In a double-blind trial in which 45 patients with acute schizophrenia took part the alpha-isomer of flupenthixol (which blocks the dopamine receptor) was found to be significantly more effective than both beta-flupenthixol (which does not) and placebo. The drug effect was confined to the "positive" symptoms--delusions, hallucinations, and though disorder--and appeared only in the 3rd and 4th weeks of the trial. It was as great in patients with evidence of deterioration (Feighner-positive patients) as in patients without deterioration and was less in patients who had affective disturbance in addition to schizophrenia symptoms. The findings are consistent with the hypothesis that dopamine-receptor blockade is the only requirement for antipsychotic activity and suggest that the antipsychotic effect occurs in patients with typically schizophrenic illnesses but may be limited to positive symptoms.