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

The following question concerns Nagura's proof that there is a prime between x and $\frac{6}{5}x$ for x > 2103. The question does not depend on a precise definition of the functions involved. We have

$$T(x) = \sum_{m=1}^{\infty}\psi(\frac{x}{m})$$

where $\frac{x}{m}$ is ordinary division. The claim reads:

$T(x) -T(\frac{x}{2})-T(\frac{x}{3})-T(\frac{x}{5}) + T(\frac{x}{30})$

$ = \psi(x) + \psi(\frac{x}{7})+\psi(\frac{x}{11})+...+\psi(\frac{x}{29})+...(*)$

$- \psi(\frac{x}{6})-\psi(\frac{x}{10})-...-\psi(\frac{x}{30})...(*) \leq\psi(x)......... (1) $

(*) is the author's note that the denominators in the remaining terms repeat under congruence mod 30.

The claim may well be true as stated, but I wonder if it shouldn't read $2 T(\frac{x}{30})$. Otherwise, if we do the bookkeeping,

$T(x) = \psi(x)+\psi(\frac{x}{2})+\psi(\frac{x}{3})... $


$-T(\frac{x}{3})=\psi(\frac{x}{3})+\psi(\frac{x}{6})+\psi(\frac{x}{9})... $

$-T(\frac{x}{5})=\psi(\frac{x}{5})+\psi(\frac{x}{10})+\psi(\frac{x}{15})... $

$+T(\frac{x}{30})=\psi(\frac{x}{30})+\psi(\frac{x}{60})+\psi(\frac{x}{90})... $

Each of the terms in the last line occur in all three of the previous (negative) sequences. For example, $\psi(\frac{x}{30})$ will occur in $T(\frac{x}{2}),T(\frac{x}{2}),T(\frac{x}{2})$. As written, it appears we should have $-2\psi(\frac{x}{30})$ in (1).

So is this likely a typo, or is the equation true as written? It seems to me the argument depends on keeping the denominators in the negative terms of (1) smaller than those in the positive terms.

I tried to preserve the question while keeping details to a minimum. More details on request, thanks for any suggestions.

share|cite|improve this question
What is T(x)? I haven't heard about a number theory function that is often called T(x). – Mayank Pandey Mar 5 '14 at 23:45
@MayankPandey: I haven't looked at this in 2 years but T(x) is defined in the first formula: a sum of $\psi$ functions. It occurs in Nagura's 1952 paper. – daniel Mar 5 '14 at 23:58
Oh. I thought T(x) was something different that happened to satisfy the first statement. – Mayank Pandey Mar 7 '14 at 0:12
up vote 3 down vote accepted

Perhaps I'm missing something obvious here, but doesn't the $\psi(\frac{x}{30})$ term appear once in each of $T(x),T(\frac{x}{2}),T(\frac{x}{3}),T(\frac{x}{5}),T(\frac{x}{30})$ meaning that it is added twice and subtracted three times in (1) giving the equation as written by the author?

share|cite|improve this answer
No, you are right! I forgot it would occur in the first line. Thanks. – daniel Dec 24 '11 at 4:19

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.