Exercises for the lecture

MA3008 – Algebraic Topology

Solutions of Mock Exam 02 Spring Semester 2017

Exercise 1 (Metric spaces).

Let (X, d) be a metric space.

a) Give the definition of the metric topology T (d).

b) Let x ∈ X and let r > 0. Show that the subset Br (x) = {x0 ∈ X | d(x0 , x) < r} is
open with respect to T (d).

c) Let d0 (x, y) = d(x, y) for all x, y ∈ X. Show that this defines a new metric on X,
p

i.e. that (X, d0 ) is also a metric space.

d) Let d0 : X × X → R be the metric from part (c). Prove that T (d0 ) = T (d).

Solution: Let Br (x) ⊂ X be the subset defined in part b). A subset U ⊂ X is called
open with respect to the metric if for every x ∈ U there is an r > 0 such that Br (x) ⊂ U .
The metric topology T (d) is defined as the set of all U ⊂ X that are open with respect to
the metric d.
For b) let y ∈ Br (x) and let r0 = r − d(x, y). Observe that d(x, y) < r, hence r0 > 0. Let
y 0 ∈ Br0 (y). Then we have

d(y 0 , x) ≤ d(y 0 , y) + d(y, x) < r0 + d(y, x) = r0 + d(x, y) = r .

Thus, y 0 ∈ Br (x) and Br0 (y) ⊆ Br (x). Altogether we have proven that for any point
y ∈ Br (x) there is r0 > 0, such that Br0 (y) ⊆ Br (x). Hence, Br (x) is open with respect to
the metric d.
To solve c) we have to show that d0 has the properties of a metric. Since d(x, y) ≥ 0 and
√
x 7→ x maps R+ to itself, we have that d0 (x, y) ≥ 0. Observe that

d0 (x, y) = 0 ⇔ 0 = d0 (x, y)2 = d(x, y) ⇔ x=y .

Moreover, d0 (x, y) = d(x, y) = d(y, x) = d0 (y, x). Hence, d0 is symmetric. To see that
p p
the triangle inequality also holds note that

d0 (x, z)2 = d(x, z) ≤ d(x, y) + d(y, z)

p p
≤ d(x, y) + d(y, z) + 2 d(x, y) d(y, z)
p p 2
= d(x, y) + d(y, z)

2
= d0 (x, y) + d0 (y, z)

Since the square root function is monotonically increasing, we obtain d0 (x, z) ≤ d0 (x, y) +
d0 (y, z). Thus, d0 is indeed a metric.
0
Concerning part d) of the question let x ∈ X, let r > 0 and let Brd (x) and Brd (x)
be the open balls of radius r around x with respect to the metric d or d0 respectively.
Let y ∈ Brd (x). Since the square root function is monotonically increasing, we have
√ d0 (x) and therefore B d (x) ⊆ B√ d0 (x). Likewise,
d0 (x, y) = d(x, y) < r and hence y ∈ B√
p
r r r
let y ∈ B√ d0 (x). The function x 7→ x2 is monotonically increasing on the non-negative
r
half-line. Hence, we have d(x, y) = d0 (x, y)2 < r. This implies B√ d0 (x) ⊆ B d (x) and
r r
d 0 d
therefore B√r (x) = Br (x). Any subset U ⊆ X that is open with respect to the metric d
is of the form
d0
[ [
U= d
Br(y) (y) = B√ (y)
r(y)
y∈U y∈U

which shows that it is also open with respect to the metric d0 . The same argument proves
that T (d) = T (d0 ).
Exercise 2 (Connectedness and continuous maps).
Let R and R2 be equipped with their standard metric topologies and let S 1 ⊂ R2 and
S 0 = {−1, 1} ⊂ R be equipped with their subspace topologies.

a) Give the definitions of path-connected and of connected.

b) Show that the subspace topology S 0 ⊂ R is the same as the discrete topology on S 0 .

c) Let f : S 1 → S 0 be a continuous map. Prove that it has to be equal either to the

constant map with value 1 or the constant map with value −1.

d) Suppose that f : S 1 → R is a continuous map. Prove that there has to be a point

x ∈ S 1 , such that f (x) = f (−x).
Hint: Suppose that there is no such point and consider the map
f (x) − f (−x)
g : S1 → S0 ; g(x) = .
|f (x) − f (−x)|

Solution A topological space (X, TX ) is path-connected if for any two points x, y ∈ X

there exists a continuous path γ : I → X with γ(0) = x and γ(1) = y. A topological space
(X, TX ) is disconnected if there are two non-empty open subsets U, V ⊆ X with U ∩ V = ∅
and U ∪ V = X. The space X is connected if it is not disconnected.
For part b) observe that J = (−2, 0) ⊂ R is open in the metric topology on R. Since
{−1} = J ∩ S 0 , we see that {−1} ⊂ S 0 is open in the subspace topology. Likewise,
(0, 2) ⊂ R is open and therefore {1} = (0, 2) ∩ S 0 is also open. This implies that the
subspace topology on S 0 induced by R is the discrete topology.
Let U = {−1} ⊂ S 0 and V = {1} ⊂ S 0 . These two sets are non-empty and open and we
have S 0 = U ∪ V and U ∩ V = ∅. Hence, S 0 is disconnected. By a Theorem 4.0.4 in the
lecture notes the continuous image of a connected space is connected. For a continuous
map f : S 1 → S 0 as in part c) we therefore must have f (S 1 ) ⊂ {−1} or f (S 1 ) ⊂ {1}. But
this is the same as saying that f is constant with value either −1 or 1.
To prove d) assume that there is no x ∈ S 1 with the property that f (x) = f (−x). This
means that f (x) − f (−x) 6= 0 for all x ∈ S 1 . The map g : S 1 → S 0 with g(x) = |ff (x)−f
(x)−f (−x)
(−x)|
described in the hint is then continuous. By part c) we have that g(x) is constant with
value either −1 or 1. But this is impossible, since
f (−x) − f (x) f (x) − f (−x)
g(−x) = =− = −g(x) .
|f (−x) − f (x)| |f (x) − f (−x)|
We arrived at a contradiction, hence there must be a point x ∈ S 1 with f (x) = f (−x).
Exercise 3 (Compact spaces).

a) Give the definition of compact topological space and the definition of Hausdorff space.

b) Let (X, TX ) be a compact topological space and let A ⊂ X be a closed subspace. Prove
that A is compact.

c) Let R be equipped with its standard metric topology. Show that if X is compact and
f : X → R is a continuous map, then f is bounded and takes on a minimum and a
maximum value.

d) Let (X, TX ) be a topological space, let A ⊂ X and B ⊂ X be compact subspaces.

Show that A ∪ B is a compact subspace as well.

Solution A family (Ui )i∈I of open subsets of a topological space (X, TX ) is called an
open cover of X if
[
X= Ui .
i∈I

A topological space (X, TX ) is called compact if for any open cover (Ui )i∈I there is a finite
subset {i1 , . . . , in } ⊂ I with the property that
n
[
X= Uik = Ui1 ∪ · · · ∪ Uin .
k=1

A topological space (X, TX ) is called a Hausdorff space if for any two points x, y ∈ X with
x 6= y there are open subsets U, V ⊂ X with x ∈ U , y ∈ V and U ∩ V = ∅. This solves a).
Concerning part b) let A ⊂ X be a closed subspace of a compact topological space (X, TX ).
Let (Ui )i∈I be an open cover of the subspace A. By the definition of the subspace topology
there is an open subset Vi ⊂ X for each i ∈ I with the property that Vi ∩ A = Ui . Observe
that X \ A is open, since A is closed. Hence,
[
X= Vi ∪ (X \ A)
i∈I

is an open cover of X. Since X is compact, there is a finite subset {i1 , . . . , in } ⊆ I such

that
X = Vi1 ∪ · · · ∪ Vin ∪ (X \ A) .

Then we have

A = X ∩ A = (Vi1 ∩ A) ∪ · · · ∪ (Vin ∩ A) ∪ ((X \ A) ∩ A) = Ui1 ∪ · · · ∪ Uin

which proves that A is covered by finitely many of the open subsets Ui . Hence, A is
compact.
Let f : X → R be a continuous map as in part c). Since X is compact, its continuous
image f (X) ⊂ R is compact as well. From the Theorem of Heine and Borel we obtain
that f (X) is closed and bounded. In particular, the boundedness implies that

s := sup{f (x) | x ∈ X} < ∞ .

To show that f takes on a maximum value suppose that s ∈ / f (X). Since f (X) is closed,
R \ f (X) is open. Thus, there is an  > 0 with (s − , s + ) ⊂ (R \ f (X)). But since s ∈ R
is the supremum of f (X), we have that s − ˜ ∈ f (X) for any ˜ > 0. This is a contradiction
since (s − , s) ∩ f (X) = ∅. Therefore we must have s ∈ f (X). To see that f also takes on
a minimum value we use the same argument as above on −f .
For part d) let (Ui )i∈I be an open cover of A ∪ B. Observe that Ui ∩ A is open in the
subspace topology induced by X on A and (Ui ∩ A)i∈I is an open cover of A. Likewise,
(Ui ∩ B)i∈I is an open cover of B. Since A is compact, there are finitely many indices
{i1 , . . . , in } ⊂ I such that A ⊂ Ui1 ∪ · · · ∪ Uin . Similarly, there are finitely many indices
{in+1 , . . . , iN } ⊂ I with B ⊂ Uin+1 ∪ · · · ∪ UiN . Therefore

A ∪ B ⊂ Ui1 ∪ · · · ∪ UiN

proving that A ∪ B is compact.

Exercise 4 (Homotopy equivalence).

a) Let (X, TX ) and (Y, TY ) be two topological spaces and let f1 , f2 : X → Y be two
continuous maps. Define what is meant by: f1 is homotopic to f2 .

b) Let (X, TX ) and (Y, TY ) be two topological spaces. Define what it means for a contin-
uous map f : X → Y to be a homotopy equivalence.

c) Let Y = R3 \ (0, 0, z) ∈ R3 | z ∈ R ⊂ R3 be equipped with the subspace topology,



i.e. Y is the complement of the z-axis. Show that the map f : Y → R2 \ {0} given by
f (x, y, z) = (x, y) is a homotopy equivalence.

d) Let Y ⊂ R3 be the subspace from part c) and let x0 = (1, 0, 0) ∈ R3 . Prove that
π1 (Y, x0 ) ∼
= Z. You may use without proof that π1 (S 1 , z0 ) ∼
= Z for any basepoint
1
z0 ∈ S as shown in the lecture.

Solution Two continuous maps f1 , f2 : X → Y between topological spaces (X, TX ) and

(Y, TY ) are called homotopic if there is a continuous map H : X ×I → Y with the property
that for Ht (x) = H(x, t) we have

H0 = f1 and H1 = f2 .

We denote this by f1 ∼h f2 .
A continuous map f : X → Y is called a homotopy equivalence if there is a continuous
map g : Y → X such that f ◦ g ∼h idY and g ◦ f ∼h idX .
Let Y ⊂ R3 be as in part c), let X = R2 \ {0} and let f : Y → X be the continuous map
given by f (x, y, z) = (x, y). Define g : X → Y by g(x, y) = (x, y, 0). This is well-defined,
since x 6= 0 or y 6= 0. Observe that

f ◦ g = idX ∼h idX

and (g ◦ f )(x, y, z) = (x, y, 0). Define H : Y × I → Y by

H((x, y, z), t) = (x, y, t · z) .

Just as above, this map is well-defined, since x 6= 0 or y 6= 0 implies (x, y, t · z) ∈ Y .

Moreover, H is continuous and we have H0 (x, y, z) = (x, y, 0) = (g ◦ f )(x, y, z) and H1 =
idY . Hence, we also have g ◦ f ∼h idY , which means that f is a homotopy equivalence.
Let x00 = (1, 0) ∈ X and note that f : Y → X and g : X → Y are based continuous maps
between the pointed topological spaces (Y, x0 ) and (X, x00 ). Moreover, we have Ht (x0 ) = x0
for all t ∈ I. Therefore H is a based homotopy equivalence between g ◦ f and idY . Thus,
we know that f induces an isomorphism

f∗ : π1 (Y, x0 ) → π1 (X, x00 )

To solve d) it therefore suffices to construct a based homotopy equivalence between (X, x00 )
and (S 1 , z0 ) for some basepoint z0 ∈ S 1 . Let z0 = (1, 0). The inclusion f 0 : S 1 → X is a
based continuous map. Define g 0 : X → S 1 to be
v
g 0 (v) =
kvk

This is well-defined, since X does not contain the origin. Moreover, it is a based continuous
map, because g 0 (x00 ) = z0 . We have g 0 ◦ f 0 = idS 1 and
v
(f 0 ◦ g 0 )(v) = .
kvk

Define H 0 : X × I → X to be
 
0 1
H (v, t) = t + (1 − t) v.
kvk

This is well-defined, since the term in brackets is never equal to 0. Now note that H00 =
f 0 ◦ g 0 , H10 = idX and Ht (x00 ) = x00 , since kx00 k = 1. In particular, f 0 ◦ g 0 ∼h,+ idX .
Therefore f 0 induces an isomorphism

f∗0 : π1 (X, x00 ) → π1 (S 1 , z0 ) .

Thus, f∗0 ◦ f∗ : π1 (Y, x0 ) → π1 (S 1 , z0 ) is an isomorphism as well. Altogether we obtain

π1 (Y, x0 ) ∼
= π1 (S 1 , z0 ) ∼
=Z.