Paper detail

A Built-in Horizontal Symmetry of $SO(10)$

In a renormalizable $SO(10)$ theory, all fermion mass matrices are linear combinations of three fundamental types, $M^{10}, M^{\overline{126}}$, and $M^{120}$, whose superscripts indicate their $SO(10)$ transformation properties. We point out that each of these fundamental mass matrices possesses a natural symmetry that can be used to generate an unbroken horizontal symmetry $\G$, if the natural symmetry is taken to be the residual symmetry. This built-in symmetry is a Coxeter group. If it is finite, it must be one of five groups, $S_4,\ Z_2\x S_4$,\ $Z_2\x A_5$, plus two `rank-4' groups. These symmetries place constraints on the fundamental mass matrices and reduce the number of parameters in an SO(10) fit. Since they are built-in and can be derived theoretically, it is hoped that they impose better constraints than those without a theoretical basis, but that is to be confirmed because there is no attempt to fit the experimental data in this article, except to count the number of free parameters. To illustrate the similarities and differences of various kinds of constraints, a comparison is made with an existing $S_4$ model, and with models possessing the Fritzsch texture.