LESSON II

Monotone Sequences

There are many situations in which it is important to know whether a sequence converges, but
the value of the limit is not relevant to the problem at hand. In this section we will study several
techniques that can be used to determine whether a sequence converges.

 Terminology
 Testing for Monotonicity
 Properties that Hold Eventually
 An Intuitive View of Convergence
 Convergence of Monotone Sequences
 The Completeness Axiom

Lecture Notes:

TERMINOLOGY
We begin with some terminology.

Some examples are given in Table 1 and their corresponding graphs are shown in
Figure 1. The first and second sequences in Table 1 are strictly monotone; the third and fourth
sequences are monotone but not strictly monotone; and the fifth sequence is neither strictly
monotone.
Table 1
TESTING FOR MONOTONICITY

Frequently, one can guess whether a sequence is monotone or strictly monotone by writing out
some of the initial terms. However, to be certain that the guess is correct, one must give a precise
mathematical argument. Table 2 provides two ways of doing this, one based

Table 1
on differences of successive terms and the other on ratios of successive terms. It is assumed in
the latter case that the terms are positive. One must show that the specified conditions hold for
all pairs of successive terms.
EXAMPLE 1 Use differences of successive terms to show that

(Figure 2) is a strictly increasing sequence.

Solution. The pattern of the initial terms suggests that the

We can obtain 𝑎𝑛+1 by replacing n by 𝑛 + 1 in this formula. This

Thus, for 𝑛 ≥ 1
Table 2

which proves that the sequence is strictly increasing.

Forming the ratio of successive terms, we obtain

from which we see that an+1/an > 1 for n ≥ 1. This proves that the sequence is strictly increasing.

The following example illustrates still a third technique for determining whether a sequence is
strictly monotone.
EXAMPLE 3 In Examples 1 and 2 we proved that the sequence

is strictly increasing by considering the difference and ratio

Table 3

so that the nth term in the given sequence is an = f(n). The function f is increasing for x ≥ 1 since
Thus,

which proves that the given sequence is strictly increasing.

In general, if f(n) = an is the nth term of a sequence, and if f is differentiable for x ≥ 1,

PROPERTIES THAT HOLD EVENTUALLY

is strictly increasing from the fifth term on, but the sequence as a whole cannot be classified as
strictly increasing because of the erratic behavior of the first four terms. To describe such
sequences, we introduce the following terminology.

For example, although we cannot say that sequence (2) is strictly increasing, we can say that it
is eventually strictly increasing.

EXAMPLE 4 Show that the sequence is eventually strictly decreasing.

Solution. We have

so

From (3), an+1/an < 1 for all n ≥ 10, so the sequence is eventually strictly decreasing, as confirmed
by the graph in Figure 3.
AN INTUITIVE VIEW OF CONVERGENCE
Informally stated, the convergence or divergence of a sequence does not depend on the behavior
of its initial terms, but rather on how the terms behave eventually. For example, the sequence

eventually behaves like the sequence

and hence has a limit of 0.

CONVERGENCE OF MONOTONE SEQUENCES

2
 Theorems 1 and 2 are examples of existence theorems; they tell us whether a limit
exists, but they do not provide a method for finding it.

EXAMPLE 5 Show that the sequence converges and find its limit.

Solution. We showed in Example 4 that the sequence is eventually strictly decreasing. Since
all terms in the sequence are positive, it is bounded below by M = 0, and hence Theorem 2
guarantees that it converges to a nonnegative limit L. However, the limit is not evident directly
from the formula 10n/n! for the nth term, so we will need some ingenuity to obtain it.
It follows from Formula (3) of Example 4 that successive terms in the given sequence are
related by the recursion formula

where an = 10n/n!. We will take the limit as n→+∞ of both sides of (4) and use the fact that

We obtain

so that

In the exercises we will show that the technique illustrated in the last example can be adapted
to obtain

for any real value of x. This result will be useful in our later work.

THE COMPLETENESS AXIOM

In this text we have accepted the familiar properties of real numbers without proof, and
indeed, we have not even attempted to define the term real number. Although this is sufficient
for many purposes, it was recognized by the late nineteenth century that the study of limits and
functions in calculus requires a precise axiomatic formulation of the real numbers analogous to
the axiomatic development of Euclidean geometry. Although we will not attempt to pursue this
development, we will need to discuss one of the axioms about real numbers in order to prove
Theorems 1 and 2. But first we will introduce some terminology.
If S is a nonempty set of real numbers, then we call u an upper bound for S if u is
greater than or equal to every number in S, and we call l a lower bound for S if l is smaller than
or equal to every number in S. For example, if S is the set of numbers in the interval (1, 3), then
u = 10, 4, 3.2, and 3 are upper bounds for S and l = −10, 0, 0.5, and 1 are lower bounds for S.
Observe also that u = 3 is the smallest of all upper bounds and l = 1 is the largest of all lower
bounds. The existence of a smallest upper bound and a largest lower bound for S is not accidental;
it is a consequence of the following axiom.

PROOF OF THEOREM 1
(a) We will prove the result for increasing sequences, and leave it for the reader to adapt the
argument to sequences that are eventually increasing. Assume there exists a number M such that
an ≤ M for n = 1, 2, . . . . Then M is an upper bound for the set of terms in the sequence. By the
Completeness Axiom there is a least upper bound for the terms; call it L. Now let Є be any positive
number. Since L is the least upper bound for the terms, L − Є is not an upper bound for the
terms, which means that there is at least one term aN such that

Moreover, since {an} is an increasing sequence, we must have

when n ≥ N. But an cannot exceed L since L is an upper bound for the terms. This observation
together with (6) tells us that L ≥ an > L− Є for n ≥ N, so all terms from the Nth on are within Є
units of L. This is exactly the requirement to have

Finally, L ≤ M since M is an upper bound for the terms and L is the least upper bound. This proves
part (a).

(b) If there is no number M such that an ≤ M for n = 1, 2, . . . , then no matter how large we
choose M, there is a term aN such that

and, since the sequence is increasing

when n ≥ N. Thus, the terms in the sequence become arbitrarily large as n increases.
That is,
We omit the proof of Theorem 2 since it is similar to that of 1.