A dexamethasone suppression test (DST) was performed on 8 schizoaffective depressed men. Cross-sectional comparisons were made with three groups: schizophrenics (n = 10), unipolar major depressives (n = 23) and healthy controls (n = 43). All were drug-free and similar in age and body weight. Evaluations utilized the Research Diagnostic Criteria (RDC) for diagnosis, and the Hamilton Rating Scale for Depression for depressive symptom rating. DST nonsuppression, defined as a blood cortisol level of ≥5.0 μg/dl at 16.00 h postdexamethasone, was observed in 43.5% of the major depressive disorder patients. This was different from the other three groups: 12.5% in schizoaffective depressed, 10.0% in schizophrenics and 9.3% in healthy controls (p < 0.01, p < 0.01, and p < 0.001 respectively). Although schizoaffective depressed patients were significantly different from major depressive disorder patients in their DST responses, both groups were similar in their total HRSD scores and different from the schizophrenics (p < 0.01 for each). These results, together with others previously reported by us on the thyrotropin-releasing hormone challenge in the same diagnostic groups, may be taken to mean that schizoaffective disorder, depressed type, is biologically distinct from major depressive disorder but not schizophrenia. On the other hand, until further corroborated, they should probably be considered a reflection of the heterogeneity of the schizoaffective syndrome and the nonspecificity of the DST.