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 was reading de Weger's paper on bounding the cardinality of the Tate-Shafarevich group and in lemma 1 (pg 111), he claims that for any $n \in \mathbb{N}$, we have that $$ c(n) << N^{((log \; 3 / 3)(1+\epsilon))/ log \; log \; n)}. $$ where $c(n)$ is the product of the exponents of the prime decomposition of $n$.

This is confusing however as in page 108, he defines $N$ as the conductor of an elliptic curve over $\mathbb{Q}$, and he doesn't seem to redefine it anywhere else.

Following his proof I computed that $$ c(n) < n^\delta exp\left(-\dfrac{log \; 3}{3}+\delta \; log \; n\right) $$ where $$ \delta = \dfrac{\frac{log \; 3}{3}(1 + \frac{\epsilon}{2})}{log \; log \; n}. $$

So my question is if this is indeed a typo or is there some connection here between $c(n)$ and $N$ that I am indeed missing here?

Also it is indeed a typo and if $$ c(n) < n^\delta exp\left(-\dfrac{log \; 3}{3}+\delta \; log \; n\right) $$ then how does the result follow for $n$ sufficiently large?

share|cite|improve this question
up vote 2 down vote accepted

Looking at the proof, it's clearly a typo. And your estimate for $c(n)$ doesn't seem to be what he writes. Simply take his estimate for $c(n)$ (which is $\log n$ times a bunch of stuff) and exponentiate --- you get $n$ raised to the power of a bunch of stuff, which is exactly what he says it should be.

share|cite|improve this answer
I continued from the point he said that $$ log \; c(n) - \delta \; log \; n < (log \; 3 / 3)(\epsilon/2) (log \; n/log \; log \; n) $$ to derive the bound I mentioned above. Maybe I'm missing something. – Eugene May 20 '12 at 7:31
@Eugene: Dear Eugene, Rewrite this as $\log c(n) < (\log n)(\delta + (\log 3 /3)(\epsilon/2)/\log\log n)$ and exponentiate, to get $c(n) < n^{\delta + (\log3 / 3)(\epsilon/2)/\log\log n}$ and substitute in the given value of $\delta$, and I think everything should be clear. Perhaps I'm missing something? Cheers, – Matt E May 20 '12 at 12:25
Thanks! Your answer helped clear things up a lot. – Eugene May 20 '12 at 13:05

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.