Old male rhesus macaques display less sexual behavior than young and middle-aged males. The decrease in sexual activity occurs without a statistically significant decline in gonadal hormones or change in diurnal patterns of serum T, DHT, or LH. Levels of sexual activity are not increased by administering T to old intact males. However, the hormone is effective in increasing sexual behavior in old long-term-castrated males. Performance can be increased to levels observed in equally old untreated intact males. Readily detectable physical disabilities of old age have been observed to impair sexual performance, but the observed general decline in sexual activity cannot be accounted for by known physical disabilities. Novelty, as represented by a change in female partner or by a change in environment, has not increased sexual performance in old rhesus males. Only when old males were paired with empirically selected preferred females is their sexual behavior increased to levels displayed by young males. Drugs reported to increase levels of sexual behavior in rats have thus far been less effective in old rhesus males than powdered rhinoceros horn has been in man. The probable absence of a placebo effect in rhesus males should increase their usefulness as an animal model for the study of sexual behavior in aging men.