Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

Let $\,f : \mathbb{R} \rightarrow \mathbb{R}$ be a infinitely differentiable function that is increasing and bounded. Then is it true that $\lim_{x\to \infty}f'(x)=0$?

share|cite|improve this question
up vote 17 down vote accepted

No. Let $h\colon\mathbb R\to\mathbb R$ be a smooth function with $h(x)>0$ for $x\in(0,1)$ and $h(x)=0$ otherwise (a smooth "bump"). Then let $$g(x)=\sum_{n=1}^\infty h(n^2(x-n))=\begin{cases}h(n^2(x-n))&\text{if }n\le x<n+1, n\in\mathbb N\\0&\text{otherwise}\end{cases}$$ And let $f(x)=\int_0^xg(t)\,\mathrm dt$. Then

  • $f$ is smooth because in each bounded interval only finitely summands are nonzero
  • $\lim_{x\to\infty}f'(x)=\lim_{x\to\infty}g(x)$ does not exist
  • $f$ is increasing because $f'\ge 0$
  • $f$ is bounded by $\sum_{n=1}^\infty\frac1{n^2}\int_0^1h(t)\,\mathrm dt<\infty$
share|cite|improve this answer

Hint: One way to look at this is by the Fundamental theorem of Calculus:

$$f(b) - f(a) = \int_a^b f'(x) dx\ .$$

As the function $f$ is bounded we have

$$\int_a^\infty f'(x) dx <\infty$$

Does that imply that $f'$ tends to $0$?

share|cite|improve this answer
This alone does not seem that useful in answering the question (see Hagen von Eitzen's answer). – Andrew Salmon May 10 '14 at 9:03
@AndrewSalmon: Um... What do you mean by not useful? – Arctic Char May 10 '14 at 9:05
How are you supposed to get a construction like Hagen von Eitzen's out of this? – Andrew Salmon May 10 '14 at 19:30
@AndrewSalmon: Note that Hagen von Eitzen's function is an explicit construction of a function I supposed in my answer. – Arctic Char May 10 '14 at 22:01

Yes, if the limit at $\infty$ exists; if this is the case, then $f$ will go to $\infty$. Say the limit of $f'(x)$ as $x \rightarrow \infty=M>0$.

Then repeatedly apply the mean value theorem, so that , for a choice of an integer $n$ large-enough, $$f(n+1)-f(n)= M(n+1-n)=M$$ . This can be done for $n+1, n+2,..$ , so that $$Lim _{n\rightarrow \infty }f(n)=\Sigma_{k=n}^{\infty} [ f(k+1)-f(k)] \rightarrow \infty$$

share|cite|improve this answer
What is the down-vote about? – Name May 22 '14 at 3:30

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.