The insulin hypoglycaemia test has been widely used for assessing the hypothalamic-pituitary-adrenocortical function. The outcome of this test has been compared to that of the 30 min ACTH test in 200 consecutive patients with proven or suspected hypothalamic-pituitary disorder. A significant correlation (R = 0.83, P less than 0.001) between the results of the two tests was found. In patients in whom blood glucose levels fell to 2.2 mmol/l (40 mg/dl) or below, the ratio between the peak plasma cortisol concentration during hypoglycaemia and plasma cortisol concentration 30 min after ACTH was 1.002. In patients in whom blood glucose concentration declined to 2.3-3.0 mmol/l (41-54 mg/dl) the ratio was significantly lower (0.828). In eight patients discordant results between the two tests were found. In two patients the ACTH test was normal despite impaired response to insulin. They were studied shortly after acute symptoms of a pituitary adenoma necrosis. Two patients had subnormal responses to insulin despite the fact that pituitary-adrenal function appeared to be intact. In both cases relative insufficiency of the hypoglycaemic stimulus was the likely cause of the discrepancy. In the remaining four patients results of one test was marginally below, and that of the other test marginally above the lower limit of normal. It is concluded that the 30 min ACTH test is reliable for assessing integrated hypothalamic-pituitary-adrenal function (except shortly after acute ACTH deprivation).