Posted by Jason Polak on 29. November 2017 · Write a comment · Categories: number-theory · Tags: , ,

Lately I’ve been thinking about primes, and I’ve plotted a few graphs to illustrate some beautiful ideas involving primes. Even though you might not always associate with primes, they are always haunting quietly in the background.

Abundance of primes in an arithmetic progression

Let’s start out with the oddest prime of all: 2. Get it? But after that, all the odd primes are either of the form $4k + 1$ or $4k + 3$. For fixed $x$, are there more primes less than $x$ of the form $4k + 1$ or the second form $4k + 3$? Let’s write $\pi(4k + r,x)$ for the number of primes less than or equal to $x$ of the form $4k + r$. Here is a graph of the difference $\pi(4k+3,x) – \pi(4k+1,x)$:

Pretty neat right? It looks like this difference is wildly erratic, reaching zero after a short while with a bit of a fight and then for a really good long while the primes of the form $4k +3$ win out. So you might be tempeted to think that primes of the form $4k + 3$ become more and more abundant as $x$ increases. That would be wrong. In fact, John E. Littlewood proved that $\pi(4k +3,x)-\pi(4k + 1,x)$ switches sign infinitely often!

Of course, that must mean there are infinitely many primes of both types, and that’s true and a special case of Dirichlet’s theorem: there are infinitely many primes in any arithmetic progression $ax + b$ whenever $a$ and $b$ are relatively prime.
More »