This paper provides an expanded discussion of issues raised in our 1990 experimental study, with particular focus upon the differential cognitive, affective, and psychomotor consequences to girls and boys of gender-typed, sex-segregated play. Evidence is provided that, as agents of socialization, teachers and schools can enlist practices and provide structure to foster cross-sex interaction and behavior. Nonstereotypical play is advocated so that each sex can enjoy the benefits typically accruing only to the opposite; correspondingly, and in light of a society not yet gender-aschematic, psychological androgyny is invoked as the developmental ideal.