Prophylactic effects of psychotropic drugs on 55 schizophrenics in remission were evaluated for 3 years in a double-blind controlled study employing a cross-over design. Patients were randomly assigned to the following drugs orally administered twice a day: placebo; diazepam 15 mg; imipramine 50 mg; chlorpromazine 75 mg; and haloperidol 3 mg. The number of days of remission for each patient was recorded. Since only two patients received all five drug treatments, the data were analyzed using the number of days allocated to the "first assigned drugs" only and the cross-over aspect of the experimental design was disregarded. All patients treated with either the placebo, diazepam or imipramine relapsed within a year. On the other hand, four patients treated with chlorpromazine, or with haloperidol, were in remission for more than 1 year. Fifty percent of the patients relapsed within 16 days with placebo; 88 days with diazepam; 30 days with imipramine; 165 days with chlorpromazine; and 74 days with haloperidol. Within a year, only chlorpromazine significantly prolonged the remission state as compared to placebo and imipramine. At the end of the 3-year trial, both chlorpromazine and haloperidol significantly prolonged the remission state as compared to the other three drugs. These data suggest that neuroleptic treatment for a longer period is vitally important to prevent relapse even in schizophrenics in remission and that such a trial seems an efficient method for investigating the prophylactic effects of neuroleptics.