To analyse further the ACTH(1-24) low-dose test, which is of clinical interest, we have examined the dose-response relationship between plasma ACTH(1-24) and cortisol concentrations after i.v. administration of increasing doses (1, 5 or 250 mg) of ACTH(1-24) as a bolus. In addition, we have measured plasma ACTH(1-39) and cortisol levels after an insulin tolerance test (ITT). Although there was a dose-response relationship between plasma ACTH(1-24) immunoreactivity and the dose injected, cortisol peaks were comparable, but lower than those reached after an ITT. Under these experimental conditions, an increase in plasma ACTH as low as 13 pmol/l (i.e. the increase obtained with the 1 mg dose) induced a near maximal cortisol response. Following injection of 1 mg ACTH(1-24), peak ACTH values were short lasting, similar to physiological daily bursts. After injection of 5 mg ACTH(1-24), plasma ACTH concentrations were higher than those reached during an ITT, but clearly shorter lasting. Injection of 250 mg ACTH(1-24) induced strikingly supraphysiological levels of plasma ACTH. We conclude that neither regular nor low-dose ACTH tests can fully reproduce the ITT. Our observations strongly suggest that the low-dose ACTH(1-24) test (1 mg) can be useful to estimate the adrenal sensitivity under basal, physiological conditions.