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

I am trying to prove that the sequence $I_n(p)=\int_1^n \frac{dx}{x^p}$ does or does not converge uniformly to the function $I(p)=\int_1^\infty \frac{dx}{x^p}$ for $p>1$.

It seems to me that $\sup |I_n(p)-I(p)|= \lim_{p \to 1} (I(p)-I_n(p))$. Is it enough, then, to show that for all $\epsilon>0$ there exists a natural number $N$ such that for $p\to 1^+$ and all $n\geq N$, $I(p)-I_n(p) < \epsilon$? Can I set $p = 1$?

$I(p)-I_n(p) = \int_1^\infty \frac{dx}{x^p} - \int_1^n \frac{dx}{x^p} = \int_n^\infty \frac{dx}{x^p}$.

$\lim_{n\to\infty} \lim_{p\to 1} \int_n^\infty \frac{dx}{x^p} = \lim_{n\to\infty} \int_n^\infty \frac{dx}{x} = 0$

Is this correct?

share|cite|improve this question
Do you mean $I_n(p)=\int_1^n\frac{dx}{x^p}$? I ask because with your definition $I_n(p)\to 0$ as $n\to\infty$ for all $p\gt1$. In any case, you can explicitly compute $I(p)-I_n(p)$. – Jonas Meyer Feb 13 '11 at 21:11
It does not, $I_n \to 0$ and $I \neq 0$. – Jonas Teuwen Feb 13 '11 at 22:45
@cardinal: (You may not be alerted of this, but maybe you'll read it anyway.) I rejected an edit of yours because in the preview it looked incorrect, but I think that I was mistaken and that it was the same edit Arturo made. Sorry about that. (On the other hand, the problem itself seems to have an error that still hasn't been clarified.) – Jonas Meyer Feb 13 '11 at 23:56
Jonas, you're right. I meant $I_n(p) = \int_1^n \frac{dx}{x^p}$. – please delete me Feb 14 '11 at 8:01
up vote 1 down vote accepted

The sequence doesn't converge uniformly -- $\lim_{p \to 1} (I(p)-I_n(p))$ doesn't exist, since


which diverges for $p\to 1^+$. That is, for given $n$ the difference is unbounded in $p$, and thus you can't find an $n_0$ such that the difference is bounded by $\epsilon$ for all $n>n_0$ and all $p$. This is because the integral diverges for $p=1$, and no matter how large you choose $n$ you're missing larger and larger contributions from the logarithmic tail as you take $p$ to $1$.

share|cite|improve this answer

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.