To investigate mechanisms of blunted adrenocortical responsiveness to exercise and mild hypercortisolism in amenorrheic runners, adrenocorticotropic hormone [ACTH-(1-24) 0.25 mg Cortrosyn] stimulation tests were performed in the presence and absence of overnight dexamethasone (1 mg) suppression (DX and NDX condition, respectively) in six eumenorrheic sedentary women (ES), nine eumenorrheic runners (ER), and nine amenorrheic runners (AR). Before the NDX stimulation test, plasma cortisol was higher (P < 0.001) in AR than in ER and ES. The cortisol response to the NDX stimulation test was blunted (P < 0.001) in AR but reached similar (P > 0.7) peak levels in all groups. Dexamethasone suppressed (P < 0.001) cortisol to similar (P > 0.5) levels (approximately 20 nmol/l) in all groups. In AR, cortisol responses to the DX test were larger (P < 0.03) than to the NDX test and similar (P > 0.6) in the three groups, again reaching comparable (P > 0.8) peak levels. The blunted cortisol response to stimulation in AR in the presence of their mild hypercortisolism appears to be due to a normal limitation in maximal adrenal secretory capacity. Extrapituitary modulators of adrenal responsiveness to ACTH may explain the mild hypercortisolism observed in AR, but limitations of these tests prevent a central negative-feedback defect or an intrinsic adrenal abnormality from being excluded until results of additional studies with even lower doses of dexamethasone and submaximal doses of ACTH-(1-24) are available.