3H-Imipramine binding was measured in freshly prepared platelet membranes from 47 drug-free major depressives and 46 healthy controls. Where possible, platelet binding in depressed subjects was repeated following treatment. A significant negative correlation was found between Bmax and assay protein concentration and Bmax values were corrected for this effect. Adjusted Bmax was significantly lower (by 14%) in female depressed patients than in female control subjects, and the difference was of similar magnitude premenopausally and postmenopausally. No such difference was found in males. Kd did not differ significantly between depressed and control subjects. Multiple regression analysis confirmed significant effects on Bmax of presence of depressive illness, age (positive correlation), and season (higher in summer). Within the depressed sample, Bmax was significantly lower in those subjects with obsessional features. Endogenicity (Research Diagnostic Criteria or Newcastle), dexamethasone suppression test result, drug-free interval, family history of depression, depressive psychosis, suicidal ideation, and past history of suicide attempts were not significantly related to Bmax. Paired comparisons revealed no significant effect on Bmax of 6 weeks' treatment with imipramine, maprotiline, or BRL 14342 or of a course of electroconvulsive therapy.