Hypercortisolemia is a frequently observed abnormality in patients with major depression. It has been hypothesized that the hippocampus, as a major feedback site for glucocorticoids, is involved in the pathophysiology of hypercortisolemia. Some have in fact posited that the hippocampus is marked by diminished size in depressed patients with hypercortisolemia. We tested this hypothesis by examining the relationship between hippocampal volume, assessed with magnetic resonance imaging, and hypercortisolemia using the dexamethasone suppression test (DST) in a group of 19 depressed patients. No differences in hippocampal volume were observed between patients and control subjects (n = 30). Within the patient group, DST suppressors did not differ from DST nonsuppressors in hippocampal volume. However, a relationship between hippocampal volume and 11 p.m. cortisol concentration was observed after covariance adjustment for age and sex. Furthermore, significant negative correlations were observed between hippocampal volume and both age of depressive onset and number of hospitalizations. The results of this study therefore provide limited support for the hypothesis regarding an essential role of the hippocampus in the neuroendocrine elevation of glucocorticoids in depression.