Quadratic pairs. (Q1879632)

This paper is a successor of [J. Algebra 258, No. 2, 442-476 (2002; Zbl 1024.20003)] due to the same author. Here he considers the case of a group having a faithful quadratic module over \(\text{GF}(p)\), \(p\) odd, and possessing a component. A module \(V\) over \(\text{GF}(p)\) is said to be quadratic if \([V,A,A]=0\) for some \(p\)-subgroup \(A\) of \(G\) with \([V,A]\neq 0\) and \(G=\langle A^G\rangle\). After some reduction the author considers the case of \(G=HA\), where \(H\) is a quasisimple normal subgroup. He then proves that either \(H\) is a group of Lie type in characteristic \(p\) or \(p=3\) and \(|A|=3\) and \(G\cong\text{PSU}(n,2)\), \(n\geq 5\), \(|Z(G)|=2\) and \(G/Z(G)\cong A_n\), \(n=5\), \(n\geq 7\), \(D_4(2)\), \(G_2(4)\), \(\text{Sp}(6,2)\), \(Co_1\), \(Suz\) or \(J_2\). The investigation of quadratic modules started with the unpublished work of \textit{J. G. Thompson} for \(p\geq 5\) [Actes Congr. internat. Math. 1970, 1, 375-376 (1971; Zbl 0236.20024)]. Then \textit{C.-Y. Ho} worked on modules for \(p=3\) [J. Algebra 43, 338-358 (1976; Zbl 0385.20006)]. He never came down with a final classification. In contrast to those papers the author uses the classification of the finite simple groups. Hence he has not to determine \(H\) by the fact allowing a quadratic representation. So he just has to find out, which still is not trivial, which quasisimple groups do posses a quadratic module. In abstract group theory quadratic modules always show up if two elementary Abelian subgroups of some group normalize each other. The paper is an important contribution to group theory and will find its place in the revision project of the finite simple groups.