Course 212: Michaelmas Term 2000

Part I: Limits and Continuity, Open and

Closed Sets, Metric Spaces
David R. Wilkins
c David R. Wilkins 19972000
Copyright

Contents
1 Review of Real Analysis
1.1 The Real Number System . . . . . . . . . . . . . .
1.2 Infinite Sequences of Real Numbers . . . . . . . . .
1.3 Monotonic Sequences . . . . . . . . . . . . . . . . .
1.4 Subsequences and the Bolzano-Weierstrass Theorem
1.5 Cauchys Criterion for Convergence . . . . . . . . .
1.6 Limits of Functions of a Real Variable . . . . . . .
1.7 Continuous Functions of a Real Variable. . . . . . .
1.8 The Intermediate Value Theorem . . . . . . . . . .
1.9 Continuous Functions on Closed Bounded Intervals

.
.
.
.
.
.
.
.
.

2
2
4
7
8
9
10
13
16
17

2 Convergence, Continuity and Open Sets in Euclidean Spaces

2.1 Open Sets in Euclidean Spaces . . . . . . . . . . . . . . . . . .
2.2 Closed Sets in a Metric Space . . . . . . . . . . . . . . . . . .
2.3 Continuous Functions and Open and Closed Sets . . . . . . .
2.4 Continuous Functions on Closed Bounded Sets . . . . . . . . .

19
25
28
29
30

3 Metric Spaces
3.1 Convergence and Continuity in Metric Spaces . .
3.2 Open Sets in Metric Spaces . . . . . . . . . . . .
3.3 Closed Sets in a Metric Space . . . . . . . . . . .
3.4 Continuous Functions and Open and Closed Sets
3.5 Homeomorphisms . . . . . . . . . . . . . . . . . .
3.6 Complete Metric Spaces . . . . . . . . . . . . . .
3.7 The Completion of a Metric Space . . . . . . . . .

32
33
35
37
39
41
41
43

.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.

.
.
.
.
.
.
.

Review of Real Analysis

1.1

The Real Number System

The system of real numbers, denoted by R, is an ordered set on which are

defined appropriate operations of addition and multiplication. The system
of real numbers is fully characterized by an axiom system consisting of the
15 axioms (listed below) which describe the algebraic structure and ordering
of the real numbers, together with one further axiom, known as the Least
Upper Bound Axiom, which distinguishes the real number system from other
number systems such as the rational number system. The 15 axioms describing the algebraic and ordering properties of the real number system are as
follows:
1. if x and y are real numbers then their sum x + y is also a real number,
2. (the Commutative Law for addition) x + y = y + x for all real numbers
x and y,
3. (the Associative Law for addition) (x + y) + z = x + (y + z) for all real
numbers x, y and z,
4. there exists a (necessarily unique) real number, denoted by 0, with the
property that x + 0 = x = 0 + x for all real numbers x,
5. for each real numbers x there exists some (necessarily unique) real
number x with the property that x + (x) = 0 = (x) + x,
6. if x and y are real numbers then their product xy is also a real number,
7. (the Commutative Law for multiplication) xy = yx for all real numbers x and y,
8. (the Associative Law for multiplication) (xy)z = x(yz) for all real numbers x, y and z,
9. there exists a (necessarily unique) real number, denoted by 1, with the
property that x1 = x = 1x for all real numbers x, and moreover 1 6= 0,
10. for each real numbers x satisfying x 6= 0 there exists some (necessarily
unique) real number x1 with the property that xx1 = 1 = x1 x,
11. (the Distributive Law ) x(y + z) = (xy) + (xz) for all real numbers x y
and z,

12. (the Trichotomy Law ) if x and y are real numbers then one and only
one of the three statements x < y, x = y and y < x is true,
13. if x, y and z are real numbers and if x < y and y < z then x < z,
14. if x, y and z are real numbers and if x < y then x + z < y + z,
15. if x and y are real numbers which satisfy 0 < x and 0 < y then 0 < xy,
The operations of subtraction and division are defined in terms of addition
and multiplication in the obvious fashion: xy = x+(y) for all real numbers
x and y, and x/y = xy 1 provided that y 6= 0. The absolute value |x| of a
real number x is defined by
n
x
if x 0;
|x| =
x if x < 0.
Note that |x| 0 for all real numbers x and that |x| = 0 if and only if x = 0.
Also |x + y| |x| + |y| and |xy| = |x||y| for all real numbers x and y.
Let D be a subset of R. A real number u is said to be an upper bound of
the set D if x u for all x D. The set D is said to be bounded above if
such an upper bound exists.
Definition Let D be some set of real numbers which is bounded above.
A real number s is said to be the least upper bound (or supremum) of D
(denoted by sup D) if s is an upper bound of D and s u for all upper
bounds u of D.
Example Indeed the real number 2 is the least upper bound of the sets
{x R : x 2} and {x R : x < 2}. Note that the first of these sets
contains its least upper bound, whereas the second set does not.
The axioms (1)(15) listed above describing the algebraic and ordering
properties of the real number system are not in themselves sufficient to fully
characterize the real number system. (Indeed any property of real numbers
that could be derived solely from these axioms would be equally valid for
rational numbers.) We require in addition the following axiom:
the Least Upper Bound Axiom: if D is any non-empty subset
of R which is bounded above then there exists a least upper
bound sup D for the set D.

A lower bound of a set D of real numbers is a real number l with the

property that l x for all x D. A set D of real numbers is said to be
bounded below if such a lower bound exists. If D is bounded below, then
there exists a greatest lower bound (or infimum) inf D of the set D. Indeed
inf D = sup{x R : x D}.
Remark We have simply listed above a complete set of axioms for the real
number system. We have not however proved the existence of a system of
real numbers satisfying these axioms. There are in fact several constructions
of the real number system: one of the most popular of these is the representation of real numbers as Dedekind sections of the set of rational numbers.
For an account of the this construction, and for a proof that these axioms
are sufficient to characterize the real number system, see chapters 2729 of
Calculus, by M. Spivak.

1.2

Infinite Sequences of Real Numbers

An infinite sequence of real numbers is a sequence of the form a1 , a2 , a3 , . . .,

where each an is a real number. (More formally, one can view an infinite
sequence of real numbers as a function from N to R which sends each natural
number n to some real number an .)
Definition A sequence a1 , a2 , a3 , . . . of real numbers is said to converge to
some real number l if and only if the following criterion is satisfied:
given any real number satisfying > 0, there exists some natural
number N such that |an l| < for all n satisfying n N .
If the sequence a1 , a2 , a3 , . . . converges to the limit l then we denote this fact
by writing an l as n +, or by writing lim an = l.
n+

Example A straightforward application of the definition of convergence

shows that 1/n 0 as n +. Indeed suppose that we are given any
real number satisfying > 0. If we pick some natural number N large
enough to satisfy N > 1/ then |1/n| < for all natural numbers n satisfying n N , as required.
Example We show that (1)n /n2 0 as n +. Indeed, given any
real number number satisfying > 0, we can find some natural number N
satisfying N 2 > 1/. If n N then |(1)n /n2 | < , as required.

Example The infinite sequence a1 , a2 , a3 , . . . defined by an = n is not convergent. To prove this formally, we suppose that it were the case that
lim an = l for some real number l, and derive from this a contradiction. On
n+

setting = 1 (say) in the formal definition of convergence, we would deduce

that there would exist some natural number N such that |an l| < 1 for all
n N . But then an < l + 1 for all n N , which is impossible. Thus the
sequence cannot converge.
Example The infinite sequence u1 , u2 , u3 , . . . defined by un = (1)n is not
convergent. To prove this formally, we suppose that it were the case that
lim un = l for some real number l. On setting = 21 in the criterion for
n+

convergence, we would deduce the existence of some natural number N such

that |un l| < 12 for all n N . But then
|un un+1 | |un l| + |l un+1 | <

1
2

1
2

=1

for all n N , contradicting the fact that un un+1 = 2 for all n. Thus the
sequence cannot converge.
Definition We say that an infinite sequence a1 , a2 , a3 , . . . of real numbers is
bounded above if there exists some real number B such that an B for all n.
Similarly we say that this sequence is bounded below if there exists some real
number A such that an A for all n. A sequence is said to be bounded if
it is bounded above and bounded below, so that there exist real numbers A
and B such that A an B for all n.
Lemma 1.1 Every convergent sequence of real numbers is bounded.
Proof Let a1 , a2 , a3 , . . . be a sequence of real numbers converging to some
real number l. On applying the formal definition of convergence (with = 1),
we deduce the existence of some natural number N such that |an l| < 1
for all n N . But then A an B for all n, where A is the minimum
of a1 , a2 , . . . , aN 1 and l 1, and B is the maximum of a1 , a2 , . . . , aN 1 and
l + 1.
Proposition 1.2 Let a1 , a2 , a3 , . . . and b1 , b2 , b3 , . . . be convergent infinite sequences of real numbers. Then the sum, difference and product of these sequences are convergent, and
lim (an + bn ) =

n+

lim (an bn ) =

lim an + lim bn ,

n+

n+

lim an lim bn ,
n+



lim (an bn ) =
lim an
lim bn .

n+

n+

n+

n+

n+

If in addition bn 6= 0 for all n and lim bn 6= 0, then the quotient of the

n+

sequences (an ) and (bn ) is convergent, and

lim an
an
n+
=
.
n+ bn
lim bn
lim

n+

Proof Throughout this proof let l = lim an and m = lim bn .

n+

n+

First we prove that an + bn l + m as n +. Let be any given

real number satisfying > 0. We must show that there exists some natural
number N such that |an + bn (l + m)| < whenever n N . Now an l
as n +, and therefore, given any 1 > 0, there exists some natural
number N1 with the property that |an l| < 1 whenever n N1 . In
particular, there exists a natural number N1 with the property that |an l| <
1
whenever n N1 . (To see this, let 1 = 12 .) Similarly there exists some
2
natural number N2 such that |bn m| < 12 whenever n N2 . Let N be the
maximum of N1 and N2 . If n N then
|an + bn (l + m)| = |(an l) + (bn m)| |an l| + |bn m|
< 12 + 12 = .
Thus an + bn l + m as n +.
Let c be some real number. We show that cbn cm as n +. The
case when c = 0 is trivial. Suppose that c 6= 0. Let > 0 be given. Then
there exists some natural number N such that |bn m| < /|c| whenever
n N . But then |cbn cm| = |c||bn m| < whenever n N . Thus
cbn cm as n +.
If we combine this result, for c = 1, with the previous result, we see
that bn m as n +, and therefore an bn l m as n +.
Next we show that if u1 , u2 , u3 , . . . and v1 , v2 , v3 , . . . are infinite sequences,
and if un 0 and vn 0 as n +, then un vn 0 as n +. Let
> 0 be
numbers N1 and N2 such that
given. Then there exist natural
|un | < whenever n N1 and |vn | < whenever n N2 . Let N be
the maximum of N1 and N2 . If n N then |un vn | < . We deduce that
un vn 0 as n +.
We can apply this result with un = an l and vn = bn m for all natural
numbers n. Using the results we have already obtained, we see that
0 =
=

lim (un vn ) = lim (an bn an m lbn + lm)

n+

n+

lim (an bn ) m lim an l lim bn + lm = lim (an bn ) lm.

n+

n+

n+

n+

Thus an bn lm as n +.
Next we show that if w1 , w2 , w3 , . . . is an infinite sequence of non-zero
real numbers, and if wn 1 as n + then 1/wn 1 as n +. Let
> 0 be given. Let 0 be the minimum of 21 and 12 . Then there exists some
natural number N such that |wn 1| < 0 whenever n N . Thus if n N
then |wn 1| < 21 and 12 < wn < 23 . But then

 

 1
  1 wn  |wn 1|

 

 wn 1 =  wn  = |wn | 2|wn 1| < .
We deduce that 1/wn 1 as n +. If we apply this result with wn =
bn /m, where m 6= 0, we deduce that m/bn 1, and thus 1/bn 1/m as
n +. The result we have already obtained for products of sequences
then enables us to deduce that an /bn l/m as n +.
Example On using Proposition 1.2, we see that
lim (6 (4/n))
6n2 4n
6 (4/n)
6
n+
lim
= lim
=
= = 2.
2
2
2
n+ 3n + 7
n+ 3 + (7/n )
lim (3 + (7/n ))
3
n+

1.3

Monotonic Sequences

An infinite sequence a1 , a2 , a3 , . . . of real numbers is said to be strictly increasing if an+1 > an for all n, strictly decreasing if an+1 < an for all n,
non-decreasing if an+1 an for all n, or non-increasing if an+1 an for all n.
A sequence satisfying any one of these conditions is said to be monotonic;
thus a monotonic sequence is either non-decreasing or non-increasing.
Theorem 1.3 A non-decreasing sequence of real numbers that is bounded
above is convergent. Similarly a non-increasing sequence of real numbers
that is bounded below is convergent.
Proof Let a1 , a2 , a3 , . . . be a non-decreasing sequence of real numbers that
is bounded above. It follows from the Least Upper Bound Axiom that there
exists a least upper bound l for the set {an : n N}. We claim that the
sequence converges to l.
Let > 0 be given. We must show that there exists some natural number N such that |an l| < whenever n N . Now l is not an upper
bound for the set {an : n N} (since l is the least upper bound), and therefore there must exist some natural number N such that aN > l . But then
l < an l whenever n N , since the sequence is non-decreasing and
7

bounded above by l. Thus |an l| < whenever n N . Therefore an l

as n +, as required.
If the sequence a1 , a2 , a3 , . . . is non-increasing and bounded below then
the sequence a1 , a2 , a3 , . . . is non-decreasing and bounded above, and
is therefore convergent. It follows that the sequence a1 , a2 , a3 , . . . is also
convergent.
Example Let a1 = 2 and
an+1 = an

a2n 2
2an

for all natural numbers n. Now

an+1

a2 + 2
= n
and a2n+1 = a2n (a2n 2) +
2an

a2n 2
2an

2


=2+

a2n 2
2an

2
.

It therefore follows by induction on n that an > 0 and a2n > 2 for all natural
numbers n. But then an+1 < an for all n, and thus the sequence a1 , a2 , a3 , . . .
is decreasing and bounded below. It follows from Theorem 1.3 that this
sequence converges to some real number . Also an > 1 for all n (since an > 0
and a2n > 2), and therefore 1. But then, on applying Proposition 1.2,
we see that


2 2
a2n 2
.
= lim an+1 = lim an
=
n+
n+
2an
2

Thus 2 = 2, and so = 2.

1.4

Subsequences and the Bolzano-Weierstrass Theorem

Let a1 , a2 , a3 , . . . be an infinite sequence of real numbers. A subsequence of

this sequence is a sequence of the form an1 , an2 , an3 , . . ., where n1 , n2 , n3 , . . .
are natural numbers satisfying n1 < n2 < n3 < . Thus, for example,
a2 , a4 , a6 , . . . and a1 , a4 , a9 , . . . are subsequences of the given sequence.
Theorem 1.4 (Bolzano-Weierstrass) Every bounded sequence of real numbers has a convergent subsequence.
Proof Let a1 , a2 , a3 , . . . be a bounded sequence of real numbers, and let
S = {n N : an ak for all k n}
8

(i.e., S is the set of all natural numbers n with the property that an is greater
than or equal to all the succeeding members of the sequence).
First let us suppose that the set S is infinite. Arrange the elements of S in
increasing order so that S = {n1 , n2 , n3 , n4 , . . .}, where n1 < n2 < n3 < n4 <
. It follows from the manner in which the set S was defined that an1
an2 an3 an4 . Thus an1 , an2 , an3 , . . . is a non-increasing subsequence
of the original sequence a1 , a2 , a3 , . . .. This subsequence is bounded below
(since the original sequence is bounded). It follows from Theorem 1.3 that
an1 , an2 , an3 , . . . is a convergent subsequence of the original sequence.
Now suppose that the set S is finite. Choose a natural number n1 which
is greater than every natural number belonging to S. Then n1 does not
belong to S. Therefore there must exist some natural number n2 satisfying
n2 > n1 such that an2 > an1 . Moreover n2 does not belong to S (since n2 is
greater than n1 and n1 is greater than every natural number belonging to S).
Therefore there must exist some natural number n3 satisfying n3 > n2 such
that an3 > an2 . We can continue in this way to construct (by induction on j)
a strictly increasing subsequence an1 , an2 , an3 , . . . of our original sequence.
This increasing subsequence is bounded above (since the original sequence is
bounded) and thus is convergent, by Theorem 1.3. This completes the proof
of the Bolzano-Weierstrass Theorem.

1.5

Cauchys Criterion for Convergence

Definition A sequence x1 , x2 , x3 , . . . of real numbers is said to be a Cauchy

sequence if the following condition is satisfied:
for every real number satisfying > 0 there exists some natural
number N such that |xm xn | < for all natural numbers m
and n satisfying m N and n N .
Lemma 1.5 Every Cauchy sequence of real numbers is bounded.
Proof Let x1 , x2 , x3 , . . . be a Cauchy sequence. Then there exists some natural number N such that |xn xm | < 1 whenever m N and n N . In
particular, |xn | |xN | + 1 whenever n N . Therefore |xn | R for all
n, where R is the maximum of the real numbers |x1 |, |x2 |, . . . , |xN 1 | and
|xN | + 1. Thus the sequence is bounded, as required.
The following important result is known as Cauchys Criterion for convergence, or as the General Principle of Convergence.
Theorem 1.6 (Cauchys Criterion for Convergence) An infinite sequence of
real numbers is convergent if and only if it is a Cauchy sequence.
9

Proof First we show that convergent sequences are Cauchy sequences. Let
x1 , x2 , x3 , . . . be a convergent sequence, and let l = lim xj . Let > 0 be
n+

given. Then there exists some natural number N such that |xn l| < 12 for
all n N . Thus if m N and n N then |xm l| < 12 and |xn l| < 21 ,
and hence
|xm xn | = |(xm l) (xn l)| |xm l| + |xn l| < .
Thus the sequence x1 , x2 , x3 , . . . is a Cauchy sequence.
Conversely we must show that any Cauchy sequence x1 , x2 , x3 , . . . is convergent. Now Cauchy sequences are bounded, by Lemma 1.5. The sequence
x1 , x2 , x3 , . . . therefore has a convergent subsequence xn1 , xn2 , xn3 , . . ., by the
Bolzano-Weierstrass Theorem (Theorem 1.4). Let l = limj+ xnj . We
claim that the sequence x1 , x2 , x3 , . . . itself converges to l.
Let > 0 be given. Then there exists some natural number N such
that |xn xm | < 21 whenever m N and n N (since the sequence is a
Cauchy sequence). Let j be chosen large enough to ensure that nj N and
|xnj l| < 12 . Then
|xn l| |xn xnj | + |xnj l| < 12 + 12 =
whenever n N , and thus xn l as n +, as required.

1.6

Limits of Functions of a Real Variable

Definition Let D be a subset of the set R of real numbers, and let s be

a real number (which may or may not belong to D). We say that s is a
limit point of D if, given any > 0, there exists some x D which satisfies
0 < |x s| < .
We now define the limit of a real-valued function at any limit point of
the domain of that function.
Definition Let f : D R be a real-valued function defined over some subset D of R, and let s be a limit point of D. A real number l is said to be the
limit of the function f as x tends to s in D if, given any > 0, there exists
some > 0 such that |f (x) l| < for all x D satisfying 0 < |x s| < .
If l is the limit of f (x) as x tends to s, for some s, then we denote this
fact either by writing f (x) l as x s or by writing lim f (x) = l.
xs

Note that lim f (x) = l if and only if lim f (s + h) = l: this follows directly
xs

h0

from the definition given above.

10

Lemma 1.7 Let f : D R be a real-valued function defined over some subset D of R, and let s be a limit point of D. Then the limit lim f (x), if it
xs
exists, is unique.
Proof Suppose that lim f (x) = l and lim f (x) = m. We must show that
xs
xs
l = m. Let > 0 be given. Then there exist 1 > 0 and 2 > 0 such that
|f (x) l| < whenever x D satisfies 0 < |x s| < 1 and |f (x) m| <
whenever x D satisfies 0 < |x s| < 2 . Choose x D satisfying 0 <
|x s| < , where is the minimum of 1 and 2 . (This is possible since s is
a limit point of D.) Then |f (x) l| < and |f (x) m| < , and hence
|l m| |l f (x)| + |f (x) m| < 2
by the Triangle Inequality. Since |l m| < 2 for all > 0, we conclude that
l = m, as required.
Example We show that lim 14 x2 = 0. Let > 0 be given. Suppose that
x0

we choose = 2 , for example. If 0 < |x| < then | 41 x2 | < 14 2 = , as

required.
Example We show that lim 3x cos(1/x) = 0. Let > 0 be given. Choose
x0

= 31 . Then > 0. Moreover if 0 < |x| < then |3x cos(1/x)| < , as
required.
Proposition 1.8 Let f : D R and g: D R be functions defined over
some subset D of R. Let s be a limit point of D. Suppose that lim f (x) and
xs

lim g(x) exist. Then lim (f (x) + g(x)), lim (f (x) g(x)) and lim (f (x)g(x))
xs
xs
xs
xs
exist, and
lim (f (x) + g(x)) = lim f (x) + lim g(x),

xs

xs

xs

lim (f (x) g(x)) = lim f (x) lim g(x),

xs

xs

xs

lim (f (x)g(x)) = lim f (x) lim g(x).

xs

xs

xs

If in addition g(x) 6= 0 for all x D and lim g(x) 6= 0, then lim f (x)/g(x)
xs
xs
exists, and
lim f (x)
f (x)
xs
=
.
lim
xs g(x)
lim g(x)
xs

11

Proof Let l = lim f (x) and m = lim g(x).

xs

xs

First we prove that lim f (x) + g(x) = l + m. Let > 0 be given. We

xs

must prove that there exists some > 0 such that |f (x) + g(x) (l + m)| <
for all x D satisfying 0 < |x s| < . Now there exist 1 > 0 and 2 > 0
such that |f (x) l| < 12 for all x D satisfying 0 < |x s| < 1 , and
|g(x) m| < 21 for all x D satisfying 0 < |x s| < 2 , since l = lim f (x)
xs

and m = lim g(x). Let be the minimum of 1 and 2 . If x D satisfies

xs

0 < |x s| < then |f (x) l| < 21 and |g(x) m| < 12 , and hence
|f (x) + g(x) (l + m)| |f (x) l| + |g(x) m| < 21 + 12 =
This shows that lim (f (x) + g(x)) = l + m.
xs

Let c be some real number. We show that lim (cg(x)) = cm. The case
xs
when c = 0 is trivial. Suppose that c 6= 0. Let > 0 be given. Then there
exists some > 0 such that |g(x) m| < /|c| whenever 0 < |x s| < .
But then |cg(x) cm| = |c||g(x) m| < whenever 0 < |x s| < . Thus
lim (cg(x)) = cm.
xs
If we combine this result, for c = 1, with the previous result, we see
that lim (g(x)) = m, and therefore lim (f (x) g(x)) = l m.
xs
xs
Next we show that if p: D R and q: D R are functions with the
property that lim p(x) = lim q(x) = 0, then lim (p(x)q(x)) = 0. Let > 0 be
xs
xs
xs

given. Then there exist real numbers 1 >0 and 2 > 0 such that |p(x)| <
whenever 0 < |x s| < 1 and |q(x)| < whenever 0 < |x s| < 2 . Let
be the minimum of 1 and 2 . If 0 < |x s| < then |p(x)q(x)| < . We
deduce that lim (p(x)q(x)) = 0.
xs

We can apply this result with p(x) = f (x) l and q(x) = g(x) m for
all x D. Using the results we have already obtained, we see that
0 = lim (p(x)q(x)) = lim (f (x)g(x) f (x)m lg(x) + lm)
xs

xs

= lim (f (x)g(x)) m lim f (x) l lim g(x) + lm = lim (f (x)g(x)) lm.

xs

xs

xs

xs

Thus lim (f (x)g(x)) = lm.

xs
Next we show that if h: D R is a function that is non-zero throughout D, and if lim h(x) 1 then lim (1/h(x)) = 1. Let > 0 be given. Let
xs

xs

0 be the minimum of 12 and 12 . Then there exists some > 0 such that
|h(x) 1| < 0 whenever 0 < |x s| < . Thus if 0 < |x s| < then
|h(x) 1| < 21 and 21 < h(x) < 32 . But then
 


 1
  h(x) 1  |h(x) 1|

 

 h(x) 1 =  h(x)  = |h(x)| < 2|h(x) 1| < .
12

We deduce that lim 1/h(x) = 1. If we apply this result with h(x) = g(x)/m,
xs

where m 6= 0, we deduce that lim m/g(x) = 1, and thus lim 1/g(x) = 1/m.
xs
xs
The result we have already obtained for products of functions then enables
us to deduce that lim (f (x)/g(x)) l/m.
xs

1.7

Continuous Functions of a Real Variable.

Definition Let D be a subset of R, and let f : D R be a real-valued

function on D. Let s be a point of D. The function f is said to be continuous
at s if, given any > 0, there exists some > 0 such that |f (x) f (s)| <
for all x D satisfying |x s| < . If f is continuous at every point of D
then we say that f is continuous on D.
Example Consider the function f : R R defined by

1 if x > 0;
f (x) =
0 if x 0.
The function f is not continuous at 0. To prove this formally we note that
when 0 < 1 there does not exist any > 0 with the property that
|f (x) f (0)| < for all x satisfying |x| < (since |f (x) f (0)| = 1 for all
x > 0).
Example Let f : R R be the function defined by
(
1
f (x) = sin x if x 6= 0;
0
if x = 0.
We show that this function is not continuous at 0. Suppose that is chosen
to satisfy 0 < < 1. No matter how small we choose , where > 0, we can
always find x R for which |x| < and |f (x) f (0)| . Indeed, given any
> 0, we can choose some integer n large enough to ensure that 0 < xn < ,
where xn satisfies 1/xn = (4n + 1)/2. Moreover f (xn ) = 1. This shows that
the criterion defining the concept of continuity is not satisfied at x = 0.
Example Let f : R R be the function defined by
(
1
f (x) = 3x sin x if x 6= 0;
0
if x = 0.
We claim that the function f is continuous at 0. To prove this, we must
apply the definition of continuity directly. Suppose we are given any real
number satisfying > 0. If = 13 then |f (x)| 3|x| < for all real
numbers x satisfying |x| < , as required.
13

The following lemma describes the relationship between limits and continuity.
Lemma 1.9 Let D be a subset of R, and let s D.
(i) Suppose that s is a limit point of D. Then a function f : D R with
domain D is continuous at s if and only if lim f (x) = f (s);
xs

(ii) Suppose that s is not a limit point of D. Then every function f : D R

with domain D is continuous at s.
Proof If s is a limit point of D belonging to D then the required result follows
immediately on comparing the formal definition of the limit of a function with
the formal definition of continuity (since the condition |f (x) f (s)| < is
automatically satisfied for any > 0 when x = s).
Suppose that s is not a limit point of D. Then there exists some > 0
such that the only element x of D satisfying |x s| < is s itself. The
definition of continuity is therefore satisfied trivially at s by any function
f : D R with domain D.
Given functions f : D R and g: D R defined over some subset D
of R, we denote by f + g, f g, f g and |f | the functions on D defined by
(f g)(x) = f (x) g(x),

(f + g)(x) = f (x) + g(x),

(f g)(x) = f (x)g(x),

|f |(x) = |f (x)|.

Proposition 1.10 Let f : D R and g: D R be functions defined over

some subset D of R. Suppose that f and g are continuous at some point s
of D. Then the functions f + g, f g and f g are also continuous at s. If
moreover the function g is everywhere non-zero on D then the function f /g
is continuous at s.
Proof This result follows directly from Proposition 1.8, using the fact that
a function f : D R is continuous at a limit point s of D belonging to D if
and only if lim f (x) = f (s) (Lemma 1.9).
xs

Remark Proposition 1.10 can also be proved directly from the formal definition of continuity by a straightforward adaptation of the proof of Proposition 1.8. Indeed suppose that f : D R and g: D R are continuous at
s, where s D. We show that f + g is continuous at s. Let > 0 be given.
Then there exist 1 > 0 and 2 > 0 such that |f (x) f (s)| < 21 for all
14

x D satisfying |x s| < 1 , and |g(x) g(s)| < 21 for all x D satisfying

|x s| < 2 . Let be the minimum of 1 and 2 . If |x s| < then
|f (x) + g(x) (f (s) + g(s))| < |f (x) f (s)| + |g(x) g(s)| < 21 + 12 = ,
showing that f + g is continuous at s. The proof of Proposition 1.8 can
be adapted in a similar fashion to show that f g, f g and f /g are also
continuous at s.
Proposition 1.11 Let f : D R and g: E R be functions defined on D
and E respectively, where D and E are subsets of R satisfying f (D) E.
Let s be an element of D. Suppose that the function f is continuous at s and
that the function g is continuous at f (s). Then the composition g f of f
and g is continuous at s.
Proof Let > 0 be given. Then there exists some > 0 such that |g(u)
g(f (s))| < for all u E satisfying |u f (s)| < . But then there exists
some > 0 such that |f (x) f (s)| < for all x D satisfying |x s| < .
Thus if |x s| < then |g(f (x)) g(f (s))| < . Hence g f is continuous
at s.
Lemma 1.12 Let f : D R be a function defined on some subset D of R,
and let a1 , a2 , a3 , . . . be a sequence of real numbers belonging to D. Suppose
that an s as n +, where s D, and that f is continuous at s. Then
f (an ) f (s) as n +.
Proof Let > 0 be given. Then there exists some > 0 such that |f (x)
f (s)| < for all x D satisfying |x s| < . But then there exists some
positive integer N such that |an s| < for all n satisfying n N . Thus
|f (an ) f (s)| < for all n N . Hence f (an ) f (s) as n +.
Proposition 1.13 Let f : D R and g: E R be functions defined on D
and E respectively, where D and E are subsets of R satisfying f (D) E. Let
s be a limit point of D, and let l be an element of E. Suppose that lim f (x) = l
xs

and that the function g is continuous at l. Then lim g(f (x)) = g(l).
xs

Proof Let > 0 be given. Then there exists some > 0 such that |g(u)
g(l)| < for all u E satisfying |u l| < . But then there exists > 0
such that |f (x) l| < for all x D satisfying 0 < |x s| < . Thus if
0 < |x s| < then |g(f (x)) g(l)| < . Hence lim g(f (x)) g(l).
xs

15

1.8

The Intermediate Value Theorem

Theorem 1.14 (The Intermediate Value Theorem) Let a and b be real numbers satisfying a < b, and let f : [a, b] R be a continuous function defined
on the interval [a, b]. Let c be a real number which lies between f (a) and f (b)
(so that either f (a) c f (b) or else f (a) c f (b).) Then there exists
some s [a, b] for which f (s) = c.
Proof We first prove the result in the special case in which c = 0 and
f (a) 0 f (b). We must show that there exists some s [a, b] for which
f (s) = 0. Let S be the subset of [a, b] defined by S = {x [a, b] : f (x) 0}.
The set S is non-empty and bounded above (since a S and b is an upper
bound for the set S). Therefore there exists a least upper bound sup S for
the set S. Let s = sup S. Then a s b, since a S and S [a, b]. We
show that f (s) = 0.
Now if it were the case that f (s) 6= 0. An application of the definition
of continuity (with 0 < |f (s)|) shows that there exists some > 0 such
that f (x) has the same sign as f (s) for all x [a, b] satisfying |x s| < .
(Thus if f (s) > 0 then f (x) > 0 whenever |x s| < , or if f (s) < 0 then
f (x) < 0 whenever |x s| < .)
In particular, suppose that it were the case that f (s) < 0. Then s < b
(since f (b) 0 by hypothesis), and hence f (x) < 0, and thus x S, for some
x [a, b] satisfying s < x < s + . But this would contradict the definition
of s.
Next suppose that it were the case that f (s) > 0. Then s > a (since
f (a) 0 by hypothesis), and hence f (x) > 0, and thus x 6 S, for all
x [a, b] satisfying x > s . But then f (x) > 0 for all x s , and
thus s would be an upper bound of the set S, which also contradicts
the definition of s. The only remaining possibility is that f (s) = 0, which is
what we are seeking to prove.
The result in the general case follows from that in the case c = 0 by
applying the result in this special case to the function x 7 f (x) c when
f (a) c f (b), and to the function x 7 cf (x) when f (a) c f (b).
Corollary 1.15 Given any positive real number b and natural number n,
there exists some positive real number a satisfying an = b.
Proof Let f (x) = xn b, and let c = max(b, 1). Then f (0) < 0 and
f (c) 0. It follows from the Intermediate Value Theorem (Theorem 1.14)
that f (a) = 0 for some real number a satisfying 0 < a c. But then an = b,
as required.

16

Corollary 1.16 Let P be a polynomial of odd degree with real coefficients.

Then P has at least one real root.
Proof Let us write P (x) = a0 + a1 x + a2 x2 + + an xn , where n is odd
and an 6= 0. Without loss of generality we may suppose that an > 0 (after
replacing the polynomial P by P if necessary). We claim that P (K) >
0 and P (K) < 0 for some sufficiently large real number K. (Indeed if
K is chosen large enough to ensure that K > 1 and |an |K 2n|aj | for
j = 0, 1, . . . , n 1 then |P (x) an xn | 12 |an xn | whenever |x| K. But
then P (x) and an xn have the same sign whenever |x| K. In particular,
P (K) > 0 and P (K) < 0.) It follows immediately from the Intermediate
Value Theorem (Theorem 1.14) that there exists some x0 [K, K] for
which P (x0 ) = 0. Thus the polynomial P has at least one real root.
A function f is said to be strictly increasing on an interval I if f (x1 ) <
f (x2 ) for all x1 , x2 I satisfying x1 < x2 .
The next theorem is useful in verifying the continuity of functions involving square roots, nth roots, inverse trigonometric and logarithm functions.
Theorem 1.17 A continuous strictly increasing function f : [a, b] R on an
interval [a, b] has a well-defined continuous inverse f 1 : [c, d] [a, b] defined
over the interval [c, d], where c = f (a) and d = f (b).
Proof The function f is injective (one-to-one) on [a, b], and maps [a, b] into
the interval [c, d], since f is strictly increasing. Moreover it follows immediately from the Intermediate Value Theorem that f maps [a, b] onto [c, d].
Thus f is a bijection from [a, b] to [c, d] and so has a well-defined inverse
f 1 : [c, d] [a, b].
Let v satisfy c < v < d, and let u = f 1 (v) (so that f (u) = v). We
show that f 1 is continuous at v. Let > 0 be given. Now a < u < b,
and hence there exist real numbers u and u+ in the interval [a, b] satisfying
u < u < u < u+ < u + . Let v = f (u ) and v+ = f (u+ ), and
let be the minimum of v+ v and v v . Then > 0, and if y [c, d]
satisfies |y v| < then v < y < v+ . But then u < f 1 (y) < u+ , and
thus |f 1 (y) f 1 (v)| < . We deduce that f 1 is continuous at v whenever
c < v < d. A similar proof shows that f 1 is continuous at both c and d.

1.9

Continuous Functions on Closed Bounded Intervals

Theorem 1.18 Let f : [a, b] R be a continuous real-valued function defined

on the interval [a, b]. Then there exists a constant M with the property that
|f (x)| M for all x [a, b].
17

We give two proofs of this theorem. The first uses the Bolzano-Weierstrass
Theorem which states that every bounded sequence of real numbers possesses
a convergent subsequence. The second makes use of the Least Upper Bound
Axiom.
1st Proof Suppose that the function were not bounded on the interval [a, b].
Then there would exist a sequence x1 , x2 , x3 , . . . of real numbers in the interval
[a, b] such that |f (xn )| > n for all n. The bounded sequence x1 , x2 , x3 , . . .
would possess a convergent subsequence xn1 , xn2 , xn3 , . . ., by the BolzanoWeierstrass Theorem (Theorem 1.4). Moreover the limit l of this subsequence
would belong to [a, b]. But then f (xnk ) f (l) as k +, by Lemma 1.12.
It follows that there exists some natural number N with the property that
|f (xnk ) f (l)| < 1 whenever k N , so that |f (xnk )| |f (l)| + 1 whenever
k N . But this is a contradiction, since |f (xnk )| > nk for all k and nk
increases without limit as k +. Thus the function f is indeed bounded
on the closed interval [a, b].
2nd Proof Define S = { [a, b] : f is bounded on [a, ]}. Clearly a S
and S [a, b]. Thus the set S is non-empty and bounded. It follows from
the Least Upper Bound axiom that there exists a least upper bound for the
set S. Let s = sup S. Then s [a, b]. The function f is continuous at s.
Therefore there exists some > 0 such that |f (x) f (s)| < 1, and thus
|f (x)| < |f (s)| + 1, for all x [a, b] satisfying |x s| < .
Now s is not an upper bound for the set S and hence s < s for
some S. But then the function f is bounded on [a, ] (since S) and
on [, s] (since |f (s)| + 1 is an upper bound on this interval). We conclude
that f is bounded on [a, s], and thus s S. Moreover if it were the case that
s < b then the function f would be bounded on [a, x], and thus x S, for
all x [a, b] satisfying s < x < s + , contradicting the definition of s as the
least upper bound of the set S. Thus s = b. But then b S, so that the
function f is bounded on the interval [a, b] as required.
Theorem 1.19 Let f : [a, b] R be a continuous real-valued function defined
on the interval [a, b]. Then there exist u, v [a, b] with the property that
f (u) f (x) f (v) for all x [a, b].
Proof Let C = sup{f (x) : a x b}. If there did not exist any v [a, b] for
which f (v) = C then the function x 7 1/(C f (x)) would be a continuous
function on the interval [a, b] which was not bounded above on this interval,
thus contradicting Theorem 1.18. Thus there must exist some v [a, b] with
the property that f (v) = C. A similar proof shows that there must exist some
u [a, b] with the property that g(u) = c, where c = inf{f (x) : a x b}.
But then f (u) f (x) f (v) for all x [a, b], as required.
18

Convergence, Continuity and Open Sets in

Euclidean Spaces

We denote by Rn the set consisting of all n-tuples (x1 , x2 , . . . , xn ) of real

numbers. The set Rn represents n-dimensional Euclidean space (with respect
to the standard Cartesian coordinate system). Let x and y be elements of
Rn , where
x = (x1 , x2 , . . . , xn ), y = (y1 , y2 , . . . , yn ),
and let be a real number. We define
x+y
xy
x
xy

=
=
=
=

(x1 + y1 , x2 + y2 , . . . , xn + yn ),
(x1 y1 , x2 y2 , . . . , xn yn ),
(x1 , x2 , . . . , xn ),
x1 y1 + x2 y2 + + xn yn ,
q
|x| =
x21 + x22 + + x2n .

The quantity x y is the scalar product (or inner product) of x and y, and
the quantity |x| is the Euclidean norm of x. Note that |x|2 = x x. The
Euclidean distance between two points x and y of Rn is defined to be the
Euclidean norm |y x| of the vector y x.
Lemma 2.1 (Schwarz Inequality) Let x and y be elements of Rn . Then
|x y| |x||y|.
Proof We note that |x + y|2 0 for all real numbers and . But
|x + y|2 = (x + y).(x + y) = 2 |x|2 + 2x y + 2 |y|2 .
Therefore 2 |x|2 + 2x y + 2 |y|2 0 for all real numbers and . In
particular, suppose that = |y|2 and = x y. We conclude that
|y|4 |x|2 2|y|2 (x y)2 + (x y)2 |y|2 0,
so that (|x|2 |y|2 (x y)2 ) |y|2 0. Thus if y 6= 0 then |y| > 0, and hence
|x|2 |y|2 (x y)2 0.
But this inequality is trivially satisfied when y = 0. Thus |x y| |x||y|, as
required.

19

It follows easily from Schwarz Inequality that |x + y| |x| + |y| for all
x, y Rn . For
|x + y|2 = (x + y).(x + y) = |x|2 + |y|2 + 2x y
|x|2 + |y|2 + 2|x||y| = (|x| + |y|)2 .
It follows that
|z x| |z y| + |y x|
for all points x, y and z of Rn . This important inequality is known as the
Triangle Inequality. It expresses the geometric fact the the length of any
triangle in a Euclidean space is less than or equal to the sum of the lengths
of the other two sides.
Definition A sequence x1 , x2 , x3 , . . . of points in Rn is said to converge to a
point p if and only if the following criterion is satisfied:
given any real number satisfying > 0 there exists some natural
number N such that |p xj | < whenever j N .
We refer to p as the limit lim xj of the sequence x1 , x2 , x3 , . . . .
j+

Lemma 2.2 Let p be a point of Rn , where p = (p1 , p2 , . . . , pn ). Then a

sequence x1 , x2 , x3 , . . . of points in Rn converges to p if and only if the ith
components of the elements of this sequence converge to pi for i = 1, 2, . . . , n.
Proof Let xji and pi denote the ith components of xj and p, where p =
lim xj . Then |xji pi | |xj p| for all j. It follows directly from the

j+

definition of convergence that if xj p as j + then xji pi as

j +.
Conversely suppose that, for each i, xji pi as j +. Let > 0 be
given.
Then there exist natural numbers N1 , N2 , . . . , Nn such that |xji pi | <
/ n whenever j Ni . Let N be the maximum of N1 , N2 , . . . , Nn . If j N
then
n
X

|xj p|2 =
(xji pi )2 < n(/ n)2 = 2 ,
i=1

so that xj p as j +.
Definition A sequence x1 , x2 , x3 , . . . of points in Rn is said to be a Cauchy
sequence if and only if the following criterion is satisfied:
given any real number satisfying > 0 there exists some natural
number N such that |xj xk | < whenever j N and k N .
20

Lemma 2.3 A sequence of points in Rn is convergent if and only if it is a

Cauchy sequence.
Proof Let x1 , x2 , x3 , . . . be a sequence of points of Rn converging to some
point p. Let > 0 be given. Then there exists some natural number N such
that |xj p| < 12 whenever j N . If j N and k N then
|xj xk | |xj p| + |p xk | < 12 + 21 = ,
by the Triangle Inequality. Thus every convergent sequence in Rn is a Cauchy
sequence.
Now let x1 , x2 , x3 , . . . be a Cauchy sequence in Rn . Then the ith components of the elements of this sequence constitute a Cauchy sequence of
real numbers. This Cauchy sequence must converge to some real number
pi , by Cauchys Criterion for Convergence (Theorem 1.6). It follows from
Lemma 2.2 that the Cauchy sequence x1 , x2 , x3 , . . . converges to the point p,
where p = (p1 , p2 , . . . , pn ).
Definition Let X and Y be a subsets of Rm and Rn respectively. A function
f : X Y from X to Y is said to be continuous at a point p of X if and
only if the following criterion is satisfied:
given any real number satisfying > 0 there exists some >
0 such that |f (x) f (p)| < for all points x of X satisfying
|x p| < .
The function f : X Y is said to be continuous on X if and only if it is
continuous at every point p of X.
Lemma 2.4 Let X, Y and Z be subsets of Rm , Rn and Rk respectively, and
let f : X Y and g: Y Z be functions satisfying f (X) Y . Suppose that
f is continuous at some point p of X and that g is continuous at f (p). Then
the composition function g f : X Z is continuous at p.
Proof Let > 0 be given. Then there exists some > 0 such that |g(y)
g(f (p))| < for all y Y satisfying |y f (p)| < . But then there exists
some > 0 such that |f (x) f (p)| < for all x X satisfying |x p| < .
It follows that |g(f (x)) g(f (p))| < for all x X satisfying |x p| < ,
and thus g f is continuous at p, as required.
Lemma 2.5 Let X and Y be a subsets of Rm and Rn respectively, and let
f : X Y be a continuous function from X to Y . Let x1 , x2 , x3 , . . . be a
sequence of points of X which converges to some point p of X. Then the
sequence f (x1 ), f (x2 ), f (x3 ), . . . converges to f (p).
21

Proof Let > 0 be given. Then there exists some > 0 such that
|f (x) f (p)| < for all x X satisfying |x p| < , since the function f is continuous at p. Also there exists some natural number N such
that |xj p| < whenever j N , since the sequence x1 , x2 , x3 , . . . converges to p. Thus if j N then |f (xj ) f (p)| < . Thus the sequence
f (x1 ), f (x2 ), f (x3 ), . . . converges to f (p), as required.
Let X and Y be a subsets of Rm and Rn respectively, and let f : X Y
be a function from X to Y . Then
f (x) = (f1 (x), f2 (x), . . . , fn (x))
for all x X, where f1 , f2 , . . . , fn are functions from X to R, referred to as
the components of the function f .
Proposition 2.6 Let X and Y be a subsets of Rm and Rn respectively. A
function f : X Y is continuous if and only if its components are continuous.
Proof Note that the ith component fi of f is given by fi = i f , where
i : Rn R is the continuous function which maps (y1 , y2 , . . . , yn ) Rn onto
its ith coordinate yi . Now any composition of continuous functions is continuous, by Lemma 2.4. Thus if f is continuous, then so are the components
of f .
Conversely suppose that the components of f are continuous at p X.
Let > 0 be given. Thenthere exist positive real numbers 1 , 2 , . . . , n such
that |fi (x) fi (p)| < / n for x X satisfying |x p| < i . Let be the
minimum of 1 , 2 , . . . , n . If x X satisfies |x p| < then
2

|f (x) f (p)| =

n
X

|fi (x) fi (p)|2 < 2 ,

i=1

and hence |f (x) f (p)| < . Thus the function f is continuous at p, as

required.
Lemma 2.7 The functions s: R2 R and p: R2 R defined by s(x, y) =
x + y and p(x, y) = xy are continuous.
Proof Let (u, v) R2 . We first show that s: R2 R is continuous at (u, v).
Let > 0 be given. Let = 21 . If (x, y) is any point of R2 whose distance
from (u, v) is less than then |x u| < and |y v| < , and hence
|s(x, y) s(u, v)| = |x + y u v| |x u| + |y v| < 2 = .
22

This shows that s: R2 R is continuous at (u, v).

Next we show that p: R2 R is continuous at (u, v). Now
p(x, y) p(u, v) = xy uv = (x u)(y v) + u(y v) + (x u)v.
for all points (x, y) of R2 . Thus if the distance from (x, y) to (u, v) is less
than then |x u| < and |y v| < , and hence |p(x, y) p(u, v)| <
2 + (|u| + |v|). Let > 0 is given. If > 0 is chosen to be the minimum of
1 and /(1 + |u| + |v|) then 2 + (|u| + |v|) < (1 + |u| + |v|) < , and thus
|p(x, y) p(u, v)| < for all points (x, y) of R2 whose distance from (u, v) is
less than . This shows that p: R2 R is continuous at (u, v).
Proposition 2.8 Let X be a subset of Rn , and let f : X R and g: X R
be continuous functions from X to R. Then the functions f + g, f g and
f g are continuous. If in addition g(x) 6= 0 for all x X then the quotient
function f /g is continuous.
Proof Note that f + g = s h and f g = p h, where h: X R2 , s: R2 R
and p: R2 R are given by h(x) = (f (x), g(x)), s(u, v) = u+v and p(u, v) =
uv for all x X and u, v R. It follows from Proposition 2.6, Lemma 2.7
and Lemma 2.4 that f + g and f g are continuous, being compositions
of continuous functions. Now f g = f + (g), and both f and g are
continuous. Therefore f g is continuous.
Now suppose that g(x) 6= 0 for all x X. Note that 1/g = r g, where
r: R \ {0} R is the reciprocal function, defined by r(t) = 1/t. Now the
reciprocal function r is continuous. Thus the function 1/g is a composition
of continuous functions and is thus continuous. But then, using the fact that
a product of continuous real-valued functions is continuous, we deduce that
f /g is continuous.
Example Consider the function f : R2 \ {(0, 0)} R2 defined by


y
x
,
f (x, y) =
.
x2 + y 2 x2 + y 2
The continuity of the components of the function f follows from straightforward applications of Proposition 2.8. It then follows from Proposition 2.6
that the function f is continuous on R2 \ {(0, 0)}.
Let X and Y be subsets of Rm and Rn respectively. A function h: X Y
from X to Y is said to be a homeomorphism if it is a bijection and both
h: X Y and its inverse h1 : Y X are continuous. If there exists a
homeomorphism h: X Y from X to Y then X and Y are said to be
homeomorphic.
23

Example The interval (1, 1) and the real line R are homeomorphic. Indeed
if h: (1, 1) R is defined by h(t) = tan(t/2), then h is a homeomorphism
from (1, 1) to R.
Example Let B n = {x Rn : |x| < 1}, and let : B n Rn be the function
defined by
1
(x) =
x.
1 |x|2
The function is a bijection from B n to Rn whose inverse 1 : Rn B is
given by

p
1 + 1 + 4|x|2
x if x 6= 0;
1 (x) =
2|x|2

0
if x = 0.
n
n
1
n
We claim that : B R and : R B are continuous. Now 1 (x) =
f (|x|)x for all x Rn , where f : R R is the continuous function defined by

(
1 + 1 + 4t2
if t 6= 0;
f (t) =
2t2
1
if t = 0.
(The continuity of this function at 0 follows from a straightforward application of lHopitals rule.) Straighforward applications of Lemma 2.4,
Proposition 2.6 and Proposition 2.8 show that and 1 are continuous.
Thus : B n Rn is a homeomorphism from B n to Rn . In particular, on
setting n = 1, we conclude that the function k: (1, 1) R defined by
k(t) = t/(1 t2 ) for all t (1, 1) is a homeomorphism from (1, 1) to R.
Example Let S n denote the n-sphere, defined by S n = {x Rn+1 : |x| = 1},
and let p be the point of S n with coordinates (0, 0, . . . , 0, 1). We show that
S n \ {p} is homeomorphic to Rn . Define a function h: S n \ {p} Rn by


x2
xn
x1
h(x1 , x2 , . . . , xn , xn+1 ) =
,
,...,
.
1 xn+1 1 xn+1
1 xn+1
The function h is a bijection whose inverse h1 : Rn S n \ {p} is given by


2y1
2y2
2yn
|y|2 1
1
h (y1 , y2 , . . . , yn ) =
,
,..., 2
,
|y|2 + 1 |y|2 + 1
|y| + 1 |y|2 + 1
(where |y|2 = y12 + y22 + + yn2 ). A straightforward application of Propositions 2.6 and 2.8 shows that the functions h and h1 are continuous. Thus
h: S n \ {p} Rn is a homeomorphism. This homeomorphism represents
stereographic projection from S n \ {p} to Rn .
24

Let X and Y be subsets of Rm and Rn , and let h: X Y be a homeomorphism. A sequence x1 , x2 , x3 , . . . of points of X converges in X to some
point p if and only if the sequence h(x1 ), h(x2 ), h(x3 ), . . . converges in Y to
h(p). (This follows on applying Lemma 2.5 to h: X Y and h1 : Y X.)
Also a function f : Y Z mapping Y into some subset Z of Rp is continuous
if and only if f h: X Z is continuous, and a function g: W X mapping
some subset W of Rk into X is continuous if and only if h g: W Y is
continuous.

2.1

Open Sets in Euclidean Spaces

Let X be a subset of Rn . Given a point p of X and a non-negative real

number r, the open ball BX (p, r) in X of radius r about p is defined to be
the subset of X given by
BX (p, r) {x X : |x p| < r}.
(Thus BX (p, r) is the set consisting of all points of X that lie within a sphere
of radius r centred on the point p.)
Definition Let X be a subset of Rn . A subset V of X is said to be open
in X if and only if, given any point p of V , there exists some > 0 such that
BX (p, ) V .
By convention, we regard the empty set as being an open subset of X.
(The criterion given above is satisfied vacuously in the case when V is the
empty set.)
In particular, a subset V of Rn is said to be an open set (in Rn ) if and only
if, given any point p of V , there exists some > 0 such that B(p, ) V ,
where B(p, r) = {x Rn : |x p| < r}.
Example Let H = {(x, y, z) R3 : z > c}, where c is some real number.
Then H is an open set in R3 . Indeed let p be a point of H. Then p = (u, v, w),
where w > c. Let = w c. If the distance from a point (x, y, z) to the
point (u, v, w) is less than then |z w| < , and hence z > c, so that
(x, y, z) H. Thus B(p, ) H, and therefore H is an open set.
The previous example can be generalized. Given any integer i between 1
and n, and given any real number ci , the sets
{(x1 , x2 , . . . , xn ) Rn : xi > ci },

{(x1 , x2 , . . . , xn ) Rn : xi < ci }

are open sets in Rn .

25

Example Let U be an open set in Rn . Then for any subset X of Rn , the

intersection U X is open in X. (This follows directly from the definitions.)
Thus for example, let S 2 be the unit sphere in R3 , given by
S 2 = {(x, y, z) R3 : x2 + y 2 + z 2 = 1}
and let N be the subset of S 2 given by
N = {(x, y, z) Rn : x2 + y 2 + z 2 = 1 and z > 0}.
Then N is open in S 2 , since N = H S 2 , where H is the open set in R3
given by
H = {(x, y, z) R3 : z > 0}.
Note that N is not itself an open set in R3 . Indeed the point (0, 0, 1) belongs
to N , but, for any > 0, the open ball (in R3 of radius about (0, 0, 1)
contains points (x, y, z) for which x2 + y 2 + z 2 6= 1. Thus the open ball of
radius about the point (0, 0, 1) is not a subset of N .
Lemma 2.9 Let X be a subset of Rn , and let p be a point of X. Then, for
any positive real number r, the open ball BX (p, r) in X of radius r about p
is open in X.
Proof Let x be an element of BX (p, r). We must show that there exists
some > 0 such that BX (x, ) BX (p, r). Let = r |x p|. Then > 0,
since |x p| < r. Moreover if y BX (x, ) then
|y p| |y x| + |x p| < + |x p| = r,
by the Triangle Inequality, and hence y BX (p, r). Thus BX (x, )
BX (p, r). This shows that BX (p, r) is an open set, as required.
Lemma 2.10 Let X be a subset of Rn , and let p be a point of X. Then, for
any non-negative real number r, the set {x X : |x p| > r} is an open set
in X.
Proof Let x be a point of X satisfying |x p| > r, and let y be any point
of X satisfying |y x| < , where = |x p| r. Then
|x p| |x y| + |y p|,
by the Triangle Inequality, and therefore
|y p| |x p| |y x| > |x p| = r.
Thus BX (x, ) is contained in the given set. The result follows.
26

Proposition 2.11 Let X be a subset of Rn . The collection of open sets in X

has the following properties:
(i) the empty set and the whole set X are both open in X;
(ii) the union of any collection of open sets in X is itself open in X;
(iii) the intersection of any finite collection of open sets in X is itself open
in X.
Proof The empty set is an open set by convention. Moreover the definition
of an open set is satisfied trivially by the whole set X. This proves (i).
Let A be any collection of open sets in X, and let U denote the union of
all the open sets belonging to A. We must show that U is itself open in X.
Let x U . Then x V for some set V belonging to the collection A. It
follows that there exists some > 0 such that BX (x, ) V . But V U ,
and thus BX (x, ) U . This shows that U is open in X. This proves (ii).
Finally let V1 , V2 , V3 , . . . , Vk be a finite collection of subsets of X that
are open in X, and let V denote the intersection V1 V2 Vk of these
sets. Let x V . Now x Vj for j = 1, 2, . . . , k, and therefore there
exist strictly positive real numbers 1 , 2 , . . . , k such that BX (x, j ) Vj for
j = 1, 2, . . . , k. Let be the minimum of 1 , 2 , . . . , k . Then > 0. (This is
where we need the fact that we are dealing with a finite collection of sets.)
Now BX (x, ) BX (x, j ) Vj for j = 1, 2, . . . , k, and thus BX (x, ) V .
Thus the intersection V of the sets V1 , V2 , . . . , Vk is itself open in X. This
proves (iii).
Example The set {(x, y, z) R3 : x2 + y 2 + z 2 < 4 and z > 1} is an open
set in R3 , since it is the intersection of the open ball of radius 2 about the
origin with the open set {(x, y, z) R3 : z > 1}.
Example The set {(x, y, z) R3 : x2 + y 2 + z 2 < 4 or z > 1} is an open set
in R3 , since it is the union of the open ball of radius 2 about the origin with
the open set {(x, y, z) R3 : z > 1}.
Example The set
{(x, y, z) R3 : (x n)2 + y 2 + z 2 <

1
4

for some n Z}

is an open set in R3 , since it is the union of the open balls of radius

the points (n, 0, 0) for all integers n.

27

1
2

about

Example For each natural number k, let

Vk = {(x, y, z) R3 : k 2 (x2 + y 2 + z 2 ) < 1}.
Now each set Vk is an open ball of radius 1/k about the origin, and is therefore
an open set in R3 . However the intersection of the sets Vk for all natural
numbers k is the set {(0, 0, 0)}, and thus the intersection of the sets Vk for all
natural numbers k is not itself an open set in R3 . This example demonstrates
that infinite intersections of open sets need not be open.
Lemma 2.12 A sequence x1 , x2 , x3 , . . . of points in Rn converges to a point p
if and only if, given any open set U which contains p, there exists some
natural number N such that xj U for all j satisfying j N .
Proof Suppose that the sequence x1 , x2 , x3 , . . . has the property that, given
any open set U which contains p, there exists some natural number N such
that xj U whenever j N . Let > 0 be given. The open ball B(p, ) of
radius about p is an open set by Lemma 2.9. Therefore there exists some
natural number N such that xj B(p, ) whenever j N . Thus |xj p| <
whenever j N . This shows that the sequence converges to p.
Conversely, suppose that the sequence x1 , x2 , x3 , . . . converges to p. Let
U be an open set which contains p. Then there exists some > 0 such that
the open ball B(p, ) of radius about p is a subset of U . Thus there exists
some > 0 such that U contains all points x of X that satisfy |x p| < .
But there exists some natural number N with the property that |xj p| <
whenever j N , since the sequence converges to p. Therefore xj U
whenever j N , as required.

2.2

Closed Sets in a Metric Space

Let X be a subset of Rn . A subset F of X is said to be closed in X if and

only if its complement X \ F in X is open in X. (Recall that X \ F = {x
X : x 6 F }.)
Example The sets {(x, y, z) R3 : z c}, {(x, y, z) R3 : z c}, and
{(x, y, z) R3 : z = c} are closed sets in R3 for each real number c, since the
complements of these sets are open in R3 .
Example Let X be a subset of Rn , and let x0 be a point of X. Then the
sets {x X : |x x0 | r} and {x X : |x x0 | r} are closed for
each non-negative real number r. In particular, the set {x0 } consisting of
the single point x0 is a closed set in X. (These results follow immediately
using Lemma 2.9 and Lemma 2.10 and the definition of closed sets.)
28

Let A be some collection of subsets of a set X. Then

[
\
\
[
X\
S=
(X \ S),
X\
S=
(X \ S)
SA

SA

SA

SA

(i.e., the complement of the union of some collection of subsets of X is the

intersection of the complements of those sets, and the complement of the
intersection of some collection of subsets of X is the union of the complements of those sets). The following result therefore follows directly from
Proposition 2.11.
Proposition 2.13 Let X be a subset of Rn . The collection of closed sets
in X has the following properties:
(i) the empty set and the whole set X are both closed in X;
(ii) the intersection of any collection of closed sets in X is itself closed in
X;
(iii) the union of any finite collection of closed sets in X is itself closed in
X.
Lemma 2.14 Let X be a subset of Rn , and let F be a subset of X which is
closed in X. Let x1 , x2 , x3 , . . . be a sequence of points of F which converges
to a point p of X. Then p F .
Proof The complement X \ F of F in X is open, since F is closed. Suppose
that p were a point belonging to X \F . It would then follow from Lemma 2.12
that xj X \ F for all values of j greater than some positive integer N ,
contradicting the fact that xj F for all j. This contradiction shows that p
must belong to F , as required.

2.3

Continuous Functions and Open and Closed Sets

Let X and Y be subsets of Rm and Rn , and let f : X Y be a function

from X to Y . We recall that the function f is continuous at a point p of X
if, given any > 0, there exists some > 0 such that |f (u) f (p)| <
for all points u of X satisfying |u p| < . Thus the function f : X Y
is continuous at p if and only if, given any > 0, there exists some > 0
such that the function f maps BX (p, ) into BY (f (p), ) (where BX (p, )
and BY (f (p), ) denote the open balls in X and Y of radius and about
p and f (p) respectively).
Given any function f : X Y , we denote by f 1 (V ) the preimage of a
subset V of Y under the map f , defined by f 1 (V ) = {x X : f (x) V }.
29

Proposition 2.15 Let X and Y be subsets of Rm and Rn , and let f : X Y

be a function from X to Y . The function f is continuous if and only if f 1 (V )
is open in X for every open subset V of Y .
Proof Suppose that f : X Y is continuous. Let V be an open set in Y .
We must show that f 1 (V ) is open in X. Let p f 1 (V ). Then f (p)
V . But V is open, hence there exists some > 0 with the property that
BY (f (p), ) V . But f is continuous at p. Therefore there exists some
> 0 such that f maps BX (p, ) into BY (f (p), ) (see the remarks above).
Thus f (x) V for all x BX (p, ), showing that BX (p, ) f 1 (V ). This
shows that f 1 (V ) is open in X for every open set V in Y .
Conversely suppose that f : X Y is a function with the property that
1
f (V ) is open in X for every open set V in Y . Let p X. We must
show that f is continuous at p. Let > 0 be given. Then BY (f (p), ) is
an open set in Y , by Lemma 2.9, hence f 1 (BY (f (p), )) is an open set
in X which contains p. It follows that there exists some > 0 such that
BX (p, ) f 1 (BY (f (p), )). Thus, given any > 0, there exists some
> 0 such that f maps BX (p, ) into BY (f (p), ). We conclude that f is
continuous at p, as required.
Let X be a subset of Rn , let f : X R be continuous, and let c be some
real number. Then the sets {x X : f (x) > c} and {x X : f (x) < c}
are open in X, and, given real numbers a and b satisfying a < b, the set
{x X : a < f (x) < b} is open in X.
Let X and Y be subsets of Rm and Rn . Any homeomorphism h: X Y
induces a one-to-one correspondence between the open sets of X and the open
sets of Y : a subset V of Y is open in Y if and only if h1 (V ) is open in X.
This result follows immediately on applying Proposition 2.15 to h: X Y
and its inverse h1 : Y X.

2.4

Continuous Functions on Closed Bounded Sets

We shall prove that continuous functions are bounded on closed bounded

sets in Rn . First we prove an n-dimensional generalization of the BolzanoWeierstrass Theorem.
A sequence x1 , x2 , x3 , . . . of points in Rn is said to be bounded if there
exists some constant K such that |xj | K for all j.
Theorem 2.16 Every bounded sequence of points in Rn has a convergent
subsequence.

30

Proof Let x1 , x2 , x3 , . . . be a bounded sequence in Rn . The first components

of the elements of the sequence constitute a bounded sequence of real numbers. It follows from the Bolzano-Weierstrass Theorem (Theorem 1.4) that
there exists a subsequence of x1 , x2 , x3 , . . . with the property that the first
components of the elements of the subsequence converge to some real number p1 . We can then extract from this subsequence a further subsequence
with the property that the second components of the elements of the new
subsequence converge to some real number p2 . By proceeding in this fashion (i.e., replacing subsequences of the original sequence with subsequences
of themselves) one can extract a subsequence xn1 , xn2 , xn3 , . . . of the original sequence with the property that the ith components of xnj converge to
some real number pi as j +. It then follows from Lemma 2.2 that the
sequence x1 , x2 , x3 , . . . converges to p, where p = (p1 , p2 , . . . , pn ).
Theorem 2.17 Let X be a closed bounded set in Rn , and let f : X R be
a continuous function on X. Then there exists some constant K such that
|f (x)| K for all x X.
Proof Suppose that the function were not bounded on the set X. Then there
would exist a sequence x1 , x2 , x3 , . . . of points of X such that |f (xj )| > j
for j = 1, 2, 3, . . .. This sequence would be bounded, since X is bounded,
and would therefore possess a convergent subsequence xj1 , xj2 , xj3 , . . ., by
Theorem 2.16). Moreover the limit p of this subsequence would belong to X
by Lemma 2.14, since X is closed. But then f (xjk ) f (p) as k +, by
Lemma 2.5. But this leads to a contradiction, since |f (xjk )| > jk for all k,
so that |f (xjk )| increases without limit as k +, whereas any convergent
sequence of real numbers is bounded. This contradiction shows that the
function f is bounded on the set X, as required.
Corollary 2.18 Let X be a closed bounded set in Rn , and let f : X R be
a continuous function on X. Then there exist points u and v in X with the
property that f (u) f (x) f (v) for all x X.
Proof Let C = sup{f (x) : x X}. If there did not exist any v X for
which f (v) = C then the function x 7 1/(C f (x)) would be a continuous
function on the set X which was not bounded above on this set, thus contradicting Theorem 2.17. Thus there must exist some v X with the property
that f (v) = C. A similar proof shows that there must exist some u X
with the property that g(u) = c, where c = inf{f (x) : x X}. But then
f (u) f (x) f (v) for all x X, as required.

31

Metric Spaces

Definition A metric space (X, d) consists of a set X together with a distance

function d: X X [0, +) on X satisfying the following axioms:
(i) d(x, y) 0 for all x, y X,
(ii) d(x, y) = d(y, x) for all x, y X,
(iii) d(x, z) d(x, y) + d(y, z) for all x, y, z X,
(iv) d(x, y) = 0 if and only if x = y.
The quantity d(x, y) should be thought of as measuring the distance between the points x and y. The inequality d(x, z) d(x, y)+d(y, z) is referred
to as the Triangle Inequality. The elements of a metric space are usually referred to as points of that metric space.
Note that if X is a metric space with distance function d and if A is a
subset of X then the restriction d|A A of d to pairs of points of A defines
a distance function on A satisfying the axioms for a metric space.
The set R of real numbers becomes a metric space with distance function d
given by d(x, y) = |x y| for all x, y R. Similarly the set C of complex
numbers becomes a metric space with distance function d given by d(z, w) =
|z w| for all z, w C, and n-dimensional Euclidean space Rn is a metric
space with with respect to the Euclidean distance function d, given by
v
u n
uX
d(x, y) = |x y| = t (xi yi )2
i=1

for all x, y Rn . Any subset X of R, C or Rn may be regarded as a metric

space whose distance function is the restriction to X of the distance function
on R, C or Rn defined above.
Example The n-sphere S n is defined to be the subset of (n+1)-dimensional
Euclidean space Rn+1 consisting of all elements x of Rn+1 for which |x| = 1.
Thus
S n = {(x1 , x2 , . . . , xn+1 ) Rn+1 : x21 + x22 + + x2n+1 = 1}.
(Note that S 2 is the standard (2-dimensional) unit sphere in 3-dimensional
Euclidean space.) The chordal distance between two points x and y of S n
is defined to be the length |x y| of the line segment joining x and y. The
n-sphere S n is a metric space with respect to the chordal distance function.
32

Example Let C([a, b]) denote the set of all continuous real-valued functions
on the closed interval [a, b], where a and b are real numbers satisfying a < b.
Then C([a, b]) is a metric space with respect to the distance function d, where
d(f, g) = supt[a,b] |f (t) g(t)| for all continuous functions f and g from X
to R. (Note that, for all f, g C([a, b]), f g is a continuous function
on [a, b] and is therefore bounded on [a, b]. Therefore the distance function d
is well-defined.)

3.1

Convergence and Continuity in Metric Spaces

Definition Let X be a metric space with distance function d. A sequence

x1 , x2 , x3 , . . . of points in X is said to converge to a point p in X if and only
if the following criterion is satisfied:
given any real number satisfying > 0, there exists some natural
number N such that d(xn , p) < whenever n N .
We refer to p as the limit lim xn of the sequence x1 , x2 , x3 , . . . .
n+

Note that this definition of convergence generalizes to arbitrary metric

spaces the standard definition of convergence for sequences of real or complex
numbers.
If a sequence of points in a metric space is convergent then the limit of
that sequence is unique. Indeed let x1 , x2 , x3 , . . . be a sequence of points in a
metric space (X, d) which converges to points p and p0 of X. We show that
p = p0 . Now, given any > 0, there exist natural numbers N1 and N2 such
that d(xn , p) < whenever n N1 and d(xn , p0 ) < whenever n N2 . On
choosing n so that n N1 and n N2 we see that
0 d(p, p0 ) d(p, xn ) + d(xn , p0 ) < 2
by a straightforward application of the metric space axioms (i)(iii). Thus
0 d(p, p0 ) < 2 for every > 0, and hence d(p, p0 ) = 0, so that p = p0 by
Axiom (iv).
Lemma 3.1 Let (X, d) be a metric space, and let x1 , x2 , x3 , . . . be a sequence
of points of X which converges to some point p of X. Then, for any point y
of X, d(xn , y) d(p, y) as n +.
Proof Let > 0 be given. We must show that there exists some natural
number N such that |d(xn , y) d(p, y)| < whenever n N . However N
can be chosen such that d(xn , p) < whenever n N . But
d(xn , y) d(xn , p) + d(p, y),

d(p, y) d(p, xn ) + d(xn , y)

33

for all n, hence

d(xn , p) d(xn , y) d(p, y) d(xn , p)
for all n, and hence |d(xn , y) d(p, y)| < whenever n N , as required.
Definition Let X and Y be metric spaces with distance functions dX and
dY respectively. A function f : X Y from X to Y is said to be continuous
at a point x of X if and only if the following criterion is satisfied:
given any real number satisfying > 0 there exists some > 0 such
that dY (f (x), f (x0 )) < for all points x0 of X satisfying dX (x, x0 ) < .
The function f : X Y is said to be continuous on X if and only if it is
continuous at x for every point x of X.
Note that this definition of continuity for functions between metric spaces
generalizes the definition of continuity for functions of a real or complex
variable.
Lemma 3.2 Let X, Y and Z be metric spaces, and let f : X Y and g: Y
Z be continuous functions. Then the composition function g f : X Z is
continuous.
Proof We denote by dX , dY and dZ the distance functions on X, Y and
Z respectively. Let x be any point of X. We show that g f is continuous at x. Let > 0 be given. Now the function g is continuous
at f (x). Hence there exists some > 0 such that dZ (g(y), g(f (x))) <
for all y Y satisfying dY (y, f (x)) < . But then there exists some > 0
such that dY (f (x0 ), f (x)) < for all x0 X satisfying dX (x0 , x) < . Thus
dZ (g(f (x0 )), g(f (x))) < for all x0 X satisfying dX (x0 , x) < , showing
that g f is continuous at x, as required.
Lemma 3.3 Let f : X Y be a continuous function between metric spaces
X and Y , and let x1 , x2 , x3 , . . . be a sequence of points in X which converges
to some point p of X. Then the sequence f (x1 ), f (x2 ), f (x3 ), . . . converges to
f (p).
Proof We denote by dX and dY the distance functions on X and Y respectively. Let > 0 be given. We must show that there exists some natural
number N such that dY (f (xn ), f (p)) < whenever n N . However there
exists some > 0 such that dY (f (x0 ), f (p)) < for all x0 X satisfying
dX (x0 , p) < , since the function f is continuous at p. Also there exists some
natural number N such that dX (xn , p) < whenever n N , since the sequence x1 , x2 , x3 , . . . converges to p. Thus if n N then dY (f (xn ), f (p)) < ,
as required.
34

3.2

Open Sets in Metric Spaces

Definition Let (X, d) be a metric space. Given a point x of X and r 0,

the open ball BX (x, r) of radius r about x in X is defined by
BX (x, r) {x0 X : d(x0 , x) < r}.
Definition Let (X, d) be a metric space. A subset V of X is said to be an
open set if and only if the following condition is satisfied:
given any point v of V there exists some > 0 such that BX (v, ) V .
By convention, we regard the empty set as being an open subset of X.
(The criterion given above is satisfied vacuously in this case.)
Lemma 3.4 Let X be a metric space with distance function d, and let x0 be
a point of X. Then, for any r > 0, the open ball BX (x0 , r) of radius r about
x0 is an open set in X.
Proof Let x BX (x0 , r). We must show that there exists some > 0
such that BX (x, ) BX (x0 , r). Now d(x, x0 ) < r, and hence > 0, where
= r d(x, x0 ). Moreover if x0 BX (x, ) then
d(x0 , x0 ) d(x0 , x) + d(x, x0 ) < + d(x, x0 ) = r,
by the Triangle Inequality, hence x0 BX (x0 , r). Thus BX (x, ) BX (x0 , r),
showing that BX (x0 , r) is an open set, as required.
Lemma 3.5 Let X be a metric space with distance function d, and let x0 be
a point of X. Then, for any r 0, the set {x X : d(x, x0 ) > r} is an open
set in X.
Proof Let x be a point of X satisfying d(x, x0 ) > r, and let x0 be any point
of X satisfying d(x0 , x) < , where = d(x, x0 ) r. Then
d(x, x0 ) d(x, x0 ) + d(x0 , x0 ),
by the Triangle Inequality, and therefore
d(x0 , x0 ) d(x, x0 ) d(x, x0 ) > d(x, x0 ) = r.
Thus BX (x, ) {x0 X : d(x0 , x0 ) > r}, as required.
Proposition 3.6 Let X be a metric space. The collection of open sets in X
has the following properties:
35

(i) the empty set and the whole set X are both open sets;
(ii) the union of any collection of open sets is itself an open set;
(iii) the intersection of any finite collection of open sets is itself an open set.
Proof The empty set is an open set by convention. Moreover the definition
of an open set is satisfied trivially by the whole set X. Thus (i) is satisfied.
Let A be any collection of open sets in X, and let U denote the union of
all the open sets belonging to A. We must show that U is itself an open set.
Let x U . Then x V for some open set V belonging to the collection A.
Therefore there exists some > 0 such that BX (x, ) V . But V U , and
thus BX (x, ) U . This shows that U is open. Thus (ii) is satisfied.
Finally let V1 , V2 , V3 , . . . , Vk be a finite collection of open sets in X, and let
V = V1 V2 Vk . Let x V . Now x Vj for all j, and therefore there
exist strictly positive real numbers 1 , 2 , . . . , k such that BX (x, j ) Vj
for j = 1, 2, . . . , k. Let be the minimum of 1 , 2 , . . . , k . Then > 0.
(This is where we need the fact that we are dealing with a finite collection
of open sets.) Moreover BX (x, ) BX (x, j ) Vj for j = 1, 2, . . . , k, and
thus BX (x, ) V . This shows that the intersection V of the open sets
V1 , V2 , . . . , Vk is itself open. Thus (iii) is satisfied.
Remark For each natural number n, let Vn denote the open set in the
complex plane C defined by
Vn = {z C : |z| < 1/n}.
The intersection of all of these sets (as n ranges over the set of natural
numbers) consists of the set {0}, and this set is not an open subset of the
complex plane. This demonstrates that an intersection of an infinite number
of open sets in a metric space is not necessarily an open set.
Lemma 3.7 Let X be a metric space. A sequence x1 , x2 , x3 , . . . of points
in X converges to a point p if and only if, given any open set U which
contains p, there exists some natural number N such that xj U for all
j N.
Proof Let x1 , x2 , x3 , . . . be a sequence satisfying the given criterion, and let
> 0 be given. The open ball BX (p, ) of radius about p is an open set
(see Lemma 3.4). Therefore there exists some natural number N such that,
if j N , then xj BX (p, ), and thus d(xj , p) < . Hence the sequence (xj )
converges to p.
36

Conversely, suppose that the sequence (xj ) converges to p. Let U be

an open set which contains p. Then there exists some > 0 such that
BX (p, ) U . But xj p as j +, and therefore there exists some
natural number N such that d(xj , p) < for all j N . If j N then
xj BX (p, ) and thus xj U , as required.
Definition Let (X, d) be a metric space, and let x be a point of X. A
subset N of X is said to be a neighbourhood of x (in X) if and only if there
exists some > 0 such that BX (x, ) N , where BX (x, ) is the open ball
of radius about x.
It follows directly from the relevant definitions that a subset V of a metric
space X is an open set if and only if V is a neighbourhood of v for all v V .

3.3

Closed Sets in a Metric Space

A subset F of a metric space X is said to be a closed set in X if and only

if its complement X \ F is open. (Recall that the complement X \ F of F
in X is, by definition, the set of all points of the metric space X that do not
belong to F .) The following result follows immediately from Lemma 3.4 and
Lemma 3.5.
Lemma 3.8 Let X be a metric space with distance function d, and let x0
X. Given any r 0, the sets
{x X : d(x, x0 ) r},

{x X : d(x, x0 ) r}

are closed. In particular, the set {x0 } consisting of the single point x0 is a
closed set in X.
Let A be some collection of subsets of a set X. Then
[
\
\
[
X\
S=
(X \ S),
X\
S=
(X \ S)
SA

SA

SA

SA

(i.e., the complement of the union of some collection of subsets of X is the

intersection of the complements of those sets, and the complement of the
intersection of some collection of subsets of X is the union of the complements of those sets, so that the operation of taking complements converts
unions into intersections and intersections into unions). The following result
therefore follows directly from Proposition 3.6.
Proposition 3.9 Let X be a metric space. The collection of closed sets in X
has the following properties:
37

(i) the empty set and the whole set X are both closed sets;
(ii) the intersection of any collection of closed sets in X is itself a closed
set;
(iii) the union of any finite collection of closed sets in X is itself a closed
set.
Lemma 3.10 Let F be a closed set in a metric space X and let (xj : j N)
be a sequence of points of F . Suppose that xj p as j +. Then p also
belongs to F .
Proof Suppose that the limit p of the sequence were to belong to the complement X \ F of the closed set F . Now X \ F is open, and thus it would
follow from Lemma 3.7 that there would exist some natural number N such
that xj X \ F for all j N , contradicting the fact that xj F for all j.
This contradiction shows that p must belong to F , as required.
Definition Let A be a subset of a metric space X. The closure A of A is
the intersection of all closed subsets of X containing A.
Let A be a subset of the metric space X. Note that the closure A of A
is itself a closed set in X, since the intersection of any collection of closed
subsets of X is itself a closed subset of X (see Proposition 3.9). Moreover if
F is any closed subset of X, and if A F , then A F . Thus the closure A
of A is the smallest closed subset of X containing A.
Lemma 3.11 Let X be a metric space with distance function d, let A be a
subset of X, and let x be a point of X. Then x belongs to the closure A of
A if and only if, given any > 0, there exists some point a of A such that
d(x, a) < .
Proof Let x be a point of X with the property that, given any > 0, there
exists some a A satisfying d(x, a) < . Let F be any closed subset of X
containing A. If x did not belong to F then there would exist some > 0
with the property that BX (x, ) F = , where BX (x, ) denotes the open
ball of radius about x. But this would contradict the fact that BX (x, ) A
is non-empty for all > 0. Thus the point x belongs to every closed subset F
of X that contains A, and therefore x A, by definition of the closure A of
A.
Conversely let x A, and let > 0 be given. Let F be the complement
X \ BX (x, ) of BX (x, ). Then F is a closed subset of X, and the point x
does not belong to F . If BX (x, ) A = then A would be contained in F ,
and hence x F , which is impossible. Therefore there exists a A satisfying
d(x, a) < , as required.
38

3.4

Continuous Functions and Open and Closed Sets

Let X and Y be metric spaces, and let f : X Y be a function from X to Y .

We recall that the function f is continuous at a point x of X if and only
if, given any > 0, there exists some > 0 such that dY (f (x0 ), f (x)) <
for all points x0 of X satisfying dX (x0 , x) < , where dX and dY denote the
distance functions on X and Y respectively. Expressed in terms of open
balls, this means that the function f : X Y is continuous at x if and only
if, given any > 0, there exists some > 0 such that f maps BX (x, ) into
BY (f (x), ) (where BX (x, ) and BY (f (x), ) denote the open balls of radius
and about x and f (x) respectively).
Let f : X Y be a function from a set X to a set Y . Given any subset V
of Y , we denote by f 1 (V ) the preimage of V under the map f , defined by
f 1 (V ) = {x X : f (x) V }.
Proposition 3.12 Let X and Y be metric spaces, and let f : X Y be a
function from X to Y . The function f is continuous if and only if f 1 (V )
is an open set in X for every open set V of Y .
Proof Suppose that f : X Y is continuous. Let V be an open set in Y .
We must show that f 1 (V ) is open in X. Let x be a point belonging
to f 1 (V ). We must show that there exists some > 0 with the property that BX (x, ) f 1 (V ). Now f (x) belongs to V . But V is open, hence
there exists some > 0 with the property that BY (f (x), ) V . But f is
continuous at x. Therefore there exists some > 0 such that f maps the
open ball BX (x, ) into BY (f (x), ) (see the remarks above). Thus f (x0 ) V
for all x0 BX (x, ), showing that BX (x, ) f 1 (V ). We have thus shown
that if f : X Y is continuous then f 1 (V ) is open in X for every open
set V in Y .
Conversely suppose that f : X Y has the property that f 1 (V ) is open
in X for every open set V in Y . Let x be any point of X. We must show
that f is continuous at x. Let > 0 be given. The open ball BY (f (x), )
is an open set in Y , by Lemma 3.4, hence f 1 (BY (f (x), )) is an open set
in X which contains x. It follows that there exists some > 0 such that
BX (x, ) f 1 (BY (f (x), )). We have thus shown that, given any >
0, there exists some > 0 such that f maps the open ball BX (x, ) into
BY (f (x), ). We conclude that f is continuous at x, as required.
Let f : X Y be a function between metric spaces X and Y . Then the
preimage f 1 (Y \ G) of the complement Y \ G of any subset G of Y is equal
to the complement X \ f 1 (G) of the preimage f 1 (G) of G. Indeed
x f 1 (Y \ G) f (x) Y \ G f (x) 6 G x 6 f 1 (G).
39

Also a subset of a metric space is closed if and only if its complement is open.
The following result therefore follows directly from Proposition 3.12.
Corollary 3.13 Let X and Y be metric spaces, and let f : X Y be a
function from X to Y . The function f is continuous if and only if f 1 (G)
is a closed set in X for every closed set G in Y .
Let f : X Y be a continuous function from a metric space X to a metric
space Y . Then, for any point y of Y , the set {x X : f (x) = y} is a closed
subset of X. This follows from Corollary 3.13, together with the fact that
the set {y} consisting of the single point y is a closed subset of the metric
space Y .
Let X be a metric space, and let f : X R be a continuous function from
X to R. Then, given any real number c, the sets
{x X : f (x) > c},

{x X : f (x) < c}

are open subsets of X, and the sets

{x X : f (x) c},

{x X : f (x) c},

{x X : f (x) = c}

are closed subsets of X. Also, given real numbers a and b satisfying a < b,
the set
{x X : a < f (x) < b}
is an open subset of X, and the set
{x X : a f (x) b}
is a closed subset of X.
Similar results hold for continuous functions f : X C from X to C.
Thus, for example,
{x X : |f (x)| < R},

{x X : |f (x)| > R}

are open subsets of X and

{x X : |f (x)| R},

{x X : |f (x)| R},

{x X : |f (x)| = R}

are closed subsets of X, for any non-negative real number R.

40

3.5

Homeomorphisms

Let X and Y be metric spaces. A function h: X Y from X to Y is said to

be a homeomorphism if it is a bijection and both h: X Y and its inverse
h1 : Y X are continuous. If there exists a homeomorphism h: X Y
from a metric space X to a metric space Y , then the metric spaces X and Y
are said to be homeomorphic.
The following result follows directly on applying Proposition 3.12 to
h: X Y and to h1 : Y X.
Lemma 3.14 Any homeomorphism h: X Y between metric spaces X and
Y induces a one-to-one correspondence between the open sets of X and the
open sets of Y : a subset V of Y is open in Y if and only if h1 (V ) is open
in X.
Let X and Y be metric spaces, and let h: X Y be a homeomorphism.
A sequence x1 , x2 , x3 , . . . of points in X is convergent in X if and only if
the corresponding sequence h(x1 ), h(x2 ), h(x3 ), . . . is convergent in Y . (This
follows directly on applying Lemma 3.3 to h: X Y and its inverse h1 : Y
X.) Let Z and W be metric spaces. A function f : Z X is continuous if
and only if hf : Z Y is continuous, and a function g: Y W is continuous
if and only if g h: X W is continuous.

3.6

Complete Metric Spaces

Definition Let X be a metric space with distance function d. A sequence

x1 , x2 , x3 , . . . of points of X is said to be a Cauchy sequence in X if and only if,
given any > 0, there exists some natural number N such that d(xj , xk ) <
for all j and k satisfying j N and k N .
Every convergent sequence in a metric space is a Cauchy sequence. Indeed
let X be a metric space with distance function d, and let x1 , x2 , x3 , . . . be a
sequence of points in X which converges to some point p of X. Given any
> 0, there exists some natural number N such that d(xn , p) < /2 whenever
n N . But then it follows from the Triangle Inequality that
d(xj , xk ) d(xj , p) + d(p, xk ) <

+ =
2 2

whenever j N and k N .
Definition A metric space (X, d) is said to be complete if every Cauchy
sequence in X converges to some point of X.
41

The spaces R and C are complete metric spaces with respect to the distance function given by d(z, w) = |z w|. Indeed this result is Cauchys
Criterion for Convergence. However the space Q of rational numbers (with
distance function d(q, r) = |q r|) is not complete. Indeed one can construct
an
infinite sequence q1 , q2 , q3 , . . . of rational numbers which converges (in R)
to 2. Such a sequence of rational numbers is a Cauchy sequence in both R
and Q. However this Cauchy
sequence does not converge to an point of the

metric space Q (since 2 is an irrational number). Thus the metric space Q

is not complete.
Lemma 3.15 Let X be a complete metric space, and let A be a subset of X.
Then A is complete if and only if A is closed in X.
Proof Suppose that A is closed in X. Let a1 , a2 , a3 , . . . be a Cauchy sequence
in A. This Cauchy sequence must converge to some point p of X, since X is
complete. But the limit of every sequence of points of A must belong to A,
since A is closed (see Lemma 3.10). In particular p A. We deduce that A
is complete.
Conversely, suppose that A is complete. Suppose that A were not closed.
Then the complement X \ A of A would not be open, and therefore there
would exist a point p of X \ A with the property that BX (p, ) A is nonempty for all > 0, where BX (p, ) denotes the open ball in X of radius
centred at p. We could then find a sequence a1 , a2 , a3 , . . . of points of A
satisfying d(aj , p) < 1/j for all natural numbers j. This sequence would be a
Cauchy sequence in A which did not converge to a point of A, contradicting
the completeness of A. Thus A must be closed, as required.
Theorem 3.16 The metric space Rn (with the Euclidean distance function)
is a complete metric space.
Proof Let p1 , p2 , p3 , . . . be a Cauchy sequence in Rn . Then for each integer m between 1 and n, the sequence (p1 )m , (p2 )m , (p3 )m , . . . is a Cauchy
sequence of real numbers, where (pj )m denotes the mth component of pj .
But every Cauchy sequence of real numbers is convergent (Cauchys criterion for convergence). Let qm = lim (pj )m for m = 1, 2, . . . , n, and let
j+

q = (q1 , q2 , . . . , qn ). We claim that pj q as j +.

Let > 0 be given. Then
there exist natural numbers N1 , N2 , . . . , Nn
such that |(pj )m qm | < / n whenever j Nm (where m = 1, 2, . . . , n).
Let N be the maximum of N1 , N2 , . . . , Nn . If j N then
2

|pj q| =

n
X

((pj )m qm )2 < 2 .

m=1

42

Thus pj q as j +. Thus every Cauchy sequence in Rn is convergent,

as required.
The following result follows directly from Lemma 3.15 and Theorem 3.16.
Corollary 3.17 A subset X of Rn is complete if and only if it is closed.
Example The n-sphere S n (with the chordal distance function given by
d(x, y) = |x y|) is a complete metric space, where
S n = {(x1 , x2 , . . . , xn+1 ) Rn+1 : x21 + x22 + + x2n+1 = 1}.

3.7

The Completion of a Metric Space

We describe below a construction whereby any metric space can be embedded

in a complete metric space.
Lemma 3.18 Let X be a metric space with distance function d, let (xj ) and
(yj ) be Cauchy sequences of points in X, and let dj = d(xj , yj ) for all natural
numbers j. Then (dj ) is a Cauchy sequence of real numbers.
Proof It follows from the Triangle Inequality that
dj d(xj , xk ) + dk + d(yk , yj )
and thus dj dk d(xj , xk ) + d(yj , yk ) for all integers j and k. Similarly
dk dj d(xj , xk ) + d(yj , yk ). It follows that
|dj dk | d(xj , xk ) + d(yj , yk )
for all integers j and k.
Let > 0 be given. Then there exists some natural number N such that
d(xj , xk ) < 21 and d(yj , yk ) < 21 whenever j N and k N , since the
sequences (xj ) and (yj ) are Cauchy sequences in X. But then |dj dk | <
whenever j N and k N . Thus the sequence (dj ) is a Cauchy sequence
of real numbers, as required.
Let X be a metric space with distance function d. It follows from Cauchys
Criterion for Convergence and Lemma 3.18 that lim d(xj , yj ) exists for all
j+

Cauchy sequences (xj ) and (yj ) in X.

43

Lemma 3.19 Let X be a metric space with distance function d, and let (xj ),
(yj ) and (zj ) be Cauchy sequences of points in X. Then
0 lim d(xj , zj ) lim d(xj , yj ) + lim d(yj , zj ).
j+

j+

j+

Proof This follows immediately on taking limits of both sides of the Triangle
Inequality.
Lemma 3.20 Let X be a metric space with distance function d, and let (xj ),
(yj ) and (zj ) be Cauchy sequences of points in X. Suppose that
lim d(xj , yj ) = 0 and lim d(yj , zj ) = 0.

j+

j+

Then lim d(xj , zj ) = 0.

j+

Proof This is an immediate consequence of Lemma 3.19.

Lemma 3.21 Let X be a metric space with distance function d, and let
(xj ), (x0j ), (yj ) and (yj0 ) be Cauchy sequences of points in X. Suppose
that lim d(xj , x0j ) = 0 and lim d(yj , yj0 ) = 0. Then lim d(xj , yj ) =
j+

j+

j+

lim d(x0j , yj0 ).

j+

Proof It follows from Lemma 3.19 that

lim d(xj , yj )

j+

lim d(xj , x0j ) + lim d(x0j , yj0 ) + lim d(yj0 , yj )

j+

lim

j+

j+

j+

d(x0j , yj0 ).

Similarly lim d(x0j , yj0 ) lim d(xj , yj ). It follows that lim d(xj , yj ) =
lim

j+

j+
0
d(xj , yj0 ), as

j+

j+

required.

Let X be a metric space with distance function d. Then there is an

equivalence relation on the set of Cauchy sequences of points in X, where
two Cauchy sequences (xj ) and (x0j ) in X are equivalent if and only if
denote the set of equivalence classes of Cauchy
lim d(xj , x0j ) = 0. Let X

j+

sequences in X with respect to this equivalence relation. Let x and y be

and let (xj ) and (yj ) be Cauchy sequences belonging to the
elements of X,
equivalence classes represented by x and y. We define
d(
x, y) = lim d(xj , yj ).
j+

44

It follows from Lemma 3.21 that the value d(

x, y) does not depend on the
choice of Cauchy sequences (xj ) and (yj ) representing x and y. We obtain
This distance function satisfies
in this way a distance function on the set X.
the Triangle Inequality (Lemma 3.19) and the other metric space axioms.
with this distance function is a metric space. We refer to the
Therefore X
as the completion of the metric space X.
space X
We can regard the metric space X as being embedded in its completion

by the equivalence class of the

X, where a point x of X is represented in X
constant sequence x, x, x, . . ..
Example The completion of the space Q of rational numbers is the space R
of real numbers.
of a metric space X is a complete metric
Theorem 3.22 The completion X
space.
of X.
Proof Let x1 , x2 , x3 , . . . be a Cauchy sequence in the completion X
For each positive integer m let xm,1 , xm,2 , xm,3 , . . . be a Cauchy sequence in X
Then,
belonging to the equivalence class that represents the element xm of X.
for each positive integer m there exists a natural number N (m) such that
d(xm,j , xm,k ) < 1/m whenever j N (m) and k N (m). Let ym = xm,N (m) .
We claim that the sequence y1 , y2 , y3 , . . . is a Cauchy sequence in X, and that
corresponding to this Cauchy sequence is the limit in X

the element y of X
of the sequence x1 , x2 , x3 , . . ..
Let > 0 be given. Then there exists some positive integer M such that
M > 3/ and d(
xp , xq ) < 31 whenever p M and q M . It follows from
that if p M and q M then
the definition of the distance function on X
1
d(xp,k , xq,k ) < 3 for all sufficiently large natural numbers k. If p M and
k N (p) then
d(yp , xp,k ) = d(xp,N (p) , xp,k ) < 1/p 1/M < 13
It follows that if p M and q M , and if k is sufficiently large, then
d(yp , xp,k ) < 13 , d(yq , xq,k ) < 13 , and d(xp,k , xq,k ) < 13 , and hence d(yp , yq ) <
. We conclude that the sequence y1 , y2 , y3 , . . . of points of X is indeed a
Cauchy sequence.
which is represented by the Cauchy sequence
Let y be the element of X
y1 , y2 , y3 , . . . of points of X, and, for each natural number m, let ym be the
represented by the constant sequence ym , ym , ym , . . . in X. Now
element of X
d(
y , ym ) = lim d(yp , ym ),
p+

45

and therefore d(
y , ym ) 0 as m +. Also
d(
ym , xm ) = lim d(xm,N (m) , xm,j )
j+

1
m

and hence d(
ym , xm ) 0 as m +. It follows from this that d(
y , xm ) 0
converges
as m +, and therefore the Cauchy sequence x1 , x2 , x3 , . . . in X
We conclude that X
is a complete metric space, since
to the point y of X.
is convergent.
we have shown that every Cauchy sequence in X
Remark In a paper published in 1872, Cantor gave a construction of the real
number system in which real numbers are represented as Cauchy sequences
of rational numbers. The real numbers represented by two Cauchy sequences
of rational numbers are equal if and only if the difference of the Cauchy
sequences converges to zero. Thus the construction of the completion of
a metric space, described above, generalizes Cantors construction of the
system of real numbers from the system of rational numbers.

46