Simple Groups, Permutation Groups, and Probability

In recent years probabilistic methods have proved useful in the solution of several problems concerning finite groups, mainly involving simple groups and permutation groups. In some cases the probabilistic nature of the problem is apparent from its very formulation (see [KL], [GKS], [LiSh1]); but in other cases the use of probability, or counting, is not entirely anticipated by the nature of the problem (see [LiSh2], [GSSh]). In this paper we study a variety of problems in finite simple groups and finite permutation groups using a unified method, which often involves probabilistic arguments. We obtain new bounds on the minimal degrees of primitive actions of classical groups, and prove the Cameron-Kantor conjecture that almost simple primitive groups have a base of bounded size, apart from various subset or subspace actions of alternating and classical groups. We use the minimal degree result to derive applications in two areas: the first is a substantial step towards the Guralnick-Thompson genus conjecture, that for a given genus g, only finitely many non-alternating simple groups can appear as a composition factor of a group of genus g (see below for definitions); and the second concerns random generation of classical groups. Our proofs are largely based on a technical result concerning the size of the intersection of a maximal subgroup of a classical group with a conjugacy class of elements of prime order. We now proceed to describe our results in detail.