We have reported a decrease in the number of arcuate nucleus (ARC)-immunoreactive β-endorphin neurons in old (24 months) female C57BL/6J mice versus young (5 months) mice. Here, we have tested by immunocytochemistry whether age-related changes in β-endorphin neuron numbers are selective for β-endorphin neurons which do or do not contain estrogen receptors (E2R). We also compared β-endorphin neuron number in mice with short- (S) and long-duration (L) ovariectomy (OVX), since the latter may protect against neuroendocrine aging. Mice were studied at 5 (young), 12 (middle-aged), or 23-24 months (old). When the mean number of neurons per tissue section (15 sections per animal) was examined, there were no significant differences between young and middle-aged S-OVX females for either β-endorphin, E2R, or β-endorphin/ E2R neuron number. However, there were significant decreases in β-endorphin-containing neurons in the oldest age group versus young females (young S-OVX: 74.4 ± 11 (±SD) immunopositive neurons per tissue section, n = 10 mice; young L-OVX: 61.6 ± 6.9, n = 6; old S-OVX: 45.7 ± 9.9, n = 7; and old L-OVX: 37.5 ± 7.3, n = 7). There were also decreases in β-endorphin neurons which contained E2R in the oldest animals (young S-OVX: 16.6 ± 6.4; young L-OVX: 13.7 ± 1.3; old S-OVX: 9.2 ± 1.8; L-OVX: 6.0 ± 1.5) (p < 0.05 ANOVA). Both age (p < 0.001, two-way ANOVA) and ovarian status (p < 0.05) independently affected neuron number for both the β-endorphin and β-endorphin E2R populations versus young mice. We tested whether the observed age and/or ovarian-related decreases were proportionally greater in the subpopulation of β-endorphin neurons which contained E2R compared to the total β-endorphin neuron population. In the oldest age group, there was no significant difference in the decrease with age in the population of β-endorphin neurons which contained E2R and the total β-endorphin population (p = 0.208). When we examined the E2R neuron population as compared to the β-endorphin neuron populations, age-related decreases in the β-endorphin neuronal population tended to be greater than the decreases seen in the E2R neuron population (p = 0.054 repeated measures ANOVA). The tyrosine hydroxylase (TH) neuron population was studied to test whether there were changes in another ARC neuron population. There was no age-related change in either TH neuron number (young S-OVX: 35.8 ± 6.2, n = 4; middle-aged S-OVX: 33.1 ± 2.8, n = 4; old S-OVX: 33.4 ± 3.4, n = 4) or in the number of TH neurons which contained E2R (young S-OVX: 7.6 ± 1.0; middle-aged S-OVX: 7.2 ± 0.8; old S-OVX: 6.9 ± 1.2). These data demonstrate that ARC β-endorphin-containing perikarya are sensitive to alterations associated both with age and ovarian status; these effects are independent of one another. Our findings fail to support the hypothesis that L-OVX has a protective effect on hypothalamic neurons in older females. The subpopulation of β-endorphin neurons that contain E2R are not affected to any greater degree by age or ovarian status than the β-endorphin neuron population as a whole in the old animal.