Serum samples were obtained from 7 subjects (6 men and 1 woman) in the morning (09.00 h) and in the afternoon (15.00 h) of each week for a period of 13 months and assayed for prolactin by RIA. In addition, 4 female monkeys were sampled once weekly in the afternoon for approximately two years. The data were analyzed for cyclicity by power spectrum analysis. In both species a serum prolactin circannual cycle was not obvious. In most subjects statstically significant (P<0.05) cycles were observed; however, there was considerable variability among individuals in the duration of the cycle, which ranged between 4 and 50 weeks. A similar result was also observed for the monkey data; here the cycles ranged between 7 and 51 weeks. The physiologic significance of these cycles is unknown at present. It is suggested that although there appears to be no diurnal serum prolactin circannual cycle, this does not rule out the possibility that there may be a nocturnal circannual cycle.An annual alteration in reproductive capacity and behaviour has been appre¬ ciated for a number of years in animals under natural conditions (Asdell 1946). The advent of sensitive and specific radioimmunoassays enabled in-