It is well documented that estrogen increases dendritic spine density in CA1 pyramidal cells of young female rats. However, this effect is attenuated in aged rats. We report here a quantitative analysis of estrogen effects on hippocampal spine number as visualized with antispinophilin in young (6–8 years old) and aged (19–23 years old) female rhesus monkeys, a species with a pattern of female endocrine senescence comparable to that of humans. Monkeys were ovariectomized and administered either vehicle or estradiol cypionate 3 months postovariectomy, followed by an additional dose 3 weeks later, with perfusion 24 hours after the last estrogen treatment. Immunolocalization of spinophilin, a spine-associated protein, was used for quantitative stereologic analyses of total spinophilin-immunoreactive spine numbers in CA1 stratum radiatum and the inner and outer molecular layers of dentate gyrus. In both young and aged female monkeys, the estrogen-treated groups had an increase in spinophilin-immunoreactive spines (37% in young, P < .005; 35% in aged, P < .05) compared with the untreated groups that amounted to more than 1 billion additional immunoreactive spines. The young group also showed a trend toward an estrogen-induced increase in immunoreactive spines in the dentate gyrus outer molecular layer, but this effect was not statistically significant (P = .097). We conclude that spine number in the rhesus monkey hippocampus is highly responsive to estrogen, yet, unlike the female rat, aged female rhesus monkeys retain the capacity for spine induction in response to estrogen. These data have important implications for cognitive effects of estrogen replacement in postmenopausal women and demonstrate that an estrogen replacement protocol that mimics normal physiological cycles with timed, intermittent peaks can have profound neurobiological effects. J. Comp. Neurol. 465:540–550, 2003. © 2003 Wiley-Liss, Inc.