Products of primes in conjugacy class sizes and irreducible character degrees. (Q2655790)

There have been many results over the last few years which relate the structure of groups to the sizes of its conjugacy classes and to the degrees of its irreducible characters. This paper is a major contribution to this study. Before describing its result it will be helpful to introduce some definitions. Let \(X\) be a set of positive integers. Associate a graph to \(X\), the `prime vertex graph', which has vertex set \(\rho(X)=\bigcup_{x\in X}\pi(x)\) where \(\pi(x)\) denotes the prime divisors of \(X\). There is an edge between \(p,q\in\rho(X)\) if \(pq\) divides \(x\) for some \(x\in X\). There are two common choices for the set \(X\): the first is \(\text{cs}(G)\) the set of conjugacy class sizes of \(G\) and the second is \(\text{cd}(G)\) the set of degrees of irreducible characters of \(G\). A consequence of Theorem A in the paper being reviewed says that the second graph is a subgraph of the first. The structure of the paper is to examine groups \(G\) which have the property that there exist primes \(p\) and \(q\) which are vertices in \(\Gamma(\text{cs}(G))\) but are not connected. Theorem B says that in this situation (i) \(G\) is \(\{p,q\}\)-soluble, (ii) the \(\{p,q\}\) length of \(G\) is \(1\) where both Sylow \(p\)-subgroups and Sylow \(q\)-subgroups are Abelian and (iii) if \(x\) is \(\{p,q\}'\)-element of \(G\) then either a Sylow \(p\)-subgroup or a Sylow \(q\)-subgroup of \(G\) is a normal subgroup of a Hall \(\{p,q\}\)-subgroup of \(C_G(x)\). This theorem depends on the classification of finite simple groups. The athors use this to prove the main theorem, Theorem A, which says that if \(G\) is a finite group and if \(p\) and \(q\) are two primes such that \(pq\) divides the degree of some irreducible character of \(G\) then \(pq\) divides the size of some conjugacy class of \(G\). This was proved in the case when \(G\) was soluble by \textit{S. Dolfi} [J. Algebra 174, No. 3, 749-752 (1995; Zbl 0833.20010)]. The proof of Theorem A follows from Theorem B by letting \(G\) be a minimal counter-example and then using the structure of \(G\) as described in Theorem B.