Through the synthesis and secretion of the hormone, melatonin, the pineal has been assigned the role of synchronizing a reproductive response to appropriate environmental conditions. Theoretical melatonin target sites may occur at several levels of the hypothalamic-pituitary-gonadal hierarchy, including a direct action on the gonads. The availability of a biologically active radioligand, 2-[125I]iodomelatonin, has provided the opportunity to examine the possible direct melatonin action on the gonads. 2-[125I]Iodomelatonin binding sites were identified in the testes and ovaries of chickens, ducks and quail but were not measurable in mammalian gonads, with the exception of tree shrew testes. The avian gonadal 2-[125I]iodomelatonin binding sites were stable, saturable, reversible, specific and of high affinity. 2-[125I]Iodomelatonin appeared to label a single class of binding sites as evidenced by the linearity of Rosenthal analysis of the specific binding data, the Hill coefficients close to unity and the monophasic competition curves. The high affinity on the gonadal 2-[125I]iodomelatonin binding sites, characterized by apparent equilibrium dissociation constants in the low picomolar range, was in accordance with circulating levels of melatonin suggesting that they may be physiologically relevant. Autoradiography indicated that these 2-[125I]iodomelatonin binding sites were widely distributed throughout the testes but localized in ovarian follicles in the birds studied. Specific inhibition of testicular 2-[125I]iodomelatonin binding by a guanine nucleotide analog has provided evidence that the 2-[125I]iodomelatonin binding sites in chicken testes may be coupled to a guanine nucleotide binding protein-effector system, thus promoting the idea that testicular 2-[125I]iodomelatonin binding sites may mediate a cascade of intracellular events. Although no circadian rhythm in the density or affinity of 2-[125I]iodomelatonin binding in chicken ovaries was found, there was a decrease in 2-[125I]iodomelatonin binding affinity at middark in chicken testes with no change in the number of testicular 2-[125I]iodomelatonin binding sites. The present evidence is in line with the hypothesis of a direct melatonin action on the gonads and further investigations on the above problem will be rewarding.