The effect of traditional tricyclic antidepressants on serum prolactin levels is controversial. In a five-week double-blind study of depressed outpatients, imipramine hydrochloride therapy did not lead to any significant change in serum prolactin levels. In contrast, amoxapine, a new antidepressant, produced significant elevations in serum prolactin levels in female and in male patients. Amoxapine may block dopamine receptors in central tuberoinfundibular pathways, which would account for its prolactin-elevating activity. On the other hand, imipramine and other traditional tricyclic antidepressants do not affect dopamine transmission, do not raise serum prolactin levels, and are not effective antipsychotic drugs.