A trace amount (5 mg) of stable isotopically labelled cortisol ([1,1,19,19,19-2H5]cortisol, cortisol-d5) was administered orally to a healthy human subject to examine the pharmacokinetic behaviour of exogenous cortisol and study the interconversion of cortisol to cortisone catalysed by 11 beta-hydroxysteroid dehydrogenase (11 beta-OHSD). The use of a gas chromatographic/mass spectrometric method allowed the simultaneous measurement of the plasma concentrations of endogenous and exogenous cortisol and cortisone. The amount administered was sufficiently small not to disturb or decrease the endogenous cortisol secretion by negative feedback. The appearance of cortisol-d5 and cortisone-d5 was very rapid. In a blood sample taken 10 min after administration of cortisol-d5, plasma concentrations of 154.8 ng ml-1 cortisol-d5 and 2.7 ng ml-1 cortisone-d5 were detected. A peak value of 196.6 ng ml-1 cortisol-d5 was observed 30 min after cortisol-d5 administration. The peak value of cortisone-d5 was 15.5 ng ml-1 40 min after cortisol-d5 administration, and thereafter cortisone-d5 declined in parallel with the slower decline phase of cortisol-d5. The plasma concentrations fell as low as 26.9 ng ml-1 for cortisol-d5 and 4.6 ng ml-1 for cortisone-d5 4 h after cortisol-d5 administration. The ratio of cortisone-d5 to cortisol-d5 approached a plateau 2 h after cortisol-d5 administration, reaching a constant value of approximately 0.15 thereafter. The higher value of the cortisone/cortisol concentration ratio (0.14-0.81; average 0.30) than the cortisone-d5/cortisol-d5 ratio may provide important information about the possibility that cortisone might be formed not only via the pathway of the cortisol conversion but also through other biotransformation processes.