 ), . . . , Df (x) gDx f (vk ))
= g ∗ ωf (x) (Dx f v1 , . . . , Dx f vk )
= f ∗ g ∗ ωx (v1 , . . . , vk )
In order to deal with smooth k-forms we are going to need to express them in local
coordinates. To this end, recall that on Rm we have the global coordinate functions
πi : Rm → R given by x 7→ xi which provide globally defined 1-forms dπ1 , . . . , dπm .
For a coordinate chart (U, ϕ) on a manifold M we define dxi := ϕ∗ (dπi ). Then
for every p ∈ U , the 1-forms (dx1 )p , . . . , (dxm )p form a basis of (Tp M )∗ . Let now
ω ∈ Γ(Λk (M )). Then by Corollary 3.7 we have for every p ∈ U that ωp can be
written as X
ωp = ai1 ,...,ik (p)(dxi1 )p ∧ · · · ∧ (dxik )p
i1 <···<ik
in a unique way. In terms of the algebra structure defined above, this reads
X
ω|U = ai1 ,...,ik dxi1 ∧ · · · ∧ dxik .
i1 <···<ik
One often writes I to refer to a multi-index i1 < · · · < ik and defines aI := ai1 ,...,ik
as well as dxI := dxi1 ∧ · · · ∧ dxik in which case the above is condensed to
X
ω|U = aI dxI
I
f ∗ (dxi ) = f ∗ (ϕ∗ (dπi )) = (ϕ◦f )∗ (dπi ) = (ϕ◦f ◦ψ −1 ◦ψ)∗ (dπi ) = ψ ∗ (ϕ◦f ◦ψ −1 )∗ (dπi )
and therefore
n
X ∂Fi
f ∗ (dxi )p = (ψ(p)) dyj
j=1
∂yj
Overall, we have (f ∗ ω)p = a(f (p)) det(Dp f ) dx1 ∧ · · · ∧ dxm . This computation
can be generalized to k-forms in which case the coefficients appearing in f ∗ ω are
determinants of minors of Dp f .
40 MARC BURGER STEPHAN TORNIER
3.3. Partition of Unity. In this section we construct the most powerful tool for
studying smooth manifolds, namely partition of unity. It allows one to construct
various global objects out of locally defined ones, such as a Riemannian R metric
and the integral of a suitable differential form over a manifold, denotes M ω. We
demonstrate the idea in the case of a Riemannian metric which is a smooth choice of
scalar products (h−, −ix )x∈M on the tangent spaces Tx M of a manifold M . Using
these we can define the length of a C 1 -path c : [0, 1] → M by
Z 1 c(t)
b
ċ(t)
l(c) := kċ(t)kc(t) dt b
0 x b
y
This in turn allows us to define the Riemannian distance of points x, y ∈ M
associated to the Riemannian metric by
d(x, y) := inf{l(c) | c : [0, 1] → M C 1 -path with c(0) = x and c(1) = y}.
The problem lies in defining the Riemannian metric in the first place. For instance,
if (U, ϕ) is a chart of M at x ∈ M one could define h−, −ix as the pullback of the
standard scalar product on Rm = Dx ϕTx M . This works well as long as there is
only one chart involved. However, if a point x ∈ M lies in two charts domains U
and V we cannot guarantee that the scalar products so defined on Tx M coincide.
U V
x
b
compact Hausdorff, every Bi is a union of open sets with compact closure and
every open set with compact closure is a union of elements of B which hence lie
in B c . Now, we define an exhaustion of M by compact sets (Kn )n∈N with nicely
behaved interiors and closures which will aid in definining the asserted locally finite
refinement of {Vα | α ∈ A}:
Set K1 := P 1 . By compactness and the fact that {Pi | i ∈ N} is a basis there is
j ∈ N such that P1 ∪· · ·∪Pj ⊇ K1 = P 1 . In fact, we must have j ≥ 2 since there are
no non-empty open compact subsets M by the assumption that it be connected and
non-compact. Let r1 := min{P 1 ⊆ P1 ∪ · · · ∪ Pj } ≥ 2 and set K2 := P 1 ∪ · · · ∪ P r1 .
Iterate this process to define Kj := P 1 ∪· · ·∪P rj−1 . Then Kj−1 ⊆ P1 ∪· · ·∪Prj−1 but
Kj−1 6⊆ P1 ∪ · · · ∪ Pl for any l < rj−1 . Also, observe K̊j ⊇ P1 ∪ · · · ∪ Prj−1 ⊇ Kj−1 .
Therefore, K̊i+2 \Ki−1 is an open set which contains the compact set Ki+1 \K̊i .
K̊i+2
Ki+1
K̊i
Ki−1
Now consider K̊i+2 \Ki−1 ∩ Vα : For every p ∈ K̊i+2 \Ki−1 ∩ Vα let (Upi , ϕip ) be a
chart at p contained in K̊i+2 \Ki−1 ∩ Vα and such that ϕip (Upi ) = C3ε m
(0). Then
set Vp := (ϕp ) (Cε (0)). For fixed i ∈ N, consider
i i −1 m
S the set of all charts (Upi , ϕip )
obtained in this way by varying α over A. Since α∈A Vα = M we have in particular
that the sets Vpi so obtained cover Ki+1 \S K̊i . By compactness we can therefore
find a finite set Si ⊂ K̊i+2 \Ki−1 such that p∈Si Vpi ⊇ Ki+1 \K̊i . Now consider the
collection {(Upi , ϕip ) | i ∈ N, p ∈ Si }. First ofSall, the Upi of this collection form a
countable refinement of {Vα | α ∈ A}. Also, i∈N,p∈Si Vpi = M by construction. As
to local finiteness, let p ∈ M and i ∈ N such that p ∈ K̊i−1 . Observe that for all
j ≥ i we have K̊j+2 \Kj−1 ∩ K̊i−1 = ∅ and conclude by noting that for all j ≥ i and
all q ∈ Sj we have Uqj ⊆ K̊j+2 \Kj−1 whence Uqj ∩ K̊i−1 = ∅. That is, only finitely
many domains of {(Upi , ϕip ) | i ∈ N, p ∈ Si } intersect K̊i−1 .
The statement about the smooth case is immediate from the above.
Theorem 3.15. Let M be a smooth manifold and let {Vα | α ∈ A} be an open cover
of M . Then there is a countable family {fi | i ∈ F } of functions on M such that
(i) fi ≥ 0, fi ∈ C ∞ (M ) for all i ∈ F with compact support,
(ii) {supp
P fi | i ∈ F } is a locally finite cover of M refining {Vα | α ∈ A}, and
(iii) i∈F i (x) = 1 for all x ∈ M .
f
... ...
Proof. (Theorem 3.15). Let {(Ui , ϕ)i | i ∈ I} be a covering as in Lemma 3.14. Now,
let g : Rm → R with 0 ≤ g(x) ≤ 1 for all x ∈ Rm be smooth such that g ≡ 1 on
42 MARC BURGER STEPHAN TORNIER
r b
b r
Therefore we have
X
ωp = fi (p) det Dϕk (p) (ϕi ◦ ϕk−1 )(ϕ∗k (dx1 ∧ · · · ∧ dxm ))p
p∈Ui ∩Uk
Now observe that since ϕ∗k (dx1 ∧ · · · ∧ dxm )p 6= 0 we only need to show that the
coefficients in the above sum are non-zero. Indeed, since A is oriented we have
X X
fi (p) det Dϕk (p) (ϕi ◦ ϕ−1
k )≥ min det Dϕk (p) (ϕi ◦ ϕ−1
k ) fi (p).
p∈Ui ∩Uk
p∈Ui ∩Uk p∈Ui ∩Uk
| {z }
1
Hence the assertion.
Note that the argument of the second part of the proof of Proposition 3.18 works
for any manifold up to the point of proving that the coefficients do not vanish, but
they will for non-orientable manifolds.
Given a volume form ω on a manifold M , we get a consistent choice of orientation
of each Tp M (p ∈ M ) by saying that a tuple (e1 , . . . , em ) of vectors in Tp M is
positively oriented if ωp (e1 , . . . , em ) > 0.
3.5. Integrating Smooth Compactly Supported Forms And More. Let M
be a smooth manifold with an oriented atlas A and let Ωpc (M ) denote the span
in Ωp (M ) of all smooth p-forms with compact support; recall that for ω ∈ Ωp (M )
we set supp(ω) = {x ∈ M | ωx 6= 0}. Now, let (U, ϕ) ∈ A. We define a linear form
c (U ) → R, where m = dim M , by
I(U,ϕ) : Ωm
Z
I(U,ϕ) (ω) := a(x1 , . . . , xm ) dµ(x1 ) · · · dµ(xm )
Rm
m
using a dx1 ∧ · · · ∧ dxm = (ϕ ) (ω) ∈ Ωm
−1 ∗
c (ϕ(U )), where a ∈ Cc (R ) has compact
∞
support contained in ϕ(U ), and the Lebesgue measure, which can be replaced by
the Jordan content as long as ω is smooth. The orientability on M is key in proving
the following compatibility lemma.
Lemma 3.19. Let M be a manifold with an oriented atlas A. Further, let (U, ϕ)
and (V, ψ) be charts, and ω ∈ Ωm
c (U ∩ V ). Then I(U,ϕ) (ω) = I(V,ψ) (ω).
Proof. Write (ϕ−1 )∗ (ω) = a dx1 ∧ · · · ∧ dxm and (ψ −1 )∗ (ω) = b dx1 ∧ · · · ∧ dxm .
Utilizing that ϕ ◦ ψ −1 : ψ(U ∩ V ) → ϕ(U ∩ V ) is a diffeomorphism between open
subsets of Rm and the calculation at the end of Section 3.2 we conclude b(x) =
a(ϕ ◦ ψ −1 (x)) det Dx (ϕ ◦ ψ −1 ). We therefore have
Z Z
I(V,ψ) (ω) = b(x) dµ(x) = a(ϕ ◦ ψ −1 (x)) det Dx (ϕ ◦ ψ −1 ) dµ(x).
Rm Rm
Since A is oriented we have det Dx (ϕ ◦ ψ −1 ) > 0 and hence, by the classical change
of variables formula with y := ϕ ◦ ψ −1 (x), we may continue the above with
Z
= a(y) dµ(y) = I(U,ϕ) (w).
Rm
We now extend the definition of integral to the whole manifold: Let (Ui , ϕi , fi )i∈N
be such that the (Ui , ϕi ) (i ∈ N) are contained in the oriented atlas A and form
a locally finite covering, and (fi )i∈N is a partition of unity subordinate to (Ui )i∈N .
Now given ω ∈ Ωm c (M ) we set
Z X
ω := I(Ui ,ϕi ) (fi ω).
M i∈N
DIFFERENTIAL GEOMETRY 45
Now,
P let k ≥ 1 and ω ∈ Ωk (U ). Then ω can be written in a unique way as ω =
I aI dxI where the sum is taken over all ordered multi-indices of length k ranging
between 1 and n. For I = (i1 , . . . , ik ) recall that dxI := dxi1 ∧ · · · ∧ dxik .
Example 3.22. To illustrate that the above formula is far from arbitrary, consider
an open subset U ⊆ R2 and a 1-form ω = P dx + Q dy where P, Q : U → R are
46 MARC BURGER STEPHAN TORNIER
m
X X ∂2f
∂2f
= dxj ∧ dxi = dxj ∧ dxi
i,j=1
∂xj ∂xi ∂xj ∂xi
i6=j
X 2
∂ f X ∂2f
= dxj ∧ dxi + dxj ∧ dxi
i<j
∂xj ∂xi i>j
∂xj ∂xi
X ∂ f 2 X
∂2f
= dxj ∧ dxi + dxi ∧ dxj
i<j
∂xj ∂xi j>i
∂xi ∂xj
X ∂2f ∂2f
= − + dxi ∧ dxj = 0.
i<j
∂xj ∂xi ∂xi ∂xj
DIFFERENTIAL GEOMETRY 47
P
Now, if ω ∈ Ωk (U ) for k ≥ 1, write ω = I aI dxI . Then
!
X (i) X (ii) X
d(dω) = d daI ∧ dxI = d(daI ∧ dxI ) = d(daI ) ∧ dxI − daI ∧ d(dxI )
| {z } | {z }
I I I
=0 =0
where d(daI ) = 0 by the above and the term d(dxI ) vanishes by recurrence: Recall
(i)
that dxI = dxi1 ∧ · · · ∧ dxik ; therefore d(dxI ) = d(dxi1 ∧ · · · ∧ dxik ) which equals
d(dxi1 ) ∧ (dxi2 ∧ · · · ∧ dxik ) − dxi ∧ d(dxi2 ∧ · · · ∧ dxik ). Here, d(dxi1 ) vanishes by
the discussion for functions and d(dxi2 ∧ · · · ∧ dxik ) by recurrence. Thus d(dω) = 0.
For part (iv) we again first consider the case g ∈ Ω0 (U ). Then
m
! m X m
X ∂g X ∂g ∂g
∗ ∗ ∗
f (dg) = f dxi = f dxi = ◦ f f ∗ (dxi ).
i=1
∂x i i=1
∂x i i=1
∂xi
Pn ∂fi
If f = (f1 , . . . , fm ) then we have f (dxi ) = i=1 ∂yj dyj . Therefore:
∗
Xm X n
∂g ∂fi
f ∗ (dg) = ◦f dyj
i=1
∂x i j=1
∂y j
by the chain rule, computing the partial derivative with respect to yj of the function
y 7→ g(f1 (y), . . . , fm (y)). Hence
m
X
∗ ∂g ◦ f
f (dg) = dyj = d(g ◦ f ) = d(f ∗ g)
j=1
∂yj
which proves
P the statement for functions g ∈ Ω (U ). Now, if ω ∈ Ω (U ) and k ≥ 1
0 k
where the term d(f ∗ (dxI )) vanishes by recurrence as before: If dxI = dxi1 ∧· · ·∧dxik
then f ∗ (dxI ) = f ∗ dπi1 ∧ · · · ∧ f ∗ dπik and applying d yields
d(f ∗ (dxI )) = d(f ∗ d(πi1 )) ∧ f ∗ (dπi2 ) ∧ · · · ∧ f ∗ (dπik )
− f ∗ (dπi1 ) ∧ d(f ∗ (dπi2 ) ∧ · · · ∧ f ∗ (dπik ))
in which d(f ∗ (dπi1 )) = f ∗ (ddπi1 ) = 0 and d(f ∗ (dπi2 ) ∧ · · · ∧ f ∗ (dπik )) vanishes by
recurrence. This proves the assertion.
We now have an understanding of how the exterior derivate behaves with respect
to the various operations for forms defined on open subsets of Euclidean space.
Suppose next that M is a manifold and ω ∈ Ωk (M ). We claim that there is a
48 MARC BURGER STEPHAN TORNIER
P
well-defined (k + 1)-form dω ∈ Ωk+1 (M ) such that if I aI dxI is the P expression
of ω in a chart (U, ϕ) then the expression of dω in the same chart is I daI ∧ dxI .
For this, recall that aI is a smooth function on U and that dxI = ϕ∗ (dπI ) where
dπI = dπi1 ∧ · · · ∧ dπik . To see that dω ∈ Ωk+1 (M ) is well-defined we rewrite
X X X
daI ∧ dxI = d(aI ◦ ϕ−1 ◦ ϕ) ∧ ϕ∗ (dπI ) = d(ϕ∗ (aI ◦ ϕ−1 )) ∧ ϕ∗ (dπI )
I I I
! !
X X
= ϕ∗ d(aI ◦ ϕ−1 ) ∧ dπI = ϕ∗ d aI ◦ ϕ−1 dπI = ϕ∗ d((ϕ−1 )∗ ω)
I I
Note that the exterior derivative d in the last expression is indeed applied to a
form (ϕ−1 )∗ ω defined on Euclidean space. In order to show that d is well-defined
on manifolds we now show that given any two charts (Uα , ϕα ) and (Uβ , ϕβ ) on M
we have
−1 ∗
ϕ∗α (d(ϕ−1 ∗ ∗
α ) (ω)) = ϕβ (d(ϕβ ) (ω)).
Let θαβ : ϕβ (Uα ∩ Uβ ) → ϕα (Uα ∩ Uβ ) denote the coordinate transformation from
α to β and set ωα := (ϕ−1
α ) ω. Then
∗
0
d−1
/ Ω0 (M ) d0
/ Ω1 (M ) d1
/ ··· / Ωk−1 (M ) dk−1/ Ωk (M ) dk
/ Ωk+1 (M ) / ···
Since dk ◦ dk−1 = 0 we have im dk−1 ⊆ ker dk . We define the k-th de Rham coho-
mology of M as the quotient vector space HkdR (M ) := ker dk / im dk−1 . For instance,
we have H0dR (M ) = ker d0 ∼ = Rπ0 (M) where π0 (M ) is the number of connected
components of M . Thus, in a sense, de Rham cohomology spaces measure higher
connectivity problems. Straight from the definition one may also view them as ob-
structions to solving certain differential equations: Given ω ∈ Ωk (M ), is there a
DIFFERENTIAL GEOMETRY 49
D
ϕ
b b
U ϕ(U )
Note that Definition 3.26 in particular implies that ∂D is a regular (m − 1)-
submanifold of M .
Example 3.27. The reader is invited to convince himself that the following are
examples of regular domains.
(i) D = {x ∈ Rm | kxk ≤ 1}.
(ii) Let N be a smooth manifold and set M := N × R. Then N × [0, 1] is a
regular domain in M with boundary ∂D = N × {0} ∪ N × {1}.
0 1
Theorem 3.28. Let M be an oriented manifold of dimension at least two and let
D be a regular domain in M . Then ∂D is orientable and the orientation of M
canonically determines an orientation on ∂D.
Proof. Let A be the set of all charts as in the definition of a regular domain.
Using a volume form on M we can orient A as in the proof of Proposition 3.18
50 MARC BURGER STEPHAN TORNIER
∂xm (0) > 0 which implies det Mm−1 > 0 by the above. Now, for every (U, ϕ) ∈ A
∂Fm
The proof of Theorem 3.29 reduces to the fundamental theorem of calculus. The
reduction however, employs all our knowledge on differential forms.
Proof. Let (Ui , ϕi )i∈I be a countable, locally finite covering of M taken from an
oriented atlas and such that (Ui , ϕi ) is as in Definition 3.26 whenever Ui ∩ ∂D 6= ∅.
Furthermore, let {fi | i ∈ I} be a partition of unity subordinate to (Ui , ϕi )i∈I .
We now give a proof in the case where ω has compact support. P It constitutes in
computing both sides of the stated equality. First of all ω = i∈I fi ω contains only
finitely many non-zero terms since supp ω is compact. Hence
Z XZ Z XZ
dω = d(fj ω) and i∗ ω = i∗ (fj ω)
D D g
∂D g
∂D
j∈I j∈I
R R
It therefore suffices to show D d(fj ω) = g
∂D
i∗ (fj ω) for all j ∈ I which amounts
to showing that Z Z
d(fj ω) = i∗ (fj ω).
D∩Uj ^j
∂D∩U
By definition, the left hand side of the above is computed as follows. Consider
ϕj : Uj → Cεm (0). Then
m
X
(ϕ−1 ∗ cl ∧ · · · ∧ dxm
gl dx1 ∧ · · · ∧ dx
j ) (fj ω) =
l=1
for smooth functions gj : Rm → R with compact support contained in Cεm (0). Using
the fact that pulling back commutes with taking the exterior derivative we compute
(ϕ−1 ∗ −1 ∗
j ) (dfj ω) = d((ϕj ) (fj ω))
m
X
= cl ∧ · · · ∧ dxm
dgl ∧ dx1 ∧ · · · ∧ dxl ∧ · · · ∧ dx
l=1
m m
!
X X ∂gl
= dxs cl ∧ · · · ∧ dxm
∧ dx1 ∧ · · · ∧ dx
s=1
∂xs
l=1
m
X ∂gl cl ∧ · · · ∧ dxm
= dxl ∧ dx1 ∧ · · · ∧ dx
∂xl
l=1
m
X ∂gl
= (−1)l−1 dx1 ∧ · · · ∧ dxm .
∂xl
l=1
We therefore obtain
Z Z Xm
∂gl
d(fj ω) = (−1)l−1 dx1 · · · dxm
D∩Uj ϕj (D∩Uj ) l=1 ∂x l
Z Z m
!
X
l−1 ∂gl
= dxm dx1 · · · dxm−1 (−1)
xm ≥0 Rm−1 ∂xl
l=1
of which we compute each summand individually: If l = m we have
Z Z
m−1 ∂gm
(−1) dxm dx1 · · · dxm
xm ≥0 R m−1 ∂xm
Z Z
m−1 ∂gm
= (−1) dx1 · · · dxm−1 dxm
R m−1 x ≥0 ∂xm
| m {z }
=−gm (x1 ,...,xm−1 ,0)
Z
= (−1)m gm (x1 , . . . , xm−1 , 0) dx1 · · · dxm−1 .
Rm−1
52 MARC BURGER STEPHAN TORNIER
where pm (x1 , . . . , xm ) := (x1 , . . . , xm−1 ) and I(x1 , . . . , xm−1 ) := (x1 , . . . , xm−1 , 0).
Then ((pm ◦ ϕj )−1 )∗ (i∗ (fj ω)) = I ∗ ((ϕj−1 )∗ (fj ω)). Recall that
m
X
(ϕ−1 ∗ cl ∧ · · · ∧ dxm
gl dx1 ∧ · · · ∧ dx
j ) (fj ω) =
l=1
and note that
cl ∧ · · · ∧ dxm ) = I ∗ (dx1 ) ∧ · · · ∧ I\
I ∗ (dx1 ∧ · · · ∧ dx ∗ (dx ) ∧ · · · ∧ I ∗ (dx )
l m
which is supported in ϕ(F ). One readily computes that dα = (ϕ−1 )∗ ω and hence
we may put η := ϕ∗ α.
The second lemma deals with adjusting the support of a general m-form to be
contained in a set F as in Lemma 3.32.
Lemma 3.33. Let M be a connected, compact manifold and let (U, ϕ) be a chart of
M such that F ⊂ U satisfies ϕ(F ) = [0, 1]m . If ω ∈ Ωm (M ) there is η ∈ Ωm−1 (M )
such that supp(ω + dη) ⊂ F ,
Proof. The proof proceedes in two steps.
(i) Claim: Let (Vi , ϕi , Fi )ni=0 be such that for every i ∈ {0, . . . , n} the tuple
(Vi , ϕi ) is a chart with Fi ⊂ Vi satisfying ϕi (Fi ) = [0, 1]m . Furthermore,
assume that F̊i−1 ∩ F̊i 6= ∅ for 1 ≤ i ≤ n and let ω0 ∈ Ωm (M ) be supported
in F0 . Then there is η ∈ Ωm−1 (M ) with supp(ω0 + dη) ⊆ Fn .
b b b
V0 , F0 V1 , F1 V2 , F2 Vn , Fn
54 MARC BURGER STEPHAN TORNIER
R Choose m-forms
R ω1 , . . . , ωn with supp ωi ⊆ F̊i−1 ∩ F̊i and such that
ω
F̊j−1 j
= ω
F̊j−1 j−1
forR all j ∈ {1, . . . , n}. Then supp(ωi+1 − ωi ) ⊆ Fi
for 0 ≤ i ≤ n − 1 and Vi ωi+1 − ωi = 0. Then by Lemma 3.32 there are
ηi ∈ Ωm−1 (M ) such that dηi = ωi+1 − ωi for all 0 ≤ i ≤ n − 1. Hence
ωn = (ωn − ωn−1 ) + (ωn−1 − ωn−2 ) + · · · + (ω1 − ω0 ) + ω0
n−1
X
= dηi + ω0
i=0
Hence the assertion.
(ii) We now finish the proof of the lemma. Let (U1 , ϕ1 , F1 ), . . . , (Un , ϕn , Fn )
be charts as in step (i) and suppose (U1 , ϕ1 , F1 ) = (U, ϕ, F ). Furthermore,
Sn
assume by compactness that i=1 F̊i = M and let (fi )ni=1 be a partition of
Pn
unity subordinate to (F̊i )ni=1 . Then we may write ω = i=1 fi ω. Now, for
i ≥ 2 there is, by connectedness and step (i),
Pn a form ηi ∈ Ω
m−1
(M ) such
that supp(fi ω + dηi ) ⊆ F1 = F . Then ω + i=1 dηi has support in F .
We are now in a position to prove the following theorem.
Theorem 3.34. Let M be a compact, connected and oriented manifold. Then the
sequence
R
dm−1 M
m−1 / Ωm (M ) /R
Ω (M )
R
is exact, i.e. ker M
= im dm−1 .
R
Proof. First, we show that im dm−1 ⊆ ker M . Let ω ∈ Ωm−1 (M ). Take any chart
(U, ϕ) of M such that the closed unit ball B is contained in ϕ(U ) and define D :=
ϕ−1 (B). Then both D and M \D are regular domains. Furthermore, they share
the same boundary ∂D = ∂(M \D) but induce opposite orientations on it. As a
consequence, Stokes’ Theorem implies
Z Z Z Z Z
dω = dω + dω = ω+ ω = 0.
g ^
M D M\D ∂D ∂(M\D)
R R
We now turn to proving that ker M ⊆ im dm−1 . To this end, let ω ∈ ker M and
let (U, ϕ) be a chart such that F ⊆ U satisfies ϕ(F ) = [0,R1]m . Then by Lemma
3.33 there is η ∈ Ωm−1 (M ) with supp(ω + dη) ⊂ F . Also, M ω + dη = 0 by the
first step. Hence Lemma 3.32 implies that there is α ∈ Ωm−1 (M ) with ω + dη = dα.
Thus ω = d(α − η).
4. De Rham Cohomology
4.1. Basic Definitions. Let M be a a smooth manifold of dimension m. Recall
the sequence
d0 d1 dm−2 dm−1 0
0 / Ω0 (M ) / Ω1 (M ) / ··· / Ωm−1 (M ) / Ωm (M ) / ··· .
Since d2 = 0, we have im dk−1 ⊆ ker dk for every k ≥ 0 and we may therefore define
HkdR (M ) := ker dk / im dk−1
Elements of im dk−1 are called exact forms and elements of ker dk are called closed
forms. The de Rham cohomology spaces are fundamental invariants of M . So far
we know that
(i) H0dR (M ) ∼
= Rπ0 , and
(ii) HdR (M ) ∼
m
= R via integration if M is compact, connected and oriented.
DIFFERENTIAL GEOMETRY 55
R
Any differential form ω ∈ Ωm (M ) with M ω = 1 is said to represent the fundamen-
tal class of M . Next, we describe the functorial properties of de Rham cohomology:
Let M, N be manifolds and let f : M → N be smooth. Combining pullback via f
and exterior differentiation we obtain the diagram
d0 d1 dk−1 dk
0 / Ω0 (N ) / Ω1 (N ) / ··· / Ωk (N ) / ···
f∗ f∗ f∗
d0
d1 dk−1
dk
0 / Ω0 (M ) / Ω1 (M ) / ··· / Ωk (M ) / ···
Since pullback and exterior differentiation commute, so does the above diagram.
As a consequence, we have
f ∗ (ker dN M
k ) ⊆ ker dk and f ∗ (im dN M
k−1 ) ⊆ im dk−1 .
I I
R ❴ ❴ ❴ ❴ ❴ ❴/ R
and hence there is a unique linear map R → R which makes the diagram commute.
Definition 4.1. Let M and N be compact, connected and oriented manifolds of
dimension m. Further, let f : M → N be a smooth map. The degree of f is the
unique real number deg f ∈ R such that the linear map R → R, x 7→ (deg f ) · x
makes the above diagram commute.
In the following, we show that the degree of a smooth map when defined is
actually an integer. First, we record the following lemma which states that f is
close to being a covering map.
Lemma 4.2. Let M and N be compact manifolds of dimension m and let f : M → N
be a smooth map. Let q ∈ f (M ) be a regular value. Then f −1 (q) is finite and there
is an open connected set Vq ⊆ N containing q and for each p ∈ f −1 (q) an open
connected set Up ⊆ M containing p such that f |Up : Up → Vq is a diffeomorphism.
Proof. Since q is regular, for all p ∈ f −1 (q) the differential Dp f : Tp M → Tq N has
full rank and hence, in our case, is an isomorphism. Hence, by the inverse function
theorem, there are open neighbourhoods Up′ of p ∈ M and Vq′ of q ∈ N such that
f |Up′ : Up′ → Vq is a diffeomorphism. In particular, Up′ ∩ f −1 (q) = {p}. Therefore
f −1 (q) is discrete in M and Thence finite by compactness of M . Now set Vq to be
the connected component of p∈f −1 (q) Vq′ and Up := (f |Up′ )−1 (Vq ).
Retain the situation of Lemma 4.2 and assume in addition that M and N are ori-
ented. For each p ∈ f −1 (q), the differential Dp f : Tp M → Tq N is an isomorphism.
We define the index I(f, p) of f at p by
(
1 if Dp f is orientation-preserving
I(f, p) :=
−1 if Dp f is orientation-reversing
56 MARC BURGER STEPHAN TORNIER
and
X m
X
dω2 = ∂bJ dt + ∂bJ
π ∗ (dxj ) ∧ π ∗ (dxJ )
∂t j=1
∂xj
|J|=k−1
X Xm
∂bJ ∗
= “stuff containing dt” + π (dxj ∧ dxJ ).
j=1
∂xj
|J|=k−1
58 MARC BURGER STEPHAN TORNIER
We therefore have
X ∂aI X Xn
∂bJ ∗
dω = “terms in dx” + dt ∧ π ∗ (dxI ) − π (dxj ∧ dxJ
∂t j=1
∂xj
|I|=k |J|=k−1
and
Z 1 X ∂aI Z 1 X X m
∂b J
I(dω) = dxI dt − dt dxj ∧ dxJ
0 |I|=k ∂t 0 j=1
∂x j
|J|=k−1
whence
Z 1 X Z 1 X X m
∂bJ
d(I(ω)) = d bJ (t, x) dxJ dt = dxj ∧ dxJ dt.
0 0 j=1
∂xj
|J|=k−1 |J|=k−1
Overall, we conclude
X X
I(dω) + d(Iω) = aI (1, x) dxI − aI (0, x) dxI = i∗1 (ω) − i∗0 (ω).
|I|=k |I|=k
As mentioned in the beginning, one important application of Poincaré’s Lemma
is that homotopic maps induce the same map in cohomology. This is made precise
in the following.
Definition 4.7. Let M and N be manifolds and let f0 , f1 : N → M be smooth
maps. A homotopy between f0 and f1 is a smooth map h : R ×N → M such that
h(0, m) = f0 (m) and h(1, m) = f1 (m) for all m ∈ N . In this situation, f0 and f1
are homotopic.
Example 4.8. Let N = M = Rn , f1 = id and f0 = 0. Then h : R ×N → M given
by h(t, x) = t · x is a homotopy between f0 and f1 .
Proposition 4.9. Let M and N be manifolds and let f0 , f1 : N → M be homotopic
maps. Then f0∗ , f1∗ : H∗dR (M ) → H∗dR (N ) coincide.
Proof. Let h : R ×N → M be a homotopy between f0 and f1 and consider the
following diagram
h∗ I
Ωk−1 (M ) / Ωk+1 (R ×N ) / Ωk (N )
O O O
d d d
Ωk (M ) / Ωk (R ×N ) / Ωk−1 (N )
h∗ I
Let α ∈ ker dk . We need to show that f1∗ (α) and f0∗ (α) differ only by an exact form.
To this end, we apply Poincaré’s Lemma 4.6 to ω := h∗ (α):
i∗1 (ω) − i∗0 (ω) = dI(ω) + I(dω).
This reads
f1∗ α−f0∗ α = (h◦i1 )∗ α−(h◦i0 )∗ α = dI(h∗ α)+I(dh∗ α) = dI(h∗ α)+Ih∗ dα = dIh∗ α
which is the assertion.
Definition 4.10. Let M and N be manifolds. Then M and N are homotopy equiv-
alent if there are smooth maps f : M → N and g : N → M such that g ◦ f is
homotopic to idM and f ◦ g is homotopic to idN .
Corollary 4.11. Let M and N be manifolds. If M and N are homotopy equivalent
then HkdR (M ) ∼
= HkdR (N ) for all degrees k.
DIFFERENTIAL GEOMETRY 59
S2 U V U ∩V
Theorem 4.16. Now, let M be a manifold and write M = U ∪ V for open sets
U, V ⊆ M . For every k ≥ 0, the sequence
i∗
U ⊕V i∗
V ⊖U
0 / Ωk (M ) / Ωk (U ) ⊕ Ωk (V ) / Ωk (U ∩ V ) / 0,
where i∗U⊕V (ω) := (i∗U,M (ω), i∗V,M (ω)) for all ω ∈ Ωk (M ), and i∗V ⊖U (α, β) :=
i∗U∩V,V (β) − i∗U∩V,U (α) for all α ∈ Ωk (U ) and β ∈ Ωk (V ), is exact.
Proof. We need to show exactness at the three slots Ωk (M ), Ωk (U ) ⊕ Ωk (V ) and
Ωk (U ∩ V ). As to the first one, it is immediate that i∗U⊕V is injective: If a k-form
on M vanishes on both U and V then it vanishes identically as U and V cover M .
Consider now the slot Ωk (U ) ⊕ Ωk (V ). It is clear that im i∗U⊕V ⊆ ker i∗V ⊖U . In
order to see the converse inclusion, let (α, β) ∈ Ωk (U ) ⊕ Ωk (V ) such that (α, β) ∈
ker i∗V ⊖U , i.e. α|U∩V = β|U∩V . Thus we may define ω ∈ Ωk (M ) by setting ω|U = α
and ω|V = β to the effect that i∗U⊕V (ω) = (α, β).
Finally, we show that i∗V ⊖U is surjective: Given ω ∈ Ωk (U ∩ V ), we construct
(ωU , ωV ) ∈ Ωk (U ) ⊕ Ωk (V ) such that i∗V ⊖U (ωU , ωV ) = ω. Let fU and fV be a
partition of unity subordinate to the cover (U, V ) of M and define ωV ∈ Ωk (V ) by
(
ωV |U∩V := fU · ω
.
ωV |V \(U∩V ) = 0
In order to verify that ωV is smooth, it suffices to cover its domain V by open sets
and show that ωV is smooth on each of these open sets. In this case, we have V =
(U ∩ V ) ∪ (V ∩ (supp fU )c ) and indeed both ωV |U∩V = fU ω and ωV |V ∩(supp fU )c = 0
are smooth. Similarly, we may define a smooth ωU ∈ Ωk (U ) by
(
ωU |U∩V = −fV · ω
.
ωU |U\(U∩V ) = 0
We obtain ωV |U∩V − ωU |U∩V = ω.
For ease of notation we abbreviate
i∗ p∗
0 / Ωk (M ) / Ωk (U ) ⊕ Ωk (V ) / Ωk (U ∩ V ) /0
The reason we do not include the zero maps in the sequence in cohomology is that
i∗ need not be injective and p∗ need not be surjective: For instance, H2dR (S 2 ) = R
but H2dR (U ) and H2dR (V ) = 0. Also, in degree one, we have H1dR (U ) and H1dR (V ) = 0
but H1dR (U ∩ V ) ∼
= H1dR (S 1 ) ∼
= R.
Nonetheless, the sequence in cohomology is exact at the middle slot.
Lemma 4.17. Retain the above notation. Then im i∗ = ker p∗ .
Proof. We know that p∗ ◦i∗ = 0. Hence im i∗ ⊆ ker p∗ . For the opposite inclusion, let
(α, β) ∈ Ωk (U )⊕Ωk (V ) with dα = 0 = dβ. The assumption that the class ([α], [β]) ∈
HkdR (U ) ⊕ HkdR (V ) is in the kernel of p∗ means that β|U∩V − α|U∩V = dγ for some
γ ∈ Ωk−1 (U ∩ V ). Let (α′ , β ′ ) ∈ Ωk−1 (U ) ⊕ Ωk−1 (V ) with β ′ |U∩V − α′ |U∩V = γ.
Hence dβ ′ |U∩V − dα′ |U∩V = dγ = β|U∩V − α|U∩V . Now let ω ∈ Ωk (M ) be such
that ω|U = α − dα′ and ω|V = β − dβ ′ and finish by observing that
(
dω|U = d(α − dα′ ) = dα = 0
dω|V = d(β − dβ ′ ) = dβ = 0
DIFFERENTIAL GEOMETRY 61
We now concern ourselves with the question to which extent the above sequence
in cohomology fails to be exact in the first and the third slot. For instance, let
ω ∈ Ωk (U ∩ V ) represent an element of HkdR (U ∩ V ), i.e. dω = 0. Since the sequence
i∗ p∗
0 / Ωk (M ) / Ωk (U ) ⊕ Ωk (V ) / Ωk (U ∩ V ) /0
p∗
Ωk (U ) ⊕ Ωk (V ) / Ωk (U ∩ V ) /0
(d,d) d
p∗
i∗
0 / Ωk+1 (M ) / Ωk+1 (U ) ⊕ Ωk+1 (V ) / Ωk+1 (U ∩ V )
p∗
Ωk−1 (U ) ⊕ Ωk−1 (V ) / Ωk−1 (U ∩ V ) /0
(d,d) d
p∗
Ωk (U ) ⊕ Ωk (V ) / Ωk (U ∩ V ) /0
(d,d) d
p∗
i∗
0 / Ωk+1 (M ) / Ωk+1 (U ) ⊕ Ωk+1 (V ) / Ωk+1 (U ∩ V )
Since its first row is exact there is (a, b) ∈ Ωk−1 (U ) ⊕ Ωk−1 (V ) with p∗ (a, b) = η.
For the same reason, we may choose pre-images (α, β) of ω and (α′ , β ′ ) of ω ′ .
Then (α, β) + (da, db) is a pre-image of ω ′ by commutativity. Therefore, (α′ −
α − da, β ′ − β − db) is in the kernel of p∗ . Hence there is a unique element c ∈
Ωk (M ) such that i∗ c = (α′ − α − da, β ′ − β − db). This implies that dc and ω e′ − ω
e
map to (dα − dα, dβ − dβ). This shows that we obtain a well-defined connecting
′ ′
,
which is actually exact.
Theorem 4.18. Let M be a manifold and assume that U, V are open subsets of M
such that M = U ∪ V . Then the Mayer-Vietoris sequence above is exact.
The proof of Theorem 4.18 is not difficult but requires to “deconfuse” oneself and
hence constitutes a good exercise which is left to the reader.
Corollary 4.19. Let m ≥ 1 and k ∈ N0 . Then
(
k m R k ∈ {0, m}
HdR (S ) = .
0 otherwise
Proof. Clearly, we have HdR (S m ) = 0 for k ≥ m + 1. For the remainder, we may
assume m ≥ 2 and write S m = U ∪ V where U = S m \{S} and V = S m \{N }.
Then U ∩ V ∼ = R ×S m−1 . Now, consider the first part of the long exact sequence
for M = S . It reads
m
Indeed, U and V are both contractible, hence the assertion about their cohomology
in degree zero and one. Furthermore, U ∩ V ∼ = R ×S m−1 has the same cohomology
as S m−1
which is connected since m ≥ 2. Now, the image of i∗ equals the kernel
of p which is hence one-dimensional. Therefore, the image of p∗ is of dimension
∗
∼
=
/ Hj (S m−1 ) δ / Hj+1 (S m )
0 dR dR
/0
As a first step towards Theorem 4.20 we show that invariant forms are closed.
Then can be checked using local coordinates or with the help of the following
trick due to É. Cartan: Consider the exponential map E : Rn → T n given by
x 7→ (e2πix1 , . . . , e2πixn ). This map is a group homomorphism and a local diffeo-
morphism. In fact, it induces an isomorphism Rn / Zn → T n . Furthermore, let
s : T n → T n , ξ 7→ ξ −1 denote the inversion which is a diffeomorphism of T n fixing
1. We have D1 s = − Id: Indeed, note that s(E(x)) = E(x)−1 = E(−x) for all
x ∈ Rn and hence (D0 E)−1 ◦ D1 s ◦ D0 E = − Id which implies D1 s = − Id.
Lemma 4.22. Let ω ∈ Ωkinv (T n ). Then dω = 0.
Proof. First, observe that s∗ preserves Ωkinv (T n ): For all ξ ∈ T n we have s ◦ Lξ =
Lξ−1 ◦ s and therefore
L∗ξ (s∗ ω) = (sLξ )∗ ω = (Lξ−1 s)∗ ω = s∗ L∗ξ−1 ω = s∗ ω.
for all ξ ∈ T n , i.e. s∗ ω is invariant.
Next, we compute s∗ ω. By invariance, it suffices to determine s∗ ω at 1 ∈ T n.
Let v1 , . . . , vk ∈ T1 T n . Then
(s∗ ω)1 (v1 , . . . , vk ) = ω1 (D1 sv1 , . . . , D1 svk ) = (−1)k ω1 (v1 , . . . , vk )
Hence the invariant forms s∗ ω and (−1)k ω conincide everywhere. On the other
hand, dω ∈ Ωk+1
inv (T ) since differential and pullback commute, therefore
n
R
L([0, 1]n ) = 1. Given ω ∈ Ωk (T n ), we define Pk ω ∈ Ωk (T n ) by “ [0,1]n L∗E(x) ω d L(x)”,
that is, for ζ ∈ T n and v1 , . . . , vk ∈ Tζ T n we set
Z
(Pk ω)ζ (v1 , . . . , vk ) := (L∗E(x) ω)ζ (v1 , . . . , vk ) d L(x).
[0,1]n
Then
Pk ω − ω = dKω + Kdω = dKω,
that is, Pk ω ∈ Ωinv (T ) and ω represent the same cohomology class. This completes
k n
using the action of S 1 on S 1 × M which acts from the left on S 1 and fixes M and
the details of the Poincaré Lemma.
5. De Rham’s Theorem
Historically, de Rham’s theorem is the first instance of a comparison of different
cohomology theories. He showed that for a manifold, de Rham cohomology and
singular cohomology, which is defined in the more general context of topological
spaces, coincide. The approach to proving this theorem is via Čech cohomology
which is combinatorial in nature and can be shown to be equal to singular and de
Rham cohomology for any manifold. A classical account of this is [Wei52]. A nice
consequence of this is the finite-dimensionality of the de Rham cohomology of a
compact manifold in every degree which follows easily from Čech cohomology.
5.1. Čech Cohomology. Let X be a set and let U = {Ui | i ∈ I} ⊆ P(X) be
a collection of subsets of X; think of X being a manifold and the Ui being open
and covering
T M . The nerve of U is the set N (U) of subsets J ⊆ I such that
UJ := j∈J Uj 6= ∅. For q ≥ 0, a q-simplex is an ordered (q + 1)-tuple of indices
σ = (i0 , . . . , iq ) such that |σ| := {i0 , . . . , iq } ∈ N (U ). The j-th face of a q-simplex
σ (q ≥ 1) is the q − 1-simplex σ j = (i0 , . . . , îj , . . . , iq ).
The nerve N (U ) is an instance of what is called a simplicial complex, a combi-
natorial object for which there is a topological realization. For instance, consider
the covering {U1 , U2 , U3 } of S 1 indicated below.
{1, 2, 4}
U4
U1
{1} {4}
b b
{1, 3}
{1, 2}
U3 U2
b b
3b
(1, 3) (2, 3)
b
(1, 2, 3) b
1 (1, 2) 2
66 MARC BURGER STEPHAN TORNIER
Then (δ1 f )((1, 2, 3)) = f ((2, 3)) − f ((1, 3)) + f ((1, 2)). Thus, in a sense, δ1 yields a
combinatorial boundary of the 2-simplex (1, 2, 3) with built-in orientation.
Lemma 5.1. Retain the above notation. The map δq is linear and δq+1 ◦ δq = 0.
As a consequence of the above lemma, we obtain a complex
δ0 δ1 δ2
C 0 (U, R) / C 1 (U, R) / C 2 (U , R) / ···
q
whose cohomology is the Čech cohomology: Ȟ (U, R) := ker δq / im δq−1 . Note that
q
if I is finite then all C q (U, R) and hence all Ȟ (U, R) are finite-dimensional.
5.2. Statements. This section collects the main statements about Čech cohomol-
ogy. They are proven in the next one. First of all we apply the previous section in
the context of manifolds.
Definition 5.2. Let M be a smooth manifold and let U = (Ui )i∈I be a covering of
M by subsets of M . Then U is admissible if
(i) Ui is open for every i ∈ I,
(ii) U is locally finite, and T
(iii) for all J ∈ N (U, M ), the intersection UJ := i∈J Ui is contractible.
De Rham’s Theorem now reads as follows.
Theorem 5.3. Let M be a smooth manifold and let U be an admissible covering of
M . Then for all k ≥ 0:
k
HkdR (M ) ∼
= Ȟ (U, R).
k
Concerning the notation Ȟ (U, R) we remark that instead of taking the real
numbers as the target of functions on simplices we could have taken any abelian
group, e.g. Z. In some cases, this yields more refined invariants. In this case however,
k
it makes sense to compare the R-vector spaces Ȟ (U, R) and HkdR (M ).
Note that the statement of Theorem 5.3 is empty without having proven the
existence of admissible coverings. This is a theorem on its own.
Theorem 5.4. Let M be a smooth manifold. Then M admits an admissible covering.
One strategy to prove the existence of admissible coverings is to apply Whitney’s
Embedding Theorem 2.15 first and then argue within Euclidean space. Another
one is based on Riemannian geometry and the existence of convex neighbourhoods.
Anyway, as a Corollary to Theorem 5.3 we immediately record the following.
Corollary 5.5. Let M be a compact manifold. Then HkdR (M ) is finite-dimensional
for all k ≥ 0.
Proof. Let U be an admissible covering of M and extract a finite subcover. Observe
that since U is finite, so is S q (U) for every q ≥ 0 and hence C q (U, R) is finite-
k
dimensional whence Ȟ (U , R) is finite-dimensional for all k ≥ 0.
We remark that in the case of compact manifolds, the above implies that de
Rham cohomology can be computed by a machine if an admissible cover and its
intersection pattern are provided. This is in sharp contrast to e.g. the fundamental
group in which case not even triviality can be decided computationally.
For the next statement recall that we computed
k n n n n−k
dim HdR (T ) = = = dim HdR (T n ).
k n−k
This can be generalized to the following extent.
DIFFERENTIAL GEOMETRY 67
Theorem 5.6 (Poincaré Duality). Let M be a compact oriented manifold. Then the
pairing Z
∧ I
Ωk (M ) × Ωn−k (M ) −
→ Ωn (M ) −
→ R, (α, β) 7→ α ∧ β.
M
is non-degenerate and hence HkdR (M ) ∼ n−k
= (HdR (M ))∗ .
Often, dim HkdR (M ) is referred to as the k-th Betti number of M , denoted bk (M ).
Ω2 (M ) / Ω(2,0)
O
/ Ω(2,1)
O
/ Ω(2,2)
O
/ Ω(2,3)
O
/ ...
O
Ω1 (M ) / Ω(1,0)
O
/ Ω(1,1)
O
/ Ω(1,2)
O
/ Ω(1,3)
O
/ ...
O
Ω0 (M ) / Ω(0,0)
O
/ Ω(0,1)
O
/ Ω(0,2)
O
/ Ω(0,3)
O
/ ...
Lemma 5.8. Retain the above notation. All squares in the above diagram commute.
Proof. Let Ω = (ωσ )σ∈S p (U ) ∈ Ω(k,p) and consider δΩ := (ωη′ )η∈S p+1 (U) . By defini-
tion,
p+1
X
ωη′ = (−1)ν ωην |U|η|
ν=0
where η ν is the ν-th face of η, i.e. η ν = (i0 , . . . , îν , . . . , ip+1 ). Now applying d to δΩ
yields dδΩ = (dωη′ )η∈S p+1 (U) where
p+1
X p+1
X
dωη′ = (−1)ν d(ωην |U|η| ) = (−1)ν (dωην )|U|η|
ν=0 ν=0
On the other hand, δdΩ = δ(dωσ )σ∈S p (U) = (ωη′′ )η∈S p+1 (U) where
p+1
X
ωη′′ = (−1)ν (dωην )|U|η| .
ν=0
Hence the assertion.
5.3.1. Homotopies. All vertical and horizontal complexes in the above diagram are
in fact exact. This is the content of the two lemmas in this section. Since for every
σ ∈ S p (U) the set U|σ| is contractible, the Poincaré Lemma 4.6 yields a linear
map I|σ| : Ωm (U|σ| ) → Ωm−1 (U|σ| ) (m ≥ 1) such that ω = dI|σ| ω + I|σ| dω for all
ω ∈ Ωm (U|σ| ). Now define
I : Ωm,p → Ωm−1,p , (ωσ )σ∈S p (U) → (I|σ| ωσ )σ∈S p (U ) .
Then the following is immediate.
Lemma 5.9. Retain the above notation. Then Ω = dIΩ + IdΩ for all Ω ∈ Ω(m,p)
with m ≥ 1 and p ≥ 0.
In particular, all vertical complexes are exact.
Now we define maps K : Ω(m,p) → Ω(m,p−1) (p ≥ 1) and K : Ω(m,0) → Ωm (M ),
playing a role analogous to the maps I above for horizontal complexes. To this end,
let (fi )i∈I be a partition
Pof unity subordinate to U. First, we define K : Ω
(m,0)
→
Ω (M ) by (ωi )i∈I 7→
m
i∈I (fi ωi ) where fi ωi is extended to the whole of M by
zero. Similarly, for p ≥ 1 and Ω = (ωσ )σ∈S p (U ) we set KΩ := (ζη )η∈S p−1 (U) where
X
ζi0 ,...,ip−1 = fk ωk,i0 ,...,ip−1
k∈I
δ
Ω(m−1,0) / Ω(m−1,1)
O
d
Ω(m−2,1)
This can be made into a well-defined map from Zm
→ Zm (U, R) and we will
dR (M )
show that it induces the sought-after isomorphism in cohomology. To this end,
define for 0 ≤ n ≤ m − 1:
F m,n := {Ω ∈ Ω(m−1−n,n) | dδΩ = 0}
and
Hm,n := {Ω ∈ Ω(m−1−n,n) | Ω = X + Y with dX = 0 and δY = 0}.
Lemma 5.11. Retain the above notation. The map
δ I
Iδ : Zm → Ω(m,0) −
dR (M ) − → Ω(m−1,0)
takes values in F (m,0) and induces an isomorphism
Hm
dR (M ) → F m,0 / Hm,0 .
Proof. For ω ∈ ZmdR (M ) we have d(δω) = δ(dω) = 0 and hence δω = dI(δω). From
this we deduce 0 = δ 2 ω = δd(Iδ(ω)) and hence Iδ(ω) ∈ F m,0 .
Now, let Ω0 ∈ F m,0 . Then δ(dΩ0 ) = 0 and hence dΩ0 ∈ Ω(m,0) is in the image
of δ: Let ω ∈ Ωm (M ) with δω = dΩ0 . Then we have
δ(dω) = dδω = d2 Ω0 = 0
and since δ : Ωm+1 (M ) → Ω(m+1,0) is injective we get dω = 0, that is ω ∈ Zm
dR (M ).
Thus dIδω = dΩ0 and hence Ω0 − Iδω ∈ Hm,0 . This shows that the composition of
Iδ with the projection Zm dR (M ) → F m,0 / Hm,0 is surjective.
Finally, let ω = dα for some α ∈ Ωm−1 (M ). Then δω = δdα = d(δα). Therefore,
d(Iδω) = δω = d(δα) and hence Iδω = δα + β with dβ = 0, i.e. Iδω ∈ Hm,0 .
Conversely, if ω ∈ Zm dR (M ) and Iδω = α + β with dα = 0 and δβ = 0 then
δω = d(Iδω) = dβ. But since δβ = 0 there is β ′ ∈ Ωm−1 (M ) with δβ ′ = β and
hence δω = dδβ ′ = δdβ ′ whence ω = dβ ′ .
Lemma 5.12. Retain the above notation. The map
δ : F m,m−1 → Ω(0,m)
takes values in Zm (U , R) and induces an isomorphism
m
F m,m−1 / Hm,m−1 → Ȟ (U, R).
Proof. Let Ω ∈ F m,m−1 ⊆ Ω(0,m−1) . Then dδΩ = 0 and δΩ = (fσ )σ∈S m (U) with fσ
constant on U|σ| . Thus δΩ ∈ C m (U, R). In addition δΩ ∈ Zm (U , R) since δ 2 Ω = 0.
Now, let c ∈ Zm (U, R). Then dc = 0 and δc = 0. Hence there is Ω ∈ Ω(0,m−1)
with δΩ = c. Clearly, dδΩ = dc = 0 and hence Ω ∈ F m,m−1 . Therefore the map
δ : F m,m−1 → Zm (U, R) is surjective.
Finally, let Ω ∈ Hm,m−1 , i.e. Ω = α + β with dα = 0 and δβ = 0. Then δΩ = δα
m−1 m
and since dα = 0 we have α ∈ Č (U, R). Hence δΩ = δα ∈ Č (U, R). Conversely,
70 MARC BURGER STEPHAN TORNIER
m−1
assume that δΩ = δγ with γ ∈ Č (U , R). Then dγ = 0 and δ(Ω − γ) = 0 and
therefore Ω ∈ Hm,m−1 .
Now, for 0 ≤ n ≤ m − 2 consider the following diagram:
δ
Ω(m−1−n,n) / Ω(m−1−n,n+1)
O
d
Ω(m−1−(n+1),n+1) .
Lemma 5.13. Retain the above notation. Then the map
δ : F m,n → Ω(m−1−(n+1)),n+1)
takes values in F m,n+1 and induces an isomorphism
F m,n / Hm,n → F m,n+1 / Hm,n+1 .
The proof of Lemma 5.13 is similar to the proofs of Lemmas 5.11 and 5.12 and
is left to the reader. Overall, this proves Theorem 5.3.
DIFFERENTIAL GEOMETRY 71
References
[Boo03] W. M. Boothby, An Introduction to Differentiable Manifolds and Riemannian Geometry,
vol. 120, Gulf Professional Publishing, 2003.
[BT03] D. Barden and Ch. Thomas, An Introduction to Differential Manifolds, vol. 10, AMC,
2003.
[dR56] G. de Rham, Variétés différentiables, 1956.
[Ker60] M. A. Kervaire, A manifold which does not admit any differentiable structure, Commen-
tarii Mathematici Helvetici 34 (1960), no. 1, 257–270.
[Mil56] John Milnor, On manifolds homeomorphic to the 7-sphere, Annals of Mathematics 64
(1956), no. 2, pp. 399–405.
[Mil97] J. W. Milnor, Topology from the Differentiable Viewpoint, Princeton University Press,
1997.
[RSN72] F. Riesz and B. Szökefalvi-Nagy, Leçons d’analyse fonctionnelle, Gauthier-Villars 1
(1972).
[Rud87] W. Rudin, Real and complex analysis, Tata McGraw-Hill Education, 1987.
[RW13] R. Remmert and H. Weyl, Die Idee der Riemannschen Fläche, vol. 5, Springer-Verlag,
2013.
[Wei52] André Weil, Sur les théorèmes de de Rham, Commentarii Mathematici Helvetici 26
(1952), no. 1, 119–145.