I am studying Zagier's paper titled Newman's Short Proof of the Prime Number Theorem. In the paper, proof ($\textbf{VI}$), we are trying to show that $\theta(x) \sim x$, where $\theta(x) = \sum_{p \leq x} ln(p)$. If I understand correctly, to do this, we take that fact that $\theta(x) = O(x)$ (as shown in proof ($\textbf{III}$)), meaning that $\lim_{x \to \infty} \theta(x) = c \cdot x$, and we then 'squeeze' $c$ from both directions, ultimately showing that $c = 1$.

My main confusion comes from line 3-4 of the proof, where the author shows that for $\lambda > 1$, and arbitrarily large $x$ so that $\theta(x) \geq \lambda x$, $$\int_x^{\lambda x} \frac{\theta(t) -t}{t^2}dt > 0, \ \ \ \ (1)$$ and then asserts that this is a contradiction for such $x$ by proof ($\textbf{V}$), which simply states that $\int_1^{\infty} \frac{\theta(x) - x}{x^2}$ is a convergent integral. I'm confused as to how (1) contradicts $(\textbf{V})$.

After poking around the internet, I believe that the contradiction not only has to do with ( V ), but it may also have to do with Newmans Tauberian Theorem, which the author refers to as the Analytic Theorem . With that being said, the author does not explicitly say he is using Newmans Tauberian Theorem, so I am not entirely sure.

Could anyone explain to me why (1) contradicts V ? Or, if not that, perhaps confirm or deny that we are using the Tauberian Theorem here? Anything to help point me in the right direction is greatly appreciated!


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.