The
fem-1 gene is required for male development of germline and somatic tissues in C. elegans . Animals lacking
fem-1 activity develop as cross-fertile females regardless of their X chromosome dose. Following a cross to wild-type males, most
fem-1 mutants produce phenotypically wild-type XO male and XX hermaphrodite heterozygous progeny. In contrast, females homozygous for any of three deficiencies that remove all or part of
fem-1 produce cross-progeny of both sexes that exhibit germline feminization. The penetrance of this effect varies from 20% to over 90%, depending upon the deficiency, and it dramatically increases in the second generation following a backcross to females homozygous for the deficiency. Feminization appears to result from maternal absence of some product of the
fem-1 locus rather than a zygotic interaction involving the deficiency chromosome, because the paternally disomic progeny of a cross to deficiency homozygotes exhibit germline feminization. It is unlikely that maternal loss of FEM-1 protein is responsible for the effect, because alleles carrying point mutations that prevent production of FEM-1 complement the deficiencies with respect to their maternal effect on the germline. Genetic evidence suggests that the affected animals fail to express
fem-1 at normal levels in the germline, which could account for their phenotype. I suggest that a
fem-1 transcript is maternally required to license
fem-1 for expression in the germline. A similar requirement might exist for the germline expression of other genes, but it would be difficult to detect in the absence of deficiency alleles that allow homozygotes to develop into fertile adults. I am investigating the relationship between mechanisms involved in maternal-effect silencing of
fem-1 and other silencing phenomena in C. elegans .