Symmetry breaking and the maximal isotropy subgroup conjecture for reflection groups (Q1813209)

\textit{M. Golubitsky} [Bifurcation theory, mechanics and physics, Proc. Colloq., 225-256 (1983; Zbl 0542.76054)] conjectured that generically the solutions to equivariant bifurcation problems have maximal isotropy type. Roughly the conjecture goes as follows: The maximal isotropy subgroup conjecture (MISC). Let the compact Lie group \(G\) act absolutely irreducibly on \(\mathbb{R}^ n\). Let \(g : \mathbb{R}^ n \times \mathbb{R} \to \mathbb{R}^ n\) be \(G\)-equivariant with a singularity at (0,0). Then generically each non trivial branch of solutions of \(g = 0\) corresponds to a maximal isotropy subgroup. Moreover each maximal isotropy subgroup corresponds to a branch of solutions to \(g = 0\) for some \(g\). Golubitsky also remarked that he believed the conjecture to be true if all of the maximal isotropy subgroups have corresponding fixed point spaces of dimension one. It is now known that this conjecture is false. In this work we have two main aims. First, we give a description of symmetry breaking in bifurcation theory in terms of transversality conditions involving certain semi-algebraic sets. As a result we are able to reformulate the MISC in geometric terms. Our second aim is to provide a relatively complete analysis of the possible symmetry breaking patterns for the finite reflection groups and, in particular, verify whether or not the MISC holds for these groups.