The evidence that hypothalamic gonadotropin-releasing hormone (GnRH) release increases during the estrogen-induced luteinizing hormone (LH) surge in castrated female macaques and that estrogen induces a similar LH surge in the male suggests that either sexual differentiation of GnRH secretion does not exist or that changes in brain GnRH are not critical for ovulation. We tested this hypothesis by using the push-pull perfusion (PPP) technique to monitor GnRH release in the mediobasal hypothalamus continuously from 10 h before to 50 h after estrogen injection in 9 castrated male rhesus macaques. Blood sampling and PPP were performed with monkeys freely moving in their own cage by using a specially designed swivel/tethering system. PPP samples were collected every 10 min and analyzed for GnRH concentration by radioimmunoassay. Blood samples were collected every hour and plasma LH was measured by bioassay. Estradiol benzoate (EB, 42 micrograms/kg, b. wt.) was subcutaneously injected after 10 h of initial PPP. The PPP was continued for 10 h after EB in 4 and 50 h after EB in 5 monkeys. Hypothalamic GnRH patterns were analyzed by the Pulsar algorithms. The results show that during the 10 h after EB, plasma LH declined rapidly, reaching low or non-detectable levels by 7-9 h, while hypothalamic GnRH releasing patterns did not change during this period (n = 9). In contrast, estrogen enhanced GnRH level and pulse amplitude, but not pulse frequency, several hours before mean peripheral plasma LH increased from suppressed values (n = 4).(ABSTRACT TRUNCATED AT 250 WORDS)