Theoretical models of gynodioecy predict a greater reproductive output by females relative to hermaphrodites. Empirical work does not consistently show greater numbers of offspring produced rather greater success of seedlings. This greater success of progeny may relate to the benefits of heterosis from outcrossing, the avoidance of inbreeding from selfing, allocation of recources from maleness into progeny, and/or pleiotrophic effects of male-sterility genes or genes at other loci after gender is established. Our understanding of gynodioecy is limited in that hypotheses have not been stated as alternatives. This is complicated by the fact that, where unequivocably documented, male sterility is the result of nuclear-cytoplasmic interactions. Our lack of knowledge of the genetics of gynodioecy makes it difficult to choose a testable hypothesis. Models are based on assumptions of relative functional gender of hermaphrodites vs. females; our knowledge of the pollination ecology of gynodioecious taxa makes such assumptions tenuous. To state truly alternative hypotheses, we need knowledge of the genetics of male-sterility, the relative contributions of these genes via pollen vs. ovules, and the relative magnitude of paternal vs. maternal components of progeny fitness.