A 24-week double-blind study was conducted to compare haloperidol and thiothixene for efficacy and safety in 46 schizophrenic outpatients. In addition to the standard psychiatric rating scales, Brief Psychiatric Rating Scale (BPRS), Nurses' Observation Scale for Inpatient Evaluation (NOSIE), and Evaluation of Social Functioning Rating (ESFR), two scales more sensitive to the incidence of treatment emergent depression were utilized. They were the Hamilton Depression Scale (HPRSD) and the Zung Self-rating Depression Scale (ZUNG). On the BPRS factors, haloperidol was significantly superior to thiothixene in Thought Disturbance and Hostility-Suspiciousness, and in Total symptomatology. Haloperidol was also significantly superior to thiothixene in Cognitive Disturbance on the HPRSD. Results of global evaluations suggested haloperidol produced slightly more rapid relief of symptoms than did thiothixene. The inclusion of the depression scales was useful in following patients who exhibited depressive symptoms; clinically significant depression was seen in 5 patients receiving haloperidol and 3 receiving thiothixene. A high incidence of akathisia in the thiothixene group was responsible for a statistically significant difference between groups in the number of central nervous system symptoms. Mean doses of test drugs were 17.5 mg/day for haloperidol an 31.8 mg/day for thiothixene. The study showed that haloperidol was equal to and in some parameters superior to thiothixene in producing improvement in the symptoms of psychosis.