Haloperidol was administered IV to 46 male psychotic inpatients and 28 male control subjects. A two-way analysis of covariance, with age as the covariate, revealed that DSM-III schizophrenics (n = 27) had a lower prolactin response to haloperidol than did the controls (n = 28). There were no significant differences between the prolactin responses in schizophrenics, patients with affective disorders (n = 7), and those with other psychoses (n = 12), which included patients with paranoia, schizophreniform, schizoaffective disorder, and atypical psychoses. These findings support the proposition that tuberoinfundibular dopaminergic dysfunction may occur in certain patients with DSM-III schizophrenia.