Roots in the mapping class groups

From MaRDI portal




Abstract: The purpose of this paper is the study of the roots in the mapping class groups. Let Sigma be a compact oriented surface, possibly with boundary, let PP be a finite set of punctures in the interior of Sigma, and let MM(Sigma,PP) denote the mapping class group of (Sigma,PP). We prove that, if Sigma is of genus 0, then each finMM(Sigma) has at most one m-root for all mge1. We prove that, if Sigma is of genus 1 and has non-empty boundary, then each finMM(Sigma) has at most one m-root up to conjugation for all mge1. We prove that, however, if Sigma is of genus ge2, then there exist f,ginMM(Sigma,PP) such that f2=g2, f is not conjugate to g, and none of the conjugates of f commutes with g. Afterwards, we focus our study on the roots of the pseudo-Anosov elements. We prove that, if partialSigmaeqemptyset, then each pseudo-Anosov element finMM(Sigma,PP) has at most one m-root for all mge1. We prove that, however, if partialSigma=emptyset and the genus of Sigma is ge2, then there exist two pseudo-Anosov elements f,ginMM(Sigma) (explicitely constructed) such that fm=gm for some mge2, f is not conjugate to g, and none of the conjugates of f commutes with g. Furthermore, if the genus of Sigma is equiv0(mod4), then we can take m=2. Finally, we show that, if Gamma is a pure subgroup of MM(Sigma,PP) and finGamma, then f has at most one m-root in Gamma for all mge1. Note that there are finite index pure subgroups in MM(Sigma,PP).











This page was built for publication: Roots in the mapping class groups

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3617280)