The pineal hormone melatonin is known to play an important role in mediating photoperiodic messages to the reproductive system in seasonal breeding animals. Our goal was to test, in a single experimental paradigm, two hypotheses that have been forwarded to describe how the circadian rhythm of pineal melatonin transmits photoperiodic information to the reproductive system: 1) induction, i.e., a short-day effect, occurs when secreted melatonin and a circadian rhythm of sensitivity to melatonin coincide in time; ?) induction occurs following exposure to elevated circulating melatonin levels for a prescribed duration. in order to determine the relative validity of these hypotheses, we investigated the testicular maturation response to I-hr daily infusions of 10, 25, and 50 ng of melatonin in pinealectomized intact and prepubertal Siberian hamsters (Phodopus sungorus). Animals received, beginning on day 15 of life, programmed subcutaneous infusions of melatonin or vehicle at one of five time points (19:00-20:00, 20:00-21:00, 21:00-22:00, 24:00-01:00, and 03:00-04:00 hr) for 15 days. In animals gestated and raised in a long photoperiod (LD16:8 = 16L, where I, is the duration of light in hours, and D that of dark), melatonin infusion right after lights off (20:00-21:00 hr) significantly retarded gonadal maturation; this dose was ineffective at other times tested. Doses of 10 and 25 ng melatonin were ineffective at all time points. Identical results were obtained in prepubertal hamsters gestated in a short photoperiod (LD10:14= 10L) and raised in 16L; these results were independent of the presence or absence of the pineal gland. In animals gestated and raised in 10L, melatonin infusions failed to suppress testicular development beyond that induced by the photoperiod: testicular development was maximally suppressed in all groups. The results of these investigations are best explained under the experimental conditions employed here: 1) the photoperiodic gonadal response in juvenile Siberian hamsters is regulated by the coincidence in time of exogenously administered melatonin with an intrinsic rhythm of sensitivity to melatonin, which, under the constraints imposed by our experimental design, occurred at 20:00 21:00 hr; and 2) the duration of the melatonin signal alone, equal in all groups, cannot explain the results.