Luteinizing hormone (LH) and prolactin (PRL) levels were compared in the serum of ovariectomized rhesus macaques (Macaca mulatta) with pituitary-intact (PI) and surgically transected pituitary stalks (SS). After stalk sectioning, LH levels declined and PRL levels rose dramatically. Chronic infusion of intermittent pulses of gonadotropin-releasing hormone (GnRH; 1 microgram/min for 6 of every 60 min) into SS females reinitiated LH secretion within a few days despite increased PRL secretion. Continuous infusion of dopamine (10 micrograms/min/kg of body weight) for 9 days suppressed PRL levels in GnRH-infused SS females to near normal without alteration in basal serum LH levels or pulsatile LH release. A comparable dopamine regimen also failed to modify serum LH patterns in PI monkeys. Immediately after cessation of dopamine, serum PRL levels rose sharply in both PI and SS animals. Injection of estradiol benzoate (42 micrograms/kg of body weight) into SS monkeys receiving GnRH infusions resulted initially in LH suppression and then in LH release. However, the magnitude of LH release was less and the latency to its onset was longer in SS animals given estrogen than in PI controls with similar serum estrogen concentrations. It is unclear whether the atypical pattern of the estrogen-induced LH release in SS monkeys was due to hyperprolactinemia or other deficiencies in this GnRH-infused model. However, these data indicate that the response of gonadotropes to GnRH stimulation was not altered by a direct action of either PRL or dopamine on the pituitary gland. Conversely, dopamine directly suppressed the lactotrope, which became hypersecretory immediately upon dopamine withdrawal. We have concluded that estrogen-induced inhibition and subsequent augmentation of LH release represents a direct action at a pituitary locus, although effects at additional, nonpituitary loci are not precluded.