Chapter IV · Limits of functions of a positive integral variable
Monotone sequences
An increasing sequence either tends to a limit or to plus infinity, and it tends to a limit exactly when it is bounded above; this proves convergence without knowing the limit.
Why this matters
Every limit proved so far was proved by guessing the limit first and then checking the definition, which mentions . But often we cannot guess. Does tend to a limit? If so, what is it? §§69–70 gives the first tool that answers the first question without the second: an increasing sequence that is bounded above must converge. Its proof is where the completeness of the real numbers (Dedekind’s theorem of §17) enters the theory of limits, and it is behind , and much of Chapter VIII.
The idea
An increasing sequence only ever moves up (or stays put). So for each number only two things can happen: either some term reaches , and then every later term stays at or above it; or no term ever reaches . Call the numbers of the first kind and those of the second kind . Everything in lies below everything in .
- If is empty, every number is eventually passed for good: the sequence tends to .
- If is not empty, and split the real line into a lower and an upper part, a Dedekind section. The section has a dividing number , and the terms must crowd up against it.
Try this. Start with . Move up slowly and find (roughly) where “reached” turns into “never reached”. Then switch to the harmonic sum and push to , , .
- ξ (move with the slider)
- terms ≥ ξ
What to notice
For a bounded increasing sequence, the change happens at one number ; only afterwards does the toggle reveal it. You found without a formula, which is the point of the theorem. The harmonic sum reaches every you try, but slowly: is first reached at . No picture of finitely many terms could tell you that it is unbounded; that needs a proof (<Sec n={77} />).
Precise statements
increases steadily (is an increasing function of ) if for every . It is strictly increasing if for every . Decreasing and strictly decreasing are defined with the inequalities reversed.
Equal consecutive values are allowed: takes the values and increases steadily. Since increases exactly when decreases, every result for increasing sequences gives one for decreasing sequences by a change of sign.
If increases steadily, then either tends to a limit or .
So an increasing sequence never oscillates: of the five possible behaviours, only two remain.
An increasing sequence tends to a limit if there is a number with for every , and tends to if there is no such .
If increases steadily and for every , then tends to a limit .
If increases steadily and tends to a limit , then for every .
A decreasing sequence tends to a limit or to ; it tends to a limit exactly when it is bounded below, and then every term is at least the limit.
Proofs
Sort every real number into (reached by some term) or (never reached). If is empty the sequence passes every bound; otherwise is a section of the reals, and the number it defines is the limit.
Define the classes
Put in if for some , and in if for every . Each is in exactly one class. Because the sequence increases, if then for every .
If is empty:
Then every is in (and so is ): some , and then for all . That is Definition II.
If is not empty: a section
is not empty either (). Every member of is less than every member of : if and , then for some while . So is a section of the real numbers, and by Dedekind’s theorem there is a number such that every number below is in and every number above is in .
The terms are trapped just below
Let . Since , for every ; as is arbitrary, for every . Since , some , and then for all .
Conclude
For : , so . Since was arbitrary, .
The corollaries are now short. Cor. 1: if for all then , so is not empty and there is a limit; if no such exists then is empty, so . Cor. 3: step 4 showed for every . Cor. 2: and every member of is at least (anything below is in ), so .
Cor. 2 says , not . The sequence is below for every , and its limit is exactly . A strict inequality between the terms and a bound becomes non-strict in the limit; this is a pattern you will see many times. In fact, Hardy notes, the terms of an increasing sequence are all strictly below the limit unless the sequence is eventually constant: if then for all .
Why this matters: convergence without the limit
§70 spells out the gain. The definition of a limit is a test for a given : it can confirm , or (by contradiction) show that no works, as for . It cannot tell us that some unknown exists. The monotone theorem can. Hardy’s motivating example is : it is not at all obvious what its limit should be, but in The limit of we show it is increasing and below , and its limit becomes the definition of the number .
Worked examples and exercises
Let . Show that tends to a limit, and that the limit is at most .
Hint
Solution
, so the sequence increases. For ,
because the sum telescopes; and . By Cor. 2 the sequence tends to a limit . (Euler showed that , but the theorem needs no such knowledge.)
Let and . Show that increases and is less than , and find its limit.
Hint
Solution
Bounded: , and if then . Increasing: , since ; both are positive, so . By Cor. 2, with . Now . The left side tends to (a shift of , then Theorem II), the right to . So , that is , and since , .
An increasing sequence satisfies for every . Which of these must be true? (a) tends to a limit; (b) the limit is less than ; (c) the limit is at most ; (d) every term is less than the limit.
Solution
(a) Yes, by Cor. 1. (b) No: has limit . (c) Yes, by Cor. 2. (d) Not necessarily: a sequence that is constant from some point on (such as ) equals its limit there. Hardy’s remark shows that (d) fails only for such sequences; Cor. 3 guarantees in every case.
Prove the theorem on decreasing sequences from the theorem on increasing ones.
Solution
If decreases, increases. So tends to a limit or to , that is, tends to or to . is bounded above by exactly when is bounded below by , which gives the criterion. And for every (Cor. 3) means .
Hardy remarks that the definition can still show indirectly that no limit exists. Show this for , and explain why the same idea cannot prove that has a limit.
Solution
If , then the even terms show and the odd terms show (each subsequence of a convergent sequence has the same limit, or argue with as in Oscillating sequences). Since there is no limit. For the sum of we cannot use the definition directly, because we do not know which to test; a proof by contradiction can rule out candidates, but not produce one. The monotone theorem fills exactly this gap.
Check yourself
Every bounded sequence tends to a limit.
False: is bounded and oscillates. Boundedness is enough only together with monotonicity (or, as in Chapter IV’s later sections, with something else such as Cauchy’s condition).
An increasing sequence can:
That is the theorem. An increasing sequence is bounded below by , so it cannot tend to ; and the Dedekind argument shows it cannot oscillate.
(the integral part of ) increases steadily in Hardy’s sense.
True. Its values are , so always, though not strictly. It is unbounded, so it tends to .
The terms of stay below for the first million values of . What follows?
Boundedness is a statement about all . The harmonic sums are in fact unbounded and tend to (§77), although they are below for and pass only after about terms.
Hardy’s language
The great importance of these theorems lies in the fact that they give us … a means of deciding, in a great many cases, whether a given function of n does or does not tend to a limit as n → ∞, without requiring us to be able to guess or otherwise infer beforehand what the limit is.
Summary
- An increasing sequence tends to a limit or to , never anything else; it converges exactly when it is bounded above.
- The proof sorts every into “reached” () or “never reached” (); if is not empty, Dedekind’s theorem supplies the limit as the dividing number.
- If for all , the limit is , not necessarily ; every term is the limit.
- Decreasing sequences behave the same way with signs reversed.
- The theorem proves that limits exist without naming them, which is how and will be defined.