Fifty-eight actively psychotic inpatients who initially met criteria for long-standing schizophrenia and subsequently met Research Diagnostic Criteria for a current episode of schizoaffective disorder (mainly schizophrenic) with a depressive syndrome, and who scored at least 30 (mean = 55, SEM = 1.6) on the Brief Psychiatric Rating Scale and 17 (mean = 23, SEM = 0.7) on the Hamilton Rating Scale for Depression, were treated for 5 weeks with haloperidol hydrochloride and benztropine. Haloperidol and benztropine treatment was continued, while those patients who consistently scored greater than 17 on the Hamilton Rating Scale for Depression were randomly assigned to the following double-blind treatment groups for 4 weeks: adjunctive amitriptyline hydrochloride, desipramine hydrochloride, or placebo. Adjunctive desipramine or amitriptyline showed no significant therapeutic advantage, when compared with haloperidol and placebo, on the Brief Psychiatric Rating Scale or the Hamilton Rating Scale for Depression. After 4 weeks of combine therapy, patients receiving adjunctive amitriptyline or desipramine, as compared with those receiving adjunctive placebo, tended to score higher on the Brief Psychiatric Rating Scale hallucinatory behavior item and on the thinking disturbance factor than patients receiving placebo. These results suggest that adjunctive antidepressants are not indicated for the treatment of depressive symptoms in actively psychotic schizophrenic inpatients. Adjunctive antidepressants may retard the rate of resolution of psychosis in this population.