Physical training has been shown to improve glucose tolerance and insulin action. In the present study, insulin action was determined using the euglycemic clamp technique in six trained male athletes compared with six untrained controls matched by age, sex, and weight at 14, 38, and 86 hours and at 6 days after cessation of exercise. The rate of insulin-mediated glucose uptake (glucose disposal) was 9.40 +/- 0.46 mg.kg-1.min-1 (mean +/- SEM) for the athletes at 14 h after the last exercise bout, compared with 6.80 +/- 0.86 mg.kg-1.min-1 obtained for the untrained controls (p less than 0.01). Glucose disposal was gradually decreased to 7.78 +/- 0.87 mg.kg-1.min-1 at 38 h, 6.82 +/- 0.49 mg.kg-1.min-1 at 86 h and to 7.11 +/- 1.00 mg.kg-1.min-1 at 6 days after cessation of physical training. At 38 h, 86 h, and 6 days of detraining, glucose disposal exhibited by training athletes did not differ significantly from untrained controls. These results suggest that physical training increases insulin action, and that this effect could be reversed to the control levels within 38 h after detraining.