Linear spaces with significant characteristic prime. (Q852442)

Given a finite linear space with \(v\) points such that each line has exactly \(k\) points it is known that the number of lines, \(b\), is equal to \(v(v-1)/(k(k-1))\). If \(b=v\) then it is known that the linear space is a projective plane. If not, it follows that there is a prime \(p\) such that \(p\mid b\) but \(p\) does not divide \(v\). This is an observation that has been made by many authors and the prime \(p\) has been called a `significant prime'. The interest when there is a group \(G\) acting on the linear space is that if \(P\) is the Sylow \(p\)-subgroup of the stabiliser of a point \(\alpha\) then \(N_G(P)\leq G_\alpha\). The main theorem of this paper concerns a group \(G\) which acts line-transitively on a linear space with a socle which is a simple group of Lie type of characteristic \(p\). If the linear socle is not a projective plane and the significant prime \(p\) then either \(U_3(q)\leq G\leq P\Gamma U(3,q)\) and the space is a Hermitian unital or \(^2G_2(q)\leq G\leq\Aut(^2G_2(q))\) and the space is a Ree unital. One useful step is to show that it can be assumed that the group is actually simple. This depends on a result of the author which has not yet appeared. Since the assumptions imply that the stabiliser of a point contains a parabolic subgroup the bulk of the paper is to show that in this case the stabiliser of a line also contains a parabolic subgroup and so \(p\) is not a significant prime. There is a misprint in the definition of \(n_s\) in the proof of Lemma 3.1. Also in Lemma 4.2 the author refers to \(P\Omega_{2m}^+(q)\) but later talks about \(D_m(q)\), which are of course the same.