Because findings regarding the prognostic significance of depressive symptoms in schizophrenia and their effect on the course and treatment of schizophrenia have been limited by the effects of previous treatment, retrospective evaluations, and differing definitions and criteria, the authors sought to determine the prevalence and prognostic significance of depressive symptoms in first-episode schizophrenia. Thirty-nine men and 31 women experiencing their first episode of schizophrenia were evaluated with behavioral and extra-pyramidal symptom scales before treatment (baseline), biweekly during acute treatment, and then monthly. Extracted scores on the Hamilton Rating Scale for Depression and a "syndromal" definition of depression based on Research Diagnostic Criteria were obtained. Patients were followed prospectively for up to 5 years and received open standardized treatment. The prevalence of depressive symptoms at baseline ranged from 75% (patients who met extracted Hamilton and/or syndromal criteria) to 22% (patients who met both criteria). Of 808 psychotic ratings given to the 70 patients over a 5-year follow-up period, 210 (26%) were concurrently rated as depressed; of the 1,754 nonpsychotic ratings, only 70 (4%) were concurrently rated as depressed. Of the 210 depressive symptoms that occurred concurrently with psychosis, 206 (98%) resolved as the psychosis remitted. Depressive symptoms were prodromal to a psychotic relapse in only two (7%) of 27 patients who relapsed. Depressive symptoms correlated more with positive and negative symptoms than with extrapyramidal symptoms. These findings suggest that depressive symptoms in patients experiencing their first episode of schizophrenia may represent a core part of the acute illness or may occur as a subjective reaction to the experience of psychotic decompensation. Since most of the depressive symptoms resolved as the psychosis remitted, antidepressant therapy should be limited to patients in whom the depression persists.