1. Mutations to r^g.

A sharp distinction cannot be drawn between mutants to r^g in groups C and D. It is expected rather, as Stadler and Roman have demonstrated in their study of A₁ mutants (Genetics 33: 1948), that an intergrading spectrum of effects could be assembled ranging from minute deficiencies to mutations indistinguishable from the bottom recessive. For the mutants of r^ch to r^g with lowered haploid viability, there is little doubt that these alterations may safely be ascribed to physical loss of the locus, or its inactivation. Earlier preliminary study of these forms by Perak, however, failed to disclose any gross abnormality in pachytene configuration. It is not implied that all mutations to r^g, though phenotypically alike, are identical in nature. There is no reason to assume that allelic compounds in this class might not have positive action in plant coloration. Appropriate tests of this hypothosis are now in progress.