In order to determine whether there are age-related changes in the supragranular layer of the dentate gyrus of the rhesus monkey, we examined this layer in monkeys 4-35 years of age, spanning the entire range of adulthood of this species. Electron microscopic analyses were conducted to determine whether there is an age-related change in the number of synapsing axon terminals, in the cross-sectional area of these terminals, or in the length of the postsynaptic density at the synaptic junction. Only asymmetrical synapses in the anterior dentate gyrus were evaluated. In a subset of our monkeys (n = 6, ages 4-31 years), we compared three different approaches to the estimation of synaptic density: (1) the conventional profile method, in which synapse numbers are expressed per unit area of the examined tissue section; (2) the empirical formula of Colonnier and Beaulieu for converting areal densities into number of synapses per unit volume, and (3) the 'disector' method, a stereological approach to the estimation of the number of synapses per unit volume that makes no assumption about the shape of the objects. Data are presented validating the small-fold method of estimating section thickness for the disector. The three methods were highly intercorrelated (rs approximately 0.89), and none of the methods revealed an age-related loss of synapses. Analysis of the thickness of the dentate gyrus molecular layer suggests that the reference volume in which these synapses were counted dose not change with age. In addition, the conventional profile method showed no age-related change in the number of axodendritic or axospinous synapses, the cross-sectional area of the synapsing terminals, or the length of the postsynaptic densities of synapses. Together, these data suggest a remarkable age-related preservation of synapses in the normal aging monkey.