Thirty-five (35) inpatients with nonpsychotic major depression were randomly assigned to either amitriptyline or nortriptyline on a double-blind basis, following a 4- to 10-day placebo run-in period. Thirty-two patients completed at least 3 weeks of active medication. The 1.0-mg dexamethasone suppression test (DST) was performed before treatment. Twelve of 32 patients were DST nonsuppressors. Both medications were equally effective. Pretreatment DST status failed to predict overall response to both medications combined. DST status also did not predict differential medication response. These data argue, along with several other studies, that pretreatment DST status may be of limited value in selecting a particular tricyclic antidepressant compound.