The authors report a study of 24-hour serial cortisol determinations, measured during baseline and after the administration of 0.25 and 0.5 mg of dexamethasone in a sample of predominantly outpatient children with major depressive disorder, nonaffective psychiatric controls, and normal controls. In this sample, 24-hour baseline cortisol and the dexamethasone suppression test (DST) do not discriminate between the three groups. In addition, the authors measured 24-hour serum dexamethasone levels. There were no significant between group differences in serum dexamethasone. These results raise questions as to the utility of this test in the diagnosis of affective disorders in children. Possible reasons for the discrepancies in the dexamethasone suppression test results between in- and outpatient studies are discussed.