Topology Exercise Sheet 2

Prof. Dr. Alessandro Sisto Due to March 6

Luca De Rosa

Ex.1:
The goal of this exercise is to give some equivalent characterizations for the interior
of a set. Let X be a topological space and let Y be a subset of X. Let:

(i) int(Y ) = {x ∈ X | there exists O open such that x ∈ O ⊆ Y } (definition in

the notes);

(ii) Y1 be the maximal open that is contained in Y ;

(iii) Y2 be the union of all the open sets that are contained in Y .

Show that int(Y ) = Y1 = Y2 .

Solution:
We show Y2 ⊆ Y1 ⊆ int(Y ) ⊆ Y2 .
For the first inclusion, let x ∈ Y2 . Then x belong to an open set O that is
contained in Y . We claim that O ⊆ Y1 . Suppose that this is not the case, then
O ∪ Y1 would be an open set contained in Y which is strictly bigger than Y1 ,
which is a contradiction.
For the second inclusion, Y1 is an open set that is contained in Y . Thus for
each x ∈ Y1 , we have that x ∈ int(Y ).
For the third inclusion, let x ∈ int(Y ). Then there exists O such that
x ∈ O ⊆ Y . By definition, O ⊆ Y2 , and thus x ∈ Y2 .

Ex.2:
The goal of this exercise is to give some equivalent characterizations for the closure
of a set. Let X be a topological space and let Y be a subset of X. Let:

(i) Y = int(Y ) ∪ {x ∈ X | for each open O that contains x, O ∩ Y 6= ∅ 6=

O ∩ (X − Y )} ;

(ii) Y1 be the minimal closed set that contains Y ;

(iii) Y2 be the intersection of all the closed sets that contain Y ;

1
(iv) Y3 = X − int(X − Y ).

Show that Y = Y1 = Y2 = Y3 .

Solution:
We will show that the first three items are the same as Y3
Case Y = Y3 . Let x be a point of X. Then exactly one of the following
three possibilities holds:

1. There exists an open set O such that x ∈ O ⊆ Y ;

2. There exists an open set O such that x ∈ O ⊆ X − Y ;

3. For every open set O that contains x, O intersects both Y and X − Y .

Indeed, it is clear that (3) holds if and only if (1) and (2) are both false. If (1)
holds, then x ∈ Y , and thus (2) cannot hold. The same apply if (2) holds. We
defined Y to be the set of points that satisfy either (1) or (3). Moreover, the set
of points that satisfy (2) is defined to be int(X −Y ). Thus Y = X −int(X −Y ).
Case Y1 = Y3 . If Y1 is the minimal closed set that contains Y , then X − Y1
is the maximal open set that does not contain Y . In particular, X − Y1 is
the maximal open set that is contained in X − Y . By the previous exercise
X − Y1 = int(X − Y ), which is equivalent to the desired equality.
Case Y2 = Y3 . If Y2 is the intersection of all the closed sets that contain Y ,
then X − Y2 is the union of all the open set that do not contain Y . Then the
conclusion follows as in the case before.

Ex.3:
Give an example of two subsets A and B of R such that:

A ∩ B = ∅, A ∩ B 6= ∅, A ∩ B 6= ∅

Bonus: can you find two (essentially different) such examples?

Solution:
We saw in the lecture that ∂Q = R. Thus A = Q and B = R − Q works.
Another example is A = [0, 1) ∪ [2, 3) and B = [1, 2).

2
Ex.4:
Let A and B be subsets of a topological space X. Show that:
(a) int(A) ∩ int(B) = int(A ∩ B).

Solution:
We start with the inclusion int(A) ∩ int(B) ⊆ int(A ∩ B). By definition,
the interior of Y is the union of all the open sets that are contained in Y .
Thus if a point x is contained in the left hand side, then there are open sets
OA , resp. OB contained in A, resp. B, that contain x. Thus O = OA ∩ OB
is an open set contained in A ∩ B that contains x. So x is a point of the
interior of A ∩ B.
For the other inclusion, let x ∈ int(A ∩ B). Then there is an open O that
contains x such that O ⊆ A ∩ B. In particular O ⊆ A and O ⊆ B. Thus
x ∈ int(A) ∩ int(B).

(b) int(A) ∪ int(B) ⊆ int(A ∪ B).

Solution:
Let x be a point in int(A) ∪ int(B). Then there is an open O that contains
x such that O is contained in either A or B. In particular, O is contained
in A ∪ B. Thus the conclusion follows.

(c) A ∪ B = A ∪ B.

Solution:
By definition, the closure of a set Y is the intersection of all the closed
sets that contain Y . Since A ∪ B is closed and contains both A and B,
we have that A ⊆ A ∪ B and similarly B ⊆ A ∪ B. Thus we obtain
the inclusion A ∪ B ⊆ A ∪ B. For the other inclusion, we show that
X − A ∪ B ⊆ X − A ∪ B. Recall that X − (A ∪ B) = (X − A) ∩ (X − B).
By the previous exercises, the inclusion that we need to show is equivalent
to the following:


int(X − A) ∩ int(X − B) ⊆ int(X − (A ∪ B)) = int (X − A) ∩ (X − B) .

The result follows from part (1).

3
(d) A ∩ B ⊆ A ∩ B.

Solution:
Since A∩B ⊆ A and A∩B ⊆ B, we have that A ∩ B ⊆ A and A ∩ B ⊆ B.
Thus the conclusion follows.

(e) Give one example where the equality in part (b) is satisfied, one where it fails,
one where the equality in part (d) is satisfied and one where it fails.

Solution:
The equality are trivially satisfied if A = B.
An example where equality (b) fails is the following: A = [0, 1] and B =
[1, 2]. Then int(A ∪ B) = (0, 2) but int(A) ∪ int(B) = (0, 1) ∪ (1, 2).
An example where equality (d) fails is the following: let A = (0, 1) and
B = (1, 2). Then A ∩ B = ∅, but A ∩ B = {1}.

Ex.5:
[There exist infinitely many primes]: Let Z be the set of integer numbers. For
every pair of integers a, b ∈ Z, with b > 0, let Ba,b be the set

Ba,b = {a + kb | k ∈ Z}.

Prove the following facts:

(a) The set B = {Ba,b | a, b ∈ Z, b > 0} forms a basis for a topology T on Z.

Solution:
We need to show that

(a) For every point n ∈ Z, there is Ba,b ∈ B such that n ∈ Ba,b . for this,
just notice that Ba,1 = Z, for every a 6= 0.
(b) Given Ba,b , Ba0 ,b0 and a point n ∈ Ba,b ∩ Ba0 ,b0 , there is Ba00 ,b00 such
that n ∈ Ba00 ,b00 ⊆ Ba,b ∩ Ba0 ,b0 .

For the first point, just notice that Ba,1 = Z, for every a 6= 0. For the
second, we claim that choosing a00 = n and b00 = bb0 works. Indeed, n ∈

4
 Bn,bb0 . We show now that Bn,bb0 ⊆ Ba,b . The other case is completely
analogous. Since n ∈ Ba,b , there exists k ∈ Z such that n = a + kb. Thus,
for every s ∈ Z, we have n + sbb0 = a + kb + sbb0 = a + (k + sb0 )b ∈ Ba,b .

(b) For every a, b, with b > 0, the set Ba,b is both open and closed in Z with
respect to B.

Solution:
For each a, b (we always assume b > 0, I will stop to write it), we have that
Ba,b is open by definition. We need to show that is also closed, namely
that it can be written as Z − O, where O is an open set. If b = 1, then
we are done, since the empty set is open. If b > 1, then Ba,b consists of all
numbers that can be reached from a adding a multiple of b. In particular,
the first number contained in Ba,b after a is going to be a + b. Thus all
the numbers a + 1, a + 2, . . . , a + (b − 1) are not in Ba,b . We claim that
for each 0 < r < b, the sets Ba,b and Ba+r,b are disjoint. Indeed, suppose
there was k and k 0 such that a + kb = a + r + k 0 b. Then we would have
r = (k − k 0 )b, which is impossible by choice of r. Moreover, it is clear that
Ba,b ∪ Ba+1,b ∪ · · · ∪ Ba+(b−1),b = Z. This implies that
b−1
[
Ba,b = Z − Ba+s,b .
s=1

That is, we wrote Ba,b as Z minus an open set. Thus Ba,b is closed.

(c) Let P = {2, 3, . . . } be the set of primes. Use theSabove facts to show that P
needs to be infinite. Hint! Consider the set Z − {B0,p | p ∈ P }.

Solution: S
If the set P was finite, then {B0,p | p ∈ P } would be the finite union of
closed sets, thus a closed set. Then, Z − {B0,p | p ∈ P } would be open.
However, by definition of prime number we have that Z − {B0,p | p ∈ P } =
{1, −1}, which is not open. Thus P has to be infinite.

5
Ex.6:
Let X, Y be topological spaces, and let f : X → Z and g : Y → W be maps. Then
we can define a map (f × g : X × Y → Z × W as (f × g)(x, y) = (f (x), g(y)).
We showed (i.e. you can find in the lecture notes) that f × g is continuous if
and only if f and g are continuous. The goal of this exercise is to show that
certain other properties are preserved/not preserved under products. We say that
a function f : X → Z is open if for every open set O ⊆ X we have that f (O) is
open. Similarly f is closed if the image of each closed set is closed.
(a) Show that if f and g are open, then so is f × g;

Solution:
Let O be an open of X × Y . By definition of the product topology, O =
S
Ui × Vi where Ui are open sets of X and Vi are open sets of Y . Then
[  [ [
f × g(O) = f × g Ui × Vi = f × g(Ui × Vi ) = f (Ui ) × g(Vi )

Since the right hand side consists of a union of open sets (product of opens
is open by defintion of product topology), we get that (f × g)(O) is open.

(b) Show with a counterexample that the product of closed functions is not nec-
essarily closed.

Solution:
Let f, g : R → R be functions defined as f (x) = x, g(x) = 0. It is easy to
see that f and g are closed function. However the function f × g : (x, y) 7→
(x, 0) is not closed. Indeed let C be the graphic in R2 of the function x1 , for
x ∈ R − {0}. Then (f × g)(C) = (R2 ) × {0}, which is not a closed subset.
Thus if we can prove that C is a closed subset of R2 , then we would have
that f × g is not closed.
To see that C is closed, consider the function η : R2 → R defined as
η((x, y)) = xy. Clearly, η is a continuous function. Note moreover that
C = η −1 ({1}), in particular, it is closed since preimage of a closed set.
The fact that the preimage (under a continuous map) of a closed set is a
closed set is a well know fact that you don’t need to prove. However, since
repetition is useful, we recall the proof of it. Let f : X → Y be continuous
and C ⊆ Y closed. Then Y − C is open in Y , thus f −1 (Y − C) is open

6
 in X. However X − f −1 (Y − C) = {x ∈ X | f (x) 6∈ Y − C} = {x ∈ X |
f (x) ∈ C} = f −1 (C). Thus f −1 (C) is closed in X.

Ex.7:
Let (X, d) be a metric space equipped with a finite number of points. Show that
in X the distance topology coincides with the discrete topology.

Solution:
Let D be the set of the possible distances between different points of X, that
is:
D = {d(x, y) : x, y ∈ X, x 6= y}.
Note that X is a finite set of positive numbers. In particular, X has a minimum,
which is a positive number. Let c be such a minimum. Then for every x ∈ X,
the ball of radius 2c around x contains only x. Thus, for every x ∈ X, we have
that {x} is open.

Ex.8:
Let p be a prime number, and d : Z × Z → [0, ∞) be a function defined by
m |x−y}
dp (x, y) = p−max{m∈N : p ,

where pm |x − y means pm divides x − y. Prove that dp is a metric on Z and that

dp (x, y) ≤ max{dp (x, z), dp (z, y)} for every x, y, z ∈ Z.

Solution:
The symmetry condition is trivial. By p > 0 and m ∈ N, we have p−m > 0
for every m ∈ N with 0 as the limit. Indeed, dp (x, y) can only be zero, if
pm|x − y for all natural numbers m. This can only happen if x − y = 0,
thus dp (x, y) = 0 if and only if x = y. For the triangle inequality condition,
consider three points x, y and z. For brevity define mx,z as − logp (dp (x, y)).
Without loss of generality we can assume that that mx,z is not larger than
my,z (i.e. dp (x, z) ≥ dp (y, z)). Then pmx,z divides both x − z and z − y
and therefore also x − z + z − y = x − y. This means that mx,y ≥ mx,z

7
 or equivalently dp (x, y) ≤ dp (x, z) = max{dp (x, z), dp (y, z)}. In particular
dp (x, y) ≤ dp (x, z) + dp (y, z) holds.

Ex.9:
Let X be a topological space equipped with a topology TX . Let be Y a subset of
X, and let TY be the subset topology on Y with respect to TX . Let Z be a subset
of Y , let TZ,Y be the subset topology on Z with respect to TY and let TZ,X be the
subset topology on Z with respect to TX . Show that TZ,Y = TZ,X .

Solution:
Let O ∈ TZ,Y . Then O = V ∩ Z, where V ∈ TY . But, by definition of
subset topology on Y , we have that V = U ∩ Y , where U ∈ TX . Thus,
O = (U ∩ Y ) ∩ Z = U ∩ Z, because Z ⊆ Y . Hence we get TZ,Y ⊆ TZ,X .
For the other inclusion, let O ∈ TZ,X . Then O = U ∩ Z, where U ∈ TX .
However, V = U ∩ Y is an open of TY , and hence V ∩ Z = O is an open of
TZ,Y .

Ex.10:
Let Y be a subspace of a topological space X (i.e. Y is a topological space equipped
with the subspace topology) and let A be a subset of Y . Let intX (A) be the interior
of A with respect to X and intY (A) be the interior of A with respect to Y . Show
that intX (A) ⊆ intY (A) and give an example of when the equality does not hold.

Solution:
Let x be a point in intX (A). Then there is an open set O of X that contains x
and that is contained in A. Note that O ∩ Y is an open set of Y that contains
x and is contained in A. Thus intX (A) ⊆ intY (A).
The other inclusion does not hold in general. An example is given by
choosing Y to be not an open set of X, and setting A = Y . Indeed, A is open
in Y (the interior of an open set is the open set itself), but is not open in X.
Concretely, let X = R and A = Y = [0, 1]. Then [0, ε) is a family of opens
subsets of Y that contains 0 and is contained in A. Thus 0 is in the Y -interior
of A. However, 0 is not in the R-interior of [0, 1].