A lower bound for the number of conjugacy classes of a finite group. (Q908091): Difference between revisions

The main result of this nice paper is easily stated: Every finite group whose order is divisible by a prime \(p\) has at least \(2\sqrt{p-1}\) conjugacy classes. Equality can occur only if \(\sqrt{p-1}\) is an integer and the group is a Frobenius group of order \(p\sqrt{p-1}\). By earlier works (starting in 2000) of Külshammer, Héthelyi, Malle, the reviewer and the author the result was known to be true with possibly finitely many non-solvable \(p\)-solvable exceptions, but the number of possible exceptions was not controlled. Thanks to the paper under review, we now know that there are no exceptions, as had, of course, been conjectured all along. The main work goes into proving the following result, to which the desired result can be quickly reduced: Theorem 2.1. Let \(V\) be an irreducible and faithful \(FG\)-module for some finite group \(G\) and finite field \(F\) of characteristic \(p\) at least 59. Suppose that \(p\) does not divide \(|G|\). Let \(n(G,V)\) be the number of orbits of \(G\) on \(V\) and \(k(G)\) the number of conjugacy classes of \(G\). Then we have \(k(G)+n(G,V)-1\geq 2\sqrt{p-1}\) with equality if and only if \(\sqrt{p-1}\) is an integer, \(|V|=|F|=p\) and \(|G|=\sqrt{p-1}\). The proof of this is long and technical, looking at many different cases separately.