The suppression of plasma corticosterone (B), measured by radioimmunoassay (RIA), was compared to simultaneous suppression of plasma cortisol (F), measured as total corticoids by a competitive protein binding (CPB) assay, in the overnight dexamethasone suppression test (DST). Baseline plasma B concentrations in 10 control subjects were 4.04 +/- 1.07 ng/ml (X +/- S.D.) at 0800 hr and 1.51 +/- 0.68 ng/ml at 1600 hr. Post-dexamethasone 1600 hr B levels in the controls were 0.46 +/- 0.29 ng/ml. An early escape of plasma B (greater than 1.2 ng/ml), like that of F (greater than 5 micrograms/dl), during the overnight 24 hr 1.0 mg dose DST was noted in patients with melancholia (endogenous depression). Half-hourly catheter samples in a normal subject stimulated to escape from dexamethasone suppression showed that in general, plasma B concentrations parallel plasma F concentrations over a 12 hr period. Repeated weekly DSTs on two patients with different psychiatric diagnoses resulted in B : F correlations of 0.74 and 0.60. Overall agreement between B- and F-DST outcomes in all categories tested at 1600 and 2300 hr was 93%; the agreement in the melancholic and non-endogenous depressed groups was 100%. Post-dexamethasone, both B and F were suppressed 55-60% below the criterion level in controls. In those patients who escaped from dexamethasone suppression, the percentage increase in plasma B above the criterion level was significantly greater (+55%) than the corresponding percentage change in plasma F. Most patients with borderline abnormal F-DSTs (3.5 - 4.9 micrograms/dl) exhibited clearly abnormal B-DSTs (greater than 1.2 ng/ml). We conclude that the use of dexamethasone suppression of plasma B (using 1.2 ng/ml as the abnormal criterion value) is an additional indicator of an abnormal DST in depressed patients.