Paracerceis sculpta breeds in intertidal sponges, Leucetta losangelensis, where males employ 1 of 3 discrete alternative reproductive behaviors. Elaborate alpha-males attract females to spongocoels where mating and brooding of young by females occurs. Variance in the number of females per alpha-male is high (N = 0-11). Smaller beta-males, resembling females, and tiny gamma-males, resembling juveniles, invade spongocoels containing alpha-males and sexually receptive females. Alpha-, beta-, and gamma-males maintained in the laboratory do not molt or grow, and the 3 morphs differ in the relative amounts of energy they invest in somatic versus gonadal tissue (gamma > beta > alpha). Alternative male reproductive behaviors may have evolved in P. sculpta, since intensifying sexual selection on alpha-males allowed only the most competitive alphas to mate. Males that obtained mates by avoiding direct competition with alphas (e.g., mimicking females or stealing mates) may have persisted, despite their reduced fitness, because they experienced greater fitness than competitively inferior alphas. Similar selective pressures and thus similar male polymorphisms probably exist in other Crustacea. Males in many animal species exhibit discontinuous variation in behaviors and morphologies that are associated with reproduction (Parker, 1970; van Rhijn, 1973; Alcock, 1979; Hamilton, 1979; Perrill et al., 1978; Dominey, 1980; Gross and Charnov, 1980; Cade, 1981; Thornhill, 1981; Eberhard, 1982; Gross, 1982; Clutton-Brock et al., 1982; Austad, 1984; Howard, 1984; Gross, 1985). These alternative reproductive behaviors (ARBs) occur primarily in polygynous species in which variance in male mating success is high and thus in which sexual selection is strong (Gadgil, 1972). Behaviors are emphasized in this terminology, since male reproductive alternatives involving only facultative shifts in reproductive tactics are by far the most common, while ARBs involving morphological differences between males are comparatively rare. Moreover, differences in reproductive morphology are usually associated with specific differences in mating behavior (Austad, 1984). Reports of ARBs are few or nonexistent for the Crustacea. This is unexpected, since many crustaceans are polygynous (Salmon, 1984) and sexual selection has been demonstrated in a variety of species (Holdich, 1968, 1971; Stein, 1976; Knowlton, 1980; Shuster, 1981; Christy, 1983; review in Salmon, 1984). The exemption of the Crustacea from what now seems a common evolutionary response to male sexual selection bears further investigation. Glynn (1968) described two male morphs in a Caribbean isopod, Dynamenella perforata, that inhabits the pallial grooves of intertidal chitons. He found small, intermediate males within isopod aggregations consisting primarily of females and of larger and ornamented males. Although intermediates possessed penes, no reproductive activity was associated with these individuals because they lacked mature appendices masculinae and did not exhibit the enlarged uropods and elaborate telsons possessed by adult males. Glynn maintained intermediates and undifferentiated individuals in a common container, and reported that some of these individuals grew into adult males after several days. The fact that intermediates lacked fully developed appendices masculinae argues in favor of Glynn's