Cubic extensions of flag-transitive planes. I: Even order (Q1599724): Difference between revisions

In this well-written paper, the central problem is to study the following situation. There is an affine translation plane \(\pi\) of even order \(q^3\) admitting an affine subplane \(\pi_0\) with the property that an automorphism group \(G\) of \(\pi\) leaves \(\pi_0\) invariant, acts flag-transitively on \(\pi_0\) and acts transitively on the parallel classes of lines of \(\pi\) outside \(\pi_0\) (the plane \(\pi\) is a cubic extension of \(\pi_0)\). If \(G\) is solvable, then the extension is called solvable. Here are a few interesting results from the paper. In the above situation, the plane \(\pi_0\) is always either Desarguesian or Lüneburg-Tits (the latter arises from the Lüneburg-Tits spread). For \(q\neq 4\), the group \(G\) contains a group acting on \(\pi_0\) isomorphic to \(\text{SL}(2,q)\) or \(\text{Suz}(\sqrt 2)\), respectively (and the involutions are all elations). If moreover \(\pi_0\) is a cubic flag-transitive extension of some plane \(\pi_1\) (we thus have a small chain of extensions), then the second extension (from \(\pi_0\) to \(\pi)\) is always nonsolvable and either the first is nonsolvable and both \(\pi_0\) and \(\pi_1\) are Desarguesian or both are Lüneburg-Tits (of orders \(2^{18}\) and \(2^6\), respectively), or the first is nonsolvable and \(\pi_2\) is Desarguesian of order 4. Also, the only cubic flag-transitive extension where both planes are Lüneburg-Tits occurs with the orders \(2^6\) and \(2^{18}\). Finally, let me mention that the authors also prove that any solvable extension of order \(q^n\) of a flag-transitive plane of order \(q\), \(q>4\) even, is always a Hall plane.