We studied pituitary corticotropin response to exogenous corticotropin-releasing hormone infusion and attempted to control for the confounding effect of variable serum cortisol levels between depressed and control subjects. If metyrapone was given during the time of day when hypothalamic pituitary adrenal activity was otherwise low, the relative increase in the corticotropin concentration was small. Pituitary response to exogenous corticotropin-releasing hormone can be defined under conditions in which the amount of glucocorticoid-mediated negative feedback present at the level of the pituitary gland is equal in all subjects. When the ambient cortisol level was equalized (and suppressed) in all subjects at the time of study with a threshold dosage of corticotropin-releasing hormone, we found an augmented response to corticotropin-releasing hormone in depressives. This raises the possibility that either increased pituitary sensitivity to corticotropin-releasing hormone or an increased intracellular pool of corticotropin is available for release in subjects with major depressive illness.