Introduction to Real Analysis

Lee Larson
University of Louisville
May 1, 2015
 About This Document

I often teach the MATH 501-502: Introduction to Real Analysis course at the
University of Louisville. These are notes Ive compiled from those experiences. They
cover the basic ideas of analysis on the real line. The course is intended for a mix of
mostly senior mathematics majors and some beginning graduate students.
Some early versions of the notes leaked out onto the Internet and they are being
offered by some of the usual download sites. The early versions were meant for
me and my classes only, and contain many typos and some gasp! outright
mistakes. The notes are updated and corrected quite often. The date of the version
youre reading is at the bottom-left of most pages. The latest version is available
for download at the Web address math.louisville.edu/lee/ira.
Feel free to use these notes for any purpose, as long as you give me blame or
credit. In return, I only ask you to tell me about mistakes. Any suggestions for
improvements and additions are appreciated. I can be contacted using the email
address on the Web page referenced above.

May 1, 2015

i
 Contents

About This Document i

Chapter 1. Basic Ideas 1-1

1. Sets 1-1
2. Algebra of Sets 1-2
3. Indexed Sets 1-4
4. Functions and Relations 1-5
5. Cardinality 1-10
6. Exercises 1-12

Chapter 2. The Real Numbers 2-1

1. The Field Axioms 2-1
2. The Order Axiom 2-3
3. The Completeness Axiom 2-5
4. Comparisons of Q and R 2-8
5. Exercises 2-10

Chapter 3. Sequences 3-1

1. Basic Properties 3-1
2. Monotone Sequences 3-4
3. Subsequences and the Bolzano-Weierstrass Theorem 3-6
4. Lower and Upper Limits of a Sequence 3-7
5. The Nested Interval Theorem 3-9
6. Cauchy Sequences 3-9
7. Exercises 3-12

Chapter 4. Series 4-1

1. What is a Series? 4-1
2. Positive Series 4-3
3. Absolute and Conditional Convergence 4-10
4. Rearrangements of Series 4-13
5. Exercises 4-15

Chapter 5. The Topology of R 5-1

1. Open and Closed Sets 5-1
2. Relative Topologies and Connectedness 5-4
3. Covering Properties and Compactness on R 5-5
4. More Small Sets 5-8
5. Exercises 5-13

Chapter 6. Limits of Functions 6-1

i
 1. Basic Definitions 6-1
2. Unilateral Limits 6-4
3. Continuity 6-5
4. Unilateral Continuity 6-8
5. Continuous Functions 6-10
6. Uniform Continuity 6-12
7. Exercises 6-13

Chapter 7. Differentiation 7-1

1. The Derivative at a Point 7-1
2. Differentiation Rules 7-2
3. Derivatives and Extreme Points 7-5
4. Differentiable Functions 7-6
5. Applications of the Mean Value Theorem 7-9
6. Exercises 7-14

Chapter 8. Integration 8-1

1. Partitions 8-1
2. Riemann Sums 8-2
3. Darboux Integration 8-4
4. The Integral 8-7
5. The Cauchy Criterion 8-9
6. Properties of the Integral 8-10
7. The Fundamental Theorem of Calculus 8-13
8. Change of Variables 8-17
9. Integral Mean Value Theorems 8-20
10. Exercises 8-21

Chapter 9. Sequences of Functions 9-1

1. Pointwise Convergence 9-1
2. Uniform Convergence 9-4
3. Metric Properties of Uniform Convergence 9-5
4. Series of Functions 9-7
5. Continuity and Uniform Convergence 9-8
6. Integration and Uniform Convergence 9-12
7. Differentiation and Uniform Convergence 9-14
8. Power Series 9-16
9. Exercises 9-22

Chapter 10. Fourier Series 10-1

1. Trigonometric Polynomials 10-1
2. The Riemann Lebesgue Lemma 10-3
3. The Dirichlet Kernel 10-4
4. Dinis Test for Pointwise Convergence 10-7
5. Gibbs Phenomenon 10-9
6. A Divergent Fourier Series 10-11
7. The Fejer Kernel 10-15
8. Exercises 10-18

Appendix. Bibliography A-1

May 1, 2015 http://math.louisville.edu/lee/ira

Index A-1

Appendix. Index A-2

May 1, 2015 http://math.louisville.edu/lee/ira

 CHAPTER 1

Basic Ideas

In the end, all mathematics can be boiled down to logic and set theory. Because
of this, any careful presentation of fundamental mathematical ideas is inevitably
couched in the language of logic and sets. This chapter defines enough of that
language to allow the presentation of basic real analysis. Much of it will be familiar
to you, but look at it anyway to make sure you understand the notation.

1. Sets
Set theory is a large and complicated subject in its own right. There is no time
in this course to touch on any but the simplest parts of it. Instead, well just look
at a few topics from what is often called naive set theory.
We begin with a few definitions.
A set is a collection of objects called elements. Usually, sets are denoted by the
capital letters A, B, , Z. A set can consist of any type and number of elements.
Even other sets can be elements of a set. The sets dealt with here usually have real
numbers as their elements.
If a is an element of the set A, we write a A. If a is not an element of the set
A, we write a / A.
If all the elements of A are also elements of B, then A is a subset of B. In this
case, we write A B or B A. In particular, notice that whenever A is a set, then
A A.
Two sets A and B are equal, if they have the same elements. In this case we
write A = B. It is easy to see that A = B iff A B and B A. Establishing that
both of these containments are true is the most common way to show that two sets
are equal.
If A B and A 6= B, then A is a proper subset of B. In cases when this is
important, it is written A ( B instead of just A B.
There are several ways to describe a set.
A set can be described in words such as P is the set of all presidents of the
United States. This is cumbersome for complicated sets.
All the elements of the set could be listed in curly braces as S = {2, 0, a}. If
the set has many elements, this is impractical, or impossible.
More common in mathematics is set builder notation. Some examples are

P = {p : p is a president of the United states}

= {Washington, Adams, Jefferson, , Clinton, Bush, Obama}

and
A = {n : n is a prime number} = {2, 3, 5, 7, 11, }.
1-1
1-2 CHAPTER 1. BASIC IDEAS

In general, the set builder notation defines a set in the form

{formula for a typical element : objects to plug into the formula}.
A more complicated example is the set of perfect squares:
S = {n2 : n is an integer} = {0, 1, 4, 9, }.
The existence of several sets will be assumed. The simplest of these is the
empty set, which is the set with no elements. It is denoted as . The natural
numbers is the set N = {1, 2, 3, } consisting of the positive integers. The set
Z = { , 2, 1, 0, 1, 2, } is the set of all integers. = {n Z : n 0} =
{0, 1, 2, } is the nonnegative integers. Clearly, A, for any set A and
N Z.

Definition 1.1. Given any set A, the power set of A, written P(A), is the set
consisting of all subsets of A; i. e.,
P(A) = {B : B A}.
For example, P({a, b}) = {, {a}, {b}, {a, b}}. Also, for any set A, it is always
true that P(A) and A P(A). If a A, it is never true that a P(A), but it
is true that {a} P(A). Make sure you understand why!
An amusing example is P() = {}. (Dont confuse with {}! The former is
empty and the latter has one element.) Now, consider
P() = {}
P(P()) = {, {}}
P(P(P())) = {, {}, {{}}, {, {}}}
After continuing this n times, for some n N, the resulting set,
P(P( P() )),
is very large. In fact, since a set with k elements has 2k elements in its power set,
22
there are22 = 65, 536 elements after only five iterations of the example. Beyond
this, the numbers are too large to print. Number sequences such as this one are
sometimes called tetrations.

2. Algebra of Sets
Let A and B be sets. There are four common binary operations used on sets.1
The union of A and B is the set containing all the elements in either A or B:
A B = {x : x A x B}.
The intersection of A and B is the set containing the elements contained in
both A and B:
A B = {x : x A x B}.
1In the following, some logical notation is used. The symbol is the logical nonexclusive or.
The symbol is the logical and. Their truth tables are as follows:
T F T F
T T F T T T
F F F F T F

May 1, 2015 http://math.louisville.edu/lee/ira

2. ALGEBRA OF SETS 1-3

A B A B

A B A B

A\B AB

A B A B

Figure 1. These are Venn diagrams showing the four standard binary
operations on sets. In this figure, the set which results from the operation
is shaded.

The difference of A and B is the set of elements in A and not in B:

A \ B = {x : x A x
/ B}.

The symmetric difference of A and B is the set of elements in one of the sets,
but not the other:

AB = (A B) \ (A B).

Another common set operation is complementation. The complement of a set A

is usually thought of as the set consisting of all elements which are not in A. But,
a little thinking will convice you this is not a meaningful definition because the
collection of elements not in A is not a precisely understood collection. To make
sense of the complement of a set, there must be a well-defined universal set U which
contains all the sets in question. Then the complement of a set A U is Ac = U \ A.
It is usually the case that the universal set U is evident from the context in which
it is used.
With these operations, an extensive algebra for the manipulation of sets can
be developed. Its usually done hand in hand with formal logic because the two
subjects share much in common. These topics are studied as part of Boolean algebra.
Several examples of set algebra are given in the following theorem and its corollary.

Theorem 1.2. Let A, B and C be sets.

(a) A \ (B C) = (A \ B) (A \ C)
(b) A \ (B C) = (A \ B) (A \ C)

May 1, 2015 http://math.louisville.edu/lee/ira

1-4 CHAPTER 1. BASIC IDEAS

Proof. (a) This is proved as a sequence of equivalences.2

x A \ (B C) x A x
/ (B C)
x A x
/ Bx
/C
(x A x
/ B) (x A x
/ C)
x (A \ B) (A \ C)
(b) This is also proved as a sequence of equivalences.
x A \ (B C) x A x
/ (B C)
x A (x
/ Bx
/ C)
(x A x
/ B) (x A x
/ C)
x (A \ B) (A \ C)

Theorem 1.2 is a version of a group of set equations which are often called
DeMorgans Laws. The more usual statement of DeMorgans Laws is in Corollary 1.3.
Corollary 1.3 is an obvious consequence of Theorem 1.2 when there is a universal
set to make the complementation well-defined.
Corollary 1.3 (DeMorgans Laws). Let A and B be sets.
(a) (A B)c = Ac B c
(b) (A B)c = Ac B c

3. Indexed Sets
We often have occasion to work with large collections of sets. For example, we
could have a sequence of sets A1 , A2 , A3 , , where there is a set An associated
with each n N. In general, let be a set and suppose for each there is a set
A . The set {A : } is called a collection of sets indexed by . In this case,
is called the indexing set for the collection.
Example 1.1. For each n N, let An = {k Z : k 2 n}. Then
A1 = A2 =A3 = {1, 0, 1}, A4 = {2, 1, 0, 1, 2}, ,
A50 = {7, 6, 5, 4, 3, 2, 1, 0, 1, 2, 3, 4, 5, 6, 7},
is a collection of sets indexed by N.
Two of the basic binary operations can be extended to work with indexed
collections. In particular, using the indexed collection from the previous paragraph,
we define [
A = {x : x A for some }

and \
A = {x : x A for all }.

DeMorgans Laws can be generalized to indexed collections.
2The logical symbol is the same as if, and only if. If A and B are any two statements,
then A B is the same as saying A implies B and B implies A. It is also common to use iff
in this way.

May 1, 2015 http://math.louisville.edu/lee/ira

4. FUNCTIONS AND RELATIONS 1-5

Theorem 1.4. If {B : } is an indexed collection of sets and A is a set,

then [ \
A\ B = (A \ B )

and \ [
A\ B = (A \ B ).

Proof. The proof of this theorem is Exercise 1.3. 

4. Functions and Relations

4.1. Tuples. When listing the elements of a set, the order in which they are
listed is unimportant; e. g., {e, l, v, i, s} = {l, i, v, e, s}. If the order in which n items
are listed is important, the list is called an n-tuple. (Strictly speaking, an n-tuple is
not a set.) We denote an n-tuple by enclosing the ordered list in parentheses. For
example, if x1 , x2 , x3 , x4 are 4 items, the 4-tuple (x1 , x2 , x3 , x4 ) is different from the
4-tuple (x2 , x1 , x3 , x4 ).
Because they are used so often, 2-tuples are called ordered pairs and a 3-tuple
is called an ordered triple.
Definition 1.5. The Cartesian product of two sets A and B is the set of all
ordered pairs
A B = {(a, b) : a A b B}.
Example 1.2. If A = {a, b, c} and B = {1, 2}, then
A B = {(a, 1), (a, 2), (b, 1), (b, 2), (c, 1), (c, 2)}.
and
B A = {(1, a), (1, b), (1, c), (2, a), (2, b), (2, c)}.
Notice that A B =
6 B A because of the importance of order in the ordered pairs.
A useful way to visualize the Cartesian product of two sets is as a table. The
Cartesian product A B from Example 1.2 is listed as the entries of the following
table.
1 2
a (a, 1) (a, 2)
b (b, 1) (b, 2)
c (c, 1) (c, 2)
Of course, the common Cartesian plane from your analytic geometry course is
nothing more than a generalization of this idea of listing the elements of a Cartesian
product as a table.
The definition of Cartesian product can be extended to the case of more than
two sets. If {A1 , A2 , , An } are sets, then
A1 A2 An = {(a1 , a2 , , an ) : ak Ak for 1 k n}
is a set of n-tuples. This is often written as
n
Y
Ak = A1 A2 An .
k=1

May 1, 2015 http://math.louisville.edu/lee/ira

1-6 CHAPTER 1. BASIC IDEAS

4.2. Relations.
Definition 1.6. If A and B are sets, then any R A B is a relation from A
to B. If (a, b) R, we write aRb.
In this case,
dom (R) = {a : (a, b) R}
is the domain of R and
ran (R) = {b : (a, b) R}
is the range of R.
In the special case when R A A, for some set A, there is some additional
terminology.
R is symmetric, if aRb bRa.
R is reflexive, if aRa whenever a dom (A).
R is transitive, if aRb bRc = aRc.
R is an equivalence relation on A, if it is symmetric, reflexive and transitive.
Example 1.3. Let R be the relation on Z Z defined by aRb a b.
Then R is reflexive and transitive, but not symmetric.
Example 1.4. Let R be the relation on Z Z defined by aRb a < b.
Then R is transitive, but neither reflexive nor symmetric.
Example 1.5. Let R be the relation on Z Z defined by aRb a2 = b2 .
In this case, R is an equivalence relation. It is evident that aRb iff b = a or b = a.
4.3. Functions.
Definition 1.7. A relation R A B is a function if
aRb1 aRb2 = b1 = b2 .
If f A B is a function and dom (f ) = A, then we usually write f : A B
and use the usual notation f (a) = b instead of af b.
If f : A B is a function, the usual intuitive interpretation is to regard f
as a rule that associates each element of A with a unique element of B. Its not
necessarily the case that each element of B is associated with something from A;
i.e., B may not be ran (f ).
Example 1.6. Define f : N Z by f (n) = n2 and g : Z Z by g(n) = n2 .
In this case ran (f ) = {n2 : n N} and ran (g) = ran (f ) {0}. Notice that even
though f and g use the same formula, they are actually different functions.

Definition 1.8. If f : A B and g : B C, then the composition of g with

f is the function g f : A C defined by g f (a) = g(f (a)).
In Example 1.6, g f (n) = g(f (n)) = g(n2 ) = (n2 )2 = n4 makes sense for all
n N, but f g is undefined at n = 0.
There are several important types of functions.
Definition 1.9. A function f : A B is a constant function, if ran (f ) has a
single element; i. e., there is a b B such that f (a) = b for all a A. The function
f is surjective (or onto B), if ran (f ) = B.

May 1, 2015 http://math.louisville.edu/lee/ira

4. FUNCTIONS AND RELATIONS 1-7

b
c
a f

A B

b
c
a g

A B
g

Figure 2. These diagrams show two functions, f : A B and g : A

B. The function g is injective and f is not because f (a) = f (c).

In a sense, constant and surjective functions are the opposite extremes. A

constant function has the smallest possible range and a surjective function has the
largest possible range. Of course, a function f : A B can be both constant and
surjective, if B has only one element.
Definition 1.10. A function f : A B is injective (or one-to-one), if
f (a) = f (b) implies a = b.
The terminology one-to-one is very descriptive because such a function
uniquely pairs up the elements of its domain and range. An illustration of this
definition is in Figure 2. In Example 1.6, f is injective while g is not.
Definition 1.11. A function f : A B is bijective, if it is both surjective and
injective.
A bijective function can be visualized as uniquely pairing up all the elements of
A and B. Some authors, favoring less pretentious language, use the more descriptive
terminology one-to-one correspondence instead of bijection. This pairing up of the
elements from each set is like counting them and finding they have the same number
of elements. Given any two sets, no matter how many elements they have, the
intuitive idea is they have the same number of elements if, and only if, there is a
bijection between them.
The following theorem shows that this property of counting the number of
elements works in a familiar way. (Its proof is left as an easy exercise.)
Theorem 1.12. If f : A B and g : B C are bijections, then g f : A C
is a bijection.
4.4. Inverse Functions.

May 1, 2015 http://math.louisville.edu/lee/ira

1-8 CHAPTER 1. BASIC IDEAS

f
1
f

f 1

A f B

Figure 3. This is one way to visualize a general invertible function.

First f does something to a and then f 1 undoes it.

Definition 1.13. If f : A B, C A and D B, then the image of C is the

set f (C) = {f (a) : a C}. The inverse image of D is the set f 1 (D) = {a : f (a)
D}.
Definitions 1.11 and 1.13 work together in the following way. Suppose f : A B
is bijective and b B. The fact that f is surjective guarantees that f 1 ({b}) 6= .
Since f is injective, f 1 ({b}) contains only one element, say a, where f (a) = b. In
this way, it is seen that f 1 is a rule that assigns each element of B to exactly one
element of A; i. e., f 1 is a function with domain B and range A.
Definition 1.14. If f : A B is bijective, the inverse of f is the function
f 1 : B A with the property that f 1 f (a) = a for all a A and f f 1 (b) = b
for all b B.
There is some ambiguity in the meaning of f 1 between 1.13 and 1.14. The
former is an operation working with subsets of A and B; the latter is a function
working with elements of A and B. Its usually clear from the context which meaning
is being used.
Example 1.7. Let A = N and B be the even natural numbers. If f : A B
is f (n) = 2n and g : B A is g(n) = n/2, it is clear f is bijective. Since
f g(n) = f (n/2) = 2n/2 = n and g f (n) = g(2n) = 2n/2 = n, we see g = f 1 .
(Of course, it is also true that f = g 1 .)
Example 1.8. Let f : N Z be defined by
(
(n 1)/2, n odd,
f (n) =
n/2, n even
Its quite easy to see that f is bijective and
(
1 2n + 1, n 0,
f (n) =
2n, n<0
Given any set A, its obvious there is a bijection f : A A and, if g : A B is a
bijection, then so is g 1 : B A. Combining these observations with Theorem 1.12,
an easy theorem follows.
Theorem 1.15. Let S be a collection of sets. The relation on S defined by
A B there is a bijection f : A B
is an equivalence relation.

May 1, 2015 http://math.louisville.edu/lee/ira

4. FUNCTIONS AND RELATIONS 1-9

4.5. Schroder-Bernstein Theorem. The following theorem is a powerful

tool in set theory, and shows that a seemingly intuitively obvious statement is
sometimes difficult to verify. It will be used in Section 5.
Theorem 1.16 (Schroder-Bernstein3). Let A and B be sets. If there are injective
functions f : A B and g : B A, then there is a bijective function h : A B.

A1 A2 A3 A4 A5

B1 B2 B3 B4
f (A)
B5
B

Figure 4. Here are the first few steps from the construction used in
the proof of Theorem 1.16.

Proof. Let B1 = B\f (A). If Bk B is defined for some k N, let Ak = g(Bk )

and Bk+1 = f (Ak ). This inductively defines Ak and Bk for all k N. Use these
S
sets to define A = kN Ak and h : A B as
(
g 1 (x), x A
h(x) = .
f (x), x A \ A
It must be shown that h is well-defined, injective and surjective.
To show h is well-defined, let x A. If x A \ A, then it is clear h(x) = f (x) is
defined. On the other hand, if x A, then x Ak for some k. Since x Ak = g(Bk ),
we see h(x) = g 1 (x) is defined. Therefore, h is well-defined.
To show h is injective, let x, y A with x 6= y.
If both x, y A or x, y A \ A, then the assumptions that g and f are injective,
respectively, imply h(x) 6= h(y).
The remaining case is when x A and y A \ A. Suppose x Ak and
h(x) = h(y). If k = 1, then h(x) = g 1 (x) B1 and h(y) = f (y) f (A) = B \ B1 .
This is clearly incompatible with the assumption that h(x) = h(y). Now, suppose
k > 1. Then there is an x1 B1 such that
x = g f g f f g (x1 ).
| {z }
k 1 f s and k gs

This implies
h(x) = g 1 (x) = f g f f g (x1 ) = f (y)
| {z }
k 1 f s and k 1 gs

3
This is often called the Cantor-Schroder-Bernstein or Cantor-Bernstein Theorem, despite the
fact that it was apparently first proved by Richard Dedekind.

May 1, 2015 http://math.louisville.edu/lee/ira

1-10 CHAPTER 1. BASIC IDEAS

so that
y = g f g f f g (x1 ) Ak1 A.
| {z }
k 2 f s and k 1 gs
This contradiction shows that h(x) 6= h(y). We conclude h is injective.
To show h is surjective, let y B. If y Bk for some k, then h(Ak ) = g 1 (Ak ) =
Bk shows y h(A). If y / Bk for any k, y f (A) because B1 = B \ f (A), and
g(y)
/ A, so y = h(x) = f (x) for some x A. This shows h is surjective. 
The Schroder-Bernstein theorem has many consequences, some of which are at
first a bit unintuitive, such as the following theorem.
Corollary 1.17. There is a bijective function h : N N N
Proof. If f : N N N is f (n) = (n, 1), then f is clearly injective. On the
other hand, suppose g : N N N is defined by g((a, b)) = 2a 3b . The uniqueness
of prime factorizations guarantees g is injective. An application of Theorem 1.16
yields h. 
To appreciate the power of the Schroder-Bernstein theorem, try to find an
explicit bijection h : N N N.

5. Cardinality
There is a way to use sets and functions to formalize and generalize how we
count. For example, suppose we want to count how many elements are in the set
{a, b, c}. The natural way to do this is to point at each element in succession and
say one, two, three. What is really happening is that were defining a bijective
function between {a, b, c} and the set {1, 2, 3}. This idea can be generalized.
Definition 1.18. Given n N, the set n = {1, 2, , n} is called an initial
segment of N. The trivial initial segment is 0 = . A set S has cardinality n, if
there is a bijective function f : S n. In this case, we write card (S) = n.
The cardinalities defined in Definition 1.18 are called the finite cardinal numbers.
They correspond to the everyday counting numbers we usually use. The idea can
be generalized still further.
Definition 1.19. Let A and B be two sets. If there is an injective function
f : A B, we say card (A) card (B).
According to Theorem 1.16, the Schroder-Bernstein Theorem, if card (A)
card (B) and card (B) card (A), then there is a bijective function f : A B. As
expected, in this case we write card (A) = card (B). When card (A) card (B), but
no such bijection exists, we write card (A) < card (B). Theorem 1.15 shows that
card (A) = card (B) is an equivalence relation between sets.
The idea here, of course, is that card (A) = card (B) means A and B have the
same number of elements and card (A) < card (B) means A is a smaller set than B.
This simple intuitive understanding has some surprising consequences when the sets
involved do not have finite cardinality.
In particular, the set A is countably infinite, if card (A) = card (N). In this case,
it is common to write card (N) = 0 .4 When card (A) 0 , then A is said to be a
4The symbol is the Hebrew letter aleph and is usually pronounced aleph nought.
0

May 1, 2015 http://math.louisville.edu/lee/ira

5. CARDINALITY 1-11

countable set. In other words, the countable sets are those having finite or countably
infinite cardinality.
Example 1.9. Let f : N Z be defined as
(
n+1
f (n) = 2 , when n is odd
.
n
1 2 , when n is even
Its easy to show f is a bijection, so card (N) = card (Z) = 0 .
Theorem 1.20. Suppose A and B are countable sets.
(a) A B is countable.
(b) A B is countable.
Proof. (a) This is a consequence of Theorem 1.17.
(b) This is Exercise 1.20. 
An alert reader will have noticed from previous examples that
0 = card (Z) = card () = card (N) = card (N N) = card (N N N) =
A logical question is whether all sets either have finite cardinality, or are countably
infinite. That this is not so is seen by letting S = N in the following theorem.
Theorem 1.21. If S is a set, card (S) < card (P(S)).
Proof. Noting that 0 = card () < 1 = card (P()), the theorem is true when
S is empty.
Suppose S 6= . Since {a} P(S) for all a S, it follows that card (S)
card (P(S)). Therefore, it suffices to prove there is no surjective function f : S
P(S).
To see this, assume there is such a function f and let T = {x S : x / f (x)}.
Since f is surjective, there is a t S such that f (t) = T . Either t T or t / T.
If t T = f (t), then the definition of T implies t / T , a contradiction. On
the other hand, if t / T = f (t), then the definition of T implies t T , another
contradiction. These contradictions lead to the conclusion that no such function f
can exist. 
A set S is said to be uncountably infinite, or just uncountable, if 0 < card (S).
Theorem 1.21 implies 0 < card (P(N)), so P(N) is uncountable. In fact, the same
argument implies
0 = card (N) < card (P(N)) < card (P(P(N))) <
So, there are an infinite number of distinct infinite cardinalities.
In 1874 Georg Cantor [5] proved card (R) = card (P(N)) and card (R) > 0 ,
where R is the set of real numbers. (A version of Cantors theorem appears in
Theorem 2.25 below.) This naturally led to the question whether there are sets S
such that 0 < card (S) < card (R). Cantor spent many years trying answer this
question and never succeeded. His assumption that no such sets exist came to be
called the continuum hypothesis.
The importance of the continuum hypothesis was highlighted by David Hilbert
at the 1900 International Congress of Mathematicians in Paris, when he put it first
on his famous list of the 23 most important open problems in mathematics. Kurt
Godel proved in 1940 that the continuum hypothesis cannot be disproved using

May 1, 2015 http://math.louisville.edu/lee/ira

1-12 CHAPTER 1. BASIC IDEAS

standard set theory, but he did not prove it was true. In 1963 it was proved by
Paul Cohen that the continuum hypothesis is actually unprovable as a theorem in
standard set theory.
So, the continuum hypothesis is a statement with the strange property that it
is neither true nor false within the framework of ordinary set theory. This means
that in the standard axiomatic development of set theory, the continuum hypothesis,
or a suitable negation of it, can be taken as an additional axiom without causing
any contradictions. The technical terminology is that the continuum hypothesis is
independent of the axioms of set theory.
The proofs of these theorems are extremely difficult and entire broad areas of
mathematics were invented just to make their proofs possible. Even today, there
are some deep philosophical questions swirling around them. A more technical
introduction to many of these ideas is contained in the book by Ciesielski [8].
A nontechnical and very readable history of the efforts by mathematicians to
understand the continuum hypothesis is the book by Aczel [1].

6. Exercises

1.1. If a set S has n elements for n , then how many elements are in P(S)?

1.2. Prove that for any sets A and B,

(a) A = (A B) (A \ B)
(b) A B = (A \ B) (B \ A) (A B) and that the sets A \ B, B \ A and
A B are pairwise disjoint.
(c) A \ B = A B c .

1.3. Prove Theorem 1.4.

1.4. For any sets A, B, C and D,
(A B) (C D) = (A C) (B D)
and
(A B) (C D) (A C) (B D).
Why does equality not hold in the second expression?

1.5. Prove Theorem 1.15.

1.6. Suppose R is an equivalence relation on A. For each x A define Cx =
{y A : xRy}. Prove that if x, y A, then either Cx = Cy or Cx Cy = . (The
collection {Cx : x A} is the set of equivalence classes induced by R.)

1.7. If f : A B and g : B C are bijections, then so is g f : A C.

1.8. Prove or give a counter example: f : X Y is injective iff whenever A and

B are disjoint subsets of Y , then f 1 (A) f 1 (B) = .

1.9. If f : A B is bijective, then f 1 is unique.

1.10. Prove that f : X Y is surjective iff for each subset A X, Y \ f (A)

f (X \ A).

May 1, 2015 http://math.louisville.edu/lee/ira

6. EXERCISES 1-13

1.11. Suppose that Ak is a set for each positive integer k.

T S
(a) Show that x Sn=1 (Tk=n Ak ) iff x Ak for infinitely many sets Ak .

(b) Show that x n=1 ( k=n Ak ) iff x Ak for all but finitely many of the
sets Ak .
T S
The S
set n=1T( k=n Ak ) from (a) is often called the superior limit of the sets Ak

and n=1 ( k=n Ak ) is often called the inferior limit of the sets Ak .
B
1.12. Given two sets A and B, it is common to let  A  denote the set of all
A
functions f : B A. Prove that for any set A, card 2 = card (P(A)). This is
why many authors use 2A as their notation for P(A).

1.13. Let S be a set. Prove the following two statements are equivalent:
(a) S is infinite; and,
(b) there is a proper subset T of S and a bijection f : S T .
This statement is often used as the definition of when a set is infinite.
1.14. If S is an infinite set, then there is a countably infinite
S collection of nonempty
pairwise disjoint infinite sets Tn , n N such that S = nN Tn .

1.15. Without using the Schroder-Bernstein theorem, find a bijection f : [0, 1]

(0, 1).

1.16. If f : [0, ) (0, ) and g : (0, ) [0, ) are given by f (x) = x + 1 and
g(x) = x, then the proof of the Schroder-Bernstein theorem yields what bijection
h : [0, ) (0, )?

1.17. Find a function f : R \ {0} R \ {0} such that f 1 = 1/f .

1.18. Find a bijection f : [0, ) (0, ).

1.19. If f : A B and g : B A are functions such that f g(x) = x for all

x B and g f (x) = x for all x A, then f 1 = g.

1.20. If A and B are sets such that card (A) = card (B) = 0 , then card (A B) =
0 .

1.21. Using the notation from the proof of the Schroder-Bernstein Theorem, let
A = [0, ), B = (0, ), f (x) = x + 1 and g(x) = x. Determine h(x).

1.22. Using the notation from the proof of the Schroder-Bernstein Theorem, let
A = N, B = Z, f (n) = n and
(
1 3n, n 0
g(n) = .
3n 1, n > 0
Calculate h(6) and h(7).

1.23. Suppose that in the statement of the Schroder-Bernstein theorem A = B = Z

and f (n) = g(n) = 2n. Following the procedure in the proof yields what function h?

May 1, 2015 http://math.louisville.edu/lee/ira

1-14 CHAPTER 1. BASIC IDEAS

S
1.24. If {An : n N} is a collection of countable sets, then nN An is countable.

1.25. If A and B are sets, the set  ofall functions f : A B is often denoted by
S
B A . If S is a set, prove that card 2 = card (P(S)).

1.26. If 0 card (S)), then there is an injective function f : S S that is not

surjective.

1.27. If card (S) = 0 , then there is a sequence

S of pairwise disjoint sets Tn , n N
such that card (Tn ) = 0 for every n N and nN Tn = S.

May 1, 2015 http://math.louisville.edu/lee/ira

 CHAPTER 2

The Real Numbers

This chapter concerns what can be thought of as the rules of the game: the
axioms of the real numbers. These axioms imply all the properties of the real
numbers and, in a sense, any set satisfying them is uniquely determined to be the
real numbers.
The axioms are presented here as rules without very much justification. Other
approaches can be used. For example, a common approach is to begin with the
Peano axioms the axioms of the natural numbers and build up to the real
numbers through several completions of the natural numbers. Its also possible to
begin with the axioms of set theory to build up the Peano axioms as theorems and
then use those to prove our axioms as further theorems. No matter how its done,
there are always some axioms at the base of the structure and the rules for the real
numbers are the same, whether theyre axioms or theorems.
We choose to start at the top because the other approaches quickly turn into
a long and tedious labyrinth of technical exercises without much connection to
analysis.

1. The Field Axioms

These first six axioms are called the field axioms because any object satisfying
them is called a field. They give the algebraic properties of the real numbers.
A field is a nonempty set F along with two binary operations, multiplication
: F F F and addition + : F F F satisfying the following axioms.1
Axiom 1 (Associative Laws). If a, b, c F, then (a + b) + c = a + (b + c) and
(a b) c = a (b c).

Axiom 2 (Commutative Laws). If a, b F, then a + b = b + a and a b = b a.

Axiom 3 (Distributive Law). If a, b, c F, then a (b + c) = (a b) + (a c).

Axiom 4 (Existence of identities). There are 0, 1 F such that a + 0 = a and

a 1 = a, for all a F.

Axiom 5 (Existence of an additive inverse). For each a F there is a F

such that a + (a) = 0.

1Given a set A, a function f : A A A is called a binary operation. In other words, a

binary operation is just a function with two arguments. The standard notations of +(a, b) = a + b
and (a, b) = a b are used here. The symbol is unfortunately used for both the Cartesian
product and the field operation, but the context in which its used removes the ambiguity.

2-1
2-2 CHAPTER 2. THE REAL NUMBERS

Axiom 6 (Existence of a multiplicative inverse). For each a F \ {0} there is

a1 F such that a a1 = 1.
Although these axioms seem to contain most properties of the real numbers we
normally use, they dont characterize the real numbers; they just give the rules for
arithmetic. There are other fields besides the real numbers and studying them is a
large part of most abstract algebra courses.
Example 2.1. From elementary algebra we know that the rational numbers
Q = {p/q : p Z q N}

form a field. It is shown in Theorem 2.12 that 2 / Q, so Q doesnt contain all the
real numbers.
Example 2.2. Let F = {0, 1, 2} with addition and multiplication calculated
modulo 3. The addition and multiplication tables are as follows.
+ 0 1 2 0 1 2
0 0 1 2 0 0 0 0
1 1 2 0 1 0 1 2
2 2 0 1 2 0 2 1
It is easy to check that the field axioms are satisfied. This field is usually called Z3 .
The following theorems, containing just a few useful properties of fields, are
presented mostly as examples showing how the axioms are used. More complete
developments can be found in any beginning abstract algebra text.
Theorem 2.1. The additive and multiplicative identities of a field F are unique.
Proof. Suppose e1 and e2 are both multiplicative identities in F. Then
e1 = e1 e2 = e2 ,
so the multiplicative identity is unique. The proof for the additive identity is
essentially the same. 

Theorem 2.2. Let F be a field. If a, b F with b 6= 0, then a and b1 are

unique.
Proof. Suppose b1 and b2 are both multiplicative inverses for b 6= 0. Then,
using Axioms 4 and 1,
b1 = b1 1 = b1 (b b2 ) = (b1 b) b2 = 1 b2 = b2 .
This shows the multiplicative inverse in unique. The proof is essentially the same
for the additive inverse. 

There are many other properties of fields which could be proved here, but they
correspond to the usual properties of the real numbers learned in beginning algebra,
so we omit them. Some of them are in the exercises at the end of this chapter.
From now on, the standard notations for algebra will usually be used; e. g., we
will allow ab instead of a b and a/b instead of a b1 . The reader may also use
the standard facts she learned from elementary algebra.

May 1, 2015 http://math.louisville.edu/lee/ira

2. THE ORDER AXIOM 2-3

2. The Order Axiom

The axiom of this section gives the order and metric properties of the real
numbers. In a sense, the following axiom adds some geometry to a field.
Axiom 7 (Order axiom.). There is a set P F such that
(a) If a, b P , then a + b, ab P .
(b) If a F, then exactly one of the following is true: a P , a P or
a = 0.
Any field F satisfying the axioms so far listed is naturally called an ordered field.
Of course, the set P is known as the set of positive elements of F. Using Axiom 7(ii),
we see that F is divided into three pairwise disjoint sets: P , {0} and {x : x P }.
The latter of these is, of course, the set of negative elements from F. The following
definition introduces familiar notation for order.
Definition 2.3. We write a < b or b > a, if b a P . The meanings of a b
and b a are now as expected.
Notice that a > 0 a = a 0 P and a < 0 a = 0 a P , so
a > 0 and a < 0 agree with our usual notions of positive and negative.
Our goal is to capture all the properties of the real numbers with the axioms.
The order axiom eliminates many fields from consideration. For example, Exercise
2.5 shows the field Z3 of Example 2.2 is not an ordered field. On the other hand,
facts from elementary algebra imply Q is an ordered field, so the first seven axioms
still dont capture the real numbers.
Following are a few standard properties of ordered fields.
Theorem 2.4. Let F be an ordered field and a F. a 6= 0 iff a2 > 0.
Proof. () If a > 0, then a2 > 0 by Axiom 7(a). If a < 0, then a > 0 by
Axiom 7(b) and a2 = 1a2 = (1)(1)a2 = (a)2 > 0.
() Since 02 = 0, this is obvious. 
Theorem 2.5. If F is an ordered field and a, b, c F, then
(a) a < b a + c < b + c,
(b) a < b b < c = a < c,
(c) a < b c > 0 = ac < bc,
(d) a < b c < 0 = ac > bc.
Proof. (a) a < b b a P (b + c) (a + c) P a + c < b + c.
(b) By supposition, both b a, c b P . Using the fact that P is closed under
addition, we see (b a) + (c b) = c a P . Therefore, c > a.
(c) Since b a P and c P and P is closed under multiplication, c(b a) =
cb ca P and, therefore, ac < bc.
(d) By assumption, b a, c P . Apply part (c) and Exercise 2.1. 
Theorem 2.6 (Two Out of Three Rule). Let F be an ordered field and a, b, c F.
If ab = c and any two of a, b or c are positive, then so is the third.
Proof. If a > 0 and b > 0, then Axiom 7(a) implies c > 0. Next, suppose
a > 0 and c > 0. In order to force a contradiction, suppose b 0. In this case,
Axiom 7(b) shows
0 a(b) = (ab) = c < 0,
which is impossible. 

May 1, 2015 http://math.louisville.edu/lee/ira

2-4 CHAPTER 2. THE REAL NUMBERS

Corollary 2.7. Let F be an ordered field and a F. If a > 0, then a1 > 0.

If a < 0, then a1 < 0.
Proof. The proof is Exercise 2.2. 
Suppose a > 0. Since 1a = a, Theorem 2.6 implies 1 > 0. Applying Theorem
2.5, we see that 1 + 1 > 1 > 0. Its clear that by induction, we can find a copy of N
embedded in any ordered field. Similarly, Z and Q also have unique copies in any
ordered field.
The standard notation for intervals will be used on an ordered field, F; i. e.,
(a, b) = {x F : a < x < b}, (a, ) = {x F : a < x}, [a, b] = {x F : a x b},
etc. Its also useful to start visualizing F as a standard number line.
2.1. Metric Properties. The order axiom on a field F allows us to introduce
the idea of the distance between points in F. To do this, we begin with the following
familiar definition.
Definition 2.8. Let F be an ordered field. The absolute value function on F is
a function | | : F F defined as
(
x, x0
|x| = .
x, x < 0
The most important properties of the absolute value function are contained in
the following theorem.
Theorem 2.9. Let F be an ordered field and x, y F. Then
(a) |x| 0 and |x| = 0 x = 0;
(b) |x| = | x|;
(c) |x| x |x|;
(d) |x| y y x y; and,
(e) |x + y| |x| + |y|.
Proof. (a) The fact that |x| 0 for all x F follows from Axiom 7(b).
Since 0 = 0, the second part is clear.
(b) If x 0, then x 0 so that | x| = (x) = x = |x|. If x < 0, then
x > 0 and |x| = x = | x|.
(c) If x 0, then |x| = x x = |x|. If x < 0, then |x| = (x) = x <
x = |x|.
(d) This is left as Exercise 2.3.
(e) Add the two sets of inequalities |x| x |x| and |y| y |y| to see
(|x| + |y|) x + y |x| + |y|. Now apply (d).

From studying analytic geometry and calculus, we are used to thinking of |x y|
as the distance between the numbers x and y. This notion of a distance between
two points of a set can be generalized.
Definition 2.10. Let S be a set and d : S S R satisfy
(a) for all x, y S, d(x, y) 0 and d(x, y) = 0 x = y,
(b) for all x, y S, d(x, y) = d(y, x), and
(c) for all x, y, z S, d(x, z) d(x, y) + d(y, z).
Then the function d is a metric on S. The pair (S, d) is called a metric space.

May 1, 2015 http://math.louisville.edu/lee/ira

3. THE COMPLETENESS AXIOM 2-5

A metric is a function which defines the distance between any two points of a
set.
Example 2.3. Let S be a set and define d : S S S by
(
1, x 6= y
d(x, y) = .
0, x = y
It can readily be verified that d is a metric on S. This simplest of all metrics is
called the discrete metric and it can be defined on any set. Its not often useful.
Theorem 2.11. If F is an ordered field, then d(x, y) = |x y| is a metric on F.
Proof. Use parts (a), (b) and (e) of Theorem 2.9. 
The metric on F derived from the absolute value function is called the standard
metric on F. There are other metrics sometimes defined for specialized purposes,
but we wont have need of them.

3. The Completeness Axiom

All the axioms given so far are obvious from beginning algebra, and, on the
surface, its not obvious they havent captured all the properties of the real numbers.
Since Q satisfies them all, the following theorem shows that were not yet done.
Theorem 2.12. There is no Q such that 2 = 2.
Proof. Assume to the contrary that there is Q with 2 = 2. Then there
are p, q N such that = p/q with p and q relatively prime. Now,
 2
p
(4) = 2 = p2 = 2q 2
q
shows p2 is even. Since the square of an odd number is odd, p must be even; i. e.,
p = 2r for some r N. Substituting this into (4), shows 2r2 = q 2 . The same
argument as above establishes q is also even. This contradicts the assumption that
p and q are relatively prime. Therefore, no such exists. 

Since we suspect 2 is a perfectly fine number, theres still something missing
from the list of axioms. Completeness is the missing idea.
The Completeness Axiom is somewhat more complicated than the previous
axioms, and several definitions are needed in order to state it.
3.1. Bounded Sets.
Definition 2.13. A subset S of an ordered field F is bounded above, if there
exists M F such that M x for all x S. A subset S of an ordered field F is
bounded below, if there exists m F such that m x for all x S. The elements
M and m are called an upper bound and lower bound for S, respectively. If S is
bounded both above and below, it is a bounded set.
There is no requirement in the definition that the upper and lower bounds for a
set are elements of the set. They can be elements of the set, but typically are not.
For example, if N = (, 0), then [0, ) is the set of all upper bounds for N , but
none of them is in N . On the other hand, if T = (, 0], then [0, ) is again the
set of all upper bounds for T , but in this case 0 is an upper bound which is also an

May 1, 2015 http://math.louisville.edu/lee/ira

2-6 CHAPTER 2. THE REAL NUMBERS

element of T . An extreme case is . A vacuous argument shows every element of F

is both an upper and lower bound, but obviously none of them is in .
A set need not have upper or lower bounds. For example N = (, 0) has no
lower bounds, while P = (0, ) has no upper bounds. The integers, Z, has neither
upper nor lower bounds. If S has no upper bound, it is unbounded above and, if
it has no lower bound, then it is unbounded below. In either case, it is usually just
said to be unbounded.
If M is an upper bound for the set S, then every x M is also an upper bound
for S. Considering some simple examples should lead you to suspect that among
the upper bounds for a set, there is one that is best in the sense that everything
greater is an upper bound and everything less is not an upper bound. This is the
basic idea of completeness.
Definition 2.14. Suppose F is an ordered field and S is bounded above in F.
A number B F is called a least upper bound of S if
(a) B is an upper bound for S, and
(b) if is any upper bound for S, then B .
If S is bounded below in F, then a number b F is called a greatest lower bound of
S if
(a) b is a lower bound for S, and
(b) if is any lower bound for S, then b .
Theorem 2.15. If F is an ordered field and A F is nonempty, then A has at
most one least upper bound and at most one greatest lower bound.
Proof. Suppose u1 and u2 are both least upper bounds for A. Since u1 and
u2 are both upper bounds for A, u1 u2 u1 = u1 = u2 . The proof of the
other case is similar. 

Definition 2.16. If A F is nonempty and bounded above, then the least

upper bound of A is written lub A. When A is not bounded above, we write
lub A = . When A = , then lub A = .
If A F is nonempty and bounded below, then the greatest lower of A is
written glb A. When A is not bounded below, we write glb A = . When A = ,
then glb A = .2
Notice that the symbol is not an element of F. Writing lub A = is just
a convenient way to say A has no upper bounds. Similarly lub = tells us
has every real number as an upper bound.
Theorem 2.17. Let A F and F. = lub A iff (, ) A = and for
all > 0, ( , ] A 6= . Similarly, = glb A iff (, ) A = and for all
> 0, [, + ) A 6= .
Proof. We will prove the first statement, concerning the least upper bound.
The second statement, concerning the greatest lower bound, follows similarly.
() If x (, ) A, then cannot be an upper bound of A, which is a
contradiction. If there is an > 0 such that ( , ] A = , then from above, we

2Some people prefer the notation sup A and inf A instead of lub A and glb A, respectively.
They stand for the supremum and infimum of A.

May 1, 2015 http://math.louisville.edu/lee/ira

3. THE COMPLETENESS AXIOM 2-7

conclude ( , ) A = . This implies /2 is an upper bound for A which

is less than = lub A. This contradiction shows ( , ] A 6= .
() The assumption that (, ) A = implies lub A. On the other hand,
suppose lub A < . By assumption, there is an x (lub A, ) A. This is clearly a
contradiction, since lub A < x A. Therefore, = lub A. 
An eagle-eyed reader may wonder why the intervals in Theorem 2.17 are (, ]
and [, + ) instead of ( , ) and (, + ). Just consider the case A = {}
to see that the theorem fails when the intervals are open. When lub A / A or
glb A
/ A, the intervals can be open, as shown in the following corollary.
Corollary 2.18. If A is bounded above and = lub A / A, then for all > 0,
( , ) A is an infinite set. Similarly, if A is bounded below and = glb A / A,
then for all > 0, (, + ) A is an infinite set.
Proof. Let > 0. According to Theorem 2.17, there is an x1 ( , ] A.
By assumption, x1 < . We continue by induction. Suppose n N and xn has
been chosen to satisfy xn ( , ) A. Using Theorem 2.17 as before to
choose xn+1 (xn , ) A. The set {xn : n N} is infinite and contained in
( , ) A. 
When F = Q, Theorem 2.12 shows there is no least upper bound for A = {x :
x2 < 2} in Q. In a sense, Q has a hole where this least upper bound should be.
Adding the following completeness axiom enlarges Q to fill in the holes.
Axiom 8 (Completeness). Every nonempty set which is bounded above has a
least upper bound.
This is the final axiom. Any field F satisfying all eight axioms is called a
complete ordered field. We assume the existence of a complete ordered field, R,
called the real numbers.
In naive set theory it can be shown that if F1 and F2 are both complete
ordered fields, then they are the same, in the following sense. There exists a unique
bijective function i : F1 F2 such that i(a + b) = i(a) + i(b), i(ab) = i(a)i(b) and
a < b i(a) < i(b). Such a function i is called an order isomorphism. The
existence of such an order isomorphism shows that R is essentially unique. More
reading on this topic can be done in some advanced texts [10, 11].
Every statement about upper bounds has a dual statement about lower bounds.
A proof of the following dual to Axiom 8 is left as an exercise.
Corollary 2.19. Every nonempty set which is bounded below has a greatest
lower bound.
In Section 4 it will be shown that there is an x R satisfying x2 = 2. This will
show R removes the deficiency of Q highlighted by Theorem 2.12. The Completeness
Axiom plugs up the holes in Q.
3.2. Some Consequences of Completeness. The property of completeness
is what separates analysis from geometry and algebra. It requires the use of approxi-
mation, infinity and more dynamic visualizations than algebra or classical geometry.
The rest of this course is largely concerned with applications of completeness.
Theorem 2.20 (Archimedean Principle ). If a R, then there exists na N
such that na > a.

May 1, 2015 http://math.louisville.edu/lee/ira

2-8 CHAPTER 2. THE REAL NUMBERS

Proof. If the theorem is false, then a is an upper bound for N. Let = lub N.
According to Theorem 2.17 there is an m N such that m > 1. But, this is a
contradiction because = lub N < m + 1 N. 
Some other variations on this theme are in the following corollaries.
Corollary 2.21. Let a, b R with a > 0.
(a) There is an n N such that an > b.
(b) There is an n N such that 0 < 1/n < a.
(c) There is an n N such that n 1 a < n.
Proof. (a) Use Theorem 2.20 to find n N where n > b/a.
(b) Let b = 1 in part (a).
(c) Theorem 2.20 guarantees that S = {n N : n > a} 6= . If n is the least
element of this set, then n 1
/ S and n 1 a < n. 
Corollary 2.22. If I is any interval from R, then I Q 6= and I Qc 6= .
Proof. Left as an exercise. 
A subset of R which intersects every interval is said to be dense in R. Corollary
2.22 shows both the rational and irrational numbers are dense.

4. Comparisons of Q and R
All of the above still does not establish that Q is different from R. In Theorem
2.12, it was shown that the equation x2 = 2 has no solution in Q. The following
theorem shows x2 = 2 does have solutions in R. Since a copy of Q is embedded in
R, it follows, in a sense, that R is bigger than Q.
Theorem 2.23. There is a positive R such that 2 = 2.
Proof. Let S = {x > 0 : x2 < 2}. Then 1 S, so S 6= . If x 2, then
Theorem 2.5(c) implies x2 4 > 2, so S is bounded above. Let = lub S. It will
be shown that 2 = 2.
Suppose first that 2 < 2. This assumption implies (2 2 )/(2 + 1) > 0.
According to Corollary 2.21, there is an n N large enough so that
1 2 2 2 + 1
0< < = 0 < < 2 2 .
n 2 + 1 n
Therefore,
 2  
1 2 1 1 1
+ = 2 +
+ 2 = 2 + 2 +
n n n n n
(2 + 1)
< 2 + < 2 + (2 2 ) = 2
n
contradicts the fact that = lub S. Therefore, 2 2.
Next, assume 2 > 2. In this case, choose n N so that
1 2 2 2
0< < = 0 < < 2 2.
n 2 n
Then
 2
1 2 1 2
= 2 + 2 > 2 > 2 (2 2) = 2,
n n n n

May 1, 2015 http://math.louisville.edu/lee/ira

4. COMPARISONS OF Q AND R 2-9

1 = .1 (1) 1 (2) 1 (3) 1 (4) 1 (5) . . .

2 = .2 (1) 2 (2) 2 (3) 2 (4) 2 (5) . . .
3 = .3 (1) 3 (2) 3 (3) 3 (4) 3 (5) . . .
4 = .4 (1) 4 (2) 4 (3) 4 (4) 4 (5) . . .
5 = .5 (1) 5 (2) 5 (3) 5 (4) 5 (5) . . .
.. .. .. .. .. ..
. . . . . .

Figure 1. The proof of Theorem 2.25 is called the diagonal argument

because it constructs a new number z by working down the main diagonal
of the array shown above, making sure z(n) 6= n (n) for each n N.

again contradicts that = lub S.

Therefore, 2 = 2. 
Theorem 2.12 leads to the obvious question of how much bigger R is than Q.
First, note that since N Q, it is clear that card (Q) 0 . On the other hand,
every q Q has a unique reduced fractional representation q = m(q)/n(q) with
m(q) Z and n(q) N. This gives an injective function f : Q Z N defined by
f (q) = (m(q), n(q)), and we conclude card (Q) card (Z N) = 0 . The following
theorem ensues.
Theorem 2.24. card (Q) = 0 .
In 1874, Georg Cantor first showed that R is not countable. The following proof
is his famous diagonal argument from 1891.
Theorem 2.25. card (R) > 0

Proof. It suffices to prove that card ([0, 1]) > 0 . If this is not true, then there
is a bijection : N [0, 1]; i.e.,
(5) [0, 1] = {n : n N}.
P
Each x [0, 1] can be written in the decimal form x = n=1 x(n)/10n where
x(n) {0, 1, 2, 3, 4, 5, 6, 7, 8, 9} for each n N. This decimal representation is not
necessarily unique. For example,

1 5 4 X 9
= = + .
2 10 10 n=2 10n
In such a case, there is a choice of x(n) so it is constantly 9 or constantly 0 from
some N onward. When given a choice, we will always opt to end the number with a
string of nines. With this convention, the decimal representation of x is unique.
Define
Pz [0, 1] by choosing z(n) {d : d 8} such that z(n) 6= n (n).
Let z = n=1 z(n)/10n . Since z [0, 1], there is an n N such that z = (n).
But, this is impossible because z(n) differs from n in the nth decimal place. This
contradiction shows card ([0, 1]) > 0 . 
Around the turn of the twentieth century these then-new ideas about infinite
sets were very controversial in mathematics. This is because some of these ideas are
very unintuitive. For example, the rational numbers are a countable set and the
irrational numbers are uncountable, yet between every two rational numbers are an

May 1, 2015 http://math.louisville.edu/lee/ira

2-10 CHAPTER 2. THE REAL NUMBERS

uncountable number of irrational numbers and between every two irrational numbers
there are a countably infinite number of rational numbers. It would seem there are
either too few or too many gaps in the sets to make this possible. Such a seemingly
paradoxical situation flies in the face of our intuition, which was developed with
finite sets in mind.
This brings us back to the discussion of cardinalities and the Continuum
Hypothesis at the end of Section 5. Most of the time, people working in real
analysis assume the Continuum Hypothesis is true. With this assumption and
Theorem 2.25 it follows that whenever A R, then either card (A) 0 or
card (A) = card (R) = card (P(N)).3 Since P(N) has many more elements than N,
any countable subset of R is considered to be a small set, in the sense of cardinality,
even if it is infinite. This works against the intuition of many beginning students
who are not used to thinking of Q, or any other infinite set as being small. But it
turns out to be quite useful because the fact that the union of a countably infinite
number of countable sets is still countable can be exploited in many ways.4
In later chapters, other useful small versus large dichotomies will be found.

5. Exercises

2.1. Prove that if a, b F, where F is a field, then (a)b = (ab) = a(b).

2.2. Prove Corollary 2.7. If a > 0, then so is a1 . If a < 0, then so is a1 .

2.3. Prove |x| y iff y x y.

2.4. If S R is bounded above, then

lub S = glb {x : x is an upper bound for S}.

2.5. Prove there is no set P Z3 which makes Z3 into an ordered field.

2.6. If is an upper bound for S and S, then = lub S.

2.7. Let A and B be subsets of R that are bounded above. Define A + B = {a + b :

a A b B}. Prove that lub (A + B) = lub A + lub B.

2.8. If A Z is bounded below, then A has a least element.

2.9. If F is an ordered field and a F such that 0 a < for every > 0, then
a = 0.
2.10. Let x R. Prove |x| < for all > 0 iff x = 0.

2.11. If p is a prime number, then the equation x2 = p has no rational solutions.

2.12. If p is a prime number and > 0, then there are x, y Q such that
x2 < p < y 2 < x2 + .
3Since is the smallest infinite cardinal, is used to denote the smallest uncountable
0 1
cardinal. You will also see card (R) = c, where c is the old-style German letter c, standing for the
cardinality of the continuum. Assuming the continuum hypothesis, it follows that 0 < 1 = c.
4See Problem 24 on page 1-13.

May 1, 2015 http://math.louisville.edu/lee/ira

5. EXERCISES 2-11

2.13. If a < b, then (a, b) Q 6= .

2.14. If q Q and a R \ Q, then q + a R \ Q. Moreover, if q 6= 0, then

aq R \ Q.

2.15. Prove that if a < b, then there is a q Q such that a < 2q < b.

2.16. Prove Corollary 2.22.

2.17. If F is an ordered field and x1 , x2 , . . . , xn F for some n N, then

 n  n
X  X
(8) xi  |xi |.
 

 
i=1 i=1

2.18. Prove Corollary 2.19.

2.19. Prove card (Qc ) = c.

2.20. If A R and B = {x : x is an upper bound for A}, then lub (A) = glb (B).

May 1, 2015 http://math.louisville.edu/lee/ira

 CHAPTER 3

Sequences

1. Basic Properties
Definition 3.1. A sequence is a function a : N R.
Instead of using the standard function notation of a(n) for sequences, it is
usually more convenient to write the argument of the function as a subscript, an .
Example 3.1. Let the sequence an = 1 1/n. The first three elements are
a1 = 0, a2 = 1/2, a3 = 2/3, etc.
Example 3.2. Let the sequence bn = 2n . Then b1 = 2, b2 = 4, b3 = 8, etc.

Example 3.3. If a and r are constants, then a sequence given by c1 = a,

c2 = ar, c3 = ar2 and in general cn = arn1 is called a geometric sequence. The
number r is called the ratio of the sequence. Staying away from the trivial cases
where a = 0 or r = 0, a geometric sequence can always be recognized by noticing
that aan+1
n
= r for all n N. Example 3.2 is a geometric sequence with a = r = 2.
Example 3.4. If a and d are constants, then a sequence of the form dn =
a + (n 1)d is called an arithmetic sequence. Another way of looking at this is that
dn is an arithmetic sequence if dn+1 dn = d for all n N.
Example 3.5. Some sequences are not defined by an explicit formula, but are
defined recursively. This is an inductive method of definition in which successive
terms of the sequence are defined by using other terms of the sequence. The most
famous of these is the Fibonacci sequence. To define the Fibonacci sequence, fn ,
let f1 = 0, f2 = 1 and for n > 2, let fn = fn2 + fn1 . The first few terms are
0, 1, 1, 2, 3, 5, 8, . . . . There actually is a simple formula that directly gives fn , but
we leave its derivation as Exercise 3.5.
Its often inconvenient for the domain of a sequence to be N, as required by
Definition 3.1. For example, the sequence beginning 1, 2, 4, 8, . . . can be written
20 , 21 , 22 , 23 , . . . . Written this way, its natural to let the sequence function be 2n
with domain . As long as there is a simple substitution to write the sequence
function in the form of Definition 3.1, theres no reason to adhere to the letter of the
law. In general, the domain of a sequence can be any set of the form {n Z : n N }
for some N Z.
Definition 3.2. A sequence an is bounded if {an : n N} is a bounded set.
This definition is extended in the obvious way to bounded above and bounded below.
The sequence of Example 3.1 is bounded, but the sequence of Example 3.2 is
not, although it is bounded below.
3-1
3-2 CHAPTER 3. SEQUENCES

Definition 3.3. A sequence an converges to L R if for all > 0 there exists

an N N such that whenever n N , then |an L| < . If a sequence does not
converge, then it is said to diverge.
When an converges to L, we write limn an = L, or often, more simply,
an L.
Example 3.6. Let an = 1 1/n be as in Example 3.1. We claim an 1. To
see this, let > 0 and choose N N such that 1/N < . Then, if n N
|an 1| = |(1 1/n) 1| = 1/n 1/N < ,
so an 1.
Example 3.7. The sequence bn = 2n of Example 3.2 diverges. To see this,
suppose not. Then there is an L R such that bn L. If = 1, there must be
an N N such that |bn L| < whenever n N . Choose n N . |L 2n | < 1
implies L < 2n + 1. But, then
bn+1 L = 2n+1 L > 2n+1 (2n + 1) = 2n 1 1 = .
This violates the condition on N . We conclude that for every L R there exists an
> 0 such that for no N N is it true that whenever n N , then |bn L| < .
Therefore, bn diverges.
Definition 3.4. A sequence an diverges to if for every B > 0 there is an
N N such that n N implies an > B. The sequence an is said to diverge to
if an diverges to .
When an diverges to , we write limn an = , or often, more simply,
an .
A common mistake is to forget that an actually means the sequence
diverges in a particular way. Dont be fooled by the suggestive notation into treating
as a number!
Example 3.8. It is easy to prove that the sequence an = 2n of Example 3.2
diverges to .
Theorem 3.5. If an L, then L is unique.
Proof. Suppose an L1 and an L2 . Let > 0. According to Definition
3.2, there exist N1 , N2 N such that n N1 implies |an L1 | < /2 and n N2
implies |an L2 | < /2. Set N = max{N1 , N2 }. If n N , then
|L1 L2 | = |L1 an + an L2 | |L1 an | + |an L2 | < /2 + /2 = .
Since is an arbitrary positive number an application of Exercise 3.10 shows
L1 = L2 . 
Theorem 3.6. an L iff for all > 0, the set {n : an
/ (L , L + )} is
finite.
Proof. () Let > 0. According to Definition 3.2, there is an N N such
that {an : n N } (L, L+). Then {n : an / (L, L+)} {1, 2, . . . , N 1},
which is finite.
() Let > 0. By assumption {n : an / (L , L + )} is finite, so let
N = max{n : an / (L , L + )} + 1. If n N , then an (L , L + ). By
Definition 3.2, an L. 

May 1, 2015 http://math.louisville.edu/lee/ira

1. BASIC PROPERTIES 3-3

Corollary 3.7. If an converges, then an is bounded.

Proof. Suppose an L. According to Theorem 3.6 there are a finite number
of terms of the sequence lying outside (L 1, L + 1). Since any finite set is bounded,
the conclusion follows. 

The converse of this theorem is not true. For example, an = (1)n is bounded,
but does not converge. The main use of Corollary 3.7 is as a quick first check to see
whether a sequence might converge. Its usually pretty easy to determine whether a
sequence is bounded. If it isnt, it must diverge.
The following theorem lets us analyze some complicated sequences by breaking
them down into combinations of simpler sequences.
Theorem 3.8. Let an and bn be sequences such that an A and bn B.
Then
(a) an + bn A + B,
(b) an bn AB, and
(c) an /bn A/B as long as bn 6= 0 for all n N and B 6= 0.
Proof. (a) Let > 0. There are N1 , N2 N such that n N1 im-
plies |an A| < /2 and n N2 implies |bn B| < /2. Define
N = max{N1 , N2 }. If n N , then
|(an + bn ) (A + B)| |an A| + |bn B| < /2 + /2 = .
Therefore an + bn A + B.
(b) Let > 0 and > 0 be an upper bound for |an |. Choose N1 , N2 N such
that n N1 = |an A| < /2(|B|+1) and n N2 = |bn B| < /2.
If n N = max{N1 , N2 }, then
|an bn AB| = |an bn an B + an B AB|
|an bn an B| + |an B AB|
= |an ||bn B| + ||B||an A|

< + |B|
2 2(|B| + 1)
< /2 + /2 = .
(c) First, notice that it suffices to show that 1/bn 1/B, because part (b) of
this theorem can be used to achieve the full result.
Let > 0. Choose N N so that the following two conditions are
satisfied: n N = |bn | > |B|/2 and |bn B| < B 2 /2. Then, when
n N,
     2 
 1  =  B bn  <  B /2  = .
1     
 bn B   bn B   (B/2)B 
Therefore 1/bn 1/B.


If youre not careful, you can easily read too much into the previous theorem
and try to use its converse. Consider the sequences an = (1)n and bn = an .
Their sum, an + bn = 0, product an bn = 1 and quotient an /bn = 1 all converge,
but the original sequences diverge.

May 1, 2015 http://math.louisville.edu/lee/ira

3-4 CHAPTER 3. SEQUENCES

It is often easier to prove that a sequence converges by comparing it with a

known sequence than it is to analyze it directly. For example, a sequence such
as an = sin2 n/n3 can easily be seen to converge to 0 because it is dominated by
1/n3 . The following theorem makes this idea more precise. Its called the Sandwich
Theorem here, but is also called the Squeeze, Pinching, Pliers or Comparison
Theorem in different texts.
Theorem 3.9 (Sandwich Theorem). Suppose an , bn and cn are sequences such
that an bn cn for all n N.
(a) If an L and cn L, then bn L.
(b) If bn , then cn .
(c) If bn , then an .
Proof. (a) Let > 0. There is an N N large enough so that when
n N , then L < an and cn < L + . These inequalities imply
L < an bn cn < L + . Theorem 3.6 shows cn L.
(b) Let B > 0 and choose N N so that n N = bn > B. Then
cn bn > B whenever n N . This shows cn .
(c) This is essentially the same as part (b).


2. Monotone Sequences
One of the problems with using the definition of convergence to prove a given
sequence converges is the limit of the sequence must be known in order to verify the
sequence converges. This gives rise in the best cases to a chicken and egg problem
of somehow determining the limit before you even know the sequence converges. In
the worst case, there is no nice representation of the limit to use, so you dont even
have a target to shoot at. The next few sections are ultimately concerned with
removing this deficiency from Definition 3.2, but some interesting side-issues are
explored along the way.
Not surprisingly, we begin with the simplest case.
Definition 3.10. A sequence an is increasing, if an+1 an for all n N. It is
strictly increasing if an+1 > an for all n N.
A sequence an is decreasing, if an+1 an for all n N. It is strictly decreasing
if an+1 < an for all n N.
If an is any of the four types listed above, then it is said to be a monotone
sequence.
Notice the and in the definitions of increasing and decreasing sequences,
respectively. Many calculus texts use strict inequalities because they seem to better
match the intuitive idea of what an increasing or decreasing sequence should do.
For us, the non-strict inequalities are more convenient.
Theorem 3.11. A bounded monotone sequence converges.
Proof. Suppose an is a bounded increasing sequence, L = lub {an : n N}
and > 0. Clearly, an L for all n N. According to Theorem 2.17, there
exists an N N such that aN > L . Because the sequence is increasing,
L an aN > L for all n N . This shows an L.
If an is decreasing, let bn = an and apply the preceding argument. 

May 1, 2015 http://math.louisville.edu/lee/ira

2. MONOTONE SEQUENCES 3-5

The key idea of this proof is the existence of the least upper bound of the sequence
when viewed as a set. This means the Completeness Axiom implies Theorem 3.11.
In fact, it isnt hard to prove Theorem 3.11 also implies the Completeness Axiom,
showing they are equivalent statements. Because of this, Theorem 3.11 is often used
as the Completeness Axiom on R instead of the least upper bound property we used
in Axiom 8.

n
Example 3.9. The sequence en = 1 + n1 converges.

Looking at the first few terms of this sequence, e1 = 2, e2 = 2.25, e3 2.37,

e4 2.44, it seems to be increasing. To show this is indeed the case, fix n N and
use the binomial theorem to expand the product as
n  
X n 1
(9) en =
k nk
k=0

and
n+1
X 
n+1 1
(10) en+1 = .
k (n + 1)k
k=0

For 1 k n, the kth term of (9) is

n(n 1)(n 2) (n (k 1))

 
n 1
=
k nk k!nk
1 n1n2 nk+1
=
k!  n n n 
k1
 
1 1 2
= 1 1 1
k! n n n
k1
    
1 1 2
< 1 1 1
k! n+1 n+1 n+1
(n + 1)n(n 1)(n 2) (n + 1 (k 1))
=
k!(n + 1)k
 
n+1 1
= ,
k (n + 1)k
which is the kth term of (10). Since (10) also has one more positive term in the
sum, it follows that en < en+1 , and the sequence en is increasing.
Noting that 1/k! 1/2k1 for k N, equation (9) implies
 
n 1 n! 1
=
k nk k!(n k)! nk
n1n2 nk+1 1
=
n n n k!
1
<
k!
1
k1 .
2

May 1, 2015 http://math.louisville.edu/lee/ira

3-6 CHAPTER 3. SEQUENCES

Substituting this into (9) yields

n  
X n 1
en =
k nk
k=0
1 1 1
<1+1+ + + + n1
2 4 2
1 21n
=1+ < 3,
1 12
so en is bounded.
Since en is increasing and bounded, Theorem 3.11 implies en converges. Of
course, you probably remember from your calculus course that en e 2.71828.
Theorem 3.12. An unbounded monotone sequence diverges to or ,
depending on whether it is increasing or decreasing, respectively.
Proof. Suppose an is increasing and unbounded. If B > 0, the fact that an is
unbounded yields an N N such that aN > B. Since an is increasing, an aN > B
for all n N . This shows an .
The proof when the sequence decreases is similar. 

3. Subsequences and the Bolzano-Weierstrass Theorem

Definition 3.13. Let an be a sequence and : N N be a function such
that m < n implies (m) < (n); i.e., is a strictly increasing sequence of natural
numbers. Then bn = a (n) = a(n) is a subsequence of an .
The idea here is that the subsequence bn is a new sequence formed from an
old sequence an by possibly leaving terms out of an . In other words, we see all the
terms of bn must also appear in an , and they must appear in the same order.
Example 3.10. Let (n) = 3n and an be a sequence. Then the subsequence
a(n) looks like
a3 , a6 , a9 , . . . , a3n , . . .
The subsequence has every third term of the original sequence.
Example 3.11. If an = sin(n/2), then some possible subsequences are
bn = a4n+1 = bn = 1,
cn = a2n = cn = 0,
and
dn = an2 = dn = (1 + (1)n+1 )/2.
Theorem 3.14. an L iff every subsequence of an converges to L.
Proof. () Suppose : N N is strictly increasing, as in the preceding
definition. With a simple induction argument, it can be seen that (n) n for all
n. (See Exercise 3.8.)
Now, suppose an L and bn = a(n) is a subsequence of an . If > 0, there
is an N N such that n N implies an (L , L + ). From the preceding
paragraph, it follows that when n N , then bn = a(n) = am for some m n. So,
bn (L , L + ) and bn L.
() Since an is a subsequence of itself, it is obvious that an L. 

May 1, 2015 http://math.louisville.edu/lee/ira

4. LOWER AND UPPER LIMITS OF A SEQUENCE 3-7

The main use of Theorem 3.14 is not to show that sequences converge, but, rather
to show they diverge. It gives two strategies for doing this: find two subsequences
converging to different limits, or find a divergent subsequence. In Example 3.11, the
subsequences bn and cn demonstrate the first strategy, while dn demonstrates the
second.
Even if the original sequence diverges, it is possible there are convergent subse-
quences. For example, consider the divergent sequence an = (1)n . In this case, an
diverges, but the two subsequences a2n = 1 and a2n+1 = 1 are constant sequences,
so they converge.
Theorem 3.15. Every sequence has a monotone subsequence.
Proof. Let an be a sequence and T = {n N : m > n = am an }. There
are two cases to consider, depending on whether T is finite.
First, assume T is infinite. Define (1) = min T and assuming (n) is defined,
set (n + 1) = min T \ {(1), (2), . . . , (n)}. This inductively defines a strictly
increasing function : N N. The definition of T guarantees a(n) is an increasing
subsequence of an .
Now, assume T is finite. Let (1) = max T + 1. If (n) has been chosen for
some n > max T , then the definition of T implies there is an m > (n) such that
am a(n) . Set (n + 1) = m. This inductively defines the strictly increasing
function : N N such that a(n) is a decreasing subsequence of an . 
If the sequence in Theorem 3.15 is bounded, then the corresponding monotone
subsequence is also bounded. Recalling Theorem 3.11, we see the following.
Theorem 3.16 (Bolzano-Weierstrass). Every bounded sequence has a convergent
subsequence.

4. Lower and Upper Limits of a Sequence

There are an uncountable number of strictly increasing functions : N N,
so every sequence an has an uncountable number of subsequences. If an converges,
then Theorem 3.14 shows all of these subsequences converge to the same limit. Its
also apparent that when an or an , then all its subsequences diverge in
the same way. When an does not converge or diverge to , the situation is a bit
more difficult.
Example 3.12. Let Q = {qn : n N} and R. Since every interval
contains an infinite number of rational numbers, it is possible to choose (1) =
min{k : |qk | < 1}. In general, assuming (n) has been chosen, choose (n +
1) = min{k > (n) : |qk | < 1/n}. Such a choice is always possible because
Q(1/n, +1/n)\{qk : k (n)} is infinite. This induction yields a subsequence
q(n) of qn converging to .
If an is a sequence and bn is a convergent subsequence of an with bn L,
then L is called an accumulation point of an . A convergent sequence has only one
accumulation point, but a divergent sequence may have many accumulation points.
As seen in Example 3.12, a sequence may have all of R as its set of accumulation
points.
To make some sense out of this, suppose an is a bounded sequence, and
Tn = {ak : k n}. Define
`n = glb Tn and n = lub Tn .

May 1, 2015 http://math.louisville.edu/lee/ira

3-8 CHAPTER 3. SEQUENCES

Because Tn Tn+1 , it follows that for all n N,

(11) `1 `n `n+1 n+1 n 1 .
This shows `n is an increasing sequence bounded above by 1 and n is a decreasing
sequence bounded below by `1 . Theorem 3.11 implies both `n and n converge. If
`n ` and n , (11) shows for all n,
(12) `n ` n .
Suppose bn is any convergent subsequence of an . From the definitions of
`n and n , it is seen that `n bn n for all n. Now (12) shows ` .
The normal terminology for ` and is given by the following definition.
Definition 3.17. Let an be a sequence. If an is bounded below, then the lower
limit of an is
lim inf an = lim glb {ak : k n}.
n
If an is bounded above, then the upper limit of an is
lim sup an = lim lub {ak : k n}.
n
When an is unbounded, the lower and upper limits are set to appropriate infinite
values, while recalling the familiar warnings about not being a number.
Example 3.13. Define
(
2 + 1/n, n odd
an = .
1 1/n, n even
Then (
2 + 1/n, n odd
n = lub {ak : k n} = 2
2 + 1/(n + 1), n even
and (
1 1/n, n even
`n = glb {ak : k n} = 1.
1 1/(n + 1), n even
So,
lim sup an = 2 > 1 = lim inf an .
Suppose an is bounded above and both n and are as in the discussion
preceding the definition. Choose (1) so a(1) > 1 1. If (n) has been chosen for
some n N, then choose (n + 1) > (n) to satisfy
n a(n+1) > lub Tn+1 1/n = un+1 1/n.
This inductively defines a subsequence a(n) = lim sup an , where the conver-
gence is guaranteed by Theorem 3.9, the Sandwich Theorem.
In the cases when lim sup an = and lim sup an = , it is left to the reader
to show there is a subsequence bn lim sup an .
Similar arguments can be made for lim inf an .
To summarize: If is an accumulation point of an , then
lim inf an lim sup an .
In case an is bounded, both lim inf an and lim sup an are accumulation points of an
and an converges iff lim inf an = limn an = lim sup an .
The following theorem has been proved.

May 1, 2015 http://math.louisville.edu/lee/ira

3: Cauchy Sequences 3-9

Theorem 3.18. Let an be a sequence.

(a) If is an accumulation point of an , then lim inf an lim sup an .
(b) There are subsequences of an converging to lim inf an and lim sup an .
This shows, whenever they are finite, lim inf an and lim sup an are accu-
mulation points of an .
(c) lim inf an = lim sup an iff an converges.

5. The Nested Interval Theorem

Definition 3.19. A collection of sets {Sn : n N} is said to be nested, if
Sn+1 Sn for all n N.
Theorem 3.20 (Nested Interval Theorem). If {In = [an , bn ] : n N} is a
closed intervals such that limn (bn an ) = 0, then there is
nested collection of T
an x R such that nN In = {x}.
Proof. Since the intervals are nested, its clear that an is an increasing sequence
bounded above by b1 and bn is a decreasing sequence bounded below by a1 . Applying
Theorem 3.11 twice, we find there are , R such that an and bn .
We claim = . To see this, let > 0 and use the shrinking condition on
the intervals to pick N N so that bN aN < . The nestedness of the intervals
implies aN an < bn bN for all n N . Therefore
aN lub {an : n N } = bN and aN glb {bn : n N } = bN .
This shows | | |bN aN | < . Since > 0 was chosen arbitrarily, we conclude
= . T
Let x = = . It remains T to show that nN In = {x}.
First, we show that x nN In . To do this, fix N N. Since an increases to
x, its clear that x aN . Similarly, x bTN . Therefore x [aN , bN ]. Because N
T x nN In .
was chosen arbitrarily, it follows that
Next, suppose there are x,Ty nN In and let > 0. Choose N N such that
bN aN < . Then {x, y} nN In [aNT, bN ] implies |x y| < . Since was
chosen arbitrarily, we see x = y. Therefore nN In = {x}. 
ExampleT3.14. If In = (0, 1/n] for all n N, then the collection {In : n N}
is nested, but nN In = . This shows the assumption that the intervals be closed
in the Nested Interval Theorem is necessary.
Example
T 3.15. If In = [n, ) then the collection {In : n N} is nested,
but nN In = . This shows the assumption that the lengths of the intervals be
bounded is necessary. (It will be shown in Corollary 5.11 that when their lengths
dont go to 0, then the intersection is nonempty, but the uniqueness of x is lost.)

6. Cauchy Sequences
Often the biggest problem with showing that a sequence converges using the
techniques we have seen so far is that we must know ahead of time to what it
converges. This is the chicken and egg problem mentioned above. An escape from
this dilemma is provided by Cauchy sequences.
Definition 3.21. A sequence an is a Cauchy sequence if for all > 0 there is
an N N such that n, m N implies |an am | < .
1May 1, 2015 Lee
c Larson ([email protected])

May 1, 2015 http://math.louisville.edu/lee/ira

3-10 CHAPTER 3. SEQUENCES

This definition is a bit more subtle than it might at first appear. It sort of says
that all the terms of the sequence are close together from some point onward. The
emphasis is on all the terms from some point onward. To stress this, first consider
a negative example.
Pn
Example 3.16. Suppose an = k=1 1/k for n N. Theres a trick for showing
the sequence an diverges. First, note that an is strictly increasing. For any n N,
consider
n j
2X 1 n1
X 2X1
1 1
a2n 1 = = j +k
k j=0
2
k=1 k=0
j
n1
X 2X 1 n1
1 X 1 n
> = =
j=0 k=0
2j+1 j=0
2 2
Hence, the subsequence a2n 1 is unbounded and the sequence an diverges. (To see
how this works, write out the first few sums of the form a2n 1 .)
On the other hand, |an+1 an | = 1/(n + 1) 0 and indeed, if m is fixed,
|an+m sn | 0. This makes it seem as though the terms are getting close together,
as in the definition of a Cauchy sequence. But, sn is not a Cauchy sequence, as
shown by the following theorem.
Theorem 3.22. A sequence converges iff it is a Cauchy sequence.
Proof. () Suppose an L and > 0. There is an N N such that n N
implies |an L| < /2. If m, n N , then
|am an | = |am L + L an | |am L| + |L am | < /2 + /2 = .
This shows an is a Cauchy sequence.
() Let an be a Cauchy sequence. First, we claim that an is bounded. To see
this, let = 1 and choose N N such that n, m N implies |an am | < 1. In this
case, aN 1 < an < aN + 1 for all n N , so {an : n N } is a bounded set. The
set {an : n < N }, being finite, is also bounded. Since {an : n N} is the union of
these two bounded sets, it too must be bounded.
Because an is a bounded sequence, Theorem 3.16 implies it has a convergent
subsequence bn = a(n) L. Let > 0 and choose N N so that n, m N implies
|an am | < /2 and |bn L| < /2. If n N , then (n) n N and
|an L| = |an bn + bn L|
|an bn | + |bn L|
= |an a(n) | + |bn L|
< /2 + /2 = .
Therefore, an L. 
The fact that Cauchy sequences converge is yet another equivalent version
of completeness. In fact, most advanced texts define completeness as Cauchy
sequences converge. This is convenient in general spaces because the definition of a
Cauchy sequence only needs the metric on the space and none of its other structure.
A typical example of the usefulness of Cauchy sequences is given below.
Definition 3.23. A sequence xn is contractive if there is a c (0, 1) such that
|xk+1 xk | c|xk xk1 | for all k > 1. c is called the contraction constant.

May 1, 2015 http://math.louisville.edu/lee/ira

3: Cauchy Sequences 3-11

Theorem 3.24. If a sequence is contractive, then it converges.

Proof. Let xk be a contractive sequence with contraction constant c (0, 1).
We first claim that if n N, then
(13) |xn xn+1 | cn1 |x1 x2 |.
This is proved by induction. When n = 1, the statement is
|x1 x2 | c0 |x1 x2 | = |x1 x2 |,
which is trivially true. Suppose that |xn xn+1 | cn1 |x1 x2 | for some n N.
Then, from the definition of a contractive sequence and the induction hypothesis,
|xn+1 xn+2 | c|xn xn+1 | c cn1 |x1 x2 | = cn |x1 x2 |.

This shows the claim is true in the case n + 1. Therefore, by induction, the claim is
true for all n N.
To show xn is a Cauchy sequence, let > 0. Since cn 0, we can choose N N
so that
cN 1
(14) |x1 x2 | < .
(1 c)
Let n > m N . Then
|xn xm | = |xn xn1 + xn1 xn2 + xn2 xm+1 + xm+1 xm |
|xn xn1 | + |xn1 xn2 | + + |xm+1 xm |
Now, use (13) on each of these terms.
cn2 |x1 x2 | + cn3 |x1 x2 | + + cm1 |x1 x2 |
= |x1 x2 |(cn2 + cn3 + + cm1 )
Apply the formula for a geometric sum.
1 cnm
= |x1 x2 |cm1
1c
cm1
(15) < |x1 x2 |
1c
Use (14) to estimate the following.
cN 1
|x1 x2 |
1c

< |x1 x2 |
|x1 x2 |
=
This shows xn is a Cauchy sequenceand must converge by Theorem 3.22. 
Pn
Example 3.17. Let 1 < r < 1 and define the sequence sn = k=0 rk . (You
no doubt recognize this as the geometric series from your calculus course.) If r = 0,
the convergence of sn is trivial. So, suppose r 6= 0. In this case,
|sn+1 sn |  rn+1 
 
=  n  = |r| < 1
|sn sn1 | r
and sn is contractive. Theorem 3.24 implies sn converges.

May 1, 2015 http://math.louisville.edu/lee/ira

3-12 CHAPTER 3. SEQUENCES

Example 3.18. Suppose f (x) = 2 + 1/x, a1 = 2 and an+1 = f (an ) for n N.

It is evident that an 2 for all n. Some algebra gives
   
 an+1 an   f (f (an1 )) f (an1 )  1 1
 1 + 2an1 5 .
 = =
 an an1   f (an1 ) an1
This shows an is a contractive sequence and, according to Theorem 3.24, an L
for some L 2. Since, an+1 = 2 + 1/an , taking the limit as n of both sides
gives L = 2 + 1/L. A bit more algebra shows L = 1 + 2.
L is called a fixed point of the function f ; i.e. f (L) = L. Many approximation
techniques for solving equations involve such iterative techniques depending upon
contraction to find fixed points.
The calculations in the proof of Theorem 3.24 give the means to approximate
the fixed point to within an allowable error. Looking at line (15), notice
cm1
|xn xm | < |x1 x2 | .
1c
Let n in this inequality to arrive at the error estimate
cm1
(16) |L xm | |x1 x2 | .
1c
In Example 3.18, a1 = 2, a2 = 5/2 and c 1/5. Suppose we want to approximate
L to 5 decimal places of accuracy. This means we need |an L| < 5 106 . Using
(16), with m = 9 shows
cm1
|a1 a2 | 1.6 106 .
1c

Some arithmetic gives a9 2.41421. The calculator value of L = 1 + 2
2.414213562, confirming our estimate.

7. Exercises

6n 1
3.1. Let the sequence an = . Use the definition of convergence for a
3n + 2
sequence to show an converges.

3.2. If an is a sequence such that a2n L and a2n+1 L, then an L.

3.3. Let an be a sequence such that a2n A and a2n a2n1 0. Then an A.

3.4. If an is a sequence of positive numbers converging to 0, then an 0.

3.5. Find examples of sequences an and bn such that an 0 and bn such

that
(a) an bn 0
(b) an bn
(c) limn an bn does not exist, but an bn is bounded.
(d) Given c R, an bn c.

3.6. If xn and yn are sequences such that limn xn = L 6= 0 and limn xn yn

exists, then limn yn exists.

May 1, 2015 http://math.louisville.edu/lee/ira

7. EXERCISES 3-13

3.7. Determine the limit of an = n
n!. (Hint: If n is even, then n! > (n/2)n/2 .)

3.8. If : N N is strictly increasing, then (n) n for all n N.

3.9. Every unbounded sequence contains a monotonic subsequence.

3.10. Find a sequence an such that given x [0, 1], there is a subsequence bn of
an such that bn x.

3.11. A sequence an converges to 0 iff |an | converges to 0.

3.12. Define the sequence an = n for n N. Show that |an+1 an | 0, but an
is not a Cauchy sequence.

3.13. Suppose a sequence is defined by a1 = 0, a1 = 1 and an+1 = 12 (an + an1 )

for n 2. Prove an converges, and determine its limit.

3.14. If the sequence an is defined recursively by a1 = 1 and an+1 = an + 1,
then show an converges and determine its limit.

3.15. If an is a sequence such that limn |an+1 /an | = < 1, then an 0.

3.16. Prove that the sequence an = n3 /n! converges.

3.17. Let an and bn be sequences. Prove that both sequences an and bn converge
iff both an + bn and an bn converge.

3.18. Let an be a bounded sequence. Prove that given any > 0, there is an
interval I with length such that {n : an I} is infinite. Is it necessary that an be
bounded?
Pn
3.19. A sequence an converges in the mean if an = n1 k=1 ak converges. Prove
that if an L, then an L, but the converse is not true.

3.20. Find a sequence xn such that for all n N there is a subsequence of xn

converging to n.

3.21. If an is a Cauchy sequence whose terms are integers, what can you say about
the sequence?
Pn
3.22. Show an = k=0 1/k! is a Cauchy sequence.

3.23. If an is a sequence such that every subsequence of an has a further

subsequence converging to L, then an L.

3.24. If a, b (0, ), then show n an + bn max{a, b}.

3.25. If 0 < < 1 and sn is a sequence satisfying |sn+1 | < |sn |, then sn 0.

3.26. If c 1 in the definition of a contractive sequence, can the sequence

converge?

May 1, 2015 http://math.louisville.edu/lee/ira

3-14 CHAPTER 3. SEQUENCES

3.27. If an is a convergent sequence and bn is a sequence such that |am an |

|bm bn | for all m, n N, then bn converges.

3.28. If an 0 for all n N and an L, then an L.

3.29. If an is a Cauchy sequence and bn is a subsequence of an such that bn L,

then an L.

3.30. Let an be a sequence. an L iff lim sup an = L = lim inf an .

3.31. Is lim sup(an + bn ) = lim sup an + lim sup bn ?

3.32. If an is a sequence of positive numbers, then lim inf an = lim sup 1/an .

3.33. an = 1/n is not contractive.

3.34. The equation x3 4x + 2 = 0 has one real root lying between 0 and 1.
Find a sequence of rational numbers converging to this root. Use this sequence to
approximate the root to five decimal places.

3.35. Approximate a solution of x3 5x + 1 = 0 to within 104 using a Cauchy

sequence.

3.36. Prove or give a counterexample: If an L and : N N is bijective, then

bn = a(n) converges. Note that bn might not be a subsequence of an . (bn is called
a rearrangement of an .)

May 1, 2015 http://math.louisville.edu/lee/ira

 CHAPTER 4

Series

Given a sequence an , in many contexts it is natural to ask about the sum of all
the numbers in the sequence. If only a finite number of the an are nonzero, this
is trivialand not very interesting. If an infinite number of the terms arent zero,
the path becomes less obvious. Indeed, its even somewhat questionable whether it
makes sense at all to add an infinite number of numbers.
There are many approaches to this question. The method given below is the
most common technique. Others are mentioned in the exercises.

1. What is a Series?
The idea behind adding up an infinite collection of numbers is a reduction to
the well-understood idea of a sequence. This is a typical approach in mathematics:
reduce a question to a previously solved problem.
Definition 4.1. Given a sequence an , the series having an as its terms is the
new sequence
Xn
sn = ak = a1 + a2 + + an .
k=1
The numbers sn are called the partial sums of the series. If sn S R, then the
series converges to S. This is normally written as

X
ak = S.
k=1

Otherwise, the series

Pdiverges.
The notation n=1 an is understood to stand for the sequence of partial sums
of the series
P with terms an . When there is no ambiguity, this is often abbreviated
to just an .
Example 4.1. If an = (1)n for n N, then s1 = 1, s2 = 1 + 1 = 0,
s3 = 1 + 1 1 = 1 and in general
(1)n 1
sn =
2
does not converge because it oscillates between 1 and 0. Therefore, the series
(1)n diverges.
P

Example 4.2 (Geometric Series). Recall that a sequence of the form an = c rn1
is called a geometric sequence. It gives rise to a series

X
c rn1 = c + cr + cr2 + cr3 +
n=1

4-1
4-2 Series

called a geometric series. The number r is called the ratio of the series.
Suppose an = rn1 for r 6= 1. Then,
s1 = 1, s2 = 1 + r, s3 = 1 + r + r2 , . . .
In general, it can be shown by induction (or even long division) that
n n
X X 1 rn
(23) sn = ak = rk1 = .
1r
k=1 k=1

The convergence of sn in (23) depends on the value of r. Letting n , its

apparent that sn diverges when |r| > 1 and converges to 1/(1 r) when |r| < 1.
When r = 1, sn = n . When r = 1, its essentially the same as Example 4.1,
and therefore diverges. In summary,

X c
c rn1 =
n=1
1r
for |r| < 1, and diverges when |r| 1. This is called a geometric series with ratio r.

Figure 1. Stepping to the wall.

steps

2 1 1/2 0
distance from wall

In some cases, the geometric series has an intuitively plausible limit. If you
start two meters away from a wall and keep stepping halfway to the wall, no number
of steps will get you to the wall, but a large number of steps will get you as close to
the wall as you want. (See Figure 1.) So, the total distance stepped has limiting
value 2. The total distance after n steps is the nth partial sum of a geometric series
with ratio r = 1/2 and c = 1.
P
Example 4.3 (Harmonic Series). The series n=1 1/n is called the harmonic
series. It was shown in Example 3.16 that the harmonic series diverges.
Example 4.4. The terms of the sequence
1
an = 2 , n N.
n +n
can be decomposed into partial fractions as
1 1
an = .
n n+1
If sn is the series having an as its terms, then s1 = 1/2 = 1 1/2. We claim that
sn = 1 1/(n + 1) for all n N. To see this, suppose sk = 1 1/(k + 1) for some
k N. Then
 
1 1 1 1
sk+1 = sk + ak+1 = 1 + =1
k+1 k+1 k+2 k+2

May 1, 2015 http://math.louisville.edu/lee/ira

Positive Series 4-3

and the claim is established by induction. Now its easy to see that
 
X 1 1
= lim 1 = 1.
n=1
n2 + n n n+2
This is an example of a telescoping series. The name is apparently based on the
idea that the middle terms of the series cancel, causing the series to collapse like a
hand-held telescope.
The following theorem is an easy consequence of the properties of sequences
shown in Theorem 3.8.
P P
Theorem 4.2. Let an and bn be convergent series.
P P
(a) If
P c R, then P n P an .
c a = c
(b) (an + bn ) = an + bn .
(c) an 0
Pn Pn
Proof. Let An = k=1 ak and Bn = k=1 bk be the sequences of partial
sums for each of the two series. By assumption, there are numbers A and B where
An AP and Bn B.P
n n
(a) k=1 c ak = c k=1 ak = cAn cA.
Pn Pn Pn
(b) k=1 (ak + bk ) = k=1 ak + k=1 bk = An + Bn A + B.
Pn Pn1
(c) For n > 1, an = k=1 ak k=1 ak = An An1 A A = 0. 
Notice that the first two parts of Theorem 4.2 show that the set of all convergent
series is closed under linear combinations.
Theorem 4.2(c) is most useful because its contrapositive provides the most basic
test for divergence.
P
Corollary 4.3 (Going to Zero Test). If an 6 0, then an diverges.
Many have made the mistake of reading too much into Corollary 4.3. It can
only be used to show divergence. When the terms of a series do tend to zero, that
does not guarantee convergence. Example 4.3, shows Theorem 4.2(c) is necessary,
but not sufficient for convergence.
Another useful observation is that the partial sums of a convergent series are a
Cauchy sequence. The Cauchy criterion for sequences can be rephrased for series as
the following theorem, the proof of which is Exercise 4.4.
P
Theorem 4.4 (Cauchy Criterion). Let an be a series. The following state-
ments are equivalent.
P
(a) an converges.
(b) For every > 0 there is an N N such that whenever n m N ,
then  n 
X 
ai  < .
 

 
i=m

2. Positive Series
Most of the time, it is very hard or impossible to determine the exact limit of
a convergent series. We must satisfy ourselves with determining whether a series
converges, and then approximating its sum. For this reason, the study of series

May 1, 2015 http://math.louisville.edu/lee/ira

4-4 Series

usually involves learning a collection of theorems that might answer whether a given
series converges, but dont tell us to what it converges. These theorems are usually
called the convergence tests. The reader probably remembers a battery of such tests
from her calculus course. There is a myriad of such tests, and the standard ones are
presented in the next few sections, along with a few of those less widely used.
Since convergence of a series is determined by convergence of the sequence of its
partial sums, the easiest series to study are those with well-behaved partial sums.
Series with monotone sequences of partial sums are certainly the simplest such
series.
P
Definition 4.5. The series an is a positive series, if an 0 for all n.
The advantage of a positive series is that its sequence of partial sums is non-
negative and increasing. Since an increasing sequence converges if and only if it is
bounded above, there is a simple criterion to determine whether a positive series
converges. All of the standard convergence tests for positive series exploit this
criterion.

2.1. The Most Common Convergence Tests. All beginning calculus courses
contain several simple tests to determine whether positive series converge. Most of
them are presented below.
2.1.1. Comparison Tests. The most basic convergence tests are the comparison
tests. In these tests, the behavior of one series is inferred from that of another
series. Although theyre easy to use, there is one often fatal catch: in order to use
a comparison test, you must have a known series to which you can compare the
mystery series. For this reason, a wise mathematician collects example series for her
toolbox. The more samples in the toolbox, the more powerful are the comparison
tests.
P P
Theorem 4.6 (Comparison Test). Suppose an and bn are positive series
with an bn for all n.
P P
(a) If P bn converges, then so doesP an .
(b) If an diverges, then so does bn .
P P
Proof. Let An and Bn be the partial sums of an and bn , respectively. It
follows from the assumptions that An and Bn are increasing and for all n N,
(24) An Bn .
P P
If bn = B, then (24) implies
P B is an upper bound for An , and an converges.
On the other hand, if an diverges, An and the Sandwich Theorem
3.9(b) shows Bn . 

1/n diverges. If p 1, then 1/np

P
Example 4.5. Example 4.3 Pshows that
p
1/n, and Theorem 4.6 implies 1/n diverges.
P 2
Example 4.6. The series sin n/2n converges because
sin2 n 1
n
2n 2
1/2n = 1.
P
for all n and the geometric series

May 1, 2015 http://math.louisville.edu/lee/ira

Positive Series 4-5

a1 + a2 + a3 + a4 + a5 + a6 + a7 + a8 + a9 + + a15 +a16 +
| {z } | {z } | {z }
2a2 4a4 8a8

a1 + a2 + a3 + a4 + a5 + a6 + a7 + a8 + a9 + + a15 +a16 +
|{z} | {z } | {z } | {z }
a2 2a4 4a8 8a16

Figure 2. This diagram shows the groupings used in inequality (25).

Theorem 4.7 (Cauchys Condensation Test1). Suppose an is a decreasing

sequence of nonnegative numbers. Then
X X
an converges iff 2n a2n converges.

Proof. Since an is decreasing, for n N,

2n+1
X1
n
2X 1
n
(25) ak 2 a2n 2 ak .
k=2n k=2n1

(See Figure 2.1.1.) Adding for 1 n m gives

2m+1
X1 m
X
m
2X 1
k
ak 2 a2k 2 ak
k=2 k=1 k=1

and the theorem follows from the Comparison Test. 

1/np is called a p-series.

P
Example 4.7 (p-series). For fixed p R, the series
The special case when p = 1 is the harmonic series. Notice
X 2n X
n
n p
= 21p
(2 )
is a geometric series with ratio 21p , so it converges only when 21p < 1. Since
21p < 1 only when p > 1, it follows from the Cauchy Condensation Test that the
p-series converges when p > 1 and diverges when p 1. (Of course, the divergence
half of this was already known from Example 4.5.)
The p-series are often useful for the Comparison Test, and also occur in many
areas of advanced mathematics such as harmonic analysis and number theory.
P P
Theorem 4.8 (Limit Comparison Test). Suppose an and bn are positive
series with
an an
(26) = lim inf lim sup = .
bn bn
P P P
(a) If (0, ) and P an converges, then so does bn , and if bn
diverges, then so does Pan . P P
(b) If (0, ) and Pbn diverges, then so does an , and if an
converges, then so does bn .

1The series P 2n a n is sometimes called the condensed series associated with P a .

2 n

May 1, 2015 http://math.louisville.edu/lee/ira

4-6 Series

Proof. To prove (a), suppose > 0. There is an N N such that

an
(27) n N = < .
2 bn
If n > N , then (27) gives
n n
X X
(28) bk < ak
2
k=N k=N

Part (a) now follows from the comparison test.

The proof of (b) is similar. 

The following easy corollary is the form this test takes in most calculus books.
Its easier to use than Theorem 4.8 and suffices most of the time.
P P
Corollary 4.9 (Limit Comparison Test). Suppose an and bn are positive
series with
an
(29) = lim .
n bn
P P
If (0, ), then an and bn either both converge or both diverge.
X 1
Example 4.8. To test the series for convergence, let
2n n
1 1
an = n and bn = n .
2 n 2
Then
an 1/(2n n) 2n 1
lim = lim n
= lim n = lim = 1 (0, ).
n bn n 1/2 n 2 n n 1 n/2n

1/2n = 1, the original series converges by the Limit Comparison Test.

P
Since
2.1.2. Geometric Series-Type Tests. The most important series is undoubtedly
the geometric series. Several standard tests are basically comparisons to geometric
series.
P
Theorem 4.10 (Root Test). Suppose an is a positive series and
= lim sup a1/n
n .
P P
If < 1, then an converges. If > 1, then an diverges.
1/n
Proof. First, suppose < 1 and r (, 1). There is an N N so that an < r
for all n N . This is the same as an < rn for all n N . Using this, it follows that
when n N ,
Xn N
X 1 n
X N
X 1 Xn
ak = ak + ak < ak + rk .
k=1 k=1 k=N k=1 k=N
Pn k
Since k=N r is a partial sum of a geometric series with ratio r < 1, it must
converge.
1/k
If > 1, there is an increasing sequence of integers kn such that akn n > 1
P
for all n N. This shows akn > 1 for all n N. By Theorem 4.3, an diverges. 

May 1, 2015 http://math.louisville.edu/lee/ira

Positive Series 4-7

P n
Example 4.9. For any x R, the series |x |/n! converges. To see this, note
that according to Exercise 4.7,
 n 1/n
|x | |x|
= 0 < 1.
n! (n!)1/n
Applying the Root Test shows the series converges.
1/n2 . The first diverges
P P
Example 4.10. Consider the p-series 1/n and
1/n 2/n
and the second converges. Since n 1 and n 1, it can be seen that when
= 1, the Root Test in inconclusive.
P
Theorem 4.11 (Ratio Test). Suppose an is a positive series. Let
an+1 an+1
r = lim inf lim sup = R.
n an n an
P P
If R < 1, then an converges. If r > 1, then an diverges.
Proof. First, suppose R < 1 and (R, 1). There exists N N such that
an+1 /an < whenever n N . This implies an+1 < an whenever n N . From
this its easy to prove by induction that aN +m < m aN whenever m N. It follows
that, for n > N ,
n
X N
X n
X
ak = ak + ak
k=1 k=1 k=N +1
N
X nN
X
= ak + aN +k
k=1 k=1
N
X nN
X
< ak + aN k
k=1 k=1
N
X aN
< ak + .
1
k=1
P P
Therefore, the partial sums of an are bounded, and an converges.
If r > 1, then choose N N so that an+1 > an for all n N . Its now apparent
that an 6 0. 

In calculus books, the ratio test usually takes the following simpler form.
P
Corollary 4.12 (Ratio Test). Suppose an is a positive series. Let
an+1
r = lim .
n an
P P
If r < 1, then an converges. If r > 1, then an diverges.
From a practical viewpoint, the ratio test is often easier to apply than the root
test. But, the root test is actually the stronger of the two in the sense that there
are series for which the ratio test fails, but the root test succeeds. (See Exercise
4.11, for example.) This happens because
an+1 an+1
(30) lim inf lim inf a1/n
n lim sup a1/n
n lim sup .
an an

May 1, 2015 http://math.louisville.edu/lee/ira

4-8 Series

To see this, note the middle inequality is always true. To prove the right-hand
1/n
inequality, choose r > lim sup an+1 /an . It suffices to show lim sup an r. As in
the proof of the ratio test, an+k < rk an . This implies
an
an+k < rn+k n ,
r
which leads to
 a 1/(n+k)
1/(n+k) n
an+k <r n .
r
Finally,
 a 1/(n+k)
1/(n+k) n
lim sup a1/n
n = lim sup an+k lim sup r = r.
k k rn
The left-hand inequality is proved similarly.

2.2. Kummer-Type Tests. Most times the simple tests of the preceding sec-
tion suffice. However, more difficult series require more delicate tests. There dozens
of other, more specialized, convergence tests. Several of them are consequences of
the following theorem.
P
Theorem 4.13 (Kummers Test). Suppose an is a positive series, pn is a
sequence of positive numbers and
   
an an
(31) = lim inf pn pn+1 lim sup pn pn+1 =
an+1 an+1
P P P
If > 0, then an converges. If 1/pn diverges and < 0, then an diverges.
Pn
Proof. Let sn = k=1 ak , suppose > 0 and choose r (0, ). There must
be an N > 1 such that
an
pn pn+1 > r, n N.
an+1
Rearranging this gives
(32) pn an pn+1 an+1 > ran+1 , n N.
For M > N , (32) implies
M
X M
X
(pn an pn+1 an+1 ) > ran+1
n=N n=N
pN aN pM +1 aM +1 > r(sM sN 1 )
pN aN pM +1 aM +1 + rsN 1 > rsM
pN aN + rsN 1
> sM
r
P
Since N is fixed, the left side is an upper bound for sM , and it follows that an
converges. P
Next suppose 1/pn diverges and < 0. There must be an N N such that
an
pn pn+1 < 0, n N.
an+1
This implies
pn an < pn+1 an+1 , n N.

May 1, 2015 http://math.louisville.edu/lee/ira

Kummer-Type Tests 4-9

Therefore, pn an > pN aN whenever n > N and

1
an > pN aN , n N.
pn
P P
Because N is fixed and 1/pn diverges, the Comparison Test shows an diverges.

Kummers test is powerful. In fact, it can be shown that, given any positive
series, a judicious choice of the sequence pn can always be made to determine whether
it converges. (See Exercise 4.18, [18] and [17].) But, as stated, Kummers test is
not very useful because choosing pn for a given series is often difficult. Experience
has led to some standard choices that work with large classes of series. For example,
Exercise 4.9 asks you to prove the choice pn = 1 for all n reduces Kummers test to
the standard ratio test. Other useful choices are shown in the following theorems.
P
Theorem 4.14 (Raabes Test). Let an be a positive series such that an > 0
for all n. Define
   
an an
= lim sup n 1 lim inf n 1 =
n an+1 n an+1
P P
If > 1, then an converges. If < 1, then an diverges.
Proof. Let pn = n in Kummers test, Theorem 4.13. 
When Raabes test is inconclusive, there are even more delicate tests, such as
the theorem given below.
P
Theorem 4.15 (Bertrands Test). Let an be a positive series such that an > 0
for all n. Define
       
an an
= lim inf ln n n 1 1 lim sup ln n n 1 1 = .
n an+1 n an+1
P P
If > 1, then an converges. If < 1, then an diverges.
Proof. Let pn = n ln n in Kummers test. 
Example 4.11. Consider the series
n
!p
X X Y 2k
(33) an = .
2k + 1
k=1
Its of interest to know for what values of p it converges.
An easy computation shows that an+1 /an 1, so the ratio test is inconclusive.
Next, try Raabes test. Manipulating
 p
2n+3

an
p 
2n+2 1
lim n 1 = lim 1
n an+1 n
n

it becomes a 0/0 form and can be evaluated with LHospitals rule.2

 p
3+2 n
n2 2+2 n p p
lim = .
n (1 + n) (3 + 2 n) 2

2See 5.2.

May 1, 2015 http://math.louisville.edu/lee/ira

4-10 Series

From Raabes test, Theorem 4.14, it follows that the series converges when p > 2
and diverges when p < 2. Raabes test is inconclusive when p = 2.
Now, suppose p = 2. Consider
   
an (4 + 3 n)
lim ln n n 1 1 = lim ln n 2 =0
n an+1 n 4 (1 + n)
and Bertrands test, Theorem 4.15, shows divergence.
The series (33) converges only when p > 2.

3. Absolute and Conditional Convergence

The tests given above are for the restricted case when a series has positive
terms. If the stipulation that the series be positive is thrown out, things becomes
considerably more complicated. But, as is often the case in mathematics, some
problems can be attacked by reducing them to previously solved cases. The following
definition and theorem show how to do this for some special cases.
P P P
Definition 4.16. Let an be a series. If |an | converges, then an is
absolutely convergent. If it is convergent, but not absolutely convergent, then it is
conditionally convergent.
P
Since |an | is a positive series, the preceding tests can be used to determine its
convergence. The following theorem shows that this is also enough for convergence
of the original series.
P
Theorem 4.17. If an is absolutely convergent, then it is convergent.
Proof. Let > 0. Theorem 4.4 yields an N N such that when n m N ,
n
 n 
X X 
> |ak |  ak  0.
 
 
k=m k=m
Another application Theorem 4.4 finishes the proof. 

(1)n+1 /n is called the alternating harmonic

P
Example 4.12. The series
series. Since the harmonic series diverges, we see the alternating harmonic series is
not absolutely convergent. Pn
On the other hand, if sn = k=1 (1)k+1 /k, then
n   X n
X 1 1 1
s2n = =
2k 1 2k 2k(2k 1)
k=1 k=1

is a positive series that converges. Since |s2n s2n1 | = 1/2n 0, its clear
that
P s2n1 must also converge to the same limit. Therefore, sn converges and
(1)n+1 /n is conditionally convergent. (See Figure 3.)
To summarize: absolute convergence implies convergence, but convergence does
not imply absolute convergence.
There are a few tests that address conditional convergence. Following are the
most well-known.
Theorem 4.18 (Abels Test). Let an and bn be sequences satisfying
Pn
(a) sn = k=1 ak is a bounded sequence.
(b) bn bn+1 , n N

May 1, 2015 http://math.louisville.edu/lee/ira

Absolute and Conditional Convergence 4-11

1.0

0.8

0.6

0.4

0.2

0 5 10 15 20 25 30 35

Figure 3. This plot shows the first 35 partial sums of the alternating
harmonic series. It converges to ln 2 0.6931, which is the level of the
dashed line. Notice how the odd partial sums decrease to ln 2 and the
even partial sums increase to ln 2.

(c) bn 0
P
Then an bn converges.

To prove this theorem, the following lemma is needed.

Lemma 4.19 (Summation by Parts). For every pair of sequences an and bn ,

n
X n
X n
X k
X
ak bk = bn+1 ak (bk+1 bk ) a`
k=1 k=1 k=1 `=1
Pn
Proof. Let sn = k=1 ak and s0 = 0. Then
n
X n
X
ak bk = (sk sk1 )bk
k=1 k=1
Xn n
X
= sk bk sk1 bk
k=1 k=1
n n
!
X X
= sk bk sk bk+1 sn bn+1
k=1 k=1
n
X n
X k
X
= bn+1 ak (bk+1 bk ) a`
k=1 k=1 `=1


Pn
Proof. To prove the theorem, suppose | k=1 ak | < M for all n N. Let
> 0 and choose N N such that bN < /2M . If N m < n, use Lemma 4.19 to

May 1, 2015 http://math.louisville.edu/lee/ira

4-12 Series

write
 n   n m1

X  X X 
a` b`  =  a` b` a` b` 
   

   
`=m `=1 `=1
 n n `
 X X X
= bn+1 a` (b`+1 b` ) ak


`=1 `=1 k=1
m1 m1 `
!
X X X 
bm a` (b`+1 b` ) ak 


`=1 `=1 k=1

Using (a) gives

n
X
(bn+1 + bm )M + M |b`+1 b` |
`=m

Now, use (b) to see

n
X
= (bn+1 + bm )M + M (b` b`+1 )
`=m

and then telescope the sum to arrive at

= (bn+1 + bm )M + M (bm bn+1 )
= 2M bm

< 2M
2M
<
Pn
This shows `=1 a` b` satisfies Theorem 4.4, and therefore converges. 
Theres one special case of this theorem thats most often seen in calculus texts.

P Corollary 4.20 (Alternating Series Test).PIf cn decreases to 0, then the series

n
(1)n+1 cn converges. Moreover, if sn = k=1 (1)
n+1
cn and sn s, then
|sn s| < cn+1 .
Proof. Let an = (1)n+1 and bn = cn inPTheorem 4.18 to see the series
n
converges to some number s. For n N, let sn = k=0 (1)k+1 ck and s0 = 0. Since
s2n s2n+2 = c2n+1 + c2n+2 0 and s2n+1 s2n+3 = c2n+2 c2n+3 0,
It must be that s2n s and s2n+1 s. For all n ,
0 s2n+1 s s2n+1 s2n+2 = c2n+2 and 0 s s2n s2n+1 s2n = c2n+1 .
This shows |sn s| < cn+1 for all n. 
A series such as that in Corollary 4.20 is called an alternating series.
P More
formally, if an is a sequence such that an /an+1 < 0 for all n, then an is an
alternating series. Informally, it just means the series alternates between positive
and negative terms.
Example 4.13. Corollary 4.20 provides another way to prove the alternating
harmonic series in Example 4.12 converges. Figures 3 and 4 show how the partial
sums bounce up and down across the sum of the series.

May 1, 2015 http://math.louisville.edu/lee/ira

Rearrangements of Series 4-13

Figure 4. Here is a more whimsical way to visualize the partial sums

of the alternating harmonic series.

4. Rearrangements of Series
We want to use our standard intuition about adding lists of numbers when
working with series. But, this intuition has been formed by working with finite sums
and does not always work with series.

(1)n+1 /n = so that (1)n+1 2/n = 2. Its

P P
Example 4.14. Suppose
easy to show > 1/2. Consider the following calculation.
X 2
2 = (1)n+1
n
2 1 2 1
= 2 1 + + +
3 2 5 3
Rearrange and regroup.
   
1 2 1 1 2 1 1
= (2 1) + + +
2 3 3 4 5 5 6
1 1 1
= 1 + +
2 3 4
=

So, = 2 with =6 0. Obviously, rearranging and regrouping of this series is a

questionable thing to do.

In order to carefully consider the problem of rearranging a series, a precise

definition is needed.
P
Definition
P 4.21. Let : N N be a bijection and an be a series. The new
series a(n) is a rearrangement of the original series.

May 1, 2015 http://math.louisville.edu/lee/ira

4-14 Series

The problem with Example 4.14 is that the series is conditionally convergent.
Such examples cannot happen with absolutely convergent series. For the most part,
absolutely convergent series behave as we are intuitively led to expect.
P P
PTheorem P 4.22. If P an is absolutely convergent and a(n) is a rearrangement
of an , then a(n) = an .
P
Pn Proof. Let an = s and > 0. Choose N N so that N m < n implies
k=m |ak | < . Choose M N such that

{1, 2, . . . , N } {(1), (2), . . . , (M )}.

If P > M , then

 
XP P
X  X
ak a(k)  |ak |
 

 
k=1 k=1 k=N +1
and both series converge to the same number. 

When a series is conditionally convergent, the result of a rearrangement is

unpredictable. This is shown by the following theorem.
P
Theorem 4.23 (Riemann Rearrangement). If an is conditionally
P convergent
and c R {, }, then there is a rearrangement such that a(n) = c.
To prove this, the following lemma is needed.
P
Lemma 4.24. If an is conditionally convergent and
( (
an , an > 0 an , an < 0
bn = and cn = ,
0, an 0 0, an 0
P P
then both bn and cn diverge.
P P
Proof. Suppose bn converges. By assumption, an converges, so Theorem
4.2 implies
X X X
cn = bn an
converges. Another application of Theorem 4.2 shows
X X X
|an | = bn + cn
P
converges. ThisPis a contradiction of the assumption that an is conditionally
convergent, so bn cannot converge. P
A similar contradiction arises under the assumption that cn converges. 

Proof. (Theorem 4.23) Let bn and cn be as in Lemma 4.24 and define the
subsequence a+n of bn by removing those terms for which bn = 0 and an = 6 0. Define

the
P +subsequence a of
Pn n c by removing those terms for which c n = 0. The series
a
n=1 n and a
n=1 n are still divergent because only terms equal to zero have
been removed from bn and cn .
Now, let c R and m0 = n0 = 0. According to Lemma 4.24, we can define the
natural numbers
Xn m1
X Xn
m1 = min{n : a+k > c} and n 1 = min{n : a+
k + a
` < c}.
k=1 k=1 `=1

May 1, 2015 http://math.louisville.edu/lee/ira

5. EXERCISES 4-15

If mp and np have been chosen for some p N, then define

p mk+1 nk+1 n
!
X X X X
mp+1 = min n : a+
` a
` + a+
` >c

k=0 `=mk +1 `=nk +1 `=mp +1

and
p mk+1 nk+1
( !
X X X
np+1 = min n : a+
` a
`
k=0 `=mk +1 `=nk +1

np+1 n
X X
+ a+
` a
` < c .
`=mp +1 `=np +1

Consider the series

(34) a+ + +
1 + a2 + + am1 a1 a2 an1

+ a+ + +
m1 +1 + am1 +2 + + am2 an1 +1 an1 +2 an2

+ a+ + +
m2 +1 + am2 +2 + + am3 an2 +1 an2 +2 an3
+
P
It is clear this series is a rearrangement of n=1 an and the way in which mp and
np were chosen guarantee that

p1 mk+1 nk mp
X X X X
0< a+
` a
` + a+
k
c a+
mp
k=0 `=mk +1 `=nk +1 k=mp +1

and
p mk+1 nk
!
X X X
0<c a+
` a
` a
np
k=0 `=mk +1 `=nk +1

Since both a+
0 and
mp anp 0, the result follows
from the Squeeze Theorem.
The argument when c is infinite is left as Exercise 4.32. 

5. Exercises

4.1. Prove Theorem 4.4.

P P 1
4.2. If n=1 an is a convergent positive series, then does n=1 1+an converge?

X
4.3. The series (an an+1 ) converges iff the sequence an converges.
n=1
P
4.4. Prove or give a counter example: If |an | converges, then nan 0.

4.5. If the series a1 + a2 + a3 + converges to S, then so does

(35) a1 + 0 + a2 + 0 + 0 + a3 + 0 + 0 + 0 + a4 + .

P
4.6.
P If n=1 an converges and bn is a bounded monotonic sequence, then
a b
n=1 n n converges.

May 1, 2015 http://math.louisville.edu/lee/ira

4-16 Series

4.7.
P Let xn be a sequence with range {0, 1, 2, 3, 4, 5, 6, 7, 8, 9}. Prove that
n
n=1 n 10
x converges.

4.8. Write 6.17272727272 as a fraction.

4.9. Prove the ratio test by setting pn = 1 for all n in Kummers test.
P P an
4.10. If n=1 an and n=1 bn are positive series and lim
exists and is not
bnn
zero, then both series converge or both series diverge. (This is called the limit
comparison test.)

4.11. Consider the series

1 1 1 1
1+1++ + + + = 4.
2 2 4 4
Show that the ratio test is inconclusive for this series, but the root test gives a
positive answer.

X 1
4.12. Does converge?
n=2
n(ln n)2

4.13. Does
1 12 123 1234
+ + + +
3 35 357 3579
converge?

4.14. For what values of p does

 p  p  p
1 13 135
+ + +
2 24 246
converge?

4.15. Find sequences an and bn satisfying:

Pn
(a) an > 0,
n
N and an 0;
(b) B
Pn= k=1 bk is a bounded sequence; and,
(c) n=1 n bn diverges.
a

4.16. Let an be a sequence such that a2n A and a2n a2n1 0. Then
an A.

4.17. Prove Bertrands test, Theorem 4.15.

P P
4.18. Let an be a positive series. Prove that an converges if and only if there
is a sequence of positive numbers pn and > 0 such that
an
lim pn pn+1 = .
n an+1
P Pn
(Hint: If s = an and sn = k=1 ak , then let pn = (s sn )/an .)
P xn
4.19. Prove that n=0 n! converges for all x R.

May 1, 2015 http://math.louisville.edu/lee/ira

5. EXERCISES 4-17

P
4.20. Find all values of x for which k=0 k 2 (x + 3)k converges.

4.21. For what values of x does the series

X (1)n+1 x2n1
(40)
n=1
2n 1

converge?

X (x + 3)n
4.22. For what values of x does converge absolutely, converge
n=1
n4n
conditionally or diverge?

X n+6
4.23. For what values of x does 2 (x 1)n
converge absolutely, converge
n=1
n
conditionally or diverge?
P
4.24. For what positive values of does n=1 n n converge?
X n
4.25. Prove that cos sin converges.
3 n
P
4.26. For a series k=1 an with partial sums sn , define
n
1X
n = sn .
n
k=1
P
P that if k=1 an = s, then n s. Find an example where n converges, but
Prove
k=1 an does not. (If n converges, the sequence is said to be Cesaro summable.)
P
P If an is a sequence with a subsequence
4.27. Pbn , then n=1 bn is a subseries
P
of n=1 an . Prove that if every subseries of n=1 an converges, then n=1 an
converges absolutely.
P P
4.28. If n=1 an is a convergent positive series, then so is n=1 a2n . Give an
example to show the converse is not true.
P P
4.29. Prove or give a counter example: If n=1 an converges, then n=1 a2n
converges.
P
4.30. For what positive values of does n=1 n n converge?

4.31. If an 0 for all n N and there is a p > 1 such that limn np an exists
P
and is finite, then n=1 an converges. Is this true for p = 1?

4.32. Finish the proof of Theorem 4.23.

4.33. Leonhard Euler started with the equation

x x
+ = 0,
x1 1x
May 1, 2015 http://math.louisville.edu/lee/ira
4-18 Series

transformed it to
1 x
+ = 0,
1 1/x 1 x
and then used geometric series to write it as
1 1
(44) + 2 + + 1 + x + x2 + x3 + = 0.
x x
Show how Euler did his calculation and find his mistake.

May 1, 2015 http://math.louisville.edu/lee/ira

 CHAPTER 5

The Topology of R

1. Open and Closed Sets

Definition 5.1. A set G R is open if for every x G there is an > 0 such
that (x , x + ) G. A set F R is closed if F c is open.
The idea is that about every point of an open set, there is some room inside the
set on both sides of the point. It is easy to see that any open interval (a, b) is an
open set because if a < x < b and = min{x a, b x}, then (x , x + ) (a, b).
Open half-lines are also open sets. For example, let x (a, ) and = x a.
Then (x , x + ) (a, ).
A singleton set {a} is closed. To see this, suppose x 6= a and = |x a|. Then
c
a/ (x , x + ), and {a} must be open. The definition of a closed set implies
{a} is closed.
A common mistake is to assume all sets are either open or closed. Most sets are
neither open nor closed. For example, if S = [a, b) for some numbers a < b, then no
matter the size of > 0, neither (a , a + ) nor (b , b + ) are contained in S
or S c .
Theorem
S 5.2. (a) If {G : } is a collection of open sets, then
G is open.
Tn
(b) If {Gk : 1 k n} is a finite collection of open sets, then k=1 Gk is
open.
(c) Both and R are open.
S
Proof. (a) If x G , then there is a x such that x Gx .
Since
S Gx is open, there S is an > 0 such that x (x , x + ) Gx
G
T . This shows G is open.
n
(b) If x k=1 Gk , then x Gk for 1 k n. For each Gk there is an k
such that (x k , x + k ) Gk . Let = min{k : T 1 k n}. Then
n
(x
Tn , x + ) G k for 1 k n, so (x , x + ) k=1 Gk . Therefore
k=1 G k is open.
(c) is open vacuously. R is obviously open.

1.1. Topological Spaces. The preceding theorem provides the starting point
for a fundamental area of mathematics called topology. The properties of the open
sets of R motivate the following definition.
Definition 5.3. For X a set, not necessarily a subset of R, let T P(X). The
set T is called a topology on X if {, X} T and T is closed under arbitrary unions
and finite intersections. The pair (X, T ) is called a topological space. The elements
of T are the open sets of the topological space. The closed sets of the topological
space are those sets whose complements are open.
5-1
5-2 CHAPTER 5. THE TOPOLOGY OF R

If O = {G R : G is open}, then Theorem 5.2 shows (R, O) is a topological

space called the standard topology on R. While the standard topology is the most
widely used one, there are many other possible topologies on R. For example,
R = {(a, ) : a R} {R, } is a topology on R called the right ray topology. The
collection F = {S R : S c is finite} {} is called the finite complement topology.
The study of topologies is a huge subject, further discussion of which would take us
too far afield. There are many fine books on the subject ([14]) to which one can
refer.
Applying DeMorgans laws to the parts of Theorem 5.2 immediately yields the
following, which is sometimes also used as the definition of the standard topology.
Corollary
T 5.4. (a) If {F : } is a collection of closed sets, then
G is closed. Sn
(b) If {Fk : 1 k n} is a finite collection of closed sets, then k=1 Fk is
closed.
(c) Both and R are closed.
Surprisingly, and R are both open and closed. They are the only subsets of
the standard topology with this dual personality. Other topologies may have many
sets that are both open and closed. Such sets in a topological space are often called
clopen.
1.2. Limit Points and Closure.
Definition 5.5. x0 is a limit point1 of S R if for every > 0,
(x0 , x0 + ) S \ {x0 } =
6 .
The derived set of S is
S 0 = {x : x is a limit point of S}.
A point x0 S \ S 0 is an isolated point of S.
Notice that limit points of S need not be elements of S, but isolated points of S
must be elements of S. In a sense, limit points and isolated points are at opposite
extremes. The definitions can be restated as follows:
x0 is a limit point of S iff > 0 (S (x0 , x0 + ) \ {x0 } =
6 )
x0 S is an isolated point of S iff > 0 (S (x0 , x0 + ) \ {x0 } = )
Example 5.1. If S = (0, 1], then S 0 = [0, 1] and S has no isolated points.
Example 5.2. If T = {1/n : n Z \ {0}}, then T 0 = {0} and all points of T
are isolated points of T .
Theorem 5.6. x0 is a limit point of S iff there is a sequence xn S \ {x0 }
such that xn x0 .
Proof. () For each n N choose xn S (x0 1/n, x0 + 1/n) \ {x0 }. Then
|xn x0 | < 1/n for all n N, so xn x0 .
() Suppose xn is a sequence from xn S \ {x0 } converging to x0 . If > 0, the
definition of convergence for a sequence yields an N N such that whenever n N ,
1This use of the term limit point is not universal. Some authors use the term accumulation
point. Others use condensation point, although this is more often used for those cases when every
neighborhood of x0 intersects S in an uncountable set.

May 1, 2015 http://math.louisville.edu/lee/ira

1. OPEN AND CLOSED SETS 5-3

then xn S (x0 , x0 + ) \ {x0 }. This shows S (x0 , x0 + ) \ {x0 } =

6 ,
and x0 must be a limit point of S. 

There is some common terminology making much of this easier to state. If

x0 R and G is an open set containing x0 , then G is called a neighborhood of x0 .
The observations given above can be restated in terms of neighborhoods.
Corollary 5.7. Let S R.
(a) x0 is a limit point of S iff every neighborhood of x0 contains an infinite
number of points from S.
(b) x0 S is an isolated point of S iff there is a neighborhood of x0
containing only a finite number of points from S.
Following is a generalization of Theorem 3.16.
Theorem 5.8 (Bolzano-Weierstrass Theorem). If S R is bounded and infinite,
then S 0 6= .
Proof. For the purposes of this proof, if I = [a, b] is a closed interval, let
I L = [a, (a + b)/2] be the closed left half of I and I R = [(a + b)/2, b] be the closed
right half of I.
Suppose S is a bounded and infinite set. The assumption that S is bounded
implies the existence of an interval I1 = [B, B] containing S. Since S is infinite,
at least one of the two sets I1L S or I1R S is infinite. Let I2 be either I1L or I1R
such that I2 S is infinite.
If In is such that In S is infinite, let In+1 be either InL or InR , where In+1 S
is infinite.
In this way, a nested sequence of intervals, In for n N, is defined such that
In S is infinite for all n N and the length of In is B/2n2 T 0. According to
the Nested Interval Theorem, there is an x0 R such that nN In = {x0 }.
To see that x0 is a limit point of S, let > 0 and choose n N so that
B/2n2 < . Then x0 In (x0 , x0 + ). Since In S is infinite, we see
S (x0 , x0 + ) \ {x0 } =
6 . Therefore, x0 is a limit point of S. 

Theorem 5.9. A set S R is closed iff it contains all its limit points.
Proof. () Suppose S is closed and x0 is a limit point of S. If x0 / S, then
S c open implies the existence of > 0 such that (x0 , x0 + ) S = . This
contradicts the fact that x0 is a limit point of S. Therefore, x0 S, and S contains
all its limit points.
() Since S contains all its limit points, if x0
/ S, there must exist an > 0
such that (x0 , x0 + ) S = . It follows from this that S c is open. Therefore S
is closed. 

Definition 5.10. The closure of a set S is the set S = S S 0 .

For the set S of Example 5.1, S = [0, 1]. In Example 5.2, T = {1/n : n
Z \ {0}} {0}. According to Theorem 5.9, the closure of any set is a closed set.
A useful way to think about this is that S is the smallest closed set containing S.
This is made more precise in Exercise 5.2.
Following is a generalization of Theorem 3.20.

May 1, 2015 http://math.louisville.edu/lee/ira

5-4 CHAPTER 5. THE TOPOLOGY OF R

Corollary 5.11. T If {Fn : n N} is a nested collection of nonempty closed

and bounded sets, then nN Fn 6= .
Proof. Form a sequence xn by choosing xn Fn for each n N. Since the Fn
are nested, {xn : n N} F1 , and the boundedness of F1 implies xn is a bounded
sequence. An application of Theorem 3.16 yields a subsequence yn of xn such that
yn y. It suffices to prove y Fn for all n N.
To do this, fix n0 N. Because yn is a subsequence of xn and xn0 Fn0 , it is
easy to see yn Fn0 for all n n0 . Using the fact that yn y, we see y Fn0 0 .
Since Fn0 is closed, Theorem 5.9 shows y Fn0 . 

2. Relative Topologies and Connectedness

2.1. Relative Topologies. Another useful topological notion is that of a
relative or subspace topology. In our case, this amounts to using the standard
topology on R to generate a topology on a subset of R. The definition is as follows.
Definition 5.12. Let X R. The set S X is relatively open in X, if there
is a set G, open in R, such that S = G X. The set T X is relatively closed in
X, if there is a set F , closed in R, such that S = F X. (If there is no chance for
confusion, the simpler terminology open in X and closed in X is sometimes used.)
It is left as exercises to show that if X R and S consists of all relatively open
subsets of X, then (X, S) is a topological space and T is relatively closed in X, if
X \ T S. (See Exercises 5.10 and 5.11.)
Example 5.3. Let X = [0, 1]. The subsets [0, 1/2) = X (1, 1/2) and
(1/4, 1] = X (1/4, 2) are both relatively open in X.

Example
5.4. If X = Q, then {x Q : 2 < x < 2} = ( 2, 2) Q =
[ 2, 2] Q is clopen relative to Q.
2.2. Connected Sets. One place where the relative topologies are useful is
in relation to the following definition.
Definition 5.13. A set S R is disconnected if there are two open intervals U
and V such that U V = , U S 6= , V S 6= and S U V . Otherwise, it is
connected. The sets U S and V S are said to be a separation of S.
In other words, S is disconnected if it can be written as the union of two disjoint
and nonempty sets that are both relatively open in S. Since both these sets are
complements of each other relative to S, they are both clopen in S. This, in turn,
implies S is disconnected if it has a proper relatively clopen subset.
Example 5.5. Let S = {x} be a set containing a single point. S is connected
because there cannot exist nonempty disjoint open sets U and V such that S U =
6
and S V 6= . The same argument shows that is connected.
Example 5.6. If S = [1, 0) (0, 1], then U = (2, 0) and V = (0, 2) are open
sets such that U V = , U S = 6 , V S = 6 and S U V . This shows S is
disconnected.

Example 5.7. The sets U = (, 2) and V = ( 2, ) are open sets such
that U V = , U Q 6= , V Q 6= and Q U V = R \ { 2}. This shows
Q is disconnected. In fact, the only connected subsets of Q are single points. Sets
with this property are often called totally disconnected.

May 1, 2015 http://math.louisville.edu/lee/ira

Section 3: Covering Properties and Compactness on R 5-5

The notion of connectedness is not really very interesting on R because the

connected sets are exactly what one would expect. It becomes more complicated in
higher dimensional spaces.
Theorem 5.14. A nonempty set S R is connected iff it is either a single
point or an interval.
Proof. () If S is not a single point or an interval, there must be numbers
r < s < t such that r, t S and s / S. In this case, the sets U = (, s) and
V = (s, ) are a disconnection of S.
() It was shown in Example 5.5 that a set containing a single point is connected.
So, assume S is an interval.
Suppose S is not connected with U and V forming a disconnection of S. Choose
u U S and v V S. There is no generality lost by assuming u < v, so that
[u, v] S.
Let A = {x : [u, x) U }.
We claim A 6= . To see this, use the fact that U is open to find (0, v u)
such that (u , u + ) U . Then u < u + /2 < v, so u + /2 A.
Define w = lub A.
Since v V it is evident u < w v and w S.
If w U , then u < w < v and there is (0, v w) such that (w , w +) U
and [u, w + ) S because w + < v. This clearly contradicts the definition of w,
so w / U.
If w V , then there is an > 0 such that (w , w] V . In particular, this
shows w = lub A w < w. This contradiction forces the conclusion that w / V.
Now, putting all this together, we see w S U V and w / U V . This is a
clear contradiction, so were forced to conclude there is no separation of S. 

3. Covering Properties and Compactness on R

3.1. Open Covers.
Definition 5.15. LetSS R. A collection of open sets, O = {G : }, is
an open cover of S if S GO G. If O0 O is also an open cover of S, then O0 is
an open subcover of S from O.
Example 5.8. Let S = (0, 1) and O = {(1/n, 1) : n N}. We claim that O is
an open cover of S. To prove this, let x (0, 1). Choose n0 N such that 1/n0 < x.
Then [ [
x (1/n0 , 1) (1/n, 1) = G.
nN GO
S
Since x is an arbitrary element of (0, 1), it follows that (0, 1) = GO G.
Suppose O0 is any infinite subset of O and x (0, 1). Since O0 is infinite, there
exists an n N such that x (1/n, 1) O0 . The rest of the proof proceeds as
above.
On the other hand, if O0 is a finite subset
S of O, then let M = max{n : (1/n, 1)
O0 }. If 0 < x < 1/M , it is clear that x / GO0 G, so O0 is not an open cover of
(0, 1).

2May 1, 2015 Lee

c Larson ([email protected])

May 1, 2015 http://math.louisville.edu/lee/ira

5-6 CHAPTER 5. THE TOPOLOGY OF R

Example 5.9. Let T = [0, 1) and 0 < < 1. If

O = {(1/n, 1) : n N} {(, )},
then O is an open cover of T .
It is evident that any open subcover of T from O must contain (, ), because
that is the only element of O which contains 0. Choose n N such that 1/n < .
Then O0 = {(, ), (1/n, 1)} is an open subcover of T from O which contains only
two elements.

Theorem 5.16 (Lindelof Property). If S R and O is any open cover of S,

then O contains a subcover with a countable number of elements.
Proof. Let O = {G : } be an open cover of S R. Since O is an open
cover of S, for each x S there is a x and numbers px , qx Q satisfying
x (px , qx ) Gx O. The collection T = {(px , qx ) : x S} is an open cover of
S.
Thinking of the collection T = {(px , qx ) : x S} as a set of ordered pairs of
rational numbers, it is seen that card (T ) card (Q Q) = 0 , so T is countable.
For each interval I T , choose a I such that I GI . Then
[ [
S I GI
IT IT

shows O = {GI : I T } O is an open subcover of S from O. Also, card (O0 )

0

card (T ) 0 , so O0 is a countable open subcover of S from O. 

Corollary 5.17. Any open subset of R can be written as a countable union of

pairwise disjoint open intervals.
Proof. Let G be open in R. For x G let x = glb {y : (y, x] G} and
x = lub {y : [x, y) G}. The fact that G is open implies x < x < x . Define
Ix = (x , x ).
Then Ix G. To see this, suppose x < w < x . Choose y (w, x ). The
definition of x guarantees w (x, y) G. Similarly, if x < w < x, it follows that
w G. S
This shows O = {Ix : x G} has the property that G = xG Ix .
Suppose x, y G and Ix Iy 6= . There is no generality lost in assuming
x < y. In this case, there must be a w (x, y) such that w Ix Iy . We know
from above that both [x, w] G and [w, y] G, so [x, y] G. It follows that
x = y < x < y < x = y and Ix = Iy .
From this we conclude O consists of pairwise disjoint open intervals.
To finish, apply Theorem 5.16 to extract a countable subcover from O. 

Corollary 5.17 can also be proved by a different strategy. Instead of using

Theorem 5.16 to extract a countable subcover, we could just choose one rational
number from each interval in the cover. The pairwise disjointness of the intervals in
the cover guarantee this will give a bijection between O and a subset of Q. This
method has the advantage of showing O itself is countable from the start.

3.2. Compact Sets. There is a class of sets for which the conclusion of
Lindelofs theorem can be strengthened.

May 1, 2015 http://math.louisville.edu/lee/ira

Section 3: Covering Properties and Compactness on R 5-7

Definition 5.18. An open cover O of a set S is a finite cover, if O has only a

finite number of elements. The definition of a finite subcover is analogous.

Definition 5.19. A set K R is compact, if every open cover of K contains a

finite subcover.

Theorem 5.20 (Heine-Borel). A set K R is compact iff it is closed and

bounded.
Proof. () Suppose K is unbounded. The collection S O = {(n, n) : n N}
is an open cover of K. If O0 is any finite subset of O, then GO0 G is a bounded
set and cannot cover the unbounded set K. This shows K cannot be compact, and
every compact set must be bounded.
Suppose K is not closed. According to Theorem 5.9, there is a limit point x of K
c
such that x S O = {[x 1/n, x + 1/n] : n N}. Then O is a collection of
/ K. Define
open sets and K GO G = R \ {x}. Let O0 = {[x 1/ni , x + 1/ni ]c : 1 i N }
be a finite subset of O and M = max{ni : 1 i NS}. Since x is a limit point of K,
there is a y K (x 1/M, x + 1/M ). Clearly, y / GO0 G = [x 1/M, x + 1/M ]c ,
so O0 cannot cover K. This shows every compact set must be closed.
() Let K be closed and bounded and let O be an open cover of K. Applying
Theorem 5.16, if necessary, we can assume O is countable. Thus, O = {Gn : n N}.
For each n N, define
[n n
\
Fn = K \ Gi = K Gci .
i=1 i=1
Then Fn is a sequence of nested, bounded and closed subsets of K. Since O covers
K, it follows that \ [
Fn K \ Gn = .
nN nN
According to the Corollary
Sn 5.11, the only way this can happen is if Fn = for some
n N. Then K i=1 Gi , and O0 = {Gi : 1 i n} is a finite subcover of K
from O. 
Compactness shows up in several different, but equivalent ways on R. Weve
already seen several of them, but their equivalence is not obvious. The following
theorem shows a few of the most common manifestations of compactness.
Theorem 5.21. Let K R. The following statements are equivalent to each
other.
(a) K is compact.
(b) K is closed and bounded.
(c) Every infinite subset of K has a limit point in K.
(d) Every sequence {an : n N} K has a subsequence converging to an
element of K.
(e) If
T Fn is a nested sequence of nonempty relatively closed subsets of K, then
nN Fn 6= .

Proof. (a) (b) is the Heine-Borel Theorem, Theorem 5.20.

That (b)(c) is the Bolzano-Weierstrass Theorem, Theorem 5.8.
(c)(d) is contained in the sequence version of the Bolzano-Weierstrass theorem,
Theorem 3.16.

May 1, 2015 http://math.louisville.edu/lee/ira

5-8 CHAPTER 5. THE TOPOLOGY OF R

(d)(e) Let Fn be as in (e). For each n N, choose an Fn . By assumption,

an has a convergent subsequence bn b. Each T Fn contains a tail of the sequence
bn , so b Fn0 Fn for each n. Therefore, b nN Fn , and (e) follows.
(e)(b). Suppose K is such that (e) is true.
Let Fn = K ((, n] [n, )). T Then Fn is a sequence of sets which
are relatively closed in K such that nN Fn = . If K is unbounded, then
Fn 6= , n N, and a contradiction of (e) is evident. Therefore, K must be
bounded.
If K is not closed, then there must be a limit point x of K such that x / K.
Define a sequence of relatively
T closed and nested subsets of K by F n = [x 1/n, x+
1/n] K for n N. Then nN Fn = , because x / K. This contradiction of (e)
shows that K must be closed. 
These various ways of looking at compactness have been given different names by
topologists. Property (c) is called limit point compactness and (d) is called sequential
compactness. There are topological spaces in which various of the equivalences do
not hold.

4. More Small Sets

This is an advanced section that can be omitted.
Weve already seen one way in which a subset of R can be considered smallif
its cardinality is at most 0 . Such sets are small in the set-theoretic sense. This
section shows how sets can be considered small in the metric and topological senses.
4.1. Sets of Measure Zero. An interval is the only subset of R for which
most people could immediately come up with some sort of measure its length.
This idea of measuring a set by length can be generalized. For example, we know
every open set can be written as a countable union of open intervals, so it is natural
to assign the sum of the lengths of its component intervals as the measure of the
set. Discounting some technical difficulties, such as components with infinite length,
this is how the Lebesgue measure of an open set is defined. It is possible to assign a
measure to more complicated sets, but well only address the special case of sets
with measure zero, sometimes called Lebesgue null sets.
Definition 5.22. A set S R has measure zero if given any > 0 there is a
sequence (an , bn ) of open intervals such that
[
X
S (an , bn ) and (bn an ) < .
nN n=1

Such sets are small in the metric sense.

Example 5.10. If S = {a} contains only one point, then S has measure zero.
To see this, let > 0. Note that S (a /4, a + /4) and this single interval has
length /2 < .
There are complicated sets of measure zero, as well see later. For now, well
start with a simple theorem.

S 5.23. If {Sn : n N} is a countable collection of sets of measure

Theorem
zero, then nN Sn has measure zero.

May 1, 2015 http://math.louisville.edu/lee/ira

4. MORE SMALL SETS 5-9

Proof. Let > 0. For each n, let {(an,k , bn,k ) : k N} be a collection of

intervals such that

[ X
Sn (an,k , bn,k ) and (bn,k an,k ) < n .
2
kN k=1
Then
X

[ [ [ X X
Sn (an,k , bn,k ) and (bn,k an,k ) < n
= .
n=1 k=1 n=1
2
nN nN kN

Combining this with Example 5.10 gives the following corollary.
Corollary 5.24. Every countable set has measure zero.
The rational numbers is a large set in the sense that every interval contains a
rational number. But we now see it is small in both the set theoretic and metric
senses because it is countable and of measure zero.
Uncountable sets of measure zero are constructed in Section 4.3.
There is some standard terminology associated with sets of measure zero. If a
property P is true, except on a set of measure zero, then it is often said P is true
almost everywhere or almost every point satisfies P . It is also said P is true on
a set of full measure. For example, Almost every real number is irrational. or
The irrational numbers are a set of full measure.
4.2. Dense and Nowhere Dense Sets. We begin by considering a way that
a set can be considered topologically large in an interval. If I is any interval, recall
from Corollary 2.22 that I Q = 6 and I Qc 6= . An immediate consequence of
this is that every real number is a limit point of both Q and Qc . In this sense, the
rational and irrational numbers are both uniformly distributed across the number
line. This idea is generalized in the following definition.
Definition 5.25. Let A B R. A is said to be dense in B, if B A.
Both the rational and irrational numbers are dense in every interval. Corollary
5.17 shows the rational and irrational numbers are dense in every open set. Its
not hard to construct other sets dense in every interval. For example, the set of
dyadic numbers, D = {p/2q : p, q Z}, is dense in every interval and dense in
the rational numbers.
On the other hand, Z is not dense in any interval because its closed and contains
no interval. If A B, where B is an open set, then A is not dense in B, if A
contains any interval-sized gaps.
Theorem 5.26. Let A B R. A is dense in B iff whenever I is an open
interval such that I B 6= , then I A 6= .
Proof. () Assume there is an open interval I such that I B 6= and
I A = . If x I B, then I is a neighborhood of x that does not intersect A.
Definition 5.5 shows x / A0 A, a contradiction of the assumption that B A.
This contradiction implies that whenever I B 6= , then I A 6= .
() If x B A = A, then x A. Assume x B \ A. By assumption, for
each n N, there is an xn (x 1/n, x + 1/n) A. Since xn x, this shows
x A0 A. It now follows that B A. 

May 1, 2015 http://math.louisville.edu/lee/ira

5-10 CHAPTER 5. THE TOPOLOGY OF R

If B R and I is an open interval with I B =6 , then I B is often called a

portion of B. The previous theorem says that A is dense in B iff every portion of B
intersects A.
If A being dense in B is thought of as A being a large subset of B, then perhaps
when A is not dense in B, it can be thought of as a small subset. But, thinking of
A as being small when it is not dense isnt quite so clear when it is noticed that A
could still be dense in some portion of B, even if it isnt dense in B. To make A
be a truly small subset of B in the topological sense, it should not be dense in any
portion of B. The following definition gives a way to assure this is true.
Definition 5.27. Let A B R. A is said to be nowhere dense in B if B \ A
is dense in B.
The following theorem shows that a nowhere dense set is small in the sense
mentioned above because it fails to be dense in any part of B.
Theorem 5.28. Let A B R. A is nowhere dense in B iff for every open
interval I such that I B 6= , there is an open interval J I such that J B =
6
and J A = .
Proof. () Let I be an open interval such that I B 6= . By assumption,
B \ A is dense in B, so Theorem 5.26 implies I (B \ A) 6= . If x I (B \ A),
then there is an open interval J such that x J I and J A = . Since A A,
this J satisfies the theorem.
() Let I be an open interval with I B 6= . By assumption, there is an open
interval J I such that J A = . It follows that J A = . Theorem 5.26 implies
B \ A is dense in B. 
Example 5.11. Let G be an open set that is dense in R. If I is any open
interval, then Theorem 5.26 implies I G =6 . Because G is open, if x I G,
then there is an open interval J such that x J G. Now, Theorem 5.28 shows
Gc is nowhere dense.
The nowhere dense sets are topologically small in the following sense.
Theorem 5.29 (Baire). If I is an open interval, then I cannot be written as a
countable union of nowhere dense sets.
Proof. Let An be a sequence of nowhere dense subsets of I. According to
Theorem 5.28, there is a bounded open interval J1 I such that J1 A1 = . By
shortening J1 a bit at each end, if necessary, it may be assumed that J1 A1 = .
Assume Jn has been chosen for some n N. Applying Theorem 5.28 again, choose
an open interval Jn+1 as above so Jn+1 An+1 = . Corollary 5.11 shows
[ \
I\ An Jn 6=
nN nN
and the theorem follows. 
Theorem 5.29 is called the Baire category theorem because of the terminology
introduced by Rene-Louis Baire in 1899.3 He said a set was of the first category, if
it could be written as a countable union of nowhere dense sets. An easy example of
3Rene-Louis Baire (1874-1932) was a French mathematician. He proved the Baire category
theorem in his 1899 doctoral dissertation.

May 1, 2015 http://math.louisville.edu/lee/ira

4. MORE SMALL SETS 5-11

such a set is any countable set, which is a countable union of singletons. All other
sets are of the second category.4 Theorem 5.29 can be stated as Any open interval
is of the second category. Or, more generally, as Any nonempty open set is of the
second category.
A set is called a G set, if it is the countable intersection of open sets. It is
called an F set, if it is the countable union of closed sets. DeMorgans laws show
that the complement of an F set is a G set and vice versa. Its evident that any
countable subset of R is an F set, so Q is an F set.
On the other hand,
T suppose Q is a G set. Then there is a sequence of open sets
Gn such that Q = nN Gn . Since Q is dense, each Gn must be dense S and Example
5.11 shows Gcn is nowhere dense. From DeMorgans law, R = Q nN Gcn , showing
R is a first category set and violating the Baire category theorem. Therefore, Q is
not a G set.
Essentially the same argument shows any countable subset of R is a first category
set. The following protracted example shows there are uncountable sets of the first
category.

4.3. The Cantor Middle-Thirds Set. One particularly interesting example

of a nowhere dense set is the Cantor Middle-Thirds set, introduced by the German
mathematician Georg Cantor in 1884.5 It has many strange properties, only a few
of which will be explored here.
To start the construction of the Cantor Middle-Thirds set, let C0 = [0, 1] and
C1 = I1 \ (1/3, 2/3) = [0, 1/3] [2/3, 1]. Remove the open middle thirds of the
intervals comprising C1 , to get
       
1 2 1 2 7 8
C2 = 0, , , ,1 .
9 9 3 3 9 9
Continuing in this way, if Cn consists of 2n pairwise disjoint closed intervals each of
length 3n , construct Cn+1 by removing the open middle third from each of those
closed intervals, leaving 2n+1 closed intervals each of length 3(n+1) . This gives a
nested sequence of closed sets Cn each consisting of 2n closed intervals of length
3n . (See Figure 1.) The Cantor Middle-Thirds set is
\
C= Cn .
nN

Corollaries 5.4 and 5.11 show C is closed and nonempty. In fact, the latter is
apparent because {0, 1/3, 2/3, 1} Cn for every n. At each step in the construction,
2n open middle thirds, each of length 3(n+1) were removed from the intervals
comprising Cn . The total length of the open intervals removed was
 n
X 2n 1X 2
= = 1.
n=0
3n+1 3 n=0 3
Because of this, Example 5.11 implies C is nowhere dense in [0, 1].

4Baire did not define any categories other than these two. Some authors call first category
sets meager sets, so as not to make students fruitlessly wait for definitions of third, fourth and
fifth category sets.
5Cantors original work [6] is reprinted with an English translation in Edgars Classics on
Fractals [9]. Cantor only mentions his eponymous set in passing and it had actually been presented
earlier by others.

May 1, 2015 http://math.louisville.edu/lee/ira

5-12 CHAPTER 5. THE TOPOLOGY OF R

Figure 1. Shown here are the first few steps in the construction of the
Cantor Middle-Thirds set.

C is an example of a perfect set; i.e., a closed set all of whose points are limit
points of itself. (See Exercise 5.23.) Any closed set without isolated points is perfect.
The Cantor Middle-Thirds set is interesting because it is an example of a perfect
set without any interior points. Many people call any bounded perfect set without
interior points a Cantor set. Most of the time, when someone refers to the Cantor
set, they mean C.
There is another way to view the Cantor set. Notice that at the nth stage of
the construction, removing the middle thirds of the intervals comprising Cn removes
those points whose base 3 representation contains the digit 1 in position n + 1.
Then,

( )
X cn
(47) C= c= : cn {0, 2} .
n=1
3n
So, C consists of all numbers c [0, 1] that can be written in base 3 without using
the digit 1.6 P
If c C, then (47) shows c = n=1 3cnn for some sequence cn with range in
{0, 2}. Moreover, every such sequence corresponds to a unique element of C. Define
: C [0, 1] by

!
X cn X cn /2
(48) (c) = n
= .
n=1
3 n=1
2n
Since cn is a sequence from {0, 2}, then cn /2 is a sequence from {0, 1} and (c) can
be considered the binary representation of a number in [0, 1]. According to (47), it
follows that is a surjection and
( ) ( )
X cn /2 X bn
(C) = : cn {0, 2} = : bn {0, 1} = [0, 1].
n=1
2n n=1
2n
Therefore, card (C) = card ([0, 1]) > 0 .
The Cantor set is topologically small because it is nowhere dense and large from
the set-theoretic viewpoint because it is uncountable.
The Cantor set is also a set of measure zero. To see this, let Cn be as in the
construction of the Cantor set given above. Then C Cn and Cn consists of 2n
pairwise disjoint closed intervals each of length 3n . Their total length is (2/3)n .
Given > 0, choose n N so (2/3)n < /2. Each of the closed intervals comprising

6Notice that 1 = P 2/3n , 1/3 = P 2/3n , etc.

n=1 n=2

May 1, 2015 http://math.louisville.edu/lee/ira

5. EXERCISES 5-13

Cn can be placed inside a slightly longer open interval so the sums of the lengths of
the 2n open intervals is less than .

5. Exercises

5.1. If G is an open set and F is a closed set, then G \ F is open and F \ G is

closed.
T
5.2. Let S R and F = {F : F is closed and S F }. Prove S = F F F . This
proves that S is the smallest closed set containing S.

5.3. If S and T are subsets of R, then S T = S T .

5.4. If S is a finite subset of R, then S is closed.

5.5. Q is neither open nor closed.

5.6. A set S R is open iff S S = . (S is the set of boundary points of S.)

T
5.7. Find a sequence of open sets Gn such that nN Gn is neither open nor
closed.

5.8. An open set G is called regular if G = (G) . Find an open set that is not
regular.

5.9. Let R = {(x, ) : x R} and T = R {R, }. Prove that (R, T ) is a

topological space. This is called the right ray topology on R.

5.10. If X R and S is the collection of all sets relatively open in X, then (X, S)
is a topological space.

5.11. If X R and G is an open set, then X \ G is relatively closed in X.

5.12. For any set S, let F = {T S : card (S \ T ) 0 } {}. Then (S, F) is a

topological space. This is called the finite complement topology.

5.13. An uncountable subset of R must have a limit point.

5.14. If S R, then S 0 is closed.

5.15. Prove that the set of accumulation points of any sequence is closed.

5.16. Prove any closed set is the set of accumulation points for some sequence.

5.17. If an is a sequence such that an L, then {an : n N} {L} is compact.

5.18. If F is closed and K is compact, then F K is compact.

T
5.19. If {K : A} is a collection of compact sets, then A K is compact.

May 1, 2015 http://math.louisville.edu/lee/ira

5-14 CHAPTER 5. THE TOPOLOGY OF R

5.20. Prove the union of a finite number of compact sets is compact. Give an
example to show this need not be true for the union of an infinite number of compact
sets.
5.21. (a) Give an example of a set S such that S is disconnected, but S {1} is
connected. (b) Prove that 1 must be a limit point of S.

5.22. If K is compact and V is open with K V , then there is an open set U

such that K U U V .
5.23. If C is the Cantor middle-thirds set, then C = C 0 .

5.24. If x R and K is compact, then there is a z K such that |x z| =

glb{|x y| : y K}. Is z unique?

5.25. If K is compact and O is an open cover of K, then there is an > 0 such

that for all x K there is a G O with (x , x + ) G.

5.26. Let f : [a, b] R be a function such that for every x [a, b] there is a
x > 0 such that f is bounded on (x x , x + x ). Prove f is bounded.

5.27. Is the function defined by (48) a bijection?

5.28. If A is nowhere dense in an interval I, then A contains no interval.

5.29. Use the Baire category theorem to show R is uncountable.

May 1, 2015 http://math.louisville.edu/lee/ira

 CHAPTER 6

Limits of Functions

1. Basic Definitions
Definition 6.1. Let D R, x0 be a limit point of D and f : D R. The
limit of f (x) at x0 is L, if for each > 0 there is a > 0 such that when x D
with 0 < |x x0 | < , then |f (x) L| < . When this is the case, we write
limxx0 f (x) = L.
It should be noted that the limit of f at x0 is determined by the values of f
near x0 and not at x0 . In fact, f need not even be defined at x0 .

Figure 1. This figure shows a way to think about the limit. The graph
of f must stay inside the box (x0 , x0 + ) (L , L + ), except
possibly the point (x0 , f (x0 )).

A useful way of rewording the definition is to say that limxx0 f (x) = L iff
for every > 0 there is a > 0 such that x (x0 , x0 + ) D \ {x0 } implies
f (x) (L , L + ). This can also be succinctly stated as
> 0 > 0 ( f ( (x0 , x0 + ) D \ {x0 } ) (L , L + ) ) .
Example 6.1. If f (x) = c is a constant function and x0 R, then for any
positive numbers and ,
x (x0 , x0 + ) D \ {x0 } |f (x) c| = |c c| = 0 < .
This shows the limit of every constant function exists at every point, and the limit
is just the value of the function.

1
May 1, 2015 Lee
c Larson ([email protected])

6-1
6-2 CHAPTER 6. LIMITS OF FUNCTIONS

-2 2

Figure 2. The function from Example 6.3. Note that the graph is a
line with one hole in it.

Example 6.2. Let f (x) = x, x0 R, and = > 0. Then

x (x0 , x0 + ) D \ {x0 } |f (x) f (x0 )| = |x x0 | < = .
This shows that the identity function has a limit at every point and its limit is just
the value of the function at that point.
2
8
Example 6.3. Let f (x) = 2xx2 . In this case, the implied domain of f is
D = R \ {2}. We claim that limx2 f (x) = 8.
To see this, let > 0 and choose (0, /2). If 0 < |x 2| < , then
 2 
 2x 8 
|f (x) 8| =   8 = |2(x + 2) 8| = 2|x 2| < .
x2

Example 6.4. Let f (x) = x + 1. Then the implied domain of f is D =
[1, ). We claim that limx1 f (x) = 0.
To see this, let > 0 and choose (0, 2 ). If 0 < x (1) = x + 1 < , then

|f (x) 0| = x + 1 < < 2 = .

Figure 3. The function f (x) = |x|/x from Example 6.5.

Example 6.5. If f (x) = |x|/x for x =

6 0, then limx0 f (x) does not exist. (See
Figure 3.) To see this, suppose limx0 f (x) = L, = 1 and > 0. If L 0 and
< x < 0, then f (x) = 1 < L. If L < 0 and 0 < x < , then f (x) = 1 > L+.
These inequalities show for any L and every > 0, there is an x with 0 < |x| <
such that |f (x) L| > .

May 1, 2015 http://math.louisville.edu/lee/ira

1. BASIC DEFINITIONS 6-3

1.0

0.5

0.2 0.4 0.6 0.8 1.0

-0.5

-1.0

Figure 4. This is the function from Example 6.6. The graph shown
here is on the interval [0.03, 1]. There are an infinite number of oscillations
from 1 to 1 on any open interval containing the origin.

There is an obvious similarity between the definition of limit of a sequence and

limit of a function. The following theorem makes this similarity explicit, and gives
another way to prove facts about limits of functions.
Theorem 6.2. Let f : D R and x0 be a limit point of D. limxx0 f (x) = L
iff whenever xn is a sequence from D \ {x0 } such that xn x0 , then f (xn ) L.
Proof. () Suppose limxx0 f (x) = L and xn is a sequence from D \ {x0 }
such that xn x0 . Let > 0. There exists a > 0 such that |f (x) L| <
whenever x (x , x + ) D \ {x0 }. Since xn x0 , there is an N N such that
n N implies 0 < |xn x0 | < . In this case, |f (xn ) L| < , showing f (xn ) L.
() Suppose whenever xn is a sequence from D \ {x0 } such that xn x0 ,
then f (xn ) L, but limxx0 f (x) 6= L. Then there exists an > 0 such that for
all > 0 there is an x (x0 , x0 + ) D \ {x0 } such that |f (x) L| . In
particular, for each n N, there must exist xn (x0 1/n, x0 + 1/n) D \ {x0 }
such that |f (xn ) L| . Since xn x0 , this is a contradiction. Therefore,
limxx0 f (x) = L. 
Theorem 6.2 is often used to show a limit doesnt exist. Suppose we want to
show limxx0 f (x) doesnt exist. There are two strategies: find a sequence xn x0
such that f (xn ) has no limit; or, find two sequences yn x0 and zn x0 such
that f (yn ) and f (zn ) converge to different limits. Either way, the theorem shows
limxx0 fails to exist.
In Example 6.5, we could choose xn = (1)n /n so f (xn ) oscillates between 1
and 1. Or, we could choose yn = 1/n = zn so f (xn ) 1 and f (zn ) 1.
1 2
Example 6.6. Let f (x) = sin(1/x), an = n and bn = (4n+1) . Then an 0,
bn 0, f (an ) = 0 and f (bn ) = 1 for all n N. An application of Theorem 6.2 shows
limx0 f (x) does not exist. (See Figure 4.)
Theorem 6.3 (Squeeze Theorem). Suppose f , g and h are all functions defined
on D R with f (x) g(x) h(x) for all x D. If x0 is a limit point of D and
limxx0 f (x) = limxx0 h(x) = L, then limxx0 g(x) = L.
Proof. Let xn be any sequence from D \ {x0 } such that xn x0 . According
to Theorem 6.2, both f (xn ) L and h(xn ) L. Since f (xn ) g(xn ) h(xn ), an
application of the Sandwich Theorem for sequences shows g(xn ) L. Now, another
use of Theorem 6.2 shows limxx0 g(x) = L. 

May 1, 2015 http://math.louisville.edu/lee/ira

6-4 CHAPTER 6. LIMITS OF FUNCTIONS

0.10

0.05

0.02 0.04 0.06 0.08 0.10

-0.05

-0.10

Figure 5. This is the function from Example 6.7. The bounding lines
y = x and y = x are also shown. There are an infinite number of
oscillations between x and x on any open interval containing the origin.

Example 6.7. Let f (x) = x sin(1/x). Since 1 sin(1/x) 1 when x 6= 0, we

see that x x sin(1/x) x for x 6= 0. Since limx0 x = limx0 x = 0, Theorem
6.3 implies limx0 x sin(1/x) = 0. (See Figure 5.)

Theorem 6.4. Suppose f : D R and g : D R and x0 is a limit point of

D. If limxx0 f (x) = L and limxx0 g(x) = M , then
(a) limxx0 (f + g)(x) = L + M ,
(b) limxx0 (af )(x) = aL, x R,
(c) limxx0 (f g)(x) = LM , and
(d) limxx0 (1/f )(x) = 1/L, as long as L 6= 0.

Proof. Suppose an is a sequence from D \ {x0 } converging to x0 . Then

Theorem 6.2 implies f (an ) L and g(an ) M . (a)-(d) follow at once from the
corresponding properties for sequences. 

Example 6.8. Let f (x) = 3x + 2. If g1 (x) = 3, g2 (x) = x and g3 (x) = 2, then

f (x) = g1 (x)g2 (x) + g3 (x). Examples 6.1 and 6.2 along with parts (a) and (c) of
Theorem 6.4 immediately show that for every x R, limxx0 f (x) = f (x0 ).

In the same manner as Example 6.8, it can be shown for every rational function
f (x), that limxx0 f (x) = f (x0 ) whenever f (x0 ) exists.

2. Unilateral Limits
Definition 6.5. Let f : D R and x0 be a limit point of (, x0 ) D.
f has L as its left-hand limit at x0 if for all > 0 there is a > 0 such that
f ((x0 , x0 ) D) (L , L + ). In this case, we write limxx0 f (x) = L.

May 1, 2015 http://math.louisville.edu/lee/ira

Section 3: Continuity 6-5

Let f : D R and x0 be a limit point of D (x0 , ). f has L as its right-hand

limit at x0 if for all > 0 there is a > 0 such that f (D (x0 , x0 +)) (L, L+).
In this case, we write limxx0 f (x) = L.2
These are called the unilateral or one-sided limits of f at x0 . When they are
different, the graph of f is often said to have a jump at x0 , as in the following
example.
Example 6.9. As in Example 6.5, let f (x) = |x|/x. Then limx0 f (x) = 1 and
limx0 f (x) = 1. (See Figure 3.)
In parallel with Theorem 6.2, the one-sided limits can also be reformulated in
terms of sequences.
Theorem 6.6. Let f : D R and x0 .
(a) Let x0 be a limit point of D (x0 , ). limxx0 f (x) = L iff whenever xn
is a sequence from D (x0 , ) such that xn x0 , then f (xn ) L.
(b) Let x0 be a limit point of (, x0 ) D. limxx0 f (x) = L iff whenever xn
is a sequence from (, x0 ) D such that xn x0 , then f (xn ) L.
The proof of Theorem 6.6 is similar to that of Theorem 6.2 and is left to the
reader.
Theorem 6.7. Let f : D R and x0 be a limit point of D.
lim f (x) = L lim f (x) = L = lim f (x)
xx0 xx0 xx0

Proof. This proof is left as an exercise. 

Theorem 6.8. If f : (a, b) R is monotone, then both unilateral limits of f
exist at every point of (a, b).
Proof. To be specific, suppose f is increasing and x0 (a, b). Let > 0
and L = lub {f (x) : a < x < x0 }. According to Corollary 2.18, there must exist an
x (a, x0 ) such that L < f (x) L. Define = x0 x. If y (x0 , x0 ), then
L < f (x) f (y) L. This shows limxx0 f (x) = L.
The proof that limxx0 f (x) exists is similar.
To handle the case when f is decreasing, consider f instead of f . 

3. Continuity
Definition 6.9. Let f : D R and x0 D. f is continuous at x0 if for
every > 0 there exists a > 0 such that when x D with |x x0 | < , then
|f (x) f (x0 )| < . The set of all points at which f is continuous is denoted C(f ).
Several useful ways of rephrasing this are contained in the following theorem.
They are analogous to the similar statements made about limits. Proofs are left to
the reader.
Theorem 6.10. Let f : D R and x0 D. The following statements are
equivalent.
(a) x0 C(f ),
2 Calculus books often use the notation lim
xx0 f (x) = limxx0 f (x) and limxx0 f (x) =
limxx0 + f (x).

May 1, 2015 http://math.louisville.edu/lee/ira

6-6 CHAPTER 6. LIMITS OF FUNCTIONS

Figure 6. The function f is continuous at x0 , if given any > 0 there

is a > 0 such that the graph of f does not cross the top or bottom of
the dashed rectangle (x0 , x0 + d) (f (x0 ) , f (x0 ) + ).

(b) For all > 0 there is a > 0 such that

x (x0 , x0 + ) D f (x) (f (x0 ) , f (x0 ) + ),
(c) For all > 0 there is a > 0 such that
f ((x0 , x0 + ) D) (f (x0 ) , f (x0 ) + ).
Example 6.10. Define
(
2x2 8
x2 , x 6= 2
f (x) = .
8, x=2
It follows from Example 6.3 that 2 C(f ).
There is a subtle difference between the treatment of the domain of the function
in the definitions of limit and continuity. In the definition of limit, the target
point, x0 is required to be a limit point of the domain, but not actually be an
element of the domain. In the definition of continuity, x0 must be in the domain of
the function, but does not have to be a limit point. To see a consequence of this
difference, consider the following example.
Example 6.11. If f : Z R is an arbitrary function, then C(f ) = Z. To see
this, let n0 Z, > 0 and = 1. If x Z with |x n0 | < , then x = n0 . It follows
that |f (x) f (n0 )| = 0 < , so f is continuous at n0 .
This leads to the following theorem.
Theorem 6.11. Let f : D R and x0 D. If x0 is a limit point of D, then
x0 C(f ) iff limxx0 f (x) = f (x0 ). If x0 is an isolated point of D, then x0 C(f ).
Proof. If x0 is isolated in D, then there is an > 0 such that (x0 , x0 +
) D = {x0 }. For any > 0, the definition of continuity is satisfied with this .
Next, suppose x0 D0 .
The definition of continuity says that f is continuous at x0 iff for all > 0 there
is a > 0 such that when x (x0 , x0 + ) D, then f (x) (f (x0 ) , f (x0 ) + ).
The definition of limit says that limxx0 f (x) = f (x0 ) iff for all > 0 there is a
> 0 such that when x (x0 , x0 +)D\{x0 }, then f (x) (f (x0 ), f (x0 )+).

May 1, 2015 http://math.louisville.edu/lee/ira

Section 3: Continuity 6-7

Comparing these two definitions, it is clear that x0 C(f ) implies

lim f (x) = f (x0 ).
xx0

On the other hand, suppose limxx0 f (x) = f (x0 ) and > 0. Choose
according to the definition of limit. When x (x0 , x0 + ) D \ {x0 }, then
f (x) (f (x0 ) , f (x0 ) + ). It follows from this that when x = x0 , then f (x)
f (x0 ) = f (x0 ) f (x0 ) = 0 < . Therefore, when x (x0 , x0 + ) D, then
f (x) (f (x0 ) , f (x0 ) + ), and x0 C(f ), as desired. 
Example 6.12. If f (x) = c, for some c R, then Example 6.1 and Theorem
6.11 show that f is continuous at every point.
Example 6.13. If f (x) = x, then Example 6.2 and Theorem 6.11 show that f
is continuous at every point.
Corollary 6.12. Let f : D R and x0 D. x0 C(f ) iff whenever xn is a
sequence from D with xn x0 , then f (xn ) f (x0 ).
Proof. Combining Theorem 6.11 with Theorem 6.2 shows this to be true. 

Example 6.14 (Dirichlet Function). Suppose

(
1, x Q
f (x) = .
0, x
/Q
For each x Q, there is a sequence of irrational numbers converging to x, and for
each y Qc there is a sequence of rational numbers converging to y. Corollary 6.12
shows C(f ) = .

Example 6.15 (Salt and Pepper Function). Since Q is a countable set, it can
be written as a sequence, Q = {qn : n N}. Define
(
1/n, x = qn ,
f (x) =
0, x Qc .
If x Q, then x = qn , for some n and f (x) = 1/n > 0. There is a sequence xn
from Qc such that xn x and f (xn ) = 0 6 f (x) = 1/n. Therefore C(f ) Q = .
On the other hand, let x Qc and > 0. Choose N N large enough so
that 1/N < and let = min{|x qn | : 1 n N }. If |x y| < , there
are two cases to consider. If y Qc , then |f (y) f (x)| = |0 0| = 0 < . If
y Q, then the choice of guarantees y = qn for some n > N . In this case,
|f (y) f (x)| = f (y) = f (qn ) = 1/n < 1/N < . Therefore, x C(f ).
This shows that C(f ) = Qc .
It is a consequence of the Baire category theorem that there is no function f
such that C(f ) = Q. Proving this would take us too far afield.
The following theorem is an almost immediate consequence of Theorem 6.4.
Theorem 6.13. Let f : D R and g : D R. If x0 C(f ) C(g), then
(a) x0 C(f + g),
(b) x0 C(f ), R,
(c) x0 C(f g), and
(d) x0 C(f /g) when g(x0 ) 6= 0.

May 1, 2015 http://math.louisville.edu/lee/ira

6-8 CHAPTER 6. LIMITS OF FUNCTIONS

Corollary 6.14. If f is a rational function, then f is continuous at each point

of its domain.
Proof. This is a consequence of Examples 6.12 and 6.13 used with Theorem
6.13. 
Theorem 6.15. Suppose f : Df R and g : Dg R such that f (Df ) Dg .
If there is an x0 C(f ) such that f (x0 ) C(g), then x0 C(g f ).
Proof. Let > 0 and choose 1 > 0 such that
g((f (x0 ) 1 , f (x0 ) + 1 ) Dg ) (g f (x0 ) , g f (x0 ) + ).
Choose 2 > 0 such that
f ((x0 2 , x0 + 2 ) Df ) (f (x0 ) 1 , f (x0 ) + 1 ).
Then
g f ((x0 2 , x0 + 2 ) Df ) g((f (x0 ) 1 , f (x0 ) + 1 ) Dg )
(g f (x0 ) 2 , g f (x0 ) + 2 ) Df ).
Since this shows Theorem 6.10(c) is satisfied at x0 with the function g f , it follows
that x0 C(g f ). 

Example 6.16. If f (x) = x for x 0,then according to Problem 6.8,
C(f ) = [0, ). Theorem 6.15 shows f f (x) = 4 x is continuous on [0, ).
n
In similar way, it can be shown by induction that f (x) = xm/2 is continuous
on [0, ) for all m, n Z.

4. Unilateral Continuity
Definition 6.16. Let f : D R and x0 D. f is left-continuous at x0 if for
every > 0 there is a > 0 such that f ((x0 , x0 ] D) (f (x0 ) , f (x0 ) + ).
Let f : D R and x0 D. f is right-continuous at x0 if for every > 0 there
is a > 0 such that f ([x0 , x0 + ) D) (f (x0 ) , f (x0 ) + ).

Example 6.17. Let the floor function be

bxc = max{n Z : n x}
and the ceiling function be
dxe = min{n Z : n x}.
The floor function is right-continuous, but not left-continuous at each integer, and
the ceiling function is left-continuous, but not right-continuous at each integer.
Theorem 6.17. Let f : D R and x0 D. x0 C(f ) iff f is both right and
left-continuous at x0 .
Proof. The proof of this theorem is left as an exercise. 
According to Theorem 6.7, when f is monotone on an interval (a, b), the
unilateral limits of f exist at every point. In order for such a function to be
continuous at x0 (a, b), it must be the case that
lim f (x) = f (x0 ) = lim f (x).
xx0 xx0

If either of the two equalities is violated, the function is not continuous at x0 .

May 1, 2015 http://math.louisville.edu/lee/ira

Section 4: Unilateral Continuity 6-9

7
6
5
4
3
x2 4
2 f (x) = x2

1 2 3 4

Figure 7. The function from Example 6.19. Note that the graph is a
line with one hole in it. Plugging up the hole removes the discontinuity.

In the case, when limxx0 f (x) 6= limxx0 f (x), it is said that a jump discontinuity
occurs at x0 .
Example 6.18. The function
(
|x|/x, x 6= 0
f (x) = .
0, x=0
has a jump discontinuity at x = 0.
In the case when limxx0 f (x) = limxx0 f (x) 6= f (x0 ), it is said that f has a
removable discontinuity at x0 . The discontinuity is called removable because in
this case, the function can be made continuous at x0 by merely redefining its value
at the single point, x0 , to be the value of the two one-sided limits.
2
4
Example 6.19. The function f (x) = xx2 is not continuous at x = 2 because 2
is not in the domain of f . Since limx2 f (x) = 4, if the domain of f is extended to
include 2 by setting f (2) = 4, then this extended f is continuous everywhere. (See
Figure 7.)
Theorem 6.18. If f : (a, b) R is monotone, then (a, b) \ C(f ) is countable.
Proof. In light of the discussion above and Theorem 6.7, it is apparent that
the only types of discontinuities f can have are jump discontinuities.
To be specific, suppose f is increasing and x0 , y0 (a, b) \ C(f ) with x0 < y0 .
In this case, the fact that f is increasing implies
lim f (x) < lim f (x) lim f (x) < lim f (x).
xx0 xx0 xy0 xy0

This implies that for any two x0 , y0 (a, b) \ C(f ), there are disjoint open intervals,
Ix0 = (limxx0 f (x), limxx0 f (x)) and Iy0 = (limxy0 f (x), limxy0 f (x)). For each
x (a, b) \ C(f ), choose qx Ix Q. Because of the pairwise disjointness of the
intervals {Ix : x (a, b) \ C(f )}, this defines an bijection between (a, b) \ C(f ) and
a subset of Q. Therefore, (a, b) \ C(f ) must be countable.
A similar argument holds for a decreasing function. 
Theorem 6.18 implies that a monotone function is continuous at nearly every
point in its domain. Characterizing the points of discontinuity as countable is the
best that can be hoped for, as seen in the following example.

May 1, 2015 http://math.louisville.edu/lee/ira

6-10 CHAPTER 6. LIMITS OF FUNCTIONS

Example 6.20. Let D = {dn : n N} be a countable set and define Jx = {n :

dn < x}. The function
(
0, Jx =
(49) f (x) = P 1
.
nJx 2n Jx 6=
is increasing and C(f ) = Dc . The proof of this statement is left as Exercise 6.9.

5. Continuous Functions
Up until now, continuity has been considered as a property of a function at a
point. There is much that can be said about functions continuous everywhere.
Definition 6.19. Let f : D R and A D. We say f is continuous on A if
A C(f ). If D = C(f ), then f is continuous.
Continuity at a point is, in a sense, a metric property of a function because it
measures relative distances between points in the domain and image sets. Continuity
on a set becomes more of a topological property, as shown by the next few theorems.
Theorem 6.20. f : D R is continuous iff whenever G is open in R, then
f 1 (G) is relatively open in D.
Proof. () Assume f is continuous on D and let G be open in R. Let
x f 1 (G) and choose > 0 such that (f (x) , f (x) + ) G. Using the
continuity of f at x, we can find a > 0 such that f ((x , x + ) D) G. This
implies that (x , x + ) D f 1 (G). Because x was an arbitrary element of
f 1 (G), it follows that f 1 (G) is open.
() Choose x D and let > 0. By assumption, the set f 1 ((f (x)
, f (x) + ) is relatively open in D. This implies the existence of a > 0 such
that (x , x + ) D f 1 ((f (x) , f (x) + ). It follows from this that
f ((x , x + ) D) (f (x) , f (x) + ), and x C(f ). 

A function as simple as any constant function demonstrates that f (G) need not
be open when G is open. Defining f : [0, ) R by f (x) = sin x tan1 x shows
that the image of a closed set need not be closed because f ([0, )) = (/2, /2).
Theorem 6.21. If f is continuous on a compact set K, then f (K) is compact.
Proof. Let O be an open cover of f (K) and I = {f 1 (G) : G O}. By
Theorem 6.20, I is a collection of sets which are relatively open in K. Since I covers
K, I is an open cover of K. Using the fact that K is compact, we can choose a finite
subcover of K from I, say {G1 , G2 , . . . , Gn }. There are {H1 , H2 , . . . , Hn } O such
that f 1 (Hk ) = Gk for 1 k n. Then

[ [
f (K) f Gk = Hk .
1kn 1kn

Thus, {H1 , H2 , . . . , H3 } is a subcover of f (K) from O. 

Several of the standard calculus theorems giving properties of continuous func-

tions are consequences of Corollary 6.21. In a calculus course, K is usually a compact
interval, [a, b].

May 1, 2015 http://math.louisville.edu/lee/ira

Section 5: Continuous Functions 6-11

Corollary 6.22. If f : K R is continuous and K is compact, then f is

bounded.
Proof. By Theorem 6.21, f (K) is compact. Now, use the Bolzano-Weierstrass
theorem to conclude f is bounded. 
Corollary 6.23 (Maximum Value Theorem). If f : K R is continuous and
K is compact, then there are m, M K such that f (m) f (x) f (M ) for all
x K.
Proof. According to Theorem 6.21 and the Bolzano-Weierstrass theorem, f (K)
is closed and bounded. Because of this, glb f (K) f (K) and lub f (K) f (K). It
suffices to choose m f 1 (glb f (K)) and M f 1 (lub f (K)). 
Corollary 6.24. If f : K R is continuous and invertible and K is compact,
then f 1 : f (K) K is continuous.
Proof. Let G be open in K. According to Theorem 6.20, it suffices to show
f (G) is open in f (K).
To do this, note that K \G is compact, so by Theorem 6.21, f (K \G) is compact,
and therefore closed. Because f is injective, f (G) = f (K) \ f (K \ G). This shows
f (G) is open in f (K). 
Theorem 6.25. If f is continuous on an interval I, then f (I) is an interval.
Proof. If f (I) is not connected, there must exist two disjoint open sets, U
and V , such that f (I) U V and f (I) U = 6 = 6 f (I) V . In this case, Theorem
6.20 implies f 1 (U ) and f 1 (V ) are both open. They are clearly disjoint and
f 1 (U ) I 6= 6= f 1 (V ) I. But, this implies f 1 (U ) and f 1 (V ) disconnect I,
which is a contradiction. Therefore, f (I) is connected. 
Corollary 6.26 (Intermediate Value Theorem). If f : [a, b] R is continuous
and is between f (a) and f (b), then there is a c [a, b] such that f (c) = .
Proof. This is an easy consequence of Theorem 6.25 and Theorem 5.14. 
Definition 6.27. A function f : D R has the Darboux property if whenever
a, b D and is between f (a) and f (b), then there is a c between a and b such
that f (c) = .
Calculus texts usually call the Darboux property the intermediate value property.
Corollary 6.26 shows that a function continuous on an interval has the Darboux
property. The next example shows continuity is not necessary for the Darboux
property to hold.
Example 6.21. The function
(
sin 1/x, x 6= 0
f (x) =
0, x=0
is not continuous, but does have the Darboux property. (See Figure 4.) It can be
seen from Example 6.6 that 0 / C(f ).
To see f has the Darboux property, choose two numbers a < b.
If a > 0 or b < 0, then f is continuous on [a, b] and Corollary 6.26 suffices to
finish the proof.

May 1, 2015 http://math.louisville.edu/lee/ira

6-12 CHAPTER 6. LIMITS OF FUNCTIONS

On the other hand, if 0 [a, b], then there must exist an n Z such that both
2 2 2 2
(4n+1) , (4n+3) [a, b]. Since f ( (4n+1) ) = 1, f ( (4n+3) ) = 1 and f is continuous
on the interval between them, we see f ([a, b]) = [1, 1], which is the entire range of
f . The claim now follows.

6. Uniform Continuity
Most of the ideas contained in this section will not be needed until we begin
developing the properties of the integral in Chapter 8.
Definition 6.28. A function f : D R is uniformly continuous if for all > 0
there is a > 0 such that when x, y D with |x y| < , then |f (x) f (y)| < .
The idea here is that in the ordinary definition of continuity, the in the
definition depends on both the and the x at which continuity is being tested. With
uniform continuity, only depends on ; i. e., the same works uniformly across
the whole domain.
Theorem 6.29. If f : D R is uniformly continuous, then it is continuous.
Proof. This proof is left as Exercise 6.31. 

The converse is not true.

Example 6.22. Let f (x) = 1/x on D = (0, 1) and > 0. Its clear that f is
continuous on D. Let > 0 and choose m, n N such that m > 1/ and n m > .
If x = 1/m and y = 1/n, then 0 < y < x < and f (y) f (x) = n m > .
Therefore, f is not uniformly continuous.
Theorem 6.30. If f : D R is continuous and D is compact, then f is
uniformly continuous.
Proof. Suppose f is not uniformly continuous. Then there is an > 0 such that
for every n N there are xn , yn D with |xn yn | < 1/n and |f (xn ) f (yn )| .
An application of the Bolzano-Weierstrass theorem yields a subsequence xnk of xn
such that xnk x0 D.
Since f is continuous at x0 , there is a > 0 such that whenever x (x0
, x0 + ) D, then |f (x) f (x0 )| < /2. Choose nk N such that 1/nk < /2 and
xnk (x0 /2, x0 + /2). Then both xnk , ynk (x0 , x0 + ) and
|f (xnk ) f (ynk )| = |f (xnk ) f (x0 ) + f (x0 ) f (ynk )|
|f (xnk ) f (x0 )| + |f (x0 ) f (ynk )| < /2 + /2 = ,
which is a contradiction.
Therefore, f must be uniformly continuous. 

The following corollary is an immediate consequence of Theorem 6.30.

Corollary 6.31. If f : [a, b] R is continuous, then f is uniformly continuous.
Theorem 6.32. Let D R and f : D R. If f is uniformly continuous and
xn is a Cauchy sequence from D, then f (xn ) is a Cauchy sequence..
Proof. The proof is left as Exercise 6.38. 

May 1, 2015 http://math.louisville.edu/lee/ira

7. EXERCISES 6-13

Uniform continuity is necessary in Theorem 6.32. To see this, let f : (0, 1) R

be f (x) = 1/x and xn = 1/n. Then xn is a Cauchy sequence, but f (xn ) = n is not.
This idea is explored in Exercise 6.33.
Its instructive to think about the converse to Theorem 6.32. Let f (x) = x2 ,
defined on all of R. Since f is continuous everywhere, Corollary 6.12 shows f maps
Cauchy sequences to Cauchy sequences. On the other hand, in Exercise 6.37, it
is shown that f is not uniformly continuous. Therefore, the converse to Theorem
6.32 is false. Those functions mapping Cauchy sequences to Cauchy sequences are
sometimes said to be Cauchy continuous. The converse to Theorem 6.32 can be
tweaked to get a true statement.
Theorem 6.33. Let f : D R where D is bounded. If f is Cauchy continuous,
then f is uniformly continuous.
Proof. Suppose f is not uniformly continuous. Then there is an > 0 and
sequences xn and yn from D such that |xn yn | < 1/n and |f (xn ) f (yn )| .
Since D is bounded, the sequence xn is bounded and the Bolzano-Weierstrass
theorem gives a Cauchy subsequence, xnk . The new sequence
(
xn(k+1)/2 k odd
zk =
ynk/2 k even
is easily shown to be a Cauchy sequence. But, f (zk ) is not a Cauchy sequence,
since |f (zk ) f (zk+1 )| for all odd k. This contradicts the fact that f is
Cauchy continuous. Were forced to conclude the assumption that f is not uniformly
continuous is false. 

7. Exercises

6.1. Prove lim (x2 + 3x) = 2.

x2

6.2. Give examples of functions f and g such that neither function has a limit at
a, but f + g does. Do the same for f g.

6.3. Let f : D R and a D0 .

lim f (x) = L lim f (x) = lim f (x) = L
xa xa xa

6.4. Find two functions defined on R such that

0 = lim (f (x) + g(x)) 6= lim f (x) + lim g(x).
x0 x0 x0

6.5. If lim f (x) = L > 0, then there is a > 0 such that f (x) > 0 when
xa
0 < |x a| < .

6.6. If Q = {qn : n N} is an enumeration of the rational numbers and

(
1/n, x = qn
f (x) =
0, x Qc
then limxa f (x) = 0, for all a Qc .

May 1, 2015 http://math.louisville.edu/lee/ira

6-14 CHAPTER 6. LIMITS OF FUNCTIONS

6.7. Use the definition of continuity to show f (x) = x2 is continuous everywhere

on R.

6.8. Prove that f (x) = x is continuous on [0, ).

6.9. If f is defined as in (49), then D = C(f )c .

6.10. If f : R R is monotone, then there is a countable set D such that

the values of f can be altered on D in such a way that the altered function is
left-continuous at every point of R.

6.11. Does there exist an increasing function f : R R such that C(f ) = Q?

6.12. If f : R R and there is an > 0 such that |f (x) f (y)| |x y| for all
x, y R, then show that f is continuous.

6.13. Suppose f and g are each defined on an open interval I, a I and

a C(f ) C(g). If f (a) > g(a), then there is an open interval J such that
f (x) > g(x) for all x J.

6.14. If f, g : (a, b) R are continuous, then G = {x : f (x) < g(x)} is open.

6.15. If f : R R and a C(f ) with f (a) > 0, then there is a neighborhood G

of a such that f (G) (0, ).

6.16. Let f and g be two functions which are continuous on a set D R. Prove
or give a counter example: {x D : f (x) > g(x)} is relatively open in D.

6.17. If f, g : R R are functions such that f (x) = g(x) for all x Q and
C(f ) = C(g) = R, then f = g.

6.18. Let I = [a, b]. If f : I I is continuous, then there is a c I such that

f (c) = c.

6.19. Find an example to show the conclusion of Problem 18 fails if I = (a, b).

6.20. If f and g are both continuous on [a, b], then {x : f (x) g(x)} is compact.

6.21. If f : [a, b] R is continuous, not constant,

m = glb {f (x) : a x b} and M = lub {f (x) : a x b},
then f ([a, b]) = [m, M ].

6.22. Suppose f : R R is a function such that every interval has points at which
f is negative and points at which f is positive. Prove that every interval has points
where f is not continuous.

6.23. If f : [a, b] R has a limit at every point, then f is bounded. Is this true
for f : (a, b) R?

6.24. Give an example of a bounded function f : R R with a limit at no point.

May 1, 2015 http://math.louisville.edu/lee/ira

7. EXERCISES 6-15

6.25. If f : R R is continuous and periodic, then there are xm , xM R such

that f (xm ) f (x) f (xM ) for all x R. (A function f is periodic, if there is a
p > 0 such that f (x + p) = f (x) for all x R.)

6.26. A set S R is disconnected iff there is a continuous f : S R such that

f (S) = {0, 1}.

6.27. If f : R R satisfies f (x + y) = f (x) + f (y) for all x and y and 0 C(f ),

then f is continuous.

6.28. Assume that f : R R is such that f (x + y) = f (x)f (y) for all x, y R. If f

has a limit at zero, prove that either limx0 f (x) = 1 or f (x) = 0 for all x R \ {0}.

6.29. If g : R R satisfies g(x + y) = g(x)g(y) for all x, y R and 0 C(g), then

g is continuous.

6.30. If F R is closed, then there is an f : R R such that F = C(f )c .

6.31. If f : [a, b] R is uniformly continuous, then f is continuous.

6.32. A function f : R R is periodic with period p > 0, if f (x + p) = f (x) for

all x. If f : R R is periodic with period p and continuous on [0, p], then f is
uniformly continuous.

6.33. Prove that an unbounded function on a bounded open interval cannot be

uniformly continuous.

6.34. If f : D R is uniformly continuous on a bounded set D, then f is bounded.

6.35. Prove Theorem 6.29.

6.36. Every polynomial of odd degree has a root.

6.37. Show f (x) = x2 , with domain R, is not uniformly continuous.

6.38. Prove Theorem 6.32.

May 1, 2015 http://math.louisville.edu/lee/ira

 CHAPTER 7

Differentiation

1. The Derivative at a Point

Definition 7.1. Let f be a function defined on a neighborhood of x0 . f is
differentiable at x0 , if the following limit exists:
f (x0 + h) f (x0 )
f 0 (x0 ) = lim .
h0 h
Define D(f ) = {x : f 0 (x) exists}.
df (x)
The standard notations for the derivative will be used; e. g., f 0 (x), dx , Df (x),
etc.
An equivalent way of stating this definition is to note that if x0 D(f ), then
f (x) f (x0 )
f 0 (x0 ) = lim .
xx0 x x0
(See Figure 1.)
This can be interpreted in the standard way as the limiting slope of the secant
line as the points of intersection approach each other.
Example 7.1. If f (x) = c for all x and some c R, then
f (x0 + h) f (x0 ) cc
lim = lim = 0.
h0 h h0 h

So, f 0 (x) = 0 everywhere.

Example 7.2. If f (x) = x, then
f (x0 + h) f (x0 ) x0 + h x0 h
lim = lim = lim = 1.
h0 h h0 h h0 h

So, f 0 (x) = 1 everywhere.

Theorem 7.2. For any function f , D(f ) C(f ).
Proof. Suppose x0 D(f ). Then
 
 f (x) f (x0 ) 
lim |f (x) f (x0 )| = lim  (x x0 )
xx0 xx0 x x0
0
= f (x0 ) 0 = 0.
This shows limxx0 f (x) = f (x0 ), and x0 C(f ). 
Of course, the converse of Theorem 7.2 is not true.

1
May 1, 2015 Lee
c Larson ([email protected])

7-1
7-2 CHAPTER 7. DIFFERENTIATION

Figure 1. These graphs illustrate that the two standard ways of writing
the difference quotient are equivalent.

Example 7.3. The function f (x) = |x| is continuous on R, but

f (0 + h) f (0) f (0 + h) f (0)
lim = 1 = lim ,
h0 h h0 h
0
so f (0) fails to exist.
Theorem 7.2 and Example 7.3 show that differentiability is a strictly stronger
condition than continuity. For a long time most mathematicians believed that
every continuous function must certainly be differentiable at some point. In the
nineteenth century, several researchers, most notably Bolzano and Weierstrass,
presented examples of functions continuous everywhere and differentiable nowhere.2
It has since been proved that, in a technical sense, the typical continuous function
is nowhere differentiable [4]. So, contrary to the impression left by many beginning
calculus classes, differentiability is the exception rather than the rule, even for
continuous functions..

2. Differentiation Rules
Following are the standard rules for differentiation learned in every beginning
calculus course.
Theorem 7.3. Suppose f and g are functions such that x0 D(f ) D(g).
(a) x0 D(f + g) and (f + g)0 (x0 ) = f 0 (x0 ) + g 0 (x0 ).
(b) If a R, then x0 D(af ) and (af )0 (x0 ) = af 0 (x0 ).
(c) x0 D(f g) and (f g)0 (x0 ) = f 0 (x0 )g(x0 ) + f (x0 )g 0 (x0 ).
(d) If g(x0 ) 6= 0, then x0 D(f /g) and
 0
f f 0 (x0 )g(x0 ) f (x0 )g 0 (x0 )
(x0 ) = .
g (g(x0 ))2

2Bolzano presented his example in 1834, but it was little noticed. The 1872 example of
Weierstrass is more well-known [2]. A translation of Weierstrass original paper [19] is presented
by Edgar [9]. Weierstrass example is not very transparent because it depends on trigonometric
series. Many more elementary constructions have since been made. One such will be presented in
Example 9.9.

May 1, 2015 http://math.louisville.edu/lee/ira

2. DIFFERENTIATION RULES 7-3

Proof. (a)
(f + g)(x0 + h) (f + g)(x0 )
lim
h0 h
f (x0 + h) + g(x0 + h) f (x0 ) g(x0 )
= lim
h0 h
f (x0 + h) f (x0 ) g(x0 + h) g(x0 )
 
= lim + = f 0 (x0 ) + g 0 (x0 )
h0 h h
(b)
(af )(x0 + h) (af )(x0 ) f (x0 + h) f (x0 )
lim = a lim = af 0 (x0 )
h0 h h0 h
(c)
(f g)(x0 + h) (f g)(x0 ) f (x0 + h)g(x0 + h) f (x0 )g(x0 )
lim = lim
h0 h h0 h
Now, slip a 0 into the numerator and factor the fraction.
f (x0 + h)g(x0 + h) f (x0 )g(x0 + h) + f (x0 )g(x0 + h) f (x0 )g(x0 )
= lim
h0 h
f (x0 + h) f (x0 ) g(x0 + h) g(x0 )
 
= lim g(x0 + h) + f (x0 )
h0 h h
Finally, use the definition of the derivative and the continuity of f and g at x0 .
= f 0 (x0 )g(x0 ) + f (x0 )g 0 (x0 )
(d) It will be proved that if g(x0 ) 6= 0, then (1/g)0 (x0 ) = g 0 (x0 )/(g(x0 ))2 . This
statement, combined with (c), yields (d).
1 1

(1/g)(x0 + h) (1/g)(x0 ) g(x0 + h) g(x0 )
lim = lim
h0 h h0 h
g(x0 ) g(x0 + h) 1
= lim
h0 h g(x0 + h)g(x0 )
g 0 (x0 )
=
(g(x0 )2
Plug this into (c) to see
 0  0
f 1
(x0 ) = f (x0 )
g g
1 g 0 (x0 )
= f 0 (x0 ) + f (x0 )
g(x0 ) (g(x0 ))2
f 0 (x0 )g(x0 ) f (x0 )g 0 (x0 )
= .
(g(x0 ))2

Combining Examples 7.1 and 7.2 with Theorem 7.3, the following theorem is
easy to prove.
Corollary 7.4. A rational function is differentiable at every point of its
domain.

May 1, 2015 http://math.louisville.edu/lee/ira

7-4 CHAPTER 7. DIFFERENTIATION

Theorem 7.5 (Chain Rule). If f and g are functions such that x0 D(f ) and
f (x0 ) D(g), then x0 D(g f ) and (g f )0 (x0 ) = g 0 f (x0 )f 0 (x0 ).
Proof. Let y0 = f (x0 ). By assumption, there is an open interval J containing
f (x0 ) such that g is defined on J. Since J is open and x0 C(f ), there is an open
interval I containing x0 such that f (I) J.
Define h : J R by
g(y) g(y0 ) g 0 (y ), y 6= y

0 0
h(y) = y y0 .
0, y = y0

Since y0 D(g), we see

g(y) g(y0 )
lim h(y) = lim g 0 (y0 ) = g 0 (y0 ) g 0 (y0 ) = 0 = h(y0 ),
yy0 yy0 y y0
so y0 C(h). Now, x0 C(f ) and f (x0 ) = y0 C(h), so Theorem 6.15 implies
x0 C(h f ). In particular
(53) lim h f (x) = 0.
xx0

From the definition of h f for x I with f (x) 6= f (x0 ), we can solve for
(54) g f (x) g f (x0 ) = (h f (x) + g 0 f (x0 ))(f (x) f (x0 )).
Notice that (54) is also true when f (x) = f (x0 ). Divide both sides of (54) by x x0 ,
and use (53) to obtain
g f (x) g f (x0 ) f (x) f (x0 )
lim = lim (h f (x) + g 0 f (x0 ))
xx0 x x0 xx0 x x0
= (0 + g 0 f (x0 ))f 0 (x0 )
= g 0 f (x0 )f 0 (x0 ).

Theorem 7.6. Suppose f : [a, b] [c, d] is continuous and invertible. If
x0 D(f ) and f 0 (x0 ) 6= 0 for some x0 (a, b), then f (x0 ) D(f 1 ) and

0
f 1 (f (x0 )) = 1/f 0 (x0 ).
Proof. Let y0 = f (x0 ) and suppose yn is any sequence in f ([a, b]) \ {y0 }
converging to y0 and xn = f 1 (yn ). By Theorem 6.24, f 1 is continuous, so
x0 = f 1 (y0 ) = lim f 1 (yn ) = lim xn .
n n
Therefore,
f 1 (yn ) f 1 (y0 ) xn x0 1
lim = lim = 0 .
n yn y0 n f (xn ) f (x0 ) f (x0 )

Example 7.4. It follows easily from Theorem 7.3 that f (x) = x3 is differentiable
everywhere with f (x) = 3x . Define g(x) = x. Then g(x) = f 1 (x). Suppose
0 2 3

g(y0 ) = x0 for some y0 R. According to Theorem 7.6,

1 1 1 1 1
g 0 (y0 ) = 0 = 2 = = = 2/3 .
f (x0 ) 3x0 3(g(y0 ))2 3( 3 y0 )2 3y 0

May 1, 2015 http://math.louisville.edu/lee/ira

3. DERIVATIVES AND EXTREME POINTS 7-5

In the same manner as Example 7.4, the following corollary can be proved.
Corollary 7.7. Suppose q Q, f (x) = xq and D is the domain of f . Then
0
f (x) = qxq1 on the set
(
D, when q 1
.
D \ {0}, when q < 1

3. Derivatives and Extreme Points

As is learned in calculus, the derivative is a powerful tool for determining the
behavior of functions. The following theorems form the basis for much of differential
calculus. First, we state a few familiar definitions.
Definition 7.8. Suppose f : D R and x0 D. f is said to have a relative
maximum at x0 if there is a > 0 such that f (x) f (x0 ) for all x (x0 , x0 +)D.
f has a relative minimum at x0 if f has a relative maximum at x0 . If f has either
a relative maximum or a relative minimum at x0 , then it is said that f has a relative
extreme value at x0 .
The absolute maximum of f occurs at x0 if f (x0 ) f (x) for all x D. The
definitions of absolute minimum and absolute extreme are analogous.
Examples like f (x) = x on (0, 1) show that even the nicest functions need
not have relative extrema. Corollary 6.23 shows that if D is compact, then any
continuous function defined on D assumes both an absolute maximum and an
absolute minimum on D.
Theorem 7.9. Suppose f : (a, b) R. If f has a relative extreme value at x0
and x0 D(f ), then f 0 (x0 ) = 0.
Proof. Suppose f (x0 ) is a relative maximum value of f . Then there must be
a > 0 such that f (x) f (x0 ) whenever x (x0 , x0 + ). Since f 0 (x0 ) exists,
f (x) f (x0 ) f (x) f (x0 )
(55) x (x0 , x0 ) = 0 = f 0 (x0 ) = lim 0
x x0 xx0 x x0
and
f (x) f (x0 ) f (x) f (x0 )
(56) x (x0 , x0 +) = 0 = f 0 (x0 ) = lim 0.
x x0 xx0 x x0
Combining (55) and (56) shows f 0 (x0 ) = 0.
If f (x0 ) is a relative minimum value of f , apply the previous argument to
f . 

Theorem 7.9 is, of course, the basis for much of a beginning calculus course. If
f : [a, b] R, then the extreme values of f occur at points of the set
C = {x (a, b) : f 0 (x) = 0} {x [a, b] : f 0 (x) does not exist}.
The elements of C are often called the critical points or critical numbers of f on
[a, b]. To find the maximum and minimum values of f on [a, b], it suffices to find its
maximum and minimum on the smaller set C, which is finite in elementary calculus
courses.

May 1, 2015 http://math.louisville.edu/lee/ira

7-6 CHAPTER 7. DIFFERENTIATION

4. Differentiable Functions
Differentiation becomes most useful when a function has a derivative at each
point of an interval.
Definition 7.10. The function f is differentiable on an open interval I if
I D(f ). If f is differentiable on its domain, then it is said to be differentiable. In
this case, the function f 0 is called the derivative of f .
The fundamental theorem about differentiable functions is the Mean Value
Theorem. Following is its simplest form.
Lemma 7.11 (Rolles Theorem). If f : [a, b] R is continuous on [a, b],
differentiable on (a, b) and f (a) = 0 = f (b), then there is a c (a, b) such that
f 0 (c) = 0.
Proof. Since [a, b] is compact, Corollary 6.23 implies the existence of xm , xM
[a, b] such that f (xm ) f (x) f (xM ) for all x [a, b]. If f (xm ) = f (xM ), then f is
constant on [a, b] and any c (a, b) satisfies the lemma. Otherwise, either f (xm ) < 0
or f (xM ) > 0. If f (xm ) < 0, then xm (a, b) and Theorem 7.9 implies f 0 (xm ) = 0.
If f (xM ) > 0, then xM (a, b) and Theorem 7.9 implies f 0 (xM ) = 0. 
Rolles Theorem is just a stepping-stone on the path to the Mean Value Theorem.
Two versions of the Mean Value Theorem follow. The first is a version more general
than the one given in most calculus courses. The second is the usual version.4
Theorem 7.12 (Cauchy Mean Value Theorem). If f : [a, b] R and g : [a, b]
R are both continuous on [a, b] and differentiable on (a, b), then there is a c (a, b)
such that
g 0 (c)(f (b) f (a)) = f 0 (c)(g(b) g(a)).
Proof. Let
h(x) = (g(b) g(a))(f (a) f (x)) + (g(x) g(a))(f (b) f (a)).
Because of the assumptions on f and g, h is continuous on [a, b] and differentiable
on (a, b) with h(a) = h(b) = 0. Theorem 7.11 yields a c (a, b) such that h0 (c) = 0.
Then
0 = h0 (c) = (g(b) g(a))f 0 (c) + g 0 (c)(f (b) f (a))
= g 0 (c)(f (b) f (a)) = f 0 (c)(g(b) g(a)).


Corollary 7.13 (Mean Value Theorem). If f : [a, b] R is continuous on

[a, b] and differentiable on (a, b), then there is a c (a, b) such that f (b) f (a) =
f 0 (c)(b a).
Proof. Let g(x) = x in Theorem 7.12. 
Many of the standard theorems of beginning calculus are easy consequences of
the Mean Value Theorem. For example, following are the usual theorems about
monotonicity.
3May 1, 2015 Lee
c Larson ([email protected])
4Theorem 7.12 is also often called the Generalized Mean Value Theorem.

May 1, 2015 http://math.louisville.edu/lee/ira

Section 4: Differentiable Functions 7-7

Figure 2. This is a picture proof of Corollary 7.13.

Theorem 7.14. Suppose f : (a, b) R is a differentiable function. f is

increasing on (a, b) iff f 0 (x) 0 for all x (a, b). f is decreasing on (a, b) iff
f 0 (x) 0 for all x (a, b).
Proof. Only the first assertion is proved because the proof of the second is
pretty much the same with all the inequalities reversed.
() If x, y (a, b) with x 6= y, then the assumption that f is increasing gives
f (y) f (x) f (y) f (x)
0 = f 0 (x) = lim 0.
yx yx yx
() Let x, y (a, b) with x < y. According to Theorem 7.13, there is a c (x, y)
such that f (y) f (x) = f 0 (c)(y x) 0. This shows f (x) f (y), so f is increasing
on (a, b). 

Corollary 7.15. Let f : (a, b) R be a differentiable function. f is constant

iff f 0 (x) = 0 for all x (a, b).
It follows from Theorem 7.2 that every differentiable function is continuous.
But, its not true that a derivative need be continuous.
Example 7.5. Let (
x2 sin x1 , x 6= 0
f (x) = .
0, x=0
We claim f is differentiable everywhere, but f 0 is not continuous.
To see this, first note that when x 6= 0, the standard differentiation formulas
give that f 0 (x) = 2x sin(1/x) cos(1/x). To calculate f 0 (0), choose any h =
6 0.
Then
 f (h)   h2 sin(1/h)   h2 
     
 =    = |h|
 h   h  h
and it easily follows from the definition of the derivative and the Squeeze Theorem
(Theorem 6.3) that f 0 (0) = 0.
Let xn = 1/2n for n N. Then xn 0 and
f 0 (xn ) = 2xn sin(1/xn ) cos(1/xn ) = 1
for all n. Therefore, f 0 (xn ) 1 6= 0 = f 0 (0), and f 0 is not continuous at 0.

May 1, 2015 http://math.louisville.edu/lee/ira

7-8 CHAPTER 7. DIFFERENTIATION

But, derivatives do share one useful property with continuous functions; they sat-
isfy an intermediate value property. Compare the following theorem with Corollary
6.26.
Theorem 7.16 (Darbouxs Theorem). If f is differentiable on an open set
containing [a, b] and is between f 0 (a) and f 0 (b), then there is a c [a, b] such that
f 0 (c) = .
Proof. If f 0 (a) = f 0 (b), then c = a satisfies the theorem. So, we may as well
assume f 0 (a) 6= f 0 (b). There is no generality lost in assuming f 0 (a) < f 0 (b), for,
otherwise, we just replace f with g = f .

Figure 3. This could be the function h of Theorem 7.16.

Let h(x) = f (x) x so that D(f ) = D(h) and h0 (x) = f 0 (x) . In particular,
this implies h0 (a) < 0 < h0 (b). Because of this, there must be an > 0 small enough
so that
h(a + ) h(a)
< 0 = h(a + ) < h(a)

and
h(b) h(b )
> 0 = h(b ) < h(b).

(See Figure 3.) In light of these two inequalities and Theorem 6.23, there must
be a c (a, b) such that h(c) = glb {h(x) : x [a, b]}. Now Theorem 7.9 gives
0 = h0 (c) = f 0 (c) , and the theorem follows. 
Heres an example showing a possible use of Theorem 7.16.
Example 7.6. Let (
0, x 6= 0
f (x) = .
1, x=0
Theorem 7.16 implies f is not a derivative.
A more striking example is the following
Example 7.7. Define
( (
sin x1 , x 6= 0 sin x1 , x 6= 0
f (x) = and g(x) = .
1, x=0 1, x=0
Since (
0, x 6= 0
f (x) g(x) =
2, x=0
does not have the intermediate value property, at least one of f or g is not a
derivative. (Actually, neither is a derivative because f (x) = g(x).)

May 1, 2015 http://math.louisville.edu/lee/ira

5. APPLICATIONS OF THE MEAN VALUE THEOREM 7-9

5. Applications of the Mean Value Theorem

In the following sections, the standard notion of higher order derivatives is used.
To make this precise, suppose f is defined on an interval I. The function f itself can
be written f (0) . If f is differentiable, then f 0 is written f (1) . Continuing inductively,
if n , f (n) exists on I and x0 D(f (n) ), then f (n+1) (x0 ) = df (n) (x0 )/dx.

5.1. Taylors Theorem. The motivation behind Taylors theorem is the at-
tempt to approximate a function f near a number a by a polynomial. The polynomial
of degree 0 which does the best job is clearly p0 (x) = f (a). The best polynomial of
degree 1 is the tangent line to the graph of the function p1 (x) = f (a) + f 0 (a)(x a).
Continuing in this way, we approximate f near a by the polynomial pn of degree n
(k)
such that f (k) (a) = pn (a) for k = 0, 1, . . . , n. A simple induction argument shows
that
n
X f (k) (a)
(57) pn (x) = (x a)k .
k!
k=0

This is the well-known Taylor polynomial of f at a.

Many students leave calculus with the mistaken impression that (57) is the
important part of Taylors theorem. But, the important part of Taylors theorem
is the fact that in many cases it is possible to determine how large n must be to
achieve a desired accuracy in the approximation of f ; i. e., the error term is the
important part.
Theorem 7.17 (Taylors Theorem). If f is a function such that f, f 0 , . . . , f (n)
are continuous on [a, b] and f (n+1) exists on (a, b), then there is a c (a, b) such
that
n
X f (k) (a) f (n+1) (c)
f (b) = (b a)k + (b a)n+1 .
k! (n + 1)!
k=0

Proof. Let the constant be defined by

n
X f (k) (a)
(58) f (b) = (b a)k + (b a)n+1
k! (n + 1)!
k=0

and define
n
!
X f (k) (x) k
F (x) = f (b) (b x) + (b x)n+1 .
k! (n + 1)!
k=0

From (58) we see that F (a) = 0. Direct substitution in the definition of F shows
that F (b) = 0. From the assumptions in the statement of the theorem, it is easy
to see that F is continuous on [a, b] and differentiable on (a, b). An application of
Rolles Theorem yields a c (a, b) such that
 (n+1) 
0 f (c) n n
0 = F (c) = (b c) (b c) = = f (n+1) (c),
n! n!
as desired. 

5May 1, 2015 Lee

c Larson ([email protected])

May 1, 2015 http://math.louisville.edu/lee/ira

7-10 CHAPTER 7. DIFFERENTIATION

4
n=4 n=8

n = 20

2 4 6 8

y = cos(x)
-2

-4 n=2 n=6 n = 10

Figure 4. Here are several of the Taylor polynomials for the function
cos(x) graphed along with cos(x).

Now, suppose f is defined on an open interval I with a, x I. If f is n + 1

times differentiable on I, then Theorem 7.17 implies there is a c between a and x
such that
f (x) = pn (x) + Rf (n, x, a),
f (n+1) (c)
where Rf (n, x, a) = (n+1)! (x a)n+1 is the error in the approximation.6

Example 7.8. Let f (x) = cos x. Suppose we want to approximate f (2) to 5

decimal places of accuracy. Since its an easy point to work with, well choose a = 0.
Then, for some c (0, 2),
|f (n+1) (c)| n+1 2n+1
(59) |Rf (n, 2, 0)| = 2 .
(n + 1)! (n + 1)!
A bit of experimentation with a calculator shows that n = 12 is the smallest n such
that the right-hand side of (59) is less than 5 106 . After doing some arithmetic,
it follows that
22 24 26 28 210 212 27809
p12 (2) = 1 + + + = 0.41614.
2! 4! 6! 8! 10! 12! 66825
is a 5 decimal place approximation to cos(2).
But, things dont always work out the way we might like. Consider the following
example.
Example 7.9. Suppose
( 2
e1/x , x 6= 0
f (x) = .
0, x=0

6There are several different formulas for the error. The one given here is sometimes called
the Lagrange form of the remainder. In Example 8.4 a form of the remainder using integration
instead of differentiation is derived.

May 1, 2015 http://math.louisville.edu/lee/ira

5. APPLICATIONS OF THE MEAN VALUE THEOREM 7-11

Figure 5 below has a graph of this function. In Example 7.11 below it is shown that
f is differentiable to all orders everywhere and f (n) (0) = 0 for all n 0. With this
function the Taylor polynomial centered at 0 gives a useless approximation.
5.2. LHospitals Rules and Indeterminate Forms. According to Theo-
rem 6.4,
f (x) limxa f (x)
lim =
xa g(x) limxa g(x)
whenever limxa f (x) and limxa g(x) both exist and limxa g(x) 6= 0. But, it
is easy to find examples where both limxa f (x) = 0 and limxa g(x) = 0 and
limxa f (x)/g(x) exists, as well as similar examples where limxa f (x)/g(x) fails
to exist. Because of this, such a limit problem is said to be in the indeterminate
form 0/0. The following theorem allows us to determine many such limits.
Theorem 7.18 (Easy LHospitals Rule). Suppose f and g are each continuous
on [a, b], differentiable on (a, b) and f (b) = g(b) = 0. If g 0 (x) 6= 0 on (a, b) and
limxb f 0 (x)/g 0 (x) = L, where L could be infinite, then limxb f (x)/g(x) = L.
Proof. Let x [a, b), so f and g are continuous on [x, b] and differentiable on
(x, b). Cauchys Mean Value Theorem, Theorem 7.12, implies there is a c(x) (x, b)
such that
f (x) f 0 (c(x))
f 0 (c(x))g(x) = g 0 (c(x))f (x) = = 0 .
g(x) g (c(x))
Since x < c(x) < b, it follows that limxb c(x) = b. This shows that
f 0 (x) f 0 (c(x)) f (x)
L = lim 0
= lim 0 = lim .
xb g (x) xb g (c(x)) xb g(x)


Several things should be noted about this proof. First, there is nothing special
about the left-hand limit used in the statement of the theorem. It could just as
easily be written in terms of the right-hand limit. Second, if limxa f (x)/g(x) is
not of the indeterminate form 0/0, then applying LHospitals rule will usually give
a wrong answer. To see this, consider
x 1
lim = 0 6= 1 = lim .
x0 x + 1 x0 1

Another case where the indeterminate form 0/0 occurs is in the limit at infinity.
That LHopitals rule works in this case can easily be deduced from Theorem 7.18.
Corollary 7.19. Suppose f and g are differentiable on (a, ) and
lim f (x) = lim g(x) = 0.
x x
If g 0 (x) 6= 0 on (a, ) and limx f 0 (x)/g 0 (x) = L, where L could be infinite, then
limx f (x)/g(x) = L.
Proof. There is no generality lost by assuming a > 0. Let
( (
f (1/x), x (0, 1/a] g(1/x), x (0, 1/a]
F (x) = and G(x) = .
0, x=0 0, x=0
Then
lim F (x) = lim f (x) = 0 = lim g(x) = lim G(x),
x0 x x x0

May 1, 2015 http://math.louisville.edu/lee/ira

7-12 CHAPTER 7. DIFFERENTIATION

so both F and G are continuous at 0. It follows that both F and G are continuous
on [0, 1/a] and differentiable on (0, 1/a) with G0 (x) = g 0 (x)/x2 = 6 0 on (0, 1/a)
and limx0 F 0 (x)/G0 (x) = limx f 0 (x)/g 0 (x) = L. The rest follows from Theorem
7.18. 
The other standard indeterminate form arises when
lim f (x) = = lim g(x).
x x

This is called an / indeterminate form. It is often handled by the following

theorem.
Theorem 7.20 (Hard LHospitals Rule). Suppose that f and g are differentiable
on (a, ) and g 0 (x) 6= 0 on (a, ). If
f 0 (x)
lim f (x) = lim g(x) = and lim = L R {, },
x x x g 0 (x)

then
f (x)
lim = L.
x g(x)
Proof. First, suppose L R and let > 0. Choose a1 > a large enough so
that  0 
 f (x) 

 g 0 (x) L < , x > a1 .

Since limx f (x) = = limx g(x), we can assume there is an a2 > a1 such
that both f (x) > 0 and g(x) > 0 when x > a2 . Finally, choose a3 > a2 such that
whenever x > a3 , then f (x) > f (a2 ) and g(x) > g(a2 ).
Let x > a3 and apply Cauchys Mean Value Theorem, Theorem 7.12, to f and
g on [a2 , x] to find a c(x) (a2 , x) such that
 
f 0 (c(x)) f (x) f (a2 ) f (x) 1 ff(a 2)
(x)
(60) = = .
g 0 (c(x))

g(x) g(a2 ) g(x) 1 g(a2 )
g(x)

If
g(a2 )
1 g(x)
h(x) = f (a2 )
,
1 f (x)
then (60) implies
f (x) f 0 (c(x))
= 0 h(x).
g(x) g (c(x))
Since limx h(x) = 1, there is an a4 > a3 such that whenever x > a4 , then
|h(x) 1| < . If x > a4 , then
   0 
 f (x)   f (c(x)) 
 g(x) L =  g 0 (c(x)) h(x) L
   
 0 
 f (c(x)) 
= 0
 h(x) Lh(x) + Lh(x) L
g (c(x))
 0 
 f (c(x)) 
 0
 L |h(x)| + |L||h(x) 1|
g (c(x))
< (1 + ) + |L| = (1 + |L| + )

May 1, 2015 http://math.louisville.edu/lee/ira

5. APPLICATIONS OF THE MEAN VALUE THEOREM 7-13



3 2 1 1 2 3

Figure 5. This is a plot of f (x) = exp(1/x2 ). Notice how the graph

flattens out near the origin.

can be made arbitrarily small through a proper choice of . Therefore

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

The case when L = is done similarly by first choosing a B > 0 and adjusting
(60) so that f 0 (x)/g 0 (x) > B when x > a1 . A similar adjustment is necessary when
L = . 
There is a companion corollary to Theorem 7.20 which is proved in the same
way as Corollary 7.19.
Corollary 7.21. Suppose that f and g are continuous on [a, b] and differen-
tiable on (a, b) with g 0 (x) 6= 0 on (a, b). If
f 0 (x)
lim f (x) = lim g(x) = and lim = L R {, },
xa xa xa g 0 (x)
then
f (x)
lim = L.
xa g(x)
Example 7.10. If > 0, then limx ln x/x is of the indeterminate form
/. Taking derivatives of the numerator and denominator yields
1/x 1
lim1
= lim = 0.
x x x x

Theorem 7.20 now implies limx ln x/x = 0, and therefore ln x increases more
slowly than any positive power of x.

Example 7.11. Let f be as in Example 7.9. (See Figure 5.) It is clear f (n) (x)
exists whenever n and x 6= 0. We claim f (n) (0) = 0. To see this, we first prove
that
2
e1/x
(61) lim = 0, n Z.
x0 xn

May 1, 2015 http://math.louisville.edu/lee/ira

7-14 CHAPTER 7. DIFFERENTIATION

When n 0, (61) is obvious. So, suppose (61) is true whenever m n for some
n . Making the substitution u = 1/x, we see
2
e1/x un+1
(62) lim n+1 = lim u2 .
x0 x u e

Since
2
(n + 1)un (n + 1)un1 n+1 e1/x
lim 2 = lim 2 = lim =0
u 2ueu u 2eu 2 x0 xn1
by the inductive hypothesis, Theorem 7.20 gives (62) in the case of the right-hand
limit. The left-hand limit is handled similarly. Finally, (61) follows by induction.
When x 6= 0, a bit of experimentation can convince the reader that f (n) (x)
2
is of the form pn (1/x)e1/x , where pn is a polynomial. Induction and repeated
applications of (61) establish that f (n) (0) = 0 for n .

6. Exercises

7.1. If (
x2 , x Q
f (x) = ,
0, otherwise
then show D(f ) = {0} and find f 0 (0).

7.2. Let f be a function defined on some neighborhood of x = a with f (a) = 0.

Prove f 0 (a) = 0 if and only if a D(|f |).

7.3. If f is defined on an open set containing x0 , the symmetric derivative of f at

x0 is defined as
f (x0 + h) f (x0 h)
f s (x0 ) = lim .
h0 2h
Prove that if f 0 (x) exists, then so does f s (x). Is the converse true?

7.4. Let G be an open set and f D(G). If there is an a G such that

limxa f 0 (x) exists, then limxa f 0 (x) = f 0 (a).

7.5. Suppose f is continuous on [a, b] and f 00 exists on (a, b). If there is an

x0 (a, b) such that the line segment between (a, f (a)) and (b, f (b)) contains the
point (x0 , f (x0 )), then there is a c (a, b) such that f 00 (c) = 0.

7.6. If = {f : f = F 0 for some F : R R}, then is closed under addition and

scalar multiplication. (This shows the differentiable functions form a vector space.)

7.7. If (
1/2, x=0
f1 (x) =
6 0
sin(1/x), x =
and (
1/2, x=0
f2 (x) = ,
sin(1/x), x 6= 0
then at least one of f1 and f2 is not in .

May 1, 2015 http://math.louisville.edu/lee/ira

6. EXERCISES 7-15

7.8. Prove or give a counter example: If f is continuous on R and differentiable

on R \ {0} with limx0 f 0 (x) = L, then f is differentiable on R.

7.9. Suppose f is differentiable everywhere and f (x+y) = f (x)f (y) for all x, y R.
Show that f 0 (x) = f 0 (0)f (x) and determine the value of f 0 (0).

7.10. If I is an open interval, f is differentiable on I and a I, then there is a

sequence an I \ {a} such that an a and f 0 (an ) f 0 (a).

d
7.11. Use the definition of the derivative to find x.
dx
7.12. Let f be continuous on [0, ) and differentiable on (0, ). If f (0) = 0 and
|f 0 (x)| < |f (x)| for all x > 0, then f (x) = 0 for all x 0.

7.13. Suppose f : R R is such that f 0 is continuous on [a, b]. If there is a

c (a, b) such that f 0 (c) = 0 and f 00 (c) > 0, then f has a local minimum at c.

7.14. Prove or give a counter example: If f is continuous on R and differentiable

on R \ {0} with limx0 f 0 (x) = L, then f is differentiable on R.

7.15. Let f be continuous on [a, b] and differentiable on (a, b). If f (a) = and
|f 0 (x)| < for all x (a, b), then calculate a bound for f (b).

7.16. Suppose that f : (a, b) R is differentiable and f 0 is bounded. If xn is a

sequence from (a, b) such that xn a, then f (xn ) converges.

7.17. Let G be an open set and f D(G). If there is an a G such that

limxa f 0 (x) exists, then limxa f 0 (x) = f 0 (a).

7.18. Prove or give a counter example: If f D((a, b)) such that f 0 is bounded,
then there is an F C([a, b]) such that f = F on (a, b).

7.19. Show that f (x) = x3 + 2x + 1 is invertible on R and, if g = f 1 , then find

g 0 (1).

7.20. Suppose that I is an open interval and that f 00 (x) 0 for all x I. If a I,
then show that the part of the graph of f on I is never below the tangent line to
the graph at (a, f (a)).

7.21. Suppose f is continuous on [a, b] and f 00 exists on (a, b). If there is an

x0 (a, b) such that the line segment between (a, f (a)) and (b, f (b)) contains the
point (x0 , f (x0 )), then there is a c (a, b) such that f 00 (c) = 0.

7.22. Let f be defined on a neighborhood of x.

(a) If f 00 (x) exists, then
f (x h) 2f (x) + f (x + h)
lim = f 00 (x).
h0 h2
(b) Find a function f where this limit exists, but f 00 (x) does not exist.

May 1, 2015 http://math.louisville.edu/lee/ira

7-16 CHAPTER 7. DIFFERENTIATION

7.23. If f : R R is differentiable everywhere and is even, then f 0 is odd. If

f : R R is differentiable everywhere and is odd, then f 0 is even.7

7.24. Prove that

3 5
  
sin x x x + x < 1
 
 6 120  5040
when |x| 1.

7.25. The exponential function ex is not a polynomial.

7A function g is even if g(x) = g(x) for every x and it is odd if g(x) = g(x) for every
x. The terms are even and odd because this is how g(x) = xn behaves when n is an even or odd
integer, respectively.

May 1, 2015 http://math.louisville.edu/lee/ira

 CHAPTER 8

Integration

Contrary to the impression given by most calculus courses, there are many ways
to define integration. The one given here is called the Riemann integral or the
Riemann-Darboux integral, and it is the one most commonly presented to calculus
students.

1. Partitions
A partition of the interval [a, b] is a finite set P [a, b] such that {a, b} P .
The set of all partitions of [a, b] is denoted part ([a, b]). Basically, a partition should
be thought of as a way to divide an interval into a finite number of subintervals by
choosing some points where it is divided.
If P part ([a, b]), then the elements of P can be ordered in a list as a = x0 <
x1 < < xn = b. The adjacent points of this partition determine n compact
intervals of the form IkP = [xk1 , xk ], 1 k n. If the partition is clear from the
context, we write Ik instead of IkP . Its clear that these intervals only intersect at
their common endpoints and there is no requirement they have the same length.
Since its inconvenient to always list each part of a partition, well use the
partition of the previous paragraph as the generic partition. Unless its necessary
within the context to specify some other form for a partition, assume any partition
is the generic partition. (See Figure 1.)
If I is any interval, its length is written |I|. Using the notation of the previous
paragraph, it follows that
n
X n
X
|Ik | = (xk xk1 ) = xn x0 = b a.
k=1 k=1

The norm of a partition P is

kP k = max{|IkP | : 1 k n}.
In other words, the norm of P is just the length of the longest subinterval determined
by P . If |Ik | = kP k for every Ik , then P is called a regular partition.
Suppose P, Q part ([a, b]). If P Q, then Q is called a refinement of P . When
this happens, we write P  Q. In this case, its easy to see that P  Q implies

I1 I2 I3 I4 I5
x0 x1 x2 x3 x4 x5

a b

Figure 1. The generic partition with five subintervals.

8-1
8-2 CHAPTER 8. INTEGRATION

kP k kQk. It also follows at once from the definitions that P Q part ([a, b])
with P  P Q and Q  P Q. The partition P Q is called the common
refinement of P and Q.

2. Riemann Sums
Let f : [a, b] R and P part ([a, b]). Choose xk Ik for each k. The set
{xk : 1 k n} is called a selection from P . The expression
n
X
R (f, P, xk ) = f (xk )|Ik |
k=1
is the Riemann sum for f with respect to the partition P and selection xk . The
Riemann sum is the usual first step toward integration in a calculus course and can
be visualized as the sum of the areas of rectangles with height f (xk ) and width |Ik |
as long as the rectangles are allowed to have negative area when f (xk ) < 0. (See
Figure 2.)
Notice that given a particular function f and partition P , there are an uncount-
ably infinite number of different possible Riemann sums, depending on the selection
xk . This sometimes makes working with Riemann sums quite complicated.
Example 8.1. Suppose f : [a, b] R is the constant function f (x) = c. If
P part ([a, b]) and {xk : 1 k n} is any selection from P , then
n
X n
X
R (f, P, xk ) = f (xk )|Ik | = c |Ik | = c(b a).
k=1 k=1

Example 8.2. Suppose f (x) = x on [a, b]. Choose any P part ([a, b]) where
kP k < 2(b a)/n. (Convince yourself this is always possible.1) Make two specific

1This is with the generic partition

y = f (x)

a x1 x1 x2 x2 x3 x3 x4 b
x0 x4

Figure 2. The Riemann sum R (f, P, xk ) is the sum of the areas of the
rectangles in this figure. Notice the right-most rectangle has negative
area because f (x4 ) < 0.

May 1, 2015 http://math.louisville.edu/lee/ira

Riemann Sums 8-3

selections lk = xk1 and rk = xk . If xk is any other selection from P , then

lk xk rk and the fact that f is increasing on [a, b] gives
R (f, P, lk ) R (f, P, xk ) R (f, P, rk ) .
With this in mind, consider the following calculation.
n
X
(68) R (f, P, rk ) R (f, P, lk ) = (rk lk )|Ik |
k=1
Xn
= (xk xk1 )|Ik |
k=1
Xn
= |Ik |2
k=1
Xn
kP k2
k=1
= nkP k2
4(b a)2
<
n
This shows that if a partition is chosen with a small enough norm, all the Riemann
sums for f over that partition will be close to each other.
In the special case when P is a regular partition, |Ik | = (b a)/n, rk =
a + k(b a)/n and
n
X
R (f, P, rk ) = rk |Ik |
k=1
n 
k(b a)
ba
X 
= a+
n n
k=1
n
!
ba ba X
= na + k
n n
k=1
ba b a n(n + 1)
 
= na +
n n 2
ba n1
 
n+1
= a +b .
2 n n
In the limit as n , this becomes the familiar formula (b2 a2 )/2, for the integral
of f (x) = x over [a, b].

Definition 8.1. The function f is Riemann integrable on [a, b], if there exists a
number R (f ) such that for all > 0 there is a > 0 so that whenever P part ([a, b])
with kP k < , then
|R (f ) R (f, P, xk ) | <
for any selection xk from P .
Theorem 8.2. If f : [a, b] R and R (f ) exists, then R (f ) is unique.

May 1, 2015 http://math.louisville.edu/lee/ira

8-4 CHAPTER 8. INTEGRATION

Proof. Suppose R1 (f ) and R2 (f ) both satisfy the definition and > 0. For
i = 1, 2 choose i > 0 so that whenever kP k < i , then
|Ri (f ) R (f, P, xk ) | < /2,
as in the definition above. If P part ([a, b]) so that kP k < 1 2 , then
|R1 (f ) R2 (f )| |R1 (f ) R (f, P, xk ) | + |R2 (f ) R (f, P, xk ) | <
and it follows R1 (f ) = R2 (f ). 

Theorem 8.3. If f : [a, b] R and R (f ) exists, then f is bounded.

Proof. Left as Exercise 8.1. 

3. Darboux Integration
As mentioned above, a difficulty with handling Riemann sums is there are an
uncountably so many different ways to choose partitions and selections that working
with them is unwieldy. One way to resolve this problem was shown in Example 8.2,
where it was easy to find largest and smallest Riemann sums associated with each
partition. However, thats not always a straightforward calculation, so to use that
idea, a little more care must be taken.
Definition 8.4. Let f : [a, b] R be bounded and P part ([a, b]). For each
Ik determined by P , let
Mk = lub {f (x) : x Ik } and mk = glb {f (x) : x Ik }.
The upper and lower Darboux sums for f on [a, b] are
n
X n
X
D (f, P ) = Mk |Ik | and D (f, P ) = mk |Ik |.
k=1 k=1

The following theorem is the fundamental relationship between Darboux sums.

Pay careful attention because its the linchpin holding everything together!
Theorem 8.5. If f : [a, b] R is bounded and P, Q part ([a, b]) with P  Q,
then
D (f, P ) D (f, Q) D (f, Q) D (f, P ) .
Proof. Let P be the generic partition and let Q = P {x}, where x
(xk0 1 , xx0 ) for some k0 . Clearly, P  Q. Let
Ml = lub {f (x) : x [xk0 1 , x]}
ml = glb {f (x) : x [xk0 1 , x]}
Mr = lub {f (x) : x [x, xk0 ]}
mr = glb {f (x) : x [x, xk0 ]}

Then
mk0 ml Ml Mk0 and mk0 mr Mr Mk0

May 1, 2015 http://math.louisville.edu/lee/ira

Darboux Integration 8-5

so that
mk0 |Ik0 | = mk0 (|[xk0 1 , x]| + |[x, xk0 ]|)
ml |[xk0 1 , x]| + mr |[x, xk0 ]|
Ml |[xk0 1 , x]| + Mr |[x, xk0 ]|
Mk0 |[xk0 1 , x]| + Mk0 |[x, xk0 ]|
= Mk0 |Ik0 |.
This implies
n
X
D (f, P ) = mk |Ik |
k=1
X
= mk |Ik | + mk0 |Ik0 |
k6=k0
X
mk |Ik | + ml |[xk0 1 , x]| + mr |[x, xk0 ]|
k6=k0
= D (f, Q)
D (f, Q)
X
= Mk |Ik | + Ml |[xk0 1 , x]| + Mr |[x, xk0 ]|
k6=k0
X n
Mk |Ik |
k=1
= D (f, P )
The argument given above shows that the theorem holds if Q has one more point
than P . Using induction, this same technique also shows the theorem holds when Q
has an arbitrarily larger number of points than P . 
The main lesson to be learned from Theorem 8.5 is that refining a partition
causes the lower Darboux sum to increase and the upper Darboux sum to decrease.
Moreover, if P, Q part ([a, b]) and f : [a, b] [B, B], then,
B(b a) D (f, P ) D (f, P Q) D (f, P Q) D (f, Q) B(b a).
Therefore every Darboux lower sum is less than or equal to every Darboux upper
sum. Consider the following definition with this in mind.
Definition 8.6. The upper and lower Darboux integrals of a bounded function
f : [a, b] R are
D (f ) = glb {D (f, P ) : P part ([a, b])}
and
D (f ) = lub {D (f, P ) : P part ([a, b])},
respectively.
As a consequence of the observations preceding the definition, it follows that
D (f ) D (f ) always. In the case D (f ) = D (f ), the function is said to be Darboux
integrable on [a, b], and the common value is written D (f ).
The following is obvious.

May 1, 2015 http://math.louisville.edu/lee/ira

8-6 CHAPTER 8. INTEGRATION

Corollary 8.7. A bounded function f : [a, b] R is Darboux integrable if and

only if for all > 0 there is a P part ([a, b]) such that D (f, P ) D (f, P ) < .
Which functions are Darboux integrable? The following corollary gives a first
approximation to an answer.
Corollary 8.8. If f C([a, b]), then D (f ) exists.
Proof. Let > 0. According to Corollary 6.31, f is uniformly continuous, so
there is a > 0 such that whenever x, y [a, b] with |x y| < , then |f (x) f (y)| <
/(b a). Let P part ([a, b]) with kP k < . By Corollary 6.23, in each subinterval
Ii determined by P , there are xi , yi Ii such that
f (xi ) = lub {f (x) : x Ii } and f (yi ) = glb {f (x) : x Ii }.
Since |xi yi | |Ii | < , we see 0 f (xi ) f (yi ) < /(b a), for 1 i n. Then
D (f ) D (f ) D (f, P ) D (f, P )
n
X n
X
= f (xi )|Ii | f (yi )|Ii |
i=1 i=1
n
X
= (f (xi ) f (yi ))|Ii |
i=1
n
X
< |Ii |
b a i=1
=
and the corollary follows. 
This corollary should not be construed to imply that only continuous functions
are Darboux integrable. In fact, the set of integrable functions is much more
extensive than only the continuous functions. Consider the following example.
Example 8.3. Let f be the salt and pepper function of Example 6.15. It was
shown that C(f ) = Qc . We claim that f is Darboux integrable over any compact
interval [a, b].
To see this, let > 0 and N N so that 1/N < /2(b a). Let
{qki : 1 i m} = {qk : 1 k N } [a, b]
and choose P part ([a, b]) such that kP k < /2m. Then
n
X
D (f, P ) = lub {f (x) : x I` }|I` |
`=1
X X
= lub {f (x) : x I` }|I` | + lub {f (x) : x I` }|I` |
qki I
/ ` qki I`

1
(b a) + mkP k
N

< (b a) + m
2(b a) 2m
= .
Since f (x) = 0 whenever x Qc , it follows that D (f, P ) = 0. Therefore, D (f ) =
D (f ) = 0 and D (f ) = 0.

May 1, 2015 http://math.louisville.edu/lee/ira

The Integral 8-7

4. The Integral
There are now two different definitions for the integral. It would be embarassing,
if they gave different answers. The following theorem shows theyre really different
sides of the same coin.2
Theorem 8.9. Let f : [a, b] R.
(a) R (f ) exists iff D (f ) exists.
(b) If R (f ) exists, then R (f ) = D (f ).
Proof. (a) (=) Suppose R (f ) exists and > 0. By Theorem 8.3, f is
bounded. Choose P part ([a, b]) such that
|R (f ) R (f, P, xk ) | < /4
for all selections xk from P . From each Ik , choose xk and xk so that

Mk f (xk ) < and f (xk ) mk < .
4(b a) 4(b a)
Then
n
X n
X
D (f, P ) R (f, P, xk ) = Mk |Ik | f (xk )|Ik |
k=1 k=1
Xn
= (Mk f (xk ))|Ik |
k=1

< (b a) = .
4(b a) 4
In the same way,
R (f, P, xk ) D (f, P ) < /4.
Therefore,
D (f ) D (f )
= glb {D (f, Q) : Q part ([a, b])} lub {D (f, Q) : Q part ([a, b])}
D (f, P ) D (f, P )
   
< R (f, P, xk ) + R (f, P, xk )
4 4

|R (f, P, xk ) R (f, P, xk )| +
2

< |R (f, P, xk ) R (f ) | + |R (f ) R (f, P, xk ) | +
2
<
Since is an arbitrary positive number, this shows D (f ) exists and equals R (f ),
which is part (b) of the theorem.

2Theorem 8.9 shows that the two integrals presented here are the same. But, there are many
other integrals, and not all of them are equivalent. For example, the well-known Lebesgue integral
includes all Riemann integrable functions, but not all Lebesgue integrable functions are Riemann
integrable. The Denjoy integral is another extension of the Riemann integral which is not the same
as the Lebesgue integral. For more discussion of this, see [10].

May 1, 2015 http://math.louisville.edu/lee/ira

8-8 CHAPTER 8. INTEGRATION

(=) Suppose f : [a, b] [B, B], D (f ) exists and > 0. Since D (f ) exists,
there is a P1 part ([a, b]), with points a = p0 < < pm = b, such that

D (f, P1 ) D (f, P1 ) < .
2
Set = /8mB. Choose P part ([a, b]) with kP k < and let P2 = P P1 . Since
P1  P2 , according to Theorem 8.5,

D (f, P2 ) D (f, P2 ) < .
2
Thinking of P as the generic partition, the interiors of its intervals (xi1 , xi )
may or may not contain points of P1 . For 1 i n, let
Qi = {xi1 , xi } (P1 (xi1 , xi )) part (Ii ) .
If P1 (xi1 , xi ) = , then D (f, P ) and D (f, P2 ) have the term Mi |Ii | in
common because Qi = {xi1 , xi }.
Otherwise, P1 (xi1 , xi ) 6= and
D (f, Qi ) BkP2 k BkP k > B.
Since P1 has m 1 points in (a, b), there are at most m 1 of the Qi not contained
in P .
This leads to the estimate
n
X
D (f, P ) D (f, P2 ) = D (f, P ) D (f, Qi ) < (m 1)2B < .
i=1
4
In the same way,

D (f, P2 ) D (f, P ) < (m 1)2B < .
4
Putting these estimates together yields
D (f, P ) D (f, P ) =



D (f, P ) D (f, P2 ) + D (f, P2 ) D (f, P2 ) + (D (f, P2 ) D (f, P ))

< + + =
4 2 4
This shows that, given > 0, there is a > 0 so that kP k < implies
D (f, P ) D (f, P ) < .
Since
D (f, P ) D (f ) D (f, P ) and D (f, P ) R (f, P, xi ) D (f, P )
for every selection xi from P , it follows that |R (f, P, xi )D (f ) | < when kP k < .
We conclude f is Riemann integrable and R (f ) = D (f ). 
From Theorem 8.9, we are justified in using a single notation for both R (f )
Rb Rb
and D (f ). The obvious choice is the familiar a f (x) dx, or, more simply, a f .
When proving statements about the integral, its convenient to switch back and
forth between the Riemann and Darboux formulations. Given f : [a, b] R the
following three facts summarize much of what we know.
Rb
(1) a f exists iff for all > 0 there is a > 0 and an R such that whenever
P part ([a, b]) and xi is a selection from P , then |R (f, P, xi ) | < .
Rb
In this case a f = .

May 1, 2015 http://math.louisville.edu/lee/ira

The Cauchy Criterion 8-9
Rb

(2) a f exists iff > 0P part ([a, b]) D (f, P ) D (f, P ) <
(3) For any P part ([a, b]) and selection xi from P ,

D (f, P ) R (f, P, xi ) D (f, P ) .

5. The Cauchy Criterion

Rb
We now face a conundrum. In order to show that a f exists, we must know
its value. Its often very hard to determine the value of an integral, even if the
integral exists. Weve faced this same situation before with sequences. The basic
definition of convergence for a sequence, Definition 3.2, requires the limit of the
sequence be known. The path out of the dilemma in the case of sequences was the
Cauchy criterion for convergence, Theorem 3.22. The solution is the same here,
with a Cauchy criterion for the existence of the integral.
Theorem 8.10 (Cauchy Criterion). Let f : [a, b] R. The following statements
are equivalent.
Rb
(a) a f exists.
(b) Given > 0 there exists P part ([a, b]) such that if P  Q1 and
P  Q2 , then
(69) |R (f, Q1 , xk ) R (f, Q2 , yk )| <
for any selections from Q1 and Q2 .
Rb
Proof. (=) Assume a f exists. According to Definition 8.1, there is a > 0
Rb
such that whenever P part ([a, b]) with kP k < , then | a f R (f, P, xi ) | < /2
for every selection. If P  Q1 and P  Q2 , then kQ1 k < , kQ2 k < and a simple
application of the triangle inequality shows

|R (f, Q1 , xk ) R (f, Q2 , yk )|
 Z b  Z b 
 
R (f, Q1 , xk ) f +  f R (f, Q2 , yk ) < .
   
 a   a 

(=) Let > 0 and choose P part ([a, b]) satisfying (b) with /2 in place of
.
We first claim that f is bounded. To see this, suppose it is not. Then it must be
unbounded on an interval Ik0 determined by P . Fix a selection {xk Ik : 1 k n}
and let yk = xk for k 6= k0 with yk0 any element of Ik0 . Then

> |R (f, P, xk ) R (f, P, yk )| = f (xk0 ) f (yk0 ) |Ik0 | .
 
2
But, the right-hand side can be made bigger than /2 with an appropriate choice
of yk0 because of the assumption that f is unbounded on Ik0 . This contradiction
forces the conclusion that f is bounded.
Thinking of P as the generic partition and using mk and Mk as usual with
Darboux sums, for each k, choose xk , yk Ik such that

Mk f (xk ) < and f (yk ) mk < .
4n|Ik | 4n|Ik |

May 1, 2015 http://math.louisville.edu/lee/ira

8-10 CHAPTER 8. INTEGRATION

With these selections,

D (f, P ) D (f, P )
= D (f, P ) R (f, P, xk ) + R (f, P, xk ) R (f, P, yk ) + R (f, P, yk ) D (f, P )
n
X n
X
= (Mk f (xk ))|Ik | + R (f, P, xk ) R (f, P, yk ) + (f (yk ) mk )|Ik |
k=1
k=1 
n
X X n 
|Mk f (xk )| |Ik | + |R (f, P, xk ) R (f, P, yk )| +  (f (yk ) mk )|Ik |
 
 
k=1 k=1
n n
X X
< |Ik | + |R (f, P, xk ) R (f, P, yk )| + |Ik |
4n|Ik | 4n|Ik |
k=1 k=1

< + + <
4 2 4
Corollary 8.7 implies D (f ) exists and Theorem 8.9 finishes the proof. 

Rb Rd
Corollary 8.11. If a
f exists and [c, d] [a, b], then c
f exists.

Proof. Let P0 = {a, b, c, d} part ([a, b]) and > 0. Using Theorem 8.10,
choose a partition P such that P0  P and whenever P  P and P  P 0 , then

|R (f, P, xk ) R (f, P 0 , yk ) | < .

Let P1 = P [a, c], P2 = P [c, d] and P3 = P [d, b]. Suppose P2  Q1 and

P2  Q2 . Then P1 Qi P3 for i = 1, 2 are refinements of P and

|R (f, Q1 , xk ) R (f, Q2 , yk ) | =
|R f, P1 Q1 P3 , xk R f, P1 Q2 P3 , yk | <

Rb
for any selections. An application of Theorem 8.10 shows a
f exists. 

6. Properties of the Integral

Rb Rb
Theorem 8.12. If a f and a g both exist, then
Rb Rb Rb Rb
(a) If , R, then a (f + g) exists and a (f + g) = a f + a g.
Rb
(b) a f g exists.
Rb
(c) a |f | exists.

Proof. (a) Let > 0. If = 0, in light of Example 8.1, it is clear f is

integrable. So, assume 6= 0, and choose a partition Pf part ([a, b]) such that
whenever Pf  P , then
 Z b 


R (f, P, xk ) f < .
 
 a  2||

May 1, 2015 http://math.louisville.edu/lee/ira

Properties of the Integral 8-11

Then
 Z b  X n Z b 

R (f, P, xk ) f =  f (xk )|Ik | f
   
 a  
k=1 a 
 n Z b  
X
= ||  f (xk )|Ik | f
 
 a 
k=1 
 Z b 
= || R (f, P, xk ) f
 
 a 

< ||
2||

= .
2
Rb Rb
This shows f is integrable and a f = a f .
Assuming 6= 0, in the same way, we can choose a Pg part ([a, b]) such that
when Pg  P , then
 Z b 


R (g, P, xk ) g < .
 
 a  2||

Let P = Pf Pg be the common refinement of Pf and Pg , and suppose P  P .

Then
!
Z b Z b
|R (f + g, P, xk ) f + g |
a a
 Z b   Z b 
 

|| R (f, P, xk ) f  + || R (g, P, xk ) g <
   
 a   a 

Rb Rb Rb
for any selection. This shows f +g is integrable and a (f +g) = a f + a g.
Rb Rb 2
(b) Claim: If a h exists, then so does a h
To see this, suppose first that 0 h(x) M on [a, b]. If M = 0, the claim is
trivially true, so suppose M > 0. Let > 0 and choose P part ([a, b]) such that

D (h, P ) D (h, P ) .
2M

For each 1 k n, let

mk = glb {h(x) : x Ik } lub {h(x) : x Ik } = Mk .

Since h 0,

m2k = glb {h(x)2 : x Ik } lub {h(x)2 : x Ik } = Mk2 .

May 1, 2015 http://math.louisville.edu/lee/ira

8-12 CHAPTER 8. INTEGRATION

Using this, we see

n

X
D h2 , P D h2 , P = (Mk2 m2k )|Ik |

k=1
Xn
= (Mk + mk )(Mk mk )|Ik |
k=1
n
!
X
2M (Mk mk )|Ik |
k=1


= 2M D (h, P ) D (h, P )
< .
2
Therefore, h is integrable when h 0.
If h is not nonnegative, let m = glb {h(x) : a x b}. Then h m 0, and
h m is integrable by (a). From the claim, (h m)2 is integrable. Since
h2 = (h m)2 + 2mh m2 ,
it follows from (a) that h2 is integrable.
Finally, f g = 14 ((f + g)2 (f g)2 ) is integrable by the claim and (a).

(c) Claim: If h 0 is integrable, then so is h.
To see this, let > 0 and choose P part ([a, b]) such that
D (h, P ) D (h, P ) < 2 .
For each 1 k n, let
p p
mk = glb { h(x) : x Ik } lub { h(x) : x Ik } = Mk .
and define
A = {k : Mk mk < } and B = {k : Mk mk }.
Then
X
(70) (Mk mk )|Ik | < (b a).
kA

Using the fact that mk 0, we see that Mk mk Mk + mk , and

X 1X
(71) (Mk mk )|Ik | (Mk + mk )(Mk mk )|Ik |

kB kB
1X
= (Mk2 m2k )|Ik |

kB
1

D (h, P ) D (h, P )

<
Combining (70) and (71), it follows that
   
D h, P D h, P < (b a) + = ((b a) + 1)

can be made arbitrarily
p small. Therefore, h is integrable.
Since |f | = f 2 an application of (b) and the claim suffice to prove (c). 

May 1, 2015 http://math.louisville.edu/lee/ira

The Fundamental Theorem of Calculus 8-13
Rb
Theorem 8.13. If a
f exists, then
Rb
(a) If f 0 on [a, b], then a f 0.
Rb Rb
(b) | a f | a |f |
Rb Rc Rb
(c) If a c b, then a f = a f + c f .
Proof. (a) Since all the Riemann sums are nonnegative, this follows at once.
Rb
(b) It is always true that |f | f 0 and |f | f 0, so by (a), a (|f | + f ) 0
Rb Rb Rb Rb Rb
and a (|f | f ) 0. Rearranging these shows a f a |f | and a f a |f |.
Rb Rb
Therefore, | a f | a |f |, which is (b).
(c) By Corollary 8.11, all the integrals exist. Let > 0 and choose Pl
part ([a, c]) and Pr part ([c, b]) such that whenever Pl  Ql and Pr  Qr , then,
 Z c   Z b 

 

R (f, Ql , xk ) f  < and R (f, Qr , yk )

f < .

a 2  c  2
If P = Pl Pr and Q = Ql Qr , then P, Q part ([a, b]) and P  Q. The triangle
inequality gives 
 Z c Z b 

R (f, Q, xk ) f f  < .
 
 a c 
Since every refinement of P has the form Ql Qr , part (c) follows. 
Rb
Theres some notational trickery that can be played here. If a f exists, then
Ra Rb
we define b f = a f . With this convention, it can be shown
Z b Z c Z b
(72) f= f+ f
a a c
no matter the order of a, b and c, as long as at least two of the integrals exist. (See
Problem 8.4.)

7. The Fundamental Theorem of Calculus

Theorem 8.14 (Fundamental Theorem of Calculus 1). Suppose f, F : [a, b] R
satisfy
Rb
(a) a f exists
(b) F C([a, b]) D((a, b))
(c) F 0 (x) = f (x), x (a, b)
Rb
Then a f = F (b) F (a).

Proof. Let > 0. According to (a) and Definition 8.1, P part ([a, b]) can
be chosen such that 
 Z b 

R (f, P, xk ) f  < .
 
 a 
for every selection from P . On each interval [xk1 , xk ] determined by P , the
function F satisfies the conditions of the Mean Value Theorem. (See Corollary 7.13.)
Therefore, for each k, there is an ck (xk1 , xk ) such that
F (xk ) F (xk1 ) = F 0 (ck )(xk xk1 ) = f (ck )|Ik |.

May 1, 2015 http://math.louisville.edu/lee/ira

8-14 CHAPTER 8. INTEGRATION

So,
Z  Z n

 b   b X 
f (F (b) F (a)) =  f (F (xk ) F (xk1 )
   

 a   a 
k=1
Z 
 b Xn 
= f f (ck )|Ik |
 
 a 
k=1
Z 
 b 
= f R (f, P, ck )
 
 a 
<
and the theorem follows. 

Example 8.4. The Fundamental Theorem of Calculus can be used to give a

different form of Taylors theorem. As in Theorem 7.17, suppose f and its first n + 1
Rb
derivatives exist on [a, b] and a f (n+1) exists. There is a function Rf (n, x, t) such
that
n
X f (k) (t)
Rf (n, x, t) = f (x) (x t)k
k!
k=0
for a t b. Differentiating both sides of the equation with respect to t, note that
the right-hand side telescopes, so the result is
d (x t)n (n+1)
Rf (n, x, t) = f (t).
dt n!
Using Theorem 8.14 and the fact that Rf (n, x, x) = 0 gives
Rf (n, x, c) = Rf (n, x, c) Rf (n, x, x)
Z c
d
= Rf (n, x, t) dt
x dt
Z x
(x t)n (n+1)
= f (t) dt,
c n!
which is the integral form of the remainder from Taylors formula.
Corollary 8.15 (Integration by Parts). If f, g C([a, b]) D((a, b)) and both
f 0 g and f g 0 are integrable on [a, b], then
Z b Z b
0
fg + f 0 g = f (b)g(b) f (a)g(a).
a a

Proof. Use Theorems 7.3(c) and 8.14. 

Rb
Suppose a f exists. By Corollary 8.11, f is integrable on every interval [a, x],
Rx
for x [a, b]. This allows us to define a function F : [a, b] R as F (x) = a f ,
called the indefinite integral of f on [a, b].
Theorem 8.16 (Fundamental Theorem of Calculus 2). Let f be integrable on
[a, b] and F be the indefinite integral of f . Then F C([a, b]) and F 0 (x) = f (x)
whenever x C(f ) (a, b).

May 1, 2015 http://math.louisville.edu/lee/ira

The Fundamental Theorem of Calculus 8-15

x x+h

Figure
Z 3. This figure illustrates a box argument showing
x+h
1
lim f = f (x).
h0 h x

Rb
Proof. To show F C([a, b]), let x0 [a, b] and > 0. Since a f exists,
there is an M > lub {|f (x)| : a x b}. Choose 0 < < /M and x
(x0 , x0 + ) [a, b]. Then
Z x 
 
|F (x) F (x0 )| =  f  M |x x0 | < M <
x0

and x0 C(F ).
Let x0 C(f )(a, b) and > 0. There is a > 0 such that x (x0 , x0 +)
(a, b) implies |f (x) f (x0 )| < . If 0 < h < , then
   Z x0 +h 
 F (x0 + h) F (x0 )  1 
f (x ) = f f (x )

0 0
 
h  h
  
x0 
 Z 
 1 x0 +h 
= (f (t) f (x0 )) dt
 
 h x0 
Z x0 +h
1
|f (t) f (x0 )| dt
h x0
1 x0 +h
Z
< dt
h x0
= .
This shows F+0 (x0 ) = f (x0 ). It can be shown in the same way that F0 (x0 ) = f (x0 ).
Therefore F 0 (x0 ) = f (x0 ). 
The right picture makes Theorem 8.16 almost obvious. Consider Figure 3.
Suppose x C(f ) and > 0. There is a > 0 such that
f ((x d, x + d) [a, b]) (f (x) /2, f (x) + /2).
Let
m = glb {f y : |x y| < } lub {f y : |x y| < } = M.

May 1, 2015 http://math.louisville.edu/lee/ira

8-16 CHAPTER 8. INTEGRATION

Apparently M m < and for 0 < h < ,

Z x+h
F (x + h) F (x)
mh f M h = m M.
x h
Since M m 0 as h 0, a squeezing argument shows
F (x + h) F (x)
lim = f (x).
h0 h
A similar argument establishes the limit from the left and F 0 (x) = f (x).
Example 8.5. The usual definition of the natural logarithm function depends
on the Fundamental Theorem of Calculus. Recall for x > 0,
Z x
1
ln(x) = dt.
1 t

Since f (t) = 1/t is continuous on (0, ), Theorem 8.16 shows

d 1
ln(x) = .
dx x
It should also be noted that the notational convention mentioned above equation
(72) is used to get ln(x) < 0 when 0 < x < 1.
Its easy to read too much into the Fundamental Theorem of Calculus. We
are tempted to start thinking of integration and differentiation as opposites of
each other. But, this is far from the truth. The operations of integration and
antidifferentiation are different operations, that happen to sometimes be tied together
by the Fundamental Theorem of Calculus. Consider the following examples.
Example 8.6. Let (
|x|/x, x 6= 0
f (x) =
0, x=0
Its
R x easy to prove that f is integrable over any compact interval, and that F (x) =
1
f = |x| 1 is an indefinite integral of f . But, F is not differentiable at x = 0
and f is not a derivative, according to Theorem 7.16.
Example 8.7. Let (
x2 sin x12 , x 6= 0
f (x) =
0, x=0
Its straightforward to show that f is differentiable and
(
0 2x sin x12 x2 cos x12 , x 6= 0
f (x) =
0, x=0

Since f 0 is unbounded near x = 0, it follows from Theorem 8.3 that f 0 is not

integrable over any interval containing 0.
Example 8.8. Let f be the salt and pepper function of Example 6.15. It was
Rb Rx
shown in Example 8.3 that a f = 0 on any interval [a, b]. If F (x) = 0 f , then
F (x) = 0 for all x and F 0 = f only on C(f ) = Qc .

May 1, 2015 http://math.louisville.edu/lee/ira

8. CHANGE OF VARIABLES 8-17

8. Change of Variables
Integration by substitution works side-by-side with the Fundamental Theorem of
Calculus in the integration section of any calculus course. Most of the time calculus
books require all functions in sight to be continuous. In that case, a substitution
theorem is an easy consequence of the Fundamental Theorem and the Chain Rule.
(See Exercise 8.13.) More general statements are true, but they are harder to prove.
Theorem 8.17. If f and g are functions such that
(a) g is strictly monotone on [a, b],
(b) g is continuous on [a, b],
(c) g is differentiable on (a, b), and
R g(b) Rb
(d) both g(a) f and a (f g)g 0 exist,
then
Z g(b) Z b
(73) f= (f g)g 0 .
g(a) a

Proof. Suppositions (a) and (b) show g is a bijection from [a, b] to an interval
[c, d]. The correspondence between the endpoints depends on whether g is increasing
or decreasing.
Let > 0.
From (d) and Definition 8.1, there is a 1 > 0 such that whenever P part ([a, b])
with kP k < 1 , then
 Z b 
 
(74) R ((f g)g 0 , P, xi ) (f g)g 0  <
 
 a  2
for any selection from P . Choose P1 part ([a, b]) such that kP1 k < 1 .
Using the same argument, there is a 2 > 0 such that whenever Q part ([c, d])
with kQk < 2 , then
 Z d 



(75) R (f, Q, xi ) f <

 c  2
for any selection from Q. As above, choose Q1 part ([c, d]) such that kQ1 k < 2 .
Setting P2 = P1 {g 1 (x) : x Q1 } and Q2 = P1 {g(x) : x P1 }, it is
apparent that P1  P2 , Q1  Q2 , kP2 k kP1 k < 1 , kQ2 k kQ1 k < 2 and
Q2 = {g(x) : x P2 }. From (74) and (75), it follows that
Z 
 b 
 (f g)g 0 R ((f g)g 0 , P2 , xi ) <
 
 a  2
(76) and
Z 
 d 

f R (f, Q2 , yi ) <

 2

 c
for any selections from P2 and Q2 .
Label the points of P2 as a = x1 < x2 < < xn = b and those of Q2 as
c = y0 < y1 < < yn = d. From (b), (c) and the Mean Value Theorem, for each i,
choose ci (xi1 , xi ) such that
(77) g(xi ) g(xi1 ) = g 0 (ci )(xi xi1 ).

May 1, 2015 http://math.louisville.edu/lee/ira

8-18 CHAPTER 8. INTEGRATION

Notice that {ci : 1 i n} is a selection from P2 .

First, assume g is strictly increasing. In this case g(xi ) = yi for 0 i n and
g(ci ) (yi1 , yi ) for 0 < i n, so g(ci ) is a selection from Q2 .
Z 
 g(b) Z b 
0
f (f g)g 


 g(a) a 
Z 
 g(b) Z b 
0
= f R (f, Q2 , g(ci )) + R (f, Q2 , g(ci )) (f g)g 

 g(a) a 
Z   
 g(b)   Z b 
0
 f R (f, Q2 , g(ci )) + R (f, Q2 , g(ci )) (f g)g 
  
 g(a)   a 
Use the triangle inequality and (76). Expand the second Riemann sum.
 
n Z b
X 
0
< + f (g(ci )) (g(xi ) g(xi1 )) (f g)g 
2  i=1 a 

Apply the Mean Value Theorem, as in (77), and then use (76).
 
n Z b
X 
= + f (g(ci ))g 0 (ci ) (xi xi1 ) (f g)g 0 

2  i=1 a 
 Z b 
 
= + R ((f g)g 0 , P2 , ci ) (f g)g 0 

2  a 

< +
2 2
=
and (73) follows.
Now assume g is strictly decreasing on [a, b]. The proof is much the same as above,
except the bookkeeping is trickier because order is reversed instead of preserved
by g. This means g(xk ) = ynk when 0 k n and g(cnk+1 ) (yk1 , yk ) for
0 < k n. Therefore, g(cnk+1 ) is a selection from Q2 . From the Mean Value
Theorem,
yk yk1 = g(xnk ) g(xnk+1 )
(78) = (g(xnk+1 ) g(xnk ))
= g 0 (cnk+1 )(xnk+1 xnk ),
where ck (xk1 , xk ) is as above. The rest of the proof is much like the case when
g is increasing.
Z 
 g(b) Z b 
0
f (f g)g 


 g(a) a 
 Z 
 g(a) Z b 
=  f + R (f, Q2 , g(cnk+1 )) R (f, Q2 , g(cnk+1 )) (f g)g 0 
 
 g(b) a 
 Z   
 g(b)   Z b 
0
 f + R (f, Q2 , g(cnk+1 )) + R (f, Q2 , g(cnk+1 )) (f g)g 
  
 g(a)   a 

May 1, 2015 http://math.louisville.edu/lee/ira

8. CHANGE OF VARIABLES 8-19

Use (76), expand the second Riemann sum and apply (78).
 n Z b 
 X 
< +  f (g(cnk+1 ))(yk yk1 ) (f g)g 0 

2  a 
k=1
 n Z b 
X 0

0 
= + f (g(cnk+1 ))g (cnk+1 )(xnk+1 xnk ) (f g)|g |
2  a 
k=1

Reverse the order of the sum and use (76).

 n Z b 
X 0

0
= + f (g(ck ))g (ck )(xk xk1 ) (f g)g 
2  a 
k=1
 
b

Z 
= + R ((f g)g 0 , P2 , ck ) (f g)g 0 

2  a 

< +
2 2
=
The theorem has been proved. 
R1
Example 8.9. Suppose we want to calculate 1 1 x2 dx. Using the notation

of Theorem 8.17, let f (x) = 1 x2 , g(x) = sin x and [a, b] = [/2, /2]. In this
case, g is an increasing function. Then (73) becomes
Z 1p Z sin(/2) p
1 x2 dx = 1 x2 dx
1 sin(/2)
Z /2 p
= 1 sin2 x cos x dx
/2
Z /2
= cos2 d dx
/2

= .
2
On the other hand, it can also be done with a decreasing function. If g(x) = cos x
and [a, b] = [0, ], then
Z 1p Z cos 0 p
1 x2 dx = 1 x2 dx
1 cos
Z cos p
= 1 x2 dx
cos 0
Z p
= 1 cos2 x( sin x) dx
0
Z p
= 1 cos2 x sin x dx
0
Z
= sin2 x dx
0

=
2

May 1, 2015 http://math.louisville.edu/lee/ira

8-20 CHAPTER 8. INTEGRATION

9. Integral Mean Value Theorems

Theorem 8.18. Suppose f, g : [a, b] R are such that
(a) g(x) 0 on [a, b],
(b) f is bounded and m f (x) M for all x [a, b], and
Rb Rb
(c) a f and a f g both exist.
There is a c [m, M ] such that
Z b Z b
fg = c g.
a a

Proof. Obviously,
Z b Z b Z b
(79) m g fg M g.
a a a
Rb
If a
g = 0, were done. Otherwise, let
Rb
fg
c = Ra b .
a
g
Rb Rb
Then a
fg = c a
g and from (79), it follows that m c M . 

Corollary 8.19. Let f and g be as in Theorem 8.18, but additionally assume

f is continuous. Then there is a c (a, b) such that
Z b Z b
f g = f (c) g.
a a

Proof. This follows from Theorem 8.18 and Corollaries 6.23 and 6.26. 

Theorem 8.20. Suppose f, g : [a, b] R are such that

(a) g(x) 0 on [a, b],
(b) f is bounded and m f (x) M for all x [a, b], and
Rb Rb
(c) a f and a f g both exist.
There is a c [a, b] such that
Z b Z c Z b
fg = m g+M g.
a a c

Proof. For a x b let

Z x Z b
G(x) = m g+M g.
a x

By Theorem 8.16, G C([a, b]) and

Z b Z b Z b
glb G G(b) = m g fg M g = G(a) lub G.
a a a
Rb
Now, apply Corollary 6.26 to find c where G(c) = a
f g. 

May 1, 2015 http://math.louisville.edu/lee/ira

10. EXERCISES 8-21

10. Exercises

8.1. If f : [a, b] R and R (f ) exists, then f is bounded.

8.2. Let (
1, xQ
f (x) = .
0, x
/Q
(a) Use Definition 8.1 to show f is not integrable on any interval.
(b) Use Definition 8.6 to show f is not integrable on any interval.
R5
8.3. Calculate 2
x2 using the definition of integration.

8.4. If at least two of the integrals exist, then

Z b Z c Z b
f= f+ f
a a c

no matter the order of a, b and c.

Rb Rb
8.5. If > 0, f : [a, b] [, ] and a
f exists, then a
1/f exists.

8.6. If f : [a, b] [0, ) is continuous and D(f ) = 0, then f (x) = 0 for all
x [a, b].
Rb Rb Rb
8.7. If a
f exists, then limxa x
f= a
f.
Rb
8.8. If f is monotone on [a, b], then a
f exists.

8.9. If f and g are integrable on [a, b], then

Z  " Z ! Z !#1/2
 b  b b
2 2
f g f g .
 

 a  a a
Rb
(Hint: Expand a
(xf + g)2 as a quadratic with variable x.)3

8.10. If f : [a, b] [0, ) is continuous, then there is a c [a, b] such that

Z b !1/2
1
f (c) = f2 .
ba a

x
dt
Z
8.11. If f (x) = for x > 0, then f (xy) = f (x) + f (y) for x, y > 0.
1 t

8.12. If f (x) = ln(|x|) for x 6= 0, then f 0 (x) = 1/x.

3This is variously called the Cauchy inequality, Cauchy-Schwarz inequality, or the Cauchy-
Schwarz-Bunyakowsky inequality. Rearranging the last one, some people now call it the CBS
inequality.

May 1, 2015 http://math.louisville.edu/lee/ira

8-22 CHAPTER 8. INTEGRATION

8.13. In the statement of Theorem 8.17, make the additional assumptions that f
and g 0 are both continuous. Use the Fundamental Theorem of Calculus to give an
easier proof.

8.14. Find a function f : [a, b] R such that

(a) f is continuous on [c, b] for all c (a, b],
Rb
(b) limxa x f = 0, and
(c) limxa f (x) does not exist.

8.15. Find a bounded function solving Exercise 8.14.

May 1, 2015 http://math.louisville.edu/lee/ira

 CHAPTER 9

Sequences of Functions

1. Pointwise Convergence
We have accumulated much experience working with sequences of numbers. The
next level of complexity is sequences of functions. This chapter explores several ways
that sequences of functions can converge to another function. The basic starting
point is contained in the following definitions.
Definition 9.1. Suppose S R and for each n N there is a function
fn : S R. The collection {fn : n N} is a sequence of functions defined on S.
For each fixed x S, fn (x) is a sequence of numbers, and it makes sense to ask
whether this sequence converges. If fn (x) converges for each x S, a new function
f : S R is defined by
f (x) = lim fn (x).
n
The function f is called the pointwise limit of the sequence fn , or, equivalently, it is
S
said fn converges pointwise to f . This is abbreviated fn f , or simply fn f , if
the domain is clear from the context.
Example 9.1. Let
0,
x<0
fn (x) = xn , 0 x < 1 .

1, x1

Then fn f where (
0, x<1
f (x) = .
1, x1
(See Figure 1.) This example shows that a pointwise limit of continuous functions
need not be continuous.

Example 9.2. For each n N, define fn : R R by

nx
fn (x) = .
1 + n2 x2
(See Figure 2.) Clearly, each fn is an odd function and lim|x| fn (x) = 0. A bit
of calculus shows that fn (1/n) = 1/2 and fn (1/n) = 1/2 are the extreme values
of fn . Finally, if x 6= 0,
   
 nx   nx   1 
|fn (x)| = 
  <  2 2  =  
1 + n2 x2  n x nx
implies fn 0. This example shows that functions can remain bounded away from
0 and still converge pointwise to 0.

9-1
9-2 CHAPTER 9. SEQUENCES OF FUNCTIONS

1.0

0.8

0.6

0.4

0.2

0.2 0.4 0.6 0.8 1.0

Figure 1. The first ten functions from the sequence of Example 9.1.

0.4

0.2

-3 -2 -1 1 2 3

-0.2

-0.4

Figure 2. The first four functions from the sequence of Example 9.2.

Example 9.3. Define fn : R R by

2n+4 1 3
2
x 2n+3 , 2n+1 < x < 2n+2
3 1
fn (x) = 22n+4 x + 2n+4 , 2n+2 x< 2n

0, otherwise

To figure out what this looks like, it might help to look at Figure 3.
The graph of fn is a piecewise linear function supported on [1/2n+1 , 1/2n ] and
the
R 1 area under the isosceles triangle of the graph over this interval is 1. Therefore,
f = 1 for all n.
0 n
If x > 0, then whenever x > 1/2n , we have fn (x) = 0. From this it follows that
fn 0.
The Rlesson to
R 1be learned from this example is that it may not be true that
1
limn 0 fn = 0 limn fn .

Example 9.4. Define fn : R R by

(
n 2 1 1
x + 2n , |x|
fn (x) = 2 n
1
.
|x|, |x| > n

May 1, 2015 http://math.louisville.edu/lee/ira

Pointwise and Uniform Convergence 9-3

64

32

16
8

1 1 1 1
1
16 8 4 2

Figure 3. The first four functions fn 0 from the sequence of Example 9.3.

1/2

1/4
1/8

1 1
Figure 4. The first ten functions of the sequence fn |x| from
Example 9.4.

(See Figure 4.) The parabolic section in the center was chosen so fn (1/n) = 1/n
and fn0 (1/n) = 1. This splices the sections together at (1/n, 1/n) so fn is
differentiable everywhere. Its clear fn |x|, which is not differentiable at 0.
This example shows that the limit of differentiable functions need not be
differentiable.

The examples given above show that continuity, integrability and differentiability
are not preserved in the pointwise limit of a sequence of functions. To have any
hope of preserving these properties, a stronger form of convergence is needed.

May 1, 2015 http://math.louisville.edu/lee/ira

9-4 CHAPTER 9. SEQUENCES OF FUNCTIONS

2. Uniform Convergence
Definition 9.2. The sequence fn : S R converges uniformly to f : S R
on S, if for each > 0 there is an N N so that whenever n N and x S, then
|fn (x) f (x)| < .
S
In this case, we write fn f , or simply fn f , if the set S is clear from the
context.

f(x) +

f(x)

fn(x)
f(x

a b

Figure 5. |fn (x) f (x)| < on [a, b], as in Definition 9.2.

The difference between pointwise and uniform convergence is that with pointwise
convergence, the convergence of fn to f can vary in speed at each point of S. With
uniform convergence, the speed of convergence is roughly the same all across S.
Uniform convergence is a stronger condition to place on the sequence fn than
pointwise convergence in the sense of the following theorem.
S S
Theorem 9.3. If fn f , then fn f .
Proof. Let x0 S and > 0. There is an N N such that when n N , then
|f (x) fn (x)| < for all x S. In particular, |f (x0 ) fn (x0 )| < when n N .
This shows fn (x0 ) f (x0 ). Since x0 S is arbitrary, it follows that fn f . 
The first three examples given above show the converse to Theorem 9.3 is false.
There is, however, one interesting and useful case in which a partial converse is true.
S
Definition 9.4. If fn f and fn (x) f (x) for all x S, then fn increases to
S
f on S. If fn f and fn (x) f (x) for all x S, then fn decreases to f on S. In
either case, fn is said to converge to f monotonically.
The functions of Example 9.4 decrease to |x|. Notice that in this case, the
convergence is also happens to be uniform. The following theorem shows Example
9.4 to be an instance of a more general phenomenon.
Theorem 9.5 (Dinis Theorem). If
(a) S is compact,
S
(b) fn f monotonically,
(c) fn C(S) for all n N, and
(d) f C(S),
then fn f .

May 1, 2015 http://math.louisville.edu/lee/ira

3. METRIC PROPERTIES OF UNIFORM CONVERGENCE 9-5

Proof. There is no loss of generality in assuming fn f , for otherwise we

consider fn and f . With this assumption, if gn = fn f , then gn is a sequence
of continuous functions decreasing to 0. It suffices to show gn 0.
To do so, let > 0. Using continuity and pointwise convergence, for each x S
find an open set Gx containing x and an Nx N such that gNx (y) < for all y Gx .
Notice that the monotonicity condition guarantees gn (y) < for every y Gx and
n Nx .
The collection {Gx : x S} is an open cover for S, so it must contain a finite
subcover {Gxi : 1 i n}. Let N = max{Nxi : 1 i n} and choose m N . If
x S, then x Gxi for some i, and 0 gm (x) gN (x) gNi (x) < . It follows
that gn 0. 

fn (x) = xn

1/2

2 1/n 1

Figure 6. This shows a typical function from the sequence of Example 9.5.

Example 9.5. Let fn (x) = xn for n N, then fn decreases to 0 on [0, 1). If

0 < a < 1 Dinis Theorem shows fn 0 on the compact interval [0, a]. On the
whole interval [0, 1), fn (x) > 1/2 when 21/n < x < 1, so fn is not uniformly
convergent. (Why doesnt this violate Dinis Theorem?)

3. Metric Properties of Uniform Convergence

If S R, let B(S) = {f : S R : f is bounded}. For f B(S), define
kf kS = lub {|f (x)| : x S}. (It is abbreviated to kf k, if the domain S is clear from
the context.) Apparently, kf k 0, kf k = 0 f 0 and, if g B(S), then
kf gk = kg f k. Moreover, if h B(S), then
kf gk = lub {|f (x) g(x)| : x S}
lub {|f (x) h(x)| + |h(x) g(x)| : x S}
lub {|f (x) h(x)| : x S} + lub {|h(x) g(x)| : x S}
= kf hk + kh gk

May 1, 2015 http://math.louisville.edu/lee/ira

9-6 CHAPTER 9. SEQUENCES OF FUNCTIONS

Combining all this, it follows that kf gk is a metric1 on B(S).

The definition of uniform convergence implies that for a sequence of bounded
functions fn : S R,
fn f kfn f k 0.
Because of this, the metric kf gk is often called the uniform metric or the sup-
metric. Many ideas developed using the metric properties of R can be carried over
into this setting. In particular, there is a Cauchy criterion for uniform convergence.
Definition 9.6. Let S R. A sequence of functions fn : S R is a Cauchy
sequence under the uniform metric, if given > 0, there is an N N such that when
m, n N , then kfn fm k < .
Theorem 9.7. Let fn B(S). There is a function f B(S) such that fn f
iff fn is a Cauchy sequence in B(S).
Proof. () Let fn f and > 0. There is an N N such that n N
implies kfn f k < /2. If m N and n N , then

kfm fn k kfm f k + kf fn k < + =
2 2
shows fn is a Cauchy sequence.
() Suppose fn is a Cauchy sequence in B(S) and > 0. Choose N N so
that when kfm fn k < whenever m N and n N . In particular, for a fixed
x0 S and m, n N , |fm (x0 ) fn (x0 )| kfm fn k < shows the sequence
fn (x0 ) is a Cauchy sequence in R and therefore converges. Since x0 is an arbitrary
point of S, this defines an f : S R such that fn f .
Finally, if m, n N and x S the fact that |fn (x) fm (x)| < gives
|fn (x) f (x)| = lim |fn (x) fm (x)| .
m

This shows that when n N , then kfn f k . We conclude that f B(S) and
fn f . 
A collection of functions S is said to be complete under uniform convergence,
if every Cauchy sequence in S converges to a function in S. Theorem 9.7 shows
B(S) is complete under uniform convergence. Well see several other collections of
functions that are complete under uniform convergence.
Example 9.6. For S R let L(S) be all the functions f : S R such that
f (x) = mx + b for some constants m and b. In particular, let fn be a Cauchy
sequence in L([0, 1]). Theorem 9.7 shows there is an f : [0, 1] R such that fn f .
In order to show L([0, 1]) is complete, it suffices to show f L([0, 1]).
To do this, let fn (x) = mn x + bn for each n. Then fn (0) = bn f (0) and
mn = fn (1) bn f (1) f (0).
Given any x [0, 1],
fn (x) ((f (1) f (0))x + f (0)) = mn x + bn ((f (1) f (0))x + f (0))
= (mn (f (1) f (0)))x + bn f (0) 0.
This shows f (x) = (f (1) f (0))x + f (0) L([0, 1]) and therefore L([0, 1]) is
complete.
1Definition 2.10

May 1, 2015 http://math.louisville.edu/lee/ira

4. SERIES OF FUNCTIONS 9-7

Pn 9.7. Let P = {p(x) : p is a polynomial}. The sequence of polynomials

Example
pn (x) = k=0 xk /k! increases to ex on [0, a] for any a > 0, so Dinis Theorem shows
pn ex on [0, a]. But, ex
/ P, so P is not complete. (See Exercise 7.25.)

4. Series of Functions
The definitions of pointwiseP
and uniform convergence are extended in the natural

way to series of functions. If k=1 fk is a series of functions defined on a set S,
then the series converges
Pnpointwise or uniformly, depending on whether the sequence
of partial sums, sn = k=1 fk converges pointwise or uniformly,
P respectively. It is
absolutely convergent or absolutely uniformly convergent, if n=1 |fn | is convergent
or uniformly convergent on S, respectively.
The following theorem is obvious and its proof is left to the reader.
P P
Theorem 9.8. Let n=1 fn be a series of functions
P defined on S. If n=1 fn
is absolutely convergent, then it is convergent. If n=1 fn is absolutely uniformly
convergent, then it is uniformly convergent.

The following theorem is a restatement of Theorem 9.5 for series.

P
Theorem 9.9. If n=1 fn is a series of nonnegative continuous functions
P
converging pointwise to a continuous function on a compact set S, then n=1 fn
converges uniformly on S.

A simple, but powerful technique for showing uniform convergence of series is

the following.

Theorem 9.10 (Weierstrass M-Test). If fn : S R is a sequence of functions

and P
P numbers such that kfn kS Mn for all n N
Mn is a sequence nonnegative
and n=1 Mn converges, then n=1 fn is absolutely uniformly convergent.
P
Proof. Let > 0 and sn be the sequence of partial sums of n=1 |fn |. Using
Pnfor convergence of a series, choose an N N such that when
the Cauchy criterion
n > m N , then k=m Mk < . So,
n
X n
X n
X
ksn sm k = k fk k kfk k Mk < .
k=m+1 k=m+1 k=m

This shows sn is a Cauchy sequence and must converge according to Theorem 9.7. 

Example 9.8. Let a > 0 and Mn = an /n!. Since

Mn+1 a
lim = lim = 0,
n Mn n n + 1
P
the Ratio Test shows n=0 Mn converges. When x [a, a],
 n
 x  an
  .
 n!  n!
P
The Weierstrass M-Test now implies n=0 xn /n! converges absolutely uniformly on
[a, a] for any a > 0. (See Exercise 9.4.)

May 1, 2015 http://math.louisville.edu/lee/ira

9-8 CHAPTER 9. SEQUENCES OF FUNCTIONS

5. Continuity and Uniform Convergence

Theorem 9.11. If fn : S R is such that each fn is continuous at x0 and
S
fn f , then f is continuous at x0 .

Proof. Let > 0. Since fn f , there is an N N such that whenever n N

and x S, then |fn (x) f (x)| < /3. Because fN is continuous at x0 , there is a
> 0 such that x (x0 , x0 + ) S implies |fN (x) fN (x0 )| < /3. Using these
two estimates, it follows that when x (x0 , x0 + ) S,

|f (x) f (x0 )| = |f (x) fN (x) + fN (x) fN (x0 ) + fN (x0 ) f (x0 )|

|f (x) fN (x)| + |fN (x) fN (x0 )| + |fN (x0 ) f (x0 )|
< /3 + /3 + /3 = .

Therefore, f is continuous at x0 . 

The following corollary is immediate from Theorem 9.11.

Corollary 9.12. If fn is a sequence of continuous functions converging uni-

formly to f on S, then f is continuous.

Example 9.1 shows that continuity is not preserved under pointwise convergence.
Corollary 9.12 establishes that if S is compact, then C(S) is complete under the
uniform metric.
The fact that C([a, b]) is closed under uniform convergence is often useful
because, given a bad function f C([a, b]), its often possible to find a sequence
fn of good functions in C([a, b]) converging uniformly to f . Following is the most
widely used theorem of this type.

Theorem 9.13 (Weierstrass Approximation Theorem). If f C([a, b]), then

there is a sequence of polynomials pn f .

To prove this theorem, we first need a lemma.

R 1
1
Lemma 9.14. For n N let cn = 1 (1 t2 )n dt and
(
cn (1 t2 )n , |t| 1
kn (t) = .
0, |t| > 1

(See Figure 7.) Then

(a) kn (t) 0 for all t R and n N;
R1
(b) 1 kn = 1 for all n N; and,
(c) if 0 < < 1, then kn 0 on (, ] [, ).

Proof. Parts (a) and (b) follow easily from the definition of kn .
To prove (c) first note that
Z 1 Z 1/n  n
2 n 2 1
1= kn
cn (1 t ) dt cn 1 .
1 1/ n n n

May 1, 2015 http://math.louisville.edu/lee/ira

5. CONTINUITY AND UNIFORM CONVERGENCE 9-9

1.2

1.0

0.8

0.6

0.4

0.2

-1.0 -0.5 0.5 1.0

Figure 7. Here are the graphs of kn (t) for n = 1, 2, 3, 4, 5.

n 1
Since 1 n1 e , it follows there is an > 0 such that cn < n.2 Letting
(0, 1) and t 1,

kn (t) kn () n(1 2 )n 0
by LHospitals Rule. Since kn is an even function, this establishes (c). 
A sequence of functions satisfying conditions such as those in Lemma 9.14 is
called a convolution kernel or a Dirac sequence.3 Several such kernels play a key
role in the study of Fourier series, as we will see in Theorems 10.5 and 10.13. The
one defined above is called the Landau kernel.4
We now turn to the proof of the theorem.
Proof. There is no generality lost in assuming [a, b] = [0, 1], for otherwise we
consider the linear change of variables g(x) = f ((ba)x+a). Similarly, we can assume
f (0) = f (1) = 0, for otherwise we consider g(x) = f (x) ((f (1) f (0))x f (0),
which is a polynomial added to f . We can further assume f (x) = 0 when x / [0, 1].
Set
Z 1
(81) pn (x) = f (x + t)kn (t) dt.
1
To see pn is a polynomial, change variables in the integral using u = x + t to arrive
at Z x+1 Z 1
pn (x) = f (u)kn (u x) du = f (u)kn (x u) du,
x1 0

2Repeated application of integration by parts shows

n + 1/2 n 1/2 n 3/2 3/2 (n + 3/2)

cn = = .
n n1 n2 1 (n + 1)

With the aid of Stirlings formula, it can be shown cn 0.565 n.
3Given two functions f and g defined on R, the convolution of f and g is the integral
Z
f ? g(x) = f (t)g(x t) dt.

The term convolution kernel is used because such kernels typically replace g in the convolution
given above, as can be seen in the proof of the Weierstrass approximation theorem.
4
It was investigated by the German mathematician Edmund Landau (18771938).

May 1, 2015 http://math.louisville.edu/lee/ira

9-10 CHAPTER 9. SEQUENCES OF FUNCTIONS

because f (x) = 0 when x / [0, 1]. Notice that kn (x u) is a polynomial in u with

coefficients being polynomials in x, so integrating f (u)kn (x u) yields a polynomial
in x. (Just try it for a small value of n and a simple function f !)
Use (81) and Lemma 9.14(b) to see for (0, 1) that
Z 1 
 
(82) |pn (x) f (x)| =  f (x + t)kn (t) dt f (x)
1
Z 1 
 
=  (f (x + t) f (x))kn (t) dt
1
Z 1
|f (x + t) f (x)|kn (t) dt
1
Z Z
= |f (x + t) f (x)|kn (t) dt + |f (x + t) f (x)|kn (t) dt.
<|t|1

Well handle each of the final integrals in turn.

Let > 0 and use the uniform continuity of f to choose a (0, 1) such that
when |t| < , then |f (x + t) f (x)| < /2. Then, using Lemma 9.14(b) again,
Z

Z
(83) |f (x + t) f (x)|kn (t) dt < kn (t) dt <
2 2
According to Lemma 9.14(c), there is an N N so that when n N and |t| ,

then kn (t) < 8(kf k+1)(1) . Using this, it follows that
Z
(84) |f (x + t) f (x)|kn (t) dt
<|t|1
Z Z 1
= |f (x + t) f (x)|kn (t) dt + |f (x + t) f (x)|kn (t) dt
1
Z Z 1
2kf k kn (t) dt + 2kf k kn (t) dt
1

< 2kf k (1 ) + 2kf k (1 ) =
8(kf k + 1)(1 ) 8(kf k + 1)(1 ) 2
Combining (83) and (84), it follows from (82) that |pn (x)f (x)| < for all x [0, 1]
and pn f . 

Corollary 9.15. If f C([a, b]) and > 0, then there is a polynomial p such
that kf pk[a,b] < .
The theorems of this section can also be used to construct some striking examples
of functions with unwelcome behavior. Following is perhaps the most famous.
Example 9.9. There is a continuous f : R R that is differentiable nowhere.
Proof. Thinking of the canonical example of a continuous function that fails to
be differentiable at a pointthe absolute value functionwe start with a sawtooth
function. (See Figure 8.)
(
x 2n, 2n x < 2n + 1, n Z
s0 (x) =
2n + 2 x, 2n + 1 x < 2n + 2, n Z

May 1, 2015 http://math.louisville.edu/lee/ira

5. CONTINUITY AND UNIFORM CONVERGENCE 9-11

Notice that s0 is continuous and periodic with period 2 and maximum value 1.
Compress it both vertically and horizontally:
 n
3
sn (x) = s0 (4n x) , n N.
4
Each sn is continuous and periodic with period pn = 2/4n and ksn k = (3/4)n .

1.0
0.8
0.6
0.4
0.2

0.5 1.0 1.5 2.0 2.5 3.0 3.5

Figure 8. s0 , s1 and s2 from Example 9.9.

Finally, the desired function is

X
f (x) = sn (x).
n=0

Since ksn k = (3/4)n , the Weierstrass M -test implies the series defining f is
uniformly convergent and Corollary 9.12 shows f is continuous on R. We will show
f is differentiable nowhere.
Let x R, m N and hm = 1/(2 4m ).
If n > m, then hm /pn = 4nm1 N, so sn (x hm ) sn (x) = 0 and
m
f (x hm ) f (x) X sk (x hm ) sk (x)
(85) = .
hm hm
k=0
On the other hand, if n < m, then a worst-case estimate is that
   n  
 sn (x hm ) sn (x) 
  3 /
1
= 3n .
 hm  4 4n

3.0

2.5

2.0

1.5

1.0

0.5

0.5 1.0 1.5 2.0

Figure 9. The nowhere differentiable function f from Example 9.9.

May 1, 2015 http://math.louisville.edu/lee/ira

9-12 CHAPTER 9. SEQUENCES OF FUNCTIONS

This gives
 
m1  m1  
 X sk (x hm ) sk (x)  X  sk (x hm ) sk (x) 

hm hm
  
 
k=0 k=0
3m 1 3m
(86) < .
31 2
Since sm is linear on intervals of length 4m = 2 hm with slope 3m on those
linear segments, at least one of the following is true:
   
 sm (x + hm ) s(x) 
 = 3m or  sm (x hm ) s(x)  = 3m .
 
(87) 
 hm   hm 
Suppose the first of these is true. The argument is essentially the same in the second
case.
Using (85), (86) and (87), the following estimate ensues
   
 f (x + hm ) f (x)  X sk (x + hm ) sk (x) 
 = 
hm hm
  
 
k=0
 
m
X sk (x + hm ) sk (x) 
=

hm

 
k=0
  m1  
 sm (x + hm ) sm (x)  X  sk (x hm ) sk (x) 


  
hm  hm 
k=0

m 3m 3m
>3 = .
2 2
Since 3m /2 , it is apparent f 0 (x) does not exist. 
There are many other constructions of nowhere differentiable continuous func-
tions. The first was published by Weierstrass [19] in 1872, although it was known
in the folklore sense among mathematicians earlier than this. (There is an English
translation of Weierstrass paper in [9].) In fact, it is now known in a technical
sense that the typical continuous function is nowhere differentiable [4].

6. Integration and Uniform Convergence

One of the recurring questions with integrals is when it is true that
Z Z
lim fn = lim fn .
n n

This is often referred to as passing the limit through the integral. At some point
in her career, any student of advanced analysis or probability theory will be tempted
to just blithely pass the limit through. But functions such as those of Example
9.3 show that some care is needed. A common criterion for doing so is uniform
convergence.
Rb
Theorem 9.16. If fn : [a, b] R such that a fn exists for each n and fn f
on [a, b], then
Z b Z b
f = lim fn
a n a
.

May 1, 2015 http://math.louisville.edu/lee/ira

6. INTEGRATION AND UNIFORM CONVERGENCE 9-13

Proof. Some care must be taken in this proof, because there are actually two
things to prove. Before the equality can be shown, it must be proved that f is
integrable.
To show that f is integrable, let > 0 and N N such that kf fN k < /3(ba).
If P part ([a, b]), then
n
X n
X
(88) |R (f, P, xk ) R (fN , P, xk ) | = | f (xk )|Ik | fN (xk )|Ik ||
k=1 k=1
N
X
=| (f (xk ) fN (xk ))|Ik ||
k=1
N
X
|f (xk ) fN (xk )||Ik |
k=1
n
X
< |Ik |
3(b a))
k=1

=
3
According to Theorem 8.10, there is a P part ([a, b]) such that whenever
P  Q1 and P  Q2 , then

(89) |R (fN , Q1 , xk ) R (fN , Q2 , yk ) | < .
3
Combining (88) and (89) yields
|R (f, Q1 , xk ) R (f, Q2 , yk )|
= |R (f, Q1 , xk ) R (fN , Q1 , xk ) + R (fN , Q1 , xk )
R (fN , Q1 , xk ) + R (fN , Q2 , yk ) R (f, Q2 , yk )|
|R (f, Q1 , xk ) R (fN , Q1 , xk )| + |R (fN , Q1 , xk ) R (fN , Q1 , xk )|
+ |R (fN , Q2 , yk ) R (f, Q2 , yk )|

< + + =
3 3 3
Another application of Theorem 8.10 shows that f is integrable.
Finally, when n N ,
Z Z b  Z b  Z
 b  b

f fn  =  (f fn ) < = <
   
3(b a) 3

 a a   a  a
Rb Rb
shows that a fn a f . 
P
Corollary 9.17. If n=1 fn is a series of integrable functions converging
uniformly on [a, b], then
Z bX X Z b
fn = fn
a n=1 n=1 a

Example 9.10. It was shown in Example 4.2 that the geometric series

X 1
tn = , 1 < t < 1.
n=0
1t

May 1, 2015 http://math.louisville.edu/lee/ira

9-14 CHAPTER 9. SEQUENCES OF FUNCTIONS

In Exercise 9.3, you are asked to prove this convergence is uniform on any compact
subset of (1, 1). Substituting t for t in the above formula, it follows that

X 1
(t)n
n=0
1+t

on [0, x], when 0 < x < 1. Corollary 9.17 implies

Z x Z x
dt X
ln(1 + x) = = (t)n dt = x x2 + x3 x4 + .
0 1+t n=0 0

The same argument works when 1 < x < 0, so

ln(1 + x) = x x2 + x3 x4 +
when x (1, 1).
Combining Theorem 9.16 with Dinis Theorem, gives the following.
Corollary 9.18. If fn is a sequence of continuous functions converging mono-
Rb Rb
tonically to a continuous function f on [a, b], then a fn a f .

7. Differentiation and Uniform Convergence

The relationship between uniform convergence and differentiation is somewhat
more complex than those weve already examined. First, because there are two
sequences involved, fn and fn0 , either of which may converge or diverge at a point;
and second, because differentiation is more delicate than continuity or integration.
Example 9.4 is an explicit example of a sequence of differentiable functions
converging uniformly to a function which is not differentiable at a point. The
derivatives of the functions from that example converge pointwise to a function
that is not a derivative. The Weierstrass Approximation Theorem and Example
9.9 push this to the extreme by showing the existence of a sequence of polynomials
converging uniformly to a continuous nowhere differentiable function.
The following theorem starts to shed some light on the situation.
Theorem 9.19. If fn is a sequence of derivatives defined on [a, b] and fn f ,
then f is a derivative.
Proof. For each n, let Fn be an antiderivative of fn . By considering Fn (x)
Fn (a), if necessary, there is no generality lost with the assumption that Fn (a) = 0.
Let > 0. There is an N N such that

m, n N = kfm fn k < .
ba
If x [a, b] and m, n N , then the Mean Value Theorem and the assumption that
Fm (a) = Fn (a) = 0 yield a c [a, b] such that
|Fm (x) Fn (x)| = |(Fm (x) Fn (x)) (Fm (a) Fn (a))|
(90) = |fm (c) fn (c)| |x a| kfm fn k(b a) < .
This shows Fn is a Cauchy sequence in C([a, b]) and there is an F C([a, b]) with
Fn F .
It suffices to show F 0 = f . To do this, several estimates are established.

May 1, 2015 http://math.louisville.edu/lee/ira

7. DIFFERENTIATION AND UNIFORM CONVERGENCE 9-15

Let M N so that

m, n M = kfm fn k < .
3
Notice this implies

(91) kf fn k , n M.
3
For such m, n M and x, y [a, b] with x = 6 y, another application of the
Mean Value Theorem gives
 
 Fn (x) Fn (y) Fm (x) Fm (y) 
 
 xy xy 
1
= |(Fn (x) Fm (x)) (Fn (y) Fm (y))|
|x y|
1
= |fn (c) fm (c)| |x y| kfn fm k < .
|x y| 3
Letting m , it follows that
 
 Fn (x) Fn (y) F (x) F (y) 
(92)   , n M.
 xy xy  3
Fix n M and x [a, b]. Since Fn0 (x) = fn (x), there is a > 0 so that
 
 Fn (x) Fn (y) 
(93)  fn (x) < , y (x , x + ) \ {x}.
 xy 3
Finally, using (92), (93) and (91), we see
 
 F (x) F (y) 
 f (x)
 xy

 F (x) F (y) Fn (x) Fn (y)
= 
xy xy

Fn (x) Fn (y) 
+ fn (x) + fn (x) f (x)
xy
 
 F (x) F (y) Fn (x) Fn (y) 

 
xy xy 
 
 Fn (x) Fn (y) 
+  fn (x) + |fn (x) f (x)|
xy

< + + = .
3 3 3
This establishes that
F (x) F (y)
lim = f (x),
yx xy
as desired. 
Corollary 9.20. If Gn C([a, b]) is a sequence such that G0n g and Gn (x0 )
converges for some x0 [a, b], then Gn G where G0 = g.
Proof. Suppose Gn (x0 ) . For each n choose an antiderivative Fn of gn
such that Fn (a) = 0. Theorem 9.19 shows g is a derivative and an argument similar
to that in the proof of Theorem 9.19 shows Fn F on [a, b], where F 0 = g. Since

May 1, 2015 http://math.louisville.edu/lee/ira

9-16 CHAPTER 9. SEQUENCES OF FUNCTIONS

Fn0 G0n = 0, Corollary (7.15) shows Gn (x) = Fn (x) + (Gn (x0 ) Fn (x0 )). Define
G(x) = F (x) + ( F (x0 )).
Let > 0 and x [a, b]. There is an N N such that

n N = kFn F k < and |Gn (x0 ) | < .
3 3
If n N ,
|Gn (x) G(x)| = |Fn (x) + (Gn (x0 ) Fn (x0 )) (F (x) + ( F (x0 )))|
|Fn (x) F (x)| + |Gn (x0 ) | + |Fn (x0 ) F (x0 )|

< + +
3 3 3
=
This shows Gn G on [a, b] where G0 = F 0 = g. 

P 9.21. If fn is a sequence of differentiable

Corollary Pfunctions defined on [a, b]
such that k=1 fk (x0 ) exists for some x0 [a, b] and k=1 fk0 converges uniformly,
then !0
X X
f = f0
k=1 k=1

Proof. Left as an exercise. 

Example 9.11. P n
Let a > 0 and fn (x) = x /n!. Note that fn0
= fn1 for n N.

Example 9.8 shows n=0 fn0 (x) is uniformly convergent on [a, a]. Corollary 9.21
shows

!0
X xn X xn
(94) = .
n=0
n! n=0
n!
on [a, a]. Since
Pa is an arbitrary positive constant, (94) is seen to hold on all of R.
If f (x) = n=0 xn /n!, then the argument given above implies the initial value
problem (
f 0 (x) = f (x)
f (0) = 1
As is well-known, the unique solution to this problem is f (x) = ex . Therefore,

X xn
ex = .
n=0
n!

8. Power Series
8.1. The Radius and Interval of Convergence. One place where uniform
convergence plays a key role is with power series. Recall the definition.
Definition 9.22. A power series is a function of the form
X
(95) f (x) = an (x c)n .
n=0
Members of the sequence an are the coefficients of the series. The domain of f is
the set of all x at which the series converges. The constant c is called the center of
the series.

May 1, 2015 http://math.louisville.edu/lee/ira

8. POWER SERIES 9-17

To determine the domain of (95), let x R \ {c} and use the root test to see
the series converges when
lim sup |an (x c)n |1/n = |x c| lim sup |an |1/n < 1
and diverges when
|x c| lim sup |an |1/n > 1.
If r lim sup |an |1/n 1 for some r 0, then these inequalities imply (95) is absolutely
convergent when |x c| < r. In other words, if
(96) R = lub {r : r lim sup |an |1/n < 1},
then the domain of (95) is an interval of radius R centered at c. The root test gives
no information about convergence when |x c| = R. This R is called the radius
of convergence of the power series. Assuming R > 0, the open interval centered
at c with radius R is called the interval of convergence. It may be different from
the domain of the series because the series may converge at one endpoint or both
endpoints of the interval of convergence.
The ratio test can also be used to determine the radius of convergence, but, as
shown in (30), it will not work as often as the root test. When it does,
 
 an+1 
(97) R = lub {r : r lim 
  < 1}.
n an 
This is usually easier to compute than (96), and both will give the same value for R.
Example 9.12. Calling to mind Example 4.2, it is apparent the geometric
P
power series n=0 xn has center 0, radius of convergence 1 and domain (1, 1).
P
Example 9.13. For the power series n=1 2n (x + 2)n /n, we compute
 n 1/n
2 1
lim sup = 2 = R = .
n 2
Since the series diverges when x = 2 12 , it follows that the interval of convergence
is (5/2, 3/2).
P
Example 9.14. The power series n=1 xn /n has interval of convergence (1, 1)
and domain [1, 1). Notice it is not absolutely convergent when x = 1.
P n 2
Example 9.15. The power series n=1 x /n has interval of convergence
(1, 1), domain [1, 1] and is absolutely convergent on its whole domain.
The preceding is summarized in the following theorem.
Theorem 9.23. Let the power series be as in (95) and R be given by either
(96) or (97).
(a) If R = 0, then the domain of the series is {c}.
(b) If R > 0 the series converges absolutely at x when |c x| < R and
diverges at x when |c x| > R. In the case when R = , the series
converges everywhere.
(c) If R (0, ), then the series may converge at none, one or both of
c R and c + R.

May 1, 2015 http://math.louisville.edu/lee/ira

9-18 CHAPTER 9. SEQUENCES OF FUNCTIONS

8.2. Uniform Convergence of Power Series. The partial sums of a power

series are a sequence of polynomials converging pointwise on the domain of the
series. As has been seen, pointwise convergence is not enough to say much about
the behavior of the power series. The following theorem opens the door to a lot
more.
Theorem 9.24. A power series converges absolutely and uniformly on compact
subsets of its interval of convergence.
Proof. There is no generality lost in assuming the series has the form of
(95) with c = 0. Let the radius of convergence be R > 0 and K be a compact
subset of (R, R) with = lub {|x| : x K}. Choose r (, R). If x K, then
P
|an xn | P
< |an rn | for n N. Since n=0 |an rn | converges, the Weierstrass M -test

shows n=0 an xn is absolutely and uniformly convergent on K. 
The following two corollaries are immediate consequences of Corollary 9.12 and
Theorem 9.16, respectively.
Corollary 9.25. A power series is continuous on its interval of convergence.
Corollary 9.26.PIf [a, b] is an interval contained in the interval of convergence

for the power series n=0 an (x c)n , then
Z bX X Z b
an (x c)n = an (x c)n .
a n=0 n=0 a

Example 9.16. Define

(
sin x
x , x 6= 0
f (x) = .
1, x=0
R
Since limx0 f (x) = 1, f is continuous everywhere. Suppose we want 0
f with an
accuracy of five decimal places.
If x 6= 0,

1 X (1)n+1 2n1 X (1)n 2n
f (x) = x = x
x n=1 (2n 1)! n=0
(2n + 1)!
The latter series converges to f everywhere. Corollary 9.26 implies

Z Z X !
(1)n 2n
f (x) dx = x dx
0 0 n=0
(2n + 1)!
Z
X (1)n
= x2n dx
n=0
(2n + 1)! 0

X (1)n
(98) = 2n+1
n=0
(2n + 1)(2n + 1)!

The latter series satisfies the Alternating Series Test. Since 1 5/(15 15!)
1.5 106 , Corollary 4.20 shows
Z 6
X (1)n
f (x) dx 2n+1 1.85194
0 n=0
(2n + 1)(2n + 1)!

May 1, 2015 http://math.louisville.edu/lee/ira

8. POWER SERIES 9-19

The next question is: What about differentiability?

Notice that the continuity of the exponential function and LHospitals Rule
give    
1/n ln n ln n
lim n = lim exp = exp lim = exp(0) = 1.
n n n n n
Therefore, for any sequence an ,
(99) lim sup(nan )1/n = lim sup n1/n a1/n
n = lim sup a1/n
n .
P n
Now, suppose the power series n=0 an x has a nontrivial interval of conver-
gence, PI. Formally differentiating the power series term-by-term gives a new power
n1
series n=1 nan x . According to (99) and Theorem 9.23, the term-by-term
differentiated series has the same interval of convergence as the original. Its partial
sums are the derivatives of the partial sums of the original series and Theorem 9.24
guarantees they converge uniformly on any compact subset of I. Corollary 9.21
shows

d X X d X
an xn = an xn = nan xn1 , x I.
dx n=0 n=0
dx n=1
This process can be continued inductively to obtain the same results for all higher
order derivatives. We have proved the following theorem.
P
Theorem 9.27. If f (x) = n=0 an (x c)n is a power series with nontrivial
interval of convergence, I, then f is differentiable to all orders on I with

X n!
(100) f (m) (x) = an (x c)nm .
n=m
(n m)!
Moreover, the differentiated series has I as its interval of convergence.
P n
8.3. Taylor Series. Suppose f (x) = n=0 an x has I = (R, R) as its
interval of convergence for some R > 0. According to Theorem 9.27,
m! f (m) (0)
f (m) (0) = am = am = , m .
(m m)! m!
Therefore,

X f (n) (0) n
f (x) = x , x I.
n=0
n!
This is a remarkable result! It shows that the values of f on I are completely
determined by its values on any neighborhood of 0. This is summarized in the
following theorem.
P n
Theorem 9.28. If a power series f (x) = n=0 an (x c) has nontrivial
interval of convergence I, then

X f (n) (c)
(101) f (x) = (x c)n , x I.
n=0
n!

The series (101) is called the Taylor series 5 for f centered at c. The Taylor
series can be formally defined for any function that has derivatives of all orders at c,
but, as Example 7.9 shows, there is no guarantee it will converge to the function
anywhere except at c. Taylors Theorem 7.17 can be used to examine the question
5When c = 0, it is often called the Maclaurin series for f .

May 1, 2015 http://math.louisville.edu/lee/ira

9-20 CHAPTER 9. SEQUENCES OF FUNCTIONS

of pointwise convergence. If f can be represented by a power series on an open

interval I, then f is said to be analytic on I.

8.4. The Endpoints of the Interval of Convergence. We have seen that

at the endpoints of its interval of convergence a power series may diverge or even
absolutely converge. A natural question when it does converge is the following:
What is the relationship between the value at the endpoint and the values inside
the interval of convergence?

Theorem 9.29 (Abel). A power series is continuous on its domain.

P
Proof. Let f (x) = n=0 (x c)n have radius of convergence R and interval
of convergence I. If I = {0}, the theorem is vacuously true from Definition 6.9. If
I = R, the theorem follows from Corollary 9.25. So, assume R (0, ). It must be
shown that if f converges at an endpoint of I = (c R, c + R), then f is continuous
at that endpoint.
It can be assumed c = 0 and R = 1. There is no loss of generality with either
of these assumptions because otherwise just replace f (x) with f ((x + c)/R). The
theorem will be proved for = c + RPsince the other case is proved similarly.
n
Set s = f (1), s1 = 0 and sn = k=0 ak for n . For |x| < 1,
n
X n
X
ak xk = (sk sk1 )xk
k=0 k=0
Xn n
X
= sk xk sk1 xk
k=0 k=1
n1
X n1
X
= sn xn + sk xk x sk xk
k=0 k=0
n1
X
= sn xn + (1 x) sk xk
k=0

When n , since sn is bounded and |x| < 1,

X
(102) f (x) = (1 x) sk xk .
k=0
P
Since (1 x) n=0 xn = 1, (102) implies

 
 X 
(103) |f (x) s| = (1 x) (sk s)xk  .
 
 
k=0

Let > 0. Choose N N such that whenever n N , then |sn s| < /2. Choose
(0, 1) so
N
X
|sk s| < /2.
k=0

May 1, 2015 http://math.louisville.edu/lee/ira

8. POWER SERIES 9-21

Suppose x is such that 1 < x < 1. With these choices, (103) becomes

   
 XN   X 
|f (x) s| (1 x) (sk s)xk  + (1 x) (sk s)xk 
   
   
k=0 k=N +1

N
 
X  X 
< |sk s| + (1 x) xk  < + =

2  2 2
k=0 k=N +1

It has been shown that limx1 f (x) = f (1), so 1 C(f ). 

Here is an example showing the power of these techniques.
Example 9.17. The series

X 1
(1)n x2n =
n=0
1 + x2
has (1, 1) as its interval of convergence. If 0 |x| < 1, then Corollary 9.17 justifies
Z x Z xX
dt n 2n
X (1)n 2n+1
arctan(x) = 2
= (1) t dt = x .
0 1+t 0 n=0 n=0
2n + 1
This series for the arctangent converges by the alternating series test when x = 1,
so Theorem 9.29 implies

X (1)n
(104) = lim arctan(x) = arctan(1) = .
n=0
2n + 1 x1 4
A bit of rearranging gives the formula
 
1 1 1
= 4 1 + + ,
3 5 7
which is known as Gregorys series for .
Finally,
PAbels theorem opens up an interesting idea for the summation of series.
Suppose n=0 an is a series. The Abel sum of this series is

X
X
A an = lim an xn .
x1
n=0 n=0
Consider the following example.
Example 9.18. Let an = (1)n so
X
an = 1 1 + 1 1 + 1 1 +
n=0
diverges. But,

X X 1 1
A an = lim (x)n = lim = .
n=0
x1
n=0
x1 1+x 2
This shows the Abel sum of a series may exist when the ordinary sum does not.
Abels theorem guarantees when both exist they are the same.
Abel summation is one of many different summation methods used in areas
such as harmonic analysis. (For another see Exercise 4.26.)
P
Theorem 9.30 (Tauber). If n=0 an is a series satisfying

May 1, 2015 http://math.louisville.edu/lee/ira

9-22 CHAPTER 9. SEQUENCES OF FUNCTIONS

n 0 and
(a) naP

(b) A n=0 an = A,
P
then n=0 an =A.
Pn
Proof. Let sn = k=0 ak . For x (0, 1) and n N,

   
 X  X n Xn X 
sn ak xk  =  ak ak xk ak xk 
   
   
k=0 k=0 k=0 k=n+1

 n 
X X 
k
= ak (1 x ) ak xk 
 
 
k=0 k=n+1

 n 
X X 
k1 k
= ak (1 x)(1 + x + + x ) ak x 

 
k=0 k=n+1
n
X
X
(105) (1 x) k|ak | + |ak |xk .
k=0 k=n+1

Let > 0. According to (a) and Exercise 3.19, there is an N N such that
n
1X
(106) n N = n|an | < and k|ak | < .
2 n 2
k=0

Let n N and 1 1/n < x < 1. Using the right term in (106),
n n
X 1X
(107) (1 x) k|ak | < k|ak | < .
n 2
k=0 k=0

Using the left term in (106) gives

X X k
|ak |xk < x
2k
k=n+1 k=n+1
n+1
x
(108) <
2n 1 x

< .
2
Combining (106) and (106) with (105) shows

 
 X 
k
sn ak x  < .

 
k=0

Assumption (b) implies sn A. 

9. Exercises

9.1. If fn (x) = nx(1 x)n for 0 x 1, then show fn converges pointwise, but
not uniformly on [0, 1].

9.2. Show sinn x converges uniformly on [0, a] for all a (0, /2). Does sinn x
converge uniformly on [0, /2)?

May 1, 2015 http://math.louisville.edu/lee/ira

9. EXERCISES 9-23

xn converges uniformly on [r, r] when 0 < r < 1, but not on

P
9.3. Show that
(1, 1).
P
9.4. Prove n=0 xn /n! does not converge uniformly on R.

9.5. The series

X cos nx
n=0
enx
is uniformly convergent on any set of the form [a, ) with a > 0.

9.6. A sequence of functions fn : S R is uniformly bounded on S if there is an

M > 0 such that kfn kS M for all n N. Prove that if fn is uniformly convergent
on S and each fn is bounded on S, then the sequence fn is uniformly bounded on S.

9.7. Let S R and c R. If fn : S R is a Cauchy sequence, then so is cfn .

9.8. If S R and fn , gn : S R are Cauchy sequences, then so is fn + gn .

9.9. Let S R. If fn , gn : S R are uniformly bounded Cauchy sequences, then

so is fn gn .

9.10. Prove or give a counterexample: If fn is a sequence of monotone functions

converging pointwise to a continuous function f , then fn f .

9.11. Prove or give a counterexample: If fn : [a, b] R is a sequence of monotone

functions converging pointwise to a continuous function f , then fn f .

9.12. Prove there is a sequence of polynomials on [a, b] converging uniformly to a

nowhere differentiable function.
9.13. Prove Corollary 9.21.

X (1)n
9.14. Prove = 2 3 . (This is the Madhava-Leibniz series which
n=0
3n (2n + 1)
was used in the fourteenth century to compute to 11 decimal places.)

X xn d
9.15. If exp(x) = , then dx exp(x) = exp(x) for all x R.
n=0
n!

X 1
9.16. Is Abel convergent?
n=2
n ln n

May 1, 2015 http://math.louisville.edu/lee/ira

 CHAPTER 10

Fourier Series

In the late eighteenth century, it was well-known that complicated functions

could sometimes be approximated by a sequence of polynomials. Some of the leading
mathematicians at that time, including such luminaries as Daniel Bernoulli, Euler
and dAlembert began studying the possibility of using sequences of trigonometric
functions for approximation. In 1807, this idea opened into a huge area of research
when Joseph Fourier used series of sines and cosines to solve several outstanding
partial differential equations of physics.1
In particular, he used series of the form

X
an cos nx + bn sin nx
n=0

to approximate his solutions. Series of this form are called trigonometric series,
and the ones derived from Fouriers methods are called Fourier series. Much of
the mathematical research done in the nineteenth and early twentieth century was
devoted to understanding the convergence of Fourier series. This chapter presents
nothing more than the tip of that huge iceberg.

1. Trigonometric Polynomials
Definition 10.1. A function of the form
n
X
(111) p(x) = k cos kx + k sin kx
k=0

is called a trigonometric polynomial. The largest value of k such that |k |+k | =

6 0 is
the degree of the polynomial. Denote by T the set of all trigonometric polynomials.
Evidently, all functions in T are 2-periodic and T is closed under addition
and multiplication by real numbers. Indeed, it is a real vector space, in the sense of
linear algebra and the set {sin nx : n N} {cos nx : n } is a basis for T .
The following theorem can be proved using integration by parts or trigonometric
identities.
Theorem 10.2. If m, n Z, then
Z
(112) sin mx cos nx dx = 0,

1
Fouriers methods can be seen in most books on partial differential equations, such as [3]. For
example, see solutions of the heat and wave equations using the method of separation of variables.

10-1
10-2 CHAPTER 10. FOURIER SERIES

Z 0,
m 6= n
(113) sin mx sin nx dx = 0, m = 0 or n = 0

m = n 6= 0

and

Z 0,
m 6= n
(114) cos mx cos nx dx = 2 m=n=0.

m = n 6= 0

If p(x) is as in (111), then Theorem 10.2 shows

Z
20 = p(x) dx,

and for n > 0,

Z Z
n = p(x) cos nx dx, n = p(x) cos nx dx.

Combining these, it follows that if

1 1
Z Z
an = p(x) cos nx dx and bn = p(x) sin nx dx

for n , then

a0 X
(115) p(x) = + an cos nx + bn sin nx.
2 n=1

(Remember that all but a finite number of the an and bn are 0!)
At this point, the logical question is whether this same method can be used to
represent a more general 2-periodic function. For any function f , integrable on
[, ], the coefficients can be defined as above; i.e., for n ,
1 1
Z Z
(116) an = f (x) cos nx dx and bn = f (x) sin nx dx.

The numbers an and bn are called the Fourier coefficients of f . The problem is
whether and in what sense an equation such as (115) might be true. This turns out
to be a very deep and difficult question with no short answer.2 Because we dont
know whether equality in the sense of (115) is true, the usual practice is to write

a0 X
(117) f (x) + an cos nx + bn sin nx,
2 n=1

indicating that the series on the right is calculated from the function on the left
using (116). The series is called the Fourier series for f .

2Many people, including me, would argue that the study of Fourier series has been the
most important area of mathematical research over the past two centuries. Huge mathematical
disciplines, including set theory, measure theory and harmonic analysis trace their lineage back to
basic questions about Fourier series. Even after centuries of study, research in this area continues
unabated.

May 1, 2015 http://math.louisville.edu/lee/ira

2. THE RIEMANN LEBESGUE LEMMA 10-3

Figure 1. This shows f (x) = |x|, s1 (x) and s3 (x), where sn (x) is
the nth partial sum of the Fourier series for f .

Example 10.1. Let f (x) = |x|. Since f is an even functions and sin nx is odd,
1
Z
bn = |x| cos nx dx = 0

for all n N. On the other hand,

1
Z , n=0
an = |x| cos nx dx = 2(cos n 1)
, nN
n2
for n Z. Therefore,
4 cos x 4 cos 3x 4 cos 5x 4 cos 7x 4 cos 9x
|x| +
2 9 25 49 81
(See Figure 10.1.)
There are at least two fundamental questions arising from (117): Does the
Fourier series of f converge to f ? Can f be recovered from its Fourier series, even
if the Fourier series does not converge to f ? These are often called the convergence
and representation questions, respectively. The next few sections will give some
partial answers.

2. The Riemann Lebesgue Lemma

We learned early in our study of series that the first and simplest convergence
test is to check whether the terms go to zero. For Fourier series, this is always the
case.
Rb
Theorem 10.3 (Riemann-Lebesgue Lemma). If f is a function such that a f
exists, then
Z b Z b
lim f (t) cos t dt = 0 and lim f (t) sin t dt = 0.
a a

May 1, 2015 http://math.louisville.edu/lee/ira

10-4 CHAPTER 10. FOURIER SERIES

Proof. Since the two limits have similar proofs, only the first will be proved.
Let > 0 and P be a generic partition of [a, b] satisfying
Z b

0< f D (f, P ) < .
a 2
For mi = glb {f (x) : xi1 < x < xi }, define a function g on [a, b] by g(x) = mi
Rb
when xi1 x < xi and g(b) = mn . Note that a g = D (f, P ), so
Z b

(118) 0< (f g) < .
a 2
Choose
n
4X
(119) > |mi |.
i=1

Since f g,
Z  Z 
 b   b Z b 
f (t) cos t dt =  (f (t) g(t)) cos t dt + g(t) cos t dt
   

 a   a a 
Z  Z 
 b   b 
 (f (t) g(t)) cos t dt +  g(t) cos t dt
   
 a   a 
 
Z b 1 Xn 
(f g) +  mi (sin(xi ) sin(xi1 ))
 
a  
i=1
Z b n
2X
(f g) + |mi |
a i=1

Use (118) and (119).

< +
2 2
=


Corollary 10.4. If f is integrable on [, ] with an and bn the Fourier

coefficients of f , then an 0 and bn 0.

3. The Dirichlet Kernel

Suppose f is integrable on [, ] and 2-periodic on R, so the Fourier series
of f exists. The partial sums of the Fourier series are written as sn (f, x), or more
simply sn (x) when there is only one function in sight. To be more precise,
n
a0 X
sn (f, x) = + (ak cos kx + bk sin kx) .
2
k=1

Notice sn is a trigonometric polynomial of degree at most n.

We begin with the following calculation.

May 1, 2015 http://math.louisville.edu/lee/ira

3. THE DIRICHLET KERNEL 10-5

15

12

- -
2 2

-3

Figure 2. The Dirichlet kernel Dn (s) for n = 1, 4, 7.

n
a0 X
sn (x) = + (ak cos kx + bk sin kx)
2
k=1
Z n
1 1
X Z
= f (t) dt + (f (t) cos kt cos kx + f (t) sin kt sin kx) dt
2
k=1
n
Z !
1 X
= f (t) 1 + 2(cos kt cos kx + sin kt sin kx) dt
2
k=1
n
Z !
1 X
= f (t) 1 + 2 cos k(x t) dt
2
k=1

Substitute s = x t and use the assumption that f is 2-periodic.

n
Z !
1 X
(120) = f (x s) 1 + 2 cos ks ds
2
k=1
The sequence of trigonometric polynomials from within the integral,
n
X
(121) Dn (s) = 1 + 2 cos ks,
k=1
is called the Dirichlet kernel. Its properties will prove useful for determining the
pointwise convergence of Fourier series.
Theorem 10.5. The Dirichlet kernel has the following properties.
(a) Dn (s) is an even 2-periodic function for each n N.
(b) Dn (0) = 2n + 1 for each n N.
(c) |DnZ(s)| 2n + 1 for each n N and all s.

1
(d) Dn (s) ds = 1 for each n N.
2

May 1, 2015 http://math.louisville.edu/lee/ira

10-6 CHAPTER 10. FOURIER SERIES

sin(n + 1/2)s
(e) Dn (s) = for each n N and s/2 not an integer multiple
sin s/2
of .
Proof. Properties (a)(d) follow from the definition of the kernel.
The proof of property (e) uses some trigonometric manipulation. Suppose n N
and s 6= m for any m Z.
n
X
Dn (s) = 1 + 2 cos ks
k=1

Use the facts that the cosine is even and the sine is odd.
n n
X cos 2s X
= cos ks + sin ks
sin 2s
k=n k=n
n 
1 X s s 
= sin cos ks + cos sin ks
sin 2s 2 2
k=n
n
1 X 1
= sin(k + )s
sin 2s 2
k=n

This is a telescoping sum.

sin(n + 21 )s
=
sin 2s

According to (120),

1
Z
sn (f, x) = f (x t)Dn (t) dt.
2

Figure 3. This graph shows D50 (t) and the envelope y = 1/ sin(t/2).
As n gets larger, the Dn (t) fills the envelope more completely.

May 1, 2015 http://math.louisville.edu/lee/ira

4. DINIS TEST FOR POINTWISE CONVERGENCE 10-7

This is similar to a situation weve seen before within the proof of the Weierstrass
approximation theorem, Theorem 9.13. The integral given above is a convolution
integral similar to that used in the proof of Theorem 9.13, although the Dirichlet
kernel isnt a convolution kernel in the sense of Lemma 9.14 because it doesnt
satisfy conditions (a) and (c) of that lemma. (See Figure 10.3.)

4. Dinis Test for Pointwise Convergence

Theorem 10.6 (Dinis Test). Let f : R R be a 2-periodic function integrable
on [, ] with Fourier series given by (117). If there is a > 0 and s R such
that Z  
 f (x + t) + f (x t) 2s 
  dt < ,
0
 t 
then

a0 X
+ (ak cos kx + bk cos kx) = s.
2
k=1

Proof. Since Dn is even,

Z
1
sn (x) = f (x t)Dn (t) dt
2
Z 0 Z
1 1
= f (x t)Dn (t) dt + f (x t)Dn (t) dt
2 2 0
Z
1
= (f (x + t) + f (x t)) Dn (t) dt.
2 0
By Theorem 10.5(d) and (e),
Z
1
sn (x) s = (f (x + t) + f (x t) 2s) Dn (t) dt
2 0
Z
1 f (x + t) + f (x t) 2s t 1
= sin(n + )t dt.
2 0 t sin 2t 2
Since t/ sin 2t is bounded on (0, ), Theorem 10.3 shows sn (x) s 0. Now use
Corollary 8.11 to finish the proof. 

Example 10.2. Suppose f (x) = x for < x and is 2-periodic on R.

Since f is odd, an = 0 for all n. Integration by parts gives bn = (1)n+1 2/n for
n N. Therefore,

X 2
f (1)n+1 sin nx.
n=1
n
For x (, ), let 0 < < min{ x, + x}. (This is just the distance from x to
closest endpoint of (, ).) Using Dinis test, we see
Z   Z  
 f (x + t) + f (x t) 2x   x + t + x t 2x 
  dt =   dt = 0 < ,
0
 t 
0
 t 
so

X 2
(122) (1)n+1 sin nx = x for < x < .
n=1
n

May 1, 2015 http://math.louisville.edu/lee/ira

10-8 CHAPTER 10. FOURIER SERIES

- - 2
2 2

-
2

Figure 4. This plot shows the function of Example 10.2 and s8 (x)
for that function.

In particular, when x = /2, (122) gives another way to derive (104). When x = ,
the series converges to 0, which is the middle of the jump for f .
This behavior of converging to the middle of a jump discontinuity is typical. To
see this, denote the one-sided limits of f at x by
f (x) = lim f (t) and f (x+) = lim f (t),
tx tx

and suppose f has a jump discontinuity at x with

f (x) + f (x+)
s= .
2
Guided by Dinis test, consider
Z  
 f (x + t) + f (x t) 2s 
  dt
0
 t 
Z  
 f (x + t) + f (x t) f (x) f (x+) 
=   dt
0
 t 
Z   Z  
 f (x + t) f (x+)   f (x t) f (x) 

 dt +
   dt
0
 t 0
 t 
If both of the integrals on the right are finite, then the integral on the left is also
finite. This amounts to a proof of the following corollary.
Corollary 10.7. Suppose f : R R is 2-periodic and integrable on [, ].
If both one-sided limits exist at x and there is a > 0 such that both
Z   Z  
 f (x + t) f (x+)   f (x t) f (x) 
  dt < and   dt < ,
0
 t  
0 t 
then the Fourier series of f converges to
f (x) + f (x+)
.
2

May 1, 2015 http://math.louisville.edu/lee/ira

5. GIBBS PHENOMENON 10-9

Figure 5. This is a plot of s19 (f, x), where f is defined by (123).

The Dini test given above provides a powerful condition sufficient to ensure the
pointwise convergence of a Fourier series. There is a plethora of ever more abstruse
conditions that can be proved in a similar fashion to show pointwise convergence.
The problem is complicated by the fact that there are continuous functions
with Fourier series divergent at a point and integrable functions with Fourier series
diverging everywhere [12]. In Section 6, a continuous function whose Fourier series
diverges is constructed.

5. Gibbs Phenomenon
For x [, ) define
(
|x|
(123) f (x) = x , 0<x<
0, x = 0,
and extend f 2-periodically to all of R. This function is often called a square wave.
A straightforward calculation gives

4 X sin(2k 1)x
f .
2k 1
k=1

Corollary 10.7 shows sn (x) f (x) everywhere. This convergence cannot be uniform
because all the partial sums are continuous and f is discontinuous at every integer
multiple of . A plot of s19 (x) is shown in Figure 5. Notice the higher peaks in
the oscillation of sn (x) just before and after the jump discontinuities of f . This
behavior is not unique to f , as it can also be seen in Figure 4. If a function is
discontinuous at some point, the partial sums of its Fourier series will always have
such higher peaks near that point. This behavior is called Gibbs phenomenon.3
Instead of doing a general analysis of Gibbs phenomenon, well only analyze
the simple case shown in the square wave f . Its basically a calculus exercise.

3It is named after the American mathematical physicist, J. W. Gibbs, who pointed it out
in 1899. He was not the first to notice the phenomenon, as the British mathematician Henry
Wilbraham had published a little-noticed paper on it it 1848. Gibbs interest in the phenomenon
was sparked by investigations of the American experimental physicist A. A. Michelson who wrote
a letter to Nature disputing the possibility that a Fourier series could converge to a discontinuous
function. The ensuing imbroglio is recounted in a marvelous book by Paul J. Nahin [15].

May 1, 2015 http://math.louisville.edu/lee/ira

10-10 CHAPTER 10. FOURIER SERIES

Figure 6. This is a plot of s9 (f, x), where f is defined by (123).

To locate the peaks in the graph, differentiate the partial sums.

n n
d 4 X sin(2k 1)x 4X
s02n1 (x) = = cos(2k 1)x
dx 2k 1
k=1 k=1
It is left as Exercise 10.10 to show this has a closed form.
2 sin 2nx
s02n1 (x) =
sin x
Looking at the numerator, we see s02n1 (x) has critical numbers at x = k/2n
for k Z. In the interval (0, ), s2n1 (k/2n) is a relative maximum for odd k and
a relative minimum for even k. (See Figure 6.) The value s2n1 (/2n) is the height
of the left-most peak. What is the behavior of these maxima?
From Figure 7 it appears they have an asymptotic limit near 1.179. To prove
this, consider the following calculation.
n


  4 X sin (2k 1) 2n
s2n1 f, =
2n 2k 1
k=1
 
(2k1)
2 X sin
n
2n
= (2k1)
n
k=1 2n

Figure 7. This is a plot of sn (f, /2n) for n = 1, 2, , 100. The dots

come in pairs because s2n1 (f, /2n) = s2n (f, /2n).

May 1, 2015 http://math.louisville.edu/lee/ira

6. A DIVERGENT FOURIER SERIES 10-11

The last sum is a midpoint Riemann sum for the function sinx x on the interval [0, ]
using a regular partition of n subintervals. Example 9.16 shows

2 sin x
Z
dx 1.17898.
0 x

Since f (0+) f (0) = 2, this is an overshoot of a bit less than 9%. There is a
similar undershoot on the other side. It turns out this is typical behavior at points
of discontinuity [20].

6. A Divergent Fourier Series

This is an advanced section that can be omitted.
Eighteenth century mathematicians, including Fourier, Cauchy, Euler, Weier-
strass, Lagrange and Riemann, had computed the Fourier series for many functions.
They believed from these examples that the Fourier series of a continuous function
must converge to that function. Fourier went far beyond this claim in his important
1822 book Theorie Analytique de la Chaleur. Cauchy even published a flawed proof
of this fact.
Finally, in 1873, Paul du Bois-Reymond, settled the question by giving the
construction of a continuous function whose Fourier series diverges at a point [16].
It was finally shown in 1966 by Lennart Carleson [7] that the Fourier series of a
continuous function converges to that function everywhere with the exception of a
set of measure zero.
The problems around the convergence of Fourier series motivated a huge amount
of research that is still going on today. In this section, we look at the tip of
that iceberg by presenting a continuous function F (t) whose Fourier series fails to
converge when t = 0.

6.1. The Conjugate Dirichlet Kernel.

Lemma 10.8. If m, n and 0 < |t| < for k Z, then

n
X cos(m 12 )t cos(m + 21 )t
sin kt = .
k=m
2 sin 2t

Proof.
n n
X 1 X 1
kt = t sin sin kt
sin 2 k=m 2
k=m
n      
1 X 1 1
= cos k t cos k + t
sin 2t k=m 2 2
cos m 12 t cos n + 12 t



=
2 sin 2t

May 1, 2015 http://math.louisville.edu/lee/ira

10-12 CHAPTER 10. FOURIER SERIES

Definition 10.9. The conjugate Dirichlet kernel 4 is

n
X
(124) Dn (t) = sin kt, n N.
k=1

Well not have much use for the conjugate Dirichlet kernel, except as a convenient
way to refer to sums of the form (124).
Lemma 10.8 immediately gives the following bound.

Corollary 10.10. If 0 < |t| < , then

  1
Dn (t)  .
 
sin t  2

6.2. A Sawtooth Wave. If the function f (x) = ( x)/2 on [0, 2) is ex-

tended 2-periodically to R, then the graph of the resulting function is often referred
to as a sawtooth wave. It has a particularly nice Fourier series:

x X sin kx
.
2 k
k=1

According to Corollary 10.7

(
X sin kx 0, x = 2n, n Z
= .
k f (x), otherwise
k=1

Were interested in various partial sums of this series.

Lemma 10.11. If m, n with m n and 0 < |t| < 2, then

 
n
X sin kt  1
  .

k  m sin 2t 


k=m

Proof.
   
n n
X sin kt   X   1
= Dk (t) Dk1 (t)
 
k k
 
  
k=m k=m

Use summation by parts.

 n   
X 1 1 Dn (t) Dn1 (t) 
= Dk (t) +

k k+1 n+1 m

 
k=m
 n  
     
X 1 1   Dn (t)   Dn1 (t) 
  
 Dk (t) + +

k k+1   n+1   m

 
k=m

4In this case, the word conjugate does not refer to the complex conjugate, but to the
harmonic conjugate. They are related by Dn (t) + iDn (t) = 1 + 2 n kit .
P
k=1 e

May 1, 2015 http://math.louisville.edu/lee/ira

6. A DIVERGENT FOURIER SERIES 10-13

Apply Corollary 10.10.

n   !
1 X 1 1 1 1
+
2 sin 2t k=m
k k+1 n+1 m
 
1 1 1 1 1
= + +
2 sin 2t m n+1 n+1 m
1
= .
m sin 2t


Proposition 10.12. If n N and 0 < |t| < , then

 n 
X sin kt 
 1 + .
 
k 


k=1

Proof.
 n   
X sin kt   X sin kt X sin kt


= +
   
k   k k
 

k=1 1k 1t 1
t <kn

   
 X   X 
 sin kt  
  sin kt 
 +
1k 1 k   1 <kn k 
  
t t
X kt 1
+ 1 t
1
k t sin 2
1k t
X 1
t+ 1 2 t
1k 1t t 2

1+


6.3. A Continuous Function with a Divergent Fourier Series. The goal

in this subsection is to construct a continuous function F such that lim sup sn (F, 0) =
. This implies sn (F, 0) does not converge.
Example 10.3. The first step in the construction is to define a sequence of
trigonometric polynomials that form the building blocks for F .

1 cos t cos 2t cos(n 1)t

fn (t) = + + +
n n1 n2 1
cos(n + 1)t cos(n + 2)t cos 2nt

1 2 n
Note that,
(
sm (fn , 0) = fn (0) = 0, m 2n
(125) .
sm (fn , 0) > 0, 1 m < 2n

May 1, 2015 http://math.louisville.edu/lee/ira

10-14 CHAPTER 10. FOURIER SERIES

Rearrange the sum in the definition of fn to see

n
X cos(n k)t cos(n + k)t
fn (t) =
k
k=1
n
X sin kt
= 2 sin nt .
k
k=1

This closed form for fn combined with Proposition 10.12 implies the sequence of
functions fn is uniformly bounded:
 
n
 X sin kt 
(126) |fn (t)| = 2 sin nt  2 + 2.

 k 
k=1

At last, the main function can be defined.

X f2n3 (t)
F (t) =
n=1
n2

The Weierstrass M-Test along with (126) implies F is uniformly convergent and
therefore continuous on R. Using (125), consider
Z
1
s2m (F, 0) =
3 F (t)D2m3 (t) dt
2
Z X
1 f2n3 (t)
= D2m3 (t) dt
2 n=1 n2

The uniform convergence allows the sum and integration to be reordered.

Z
X 1 1
= f n3 (t)D2m3 (t) dt
n=1
n2 2 2

X 1

= 2
s2m3 f2n3 , 0
n=1
n

Use (125).

X 1

= s 3 f2n3 , 0
2 2m
n=m
n
1

> 2 s2m3 f2m3 , 0
m
m3
2
1 X1
= 2
m k
k=1
1 3
> 2 ln 2m
m
= m ln 2

This implies, lim sup sn (F, 0) limm m ln 2 = , so sn (F, 0) does not converge.

May 1, 2015 http://math.louisville.edu/lee/ira

7. THE FEJER KERNEL 10-15

7. The Fejer Kernel

Since pointwise convergence of the partial sums seems complicated, why not
change the rules of the game? Instead of looking at the sequence of partial sums,
consider a rolling average instead:
n
1 X
n (f, x) = sn (f, x).
n+1
k=0

The trigonometric polynomials n (f, x) are called the Cesaro means of the partial
sums. If limn n (f, x) exists, then the Fourier series for f is said to be (C, 1)
summable at x. The idea is that this averaging will smooth out the partial sums,
making them more nicely behaved. It is not hard to show that if sn (f, x) converges
at some x, then n (f, x) will converge to the same thing. But there are sequences
for which n (f, x) converges and sn (f, x) does not. (See Exercises 3.19 and 4.26.)
As with sn (x), well simply write n (x) instead of n (f, x), when it is clear
which function is being considered.
We start with a calculation.

n
1 X
n (x) = sk (x)
n+1
k=0
n Z
1 X 1
= f (x t)Dk (t) dt
n+1 2
k=0
Z n
1 1 X
(*) = f (x t) Dk (t) dt
2 n+1
k=0
Z n
1 1 X sin(k + 1/2)t
= f (x t) dt
2 n+1 sin t/2
k=0
Z n
1 1 X
= f (x t) sin t/2 sin(k + 1/2)t dt
2 (n + 1) sin2 t/2 k=0

Use the identity 2 sin A sin B = cos(A B) cos(A + B).

n
1 1/2
Z X
= f (x t) (cos kt cos(k + 1)t) dt
2 (n + 1) sin2 t/2 k=0

The sum telescopes.

Z
1 1/2
= f (x t) (1 cos(n + 1)t) dt
2 (n + 1) sin2 t/2

Use the identity 2 sin2 A = 1 cos 2A.

2

sin n+1
2 t

1 1
Z
(**) = f (x t) dt
2 (n + 1) sin 2t

May 1, 2015 http://math.louisville.edu/lee/ira

10-16 CHAPTER 10. FOURIER SERIES

10

- -
2 2

Figure 8. A plot of K5 (t), K8 (t) and K10 (t).

The Fejer kernel is the sequence of functions highlighted above; i.e.,

2
sin n+1
2 t

1
(127) Kn (t) = , n N.
(n + 1) sin 2t
Comparing the lines labeled (*) and (**) in the previous calculation, we see another
form for the Fejer kernel is
n
1 X
(128) Kn (t) = Dk (t).
n+1
k=0

Once again, were confronted with a convolution integral containing a kernel:

Z
1
n (x) = f (x t)Kn (t) dt.
2

Theorem 10.13. The Fejer kernel has the following properties.5

(a) Kn (t) is an even 2-periodic function for each n N.
(b) Kn (0) = n + 1 for each n .
(c) R 0 for each n N.
Kn (t)
1
(d) 2 Kn (t) dt = 1 for each n .
(e) If 0 < < , then Kn 0 on [, ] [, ].
R R
(f) If 0 < < , then Kn (t) dt 0 and Kn (t) dt 0.
Proof. Theorem 10.5 and (128) imply (a), (b) and (d). Equation (127) implies
(c).

5Compare this theorem with Lemma 9.14.

May 1, 2015 http://math.louisville.edu/lee/ira

7. THE FEJER KERNEL 10-17

Let be as in (e). In light of (a), it suffices to prove (e) for the interval [, ].
Noting that sin t/2 is decreasing on [, ], it follows that for t ,
2
sin n+1
2 t

1
Kn (t) =
(n + 1) sin 2t
 2
1 1

(n + 1) sin 2t
1 1
0
(n + 1) sin2 2
It follows that Kn 0 on [, ] and (e) has been proved.
Theorem 9.16 and (e) imply (f). 

Theorem 10.14 (Fejer). If f : R R is 2-periodic, integrable on [, ] and

continuous at x, then n (x) f (x).
R R R
Proof. Since f is 2-periodic and f (t) dt exists, so does (f (x t)
f (x)) dt. Theorem 8.3 gives an M > 0 so |f (x t) f (x)| < M for all t.
Let > 0 and choose > 0 such that |f (x) f (y)| < /3 whenever |x y| < .
By Theorem 10.13(f), there is an N N so that whenever n N ,
Z Z
1 1
Kn (t) dt < and Kn (t) dt < .
2 3M 2 3M
We start calculating.
 Z Z 
 1 1 
|n (x) f (x)| =  f (x t)Kn (t) dt f (x)Kn (t) dt
2 2
1 
Z 

= (f (x t) f (x))Kn (t) dt
2 
Z
Z
1 

= (f (x t) f (x))Kn (t) dt + (f (x t) f (x))Kn (t) dt
2 


Z 

+ (f (x t) f (x))Kn (t) dt

   
 1 Z   1 Z 
 (f (x t) f (x))Kn (t) dt +  (f (x t) f (x))Kn (t) dt
   
 2   2 
 Z 
 1 
+  (f (x t) f (x))Kn (t) dt
2
Z Z
M 1 M
Z
< Kn (t) dt + |f (x t) f (x)|Kn (t) dt + Kn (t) dt
2 2 2
Z
1
<M + Kn (t) dt + M <
3M 3 2 3M
This shows n (x) f (x). 
Theorem 10.14 gives a partial solution to the representation problem.
Corollary 10.15. Suppose f and g are continuous and 2-periodic on R. If f
and g have the same Fourier coefficients, then they are equal.

May 1, 2015 http://math.louisville.edu/lee/ira

10-18 CHAPTER 10. FOURIER SERIES

2
2

- - 2 - - 2
2 2 2 2

- -
2 2

- -

Figure 9. These plots illustrate the functions of Example 10.4. On

the left are shown f (x), s8 (x) and 8 (x). On the right are shown f (x),
3 (x), 5 (x) and 20 (x). Compare this with Figure 4.

Proof. By assumption, n (f, t) = n (g, t) for all n and Theorem 10.14 implies
0 = n (f, t) n (g, t) f g.


In the case of continuous functions, the convergence is uniform, rather than

pointwise.
Theorem 10.16 (Fejer). If f : R R is 2-periodic and continuous, then
n (x) f (x).
Proof. By Exercise 32, f is uniformly continuous. This can be used to show
the calculation within the proof of Theorem 10.14 does not depend on x. The details
are left as Exercise 12. 

A perspicacious reader will have noticed the similarity between Theorem 10.16
and the Weierstrass approximation theorem, Theorem 9.13. In fact, the Weierstrass
approximation theorem can be proved from Theorem 10.16 using power series and
Theorem 9.24.
Example 10.4. As in Example 10.2, let f (x) = x for < x and extend
f to be periodic on R with period 2. Figure 9 shows the difference between the
Fejer and classical methods of summation. Notice that the Fejer sums remain much
more smoothly affixed to the function and do not show Gibbs phenomenon.

8. Exercises

10.1. Prove Theorem 10.2.

10.2. Suppose f : R R is 2-periodic and integrable on [, ]. If b a = 2,
Rb R
then a f = f .

10.3. Let f : [, ) be a function with Fourier coefficients an and bn . If f is odd,

then an = 0 for all n . If f is even, then bn = 0 for all n N.

10.4. If f (x) = sgn(x) on [, ), then find the Fourier series for f .

May 1, 2015 http://math.louisville.edu/lee/ira

8. EXERCISES 10-19

X
10.5. Is sin nx the Fourier series of some function?
n=1

10.6. Let f (x) = x2 when x < and extend f to be 2-periodic on all of R.

Use Dinis Test to show sn (f, x) f (x) everywhere.

2 X 1
10.7. Use Exercise 10.6 to prove = 2
.
6 n=1
n

10.8. Suppose f is integrable on [, ]. If f is even, then the Fourier series for f

has no sine terms. If f is odd, then the Fourier series for f has no cosine terms.

10.9. If f : R R is periodic with period p and continuous on [0, p], then f is

uniformly continuous.

10.10. Prove
n
X sin 2nt
cos(2k 1)t =
2 sin t
k=1
Pn
for t 6= k, k Z and n N. (Hint: 2 k=1 cos(2k 1)t = D2n (t) Dn (2t).)

10.11. The function g(t) = t/ sin(t/2) is undefined whenever t = 2n for some

n Z. Show that it can be redefined on the set {2n : n Z} to be periodic and
uniformly continuous on R.

10.12. Prove Theorem 10.16.

10.13. Prove the Weierstrass approximation theorem using Fourier series and
Taylors theorem.

10.14. If f (x) = x for x < , can the Fourier series of f be integrated

term-by-term on (, ) to obtain the correct answer?

10.15. If f (x) = x for x < , can the Fourier series of f be differentiated

term-by-term on (, ) to obtain the correct answer?

May 1, 2015 http://math.louisville.edu/lee/ira

 Bibliography

[1] A.D. Aczel. The mystery of the Aleph: mathematics, the Kabbalah, and the search for infinity.
Washington Square Press, 2001.
[2] Carl B. Boyer. The History of the Calculus and Its Conceptual Development. Dover Publica-
tions, Inc., 1959.
[3] James Ward Brown and Ruel V Churchill. Fourier series and boundary value problems.
McGraw-Hill, New York, 8th ed edition, 2012.
[4] Andrew Bruckner. Differentiation of Real Functions, volume 5 of CRM Monograph Series.
American Mathematical Society, 1994.
[5] Georg Cantor. Uber eine Eigenschaft des Inbegriffes aller reelen algebraishen Zahlen. Journal
fur die Reine und Angewandte Mathematik, 77:258262, 1874.
[6] Georg Cantor. De la puissance des ensembles parfait de points. Acta Math., 4:381392, 1884.
[7] Lennart Carleson. On convergence and growth of partial sums of Fourier series. Acta Math.,
116:135157, 1966.
[8] Krzysztof Ciesielski. Set Theory for the Working Mathematician. Number 39 in London
Mathematical Society Student Texts. Cambridge University Press, 1998.
[9] Gerald A. Edgar, editor. Classics on Fractals. Addison-Wesley, 1993.
[10] James Foran. Fundamentals of Real Analysis. Marcel-Dekker, 1991.
[11] Nathan Jacobson. Basic Algebra I. W. H. Freeman and Company, 1974.
[12] Yitzhak Katznelson. An Introduction to Harmonic Analysis. Cambridge University Press,
Cambridge, UK, 3rd ed edition, 2004.
[13] Charles C. Mumma. n! and The Root Test. Amer. Math. Monthly, 93(7):561, August-
September 1986.
[14] James R. Munkres. Topology; a first course. Prentice-Hall, Englewood Cliffs, N.J., 1975.
[15] Paul J Nahin. Dr. Eulers Fabulous Formula: Cures Many Mathematical Ills. Princeton
University Press, Princeton, NJ, 2006.
[16] Paul du Bois Reymond. Ueber die Fourierschen Reihen. Nachr. Kon. Ges. Wiss. Gttingen,
21:571582, 1873.
[17] Hans Samelson. More on Kummers test. The American Mathematical Monthly, 102(9):817
818, 1995.
[18] Jingcheng Tong. Kummers test gives characterizations for convergence or divergence of all
positive series. The American Mathematical Monthly, 101(5):450452, 1994.
[19] Karl Weierstrass. Uber continuirliche Functionen eines reelen Arguments, di fur keinen Werth
des Letzteren einen bestimmten Differentialquotienten besitzen, volume 2 of Mathematische
werke von Karl Weierstrass, pages 7174. Mayer & Muller, Berlin, 1895.
[20] Antoni Zygmund. Trigonometric Series. Cambridge University Press, Cambridge, UK, 3rd ed.
edition, 2002.

A-1
 Index

A
P
n=1 , Abel sum, 9-21 , symmetric difference, 1-3
0 , cardinality of N, 1-10 , product (Cartesian or real), 1-5, 2-1
S c , complement of S, 1-3 T , trigonometric polynomials, 10-1
c, cardinality of R, 2-10 uniform convergence, 9-4
D (.) Darboux integral, 8-5 , union, 1-2
D (.) . lower Darboux integral, 8-5 Z, integers, 1-2
D (., .) lower Darboux sum, 8-4 Abels test, 4-10
D (.) . upper Darboux integral, 8-5 absolute value, 2-4
D (., .) upper Darboux sum, 8-4 almost every, 5-9
\, set difference, 1-3 alternating harmonic series, 4-10
Dn (t) Dirichlet kernel, 10-5 Alternating Series Test, 4-12
Dn (t) conjugate Dirchlet kernel, 10-11 and , 1-2
, element, 1-1 Archimedean Principle, 2-7
6, not an element, 1-1 axioms of R
, empty set, 1-2 additive inverse, 2-1
, logically equivalent, 1-4 associative laws, 2-1
Kn (t) Fejer kernel, 10-16 commutative laws, 2-1
F , F sigma set, 5-11 completeness, 2-7
B A , all functions f : A B, 1-13 distributive law, 2-1
G , G delta set, 5-11 identities, 2-1
glb , greatest lower bound, 2-6 multiplicative inverse, 2-1
order, 2-3
iff, if and only if, 1-4
= , implies, 1-4
Baire category theorem, 5-10
<, , >, , 2-3
Baire, Rene-Louis, 5-10
, infinity, 2-6
Bertrands test, 4-9
, intersection, 1-3 Bolzano-Weierstrass Theorem, 3-7, 5-3
, logical and, 1-2 bound
, logical or, 1-2 lower, 2-5
lub , least upper bound, 2-6 upper, 2-5
n, initial segment, 1-10 bounded, 2-5
N, natural numbers, 1-2 above, 2-5
, nonnegative integers, 1-2 below, 2-5
part ([a, b]) partitions of [a, b], 8-1
pointwise convergence, 9-1 Cantor, Georg, 1-11
P(A), power set, 1-2 diagonal argument, 2-9
, indexed product, 1-5 middle-thirds set, 5-11
R, real numbers, 2-7 cardinality, 1-10
R (.) Riemann integral, 8-3 countably infinite, 1-10
R (., ., .) Riemann sum, 8-2 finite, 1-10
, subset, 1-1 uncountably infinite, 1-11
(, proper subset, 1-1 Cartesian product, 1-5
, superset, 1-1 Cauchy
), proper superset, 1-1 condensation test, 4-4

A-2
Index A-3

continuous, 6-13 kernel, 10-16

criterion, 4-3 theorem, 10-17
Mean Value Theorem, 7-6 field, 2-1
sequence, 3-9 complete ordered, 2-7
Cauchy-Schwarz Inequality, 8-21 field axioms, 2-1
ceiling function, 6-8 finite cover, 5-6
clopen set, 5-2 floor function, 6-8
closed set, 5-1 Fourier series, 10-2
closure of a set, 5-3 full measure, 5-9
Cohen, Paul, 1-12 function, 1-6
compact, 5-7 bijective, 1-7
equivalences, 5-7 composition, 1-6
comparison test, 4-4 constant, 1-6
completeness, 2-5 differentiable, 7-6
composition, 1-6 even, 7-15
connected set, 5-4 image of a set, 1-7
continuous, 6-5 injective, 1-7
Cauchy, 6-13 inverse, 1-8
left, 6-8 inverse image of a set, 1-7
right, 6-8 odd, 7-15
uniformly, 6-12 one-to-one, 1-7
continuum hypothesis, 1-11, 2-10 onto, 1-6
convergence salt and pepper, 6-7
pointwise, 9-1 surjective, 1-6
uniform, 9-4 Fundamental Theorem of Calculus, 8-13,
convolution kernel, 9-9 8-14
critical number, 7-5
critical point, 7-5 geometric sequence, 3-1
Gibbs phenomenon, 10-9
Darboux Godel, Kurt, 1-11
integral, 8-5 greatest lower bound, 2-6
lower integral, 8-5
lower sum, 8-4 Heine-Borel Theorem, 5-7
Theorem, 7-8 Hilbert, David, 1-11
upper integral, 8-5
upper sum, 8-4 indexed collection of sets, 1-4
DeMorgans Laws, 1-4 indexing set, 1-4
dense set, 2-8, 5-9 infinity , 2-6
irrational numbers, 2-8 initial segment, 1-10
rational numbers, 2-8 integers, 1-2
derivative, 7-1 integral
chain rule, 7-3 Cauchy criterion, 8-9
rational function, 7-3 change of variables, 8-17
derived set, 5-2 integration by parts, 8-14
diagonal argument, 2-9 intervals, 2-4
differentiable function, 7-6 irrational numbers, 2-8
Dini isolated point, 5-2
Test, 10-7
Theorem, 9-4 Kummers test, 4-8
Dirac sequence, 9-9
LHospitals Rule
Dirichlet
easy, 7-11
conjugate kernel, 10-11
hard, 7-12
function, 6-7
least upper bound, 2-6
kernel, 10-5
left-hand limit, 6-4
disconnected set, 5-4
limit
extreme point, 7-5 left-hand, 6-4
right-hand, 6-4
Fejer unilateral, 6-4

May 1, 2015 http://math.louisville.edu/lee/ira

A-4 Index

limit comparison test, 4-5 symmetric, 1-6

limit point, 5-2 transitive, 1-6
limit point compact, 5-7 relative maximum, 7-5
Lindelof property, 5-6 relative minimum, 7-5
relative topology, 5-4
Maclaurin series, 9-19 relatively closed set, 5-4
meager, 5-11 relatively open set, 5-4
Mean Value Theorem, 7-6 Riemann
metric, 2-5 integral, 8-3
discrete, 2-5 Rearrangement Theorem, 4-14
space, 2-4 sum, 8-2
standard, 2-5 Riemann-Lebesgue Lemma, 10-3
right-hand limit, 6-4
n-tuple, 1-5
Rolles Theorem, 7-6
natural numbers, 1-2
root test, 4-6
Nested Interval Theorem, 3-9
nested sets, 3-9, 5-3 salt and pepper function, 6-7, 8-16
nowhere dense, 5-10 Sandwich Theorem, 3-4
Schroder-Bernstein Theorem, 1-9
open cover, 5-5
sequence, 3-1
finite, 5-6
accumulation point, 3-7
open set, 5-1
bounded, 3-1
or , 1-2
bounded above, 3-1
order isomorphism, 2-7
bounded below, 3-1
ordered field, 2-3
Cauchy, 3-9
ordered pair, 1-5
contractive, 3-10
ordered triple, 1-5
convergent, 3-2
partition, 8-1 decreasing, 3-4
common refinement, 8-1 divergent, 3-2
generic, 8-1 Fibonacci, 3-1
norm, 8-1 functions, 9-1
refinement, 8-1 geometric, 3-1
selection, 8-2 increasing, 3-4
Peano axioms, 2-1 lim inf, 3-8
perfect set, 5-11 limit, 3-2
portion of a set, 5-10 lim sup, 3-8
power series, 9-16 monotone, 3-4
analytic function, 9-19 recursive, 3-1
center, 9-16 subsequence, 3-6
domain, 9-16 sequentially compact, 5-7
geometric, 9-17 series, 4-1
interval of convergence, 9-17 Abels test, 4-10
Maclaurin, 9-19 absolutely convergent, 4-10
radius of convergence, 9-17 alternating, 4-12
Taylor, 9-19 alternating harmonic, 4-10
power set, 1-2 Alternating Series Test, 4-12
Bertrands test, 4-9
Raabes test, 4-9 Cauchy Criterion, 4-3
ratio test, 4-7 Cauchys condensation test, 4-4
rational function, 6-8 Cesaro summability, 4-17
rational numbers, 2-2, 2-8 comparison test, 4-4
real numbers, R, 2-7 conditionally convergent, 4-10
relation, 1-6 convergent, 4-1
domain, 1-6 divergent, 4-1
equivalence, 1-6 Fourier, 10-2
function, 1-6 geometric, 4-1
range, 1-6 Gregorys, 9-21
reflexive, 1-6 harmonic, 4-2

May 1, 2015 http://math.louisville.edu/lee/ira

Index A-5

Kummers test, 4-8 unilateral limit, 6-4

limit comparison test, 4-5
p-series, 4-5 Weierstrass
partial sums, 4-1 Approximation Theorem, 9-8, 10-18
positive, 4-4 M-Test, 9-7
Raabes test, 4-9
ratio test, 4-7
rearrangement, 4-13
root test, 4-6
subseries, 4-17
summation by parts, 4-11
telescoping, 4-3
terms, 4-1
set, 1-1
clopen, 5-2
closed, 5-1
compact, 5-7
complement, 1-3
complementation, 1-3
dense, 5-9
difference, 1-3
element, 1-1
empty set, 1-2
equality, 1-1
F , 5-11
G , 5-11
intersection, 1-2
meager, 5-11
nowhere dense, 5-10
open, 5-1
perfect, 5-11
proper subset, 1-1
subset, 1-1
symmetric difference, 1-3
union, 1-2
square wave, 10-9
subcover, 5-5
subspace topology, 5-4
summation
Abel, 9-21
by parts, 4-11
Cesaro, 4-17

Taylor series, 9-19

Taylors Theorem, 7-9
integral remainder, 8-14
topology, 5-2
finite complement, 5-13
relative, 5-13
right ray, 5-2, 5-13
standard, 5-2
totally disconnected set, 5-4
trigonometric polynomial, 10-1

unbounded, 2-6
uniform continuity, 6-12
uniform metric, 9-6
Cauchy sequence, 9-6
complete, 9-6

May 1, 2015 http://math.louisville.edu/lee/ira