Plasma cortisol and adrenocorticotrophin hormone (ACTH) profiles were estimated in twelve patients with Addison's disease following randomized oral administration of either cortisone acetate (25 mg) or hydrocortisone (20 mg) alternately, at 0900 h on consecutive days. Normal corticosteroid replacement therapy was discontinued from 1200 h on the day prior to the study period. In four patients elevated basal plasma ACTH concentrations were not suppressed to the limit of detection following the administration of either drug, and in three of these no suppression was found following the prolonged administration of pharmacological doses of dexamethasone. Diminished sensitivity of pituitary ACTH secretion to cortisol inhibition may result from chronic loss of negative feedback before and/or after diagnosis and treatment. In three patients elevated basal plasma ACTH concentrations were suppressed adequately during the administration of either drug, but in five, low basal ACTH concentrations following corticosteroid withdrawal suggested chronic inhibition of anterior pituitary corticotrophs by over-replacement with glucocorticoid. However, further study is necessary to determine whether the estimation of ACTH profiles is a more accurate reflection of the adequacy of corticosteroid replacement than the estimation of cortisol profiles alone, and whether this estimation leads to an improvement in patient management. Hydrocortisone (20 mg) achieved higher mean cortisol levels and lower mean ACTH levels than cortisone acetate (25 mg), but either drug may be suitable for glucocorticoid replacement provided the dose is tailored to the individual needs.