During the light phase of each of 3 photoperiods tested, plasma melatonin concentrations were less than 16 to 62 pg/ml and during the dark phase they were 31 to 169 pg/ml. When the photoperiod to which the tammars were exposed was altered from 15 h light:9 h dark to 12L:12D the onset of the nocturnal rise in melatonin was advanced from the first day, thereby extending its duration, and the females gave birth 32 +/- 0.4 (mean +/- s.e.m.) days later. To test whether melatonin mediated this effect of photo-period change, tammars in a second group were injected s.c. with melatonin (400 ng/kg, N = 6) or the arachis oil vehicle (N = 6), 2.5 to 2.25 h before dark during 15L:9D for 15 days before exposure to 12.5L:11.5D. The melatonin injections mimicked the endogenous melatonin profile of 12L:12D and the melatonin-injected tammars gave birth 32 +/- 0.8 days after the start of injections, which was the same as the interval from photoperiod change in Exp. 1 but was significantly different (P less than 0.005) from the interval in the control group (46.0 +/- 1.1 days). These results show that exogenous melatonin given 2.5 to 2.25 h in advance of the endogenous rise fully mimics the response of the tammar to photoperiod change.