Rough-and-tumble play is sexually dimorphic in many species of mammals. Effects of androgen exposure on this behavior have been well documented, although the possible role of other hormones is less well known. This paper reports experiments in a series designed to investigate effects of neonatal exposure to progestins on the development of juvenile play behavior in rats. Specifically, the effects of altering levels of progestins and androgens in combination on the subsequent development of play behavior was studied, in order to investigate further the mechanism of progestogenic effects. We have previously [ L. I. A. Birke and D. Sadler (1984). Physiol. Behav. 33, 217–219]reported an apparently anomalous effect of lowering endogenous progesterone using the antiserum given postnatally; this was predicted to raise levels of juvenile play. The prediction was, however, supported only for males; treated females showed lower levels of play than controls. The present study was designed to investigate these effects further, and to test hypotheses about the mode of action of neonatal progestins. Using various combinations of neonatal progestins (progesterone or medroxy-progesterone acetate, MPA), and testosterone or the antiserum to testosterone, we analyzed the amounts of play and other social behavior shown by juvenile rats. The data support a hypothesis that the previously reported effects of progestin exposure on play result from weakly androgenic actions of the progestins, which presumably compete with other androgens in males. The cellular mechanisms involved are, however, unknown.