Math 752 Topology Lecture Notes

Laureniu Maxim
March 8, 2013

Contents
1 Selected topics in Homology
1.1 Cellular Homology . . . . . . . . . . . . . . . . . . . . . . . .
1.1.1 Degrees . . . . . . . . . . . . . . . . . . . . . . . . . .
1.1.2 How to Compute Degrees? . . . . . . . . . . . . . . . .
1.1.3 CW Complexes . . . . . . . . . . . . . . . . . . . . . .
1.1.4 Cellular Homology . . . . . . . . . . . . . . . . . . . .
1.2 Euler Characteristic . . . . . . . . . . . . . . . . . . . . . . . .
1.3 Lefschetz Fixed Point Theorem . . . . . . . . . . . . . . . . .
1.4 Homology with General Coefficients . . . . . . . . . . . . . . .
1.5 Universal Coefficient Theorem for Homology . . . . . . . . . .
1.5.1 Tensor Products . . . . . . . . . . . . . . . . . . . . . .
1.5.2 The Tor functor and the Universal Coefficient Theorem

.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.

3
3
3
6
8
9
18
20
23
26
26
28

2 Basics of Cohomology
2.1 Cohomology of a chain complex: definition . .
2.2 Relation between cohomology and homology .
2.2.1 Ext groups . . . . . . . . . . . . . . . .
2.2.2 Universal Coefficient Theorem . . . . .
2.3 Cohomology of spaces . . . . . . . . . . . . .
2.3.1 Definition and immediate consequences
2.3.2 Reduced cohomology groups . . . . . .
2.3.3 Relative cohomology groups . . . . . .
2.3.4 Induced homomorphisms . . . . . . . .
2.3.5 Homotopy invariance . . . . . . . . . .
2.3.6 Excision . . . . . . . . . . . . . . . . .
2.3.7 Mayer-Vietoris sequence . . . . . . . .
2.3.8 Cellular cohomology . . . . . . . . . .

.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.

33
33
34
34
35
36
36
38
39
40
40
41
42
43

3 Cup Product in Cohomology

3.1 Cup Products: definition, properties, examples
3.2 Application: Borsuk-Ulam Theorem . . . . . .
3.3 Knneth Formula . . . . . . . . . . . . . . . .
3.3.1 Cross product . . . . . . . . . . . . . .

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

49
49
60
64
64

3.3.2
3.3.3

Knneth theorem in cohomology. Examples . . . . . . . . . . . . . .

Knneth exact sequence and applications . . . . . . . . . . . . . . . .

4 Poincar Duality
4.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . .
4.2 Manifolds. Orientation of manifolds . . . . . . . . . . . . . .
4.3 Cohomolgy with Compact Support . . . . . . . . . . . . . .
4.4 Cap Product and the Poincar Duality Map . . . . . . . . .
4.5 The Poincar Duality Theorem . . . . . . . . . . . . . . . .
4.6 Immediate applications of Poincar Duality . . . . . . . . . .
4.7 Addendum to orientations of manifolds . . . . . . . . . . . .
4.8 Cup product and Poincar Duality . . . . . . . . . . . . . .
4.9 Manifolds with boundary: Poincar duality and applications
4.9.1 Signature . . . . . . . . . . . . . . . . . . . . . . . .
4.9.2 Connected Sums . . . . . . . . . . . . . . . . . . . .
5 Basics of Homotopy Theory
5.1 Homotopy Groups . . . . . .
5.2 Relative Homotopy Groups . .
5.3 Homotopy Groups of Spheres
5.4 Whiteheads Theorem . . . .
5.5 Fibrations and Fiber Bundles

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.

65
70

.
.
.
.
.
.
.
.
.
.
.

73
. 73
. 74
. 79
. 82
. 85
. 90
. 92
. 96
. 99
. 101
. 102

.
.
.
.
.

104
104
110
113
116
120

.
.
.
.
.

Chapter 1
Selected topics in Homology
Note: Knowledge of simplicial and singular homology will be assumed.

1.1

Cellular Homology

1.1.1

Degrees

Definition 1. The degree of continuous map f : S n S n is defined as:

deg f = f (1)

(1.1)

e n (S n ) = Z H
e n (S n ) = Z is the homomorwhere 1 Z denotes the generator, and f : H
phism induced by f in homology.
The degree has the following properties:
1. deg idS n = 1.
Proof. This is because (idS n ) = id which is multiplication by the integer 1.
2. If f is not surjective, then deg f = 0.

>

Proof. Indeed, suppose f is not surjective, then there is a y

/ Imagef . Then we can
factor f in the following way:
f
Sn
> Sn
g

>

S n \ {y}
Since S n {y}
= Rn which is contractible, Hn (S n \ {y}) = 0. Therefore f = h g = 0,
so deg f = 0.

3. If f
= g, then deg f = deg g.
Proof. This is because f = g . Note that the converse is also true (by a theorem of
Hopf).
4. deg(g f ) = deg g deg f .
Proof. Indeed, we have that (g f ) = g f .
5. If f is a homotopy equivalence (so there exists a g so that gf ' idS n ), then deg f = 1.
Proof. This follows directly from 1, 3, and 4 above, since f g
= idS n implies that
deg f deg g = deg idS n = 1.
6. If r : S n S n is a reflection across some n-dimensional subspace of Rn+1 , that is,
r(x0 , . . . xn ) 7 (x0 , x1 , . . . , xn ), then deg r = 1.
Proof. Without loss of generality we can assume the subspace is Rn {0} Rn1 .
Choose a CW structure for S n whose n-cells are given by n1 and n2 , the upper
and lower hemispheres of S n , attached by identifying their boundaries together in the
standard way. Then consider the generator of Hn (S n ): [n1 n2 ]. The reflection map
r maps the cycle n1 n2 to n2 n1 = (n1 n2 ). So
r ([n1 n2 ]) = [n2 n1 ] = [(n1 n2 )] = 1 [n1 n2 ]
so deg r = 1.
7. If a : S n S n is the antipodal map (x 7 x), then deg a = (1)n+1
Proof. Note that a is a composition of n+1 reflections, since there are n+1 coordinates
in x, each getting mapped by an individual reflection. From 4 above we know that
composition of maps leads to multiplication of degrees.
8. If f : S n S n and Sf : S n+1 S n+1 is the suspension of f then deg Sf = deg f .
Proof. Recall that if f : X X is a continuos map and
X = X [1, 1]/(X {1}, X {1})
denotes the suspension of X, then Sf := f id[1,1] / , with the same equivalence as
in X. Note that S n = S n+1 .
The Suspension Theorem states that
e i (X)
e i+1 (X).
H
=H
4

This can be proved by using the Mayer-Vietoris sequence for the decomposition
X = C+ X X C X,
where C+ X and C X are the upper and lower cones of the suspension joined along
their bases:
e n+1 (C+ X) H
e n+1 (C X) H
e n+1 (X) H
e n (X) H
e n (C+ X) H
e n (C X)
H
Since C+ X and C X are both contractible, the end groups in the above sequence are
e i (X)
e i+1 (X), as desired.
both zero. Thus, by exactness, we get H
=H
n
n
Let C+ S denote the upper cone of S . Note that the base of C+ S n is S n {0} S n .
Our map f induces a map C+ f : (C+ S n , S n ) (C+ S n , S n ) whose quotient is Sf . The
long exact sequence of the pair (C+ S n , S n ) in homology gives the following commutative
diagram:
0

e i+1 (C+ S n , S n ) ' H

e i+1 (C+ S n /S n ) > H
e i (S n )
>H

(Sf )

e
Hi+1 (S n+1 )

>0

e
> Hi (S n )

Note that C+ S n /S n
= S n+1 so the boundary map at the top and bottom of the
diagram are the same map. So by the commutativity of the diagram, since f is
defined by multiplication by some integer m, then (Sf ) is multiplication by the same
integer m.
Example 1. Consider the reflection map: rn : S n S n defined by (x0 , . . . , xn ) 7
(x0 , x1 , . . . , xn ). Since rn leaves x1 , x2 , . . . , xn unchanged we can unsuspend one at a
time to get
deg rn = deg rn1 = = deg r0 ,
where ri : S i S i by (x0 , x1 , . . . , xi ) 7 (x0 , x1 , . . . , xi ). So r0 : S 0 S 0 by
x0 7 x0 . Note that S 0 is two points but in reduced homology we are only looking
at one integer. Consider


e 0 (S 0 ) H0 (S 0 )
0H
Z0
e 0 (S 0 ) = {(a, a) | a Z}, H0 (S 0 ) = Z Z, and  : (a, b)
where H
7 a + b. Then
0
0
e 0 (S ) H
e 0 (S ) is given by (a, a) 7 (a, a) = (1)(a, a). So deg rn = 1.
(r0 ) : H
9. If f : S n S n has no fixed points then deg f = (1)n+1 .
Proof. Consider the above figure. Since f (x) 6= x, the segment (1t)f (x)+t(x) from
x to f (x) does not pass through the origin in Rn+1 so we can normalize to obtain a
homotopy:
(1 t)f (x) + t(x)
: S n S n.
gt (x) :=
|(1 t)f (x) + t(x)|
5

Note that this homotopy is well defined since (1 t)f (x) tx 6= 0 for any x S n and
t [0, 1], because f (x) 6= x for all x. Then gt is a homotopy from f to a, the antipodal
map.
Exercises
1. Let f : S n S n be a map of degree zero. Show that there exist points x, y S n with
f (x) = x and f (y) = y.
2. Let f : S 2n S 2n be a continuous map. Show that there is a point x S 2n so that
either f (x) = x or f (x) = x.
3. A map f : S n S n satisfying f (x) = f (x) for all x is called an even map. Show that
an even map has even degree, and this degree is in fact zero when n is even. When n is odd,
show there exist even maps of any given even degree.

1.1.2

How to Compute Degrees?

Assume f : S n S n is surjective, and that f has the property that there exists some
y Image(S n ) so that f 1 (y) is a finite number of points, so f 1 (y) = {x1 , x2 , . . . , xm }.
Let Ui be a neighborhood of xi so that all Ui s get mapped to some neighborhood V of y.
So f (Ui xi ) V y. We can choose the Ui to be disjoint. We can do this because f is
continuous.

Let f |Ui : Ui V be the restriction of f to Ui . Then:

Hn (V, V y)
'

'

Hn (Ui , Ui xi )
(excision)

(excision)

'

Hn (S n , S n y)
l.e.s.

e n (S n )
H
'

l.e.s.

e n (S n )
H
'

'

Hn (S n , S n xi )

Define the local degree of f at xi , deg fxi , to be the effect of f : Hn (Ui , Ui xi ) Hn (V, V y).
We then have the following result:
Theorem 1. The degree of f equals the sum of local degrees at points in a generic fiber, that
is,
m
X
deg f =
deg f |xi .
i=1

Proof. Consider the commutative diagram, where the isomorphisms labelled by exc" follow
from excision, and l.e.s" stands for a long exact sequence.
f

Z
= Hn (Ui , Ui xi )

> Hn (V, V y)
=Z

deg f |xi

=, exc

= exc

ki

<
Pi

Hn (S n , S n f 1 (y))
'

Z
= Hn (S n , S n xi ) <
<

=, l.e.s

> Hn (S n , S n y)

exc

m
i=1 Hn (Ui , Ui xi )

= l.e.s.

l.e.s. j
f

Z
= Hn (S n )

deg f |xi

> Hn (S n )
=Z

By examining the diagram above we have:

ki (1) = (0, . . . , 0, 1, 0, . . . , 0)
where the entry 1 is in the ith place. Also, Pi j(1) = 1, for all i, so
j(1) = (1, 1, . . . , 1) =

m
X

ki (1).

i=1

The commutativity of the lower rectangle gives:

X
 X
m
m
m
X
deg f = f j(1) = f
ki (1) =
f (0, . . . , 0, 1, 0, . . . , 0) =
deg f |xi
i=1

i=1

i=1

Thus we have shown that the degree of a map f is the sum of its local degrees.
Example 2. Let us consider the power map f : S 1 S 1 , f (x) = xk , k Z. We claim that
deg f = k. We distinguish the following cases:
If k = 0 then f is the constant map which has degree 0.
If k < 0 we can compose f with a reflection r : S 1 S 1 by (x, y) (x, y). This
reflection has degree 1. So since composition leads to multiplication of degrees, we
can assume that k > 0.
If k > 0, then for all y S 1 , f 1 (y) has k points (the k roots), call them x1 , x2 , . . . , xk ,
and f has local degree 1 at each of these points. Indeed, for the above y S n we can
find a small open neighborhood centered at y, call this neighborhood V , so that he
pre-images of V are open neighborhoods Ui centered at each xi , with f |Ui : Ui V a
homeomorphism (which has possible degree 1). In this case, these homeomorphisms
are a restriction of a rotation, which is homotopic to the identity, and thus the degree
of f |Ui is 1 for each i.
So the degree of f is indeed k. Note that this implies that we can construct maps S n S n
of arbitrary degrees for any n, simply by suspending the power map f .

1.1.3

CW Complexes

Let us recall some notation from the theory of CW complexes. A CW complex X can be
written as
X = n Xn ,
where Xn is the n-skeleton, which contains all cells up to and including dimension n. Then

Xn = Xn1 q Dn
n

with the identification x Dn n (x), for Dn = S n1

Xn1 the attaching map of
the n-cell. So we are gluing the boundary of n-cells to Xn1 according to the attaching map
. A CW complex is endowed the weak topology: A X is open A Xn is open for
n
any n. An n-cell will be denoted by en = Int(D
). One can think of X as a disjoint union

of cells of various dimensions, or as qn, Dn , where means that we are attaching the
cells via their respective attaching maps.
A CW complex X is finite if it has finitely many cells, so there is a cell of maximum
dimension and the dimension of X can be defined. A CW complex is of finite type if it has
finitely many cells in each dimension. Note that a CW complex of finite type may have cells
in infinitely many dimensions. If X = n Xn and Xm = Xn for all m > n for some n, then
X = Xn and we say that the skeleton stabilizes.

Example 3. On the n-sphere S n we have a CW structure with one 0-cell (e0 ) and one n-cell
(en ). The attaching map for the n-cell is : S n1 = Dn point. There is only one such
map, the collapsing map. Think of taking the disk Dn and collapsing the entire boundary
to a single point, giving S n .
Example 4. A different CW structure on S n can be constructed so that there are two cells
in each dimension from 0 to n. Start with X0 = S 0 = {e01 , e02 }. Then X1 = S 1 where the two
1-cells D11 , D21 are attached to the 0-cells by homeomorphisms on the boundary. Similarly,
two 2-cells can be attached to X1 = S 1 by homeomorphism on the boundary giving X2 = S 2 .
Keep working in this manner adding two cells in each new dimension. Note that if we identify
each pair of cells in the same dimension by the antipodal map, we get a CW structure on
RP n with one cell in each dimension from 0 to n.
Example 5. The complex projective space CP n = Cn+1 /C is identified with the collection
of complex lines through the origin. So we write CP n = {[z0 , . . . , zn ]} where [z0 , . . . , zn ] =
(z0 : . . . : zn ) (z0 : . . . : zn ). We have that
CP n
= CP n1 t D2n ,
where : D2n CP n is given by
v

u
n
X
u
|zi |2 .
(z1 : . . . : zn ) 7 z1 , : . . . : zn : t1

i=1

The attaching map of the 2n-cell is = |S 2n1 : S 2n1 CP n1 . It follows that CP n has
a CW structure with one cell in each even dimension 0, 2, . . . , 2n.

1.1.4

Cellular Homology

Let us start with the following preliminary result:

Lemma 2. If X is a CW complex, then:
(
0
if k 6= n
(a) Hk (Xn , Xn1 ) =
Z # of n-cells if k = n.
(b) Hk (Xn ) = 0 if k > n.
(c) The inclusion i : Xn , X induces an isomorphism Hk (Xn ) Hk (X) if k < n.
Proof. (a) We know that Xn is obtained from Xn1 by attaching the n-cells (en ) . Pick a
point x at the center of each n-cell en . Let A := Xn {x } . Then A deformation retracts
to Xn1 , so we have that
Hk (Xn , Xn1 )
= Hk (Xn , A).
9

By excising Xn1 , the latter group is isomorphic to Hk (Dn , Dn {x }). Moreover, the
homology long exact sequence of the pair (Dn , Dn {x }) yields that
(
Z if k = n
e k1 (S n1 )
Hk (Dn , Dn {x })
=H
=

0 if k 6= n
So the claim follows.
(b) Consider the following portion of the long exact sequence of the pair for (Xn , Xn1 ):
Hk+1 (Xn , Xn1 ) Hk (Xn1 ) Hk (Xn ) Hk (Xn , Xn1 )
If k + 1 6= n and k 6= n, we have from part (a) that Hk+1 (Xn , Xn1 ) = 0 and Hk (Xn , Xn1 ) =
0. Thus Hk (Xn1 )
= Hk (Xn ). Hence if k > n (so in particular, n 6= k + 1 and n 6= k), we
get by iteration that
Hk (Xn )
= Hk (Xn1 )
=
= Hk (X0 ).
Note that X0 is a collection of points, so Hk (X0 ) = 0. Thus when k > n we have Hk (Xn ) = 0
as desired.
(c) We only prove the statement for finite dimensional CW complexes. Let k < n and
consider the long exact sequence for the pair (Xn+1 , Xn ):
Hk+1 (Xn+1 , Xn ) Hk (Xn ) Hk (Xn+1 ) Hk (Xn+1 , Xn )
Since k < n we have k+1 6= n+1 and k 6= n+1, so by part (a) we get that Hk+1 (Xn+1 , Xn ) = 0
and Hk (Xn+1 , Xn ) = 0. Thus Hk (Xn )
= Hk (Xn+1 ). By repeated iterations, we obtain:
Hk (Xn )
= Hk (Xn+1 )
= Hk (Xn+2 )
=
= Hk (Xn+l ) = Hk (X).
Since X is finite dimensional we know that X = Xn+l for some l. This proves the claim.
In what follows we defined the cellular homology of a CW complex X in terms of a given
cell structure, then we show that it coincides with the singular homology, so it is in fact
independent on the cell structure. Cellular homology is very useful for computations.
Definition 2. The cellular homology HCW (X) of a CW complex X is the homology of the
cellular chain complex (C (X), d ) indexed by the cells of X, i.e.,
Cn (X) := Hn (Xn , Xn1 ) = Z#ncells ,

10

(1.2)

and with differentials dn : Cn (X) Cn1 (X) defined by the following diagram:
Hn (Xn+1 , Xn ) = 0
>
Hn (Xn+1 )
= Hn (X)
>
in

Hn (Xn1 ) = 0

n+1

>
Hn (Xn )
>

Hn+1 (Xn+1 , Xn )

dn+1

jn

>
> Hn (Xn , Xn1 )

dn

>
Hn1 (Xn1 )
>

> Hn1 (Xn1 , Xn1 )

>
jn1

Hn1 (Xn2 ) = 0
The diagonal arrows are induced from long exact sequences of pairs, and we use Lemma 2 for
the identifications Hn (Xn1 ) = 0, Hn1 (Xn2 ) = 0 and Hn (Xn+1 )
= Hn (X) in the diagram.
In the notations of the above diagram, we now set:
dn = jn1 n : Cn (X) Cn1 (X),

(1.3)

dn dn+1 = 0.

(1.4)

and note that we have

Indeed,
dn dn+1 = jn1 n jn n+1 = 0
since n jn = 0 as the composition of two consecutive maps in a long exact sequence. So
(C (X), d ) is a chain complex.
The following result asserts that cellular homology is independent on the cell structure
used for its definition:
Theorem 3. There are isomorphisms
HnCW (X)
= Hn (X)
for all n, where Hn (X) is the singular homology of X.
Proof. Since Hn (Xn+1 , Xn ) = 0 and Hn (X)
= Hn (Xn+1 ), we get from the diagram above
that
Hn (X)
= Hn (Xn )/ ker in
= Hn (Xn )/Image n+1 .
11

Now, Hn (Xn )
= Image jn
= ker n
= ker dn . The first isomorphism comes from jn being
injective, while the second follows by exactness. Finally, ker n = ker dn since dn = jn1 n
and jn1 is injective. Also, we have Image n+1 = Image dn+1 . Indeed, dn+1 = jn n+1 and
jn is injective.
Altogether, we have
Hn (X)
= Hn (Xn )/Image n+1 = ker dn /Image dn+1 = HnCW (X)
So we have proved the theorem.
Let us now discuss some immediate consequences of the above theorem.
(a) If X has no n-cells, then Hn (X) = 0.
Indeed, in this case we have Cn = Hn (Xn , Xn1 ) = 0, so HnCW (X) = 0.
(b) If X is connected and has a single 0-cell then d1 : C1 C0 is the zero map.
Indeed, since X contains only a single 0-cell, C0 = Z. Also, since X is connected,
H0 (X) = Z. So by the theorem above Z = H0 (X) = ker d0 /Image d1 = Z/Image d1 .
This implies that Image d1 = 0, so d1 is the zero map as desired.
(c) If X has no two cells in adjacent dimensions then dn = 0 for all n and Hn (X)
= Z#ncells
for all n.
Indeed, in this case all maps dn vanish. So for any n, HnCW (X)
= Cn
= Z#ncells .
Example 6. Recall that CP n has one cell in each dimension 0, 2, 4, . . . , 2n. So CP n has no
two cells in adjacent dimensions, meaning we can apply Consequence (c) above to say:
(
Z
if i = 0, 2, 4, . . . , 2n
Hi (CP n ) =
0
otherwise.
Example 7. When n > 1, S n S n has one 0-cell, two n-cells, and one 2n-cell. Since n > 1,
these cells are not in adjacent dimensions so again Consequence (c) above applies to give:

i = 0, 2n
Z
n
n
2
Hi (S S ) = Z
i=n

0
otherwise.
In the remaining of this section, we discuss how to compute in general the maps
dn : Cn (X) = Z#ncells Cn1 (X) = Z#(n1)cells
of the cellular chain complex. Let us consider the n-cells {en } as the basis for Cn (X) and
the (n 1)-cells {en1
} as the basis for Cn1 (X). In particular, we can write:
X
dn (en ) =
d en1 ,

with d Z. The following result provides a way of computing the coefficients d :

12

Theorem 4. The coefficient d is equal to the degree of the map , : Sn1 Sn1
defined by the composition:
n

collapse

Sn1 = en
Xn1 = Xn2 q en1 Xn1 /(Xn2 t6= en1 ) = Sn1 ,
`
where n is the attaching map of en , and the collapsing map sends Xn2 6= en1
to a

point.

Proof. We will proceed with the proof by chasing the following diagram, and we note that
the map is defined so that the top right square commutes.
Hn (Dn , Sn1 )


'

Hn1 (Sn1 )

Cn (X)

Hn1 (Sn1 )
q

n1 (Xn1 /Xn2 )
H

jn1

&

n1

Hn1 (Xn1 )

dn

n1 (en1 /en1 )
= H

'

Xn1 Xn2
,
)
Xn2 Xn2
Recall that our goal is to compute dn (en ). The upper left square is natural and therefore
commutes (it is induced by the characteristic map : (D , S 1 ) (X , X1 ) of a cell),
while the lower left triangle is part of the exact diagram defining the chain complex C (X)
and is defined to commute as well. Appealing to naturality, the map gives a unique Dn
so that n (Dn ) = en . Since the top left square and the bottom left triangle both commute,
this gives that
n
dn (en ) = jn1 n1
(D ).
Cn1 (X)

'

/ Hn (

Looking to the bottom right square, recall that since X is a CW complex, (Xn , Xn1 ) is a
n1 (Xn1 /Xn2 )
good pair. This gives the isomorphism Cn1 (X) = Hn1 (Xn1 , Xn2 ) ' H
n1 (Xn1 /Xn2 ) ' Hn1 (Xn1 /Xn2 , Xn2 /Xn2 ), where the isoBut, we similarly have H
morphism is induced by the quotient map q collapsing Xn2 .
The bottom right square commutes by the definition of jn1 and q , from which it follows
that

dn (en ) = q n1
(Dn ),
13

where formally we should precompose in the left hand side with the isomorphism between
n1 (Xn1 /Xn2 ) so that everything is in the same space. This last map takes
Cn1 (X) and H
n1 (en1 /en1 ). To see
the generator Dn to some linear combination of generators in H

which generators it maps to, we project down to the summands to obtain

X
n1
dn (en ) =
q q
Dn .

n1
As noted before, we have defined = q q
. So writing
X
dn (en ) =
Dn ,

we see from the definition of the above maps that is multiplication by d .

Example 8. Let Mg be the close oriented surface of genus g, with its usual CW structure:
one 0-cell, 2g 1-cells {a1 , b1 , ag , bg }, and one 2-cell attached by product of commutators
[a1 , b1 ] [ag , bg ]. The associated cellular chain complex of Mg is:
0

d3

d2

Z2g

d1

d0

Since Mg is connected and has only one 0-cell, we get that d1 = 0. We claim that d2 is also
the zero map. This amounts to showing that d2 (e) = 0, where e denotes the 2-cell. Indeed,
let us compute the coefficients deai and debi in our degree formula. As the attaching map
1
1 1
sends the generator to a1 b1 a1
1 b1 ...ag bg ag bg , when we collapse all 1-cells (except ai , resp.
1
1
1 1
bi ) to a point, the word defining the attaching map a1 b1 a1
1 b1 ...ag bg ag bg reduces to ai ai
and resp. bi b1
i . Hence deai = 1 1 = 0. Similarly, debi = 1 1 = 0, for each i. Altogether,
d2 (e) = a1 + b1 a1 b1 + ag + bg ag bg = 0.
So the homology groups of Mg are given by

Z
Hn (Mg ) = Z2g

i=0,2
i=1
otherwise.

Example 9. Let Ng be the closed nonorientable surface of genus g, with its cell structure
consisting of one 0-cell, g 1-cells {a1 , , ag }, and one 2-cell e attached by the word a21 a2g .
The cellular chain complex of Ng is given by
0

d3

d2

Zg

d1

d0

As before, d1 = 0 since Ng is connected and there is only one cell in dimension zero. To
compute d2 : Z Zg we again apply the cellular boundary formula, and obtain
d2 (1) = (2, 2, , 2)
14

since each a1 appears in the attaching word with total exponent 2, which means that each
map is homotopic to the map z 7 z 2 of degree 2. In particular, d2 is injective, hence
H2 (Ng ) = 0. If we change the standard basis for Zg by replacing the last standard basis
element en = (0, , 0, 1) by e0n (1, , 1), then d2 (1) = 2 e0n , so
H1 (Ng )
= Zg /Image d2
= Zg /2Z
= Zg1 Z/2.
Altogether,

Z
Hn (Ng ) = Zg1 Z2

i=0
i=1
otherwise.

Example 10. Recall that RPn has a CW structure with one cell ek in each dimension
0 k n. Moreover, the attaching map of ek in RPn is the two-fold cover projection
: S k1 RPk1 . The cellular chain complex for RPn looks like:
0

dn+1

/Z

dn

/ ...Z d1

/Z

d0

To compute the differential dk , we need to compute the degree of the composite map

: S k1 RPk1 RPk1 /RPk2 = S k1 .

The map is a homeomorphism when restricted to each component of S k1 \ S k2 , and
these homeomorphisms are obtained from each other by precomposing with the antipodal
map a of S k1 , which has degree (1)k . Hence, by our local degree formula, we get that:
deg = deg id + deg a = 1 + (1)k .
In particular,
(
0 if k is odd
dk =
2 if k is even,
and therefore we obtain that

Z2
n
Hk (RP ) = Z

if k is odd , 0 < k < n

k = 0, and k = n odd
otherwise.

Finally, note that an equivalent definition of the above map is obtained by first collapsing
the equatorial S k2 to a point to get S k1 S k1 , and then mapping the two copies of S k1
onto S k1 , the first one by the identity map, and the second by the antipodal map.

15

Exercises
1. Describe a cell structure on S n S n S n and calculate H (S n S n S n ).
2. Let f : S n S n be a map of degree m. Let X = S n f Dn+1 be a space obtained from
S n by attaching a (n + 1)-cell via f . Compute the homology of X.
3. Let G be a finitely generated abelian group, and fix n 1. Construct a CW-complex X
i (X) = 0 for all i 6= n. (Hint: Use the calculation of the previous
such that Hn (X)
= G and H
exercise, together with know facts from Algebra about the structure of finitely generated
abelian groups.) More generally, given finitely generated abelian groups G1 , G2 , , Gk ,
construct a CW-complex X whose homology groups are Hi (X) = Gi , i = 1, , k, and
i (X) = 0 for all i
H
/ {1, 2, , k}.
4. Show that RP5 and RP4 S 5 have the same homology and fundamental group. Are these
spaces homotopy equivalent?
5. Let 0 m < n. Compute the homology of RPn /RPm .
6. The mapping torus Tf of a map f : X X is the quotient of X I
Tf =

X I
.
(x, 0) (f (x), 1)

Let A and B be copies of S 1 , let X = A B, and let p be the wedge point of X. Let
f : X X be a map that satisfies f (p) = p, carries A into A by a degree3 map, and carries
B into B by a degree5 map.
(a) Equip Tf with a CW structure by attaching cells to X S 1 .
(b) Compute a presentation of 1 (Tf ).
(c) Compute H1 (Tf ; Z).
16

7. The closed oriented surface Mg of genus g, embedded in R3 in the standard way, bounds
a compact region R. Two copies of R, glued together by the identity map between their
boundary surfaces Mg , form a space X. Compute the homology groups of X and the relative
homology groups of (R, Mg ).
8. Let X be the space obtained by attaching two 2-cells to S 1 , one via the map z 7 z 3 and
the other via z 7 z 5 , where z denotes the complex coordinate on S 1 C.
(a) Compute the homology of X with coefficients in Z.
(b) Is X homeomorphic to the 2-sphere S 2 ? Justify your answer!

17

1.2

Euler Characteristic

Definition 3. Let X be a finite CW complex of dimension n and denote by ci the number

of i cells of X. The Euler characteristic of X is defined as:
n
X
(X) =
(1)i ci .
(1.5)
i=0

It is natural to question whether or not the Euler characteristic depends on the cell
structure chosen for the space X. As we will see below, this is not the case. It suffices to
show that the Euler characteristic depends only on the cellular homology of the space X.
Indeed, cellular homology is isomorphic to singular homology, and the latter is independent
of the cell structure on X.
Recall that if G is a finitely generated abelian group, then G decomposes into a free part
and a torsion part, i.e.,
G ' Zr Zn1 Znk .
The integer r := rk(G) is the rank of G. The rank is additive in short exact sequences of
finitely generated abelian groups.
Theorem 5. The Euler characteristic can be computed as:
n
X
(X) =
(1)i bi (X)

(1.6)

i=0

with bi (X) := rkHi (X) the i-th Betti number of X. In particular, (X) is independent of
the chosen cell structure on X.
Proof. We will follow the following notation: Bi = Image(di+1 ), Zi = ker(di ), and Hi =
Zi /Bi . Consider a chain complex of finitely generated abelian groups and the short exact
sequences defining homology:
0

dn+1

dn

/ Cn
/

Zi 
/

Bi

d2

...

/ Ci

di+1

Zi

d1

/ C1
di
q

//

/ C0 d0 /
/

Bi1
/

Hi

The additivity of rank yields that

ci := rkCi = rkZi + rkBi1
and
rkZi = rkBi + rkHi .
Substitute the second equality P
into the first, multiply the resulting equality by (1)i , and
sum over i to get that (X) = ni=0 (1)i rkHi .
Finally, apply this result to the cellular chain complex Ci = Hi (Xi , Xi1 ) and use the
identification between cellular and singular homology.
18

Example 11. If Mg and Ng denote the orientable and resp. nonorientable closed surfaces
of genus g, then (Mg ) = 1 2g + 1 = 2(1 g) and (Ng ) = 1 g + 1 = 2 g. So all the
orientable and resp. non-orientable surfaces are distinguished from each other by their Euler
characteristic, and there are only the relations (Mg ) = (N2g ).

Exercises
1. A graded abelian group is a sequence of abelian groups A := (An )n0 . We say that A
is of finite type if
X
rankAn < .
n0

The Euler characteristic of a finite type graded abelian group A is the integer
X
(A ) :=
(1)n rankAn .
n0

A short exact sequence of graded groups A , B , C , is a sequence of short exact sequences

0 An Bn Cn 0, n 0.
Prove that if 0 A B C 0 is a short exact sequence of graded abelian groups of
finite type, then
(B ) = (A ) + (C ).
2. Suppose we are given three finite type graded abelian groups A , B , C , which are part
of a long exact sequence
i

k
k
k
Ak
Bk
Ck
Ak1 A0 B0 C0 0.

Show that
(B ) = (A ) + (C ).
3. For finite CW complexes X and Y , show that
(X Y ) = (X) (Y ).
4. If a finite CW complex X is a union of subcomplexes A and B, show that
(X) = (A) + (B) (A B).
5. For a finite CW complex and p : Y X an n-sheeted covering space, show that
(Y ) = n (X).
6. Show that if f : RP2n Y is a covering map of a CW -complex Y , then f is a homeomorphism.
19

1.3

Lefschetz Fixed Point Theorem

Recall that if G is a finitely generated abelian group, then G decomposes into a free part
and a torsion part, i.e.,
G ' Zr Zn1 Znk .
Here r = rk(G) and Torsion(G) := ki=1 Zni . Given an endomorphism : G G, define its
trace by
Tr() = Tr ( : G/Torsion(G) G/Torsion(G))
(1.7)
where the latter trace is the linear algebraic trace of the map : Zr Zr .
Definition 4. If X has the homotopy type of a finite simplicial or cellular complex and
f : X X, then the Lefschetz number of f is defined to be
X
(f ) =
(1)i Tr(f : Hi (X) Hi (X)).
(1.8)
i

Remark 1. Notice that homotopic maps have the same Lefschetz number since they induce
the same maps on homology.
Example 12. If f ' idX , then (f ) = (X). This follows from the fact the map induced
in homology by the identity map is the identity matrix and that the trace of the identity
matrix in this case is the corresponding Betti number of X.
Theorem 6. (Lefschetz)
If X is a retract of a finite simplicial (or cellular) complex and if f : X X satisfies
(f ) 6= 0, then f has a fixed point.
Before proving this theorem, let us consider a few examples.
Example 13. Suppose that X has the homology of a point (up to torsion). Then
(f ) = Tr (f : H0 (X) H0 (X)) = 1.
This follows from the fact that all the other homology groups are zero and that the map
induced on H0 is the identity.
This example leads immediately to two nontrivial results, the first of which is the Brouwer
fixed point theorem.
Example 14. (Brouwer) If f : Dn Dn is continuous then f has a fixed point.
Example 15. If X = RP2n then modulo torsion X has the homology of a point. Therefore
any continuos map f : RP2n RP2n has a fixed point.
Finally we are led to an example which does not follow from the computation for a point.

20

Example 16. If f : S n S n is a continuos map and deg(f ) 6= (1)n+1 , then f has a fixed
point. To verify this, we compute
(f ) = Tr(f : H0 (S n ) H0 (S n )) + (1)n Tr(f : Hn (S n ) Hn (S n ))
= 1 + (1)n deg(f )
6= 0.
Corollary 7. If a : S n S n is the antipodal map, then deg(a) = (1)n+1 .
Now we return to outlining the proof:
Definition 5. If K and L are simplicial complexes and f : K L is a linear map which
sends each simplex of K to a simplex in L so that vertices map to vertices, then f is said to
be simplicial.
Note that a simplicial map is uniquely determined by its values on vertices. The simplicial
approximation theorem asserts that given any map f from a finite simplicial complex to an
arbitrary simplicial complex, we can find a map g in the homotopy class of f so that g is
simplicial in the above sense with respect to some finite iteration of barycentric subdivisions
of the domain.
Theorem 8. If K is a finite simplicial complex and L is an arbitrary simplicial complex,
then for any map f : K L there is a map in the homotopy class of f which is simplicial
with respect to some interated barycentric subdivision of K.
The proof of this result is omitted. We now proceed to the Lefschetz theorem.
Proof. (sketch)
Let us suppose that f has no fixed points. The general case reduces to the case when X is a
finite simplicial complex. Indeed, if r : K X is a retraction of a finite simplicial complex K
onto X, the composition f r : K X K has exactly the same fixed points as f and since
r : Hi (K) Hi (X) is projection onto a direct summand, we have that Tr(f r ) = Tr(f ),
so (f r) = (f ). We therefore take X to be a finite simplicial complex.
X is compact and there exists a metric d on X so that d restricts to the Euclidean metric
on each simplex of X; choose such a metric. If f has no fixed points, we can find a uniform 
for which d(x, f (x)) >  by the standard covering trick. Via repeated barycentric subdivision
of X we can construct L so that for each vertex, the union of all simplicies containing that
vertex has diameter less than 2 . Applying the simplicial approximation theorem we can find
a subdivision K of L and a simplicial map g : K L so that g lies in the homotopy class
of f . Moreover, we may take g so that f () lies in the subcomplex of X consisting of all
simplicies containing . Again, by repeated barycentric subdivision we may choose K so
that each simplex in K has diameter less than 2 . In particular then g() = for each
K. Notice (g) = (f ) since f and g are homotopic.
Since g is simplicial, Kn maps to Ln (that is, g sends n-skeletons to n-skeletons). We
constructed K as a subdivision of L so that g(Kn ) Kn for each n.
21

We will use the algebraic fact that trace is additive for short exact sequences to show
that we can replace Hi (X) with Hi (Ki , Ki1 ) in our computation of the Lefschetz number. By essentially the same argument as was used above in the computation of the Euler
characteristic and using this fact we obtain that
X
(g) =
(1)i Tr(g : Hi (Ki , Ki1 ) Hi (Ki , Ki1 ))
i

We have a natural basis for Hi (Ki , Ki1 ) coming from the simplicies i in Ki . But since
g() = it follows that Tr(g : Hi (Ki , Ki1 ) Hi (Ki , Ki1 )) = 0 for each i. So
(f ) = (g) = 0.
The cellular case is proved similarly, using instead a corresponding cellular approximation
theorem.

Exercises
1. Is there a continuous map f : RP2k1 RP2k1 with no fixed points? Explain.
1. Is there a continuous map f : CP2k1 CP2k1 with no fixed points? Explain. We will
see later that any map f : CP2k CP2k has a fixed point.

22

1.4

Homology with General Coefficients

Let G be an abelian group and X a topological space. We define the homology of X with G
coefficients, denoted H (X; G), as the homology of the chain complex
Ci (X; G) = Ci (X) G
(1.9)
P
consisting of finite formal sums i i i ( : i X, i G), and with boundary maps
given by
iG := i idG .
G
Since i satisfies i i+1 = 0 it follows that iG i+1
= 0, so (C (X; G), G ) forms indeed a chain complex. We can construct versions of the usual modified homology groups
(relative, reduced, etc.) in the natural way. Define relative chains by Ci (X, A; G) :=
Ci (X; G)/Ci (A; G), and reduced homology via the augmented chain complex
G
i+1

1
2
i
C0 (X; G) G 0.
C1 (X; G)

Ci (X; G)
P
P
where ( i i i ) = i i . Notice that Hi (X) = Hi (X, Z) by definition.
By looking directly at the chain maps, it follows that
(
G i=0
Hi (pt; G) =
0 i 6= 0.

Nothing (other than coefficients) needs to change in our previous proofs about the relationships between relative homology and reduced homology of quotient spaces so we can compute
the homology of a sphere as before by induction and using the long exact sequence of the
pair (Dn , S n ) to be
(
G i = 0, n
Hi (S n ; G) =
0 otherwise.
Finally, we can build cellular homology in the same way, defining
CiG (X) = Hi (Xi , Xi1 ; G) = G#

ncells

The cellular boundary maps are given by:

X
X
dn (
n en ) =
d n en1 ,

where d is as before the degree of a map : S n1 S n1 . This follows from the easy
fact that if f : S k S k has degree m, then f : Hk (S k ; G) ' G Hk (S k ; G) ' G is the
multiplication by m. As it is the case for integers, we get
HiCW (X; G) ' Hi (X; G)
for all i.
23

Example 17. We compute Hi (RPn ; Z2 ) using the calculation above. Notice that over Z the
cellular boundary maps are di = 0 or di = 2 depending on the parity of i, and therefore with
Z2 -coefficients all of boundary maps vanish. Therefore,
(
Z2 0 i n
Hi (RPn ; Z2 ) =
0
otherwise.
Example 18. Fix n > 0 and let g : S n S n be a map of degree m. Define the CW complex
X = S n g en+1 ,
where the (n + 1)-cell en+1 is attached to S n via the map g. Let f be the quotient map
f : X X/S n . Define Y = X/S n = S n+1 . The homology of X can be easily computed by
using the cellular chain complex:
0

dn+2

dn+1
m

dn

...

d1

d1

d0

Therefore,

Z
Hi (X; Z) = Zm

i=0
i=n
otherwise.

Moreover, as Y = S n+1 , we have

(
Z i = 0, n + 1
Hi (Y ; Z) =
0 otherwise.
It follows that f induces the trivial homomorphisms in homology with Z-coefficients (except
in degree zero, where f is the identity). So it is natural to ask if f is homotopic to the
constant map. As we will see below, by considering Zm -coefficients we can show that this is
not the case.
Let us now consider H (X; Zm ) where m is, as above, the degree of the map g. We return
to the cellular chain complex level and observe that we have
0

dn+2

Zm

dn+1
m

Zm

dn

/ ... d1 / 0

d1

Zm

Multiplication by m is now the zero map, so we get

(
Zm i = 0, n, n + 1
Hi (X; Zm ) =
0
otherwise.
Also, as already discussesd,
(
Zm
Hi (Y ; Zm ) =
0
24

i = 0, n + 1
otherwise.

d0

We next consider the induced homomorphism f : Hn+1 (X; Zm ) Hn+1 (X; Zm ). The
claim is that this map is injective, thus non-trivial map, so f cannot be homotopic to the cone n+1 (Y ; Zm ) ' Hn+1 (X, S n ; Zm ).
stant map. As noted before, we still have an isomorphism H
This leads us to consider the long exact sequence of the pair (X, S n ) in dimension n + 1. We
have
f
Hn+1 (S n ; Zm ) Hn+1 (X; Zm ) Hn+1 (X, S n ; Zm )
But, Hn+1 (S n ; Zm ) = 0 and so f is injective on Hn+1 (X; Zm ). Since Hn+1 (X; Zm ) = Zm 6= 0
e n+1 (Y ; Zm ) it follows that f is not trivial on Hn+1 (X; Zm ), which
and Hn+1 (X, S n ; Zm ) ' H
proves our claim.

Exercises
1. Calculate the homology of the 2-torus T 2 with coefficients in Z, Z2 and Z3 , respectively.
Do the same calculations for the Klein bottle.

25

1.5
1.5.1

Universal Coefficient Theorem for Homology

Tensor Products

Let A, B be abelian groups. Define the abelian group

A B = {a b | a A, b B}/

(1.10)

where is generated by the relations (a + a0 ) b = a b + a0 b and a (b + b0 ) =

a b + a b0 . The zero element of A B is 0 b = a 0 = 0 0 = 0AB since, e.g.,
0 b = (0 + 0) b = 0 b + 0 b so 0 b = 0AB . Similarly, the inverse of an element a b is
(a b) = (a) b = a (b) since, e.g., 0AB = 0 b = (a + (a)) b = a b + (a) b.
Lemma 9. The tensor product satisfies the following universal property which asserts that
if : A B C is any bilinear map, then there exists a unique map : A B C such
that = i, where i : A B A B is the natural map (a, b) 7 a b.
AB

AB

% 

Proof. Indeed, : A B C can be defined by a b 7 (a, b).

Proposition 10. The tensor product satisfies the following properties:
(1) A B
= B A via the isomorphism a b 7 b a.
L
L
(2) ( i Ai ) B
= i (Ai B) via the isomorphism (ai )i b 7 (ai b)i .
(3) A (B C)
= (A B) C via the isomorphism a (b c) 7 (a b) c.
(4) Z A
= A via the isomorphism n a 7 na.
(5) Z/nZ A
= A/nA via the isomorphism l a 7 la.
Proof. These are easy to prove by using the above universal property. We sketch a few.
(1) The map : AB B A defined by (a, b) 7 ba is clearly bilinear and therefore
induces a homomorphism : A B B A with a b 7 b a. Similarly, there is the
reverse map : B A A B defined by (b, a) 7 a b which induces a homomorphism
: B A A B with b a 7 a b. Clearly, = idBA and = idAB and
AB
= B A.
(4) The map : Z A A defined by (n, a) 7 na is a bilinear map and therefore
induces a homomorphism : Z A A with n a 7 na. Now suppose (n a) = 0.
Then na = 0 and n a = 1 (na) = 1 0 = 0ZA . Thus is injective. Moreover, if a A,
then (1 a) = a and is surjective as well.

26

(5) The map : Z/nZA A/nA defined by (l, a) 7 la is a bilinear map and therefore
: Z/nZ A A/nA with P
l a 7 la. Now
induces a homomorphism
Pk
Pk suppose (l a) =
k
la = 0. Then la = i=1 nai and l a = 1 (la) = 1 ( i=1 nai ) = i=1 (n ai ) = 0Z/nZA ,
so is injective. Now let a A/nA. Then (1 a) = a and is surjective as well.
More generally, if R is a ring and A and B are R-modules, a tensor product A R B can
be defined as follows:
(1) if R is commutative, define the R-module A R B := A B/ , where is the relation
generated by ra b = a rb = r(a b).
(2) if R is not commutative, we need A a right R-module and B a left R-module and the
relation is ar b = a rb. In this case A R B is only an abelian group.
In both cases, A R B is not necessarily isomorphic to A B.

Example 19. Let R = Q[ 2] = {a + b 2 | a, b Q}. Now R R R

= R which is a
2-dimensional Q-vector space. However, R R as a Z-module is a 4-dimesnional Q-vector
space.
Lemma 11. If G is an abelian group, then the functor G is right exact, that is, if
j
j1G
i1G
i
A
B
C 0 is exact, then A G
B G
C G 0 is exact.
Proof. Let c g C G. Since j is onto, there exists, b B such that j(b) = c. Then
(j 1G )(b g) = c g and j 1G is onto.
Since ji = 0, we have (j1G )(i1G ) = (ji)1G = 0 and thus, Image(i1G ) ker(j1G ).
It remains to show that ker(j 1G ) Image(i 1G ). It is enough to show that

: B G/Image(i 1G )
C G,
where is the map induced by j 1G . Construct an inverse of , induced from the homomorphism
: C G B G/Image(i 1G )
defined by (c, g) 7 b g, where j(b) = c. We must show that is a well-defined bilinear
map and that the induced map satisfies = id and = id.
If j(b) = j(b0 ) = c, then b b0 ker j = Image i, so b b0 = i(a) for some a A.
Thus, b g b0 g = (b b0 ) g = i(a) g Image(i 1G ). So is well defined.
Now ((c + c0 , g)) = d g where j(d) = c + c0 . Since j is surjective, choose b, b0 B
such that j(b) = c and j(b0 ) = c0 . Then d (b + b0 ) ker j = Image i and so there
exists a A such that i(a) = d (b + b0 ). Thus, ((c + c0 , g)) = d g = (b + b0 ) g =
27

b g + b0 g = (c, g) + (c0 , g) and is linear in the first component. For the second
component, (c, g + g 0 ) = b (g + g 0 ) = b g + b g 0 = (c, g) + (c, g 0 ). Thus, is bilinear.
Now by the universal property of the tensor product, the bilinear map induces a homomorphism
: C G B G/Image(i 1G )
defined by c g 7 (c, g) = b g, where j(b) = c. For c g C G,
(c g) = (b g) = j(b) g = c g,
so = idCG . Similarly, for b g B G/Image(i 1G ), (b g) = (j(b) g) =
(j(b), g) = b g. Thus = id.

1.5.2

The Tor functor and the Universal Coefficient Theorem

In this section we explain how to compute H (X; G) in terms of H (X; Z) and G. More
generally, given a chain complex

n
C : Cn
Cn1 C0 0

of free abelian groups and G an abelian group, we aim to compute H (C ; G) = H (C G)

in terms of H (C ; Z) and G. The answer is provided by the following result:
Theorem 12. (Universal Coefficient Theorem)
There are natural short exact sequences:
0 Hn (C ) G Hn (C ; G) Tor(Hn1 (C ), G) 0 for all n.

(1.11)

Naturality here means that if C C0 is a chain map, then there is an induced map of short
exact sequences with commuting squares. Moreover, these short exact sequences split, but
not naturally.
In particular, if C = C (X, A) is the relative singular chain complex, then there are
natural short exact sequences
0 Hn (X, A) G Hn (X, A; G) Tor(Hn1 (X, A), G) 0.

(1.12)

Naturality is with respect to maps of pairs (X, A)

(Y, B). The exact sequence (1.12)
splits, but not naturally.
Indeed, if we assume that A = B = , then

we have splittings
Hn (X; G) = Hn (X)G Tor(Hn1 (X), G), Hn (Y ; G) = Hn (Y )G Tor(Hn1 (Y ), G).
If these splittings were natural, and f induces the trivial map f = 0 on H (; Z) then f
induces the trivial map on H (; G), for any coefficient group G. But this is in contradiction
with Example 18.
Let us next explain the Tor functor appearing in the statement of the universal coefficient
theorem.
28

Definition 6. A free resolution of an abelian group H is an exact sequence:

f2

f1

f0

F2
F1
F0
H 0,
with each Fn free abelian.
Given an abelian group G, from a free resolution F of H, we obtain a modified chain
complex:
F G : F2 G F1 G F0 G 0.
We define
Torn (H, G) := Hn (F G).

(1.13)

Note here that we have removed the final term of the complex to account for the fact that
G is right exact.
Moreover, the following holds:
Lemma 13. For any two free resolutions F and F0 of H there are canonical isomorphisms
Hn (F G)
= Hn (F0 G) for all n. Thus, Torn (H, G) is independent of the free resolution
F .
Proposition 14. For any abelian group H, we have that

and

Torn (H, G) = 0 if n > 1,

(1.14)

Tor0 (H, G)
= H G.

(1.15)

Proof. Indeed, given an abelian group H, take F0 to be the free abelian group on a set of
f0

generators of H to get F0  H 0. Let F1 := ker(f0 ), and note that F1 is a free group, as

it is a subgroup of a free abelian group F0 . Let F1 , F0 be the inclusion map. Then
0 F1 , F0  H 0
is a free resolution of H. Thus, Torn (H, G) = 0 if n > 1. Moreover, it follows readily that
Tor0 (H, G)
= H G.
Definition 7. In what follows, we adopt the notation:
Tor(H, G) := Tor1 (H, G).
Proposition 15. The Tor functor satisfies the following properties:
(1) Tor(A, B)
= Tor(B, A).
L
L
(2) Tor( i Ai , B)
= i Tor(Ai , B).
(3) Tor(A, B) = 0 if A or B is free or torsion-free.
29

(4) Tor(A, B)
= Tor(Torsion(A), B), where Torsion(A) is the torsion subgroup of A.
n

(5) Tor(Z/nZ, A)
A).
= ker(A
(6) For a short exact sequence: 0 B C D 0 of abelian groups, there is a natural
exact sequence:
0 Tor(A, B) Tor(A, C) Tor(A, D) A B A C A D 0.
Proof. (2) Choose a free resolution for

Ai as the direct sum of free resolutions for the Ai s.

(5) The exact sequence 0 Z

Z Z/nZ 0 is a free resolution of Z/nZ. Now A
n1A
n
gives Z A Z A 0 which by property (4) of the tensor product is A
A 0.
n
Thus, Tor(Z/nZ, A) = ker(A
A).
(3) If A is free, we can choose the free resolution:
F1 = 0 F0 = A A 0
which implies that Tor(A, B) = 0. On the other hand, if B is free, tensoring the exact sequence 0 F1 F0 A 0 with B = Zs gives a direct sum of copies of
0 F1 F0 A 0. Hence, it is an exact sequence and so H1 of this complex is
0. For the torsion free case, see below.
(6) Let 0 F1 F0 A 0 be a free resolution of A, and tensor it with the short exact
sequence 0 B C D 0 to get a commutative diagram:
0
0
0

0 F1 B F1 C F1 D 0

0 F0 B F0 C F0 D 0

0
0
0
Rows are exact since tensoring with a free group preserves exactness. Thus we get a short
exact sequence of chain complexes. Recall now that for any short exact sequence of chain
complexes 0 B C D 0 (which means exactness for each level n: 0 Bn
Cn Dn 0, commuting with differential ), there is an associated long exact sequence of
homology groups
Hn (B ) Hn (C ) Hn (D ) Hn1 (B ) . . .
So in our situation we obtain the homology long exact sequence:
0 H1 (F B) H1 (F C) H1 (F D) H0 (F B) H0 (F C) H0 (F D) 0
30

Since H1 (F B) = Tor(A, B) and H0 (F B) = A B, the above long exact sequence

reduces to:
0 Tor(A, B) Tor(A, C) Tor(A, D) A B A C A D 0.
(1) Apply (6) to a free resolution 0 F1 F0 B 0 of B, and get a long exact
sequence:
0 Tor(A, F1 ) Tor(A, F0 ) Tor(A, B) A F1 A F0 A B 0.
Because F1 , F0 are free, by (3) we have that Tor(A, F1 ) = Tor(A, F0 ) = 0, so the long exact
sequence becomes:
0 Tor(A, B) A F1 A F0 A B 0.
Also, by definition of Tor, we have a long exact sequence:
0 Tor(B, A) F1 A F0 A B A 0.
So we get a diagram:
0 Tor(A, B) A F1 A F0 A B 0

'
'
'
0 Tor(B, A) F1 A F0 A B A 0
with the arrow labeled defined as follows. The two squares on the right commute since
is naturally commutative. Hence, there exists : Tor(A, B) Tor(B, A) which makes the
left square commutative. Moreover, by the 5-lemma, we get that is an isomorphism.
f

We can now prove the torsion free case of (3). Let 0 F1 F0 A 0 be a

free resolution of A. The claim about the vanishing of Tor(A,
P B) is equivalent to the injectivity
of
the
map
f

id
:
F

B.
Assume
B
1
0P
i xi bi ker(f idB ). So
P
i f (xi ) bi can be reduced to zero by a finite
i f (xi ) bi = 0 F1 B. In other words,
number of applications of the defining relations for tensor products. Only a finite number of
elemnts ofP
B, generating a finitely generated subgroup B0 of B, are involved in this process,
so in fact i xi bi ker(f
P idB0 ). But B0 is finitely generated and torsion free, hence free,
so Tor(A, B0 ) = 0. Thus i xi bi = 0, which proves the claim. The case when A is torsion
free follows now by using (1) to reduce to the previous case.
(4) Apply (6) to the short exact sequence: 0 Torsion(A) A A/Torsion(A) 0 to
get :
0 Tor(G, Torsion(A)) Tor(G, A) Tor(G, A/Torsion(A))
Because A/Torsion(A) is torsion free, Tor(G, A/Torsion(A)) = 0 by (3), so:
Tor(G, Torsion(A)) ' Tor(G, A)
Now by (1), we get that Tor(A, G) ' Tor(Torsion(A), G).

31

Remark 2. It follows from (5) that

Tor(Z/nZ, Z/mZ) =

Z
= Z/nZ Z/mZ,
(n, m)Z

where (n, m) is the greatest common divisor of n and m. More generally, if A and B are
finitely generated abelian groups, then
Tor(A, B) = Torsion(A) Torsion(B)

(1.16)

where Torsion(A) and Torsion(B) are the torsion subgroups of A and B respectively.
Let us conclude with some examples:
Example 20. Suppose G = Q, then Tor(Hn1 (X), Q) = 0, so
Hn (X; Q) ' Hn (X) Q.
It follows that the n-th Betti number of X is given by
bn (X) := rkHn (X) = dimQ Hn (X; Q).
Example 21. Suppose X = T 2 , and G = Z/4. Recall that H1 (T 2 ) = Z2 . So:
H0 (T 2 ; Z/4) = H0 (T 2 ) Z/4 = Z/4


H1 (T 2 ; Z/4) = H1 (T 2 ) Z/4 Tor(H0 (T 2 ), Z/4) = Z2 Z/4 = (Z/4)2


H2 (T 2 ; Z/4) = H2 (T 2 ) Z/4 Tor(H1 (T 2 ), Z/4) = Z/4.
Example 22. Suppose X = K is the Klein bottle, and G = Z/4. Recall that H1 (K) =
Z Z/2, and H2 (K) = 0, so:


H2 (K; Z/4) = H2 (K)Z/4 Tor(H1 (K), Z/4) = Tor(Z, Z/4)Tor(Z/2, Z/4) = 0Z/2 = Z/2.

Exercises
1. Prove Lemma 13.
e n (X; Z) = 0 for all n if, and only if, H
e n (X; Q) = 0 and H
e n (X; Z/p) = 0 for
2. Show that H
all n and for all primes p.

32

Chapter 2
Basics of Cohomology
Given a space X and an abelian group G, we will first define cohomology groups H i (X; G).
In
L thei next chapter we will show that, via the cup product operation, the graded group
i H (X; G) becomes a ring. The ring structure will help us distinguish spaces X and Y
which have isomorphic homology and cohomology groups but non-isomorphic cohomology
rings, for example X = CP2 and Y = S 2 S 4 .

2.1

Cohomology of a chain complex: definition

Let G be an abelian group, and let (C , ) be a chain complex of free abelian groups:
n+1

n1

n
Cn+1 Cn
Cn1

(2.1)

Dualize the chain complex (2.1), i.e., apply Hom(; G) to it, to get the cochain complex:
n+1

n1

C n+1 C n C n1

(2.2)

C n := Hom(Cn , G),

(2.3)

n : C n C n+1

(2.4)

( n )() = (n+1 ), for C n and Cn+1 .

(2.5)

( n+1 n )() = n+1 n+2 = 0,

(2.6)

with
and where the coboundary map
is defined by
It follows that
since n+1 n+2 = 0 in the chain complex (2.1).
Definition 8. The n-th cohomology group H n (C ; G) with G-coefficients of the chain complex C is defined by:
H n (C ; G) := Hn (C ; ) := ker( : C n C n+1 )/Image( : C n1 C n ).
33

(2.7)

2.2

Relation between cohomology and homology

In this section, we explain how each cohomology group H n (C ; G) can be computed only in
terms of the coefficients G and the integral homology groups H (C ) of (C , ).

2.2.1

Ext groups

Let H and G be given abelian groups. Consider a free resolution of H:

f2

f1

f0

F : F1 F0 H 0
Dualize it with respect to G, i.e., apply Hom(, G) to it, to get the cochain complex
f

2
1
0

F1
F0
H 0

where we set H = Hom(H, G) and similarly for Fi . After discarding H , we get the cochain
complex involving only the Fi s, and we consider its cohomology groups.

H n (F ; G) = ker fn+1
/Imagefn

The Ext groups are defined as:

Extn (H, G) := H n (F ; G).

(2.8)

The following result holds:

Lemma 16. The Ext groups are well-defined, i.e., independent of the choice of resolution
F of H.
As in the case of the Tor functor, one can thus work with the free resolution of H given
by
0 F1 F0 H 0,
where F0 is the free abelian group on the generators of H, while F1 is the free abelian group
on the relations of H. In particular, it follows that
Extn (H, G) = 0 , n 1.
We also get that
Ext0 (H, G) = Hom(H, G).
For simplicity, we set:
Ext(H, G) := Ext1 (H, G).
Proposition 17. The Ext group Ext(H, G) satisfies the following properties:
(a) Ext(H H 0 , G) = Ext(H, G) Ext(H 0 , G).
34

(2.9)

(b) If H is free, then Ext(H, G) = 0.

(c) Ext(Z/n, G) = G/nG.
Proof. For (a) use the fact that a free resolution of H H 0 is a direct sum of free resolutions
of H and resp. H 0 . For (b), if H is free, then 0 H H 0 is a free resolution of H,
so Ext(H, G) = 0. For part (c), start with the free resolution of Z/n given by
n

0 Z Z Z/n 0,
dualize it and use the fact that Hom(Z, G) = G to conclude that Ext(Z/n, G) = G/nG.
As an immediate consequence of these properties, we get the following:
Corollary 18. If H is a finitely generated abelian group, then :
Ext(H, G) = Ext(Torsion(H), G) = Torsion(H) Z G.

(2.10)

Proof. Indeed, H decomposes into a free part and a torsion part, and the claim follows by
Proposition 17.

2.2.2

Universal Coefficient Theorem

The following result shows that cohomology is entirely determined by its coefficients and the
integral homology:
Theorem 19. Given an abelian group G and a chain complex (C , ) of free abelian groups
with homology H (C ), the cohomology group H n (C ; G) fits into a natural short exact sequence:
h
0 Ext(Hn1 (C ), G) H n (C ; G) Hom(Hn (C ), G) 0
(2.11)
In addition, this sequence is split, that is,
H n (C ; G)
= Ext(Hn1 (C ), G) Hom(Hn (C ), G).

(2.12)

Proof. (Sketch)
The homomorphism h : H n (C ; G) Hom(Hn (C ), G) is defined as follows. Let Zn = ker n ,
Bn = Image n+1 , in : Bn , Zn the inclusion map, and Hn = Zn /Bn . Let [] H n (C ; G).
Then is represented by a homomorphism : Cn G, so that n := n+1 = 0, which
implies that |Bn = 0. Let 0 := |Zn , then 0 vanishes on Bn , so it induces a quotient
homomorphism 0 : Zn /Bn G, i.e., 0 Hom(Hn (C ), G). We define h by
h([]) = 0 .
Notice that if Image n1 , i.e., = n1 = n , then |Zn = 0, so 0 = 0, which shows
that h is well-defined. It is not hard to show that h is an epimorphism, and

ker h = Coker(in1 : Zn1

Bn1
) = Ext(Hn1 (C ), G),

35

(2.13)

where the Ext group is defined with respect to the free resolution of Hn1 (C ) given by
in1

0 Bn1 Zn1 Hn1 (C ) 0.

Remark 3. The splitting in the above universal coefficient theorem is not natural; see
Exercise 8 at the end of this chapter for an example.
The following special case of Theorem 19 is very useful in calculations:
Corollary 20. Let (C , ) be a chain complex so that its (integral) homology groups H are
finitely generated, and let Tn = Torsion(Hn ). Then we have natural short exact sequences:

This sequence splits, so:

0 Tn1 H n (C ; Z) Hn /Tn 0

(2.14)

H n (C ; Z)
= Tn1 Hn /Tn .

(2.15)

Finally, we have the following easy application of Theorem 19:

Proposition 21. If a chain map : C C0 between chain complexes C and C0 induces isomorphisms on integral homology groups, then induces isomorphisms on the
cohomology groups H (; G) for any abelian group G.
Proof. By the naturality part of Theorem 19, we have a commutative diagram:
0 Ext(Hn1 (C ), G) H n (C ; G) Hom(Hn (C ), G) 0
( )

( )
0 Ext(Hn1 (C0 ), G) H n (C0 ; G) Hom(Hn (C0 ), G) 0
The claim follows by the five-lemma, since and its dual are isomorphisms.

2.3
2.3.1

Cohomology of spaces
Definition and immediate consequences

Suppose X is a topological space with singular chain complex (C (X), ). The group of
singular n-cochains of X is defined as:
C n (X; G) := Hom(Cn (X), G).
So n-cochains are functions from singular n-simplices to G.
The coboundary map
n : C n (X; G) C n+1 (X; G)
36

(2.16)

is defined as the dual of the corresponding boundary map n+1 : Cn+1 Cn , i.e., for
C n (X; G), we let
n+1

n := n+1 : Cn+1 (X) Cn (X) G.

(2.17)

n+1 n = 0,

(2.18)

It follows that
and for a singular (n + 1)-simplex : n+1 X we have:
n

() =

n+1
X

(1)i (|[v0 , ,vi , ,vn+1 ] ).

(2.19)

i=0

Definition 9. The cohomology groups of X with G-coefficients are defined as:

H n (X; G) := ker( n : C n (X; G) C n+1 (X; G))/Image( n1 : C n1 (X; G) C n (X; G)).
(2.20)
n
n1
Elements of ker are called n-cocycles, and elements of Image
are called n-coboundaries.
Remark 4. Nota that is an n-cocycle if, by definition, it vanishes on n-boundaries.
Since the groups Cn (X) of singular chains are free, we can employ Theorem 19 to compute
the cohomology groups H n (X; G) in terms of the coefficients G and the integral homology
of X. More precisely, we have natural short exact sequences:
0 Ext(Hn1 (X), G) H n (X; G) Hom(Hn (X), G). 0

(2.21)

Moreover, these sequences split, though not naturally.

Let us now derive some immediate consequences from (2.21):
(a) If n = 0, (2.21) yields that
H 0 (X; G) = Hom(H0 (X), G),

(2.22)

or equivalently, H 0 (X; G) consists of all functions from the set of path-connected components of X to the group G.
(b) If n = 1, the Ext-term in (2.21) vanishes since H0 (X) is free, so we get:
H 1 (X; G) = Hom(H1 (X), G).

(2.23)

Remark 5. Theorem 19 also works for modules over a PID. In particular, if G = F is a

field, then
H n (X; F ) ' Hom(Hn (X), F ) ' HomF (Hn (X; F ), F ) = Hn (X, F )
Thus, with field coefficients, cohomology is the dual of homology.
37

Example 23. Let X be a point space. From (2.21), we have:

H i (X; G) = Hom(Hi (X), G) Ext(Hi1 (X), G).
And since

(
Z, i = 0
Hi (X) =
0, otherwise,

we get
(
G, i = 0
Hom(Hi (X), G) =
0, otherwise.
Furthermore, since Hi (X) is free for all i, we also have that Ext(Hi1 (X), G) = 0, for all i.
Altogether,
(
G, i = 0
H i (X; G) =
0, otherwise.
Example 24. Let X = S n . Then we have

Z, i = 0, n
Hi (X) =
0, otherwise.
Thus the Ext-term in the universal coefficient theorem vanishes and we get:

G, i = 0 or n
i
H (X; G) = Hom(Hi (X), G) =
0, otherwise.

2.3.2

Reduced cohomology groups

We start with the augmented singular chain complex for X:

C1 (X) C0 (X) Z 0
P
P
with ( i ni xi ) = i ni . After dualizing it (i.e., applying Hom(; G)), we get the augmented
cochain complex


C 1 (X; G) C 0 (X; G) G 0.
Note that since  = 0, we get by dualizing that  = 0. The homology of this augmented
e i (X; G).
cochain complex is the reduced cohomology of X with G-coefficients, denoted by H
e i (X; G) = H i (X; G), if i > 0, and by the universal coeffiIt follows by definition that H
e 0 (X; G) = Hom(H
e 0 (X), G).
cient theorem (applied to the augmented chain complex), we get H

38

2.3.3

Relative cohomology groups

To define relative cohomology groups H n (X, A; G) for a pair (X, A), we dualize the relative
chain complex by setting
C n (X, A; G) := Hom(Cn (X, A), G).

(2.24)

The group C n (X, A; G) can be identified with functions from n-simplices in X to G that
vanish on simplices in A, so we have a natural inclusion
C n (X, A; G) , C n (X; G).

(2.25)

: C n (X, A; G) C n+1 (X, A; G)

(2.26)

The relative coboundary maps

are obtained by restrcting the absolute ones, so they satisfy 2 = 0. So the relative cohomology groups H n (X, A; G) are defined.
We next dualize the short exact sequence
j

0 Cn (A) Cn (X) Cn (X, A) 0

to get another short exact sequence
j

0 C n (A; G) C n (X; G) C n (X, A; G) 0,

(2.27)

where the exactness at C n (A; G) follows by extending a cochain in A by zero". More

precisely, for C n (A; G), we define a function b : Cn (X) G by

(), if Cn (A)
b
()
=
0,
if Image() A =
b is a well-defined element of C n (X; G) since Cn (X) has a basis made of simplices contained
b = .
in A and those contained in X \ A. It is clear that i ()
Since i and j commute with , it follows that i and j commute with . So we obtain a
short exact sequence of cochain complexes:
j

0 C (A; G) C (X; G) C (X, A; G) 0.

(2.28)

By taking the associated long exact sequence of homology groups, we get the long exact
sequence for the cohomology groups of the pair (X, A):

/ H n (X, A; G) j

/ H n (X; G) i / H n (A; G)

/ H n+1 (X, A; G)

(2.29)

We can also consider above the augmented chain complexes on X and A, and get a long
e n (X, A; G) = H n (X, A; G):
exact sequence for the reduced cohomology groups, with H

/ H n (X, A; G)

e n (X; G)
/H

e n (A; G)
/H

/ H n+1 (X, A; G)

(2.30)

In particular, if A = x0 is a point in X, we get by (2.30) that

e n (X; G)
H
= H n (X, x0 ; G).
39

(2.31)

2.3.4

Induced homomorphisms

If f : X Y is a continuous map, we have induced chain maps

f# :

/ Cn (Y )

Cn (X)
( : n X) 

/ (f

: n X Y )

satisfying f# = f# .
Dualizing f# with respect to G, we get maps
f # : C n (Y ; G) C n (X; G),
with f # () = (f# ) and f # = f # (which is obtained by dualizing f# = f# ). Thus, we
get induced homomorphisms on cohomology groups:
f : H n (Y, G) H n (X, G).
In fact, we can repeat the above for maps of pairs, say f : (X, A) (Y, B). And note that
the universal coefficient theorem also works for pairs because Cn (X, A) = Cn (X)/Cn (A) is
free abelian. So, by naturality, we get a commutative diagram for a map of pairs f : (X, A)
(Y, B):
/

/ H n (X, A; G)
O

Ext(Hn1 (X,
A), G)
O
(f )

2.3.5

H n (Y, B; G)

0
/

(f )

Ext(Hn1 (Y, B), G)

Hom(Hn (X,
A), G)
O
/

Hom(Hn (Y, B), G)

Homotopy invariance

Theorem 22. If f ' g : (X, A) (Y, B) and G is an abelian group, then

f = g : H n (Y, B; G) H n (X, A; G).
Proof. Recall from the proof of the similar statement for homology that there is a prism
operator
P : Cn (X, A) Cn+1 (Y, B)
(2.32)
satisfying
f# g# = P + P

(2.33)

with f# and g# the induced maps on singular chain complexes. (In fact, if F : X I Y
denotes the homotopy, with F (x, 0) = f (x) and F (x, 1) = g(x), then the prism operator is
defined on generators ( : n X) Cn (X) by pre-composing F ( id) : n I Y
with an appropriate decomposition of n I into (n + 1)-dimensional simplices. Then one
notes that such a P takes Cn (A) to Cn+1 (B), hence it induces the relative prism operator of
(2.32).)
So the difference of the middle maps in the following diagram equals to the sum of the
two side paths":
40

Cn (X, A)
P

f#

Cn+1 (Y, B)

 

g#

Cn1 (X, A)

Cn (Y, B)

Then it follows from (2.33) that f = g on relative homology groups.

The claim about cohomology follows by dualizing the prism operator (2.32) to get
P : C n+1 (Y, B; G) C n (X, A; G)

(2.34)

which satisfies an identity dual to (2.33), that is,

f # g # = P + P .

(2.35)

This implies readily that f = g on relative cohomology groups.

The following is an immediate consequence of Theorem 22:
Corollary 23. If f : X Y is a homotopy equivalence, then f : H n (Y ; G) H n (X; G)
is an isomorphism, for any coefficient group G.
Example 25. We have:
i

H (R ; G) =

G, i = 0
0, otherwise.

This follows immediately by the homotopy invariance of cohomology groups, since Rn is

contractible.

2.3.6

Excision

Theorem 24. Given a topological space X, suppose that Z A X, with cl(Z) int(A).
Then the inclusion of pairs i : (X \ Z, A \ Z) , (X, A) induces isomorphisms
i : H n (X, A; G) H n (X \ Z, A \ Z; G)

(2.36)

for all n. Equivalently, if A and B are subsets of X with X = int(A) int(B), then the
inclusion map (B, A B) , (X, A) induces isomorphisms in cohomology.
Proof. By the naturality of universal coefficient theorem, we have the commutative diagram:
/

Ext(Hn1 (X, A), G)

O

(i )

Ext(Hn1 (X \ Z, A \ Z), G)

H n (X, A; G)

H n (X \ Z, A \ Z; G)

(i )

Hom(Hn (X, A), G)

Hom(Hn (X \ Z, A \ Z), G)

By excision for homology, the maps i , hence (i ) , are isomorphisms. So by the five-lemma,
it follows that i is also an isomorphism.
41

2.3.7

Mayer-Vietoris sequence

Theorem 25. Let X be a topological space, and A and B be subsets of X so that

X = int(A) int(B).
Then there is a long exact sequence of cohomology groups:

H n (X; G) H n (A; G) H n (B; G) H n (A B; G) H n+1 (X; G)

(2.37)
Proof. There is a short exact sequence of cochain complexes, which at level n is given by:
0

/ C n (A + B; G)

C n (A; G) C n (B; G)

C n (A B; G)

Hom(Cn (A + B), G)
where Cn (A + B) is the set of simplices in X which are sums of simplices in either A or B,
and the maps are defined by
() = (|Cn (A) , |Cn (B) )
and
(, ) = |Cn (AB) |Cn (AB) .
Moreover, since C (A + B) , C (X) is a chain homotopy, it follows by dualizing that
C (A + B; G) and C (X; G) are chain homotopic, and thus H (A + B; G)
= H (X; G). The
cohomology Mayer-Vietoris sequence (2.37) is the long exact cohomology sequence of the
above short exact sequence of cochain complexes.
Remark 6. A similar Mayer-Vietoris sequence holds can be obtained for the reduced cohomology groups.
Example 26. Let us compute the cohomology groups of S n by using the above MayerVietoris sequence. Cover S n by two open sets A = S n \ N and B = S n \ S, where N and S
are the North and resp. South pole of S n . Then we have A B ' S n1 and A ' B ' Rn .
Thus by the Mayer-Vietoris sequence for reduced cohomology, together with Example 25,
homotopy invariance and induction, we get:

G, i = n
i
n
i1
n1
in
0

e
e
e
H (S ; G) = H (S ; G) = = H (S ; G) =
0, otherwise.

42

2.3.8

Cellular cohomology

Definition 10. Let X be a CW complex. The cellular cochain complex of X, (C (X; G), d ),
is defined by setting:
C n (X; G) := H n (Xn , Xn1 ; G),
for Xn the n-skeleton of X, and with coboundary maps
dn = n j n
fitting in the following diagram (where the coefficient group for cohomology is by default G):
H n1 (Xn1 )
j n1

H n1 (Xn1 , Xn2 )

n1
dn1

&

dn

H n (Xn , Xn1 )
jn

H n+1 (Xn+1 , Xn )
8

n
n

H (Xn )
Here, the diagonal arrows are part of cohomology long exact sequences for the relevant pairs.
For this reason, it follows that j n n1 = 0, and therefore
dn dn1 = n j n n1 j n1 = 0.
So (C (X; G), d ) is indeed a cochain complex.
The cellular cohomology of X with G-coefficients is by definition the cohomology of the
cellular cochain complex (C (X; G), d )
Just like in the case of cellular homology, we have the following identification:
Theorem 26. The singular and cellular cohomology of X are isomorphic, i.e.,
H n (X; G)
= H n (C (X; G))

(2.38)

for all n and any coefficient group G. Moreover, the cellular cochain complex (C (X; G), d )
is isomorphic to the dual of the cellular chain complex (C (X), d ), obtained by applying
Hom(; G).
Proof. Recall from Section 1.1.4 that for the cellular chain complex of X we have that
Cn (X) := Hn (Xn , Xn1 )
= Z# of n-cells
and Hi (Xn , Xn1 ) = 0 whenever i 6= n. So by the universal coefficient theorem, we obtain:
C n (X; G) := H n (Xn , Xn1 ; G)
= Hom(Cn (X), G)
43

(2.39)

since the Ext term vanishes. The universal coefficient theorem also yields that
H i (Xn , Xn1 ; G) = 0 if i 6= n,

(2.40)

since the groups Hi (Xn , Xn1 ) are either free or trivial.

From the long exact sequence of the pair (Xn , Xn1 ), that is,
H k (Xn , Xn1 ; G) H k (Xn ; G) H k (Xn1 ; G) H k+1 (Xn , Xn1 ; G) ,
we thus get for k 6= n, n 1 the isomorphisms
H k (Xn ; G)
= H k (Xn1 ; G).

(2.41)

Therefore, if k > n, we obtain by induction:

H k (Xn ; G)
= H k (Xn1 ; G)
= H k (Xn2 ; G)
=
= H k (X0 ; G) = 0

(2.42)

since X0 is just a set of points.

We next claim that there is an isomorphism
H n (Xn+1 ; G)
= H n (X; G).

(2.43)

First recall from Lemma 2(c) that the inclusion Xn+1 , X induces isomorphisms on homology groups Hk , for k < n + 1. So by the naturality of the universal coefficient theorem, we
get the following diagram with commutative squares:
/

Ext(Hn1 (X), G)

= (i )

Ext(Hn1 (Xn+1 ), G)

H n (X; G)

/ Hom(Hn (X), G)

0
/

= (i )

/ H n (Xn+1 ; G)

h/

Hom(Hn (Xn+1 ), G)

Then, by using the five-lemma, it follows that the middle map

i : H n (X; G) H n (Xn+1 ; G)
is also an isomorphism.
Altogether, by using (2.42) and (2.43), we get the following diagram (where the diagonal
arrows are part of long exact sequences of pairs):
H n1 (Xn2 )
=0
7

H n1 (Xn1 )
j n1

/ H n1 (Xn1 , Xn2 )

n1
dn1

'
/ H n (Xn , Xn1 )

dn

jn

'

/ H n+1 (Xn+1 , Xn ) /
9
n

H n (Xn )
7

H (X)
= H n (Xn+1 )
n

44

&

H n (Xn1 )
=0

Thus, by using the definition dn = n j n of the cellular coboundary maps, and after noting
that j n1 and j n are onto and is injective, we obtain the following sequence of isomorphisms:
H n (X; G)
= H n (Xn+1 ; G)

= Image()

= ker( n )

= ker(dn )/ ker(j n )

= ker(dn )/Image( n1 )

= ker(dn )/Image( n1 j n1 )

= ker(dn )/Image(dn1 ).

(2.44)

The only claim left to prove is that

dn = (dn+1 ) .

(2.45)

By definition, the cellular coboundary map dn is the composition:

jn

dn : H n (Xn , Xn1 ; G) H n (Xn : G) H n+1 (Xn+1 , Xn ; G),

and, similarly, the boundary map dn+1 of the cellular chain complex is given by:
n+1

jn

dn+1 : Hn+1 (Xn+1 , Xn ) Hn (Xn ) Hn (Xn , Xn1 ).

Let us now consider the following diagram:
dn :

jn

H n (Xn , Xn1 ; G)

= h

(dn+1 ) :

Hom(Hn (Xn , Xn1 ), G)

(jn

H n (Xn ; G)


H n+1 (Xn+1 , Xn ; G)

= h

h
)

/ Hom(Hn (Xn ), G) (n+1 /

Hom(Hn+1 (Xn+1 , Xn ), G)

The composition across the top is the cellular coboundary map dn , and we want to conclude
that it is the same as the composition (dn+1 ) across the bottom row. The extreme vertical
arrows labelled h are isomorphisms by the universal coefficient theorem, since the relevant
Ext terms vanish (by using (2.40)). So it suffices to show that the diagram commutes. The
left square commutes by the naturality of universal coefficient theorem for the inclusion map
(Xn , ) , (Xn , Xn1 ), and the right square commutes by a simple diagram chase.
Example 27. Let X = RP2 . Then X has one cell in each dimension 0, 1, and 2, and the
cellular chain complex of X is:
0

45

To compute the (cellular) cohomology H (X; Z), we dualize (i.e., apply Hom(, Z)) the
above cellular chain complex, and get:
0o

Zo

Zo

Zo

Thus, we have

i=0
Z,
i
2
Z/2, i = 2
H (RP ; Z) =

0,
otherwise.
Similarly, in order to calculate H (X; Z/2), we dualize the cellular chain complex of X with
respect to Z/2 (i.e., by applying Hom(, Z/2)) to get:
0o

Z/2 o

We then have:
H (RP ; Z/2)
=
i

Z/2 o


Z/2 o

Z/2, i = 0, 1, or 2
0,
otherwise.

Example 28. Let K be the Klein bottle and let us compute H (K; Z/3) and H (K; Z/3).
The cellular chain complex of K is given by:
/Z

(2,0)

/Z Z 0

/Z

/0

So the cellular chain complex of K with Z/3-coefficients is given by:

/ Z/3 (2,0) / Z/3 Z/3

/ Z/3

/0

Note that the map (2, 0) : Z/3 Z/3 Z/3 is an isomorphism on the first component, so
we get:

Z/3, i = 0 or 1
Hi (K; Z/3) =
0,
otherwise.
In order to compute the cohomology with Z/3-coefficients, we dualize the cellular chain
complex of K with respect to Z/3 to get:
0o

Z/3 o

(2,0)

Z/3 Z/3 o

Z/3 o

Therefore, we have
i

H (K; Z/3) =

Z/3, i = 0 or 1
0,
otherwise.

46

Exercises
1. Prove Lemma 16.
2. Show that the functor Ext(, ) is contravariant in the first variable, that is, if H,
H 0 and G are abelian groups, a homomorphism : H H 0 induces a homomorphism
: Ext(H 0 , G) Ext(H, G).
3. For a topological space X, let
h , i : C n (X) Cn (X) Z
be the Kronecker pairing given by h, i := (). In terms of this pairing, the coboundary
map : C n (X) C n+1 (X) is defined by h(), i = h, i for all Cn+1 (X). Show that
this pairing induces a pairing between cohomology and homology:
h , i : H n (X; Z) Hn (X; Z) Z.
4. Compute H (S n ; G) by using the long exact sequence of a pair, coupled with excision.
5. Compute the cohomology of the spaces S 1 S 1 , RP2 and the Klein bottle first with Z
coefficients, then with Z/2 coefficients.
6. Show that if f : S n S n has degree d, then f : H n (S n ; G) H n (S n ; G) is multiplication
by d.
7. Show that if A is a closed subspace of X that is a deformation retract of some neighborhood, then the quotient map X X/A induces isomorphisms
e n (X/A; G)
H n (X, A; G)
=H
for all n.
8. Let X be a space obtained from S n by attaching a cell en+1 by a degree m map.
e i (; Z)
Show that the quotient map X X/S n = S n+1 induces the trivial map on H
n+1
for all i, but not on H (; Z). Conclude that the splitting in the universal coefficient
theorem for cohomology cannot be natural.
e i (; Z) for all i, but not
Show that the inclusion S n , X induces the trivial map on H
on Hn (; Z).
9. Let X and Y be path-connected and locally contractible spaces such that H 1 (X; Q) 6= 0
and H 1 (Y ; Q) 6= 0. Show that X Y is not a retract of X Y .
10. Let X be the space obtained by attaching two 2-cells to S 1 , one via the map z 7 z 3
and the other via z 7 z 5 , where z denotes the complex coordinate on S 1 C. Compute the
cohomology groups H (X; G) of X with coefficients:
47

(a) G = Z.
(b) G = Z/2.
(c) G = Z/3.

48

Chapter 3
Cup Product in Cohomology
Let us motivate this chapter with the following simple, but hopefully convincing example.
Consider the spaces X = CP 2 and Y = S 2 S 4 . As CW complexes, both X and Y have
one 0-cell, one 2-cell and one 4-cell. Hence the cellular chain complex for both X and Y is:
0

0 Z 0 Z 0 Z 0
So X and Y have the same homology and cohomology groups. Note that X and Y also have
the same fundamental groups: 1 (X) = 1 (Y ) = 0. A natural question is then whether X
and Y are homotopy equivalent. Similarly, one can ask if there is a map f : X Y inducing
isomorphisms on (co)homology groups. We will see below that by using cup products in
cohomology, we can show that the answer to both questions is negative.

3.1

Cup Products: definition, properties, examples

Definition 11. Let X be a topological space, and fix a coefficient ring R (e.g., Z, Z/nZ,
Q). Let C k (X; R) and C l (X; R). The cup product ^ C k+l (X; R) is defined
by:
(3.1)
( ^ )( : k+l X) = (|[v0 , ,vk ] ) (|[vk , ,vk+l ] ),
where denotes the multiplication in ring R.
The aim is to show that this cup product of cochains induces a cup product of cohomology
classes. We need the following result which relates the cup product to coboundary maps.
Lemma 27.
( ^ ) = ^ + (1)k ^
for C k (X; R), and C l (X; R).
Proof. For : k+l+1 X we have
( ^ )() =

k+1
X

(1)i (|[v0 , ,bvi , ,vk+1 ] ) (|[vk+1 , ,vk+l+1 ] )

i=0

49

(3.2)

and
k

(1) ( ^ )() =

k+l+1
X

(1)i (|[v0 , ,vk ] ) (|[vk , ,bvi , ,vk+l+1 ] ).

i=k

When we add these two expressions, the last term of the first sum cancels with the first term
of theP
second sum, and the remaining terms are exactly ( ^ )() = ( ^ )() since
i
= k+l+1
vi , ,vk+l+1 ] .
i=0 (1) |[v0 , ,b
As immediate consequences of the above Lemma, we have:
Corollary 28. The cup product of two cocycles is again a cocycle. That is, if , are
cocycles, then ( ^ ) = 0.
Proof. This is true, since = 0 and = 0 imply by (3.2) that ( ^ ) = 0.
Moreover,
Corollary 29. If either one of or is a cocycle and the other a coboundary, then ^
is a coboundary.
Proof. Say = 0 and = . Then ^ = ^ = ( ^ ). Similarly, if
= 0, = then ^ = ^ = ( ^ ).
It follows from Corollary 28 and Corollary 29 that we get an induced cup product on
cohomology:
^
H k (X; R) H l (X; R) H k+l (X; R).
(3.3)
It is distributive and associative since it is so on the cochain level. If R has an identity
element, then there is an identity element for the cup product, namely the class 1 H 0 (X; R)
defined by the 0-cocycle taking the value 1 on each singular 0-simplex.
Considering the cup product as an operation on the the direct sum of all cohomology
groups, we get a (graded) ring structure on the cohomology i H i (X; R). We will elaborate
on the ring structure on cohomology groups induced by the cup product after looking at a
few examples and properties of the cup product.
Example 29. Let us consider the real projective plane RP2 . Its Z/2Z-cohomology is computed by:
(
Z/2Z for i = 0, 1, 2
H i (RP2 ; Z/2Z) =
0
otherwise.
Let H 1 (RP2 ; Z/2Z) = Z/2Z be the generator, and consider
2 := ^ H 2 (RP2 ; Z/2Z).
We claim that 2 6= 0, so 2 is in fact the generator of H 2 (RP2 ; Z/2Z).

50

Consider the cell structure on RP2 with two 0-cells v and w, three 1-cells e, e1 and e2 , and
two 2-cells T1 and T2 . The 2-cell T1 is attached by the word e1 ee1
2 , and the 2-cell T2 is
1
attached by the word e2 ee1 (see the figure below).
Since is a generator of H 1 (RP2 ; Z/2Z)
= Hom(H1 (RP2 ), Z/2Z), it is represented by a
cocycle
: C1 (RP2 ) Z/2Z
with (e) = 1, where we use the fact that e represents the generator of H1 (RP2 ). The cocycle
condition for translates into the identities:
0 = ()(T1 ) = (T1 ) = (e1 ) + (e) (e2 ).
0 = ()(T2 ) = (T2 ) = (e2 ) + (e) (e1 ).
As (e) = 1, without loss of generality we may take (e1 ) = 1 and (e2 ) = 0.
Next, note that 2 = ^ is represented by ^ , and we have:
( ^ )(T1 ) = (e1 ) (e) = 1
since T1 : [vww] RP2 . Similarly,
( ^ )(T2 ) = (e2 ) (e) = 0.
Since the generator of H2 (RP2 ; Z/2Z) is T1 + T2 , and we have
( ^ )(T1 + T2 ) = ( ^ )(T1 ) + ( ^ )(T2 ) = 1 + 0 = 1,
it follows that 2 (which is represented by ^ ) is the generator of H 2 (RP2 ; Z/2Z).
The cup product on cochains
C k (X; R) C l (X; R) C k+l (X; R)

51

restricts to cup products:

C k (X, A; R) C l (X; R) C k+l (X, A; R),
C k (X, A; R) C l (X, A; R) C k+l (X, A; R),
and
C k (X; R) C l (X, A; R) C k+l (X, A; R)
since C i (X, A; R) can be regarded as the set of cochains vanishing on chains in A, and if
or vanishes on chains in A, then so does ^ . So there exist relative cup products:
^

H k (X, A; R) H l (X; R) H k+l (X, A; R),

^

H k (X, A; R) H l (X, A; R) H k+l (X, A; R),

and

H k (X; R) H l (X, A; R) H k+l (X, A; R).

e (X; R).
In particular, if A is a point, we get a cup product on the reduced cohomology H
More generally, there is a cup product
^

H k (X, A; R) H l (X, B; R) H k+l (X, A B; R)

when A and B are open subsets of X or subcomplexes of the CW complex X. Indeed, the
absolute cup product restricts first to a cup product
C k (X, A; R) C l (X, B; R) C k+l (X, A + B; R),
where C k+l (X, A + B; R) is the subgroup of C k+l (X; R) consisting of cochains vanishing on
sums of chains in A and chains in B. If A and B are opens in X, then C k+l (X, A B; R) ,
C k+l (X, A + B; R) induces an isomorphism in cohomology, via the five-lemma and the fact
that the restriction maps C i (A B; R) C i (A + B; R) induce cohomology isomorphisms.
Let us now prove the following simple but important fact:
Lemma 1. The cup product is functorial, i.e., for a map f : X Y the induced maps
f : H i (Y ; R) H i (X; R) satisfy
f ( ^ ) = f () ^ f (),
and similarly in the relative case.
Proof. It sufficies to show the following cochain formula
f # ( ^ ) = f # () ^ f # ().

52

(3.4)

For C k (X; R) and C l (X; R) we have:

f # () ^ f # ()( : k+l X) = (f # )(|[v0 , ,vk ] ) (f # )(|[vk , ,vk+l ] )
= ((f# )|[v0 , ,vk ] ) ((f# )|[vk , ,vk+l ] )
= ( ^ )(f# )
= (f # ( ^ ))().

Definition 12. A graded ring is a ring A with a sum decomposition A = k Ak where the
Ak are additive subgroups so that the multiplication of A takes Ak Al to Ak+l . Elements
of Ak are called elements of degree k.
Definition 13. The cohomology ring of a topological space X is the graded ring
M
H (X; R) := (
H k (X; R), ^),
k0

with respect to the cup product operation. If R has an identity, then so does H (X; R).
Similarly, we define the cohomology ring of a pair H (X, A; R) by using the relative cup
product.
Remark 7. By scalar multiplication with elements of R, we can regard these cohomology
rings as R-algebras.
The following is an immediate consequence of Lemma 1:
Corollary 30. If f : X Y is a continuous map then we get an induced ring homomorphism
f : H (Y ; R) H (X; R).
Example 30. The isomorphisms
G
Y

=
H ( X ; R)
H (X ; R)

(3.5)

F
whose coordinates are induced by the inclusions i : X , X is a ring isomorphism
with respect to the coordinatewise multiplication in a ring product, since each coordinate
function i is a ring homomorphism. Similarly, the group isomorphism
_
Y
e ( X ; R)
e (X ; R)
H
H
(3.6)
=

is a ring isomorphism. Here the reduced cohomology is identified to cohomology relative to

a basepoint, and we use relative cup products. (We also assume the basepoints x X are
deformation retracts of neighborhoods.)
53

Example 31. From our calculations in Example 29 we have that:

H (RP2 ; Z/2Z) = {a0 + a1 + a2 2 | ai Z/2Z}
= (Z/2Z)[]/(3 ),
where is a generator of H 1 (RP2 ; Z/2Z).
Example 32.
H (S n , Z) = Z[]/(2 )
where is a generator of H n (S n ; Z). Indeed, we have
(
Z for i = 0, n
H i (S n ; Z) =
0 otherwise.
So if is a generator of H n (S n ; Z), then the only possible cup products are ^ 1 and
^ . However, ^ H 2n (S n ; Z) = 0. Hence 2 = 0.
Let us now recall that the cell structure on
RP = n0 RPn
consists of one cell in each non-negative dimension. The following result will be proved later
on in this section:
Theorem 31. The cohomology rings of the real (resp. complex) projective spaces are given
by:
(a)

H (RPn ; Z/2Z)
= Z/2[]/(n+1 )
where is the generator of H 1 (RPn ; Z/2Z).

(b)

H (RP ; Z/2Z)
= Z/2[]
where is the generator of H 1 (RPn ; Z/2Z).

(c)
H (CPn ; Z) = Z[]/( n+1 )
where is the generator of H 2 (CPn ; Z).
(d)
H (CP ; Z) = Z[]
where is the generator of H 2 (CPn ; Z).
Before discussing the proof of the above theorem, let us get back to the following motivating example:
54

Example 33. We saw at the beginning of this chapter that the spaces X = CP2 and
Y = S 2 S 4 have the same homology and cohomology groups, and even the same CW
structure. The cup products can be used to decide whether these spaces are homotopy
equivalent. Indeed, let us consider the cohomology rings H (X; Z) and H (Y ; Z). From the
above theorem, we have that:
H (CP2 ; Z) = Z[]/( 3 ),
where is the generator of H 2 (CP2 ; Z). We also have a ring isomorphism
e (S 2 S 4 ; Z)
e (S 2 ; Z) H
e (S 4 ; Z),
H
=H
where H (S 2 ; Z) = Z[]/(2 ) and H (S 4 ; Z) = Z[]/( 2 ), with degree of equal to 2 and
degree of equal to 4. Moreover, 2 = 0 = 2 and ^ = 0.
Consider the cohomology generators in degree 2 and square them. In the case of H (CP2 ; Z),
2 is a generator of H 4 (CP2 ; Z), hence 2 6= 0. However, in the case of H (S 2 S 4 ; Z),
2 H 4 (S 2 ; Z) = 0. Hence the two cohomology rings of the two spaces are not isomorphic,
hence the two spaces are not homotopy equivalent.
Let us now get back to the proof of Theorem 31. We will discuss below the proof in
the case of RPn . The result in the case of RP follows from the finite-dimensional case
since the inclusion RPn , RP induces isomorphisms on H i (; Z/2) for i n by cellular
cohomology. The complex projective spaces are handled in precisely the same manner, using
Z-coefficients and replacing H k by H 2k and R by C.
We next prove the following result:
Theorem 32.
H (RPn ; Z/2) = Z/2[]/(n+1 ),

(3.7)

where is the generator of H 1 (RPn ; Z/2).

Proof. For simplicity, let us use the notation Pn := RPn and all coefficients for the cohomology groups are understood to be Z/2-coefficients.
We prove (3.7) by induction on n. Let i be a generator for H i (Pn ) and j be a generator
for H j (Pn ), with i + j = n. Since for any k < n the inclusion map u : Pk , Pn induces
isomorphisms on cohomology groups H l , for l k, it suffices by induction on n to show that
i ^ j 6= 0.
Recall now that Pn = S n /(Z/2), with
n

n+1

S = {(x0 , , xn ) R

n
X

x2l = 1}.

l=0

Let
i

S = {(x0 , , xi , 0, , 0) |

i
X
l=0

55

x2l = 1}

and
j

S = {(0, , 0, xnj , xn ) |

n
X

x2l = 1}

l=nj

be the i-th and j-th (sub)sphere respectively. Note that since i + j = n, we have that
xnj = xi . Hence S i S j = {(0, , 0, 1, 0, , 0)} with 1 is in the i-th position, i.e.,
the intersection consists of the two antipodal points with i-th coordinate 1 and all other
coordinates zero.

Hence, Pi = S i /(Z/2) and Pj = S j /(Z/2) are subsets of Pn = S n /(Z/2) so that

Pi Pj = {p} = (0 : : 0 : 1 : 0 : : 0)
with 1 is in the i-th place.
Let U Pn be the open subset consisting of points (x0 : : xn ) with xi 6= 0, i.e.,
U = {(x0 : : xi1 : 1 : xi+1 : : xn )},
and notice that the map


(x0 : : xi1 : 1 : xi+1 : : xn ) = (x0 , , xi1 , xi+1 , , xn )
is a homeomorphism U
= Rn which takes p to 0 Rn .
We clearly have that Pn = Pn1 U , where Pn1 is identified to the set of points in Pn with
the i-th coordinate equal to zero. Regarding U as the interior of the n-cell of Pn (attached
to Pn1 ), it follows that Pn {p} deformation retracts to Pn1 . Similarly, as {p} = Pi Pj ,
we have that Pi {p} ' Pi1 and Pj {p} ' Pj1 . All of this is represented schematically
in the figure below, where Pn is represented by a disc with its antipodal boundary points
identified.
Let us now write Rn = Ri Rj , with coordinates of factors denoted by (x0 , , xi1 ) and
(xi+1 , , xn ), respectively. Consider the following commutative diagram with horizontal

56

Figure 3.1:
arrows given by the (relative) cup product:
/

H i (Pn ) O H j (Pn )
/

H i (Pn , Pn Pj ) H j (Pn , Pn Pi )


H i (Rn , Rn Rj ) H j (Rn , Rn Ri )

n
H n (P
)
O

H n (Pn , Pn {p})


H n (Rn , Rn {0})

The diagram commutes by naturality of the cup product. Let us examine the bottom row
in the above diagram. Let Di denote a small closed i-disc in Ri with boundary S i1 . Then
by homotopy equivalence and excision we have:
H i (Rn , Rn Rj )
= H i (Rn , Rn int(Di ) Rj )

= H i (Di Rj , S i1 Rj )

= H i (Di Dj , S i1 Dj )

= H i ((Di , S i1 ) Dj )

= H i (Di , S i1 ).
Similarly,

H j (Rn , Rn Ri )
= H j ((Dj , S j1 ) Di )
= H j (Dj , S j1 )

and
H n (Rn , Rn {0})
= H n (Dn , S n1 )
= H n (Di Dj , S i1 Dj S j1 Di ).
Since Dn is an n-cell, its class [Dn ] (in the Z/2-cellular cohomology) generates H n (Dn , S n1 ),
and similar considerations apply to [Di H i (Di , S i1 ) and [Dj ] H j (Dj , S j1 ). So the
above isomorphisms and cellular cohomology show that the cup product of the bottom
57

arrow in the above commutative diagram takes the product of generators to a generator, i.e.,
it is given by
[Di ] [Dj ] 7 [Dn ].
The same will be true for the top row, provided we show that the four vertical maps in the
above diagram are isomorphisms.
For the bottom right vertical arrow, we have by excision that
H n (Pn , Pn {p})
= H n (U, U {p})
= H n (Rn , Rn {0}),

(3.8)

where the last isomorphism follows by using the homeomorphism : U Rn .

For the top right vertical arrow, we already noted that Pn {p} deformation retracts to
Pn1 , so we have
H n (Pn , Pn {p})
(3.9)
= H n (Pn , Pn1 )
= Z/2,
where the second isomorphism follows by cellular cohomology. Moreover, by using the long
exact sequence for the cohomology of the pair (Pn , Pn1 ) and the fact that H n (Pn1 ) = 0,
we get that the map Z/2 = H n (Pn , Pn1 ) H n (Pn )
= Z/2 is onto, hence an isomorphism.
Thus we get:
H n (Pn , Pn {p})
(3.10)
= H n (Pn )
To show that the two left vertical arrows are isomorphisms, consider the following commutative diagram.
H i (Pn ) o


(2)

(1)

H i (Pi ) o

H i (Pn , Pi1 ) o


(8)

(4)

H i (Pn , Pn Pj )

(3)

H i (Pi , Pi1 ) o

(9)

(5)

H i (Rn , Rn Rj )

(6)

H i (Pi , Pi {p})

(10)

(7)

/ H i (Ri , Ri

{0})

It suffices to show that all these maps are isomorphisms. (Then to finish the proof of
the theorem, just interchange i and j.) First note that (Rn , Rn Rj ) = (Ri , Ri {0}) Rj
deformation retract to (Ri , Ri {0}), so the arrow (7) is an isomorphism. As already pointed
out, (10) is an isomorphism by (3.8). Moreover, (9) is an isomorphism as in (3.9), and (8)
is an isomorphism as in (3.10). The arrow (1) is an isomorphism by cellular homology, and
the arrow (3) is an isomorphism by cellular homology and the naturality of the cohomology
long exact sequence. By commutativity of the left square, it then follows that (2) is an
isomorphism. In order to show that (4) is an isomorphism, we note that Pn Pj deformation
retracts onto Pi1 . Indeed, a point v = (x0 : : xn ) Pn Pj has at least one of the first
i coordinates non-zero, so the function
ft (v) := (x0 : : xi1 : txi : : txn )
gives, as t decreases from 1 to 0, a deformation retract from Pn Pj onto Pi1 .
Since (3), (4) and (9) are isomorphisms, the commutativity of the middle square yields that
(6) is an isomorphism. Finally, since (6), (7) and (10) are isomorphisms, the commutativity
of the right square yields that (5) is an is0morphism, which completes the proof of the
theorem.
58

Example 34. Let us consider the spaces RP2n+1 and RP2n S 2n+1 . First note that these
spaces have the same CW structure and the same cellular chain complex, so they have the
same homology and cohomology groups. However, we claim that RP2n+1 and RP2n S 2n+1
are not homotopy equivalent. In order to justify the claim, we first compute their Z/2Zcohomology rings.
From the above theorem, the cohomology ring of RP2n+1 is:
H (RP2n+1 ; Z/2Z) = Z/2Z[]/(2n+2 ),
where is a degree one element, namely the generator of H 1 (RP2n+1 ; Z/2Z).
We also have a ring isomorphism
e (RP2n S 2n+1 ; Z/2Z)
e (RP2n ; Z/2Z) H
e (S 2n+1 ; Z/2Z)
H
=H
with H (RP2n ; Z/2Z)
= Z/2Z[]/( 2n+1 ) for the degree 1 generator of H 1 (RP2n ; Z/2Z),
and H (S 2n+1 ; Z/2Z)
= Z/2Z[]/( 2 ) for the generator of H 2n+1 (S 2n+1 ; Z/2Z) of degree
2n + 1.
If there was a homotopy equivalence f : RP2n+1 RP2n S 2n+1 , then the generators of
degree one would correspond isomorphically to each other, i.e., we would get f () = . But
as f is a ring isomorphism, this would then imply that: f ( 2n+1 ) = (f ())2n+1 = 2n+1 .
However, this yields a contradiction, since 2n+1 = 0, thus f ( 2n+1 ) = 0, while 2n+1 6= 0
since 2n+1 generates H 2n+1 (RP2n+1 ; Z/2Z).

59

3.2

Application: Borsuk-Ulam Theorem

In this section we use cup products in order to prove the following result:
Theorem 33 (Borsuk-Ulam). If n > m 1, there are no maps g : S n S m commuting
with the antipodal maps, i.e., for which g(x) = g(x), for all x S n .
Proof. We prove the theorem by contradiction. Assume that there is a map g : S n S m
commuting with the antipodal maps. Then g carries
of antipodal points (x, x) in S n

pairs
m
to pairs of antipodal points g(x), g(x) = g(x) in S . So, by passage to the quotient, g
induces a map
f : RPn RPm
[x] 7 [g(x)]
which makes the following diagram commutative:
g

Sn
p0

Sm

RPn

RPm

Here p and p0 are the two-sheeted covering maps.

We claim that there exists a lift f 0 of f , i.e., f = pf 0 in the following diagram:
;S

f0

RPn

RPm

Let us for now assume the claim and complete the proof of the theorem. Consider the
following diagram:
g

Sn

p0

RPn

f0

5 ;S

RPm

We have pg = f p0 = pf 0 p0 , the second equality following from the above claim. This implies
that both g and f 0 p0 are lifts of f p0 . Under the two-sheeted covering map p, antipodal points
in S m are mapped to the same point in RPm . Therefore, pg = pf 0 p0 implies that at a point
x S n , we have g(x) = f 0 p0 (x) or ag(x) = f 0 p0 (x), where a : S m S m is the antipodal
map. But ag(x) = g(x) = g(x) and f 0 p0 (x) = f 0 p0 (x). Thus at x S n , one of following
equalities holds: g(x) = f 0 p0 (x) or g(x) = f 0 p0 (x). Since g and f 0 p0 are lifts of f p0 and
they coincide at a point, it follows by the uniqueness of the lift that g = f 0 p0 . But this is a
contradiction since p0 (x) = p0 (x), hence f 0 p0 (x) = f 0 p0 (x), while g(x) 6= g(x) = g(x).
It remains to prove the claim. A lift for f exists iff
f (1 (RPn )) p (1 (S m )).
60

(3.11)

If m = 1, the only homomorphism

f : 1 (RPn )
= Z/2 1 (RP1 )
=Z
is the trivial one, so (3.11) is satisfied.
If m > 1, both groups 1 (RPn ) and 1 (RPm ) are Z/2. We will use cup products to show
that the induced map f : Z/2 Z/2 on fundamental groups is the trivial map. Let
m H (RPm ; Z/2) and n H (RPn ; Z/2) be the generators of degree 1, and consider the
induced ring homomorphism
f : H (RPm ; Z/2) H (RPn ; Z/2).
We have:
m+1
0 = f (m
) = f (m )m+1 ,

so f (m ) H 1 (RPn ; Z/2) has order m + 1 < n + 1. Therefore,

f (m ) 6= n .
Since H 1 (RPn ; Z/2) = Z/2 = hn i, this implies that
f (m ) = 0.
Let i : RP1 , RPn and j : RP1 , RPm be the inclusions obtained by setting all
but the first two homogeneous coordinates equal to zero. By cellular cohomology, the map
j : H 1 (RPm ) H 1 (RP1 ) is an isomorphism, so j (m ) is the generator of H 1 (RP1 ), and in
particular,
j (m ) 6= 0.
On the other hand,
(f i) (m ) = i (f (m )) = 0.
So (f i) 6= j , hence the maps f i and j are not homotopic.
But the homotopy classes of i and j generate 1 (RPn ) and 1 (RPm ), respectively. So the
homomorphisms
f : 1 (RPn ) ' Z/2 1 (RPm ) ' Z/2
[i] 7 [f i] 6= [j]
maps the generator [i] to an element of Z/2 other than the generator [j], i.e., f = 0. This
proves the claim, and completes the theorem.

61

Exercises
1. Show that if X is the union of contractible open subsets A and B, then all cup products
of positive-dimensional classes in H (X) are zero. In particular, this is the case if X is a
suspension. Conclude that spaces such as RP2 and T 2 cannot be written as unions of two
open contractible subsets.
2. Is the Hopf map
f : S 3 C2 S 2 = C {}, (z, w) 7

z
w

nullhomotopic? Explain.
3. Is there a continuous map f : X Y inducing isomorphisms on all of the cohomology

=
groups (i.e., f : H i (Y ; Z) H i (X; Z), for all i) but X and Y do not have isomorphic
cohomology rings (with Z coefficients)? Explain your answer.
4. Show that RP3 and RP2 S 3 have the same cohomology rings with integer coefficients.
5.
(a) Show that H (CPn ; Z)
= Z[x]/(xn+1 ), with x the generator of H 2 (CPn ; Z).
(a) Show that the Lefschetz number f of a map f : CPn CPn is given by
f = 1 + d + d2 + + dn ,
where f (x) = dx for some d Z, and with x as in part (a).
(c) Show that for n even, any map f : CPn CPn has a fixed point.
(d) When n is odd, show that there is a fixed point unless f (x) = x, where x denotes
as before a generator of H 2 (CPn ; Z).
6. Use cup products to compute the map H (CPn ; Z) H (CPn ; Z) induced by the map
CPn CPn that is a quotient of the map Cn+1 Cn+1 raising each coordinate to the d-th
power, (z0 , , zn ) 7 (z0d , , znd ), for a fixed integer d > 0. (Hint: First do the case n = 1.)
7. Describe the cohomology ring H (X Y ) of a join of two spaces.
8. Let H = R1RiRj Rk be the skew-field of quaternions, where i2 = j 2 = k 2 = 1
and ij = k = ji, jk = i = kj, ki = j = ik. For a quaternion q =
a + bi + cj + dk,
a, b, c, d R, its conjugate is defined by q = a bi cj dk. Let |q| := a2 + b2 + c2 + d2 .
(a) Verify the following formulae in H: q q = |q|2 , q1 q2 = q2 q1 , |q1 q2 | = |q1 | |q2 |.

62

(b) Let S 7 H H be the unit sphere, and let f : S 7 S 4 = HP1 = H {} be given

by f (q1 , q2 ) = q1 q2 1 . Show that for any p S 4 , the fiber f 1 (p) is homeomorphic to
S 3.
(c) Let HPn be the quaternionic projective space defined exactly as in the complex case as
the quotient of Hn+1 \ {0} by the equivalence relation v v, for H \ {0}. Show
that the CW structure of HPn consists of only one cell in each dimension 0, 4, 8, , 4n,
and calculate the homology of HPn .
(d) Show that H (HPn ; Z)
= Z[x]/(xn+1 ), with x the generator of H 4 (HPn ; Z).
(e) Show that S 4 S 8 and HP2 are not homotopy equivalent.
9. For a map f : S 2n1 S n with n 2, let Xf = S n f D2n be the CW complex obtained
by attaching a 2n-cell to S n by the map f . Let a H n (Xf ; Z) and b H 2n (Xf ; Z) be the
generators of respective groups. The Hopf invariant H(f ) Z of the map f is defined by
the identity a2 = H(f )b.
(a) Let f : S 3 S 2 = C {} be given by f (z1 , z2 ) = z1 /z2 , for (z1 , z2 ) S 3 C2 . Show
that Xf = CP2 and H(f ) = 1.
(b) Let f : S 7 S 4 = H {} be given by f (q1 , q2 ) = q1 q2 1 in terms of quaternions
(q1 , q2 ) S 7 , the unit sphere in H2 . Show that Xf = HP2 and H(f ) = 1.

63

3.3
3.3.1

Knneth Formula
Cross product

Let us motivate this section by consider the spaces S 2 S 3 and S 2 S 3 S 5 . Both spaces are
CW complexes with cells {e0 , e2 , e3 , e5 } in degrees, 0, 2, 3 and 5, respectively. So the cellular
chain complex for both spaces is:
0

0 Z 0 Z Z 0 Z 0
Hence both spaces have the same homology and cohomology groups. It is then natural to
ask the following:
Question 34. Are the spaces S 2 S 3 and S 2 S 3 S 5 homotopy equivalent?
The aim of this section is to convince the reader that the answer is No. More precisely,
we will show that the two spaces have different cohomology rings.
The cohomology ring H (S 2 S 3 S 5 ; Z) can be computed from the ring isomorphism
e (S 2 S 3 S 5 ; Z)
e (S 2 ; Z) H
e (S 3 ; Z) H
e (S 5 ; Z),
H
=H
with H (S 2 ; Z)
= Z[]/(2 ), H (S 3 ; Z)
= Z[]/( 2 ) and H (S 5 ; Z)
= Z[]/( 2 ), where
2
2
3
3
is the generator of H (S ; Z), is the generator of H (S ; Z) and is the generator of
H 5 (S 5 ; Z). Moreover, we have that ^ = 0. Indeed, let
p : S2 S3 S5 S2 S3
be the natural retraction map. Then p induces isomorphisms on H 2 and H 3 . So if
and
2
2
3
3
2
3

are the generators of H (S S ) and H (S S ), then = p

and = p . So
=0
^ )
^ = p
^ p = p (
since
^ = 0.
By the end of this section, we will show that the product of the generators of degree 2 and
degree 3 in the cohomology ring of S 2 S 3 is the generator in degree 5, so it is non-zero.
This will then completely answer the above question.
The following result is proved in [Hatcher, Theorem 3.14]:
Theorem 35. Let R be a commutative ring, and H k (X, A; R) and H l (X, A; R).
Then the following holds:
^ = (1)kl ^ .
(3.12)
Definition 14. A graded ring which satisfies a condition as in the above theorem is called
graded commutative. Hence the cohomology ring H (X, A; R) is a graded commutative ring.

64

Corollary 36. If H (X; R) is of odd degree and if H (X; R) has no elements of order
two, then ^ = 0.
Definition 15. Cross product or External cup product
Let X and Y be topological spaces, and denote by p and q the projections p : X Y X
and q : X Y Y . By using the cohomology maps defined by these projections, we have
an induced map denoted by :

H (X; R) H (Y ; R)
H (X Y ; R)
a
b
7 a b := p (a) ^ q (b)
All cohomology groups H i (X; R) and H i (Y ; R) have an R-module structure, hence so do
the corresponding cohomology rings H (X; R) and H (Y ; R). Since the map is bilinear,
the universal property for tensor products yields a group homomorphism called the cross
product, which we again denote by :

H (X; R) R H (Y ; R) H (X Y ; R)

(3.13)

So, by definition, we have that:

(a b) := a b.
The cross-product becomes a ring isomorphism if we put a ring structure on H (X; R) R
H (Y ; R) by the following multiplication operation:
(a b) (c d) = (1)deg(b)deg(c) (ac bd)

(3.14)

Indeed, we have:
((a b) (c d)) = (1)deg(b)deg(c) (ac bd)
= (1)deg(b)deg(c) (ac bd)
= (1)(deg b)deg(c) p (a ^ c) ^ q (b ^ d)
= (1)deg(b)deg(c) p (a) ^ p (c) ^ q (b) ^ q (d)
(3.12)

= p (a) ^ q (b) ^ p (c) ^ q (d)

= (a b) ^ (c d).

3.3.2

Knneth theorem in cohomology. Examples

The following result is very helpful in finding the cohomology ring of a product of CW
complexes:
Theorem 37. Knneth Formula
If X and Y are CW complexes, and H k (Y ; R) is a finitely generated free R-module for all
k, then the cross product

H (X; R) R H (Y ; R) H (X Y ; R)
65

is a ring isomorphism. Moreover, we have the following isomorphism of groups:

M
H n (X Y ; R)
H i (X; R) R H j (Y ; R)
=

(3.15)

i+j=n

In the next section, we will explain the content of Theorem 37 in a more general context.
Let us now work out some examples.
Example 35. Let us find the cohomology ring of S 2 S 3 , which appeared at the beginning
of this section. According to the Knneth formula, we have the following ring isomorphism:
H (S 2 S 3 ; Z)
= H (S 2 ; Z) Z H (S 3 ; Z)
If we let a H (S 2 ; Z) denote the degree 2 element which generates H 2 (S 2 ; Z) and b
H (S 3 ; Z) the degree 3 element which generates H 3 (S 3 ; Z), then (a 1) and (1 b)
(where 1 denotes the identity in the respective cohomology rings) will be the generators in
H (S 2 S 3 ; Z) of degree 2 and 3, respectively. Moreover, (a 1) ^ (1 b) = (a b)
will be a generator of degree 5 in H (S 2 S 3 ; Z).
In order to simplify the notations, we make the following definition.
Definition 16. Exterior Algebra
Let R be a commutative ring with identity. The exterior algebra over R, denoted
R [1 , 2 , . . .],
is the free R-module generated by products of the form:
i1 i2 ik , with i1 < i2 < < ik ,
and with associative and distributive multiplication defined by the rules:
i j = j i , if i 6= j
i2 = 0.
The empty product of i s is allowed and it gives the identity element 1 R [1 , 2 , . . .].
Example 36. Let us now show that
H (S 3 S 5 S 7 ; Z)
= Z [a3 , a5 , a7 ],

(3.16)

where ai is the generator of degree i in H (S 3 S 5 S 7 ; Z), for i = 3, 5, 7.

By the Knneth formula applied to the product of CW complexes S 3 S 5 S 7 , we have the
following ring isomorphism:
H (S 3 S 5 S 7 ; Z)
= H (S 3 ; Z) Z H (S 5 ; Z) Z H (S 7 ; Z).
Let i be the generator of degree i in H (S i ; Z) for i = 3, 5, 7. Then the generators of degree
3, 5 and 7 in H (S 3 S 5 S 7 ; Z) are given respectively by:
66

 a3 = (3 1 1)
a5 = (1 5 1)
a7 = (1 1 7 )
The product of these generators produce generators of higher degrees, i.e., 8, 10, 12 and 15,
in the cohomology ring H (S 3 S 5 S 7 ; Z). Let us compute some products of the elements:
a23 =
=
=
=

(3 1 1) ^ (3 1 1)


(3 1 1) (3 1 1)
(32 1 1)
0

and a similar result for a25 and a27 .

a3 a5 =
=
=
=

(3 1 1) ^ (1 5 1)


(3 1 1) (1 5 1)
(1)00 (3 5 1)
(3 5 1)

a5 a3 =
=
=
=

(1 5 1) ^ (3 1 1)


(1 5 1) (3 1 1)
(1)35 (3 5 1)
a3 a5

We have similar results for the other products too. The above calculations show that we
have an isomorphism H (S 3 S 5 S 7 ; Z)
= Z [a3 , a5 , a7 ].
Remark 8. It is easy to see that a similar result holds for the cohomology ring of any (finite)
product of odd dimensional spheres.
Example 37. By the Knneth formula we have the following ring isomorphism:
H (RP RP ; Z/2) = H (RP ; Z/2) Z2 H (RP ; Z/2)
= Z/2[] Z/2 Z/2[]
= Z/2[, ]
where and are generators of degree 1, and they commute since we work with Z/2coefficients.

67

Example 38. Let us now investigate if the spaces CP 6 and S 2 S 4 S 6 are homotopy
equivalent. Fortunately, there is an easy answer to this question. Consider the usual CW
structure for CP 6 and the product CW structure for S 2 S 4 S 6 . Both spaces have cells
only in even dimensions, but CP 6 has one cell in dimension 6, whereas S 2 S 4 S 6 has two
cells in dimension 6. It follows that H6 (CP 6 ) = Z, whereas H6 (S 2 S 4 S 6 ) = Z Z. So
CP 6 and S 2 S 4 S 6 are not homotopy equivalent. A more difficult approach to answer
the question would be to show that the cohomolgy rings for these spaces are not isomorphic.
We will do this in the following example.
Example 39. Let us show that if n > 1, the spaces CP
not homotopy equivalent. Consider the following cases:

n(n+1)
2

and S 2 S 4 S 2n are

If n = 1, then CP 1 is homeomorphic to S 2 .
If n = 2, then both the spaces CP 3 and S 2 S 4 have one cell in each of the dimensions {0, 2, 4, 6}. Thus they also have the same cellular chain/cochain complex and,
in particular, their homology/cohomology groups are isomorphic. We will, however,
distinguish these spaces by their cohomology rings.
If n 3, then CP n has one cell in each of the dimensions {0, 2, 4, . . . , 2n}, but the
cell structure of S 2 S 4 S 2n is different from that of CP n since, for example,
S 2 S 4 S 2n has two 6-cells. As both spaces have cells only in even dimensions,
we can already conclude that they have different homology and cohomology groups
since they have different cell structures.
We will now show that for n > 1 the two spaces have non-isomorphic cohomology rings.
First, the Knneth formula yields that:
H (S 2 S 4 S 2n ; Z)
= H (S 2 ; Z) Z H (S 4 ; Z) Z Z H (S 2n ; Z)
So a degree 2 element in this ring looks like (a 1 1 1), where a H 2 (S 2 ). The
square of this element is:
[(a 1 1 1)]2 = [(a 1 1 1)2 ]
= (a2 1 1 1)
= 0
n(n+1)

since a2 H 4 (S 2 ) = 0. However, in the case of CP 2 , we know that square of a non-zero

degree 2 element is a non-zero degree 4 element. Hence the cohomology rings of the two
spaces are not isomorphic.
Example 40. Let us use cup products and the Knneth formula in order to show that
S n S m is not a retract of S n S m , for n, m 1. First, consider the product CW structure
on S n S m : it consists of cells {e0 , em , en , em+n } with attaching maps : em e0 and

68

0 : en e0 coming from the factors. Hence S n S m is a subset of S n S m . (Note that

we also allow the case n = m.) Next, suppose by contradiction that there is a retract
r : S n S m S n S m.
So, if i : S n S m , S n S m denotes the inclusion, then the composition r i is the identity
map on S n S m . It follows that the cohomology map (r i) = i r is the identity, so
r : H (S n S m ) H (S n S m )
is a monomorphism. By the Knneth formula, we have a ring isomorphism

H (S n ) H (S m )
= H (S n S m ).
Hence, a non-zero element in H n (S n S m ) is of the form a1 := (a1), with a H n (S n ) a
non-zero class. Similarly, a non-zero element in H m (S n S m ) is of the form 1b := (1b),
for some non-zero class b H m (S m ). Let us now consider the product of non-zero elements
a 1 H n (S n S m ) and 1 b H m (S n S m ) in the ring H (S n S m ). We get:
(a 1) ^ (1 b) = (a 1) ^ (1 b)
= [(a 1) (1 b)]
= (a b)
=ab
6= 0,

(3.17)

since a b 6= 0 in H (S n ) H (S m ). We also have a ring isomorphism

e (S n S m )
e (S n ) H
e (S m ).
H
=H
Let , H (S n S m ) be the generators of degree n and m, respectively. Then
^ H n+m (S n S m ) = 0.
On the other hand, since r is a monomorphism, the classes r () and r () are non-zero
elements of degree n and resp. m in the cohomology ring H (S n S m ), so by the above
calculation, their product is non zero. But
r () ^ r () = r ( ^ ) = r (0) = 0,
which gives us a contradiction.

69

3.3.3

Knneth exact sequence and applications

In this section, we aim to provide the necessary background for Knneth-type theorems.
Let us fix coefficients in a PID ring R.
Given two chain complexes (C , ) and (C0 , 0 ), we define (C C 0 ) to be the complex
with:
n
M
0
0
(C C )n =
(Cp Cnp
)
(3.18)
p=0
0
and boundary map dn : (C C 0 )n (C C 0 )n1 which on Cp Cnp
is given by:
0
dn (a b) = (p a) b + (1)p (a np
b).

(3.19)

Then we have:


(d d)(a b) = d (a) b + (1)p (a 0 b)
= ( 2 a) b + (1)p1 (a) ( 0 b) + (1)p (a) ( 0 b) + (1)p a ( 02 b)
= 0.


So (C C 0 ) , d is a chain complex. It is therefore natural to ask the following question:
Question 38. How is the homology H ((C C 0 ) ) related to H (C ) and H (C0 )?
The answer is provided by the following result from homological algebra:
Theorem 39. Knneth exact sequence
Let R be a PID, and assume that for each i, Ci is a free R-module. Then for all n, there is
a split short exact sequence:
M
M




TorR Hp (C ), Hnp1 (C0 ) 0
Hp (C )R Hnp (C0 ) Hn ((CC) )
0
p

(3.20)
In what follows we discuss several applications of Theorem 39.
Knneth Formula for homology.
Let X and Y be two spaces, and let C and C0 denote the singular chain complexes of X
and Y , respectively. Then it is not hard to see that the singular chain complex C (X Y ) of
X Y is chain homotopy equivalent to (C C 0 ) , so they have the same homology groups.
We thus have the following important consequence of Theorem 39:
Corollary 40. Knneth Formula for homology
If X and Y are topological spaces, then the following holds:
Hn (X Y )
=

n
M

n1

M


Hp (X) Hnp (Y )
Tor Hp (X), Hnp1 (Y ) .

p=0

p=0

70

(3.21)

In particular, if all homology groups of X or Y are free R-modules, then:

Hn (X Y )
=

n
M

Hp (X) Hnp (Y ).

(3.22)

p=0

As a consequence of Corollary 40, we have:

Corollary 41. If the Euler characteristics (X) and (Y ) are defined, then (X Y ) is
defined, and:
(X Y ) = (X) (Y ).
(3.23)
Universal Coefficient Theorem for homology
The Universal Coefficient Theorem for homology can be seen as a consequence of Theorem
39 as follows: take C to be the singular chain complex of X and let C0 to be the chain
complex defined by: Cn0 = 0 if n 6= 0, C00 = R, and n0 = 0 for all n 0. We then get by
Theorem 39 that:


Hn (X; R)
(3.24)
= Hn (X) R Tor(Hn1 (X), R).
Remark 9. Note that (3.24) can also be obtained from (3.21) by taking Y to be a point.
Knneth formula for cohomology
Finally, we also have the following cohomology Knneth formula:
Corollary 42. Knneth formula for cohomology
If R is a PID, and all homology groups Hi (X; R) are finitely generated, then there is a split
exact sequence (with R-coefficients):
0

n
M

H (X) H

np

n+1
M




n
Tor H p (X), H np+1 (Y ) 0.
(Y ) H (X Y )
p=0

p=0

(3.25)
Moreover, if all cohomology groups H (X) of X (or Y ) are free over R, we get the following
isomorphism:
n
M
H n (X Y )
H p (X) H np (Y ).
(3.26)
=
i

p=0

Proof. (Sketch.) Let us indicate how this result is obtained from Theorem 39. We would
like to apply the Knneth exact sequence to the chain complexes defined by:
n
Cn := C n (X; R), n := X

and
0
0
Cn
:= C n (Y ; R), n
:= Yn .

71

However, note that Ci and Ci0 are not necessarily R-free. Indeed,
C n (X; R) = HomR (Cn (X; R), R),
but Cn (X; R) is not necessarily a finitely generated R-module. In order to get around this
problem, the idea is to replace the chain complex C (X; R) by a chain homotopic one, which
has finitely generated components. Here is where the assumption that Hi (X; R) are finitely
generated is used.

Exercises
1. Are the spaces S 2 RP4 and S 4 RP2 homotopy equivalent? Justify your answer!
2. Using cup products, show that every map S k+l S k S l induces the trivial homomorphism Hk+l (S k+l ) Hk+l (S k S l ), assuming k > 0 and l > 0.
3. Describe H (CP /CP1 ; Z) as a ring with finitely many multiplicative generators. How
does this ring compare with H (S 6 HP ; Z)?
4. Show that if Hn (X; Z) is finitely generated and free for each n, then H (X; Zp ) and
H (X; Z) Zp are isomorphic as rings, so in particular the ring structure with Z-coefficients
determines the ring structure with Zp -coefficients.
5. Show that the cross product map H (X; Z) H (Y ; Z) H (X Y ; Z) is not an
isomorphism if X and Y are infinite discrete sets.
6. Show that for n even S n is not an H-space, i.e., there is no map : S n S n S n so
that i1 = idS n and i2 = idS n , where i1 , i2 are the inclusions on factors.

72

Chapter 4
Poincar Duality
4.1

Introduction

In this chapter, we show that oriented n-manifolds enjoy a very special symmetry on their
(co)homology groups:
Theorem 43. Let M be a closed (i.e., compact without boundary), oriented and connected
manifold of dimension n. Then for all i 0 we have isomorphisms:
Hi (M ; Z)
= H ni (M ; Z).

(4.1)

In particular, we get:
Corollary 44. For all i 0, the isomorphisms
(4.1)

(U CT )

Hi (M ; Q)
= H ni (M ; Q)
= Hom(Hni (M ; Q), Q)

(4.2)

yield a non-degenerate bilinear pairing

Hi (M ; Q) Hni (M ; Q) Q.
Moreover, the complementary Betti numbers are equal, i.e.,
i (M ) = ni (M ).
In the next section we will explain in more detail the notion of orientability of manifolds.
Later on, we will describe explicitly the nature of the isomorphism (4.1) by using the cap
product operation _, i.e., we will show that it is realized by
_ [M ] : H ni (M ; Z) Hi (M ; Z),
where [M ] Hn (M ) is the fundamental (orientation) class" of the manifold M .
73

(4.3)

4.2

Manifolds. Orientation of manifolds

Definition 17. A Hausdorff space M is a (topological) manifold if any point x M has a

neighborhood Ux homeomorphic to Rn (where such a homeomorphism takes x to 0).
Let us now compute the local homology groups of a manifold M at some point x M :
(1)

Hi (M, M \ {x}; Z)
= Hi (Ux , Ux \ {x}; Z)
(2)

= Hi (Rn , Rn \ {0}; Z)

(3)

e i1 (Rn \ {0}; Z)
=H

(4.4)

(4)

e i1 (S n1 ; Z)
=H
(
Z, if i = n
=
0 , otherwise,

where (1) follows by excision, (2) by using the homeomorphism Ux

= Rn , (3) by the homology
long exact sequence of a pair, and (4) by using a deformation retract.
Definition 18. The dimension of a manifold M , denoted dim(M ), is the only non-vanishing
degree of the local homology groups of M .
Definition 19. A local orientation of an n-manifold M at x M is a choice x of one of
the two generators of the local homology group Hn (M, M \ {x}; Z) = Z.
Remark 10. A local orientation x at x M induces local orientations at all nearby points
y, i.e., if x and y are contained in a small ball B, then we have induced isomorphisms:

x Z = Hn (M, M \ {x})
Hn (M, M \ B) = Z
Hn (M, M \ {y}) = Z y ,

(4.5)

where the above isomorphisms are induced by deformation retracts.

Definition 20. A (global) orientation on an n-manifold M is a continuous choice of local
orientations, i.e., for every x M there exists a compact subset K Ux
= Rn and K
Hn (M, M \ K) such that y : Hn (M, M \ K) Hn (M, M \ {y}) takes K to y for all y K.
Notation: Let M be an n-manifold and K L M be compact subsets. Consider the
map induced by inclusion of pairs:
K : Hi (M, M \ L) Hi (M, M \ K).
Then for a Hi (M, M \ L), K (a) is called the restriction of a to K.
In the above notations, we have the following important result:
74

Theorem 45. For any oriented manifold M of dimension n and any compact K M , there
is a unique K Hn (M, M \ K; Z) such that x (K ) = x for all x K.
An immediate corollary of the above theorem is the existence of the fundamental class of
compact oriented manifolds. More precisely, by taking K = M in Theorem 45, we get the
following:
Corollary 46. If M is a compact oriented n-manifold, there exists a unique M Hn (M ; Z)
so that x (M ) = x for all x M .
Definition 21. The homology class [M ] := M of Corollary 46 is called the fundamental
class of M .
The proof of Theorem 45 uses the following:
Lemma 47. If K is a compact subset of an oriented n-manifold M , we have:
(i) Hi (M, M \ K) = 0 if i > n.
(ii) a Hn (M, M \ K) is equal to 0 if and only if x (a) = 0 for all x K.
Before proving the above lemma, let us finish the proof of Theorem 45.
Proof. (of Theorem 45)
For the uniqueness part, if 1K and 2K are as in the statement of the theorem, then for all
x K we have x (1K 2K ) = x x = 0. Then by using Lemma 47(ii), we get that
1K 2K = 0, or 1K = 2K .
We prove the existence part in several steps:
Step I: If K is contained in an euclidean neighborhood Rn of a point y M , then for all
x K, the composition
x

K
Hn (M, M \ K) Hn (M, M \ {x})
Hn (M, M \ Rn )

(4.6)

is the isomorphism x K , and therefore x is onto.

Step II: If the theorem holds for compact subsets K1 and K2 and for their intersection
K1 K2 , we show that it holds for their union K = K1 K2 . Indeed, the Mayer-Vietoris
sequence for the open cover
M \ (K1 K2 ) = (M \ K1 ) (M \ K2 ),
with intersection
M \ K = (M \ K1 ) (M \ K2 )
gives the long exact sequence:

0 Hn (M, M \K)
Hn (M, M \K1 )Hn (M, M \K2 )
Hn (M, M \(K1 K2 )) . . . (4.7)
75

where (a) = K1 (a) K2 (a) and (b c) = K1 K2 (b) K1 K2 (c). By our assumption,

there exist unique K1 Hn (M, M \ K1 ) and K2 Hn (M, M \ K2 ) restricting to local
orientations at points x K1 and resp. x K2 , hence
x K1 K2 (Ki ) = x (Ki ) = x

(4.8)

for all x K1 K2 and i = 1, 2. Then we have

x (K1 K2 (K1 ) K1 K2 (K2 )) = x x = 0

(4.9)

for all x K1 K2 . So by Lemma 47 we get that

(K1 K2 ) = K1 K2 (K1 ) K1 K2 (K2 ) = 0,

(4.10)

i.e., K1 K2 ker = Image . Since is injective, there exists a unique

K Hn (M, M \ K)
such that (K ) = K1 K2 . By the uniqueness part, we also have that K restricts to
local orientations at points x K.
Step III: For an arbitrary compact K, we write K as a finite union K = K1 K2 . . . Kr
with each Ki as in Step I. Then the claim follows by induction on r by using Step II.
Let us now get back to proving Lemma 47:
Proof. (of Lemma 47)
The proof is done in several steps, as indicated below.
Step I: Assume that M = Rn and K is a convex compact subset. Let B be a large ball in
Rn with K B, and let S = B be the bounding sphere. Then for all x K, both M \ K
and M \ {x} deformation retract to S. So we have:
Hi (M, M \ K)
= Hi (M, M \ {x})

= Hi (Rn , S n1 )

e i1 (S n1 )
=H
(
Z for i = n
=
0 otherwise.

(4.11)

Step II: We next show that if the Lemma holds for compact sets K1 , K2 and for their
intersection K1 K2 , then it holds for K := K1 K2 . Indeed, we have the Mayer-Vietoris
sequence

Hi+1 (M, M \ (K1 K2 )) Hi (M, M \ K)

Hi (M, M \ K1 ) Hi (M, M \ K2 )

Hi (M, M \ (K1 K2 ))
(4.12)
76

If i > n, we have by our assumption that Hi+1 (M, M \ (K1 K2 )) = 0, Hi (M, M \ K1 ) = 0

and Hi (M, M \ K2 ) = 0. Therefore, Hi (M, M \ K) = 0.
If i = n, the Mayer-Vietoris sequence takes the form

0 Hn (M, M \ K)
Hn (M, M \ K1 ) Hn (M, M \ K2 )
Hn (M, M \ (K1 K2 )) . . .
(4.13)
with injective. So for a Hn (M, M \ K), we have the following sequence of equivalences:
a = 0 0 = (a) = K1 (a) K2 (a)
K1 (a) = 0 and K2 (a) = 0
x K1 (a) = 0 x K1 , and y K2 (a) = 0 y K2
(since, by assumption, the lemma holds for K1 and K2 )
x (a) = 0, x K1 K2 .

(4.14)

Step III: If M = Rn and K = K1 K2 Kr with each Ki convex and compact (which

also implies that K1 K2 is convex and compact), then the lemma holds for K by Step I
and Step II.
Step IV: Assume that M = Rn and K is an arbitrary compact subset in Rn . Choose
a compact neighborhood N of K in Rn . Then for any a Hi (M, M \ K) there exists
a0 Hi (M, M \ N ) such that K (a0 ) = a. Indeed, if is a cycle representative of a, we
have that Ci (Rn ) and Ci (Rn \ K). So K = . Choose N small enough so
that N = . Next, we cover K by a union of closed balls Bi such that Bi N and
Bi K 6= . Then K factors as
Hi (Rn , Rn \ N )

> Hi (Rn , Rn \ K)
>

K
>
Hi (Rn , Rn \ i Bi )

i Bi

If i > n, then Hi (Rn , Rn \ i Bi ) = 0 by Step III. So for any a Hi (Rn , Rn \ K), we have
that
a = K (a0 ) = K (i Bi (a0 )) = 0.
If i = n, then x (a) = 0 for all x K implies by a deformation retract argument that
x (a) = 0 for all x i Bi . By using Step III, we then get that i Bi (a0 ) = 0. Hence we have
a = K (i Bi (a0 )) = 0.
Step V: If K is contained in some euclidean neighborhood in (arbitrary) M , we have by
excision
Hi (M, M \ K)
(4.15)
= Hi (Rn , Rn \ K).
So the Lemma holds for K by Step IV.

77

Step VI: Finally, note that any compact subset K of M can be written as a union K =
K1 K2 . . . Kr with each Ki as in Step V. Then the Lemma follows by using Step V,
Step II and induction.

Exercises
1. Show that every covering space of an orientable manifold is an orientable manifold.
2. Given a covering space action of a group G on an orientable manifold M by orientationpreserving homeomorphisms, show that M/G is also orientable.
3. For a map f : M N between connected closed orientable n-manifolds with fundamental
classes [M ] and [N ], the degree of f is defined to be the integer d such that f ([M ]) = d[N ],
so the sign of the degree depends on the choice of fundamental classes. Show that for any
connected closed orientable n-manifold M there is a degree 1 map M S n .
4. Show that a p-sheeted covering space projection M N has degree p, when M and N
are connected closed orientable manifolds.
5. Given two disjoint connected n-manifolds M1 and M2 , a connected n-manifold M1 #M2 ,
their connected sum, can be constructed by deleting the interiors of closed n-balls B1
M1 and B2 M2 and identifying the resulting boundary spheres B1 and B2 via some
homeomorphism between them. (Assume that each Bi embeds nicely in a larger ball in Mi .)
(a) Show that if M1 and M2 are closed then there are isomorphisms
Hi (M1 #M2 ; Z) ' Hi (M1 ; Z) Hi (M2 ; Z), for 0 < i < n,
with one exception: If both M1 and M2 are non-orientable, then Hn1 (M1 #M2 ; Z) is
obtained from Hn1 (M1 ; Z) Hn1 (M2 ; Z) by replacing one of the two Z2 -summands
by a Z-summand.
(b) Show that (M1 #M2 ) = (M1 ) + (M2 ) (S n ) if M1 and M2 are closed.

78

4.3

Cohomolgy with Compact Support

Let X be a topological space and define compactly supported i-cochains on X by:

[
Cci (X) :=
C i (X, X \ K) C i (X).

(4.16)

K compact in X

Equivalently,
Cci (X) = { : Ci (X) Z | compact K X s.t. = 0 on chains in X \ K }.

(4.17)

Define a coboundary operator

() := (),
and note that if Cci (X) vanishes on chains in X \ K then is also zero on all chains
in X \ K , and so Cci+1 (X). Therefore we get a cochain (sub)complex (Cc (X), ).
Definition 22. Hci (X) := H i (Cc (X)) is called the cohomology of X with compact support.
In what follows, we give an alternative characterization of the cohomology with compact
support, which is more useful for calculations. We begin by recalling the notion of direct
limit of groups.
Definition 23. Let G be abelian groups indexed by some directed set I, i.e., I has a partial
order and for any , I, there exists I such that and . Suppose also
that for each pair there is a homomorphism f : G G such that f = idG and
f = f f . Consider the set
q G /
where the equivalence relation is defined as: if x G , x0 G0 , then x x0 if f (x) =
f0 (x0 ) with , 0 . Any two equivalence classes [x] and [x0 ] have representatives lying
in the same G , with , 0 , so we can define
[x] + [x0 ] = [f (x) + f0 (x0 )].
This is a well-defined binary operation, and it gives an abelian group structure on the set
q G / . The direct limit of the groups G is then the group defined as:
lim G := q G / .

(4.18)

Remark 11. If J I so that I, J with , then lim G = lim G . In

particular, if J = {} (i.e, I contains a maximal element), then lim G = G .

We can now prove the following result:

79

Proposition 48. There is an isomorphism

Hci (X)
H i (X, X \ K)
= lim

(4.19)

KI

where I := {K X| K compact}.
Proof. First note that I is a directed set since it is partially ordered by inclusion, and the
union of two compact sets is also compact. Moreover, if K L are compact subsets of X,
then there is a homomorphism fKL : H i (X, X \ K) H i (X, X \ L) induced by inclusion.
Hence the direct limit group limKI H i (X, X \ K) is well-defined.

Each element of limKI H i (X, X \ K) is represented by some cocyle C i (X, X \ K)

for some compact subset K of X. Regarding as an i-cochain with compact support, its
cohomology class yields an element [] Hci (X). Moreover, such a cocycle C i (X, X \K)
is the zero element in lim H i (X, X \ K) iff = for some C i (X, X \ L) with L K,

and so [] = 0 in Hci (X).

Remark 12. If X is compact, then Hci (X) = H i (X), for all i 0, since in this case there
is a unique maximal compact set K X, namely X itself.
Example 41. Let us compute the cohomology with compact support of Rn . By the above
proposition,
Hci (Rn ) = lim H i (Rn , Rn \ K).

Note that it suffices to let K range over closed balls Bk of integer radius k centered at the
origin since each compact K Rn is contained in such a ball. So we have that
lim H i (Rn , Rn \ K) = lim H i (Rn , Rn \ Bk )

kZ0

Moreover, we have isomorphisms

H n (Rn , Rn \ Bk )
= H n (Rn , Rn \ Bk+1 )
induced by inclusion, since for all k:
(
Z if i = n
H i (Rn , Rn \ Bk )
= H i (Rn , Rn \ pt)
=
0 otherwise.
Altogether,
(
Z if i = n
Hci (Rn )
H i (Rn , Rn \ K) = lim H i (Rn , Rn \ Bk ) =
= lim

0 otherwise.
kZ0
Remark 13. It follows from the previous example that the cohomology with compact support Hc () is not a homotopy invariant.
80

b = X x be the one point compactification of X. Then

Remark 14. Let X
b x)
e i (X).

Hci (X) = H i (X,

=H

(4.20)

e i (S n ). This follows from the following general fact. If U is an

For example, Hci (Rn )
= H
open subset of a topological space V , with closed complement Z := V \ U , then there exists
a long exact sequence for the cohomology with compact support
Hci (U ) Hci (V ) Hci (Z) Hci+1 (U )
b = X x, we get a long exact sequence
If we apply this fact to the case X
b Hci (
x)
Hci (X) Hci (X)
b and x are compact, this yields that H i (X) = H i (X,
b x)
e i (X),
b as claimed.
Since X
=H
c

81

4.4

Cap Product and the Poincar Duality Map

Definition 24. We define the cap product operation

_

C i (X) Cn (X)
Cni (X)

(4.21)

as follows: for b C i (X) and Cn (X), b _ Cni (X) is defined by

a(b _ ) := (a ^ b)

(4.22)

where a C ni (X).
Remark 15. It is not hard to see that if : n X is an n-simplex and b C i (X), then
b _ = b(|[ni , ,n ] ) |[0 , ,ni ] .
|
{z
} | {z }
Z

(4.23)

Cni (X)

The reader is encouraged to show that these two notions of cap product are equivalent.
The following result is a direct consequence of the definition:
Lemma 49. For any b C i (X) and Cn (X), we have:
(b _ ) = b _ + (1)i b _ .

(4.24)

As a consequence, the cap product descends to (co)homology:

Corollary 50. There is an induced cap product operation
_

H i (X) Hn (X)
Hni (X).

(4.25)

Remark 16. A relative cap product

_

H i (X, A) Hn (X, A)
Hni (X)

(4.26)

can be defined as follows. First note that the restriction

_

C i (X, A) Cn (X)
Cni (X)
of absolute cap product (4.21) vanishes on C i (X, A) Cn (A), so it induces:
_

C i (X, A) Cn (X, A)
Cni (X).
Since (4.24) still holds in this relative setting, we get a relative cap product operation:
_

H i (X, A) Hn (X, A)
Hni (X).
The following result states that the cap product _ is functorial. Its proof is a direct
consequence of the definition of cap products and is left as an exercise:
82

Lemma 51. If f : X Y is a continuous map, then

_ f = f ((f ) _ )

(4.27)

for all H i (Y ) and Hn (X). This fact is illustrated in the following diagram:
_

H i (X) Hn (X) Hni (X)

f y
f y
f
_

H i (Y ) Hn (Y ) Hni (Y )
Let us next move towards the definition of the Poincar duality isomorphism. Let M be a
n-dimensional orientable connected manifold (not necessarily compact), and let K L M
where K, L are compact subsets. Consider the diagram:
_

H i (M, M \ L) Hn (M, M \ L) Hni (M )

x

i y
i

_

H i (M, M \ K) Hn (M, M \ K) Hni (M )

By the functoriality of the cap product, we have for any H i (M, M \ K) that:
(i ) _ L = _ i (L ),

(4.28)

where K and L denote the orientation classes of Theorem 45. Moreover, the following
identification holds:
Lemma 52. For compact subsets K L of M , we have:
i (L ) = K .

(4.29)

Proof. The claim follows from the commutativity of the following diagram and the uniqueness
of K in Hn (M, M \ K) which restricts to local orientations x , x K.
K Hn (M, M \ K)

Hn (M, M \ x)

i
L Hn (M, M \ L)
Therefore, we have from (4.28) and (4.29) that:
(i ) _ L = _ i (L ) = _ K ,

(4.30)

for all H i (M, M \ K). Let us now recall from Proposition 48 that we have an isomorphism:
Hci (M )
H i (M, M \ K),
(4.31)
= lim

83

where the direct limit on the right-hand side is taken over all compact subsets K of M . We
can now define the Poincar duality map
_

Hci (M ) Hni (M )

(4.32)

as follows: its value on Hci (M ) is defined as K _ K , where K H i (M, M \ K)

is a representative of and K Hn (M, M \ K) is the orientation class defined by K (cf.
Theorem 45). Note that the Poincar duality map (4.32) is well-defined (i.e., independent of
the choice of the representative K ) by the commutativity of the following diagram (which
follows from the identity (4.30)):
i

H i (M, M \ K)
_ K

H i (M, M \ L)
_ L

Hni (M )
We have now all the necessary ingredients to formulate the main theorem of this chapter:
Theorem 53. (Poincar Duality)
If M is an n-dimensional oriented connected manifold, then the Poincar duality map:
_

Hci (M ) Hni (M )
is an isomorphism for all i.
An an immediate corollary, we get the following:
Corollary 54. If M is an n-dimensional closed oriented connected manifold, then the map
_

H i (M ) Hni (M )
defined by the cap product with the fundamental class of M , that is, 7 [M ], is an
isomorphism for all i.

84

4.5

The Poincar Duality Theorem

This section is devoted to proving The Poincar Duality Theorem, which we recall below for
the convenience of the reader.
Theorem 55. (Poincar Duality)
If M is an n-dimensional oriented connected manifold, then the Poincar duality map:
_

Hci (M ) Hni (M )
is an isomorphism for all i.
Proof. Recall that on an element Hci (M )
=

lim

H i (M, M K) the Poincar duality

KX
Kcompact

map takes the value K _ K , where K H i (M, M K) is a representative of , and K

is the generator of Hn (M, M K).
The proof of the theorem will be divided into several steps. We first show that the statement holds locally, then we glue the local isomorphisms by a Meyer-Vietoris argument.
Step I: We first show that the theorem holds for M = Rn .
Let Bk denote the closed ball of integer radius k in Rn . Then
(
Z if i = n
i
n
n

Hci (Rn )
H
(R
,
R

B
)
= lim
=
k

0 otherwise
Bk
and

(
Z if i = n
Hni (Rn ) '
0 otherwise.

The Universal Coefficient Theorem yields that

H n (Rn , Rn Bk ) ' Hom(Hn (Rn , Rn Bk ); Z).
So H n (Rn , Rn Bk ) is generated by some class ak so that ak (Bk ) = 1 Z. Let 1
H 0 (Rn ) = Z be the generator. Then:
1 = ak (Bk ) = (1 ^ ak )(Bk ) = 1(ak _ Bk )
Hence ak _ Bk is a generator of H0 (Rn ). In particular, the map
_ Bk : H n (Rn , Rn Bk ) H0 (Rn )
is an isomorphism. Taking the direct limit over the Bk s, we get an isomorphism

Hcn (Rn ) H0 (Rn ).

85

Also, both groups are trivial for i 6= n, so the claim follows.

Step II: Assuming the theorem holds for opens U, V M and for their intersection U V ,
we show that it holds for the union U V .
For this purpose, we construct a commutative diagram
Hci (U V )
Hci (U ) Hci (V )
Hci (U V ) Hci+1 (U V )

Hni (U V ) Hni (U ) Hni (V ) Hni (U V ) Hni1 (U V )

(4.33)

Once the diagram is constructed, the claim follows by the 5-lemma.

The bottom row in (4.33) is just the Mayer-Vietoris homology sequence. The top row of
the above diagram can be constructed as follows. For compact subsets K U and L V ,
consider the cohomology Mayer-Vietoris sequence for the pairs (M, M \ K) and (M, M \ L):
H i (M, M \ (K L)) H i (M, M \ K) H i (M, M \ L) H i (M, M \ (K L))
By excision, we get a long exact sequence:
H i (U V, U V \K L) H i (U, U \K)H i (V, V \L) H i (U V, U V \K L)
Taking direct limits over K U and L V , we get the top long exact sequence in (4.33):
Hci (U V ) Hci (U ) Hci (V ) Hci (U V )
The commutativity follows by using the definition of the Poincar duality map.
Step III: Assume M is a union of nested open subsets U so that the theorem holds for each
U . We show that the theorem holds form M .
First note that any compact subset in M (in particular, the support of a singular (co)chain)
is contained in some U . Then we claim that the following identifications hold:
Hi (M ) = lim
Hi (U )

(4.34)

Hci (U ).
Hci (M ) = lim

(4.35)

and

This claim and Poincar duality for each U imply the Poincar duality isomorphism for M .
In order to prove the claim, we note that the inclusions i : U , M induce homomorphisms
i : Hi (U ) Hi (M ) so that for U , U the following diagram commutes:
i

Hi (U )
&

Hi (M )
%

Hi (U )

86

We therefore get a well-defined map

f : lim
Hi (U ) Hi (M ).

We next show that f is an isomorphism.

f is onto: any [] Hi (M ) is represented by a cycle whose support is contained in a
compact subset of M , thus in some U . The corresponding homology class in Hi (U )
maps onto [].
f is one-to-one: if = , for Ci+1 (M ), then is a cycle in some U , but not
necessarily a boundary in U . On the other hand, is contained in some larger U , so
can be regarded as a boundary in U . Therefore, [] = 0 Hi (U ), hence it represents
the zero class in lim
Hi (U ).

So (4.34) follows. The identification in (4.35) is obtained similarly.

Step IV: We next show that the theorem holds when M is an open subset of Rn .
If M is convex, then M is homeomorphic to Rn , so the theorem holds by Step I. If M is not
convex, then M = k Vk , with each Vk open and convex in Rn . By induction and Step II,
the theorem holds for the sets Uk = V1 Vk . Note that {Uk }k forms a nested cover of
opens for M , hence the theorem follows by Step III.
Step V: Finally, we show that the Poincar duality isomorphism holds for an arbitrary M .
We first cover M by open sets V , each of which is homeomorphic to Rn . We next choose a
well ordering < of the index set, which exists by Zorns lemma. Then the sets
U := < V .
form a nested open cover of M . So by Step III, it suffices to show that the theorem holds for
each U . But U = < V , with V
= Rn for each , and the theorem holds for each V . By
Step II and transfinite induction, the theorem holds for each U , and the claim follows.
Remark 17. By taking coefficients in any commutative ring R, we can prove the Poincar
duality isomorphism over R via the coefficient map Z R. Moreover, for R = Z/2, Poincar
duality holds even without the orientability assumption.

Exercises
1. Show that if M n is connected, non-compact manifold, then Hi (M ; Z) = 0 for i n.
2. Show that the Euler characteristic of a closed, oriented, (4n + 2)-dimensional manifold is
even.

87

3. Let M be a closed oriented manifold with fundamental class [M ]. Consider the following
cup product pairing between cohomology groups of complementary dimensions (after moding
out by the corresponding torsion subgroups):
( , ) : H i (M ; Z)/Torsion H ni (M ; Z)/Torsion Z
given by (, ) = h , [M ]i. Here h , i : H n (X; Z) Hn (X; Z) Z is the Kronecker
pairing defined in Homework #1.
(i) Show that the cup product pairing is nonsingular in the following sense: for each choice
of a Z-basis {1 , , r } of H ni (M ; Z)/Torsion, there exists a Z-basis {1 , , r }
of H i (M ; Z)/Torsion such that (i , j ) = ij . (Hint: Use the Universal Coefficient
Theorem and Poincar Duality.)
(ii) As an application, re-prove the following facts about the ring structures on the cohomology of projective spaces:
(a) H (RPn ; Z2 )
= Z2 [x]/(xn+1 ), |x| = 1,
(b) H (CPn ; Z)
= Z[y]/(y n+1 ), |y| = 2,
(c) H (HPn ; Z)
= Z[w]/(wn+1 ),

|w| = 4.

4. Let M be a closed, oriented 4n-dimensional manifold, with fundamental class [M ]. The

middle intersection pairing
( , ) : H 2n (M ; Z)/Torsion H 2n (M ; Z)/Torsion Z
given by (, ) = h , [M ]i is symmetric and nondegenerate. Let {1 , , r } be a Zbasis of H 2n (M ; Z)/Torsion, and let A = (aij ) for aij := (i , j ) Z. Then A is a symmetric
matrix with det(A) = 1, so it is diagonalizable over R. Define the signature of M to be
(M ) := (the number of positive eigenvalues) (the number of negative eigenvalues)
(a) Compute (CPn ), (S 2 S 2 ).
(b) Show that the signature (M ) is congruent mod 2 to the Euler characteristic (M ).
5. Show that if a connected manifold M is the boundary of a compact manifold, then the
Euler characteristic of M is even. Conclude that RP2n , CP2n , HP2n cannot be boundaries.
6. Show that if M 4n is a connected manifold which is the boundary of a compact oriented
(4n + 1)-dimensional manifold V , then the signature of M is zero.
7. Show that if M is a compact contractible n-manifold then M is a homology (n 1)sphere, that is, Hi (M ; Z) ' Hi (S n1 ; Z) for all i.
8. Let M be a closed, connected, orientable 4-manifold with fundamental group 1 (M )
=
Z/3 Z/3 and Euler characteristic (M ) = 5.
88

(a) Compute Hi (M, Z) for all i.

(b) Prove that M is not homotopy equivalent to any CW complex with no 3-cells.
9. Let M be a closed, connected, oriented n-manifold and let f : S n M be a continuous
map of non-zero degree, i.e., the morphism
f : Hn (S n ; Z) Hn (M ; Z)
is non-trivial. Show that M and S n have the same Q-homology.
10. Show that there is no orientation-reversing self-homotopy equivalence CP2n CP2n .

89

4.6

Immediate applications of Poincar Duality

In this section we derive several applications of the Poincar duality isomorphism of Theorem
55. (In particular, we provide answers to some of the exercises listed in the previous section.)
Proposition 56. If M n is a closed odd dimensional manifold, then (M ) = 0.
Proof. Let n = 2k + 1.
If M is oriented, then
(P.D.)

(U CT )

Hi (M )
= free(Hni (M )) Torsion(Hni1 (M )).
= H ni (M )
Therefore,
rkHi (M ) = rkHni (M ).
So:
(M ) =

2k+1
X

(1) rkHi (M ) =

i=0

k
X

(1)i + (1)ni rkHi (M ) = 0.

i=0

If M is not oriented, the Poincar duality isomorphism holds with Z/2-coefficients, and
we get:
(M ) :=

2k+1
X

()

(1) rkHi (M ; Z) =

n=0

2k+1
X

(1)i dimZ/2 Hi (M ; Z/2) = 0,

n=0

where the vanishing follows as before by Poincar duality (over Z/2). The equality () follows
from the Universal Coefficient Theorem as follows:
H i (M, Z/2) = Hom(Hi (M ), Z) Ext(Hi1 (M ), Z/2).
a Z-summand of Hi (M ; Z) contributes
Hom(Z, Z/2) = Z/2 to H i (M ; Z/2), and
Ext(Z, Z/2) = 0 to H i+1 (M ; Z/2).
a Z/m summand of Hi (M ; Z), with m odd, contributes:
Hom(Z/m, Z/2) = 0 to H i (M ; Z/2), and
Ext(Z/m, Z/2) = 0 to H i+1 (M ; Z/2).
a Z/m summand of Hi (M ; Z), with m even, contributes:
Hom(Z/m, Z/2) = Z/2 to H i (M ; Z/2), and
Ext(Z/m,
Z/2) = Z/2 to H i+1 (M ; Z/2), so these Z/2 contributions cancel out in
P
i
i
i (1) dimZ/2 H (M ; Z/2).
90

Finally, note that dimZ/2 Hi (M ; Z/2) = dimZ/2 H i (M ; Z/2), so the claim follows.
Proposition 57. If M n is a closed, oriented, connected manifold, then
Torsion(Hn1 (M )) = 0.
Proof. Indeed,
(P.D.)

(U CT )

Torsion(Hi (M )) = Torsion(H ni (M )) = Ext(Hn1i (M ), Z) = Torsion(Hn1i (M ))

Since M is connected, H0 (M ) is free, so the claim follows.
We will show later the following:
Proposition 58. If M n is a closed, connected, non-orientable manifold, then
Torsion(Hn1 (M )) = Z/2
and
H n (M ) = Z/2.
The second part of Proposition 58 follows from the Universal Coefficient Theorem and the
following consequence of Poincar duality:
Lemma 59. if M n is closed and connected then
(
Z , if M -oriented
Hn (M ) =
0 , if M -non-oriented

91

4.7

Addendum to orientations of manifolds

Before we explain the proof of Proposition 58, we need to elaborate on orientations of manifolds.
Recall that if M n is a n-manifold, a local orientation at x M is a generator x
Hn (M, M x)
= Z. We say that M is oriented if there exists a global orientation, i.e., a
continuous function x x . This means that for all x M , there is a closed ball Bx around
x so that
x
Z
= Hn (M, M Bx ) Hn (M, M y)
sends the generator Bx to the local orientation class y , for all y Bx .
f.
Proposition 60. Any manifold M (oriented or not) has an oriented double cover M
Proof. (Sketch)
Define
f := {x M, x a local orientation at x M }
M
f M by x x. Clearly, is a 2 : 1 map.
and : M
f so that it becomes a manifold and is a covering map.
We need to put a topology on M
For a ball B Rn M with a generator B Hn (M, M B), define
f| x B, x = x (B )},
U (B ) = {x M
where x denotes the natural map Hn (M, M \ B) Hn (M, M \ x). Then
1 (B) = U (B ) t U (B )
and both U (B ) and U (B ) are in bijection to B. Moreover, it can be shown that the sets
f so that is continuous. So is 2-fold
{U (B )}B form basis of opens for the topology M
f is manifold.
covering and M
f is orientable. Indeed, we have,
Moreover, M
f, M
f \ x )
Hn (M
= Hn (U (B ), U (B ) \ x )
= Hn (B, B \ x)
= Hn (M, M \ x).

(4.36)

f there exists a canonical local orientation

f, M
f \ x )
So at the point x M
fx Hn (M
=Z
corresponding to x under the above isomorphism (4.36). The consistency of such local
orientations follows by construction.
Example 42. Let us look at a few simple examples:
f = S 2.
(a) The oriented double cover of M = RP2 is M
(b) The oriented double cover of the Klein bottle K is the 2-torus T 2 .
f = T 2.
(c) If M = T 2 is the 2-torus, its oriented double cover is M
92

f has 2 components.
Proposition 61. If M is connected, then M is oriented if, and only if, M
In particular, if 1 (M ) = 0 or has no index 2 subgroup, then M is oriented.
f can have one or two components. If M
f has 2 comProof. The oriented double cover M
ponents, each is oriented and homeomorphic to M , so M is oriented. Conversely, if M is
f.
oriented, it can have exactly 2 orientations at each point, each defining a sheet of M
f can be embedded in a larger covering space MZ of M as
The oriented double cover M
follows. Let
MZ = {x | x M, x Hn (M, M \ x) = Z}.
We then have the Z-fold cover map
Z : MZ M
defined by x x. A basis of opens {U (B)} for MZ can be defined by the following recipe:
for an open ball B Rn M , set
U (B) = {x | x = x (B ) for B Hn (M, M \ B)
= Z)}

with x : Hn (M, M \ B) Hn (M, M \ x) induced by inclusion as before. For any k Z, we

then get a subcover Mk MZ by selecting kx in the fibre above x. So
[
Mk
MZ =
k0

, for any integer k.

with M0
= M, Mk
= Mk , and Mk
=M
Definition 25. A section of Z : MZ M is a continuous map : M MZ defined by
x 7 x Hn (M, M \ x) = Z. An orientation of M is a section of Z assigning x to each
x M.
One can generalize the definition of orientability by replacing Z any commutative ring R
with unit. Note that by the universal coefficient theorem for homology, we have:
Hn (M, M \ x; R)
= Hn (M, M \ x) R
=ZR
= R.
The covering MZ can be generalized to:
MR = {x | x Hn (M, M \ x; R)
= R}.
The corresponding covering map R : MR M is defined by x 7 x (so the fibre over
x M is R). Each r R determines a subcovering Mr by selecting the points x r
Hn (M, M \ x; R) in each fibre. If r is an element of order 2 in R, then Mr is a copy of M .
(Indeed, x r = x r = x r.) Otherwise, Mr is homeomorphic to the oriented
f. We have
double cover M
[
MR =
Mr ,
rR

with all Mr being disjoint except for Mr = Mr , and Mr = M if 2r = 0.

93

Definition 26. An R-orientation of an n-dimensional manifold M is a section of MR assigning to each x M a generator u of Hn (M, M \ x; R)
= R.
Remark 18. Note that a generator of R is an element u so that Ru = R. Since R has a
unit, this is equivalent to saying that u is invertible in R
Remark 19. An orientable manifold is R-orientable, for all commutative rings R with unit.
A non-orientable manifold is R-orientable iff R contains a unit of order 2. Thus every
manifold is Z/2-orientable.
We are now ready to prove the following result:
Proposition 62. Let M be a closed connected n-manifold. Then:
(a) if M is orientable, then Hn (M ; R) Hn (M, M \ x; R)
= R is an isomorphism for any
x M.
(b) if M is not orientable, then Hn (M ; R) Hn (M, M \ x; R)
= R is one-to-one, with
image the set of elements of order 2 in R.
(c) Hi (M ; R) = 0, for all i > n.
The proof of Proposition 62 is based on the Theorem 45 and Lemma 47, which we
formulated here with R-coefficients:
Lemma 63. Let M be a connected n-manifold and K a compact subset of M . Then:
(a) if M is R-oriented, then there exists a unique K Hn (M, M \ K; R) such that
x (K ) = x Hn (M, M \ x; R), for all x M .
(b) Hi (M, M \ K; R) = 0 for i > n, and a class K Hn (M, M \ K; R) is zero iff
x (K ) = 0 for any x K.
To deduce Proposition 62, choose K = M , and let R (M ) be the set of sections of MR .
With respect to the addition of functions and multiplication by scalars in R, R (M ) becomes
an R-module. Moreover, there exists a homomorphism
Hn (M ; R) R (M )
defined by
(x 7 x ),
where x is the image of under the map Hn (M ; R) Hn (M, M \ {x}; R). The above
lemma asserts that this is an isomorphism. Note that if M is connected, then each section
in R (M ) is determined by its value at one point x M .
Next, we translate the statements about Hn (M ; R) in Proposition 62 into statements about
R (M ).
For the oriented case: Hn (M ; R)
= R (M ) Hn (M, M \ x; R) is an isomorphism, defined
94

by 7 (x 7 x ) 7 x for a given x.
For the non-oriented case: Hn (M ; R)
= R (M ) Hn (M, M \ x; R) is a homomorphism,
and units x of order 2 correspond to R-orientations, i.e., sections x 7 x , with 2x = 0.
So the proposition follows now from the structure of MR .
Corollary 64. If M is orientable, then Hn (M ; Z)
= Z. If M is non-orientable, then
Hn (M ; Z) = 0. In either case, Hn (M ; Z/2) = Z/2.
We can now prove the following:
Corollary 65. Let M be a closed and connected n-manifold. If M is oriented, then
Torsion(Hn1 (M )) = 0.
Otherwise,
Torsion(Hn1 (M )) = Z/2.
Proof. By the universal coefficient theorem for homology, we have:
Hn (M ; Z/p) = Hn (M ; Z) Z/p Tor(Hn1 (M ; Z), Z/p)
= Hn (M ; Z) Z/p Torsion(Hn1 (M ; Z)) Z/p.
In the orientable case, if Hn1 (M ) contained torsion, then for some prime p, the group
Hn (M ; Z/p) = Z/p would be larger than the Z/p coming from the first summand (here we
use that Hn (M ) = Z), which is impossible. This means Torsion(Hn1 (M )) = 0.
In the non-orientable case, we have by Proposition 62 that Hn (M ; Z/m) is either Z/2 or 0
depends on whether m is even or odd. (Indeed, in this case the map Hn (M ; Z/m) Z/m is
injective with image the elements of order 2 in Z/m. So, if m is odd, there are no elements
of order 2 in Z/m, while if m = 2k is even, then k is the only element of order 2 in Z/m.)
Since in this case we have Hn (M ; Z) = 0, this forces the torsion subgroup of Hn1 (M ) to be
Z/2.
Remark 20. By the universal coefficient theorem for cohomology, we have:
H n (M ) = Free(Hn (M )) Torsion(Hn1 (M )).
By using the result of the previous corollary, we get that if M is oriented, then H n (M ) = Z.
Otherwise, H n (M ) = Z/2.

95

4.8

Cup product and Poincar Duality

Let R a commutative ring and fix C l (M ; R), C k (M ; R) and Ck+l (M ; R). Then
_ Cl is defined by
( _ ) = ( ^ )() R.
(4.37)
Alternatively, if is a (k + l)-simplex, then
_ = (|[vl ,vl+1 ,...,vk+l ] ) |[v0 ,v1 ,...,vl ] .

(4.38)

( _ ) = (|[vl ,vl+1 ,...,vk+l ] ) (|[v0 ,v1 ,...,vl ] ) = ( ^ )().

(4.39)

Indeed,
This means that ^ : C l (M ) C k+l (M ) is dual to _ : Ck+l (M ) Cl (M ).
Passing to (co)homology, we get the following commutative diagram:
h

H l (M ; R) HomR (Hl (M ; R), R)

^ y
(_) y
h

H k+l (M ; R) HomR (Hk+l (M ; R), R)

In particular, if h is an isomorphism (e.g., R is a field, or we work over Z but H is torsionfree), then ^ and _ determine each other.
Definition 27. Let M be a closed connected R-oriented manifold. Then the cup product
pairing
_[M ]

H k (M ; R) H nk (M ; R) H n (M ; R) H0 (M ; R) = R

(4.40)

is defined by
(, ) 7 ( ^ ) 7 ( ^ ) _ [M ].
Definition 28. Let A and B be R-modules. A pairing : A B R is non-singular if
f : A Hom(B, R) is an isomorphism, with f defined by f (a)(b) = (a, b), and g : B
Hom(A, R) is an isomorphism, with g(b)(a) = (a, b).
We then have the following:
Proposition 66. The cup product pairing is non-singular if R is a field, or if R = Z and
torsion is factored out.
Proof. Consider the composition
(P.D.)

f : H k (M ; R) Hom(Hk (M ), R) Hom(H nk (M ), R),

where (P.D.) denotes the dual of the Poincar duality isomorphism. Under our assumptions
on R, h is isomorphism. Moreover, by Poincar Duality, (P D) is also an isomorphism, hence

96

f is an isomorphism. For H k (M ; R) and H nk (M ; R), we have:

f ()() =
=
=
=
=

((P.D.) h())()
h()(P.D.())
h()( _ [M ])
( _ [M ])
( ^ )[M ].

We obtained a similar isomorphism by interchanging k with n k, so the claim follows.

Corollary 67. Let M be a closed connected Z-oriented n-manifold. Then for any
H k (M ) a generator of a Z-summand, there exists H nk (M ) such that ^ generates
H n (M )
= Z.
Proof. By hypothesis, there exists a homomorphism (i.e., the projection to some Z-summand)
: H k (M ) Z
such that () = 1. By the non-singularity of the cup product pairing, is realized by
taking the cup product with some H nk (M ) and evaluating on the fundamental class
[M ]. We therefore get
1 = () = ( ^ )[M ].
This means ^ is the generator of H n (M ).
Corollary 68. H (CPn ; Z)
= Z[]/(n+1 ), with deg() = 2.
Proof. Let be the generator of H 2 (CPn ) = Z. By induction, we can assume that n1
generates H 2n2 (CPn ) = Z. Using the previous corollary, there exists H 2 (CPn ) so that
n1 generates H 2n (CPn ) = Z. Note that since is the generator of H 2 (CPn ) = Z, it
follows that = m, for some m Z. This means that n1 = mn generates Z. Thus
m = 1, whence n generates H 2n (CPn ).
We can now ask the following:
Question 69. Does there exist a 2n-dimensional closed manifold whose cohomology is additively isomorphic to that of CPn , but with a different cup product structure?
If n = 2, the answer is No. Indeed, H (CP2 ; Z) = Z[]/(3 ), with deg() = 2. If there is
such manifold M , then also generates H 2 (M ) = H 2 (CP2 ) = Z, so there exists H 2 (M )
such that ^ generates H 4 (M ) = Z. So, = m, for some m Z. Hence ^ = m2
generates H 4 (M ), which yields m = 1. This means that M has the same cup product
structure as CP2 .
If n 3, the answer is Yes. Indeed, as we will see in the next section, S 2 S 4 and CP3 have
isomorphic cohomology groups, but different cup product structures on their cohomology
rings.
97

Remark 21. Dualizing the cup product pairing of Proposition 66, we get the non-singular
intersection pairing
Hk (M ) Hnk (M ) Z
defined by
([], []) ]( 0 ),
where 0 is chosen so that it is homologous to but transversal to (so 0 is a finite
number of points).
Example 43. Let T be the 2-dimensional torus and S be a meridian of T . Let M be
the pinched torus T /S. Then Poincar duality fails for M. If not, let be the longitude
of M (and T ) and be the a median of M . Then Poincar duality for M would yield
([], []) ]( ) = 1. However, [] = 0. This is impossible since the intersection pairing
is non-degenerate. The reason for the failure of Poincar duality is that the pinched torus
M := T /S is not a manifold. Indeed, a neighborhood of the pinch point is a join of two
2-disks, thus it is not homeomorphic to R2 .
Another application of Poincar duality is the following:
Corollary 70. If M is a closed oriented manifold of dimension m = 4n + 2, then (M ) is
even.
Proof. By the definition of the Euler characteristic, (M ), we have
(M ) =

4n+2
X

(1)i rk (Hi (M )).

i=0

By Poincar duality, we obtain

rk(Hi (M )) = rk(Hmi (M )).
Therefore,
(M ) rk(H2n+1 (M )) (mod 2).
Let us now consider the following cup product pairing
^

_[M ]

H 2n+1 (M ) H 2n+1 (M ) H 4n+2 (M ) Z

defined by
(, ) 7 ( ^ ) 7 ( ) _ [M ].
By Poincar Duality, after moding out by torsion, this pairing is non-singular. As a result,
the matrix A of the cup product pairing is non-singular and anti-symmetric. By linear
algebra, A is similar to a matrix with diagonal blocks


0 1
1 0
Therefore,
rk(H 2n+1 (M )) = rk(A),
which is clearly even.
98

4.9

Manifolds with boundary: Poincar duality and applications

In this section we discuss the Poincar duality for manifolds with boundary. The proofs are
routine adaptation of those for closed manifolds.
Definition 29. A Hausdorff topological space M is an n-manifold with boundary if any
point x M has a neighborhood Ux homeomorphic to either Rn or Rn+ := {(x1 , , xn )
Rn | xn 0}. In particular,
(a) if Ux
= Rn , then Hn (M, M \ x) = Hn (Ux , Ux \ x)
= Z.
(b) if Ux
= Rn+ , then
Hn (M, M \ x) = Hn (Ux , Ux \ x)
= Hn (Rn+ , Rn+ {0})
= 0.
The boundary of M is defined to be
M := {x M | Hn (M, M \ x) = 0}.
Example 44. (Dn ) = S n1 , (Rn+ ) = Rn1 .
Remark 22. The boundary set M is a manifold of dimension n 1 with no boundary.
Definition 30. We say that a manifold with boundary (M, M ) is orientable if M \ M is
orientable as a manifold with no boundary.
We have the following:
Proposition 71. If (M, M ) is a compact, orientable n-manifold with boundary, then there
exists a unique class M Hn (M, M ) inducing local orientations x Hn (M, M \ x) at all
points x M \ M .
Remark 23. In the long exact sequence for the pair (M, M ), we have

Hn (M, M ) Hn1 (M )
[M ] = M 7 [M ]
if M is oriented.
Theorem 72 (Poincar Duality). If (M, M ) is a connected, oriented n-manifold with
boundary, then there are isomorphisms
_

Hci (M ) M
Hni (M, M )
=

and

Hci (M, M ) M
Hni (M )
=

(4.41)
(4.42)

where Hci (M, M ) := lim Kcompact H i (M, (M \ K) M ) is the cohomology with compact
KM \M
support.
99

Let us now describe some applications of Poincar duality for manifolds with boundary.
Proposition 73. If M n = V n+1 is a connected manifold with V compact, then the Euler
characteristic (M ) is even.
An immediate consequence of Proposition 73 is the following:
Corollary 74. RP2n , CP2n , HP2n cannot be boundaries of compact manifolds.
In order to prove Proposition 73, we need the following result:
Proposition 75. Assume V 2n+1 is an oriented, (2n + 1)-dimensional compact manifold with
connected boundary V = M 2n . If R is a field (e.g., Z/2Z if M is non-orientable), then
dimR H n (M ; R) is even.
Proof of Proposition 75. Consider the long exact sequence for the pair (V, M ):
i

H n (V ; R) > H n (M ; R) > H n+1 (V, M ; R)

= _[M ]

Hn (M ; R)

= _[V ]

> Hn (V ; R)

where i , i are induced by the inclusion i : M = V , V . By exactness, Image i

=
P.D.
ker
= ker i , so
dim(Image i ) = dim(ker i ) = dim Hn (M ; R) dim(Image i ).
Since i , i are Hom-dual, we have that dim(Image i ) = dim(Image i ). Altogether,
dim H n (M ; R) = dim Hn (M ; R) = 2 dim(Image i )
is even.
Proof of Proposition 73. If n = dim M is odd, then Proposition 56 yields that (M ) = 0,
thus even. If n = 2m is even, then we work with Z/2Z-coefficients and get:
(M ) =
(1)

2m
X

(1)i dimZ/2 Hi (M ; Z/2)

i=0
m1
X

(1)i dimZ/2 Hi (M ; Z/2) + (1)m dimZ/2 Hm (M ; Z/2)

= 2

i=0

dimZ/2 Hm (M ; Z/2) (mod 2)

(2)

0 (mod 2),

where equation (1) follows by Poincar Duality, and congruence (2) is by Proposition 75.
100

Remark 24. The proof of Proposition 75 also yields the following:

(a) Image i H n (M 2n ) is self-annihilating with respect to cup product ^, i.e., if ,
Image i , then ^ = 0.
(b) dim(Image i ) = 12 dim H n (M 2n ).
Proof. For any = i (), = i () with , H 2n (V ), we have
( ^ ) = (i () ^ i ()) = i ( ^ ) = 0


Hence, ^ ker : H 4n (M ) H 4n+1 (V, M )
= 0, where the last isomorphism follows
by the following commutative diagram

H 4n (M ) > H 4n+1 (V, M )

= P.D.

H0 (M )

= P.D.

> H0 (V )

with the bottom arrow an injection.

4.9.1

Signature

Definition 31. Let M be a closed oriented manifold. If dim M = 4k, the signature of M ,
(M ), is defined to be the signature of the symmetric non-degenerate cup product pairing
H 2k (M ; R) H 2k (M ; R) R
(, ) 7 ( ^ )[M ]
Otherwise, if dim M is not divisible by 4, we let (M ) = 0.
Remark 25. Recall that a symmetric non-degenerate bilinear pairing has only real eigenvalues, and its signature is defined by subtracting the number of negative eigenvalues from
the number of positive eigenvalues.


0 1
2
2
Example 45. (S S ) =
= 0, (CP2n ) = 1, (CP2 #CP2 ) = 2.
1 0
The signature is a cobordism invariant, i.e. if W = M t N , then (M ) = (N ).
Here N denotes the manifold N but with the opposite orientation.

101

Here we prove the following special case of this fact:

Theorem 76. If in the above notations M 4k = V 4k+1 is connected with V compact and
orientable, then (M ) = 0.
Proof. Let A = H 2k (M ) and we work with R-coefficients. The cup product yields a nondegenerate and symmetric pairing : A A R.
Let A+ be the subspace on which the pairing is positive-definite, and A the subspace
on which the pairing is negative-definite. Let r = dim A+ , 2l = dim A (which is even by
Proposition 75). Then, dim A = 2l r since the pairing is non-degenerate, and (M ) =
r (2l r) = 2r 2l. In order to prove that (M ) = 0, it suffices to show that r = l.
Let B A be the self-annihilating l-dimensional subspace given by Proposition 73. Then
A+ B = {0} and A B = {0}. Hence,
dim A+ + dim B dim A = 2l, i.e., r + l 2l i.e., r l
dim A + dim B dim A = 2l, i.e., 2l r + l 2l i.e., r l
In conclusion, r = l and (M ) = 0.

4.9.2

Connected Sums

Definition 32. Let M n , N n be closed, connected, oriented n-manifolds. Their connected

sum

is defined to be
M #N := (M \ D1n ) f (N \ D2n )
where f : D1n = S n1 D2n = S n1 is an orientation-reversing homeomorphism.
Remark 26. The connected sum M #N of closed, connected, oriented n-manifolds is itself a
closed, connected, oriented n-manifold. The cohomology ring H (M #N ) is isomorphic to the
ring resulting from the direct product of H (M ) and H (N ), with the unity elements identified, and the orientation classes identified. In particular, H 0 (M #N ) = Z, H n (M #N ) = Z
and H k (M #N )
= H k (M ) H k (N ), 0 < k < n. Moreover, cup products of positive dimensional classes, one from each of the two original manifolds, are zero, i.e., ^ = 0 for any
H k (M ) and H l (N ) with k, l > 0.
102

Example 46. By the above description of cup products of a connected sum, we get:
(CP2 # CP2 ) = 0.
In fact, it can be shown that CP2 #CP2 is the boundary of a connected, oriented 5-manifold;
Example 47. The spaces S 2 S 2 and CP2 #CP2 have the same cohomology groups,
H 0 = Z, H 2 = Z Z = Z Z, H 4 = Z,
but different cohomogy rings, since ^ 6= 0 in H (S 2 S 2 ), but ^ = 0 in
H (CP2 #CP2 ).
Example 48. We have
(CP2 #CP2 ) = 2 6= 0,
so in view of Theorem 76, CP2 #CP2 cannot be the boundary of a compact, oriented 5manifold. However, CP2 #CP2 = W 5 , where W 5 is a compact non-orientable 5-manifold.
The compact manifold W can be constructed as follows:
(a) Start with (CP2 I)#(RP2 S 3 ).
(b) Run an orientation reversing path from one CP2 to the other, by traveling along an
orientation reversing path in RP2 .
(c) Enlarge the path to a tube and remove its interior. What is left is a 5-dimensional
non-orientable manifold with W = CP2 #CP2 .

103

Chapter 5
Basics of Homotopy Theory
5.1

Homotopy Groups

Definition 33. For each n 0 and X a topological space with x0 X, the n-th homotopy
group of X is defined as


n (X, x0 ) = f : (I n , I n ) (X, x0 ) /
where is the usual homotopy of maps.
Remark 27. Note that we have the following diagram of sets:
(I n , I n )

f
g

/
6

(X, x0 )

(I n /I n , I n /I n )
with (I n /I n , I n /I n ) ' (S n , s0 ). So we can also define


n (X, x0 ) = g : (S n , s0 ) (X, x0 ) / .
Remark 28. If n = 0, then 0 (X) is the set of connected components of X. Indeed, we
have I 0 = pt and I 0 = , so 0 (X) consists of homotopy classes of maps from a point into
the space X.
Now we will prove several results analogous to the case n = 1, which corresponds to the
fundamental group.
Proposition 77. If n 1, then n (X, x0 ) is a group with respect to the operation + defined
as:
(
f (2s1 , s2 , . . . , sn )
0 s1 21
(f + g)(s1 , s2 , . . . , sn ) =
g(2s1 1, s2 , . . . , sn ) 12 s1 1

104

(Note that if n = 1, this is the usual concatenation of paths/loops.)

Proof. First note that since only the first coordinate is involved in this operation, the same
argument used to prove that 1 is a group is valid here as well. Then the identity element is
the constant map taking all of I n to x0 and the inverse element is given by
f (s1 , s2 , . . . , sn ) = f (1 s1 , s2 , . . . , sn ).

Proposition 78. If n 2, then n (X, x0 ) is abelian.

Intuitively, since the + operation only involves the first coordinate, if n 2, there is enough
space to slide f past g ".
Proof. Let n 2 and let f, g n (X, x0 ). We wish to show f + g ' g + f . Consider the
following figures:

We first shrink the domains of f and g to smaller cubes inside I n and map the remaining
region to the base point x0 . Note that this is possible since both f and g map to x0 on
the boundaries, so the resulting map is continuous. Then there is enough room to slide f
past g inside I n . We then enlarge the domains of f and g back to their original size and get
g + f . So we have constructed a homotopy between f + g and g + f and hence n (X, x0 ) is
abelian.
Remark 29. If we view n (X, x0 ) as homotopy classes of maps (S n , s0 ) (X, x0 ), then we
have the following visual representation of f + g (one can see this by collapsing boundaries
in the above cube interpretation).

105

Next recall that if X is path-connected and x0 , x1 X, then there is an isomorphism

: 1 (X, x0 ) 1 (X, x1 )
where is a path from x0 to x1 , i.e., : [0, 1] X with (0) = x0 and (1) = x1 . The
isomorphism is given by
([f ]) = [ 1 f ]
for any [f ] 1 (X, x0 ).

We next show a similar fact holds for all n 1.

Proposition 79. If n 1 and X is path-connected, then there is an isomorphism :
n (X, x1 ) n (X, x0 ) given by
([f ]) = [ f ],
where is a path in X from x1 to x0 , and f is constructed by first shrinking the domain
of f to a smaller cube inside I n , and then inserting the path radially from x1 to x0 on the
boundaries of these cubes.

Proof. It is easy to check that the following properties hold:

1. (f + g) ' f + g
2. ( ) f ' ( f ), for a path from x0 to x1
106

3. cx0 f ' f , where cx0 denotes the constant path based at x0 .

4. is well-defined with respect to homotopies of or f .
Note that (1) implies that is a group homomorphism, while (2) and (3) show that
is invertible. Indeed, if (t) = (1 t), then 1 = .
So, as in the case n = 1, if the space X is path-connected, then n is independent of the
choice of base point. Further, if x0 = x1 , then (2) and (3) also imply that 1 (X, x0 ) acts on
n (X, x0 ):
1 n n
(, [f ]) 7 [ f ]
Definition 34. We say X is an abelian space if 1 acts trivially on n for all n 1.
In particular, this means 1 is abelian, since the action of 1 on 1 is by inner-automorphisms,
which must all be trivial.
We next show that n is a functor.
Proposition 80. A map : X Y induces group homomorphisms : n (X, x0 )
n (Y, (x0 )) given by [f ] 7 [ f ], for all n 1.
Proof. First note that, if f ' g, then f ' g. Indeed, if t is a homotopy between f and
g, then t is a homotopy between f and g. So is well-defined. Moreover, from the
definition of the group operation on n , it is clear that we have (f + g) = ( f ) + ( g).
So ([f + g]) = ([f ]) + ([g]). Hence is a group homomorphism.
The following is a consequence of the definition of the above induced homomorphisms:
Proposition 81. The homomorphisms induced by : X Y on higher homotopy groups
satisfy the following two properties:
1. ( ) = .
2. (idX ) = idn (X,x0 ) .
We thus have the following important consequence:
Corollary 82. If : (X, x0 ) (Y, y0 ) is a homotopy equivalence, then : n (X, x0 )
n (Y, (x0 )) is an isomorphism, for all n 1.
Example 49. Consider Rn (or any contractible space). We have i (Rn ) = 0 for all i 1,
since Rn is homotopy equivalent to a point.
The following result is very useful for computations:
e X is a covering map, then p : n (X,
e x
Proposition 83. If p : X
e) n (X, p(e
x)) is an
isomorphism for all n 2.
107

Proof. First we claim p is surjective. Let x = p(e

x) and consider f : (S n , s0 ) (X, x). Since
e x
n 2, we have that 1 (S n ) = 0, so f (1 (S n , s0 )) = 0 p (1 (X,
e)). So f admits a lift,
n
e x
i.e., there is fe : (S , s0 ) (X,
e) such that p fe = f . Then [f ] = [p fe] = p ([fe]). So p is
surjective.
(S n , s0 )

(X,O x)
p

fe

e x
(X,
e)
Next, we show that p is injective. Suppose [fe] ker p . So p ([fe]) = [p fe] = 0. Let
p fe = f . Then f ' cx via some homotopy t : (S n , s0 ) (X, x0 ) with 1 = f and 0 = cx .
e x
Again, by the lifting criterion, there is a unique et : (S n , s0 ) (X,
e) with p et = t .
(S n , s0 )

et

(X,O x)
p

e x
(X,
e)
Then we have p e1 = 1 = f and p e0 = 0 = cx , so by the uniqueness of lifts, we must
have e1 = fe and e0 = cxe. Then et is a homotopy between fe and cxe. So [fe] = 0. Thus p is
injective.
Example 50. Consider S 1 with its universal covering map p : R S 1 given by p(t) = e2it .
We already know 1 (S 1 ) = Z. If n 2, Proposition 83 yields that n (S 1 ) = n (R) = 0.
Example 51. Consider T n = S 1 S 1 S 1 , the n-torus. We have 1 (T n ) = Zn . By using
the universal covering map p : Rn T n , we have by Proposition 83 that i (T n ) = i (Rn ) = 0
for i 2.
Definition 35. If n (X) = 0 for all n 2, the space X is called aspherical.
Proposition 84. Let {X } be a collection of path-connected spaces. Then
!
Y
Y
n
X
n (X )
=

for all n.
Q
Proof. First note that a map f : Y X is a collection of maps f : Y X . For
elements of n , take Y = S n (note that since all spaces are path-connected, we may drop
the reference to base points). For homotopies, take Y = S n I.
Example 52. It is a natural question to find two spaces X and Y such that n (X)
= n (Y )
for all n, but with X and Y not homotopy equivalent. Whiteheads Theorem (to be discussed
108

later on) states that if a map of CW complexes f : X Y induces isomorphisms on all

n , then f is a homotopy equivalence. So we must find X and Y so that there is no
continuous map f : X Y inducing the isomorphisms on n s. Consider X = S 2 RP 3
and Y = RP 2 S 3 . Then n (X) = n (S 2 RP 3 ) = n (S 2 ) n (RP 3 ). Since S 3 is a
covering of RP 3 , for all n 2 we have that n (X) = n (S 2 ) n (S 3 ). We also have 1 (X) =
1 (S 2 ) 1 (RP 3 ) = Z/2. Similarly, we have n (Y ) = n (RP 2 S 3 ) = n (RP 2 ) n (S 3 ).
And since S 2 is a covering of RP 2 , for n 2 we have that n (Y ) = n (S 2 ) n (S 3 ). Finally,
1 (Y ) = 1 (RP 2 ) 1 (S 3 ) = Z/2. So n (X) = n (Y ) for all n. By considering homology
groups, however, we see that X and Y are not homotopy equivalent. Indeed, by the Knneth
formula, we get that H5 (X) = Z while H5 (Y ) = 0 (since RP 3 is oriented while RP 2 is not).
Just like there is a homomorphism 1 (X) H1 (X), we can also construct homomorphisms
n (X) Hn (X)
defined by [f : S n X] 7 f [S n ], where [S n ] is the fundamental class of S n . A very
important result in homotopy theory is the following:
Theorem 85. (Hurewicz)
If n 2 and i (X) = 0 for all i < n, then Hi (X) = 0 for i < n and n (X)
= Hn (X).
Moreover, there is also a relative version of the Hurewicz theorem (see the next section
for a definition of the relative homotopy groups), which can be used to prove the following:
Corollary 86. If X and Y are CW complexes with 1 (X) = 1 (Y ) = 0, and a map f :
X Y induces isomorphisms on all integral homology groups Hn , then f is a homotopy
equivalence.

109

5.2

Relative Homotopy Groups

Given a triple (X, A, x0 ) where x0 A X, we define relative homotopy groups as follows:

Definition 36. Let X be a space and let A X and x0 A. Let
I n1 = {(s1 , . . . , sn ) I n | sn = 0}
and set
J n1 = I n \I n1 .
Then define the n-th homotopy group of the pair (X, A) as:


n (X, A, x0 ) = f : (I n , I n , J n1 ) (X, A, x0 ) /
where, as before, is the homotopy equivalence relation.

Alternatively, by collapsing J n1 to a point, we can take



n (X, A, x0 ) = g : (Dn , S n1 , s0 ) (X, A, x0 ) / .

A sum operation is defined in n (X, A, x0 ) by the same formulas as for n (X, x0 ), except that the coordinate sn now plays a special role and is no longer available for the sum
operation. Thus, we have:
Proposition 87. If n 2, then n (X, A, x0 ) forms a group under the usual sum operation.
Further, if n 3, then n (X, A, x0 ) is abelian.
110

Remark 30. Note that the proposition fails in the case n = 1. Indeed, we have that


1 (X, A, x0 ) = f : (I, {0, 1}, {1}) (X, A, x0 ) / .
Then 1 (X, A, x0 ) consists of homotopy classes paths starting anywhere A and ending at x0 ,
so we cannot always concatenate two paths.

Just as in the absolute case, a map of pairs : (X, A, x0 ) (Y, B, y0 ) induces homomorphisms (X, A, x0 ) (Y, B, y0 ) for all n.
A very important feature of the relative homotopy groups is the following:
Proposition 88. The relative homotopy groups of (X, A, x0 ) fit into a long exact sequence
i

n1 (A, x0 ) 0 (X, x0 ) 0,
n (A, x0 )
n (X, x0 ) n (X, A, x0 )

where the map is defined by [f ] = [f |I n1 ] and all others are induced by inclusions.
Remark 31. Near the end of the above sequence, where group structures are not defined,
exactness still makes sense: the image of one map is the kernel of the next, those elements
mapping to the homotopy class of the constant map.
For what follows, it will be important to have a good description of the zero element
0 n (X, A, x0 ). Recall that for [f ] n (X, x0 ), we had [f ] = 0 iff f is homotopic to the
constant map cx0 . Equivalently, for f : (S n , s0 ) (X, x0 ), we have [f ] = 0 iff f extends to
a map Dn+1 X. (The proof of this fact is the same as that for 1 ).
Lemma 89. Let [f ] n (X, A, x0 ). Then [f ] = 0 iff f ' g for some map g with image
contained in A.
Proof. () Suppose f ' g for some g with Image g A.

111

Then we can deform I n to J n1 as indicated in the above picture, and so g ' cx0 . Since
homotopy is a transitive relation, we then get that f ' cx0 .
() Suppose [f ] = 0 in n (X, A, x0 ). So f ' cx0 via some homotopy F : I n+1 X. Then
we may deform I n inside I n+1 (while fixing the boundary) to J n . Composing with F , we get
a homotopy from f to a map g with Image g A.

Recall that if X is path-connected, then n (X) is independent of our choice of base point
and 1 (X) acts on n (X) for all n. In the relative case, we have:
Lemma 90. If A is path-connected, then : n (X, A, x0 ) n (X, A, x1 ) is an isomorphism, where is a path in A from x0 to x1 .

Remark 32. In particular, if x0 = x1 , we get an action of 1 (A) on n (X, A).

Definition 37. We say that the pair (X, A) is n-connected if i (X, A) = 0 for i n and X
is n-connected if i (X) = 0 for i n.

112

5.3

Homotopy Groups of Spheres

We now turn our attention to computing (some of) the homotopy groups i (S n ). For i
n, i = n + 1, n + 2, n + 3 and a few more cases, this is known. In general, however, this is
a very difficult problem. For i = n, we would expect to have n (S n ) = Z by associating
to each (homotopy class of a) map f : S n S n its degree. For i < n, we will show that
i (S n ) = 0. Note that if f : S i S n is not surjective, i.e., there is y S n \f (S i ), then f
factors through Rn , which is contractible. By composing f with the retraction Rn x0 we
get that f ' cx0 . However, there are surjective maps S i S n for i < n, in which case the
above proof fails. To make things work, we "alter" f to make it cellular.
Definition 38. Let X and Y be CW-complexes. A map f : X Y is called cellular if
f (Xn ) Yn for all n, where Xn denotes the n-skeleton of X and similarly for Y .
Theorem 91. Any map between CW-complexes is homotopic to a cellular map. A similar
statement holds for maps of pairs.
Corollary 92. For i < n, we have i (S n ) = 0.
Proof. Choose the standard CW-structure on S i and S n . For [f ] i (S n ), we may assume
by the above theorem that f : S i S n is cellular. Then f (S i ) (S n )i . But (S n )i is a point,
so f is a constant map.
Corollary 93. Let A X and suppose that all cells of X \ A have dimension > n. Then
i (X, A) = 0 for i n.
Proof. Let [f ] i (X, A). By the relative version of the cellular approximation, the map of
pairs f : (Di , S i1 ) (X, A) is homotopic to a map g with g(Di ) Xi . But for i n, we
have that Xi A, so Image g A. Therefore, [f ] = [g] = 0.
Corollary 94. i (X, Xn ) = 0 for all i n.
Therefore, the long exact sequence for the homotopy groups of the pair (X, Xn ) yields
the following:
Corollary 95. For i < n, we have i (X)
= i (Xn ).
Theorem 96. (Suspension Theorem)
Let f : S i S n be a map, and consider its suspension,
f : S i = S i+1 S n = S n+1 .
The assignment
[f ] i (S n ) 7 [f ] i+1 (S n+1 )
defines a homomorphism i (S n ) i+1 (S n+1 ). Moreover, this is an isomorphism i (S n )
=
i+1 (S n+1 ) for i < 2n 1 and a surjection for i = 2n 1.
113

Corollary 97. n (S n ) is either Z or a finite quotient of Z (for n 2), generated by the

degree map.
Proof. By the Suspension Theorem, we have the following:
Z
= 1 (S 1 )  2 (S 2 )
= 3 (S 3 )
= 4 (S 4 )
=
To show that 1 (S 1 )
= 2 (S 2 ), we can use the long exact sequence for the homotopy groups
of a fibration. (Note: Covering maps are a good example of a fibration with F discrete).
Fo


E


B
i (F ) i (E) i (B) i1 (F )

(5.1)
f

Applying the above long exact sequence to the Hopf fibration S 1 , S 3

S 2 , we obtain:
2 (S 1 ) 2 (S 3 ) 2 (S 2 ) 1 (S 1 ) 1 (S 3 )
Using the fact that 2 (S 3 ) = 0 and 1 (S 3 ) = 0, we therefore have an isomorphism:
2 (S 2 )
= 1 (S 1 )
= Z.
Note that by using the vanishing of the higher homotopy groups of S 1 , the above long exact
sequence also yields that
3 (S 2 )
= 2 (S 2 )
= Z.
Remark 33. Unlike the homology and cohomology groups, the homotopy groups of a finite
CW-complex can be infinitely generated. This fact is discussed in the next example.
Example 53. For n 2, consider the finite CW complex S 1 S n . We then have that
1 S n ),
n (S 1 S n ) = n (S^
1 S n is the universal cover of S 1 S n , depicted below:
where S^

114

By contracting the segments between integers, we have that

_
1 Sn '
S^
Skn .
kZ

So for any n 2, we have:

n (S 1 S n ) = n (

Skn ),

kZ

W
which is the free abelian generated by the inclusions Skn , kZ Skn . Indeed, we have the
following:
W
Lemma 98. n ( Sn ) is free abelian and generated by the inclusions of factors.
W
Proof. First note that, since the image of any f : S n Sn is compact hence contained
n
in the wedge
can assume that there are only
many Sn s in
Q finitely
W of nfinitely many S s, we W
n
n
the wedge S . Then we can regard S as the n-skeleton of S . The cell structure
of a particular Sn consists of aQsingle 0-cell e0 and a single n-cell, en . Thus, in the product
Q
n
0
0
S there is one 0-cell e =
e , which, together with the n-cells
[ Y
(
e0 ) en ,

6=

W n
cells of dimension at least
form the n-skeleton of Sn . Hence Sn \ Q
S has
W only
n
n
2n, which by Corollary 94 yields that the pair ( S , S ) is (2n 1)-connected. In
particular, as n 2, we get:
Y  Y
_
M
M
n ( Sn )

Sn
n (Sn ) =
n (Sn ) =
Z.
= n
=
Q

W
W
To conclude our example, we showed that n (S 1 S n )
= n ( kZ Skn ), and n ( kZ Skn ) is
free abelian generated by the inclusion of each of the infinite number of n-spheres. Therefore,
n (S 1 S n ) is infinitely generated.
Remark 34. Under the action of on n , we can regard n as a Z[1 ]-module, with
X
Z[1 ] = {
na | n Z, 1 }.

Skn

Since all
in the universal cover
Z[1 ]-module of rank 1, i.e.,

kZ

Skn are identified under the 1 -action, n is a free

n
= Z[1 ]Z[Z]
= Z[t, t1 ],
1 7 t
1 7 t1
n 7 tn ,
which is infinitely generated (by the powers of t) over Z (i.e., as an abelian group).
115

5.4

Whiteheads Theorem

In this section, we discuss the following important result:

Theorem 99. (Whitehead)
If X and Y are CW complexes, and a map f : X Y induces isomorphisms on the homotopy
groups n for all n, then f is a homotopy equivalence. Moreover, if X is a subcomplex of Y ,
and f is the inclusion map, then X is a deformation retract of Y .
The following consequence is very useful in practice:
Corollary 100. If X and Y are CW complexes with 1 (X) = 1 (Y ) = 0, and f : X Y
induces isomorphisms on homology groups Hn for all n, then f is a homotopy equivalence.
The above corollary follows from Whiteheads theorem and the following relative version
of the Hurewicz theorem:
Theorem 101. (Hurewicz)
If n 2, and i (X, A) = 0 for i < n, with A simply-connected and non-empty, then
Hi (X, A) = 0 for i < n and n (X, A)
= Hn (X, A).
Before discussing the proof of Whiteheads theorem, let us give an example that shows
that having induced isomorphisms on all homology groups is not sufficient for having a
homotopy equivalence (so the simply-connectedness assumption in Corollary 100 is really
important):
Example 54. Let
X = S 1 , (S 1 S n ) en+1 = Y

(n 2),

where f is the inclusion of X in Y . We know from Example 53 that n (S 1 S n )

= Z[t, t1 ].
We define Y by attaching the (n + 1)-cell en+1 to S 1 S n by a map g : S n = en+1 S 1 S n
so that [g] n (S 1 S n ) corresponds to the element 2t 1 Z[t, t1 ]. We then see that
n (Y ) = Z[t, t1 ]/(2t 1) 6= 0 = n (X),
since by setting t = 12 we get that Z[t, t1 ]/(2t 1)
= Z[ 21 ] = { 2ak | k Z0 } Q. Moreover,
from the long exact sequence of homotopy groups for the (n 1)-connected pair (Y, X), the
inclusion X , Y induces an isomorphism on homotopy groups i for i < n. Finally, this
inclusion map also induces isomorphisms on all homology groups, Hk (X)
= Hk (Y ) forr all
k, as can be seen from cellular homology. Indeed, the cellular boundary map
Hn+1 (Yn+1 , Yn ) Hn (Yn , Yn1 )
is an isomorphism since the degree of the composition of the attaching map S n S 1 S n
of en+1 with the collapse map S 1 S n S n is 2 1 = 1.
Let us now get back to the proof of Whiteheads Theorem 99:
116

Proof. (of Whiteheads theorem) To prove Whiteheads theorem, we will use the following
compression lemma:
Lemma 102. (Compression Lemma)
Let (X, A) be a CW pair, and (Y, B) be a pair with Y path-connected and B 6= 0. Suppose
that for each n > 0 for which X \ A has cells of dimension n, n (Y, B, b0 ) = 0 for all b0 B.
Then any map f : (X, A) (Y, B) is homotopic to some map f 0 : X B fixing A (i.e.
f 0 |A = f |A ).
We can then split the proof of Whiteheads theorem into two cases:
Case 1: If f is an inclusion X , Y , since n (X) = n (Y ) for all n, we get by the long exact
sequence for the homotopy groups of the pair (Y, X) that n (Y, X) = 0 for all n. Applying
the above compression lemma to the identity map id : (Y, X) (Y, X), we get that the
identity map idY is homotopic to a deformation retract r : Y X.
Case 2: The general case of a map f : X Y can be reduced to the above case of an
inclusion by using the mapping cylinder of f , i.e.,
Mf := (X I) t Y /(x, 0) f (x).

Figure 5.1: The mapping cylinder of f , Mf

Note that Mf contains both X = X {0} and Y as subspaces, and Mf deformation retracts
onto Y . Moreover, the map f can be written as the composition of the inclusion i of X into
Mf , and the retraction r from Mf to Y :
r

X = X {1} , Mf
Y,

(f = r i, for i : X {1} , Mf ).

Since Mf is homotopy equivalent to Y via r, it suffices to show that Mf deformation retracts

onto X, so we can replace f with the inclusion map i. If f is a cellular map (i.e. f (X(n) )
Y(n) for all n), then Mf is a CW complex having X as a subcomplex. So we can apply Case
1. If f is not cellular, than f is homotopic to some cellular map g, so we may work with g
and the mapping cylinder Mg and again reduce to Case 1.
117

We can now prove Corollary 100:

Proof. (of Corollary 100)
After replacing Y by the mapping cylinder Mf , we may take f to be an inclusion X , Y .
As Hn (X)
= Hn (Y ) for all n, we have by the long exact sequence for the homology of the
pair (Y, X) that Hn (Y, X) = 0 for all n.
Since X and Y are simply-connected, we have 1 (Y, X) = 0. So by the relative Hurewicz
Theorem 101, the first non-zero n (Y, X) is isomorphic to the first non-zero Hn (Y, X). So
n (Y, X) = 0 for all n. Then, by the homotopy long exact sequence for the pair (Y, X), we
get that
n (X)
= n (Y )
for all n, with isomorphisms induced by the inclusion map f . Finally, Whiteheads theorem
yields that f is a homotopy equivalence.
Example 55. Let X = RP2 and Y = S 2 RP . First note that 1 (X) = 1 (Y )
= Z/2.
2
2
Also, since S is a covering of RP , we have that
i (X)
= i (S 2 ), i 2.
S
Moreover, i (Y )
= i (S 2 ) i (RP ), and as RP is covered by S = n0 S n , we get that
i (Y )
= i (S 2 ) i (S ), i 2.
To calculate i (S ), we use cellular approximation. More precisely, we can approximate any
f : S i S by a cellular map g so that Image g S n for i  n. Thus, [f ] = [g] i (S n ) =
0, and we see that
i (X)
= i (S 2 )
= i (Y ), i 2.
Altogether, we have that X and Y have the same homotopy groups. However, as can
be easily seen by considering homology groups, X and Y are not homotopy equivalent. In
particular, by Whiteheads theorem, there cannot exist a map f : RP2 S 2 RP inducing
isomorphisms on n for all n. Indeed, if such a map existed, it would have to be a homotopy
equivalence.
Example 56. As we will see later on, the CW complexes S 2 and S 3 CP have isomorphic
homotopy groups, but they are not homotopy equivalent.
Let us now prove another important result:
Theorem 103. If f : X Y induces isomorphisms on homotopy groups n for all n, then
it induces isomorphisms on homology and cohomology groups with G coefficients, for any
group G.
Proof. By the universal coefficient theorems, it suffices to show that f induces isomorphisms
on integral homology groups H (; Z).
118

We only prove the assertion here under the extra condition that X is simply connected
(the general case follows easily from spectral sequence theory). As before, we can also
assume that f by an inclusion (by replacing Y with the homotopy equivalent space defined
by mapping cylinder Mf of f ). Since by the hypothesis, n (X)
= n (Y ) for all n, with
isomorphisms induced by the inclusion f , the homotopy long exact sequence of the pair
(Y, X) yields that n (Y, X) = 0 for all n. By the relative Hurewicz theorem (as 1 (X) = 0),
this gives that Hn (Y, X) = 0 for all n. Hence, by the long exact sequence for homology,
Hn (X)
= Hn (Y ) for all n, and the proof is complete.
Example 57. Take X = RP2 S 3 and Y = S 2 RP3 . They have isomorphic homotopy
groups n for all n, but H5 (X) 6
= H5 (Y ). So there cannot exist a map f : X Y inducing
the isomorphisms on the n .
Example 58. Any abelian group G can be realized as n (X) with n 2 for some space X.
In fact, for a finitely generated group G = hg1 , . . . , gs | r1 , . . . , rk i, we can can take
X=

s
_

k
[

n=1

for en+1
attached to
j

Ws

n=1

en+1
,
j

j=1

S n by the map f : Sjn

Ws

n=1

S n with [f ] = rj .

Example 59. Eilenberg-MacLane spaces

For any group G and n Z, one can define a space K(G, n) = X with n (X) = G and
i (X) = 0 for all i 6= n. (These spaces unique up to homotopy!) Some familiar such spaces
are:
K(Z/2, 1) = RP
K(Z, 1) = S 1
K(Z, 2) = CP
For the last example, we can see that 2 (CP )
= 1 (S 1 ) = Z by using the fibration (see
next section for a definition)
S 1 , S CP
which gives the long exact sequence of homotopy groups
2 (S ) 2 (CP ) 1 (S 1 ) 1 (S )
together with the fact that 2 (S ) = 1 (S ) = 0, giving 2 (CP )
= 1 (S 1 ) by exactness.

119

5.5

Fibrations and Fiber Bundles

Definition 39. Homotopy Lifting Property

A map p : E B has the homotopy lifting property with respect to X if, given a homotopy
gt : X B, and a lift g0 : X E, of g0 , there exists a homotopy gt : X E lifting gt and
extending g0 .
g0

/E
?
p

gt

gt


/B

Remark 35. This is a special case of the lift extension property. A map p : E B has the
lift extension property with respect to a pair (Z, A) if for all maps f : Z B and g : A E,
there exists a lift f : Z E of f extending g. (Think of Z = X [0, 1], and A = X {0}.)
7?E
g

p
f

A

/Z


/B

Definition 40. A fibration p : E B is a map having the homotopy lifting property with
respect to all spaces X.
Theorem 104. Given a fibration p : E B, points b B and e p1 (b), and fiber
F := p1 (b), with B path connected, we have a long exact sequence of homotopy groups:
i (F, e) i (E, e) i (B, b) i1 (F, e)
In fact, there is an isomorphism i (B, b)
= i (E, F, e).
Definition 41. Fiber Bundle
A map p : E B is a fiber bundle with fiber F if, for all points b B, there exists
neighborhood Ub of b such that p1 (Ub ) = Ub F and the following diagram commutes:

p1 (Ub )
p

/ Ub

pr

"

Ub

120

Remark 36. Fibers of fibrations are homotopy equivalent, while fibers of fiber bundles are
homeomorphic.
Theorem 105. (Hurewicz)
Fiber bundles over paracompact spaces are fibrations.
Example 60. Examples of fiber bundles
1. If F is discrete, a fiber bundle with fiber F is a covering map.
2. The Mbius band I [1, 1]/(0, x) (1, x) S 1 is a fiber bundle with fiber [1, 1],
induced from the projection map I [1, 1] I.

3. By glueing the unlabeled edges of a Mbius band, we get K S 1 (where K is the

Klein bottle), a fiber bundle with fiber S 1 .
4. The following is a fiber bundle with fiber S 1 :
S 1 , S 2n+1 ( Cn+1 ) CPn
(z0 , . . . , zn ) 7 [z0 : . . . : zn ]
For [z] CPn , there is an i such that zi 6= 0. Then we have
U[z] = {[z0 : . . . : 1 : . . . : zn ]}
= Cn
(with the entry 1 in place of the ith coordinate), with a homeomorphism
p1 (U[z] ) U[z] S 1
(z0 , . . . , zn ) 7 ([z0 : . . . : zn ], zi /|zi |).
From this we get the fibration diagram from our discussion of Eilenberg-MacLane
spaces,
S1  q

S1  q
"

CP

S1  p

S 2n+3

...

"

S 2n+1


CP

n+1

121

...

"

S


CP

' {pt}

In particular, from the long exact sequence of the fibration

S 1 , S CP
with S contactible, we obtain that
i (CP )
= i1 (S 1 ) =

Z i=2
0 i=
6 2

i.e.,
CP = K(Z, 2).
Remark 37. For any topological group G, there exists a universal fiber bundle"
classifying the space of (principal) G-bundles. That is, any G-bundle over a space X
is determined (by pull-back) by (the homotopy class of) a map f : X BG.
G

G


EG

' {pt}

BG

5. Other examples of fibrations are provided by the orthogonal and unitary groups:
O(n 1) , O(n) S n1
A 7 Ax.
(If we assume n is large, the associated long exact sequence will give us that i (O(n))
is independent of n.)
U (n 1) , U (n) S 2n1
A 7 Ax.

In the remaining of this section, we show that any map is homotopic to a fibration.
Given f : A B, define
Ef := {(a, ) | a A, : [0, 1] B with (0) = f (a)}.
Then A can be regarded as a subset of Ef , by mapping a A to (a, cf (a) ), where cf (a) is the
constant path based at the image of a under f . Define
p

Ef B
(a, ) 7 (1)
Then p|A = f , and f = p i where i is the inclusion of A in Ef . Moreover, i is a homotopy
equivalence, and p is a fibration with fiber A.
122

Remark 38. If A = {b} , B, where f is the inclusion of b in B, then Ef is the contractible

space of paths in B starting at b:

In this case, the above construction yields a fibration

B = p1 (b) , Ef B,
where B is the space of all loops based at b. Since Ef is contractible, the associated long
exact sequence of the fibration yields that
i (B)
= i1 (B).

Exercises
1. Let f : X Y be a homotopy equivalence. Let Z be any other space. Show that f
induces bijections:
f : [Z, X] [Z, Y ] and f : [Y, Z] [X, Z] ,
where [A, B] denotes the set of homotopy classes of maps from the space A to B.
2. Find examples of spaces X and Y which have the same homology groups, cohomology
groups, and cohomology rings, but with different homotopy groups.
3. Use homotopy groups in order to show that there is no retraction RPn RPk if n > k > 0.
4. Show that an n-connected, n-dimensional CW complex is contractible.
5. (Extension Lemma)
Given a CW pair (X, A) and a map f : A Y with Y path-connected, show that f can be
extended to a map X Y if n1 (Y ) = 0 for all n such that X \ A has cells of dimension n.
6. Show that a CW complex retracts onto any contractible subcomplex. (Hint: Use the
above extension lemma.)
123

 A,
x0 ) (X, A, x0 ) is a covering space with A = p1 (A), show that the map
7. If p : (X,
A,
x0 ) n (X, A, x0 ) is an isomorphism for all n > 1.
p : n (X,
8. Show that a CW complex is contractible if it is the union of an increasing sequence
of subcomplexes X1 X2 such that each inclusion Xi , Xi+1 is nullhomotopic.

Conclude that
S S nis contractible, and more generally, this is true for the infinite suspension

(X) := n0 (X) of any CW complex X.

9. Use cellular approximation to show that the n-skeletons of homotopy equivalent CW
complexes without cells of dimension n + 1 are also homotopy equivalent.
10. Show that a closed simply-connected 3-manifold is homotopy equivalent to S 3 . (Hint:
Use Poincar Duality, and also the fact that closed manifolds are homotopy equivalent to
CW complexes.)
11. Show that a map f : X Y of connected CW complexes is a homotopy equivalence if
Y to the universal covers induces an
it induces an isomorphism on 1 and if a lift f : X
isomorphism on homology.
12. Show that 7 (S 4 ) is non-trivial. [Hint: It contains a Z-summand.]
13. Prove that the space SO(3) of orthogonal 3 3 matrices with determinant 1 is homeomorphic to RP3 .
14. Show that if S k S m S n is a fiber bundle, then k = n 1 and m = 2n 1.
15. Show that if there were fiber bundles S n1 S 2n1 S n for all n, then the groups
i (S n ) would be finitely generated free abelian groups computable by induction, and non-zero
if i n 2.

124