This study tested the hypothesis that puberty in primates is triggered by a remodeling of synaptic inputs and/or glial coverage of hypothalamic gonadotropin releasing hormone (GnRH) neurons. Male rhesus monkeys were prepubertally castrated at 16 months of age and were killed and perfused either 1 month later (n = 4, juvenile group) or at 30 months of age, shortly after initiation of the pubertal increase in pulsatile GnRH release (n = 4, adult group). Hypothalami were sectioned, immunocytochemically stained for GnRH, and processed for electron microscopy. Cross-sectional profiles of 77 GnRH cells from the medial basal hypothalamus (MBH) and the region of the organum vasculosum of the lamina terminalis (OVLT) were compared between the two developmental stages. GnRH cell and nucleolus size in the two groups were the same. The percentage of GnRH perikaryal membrane occupied by synaptic density in the MBH of juveniles was significantly greater (P < 0.05) than that of adults. Differences in the percentage of GnRH perikaryal membrane occupied by synaptic density were not observed in the OVLT nor on GnRH dendrites in either brain region. Qualitative analysis, based on synaptic vesicle shape, failed to reveal developmental differences in putatively excitatory or inhibitory synapses on GnRH cells. The degree of glial ensheathment of GnRH neurons did not change significantly during the two developmental stages. These findings provide ultrastructural evidence for the view that, in primates, neuronal plasticity, and specifically a decrease in synaptic input to GnRH perikarya, may underlie the initiation of the pubertal mode of release of this neuropeptide, and therefore, the onset of puberty in these species.