It is well known that the gap between consecutive primes is unbounded. Is this still true for a chain of consecutive primes ?

More Formally : Is the following statement true for all natural numbers m and n ?

There are m consecutive primes $a_1,...,a_m$ , such that all the gaps are greater than n (this means $a_{k+1}-a_k>n$ for all k with 1 <= k <= m-1) ?

I also heard about primes in arithmetic progressions, but I always wondered if the primes must be consecutive in such progressions.

Can any of the known properties of the prime-numbers help to answer this question ?


Yes, this is known. It feels a bit like squashing a fly with a B-52, but there is a beautiful theorem of Shiu (Strings of Congruent Primes, 2000) which seems to cover everything you want. A direct consequence is that for any $n,m$ the following is true:

There are $m$ consecutive primes $a_1,\ldots,a_m$ that are all congruent to $1$ mod $n$.

So not only are the gaps between these primes at least $n$, but they all lie within a prescribed arithmetic progression. (It's important not to confuse this with $a_1,\ldots,a_m$ forming an arithmetic progression: a result of such strength is not yet known.)

EDIT: Another approach is the following: for any fixed $n$ let $P_n$ be the set of primes whose gap to the next prime is $\le n$. It is known from sieve theory that the number of primes in $P_n$ up to $x$ is $O(x/\log^2 x)$, with the constant certainly dependent on $n$.

Since this set has asymptotic density zero relative to the sequence of primes, there must be arbitrarily long clusters of primes that do not lie in $P_n$ (if there were no clusters of length $m$ then the lower relative density of $P_n$ would be at least $1/m$).

| cite | improve this answer | |
  • $\begingroup$ Erick, do you have a reference for the theorem of Shiu ? Thank you in advance. $\endgroup$ – Rodney Coleman Jan 5 '14 at 19:59
  • 2
    $\begingroup$ @RodneyColeman See D. K. L. Shiu, "Strings of Congruent Primes", J. London Math. Soc. (2000) 61 (2): 359--373. jlms.oxfordjournals.org/content/61/2/359.abstract $\endgroup$ – Erick Wong Jan 5 '14 at 20:09

Your Answer

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

Not the answer you're looking for? Browse other questions tagged or ask your own question.