The in vivo deactivation of insulin action has been studied in 10 lean, nondiabetic subjects using a modification of the euglycemic glucose clamp technique. Following cessation of 40- and 120-mU/m2/min insulin infusions, the serum insulin levels fell to one-half their initial values (mean ± SE) of 126 ± 7 and 350 ± 14 μU/ml in 7 ± 1 and 8 ± 1 min, respectively. The mean incremental glucose disposal rates (IGDR) fell more slowly following discontinuation of the 40- and 120-mU/m2/min insulin infusions, so that the time required for the IGDRs to fall to one-half their initial values (D50 IGDR) were were 42 ± 5 and 78 ± 5 min, respectively. Mean hepatic glucose output was totally suppressed during the 40- and 120-mU/m2/min insulin infusions, remained completely suppressed following cessation of the infusions for 50 and 80 min, and subsequently returned to basal levels. The times required for the HGOs to return to one-half their basal levels (R50 HGO) were 59 ± 8 and 119 ± 6 min, respectively. The times required for insulin action to decrease to one-half the initial values in the periphery (D50 IGDR) and in the liver (R50 HGO) were correlated with the preceding steady-state glucose disposal rates in individual subjects (r = 0.75, P < 0.001 and r = 0.58, P < 0.05, respectively). The suppression of endogenous insulin secretion by exogenous insulin infusions was also studied in 4 subjects during a total of 5 euglycemic glucose clamps; the mean basal serum C-peptide level was 0.67 ± 0.24 pmol/ml before administration of the exogenous insulin, fell to 0.34 ± 0.17 pmol/ml during the steady-state phase of the study, and remained suppressed throughout the duration of the deactivation phase of the glucose clamp. Residual pancreatic insulin secretory capacity was demonstrated by a rise in the serum C-peptide level to 1.77 ± 0.50 pmol/ml at 120 min following a standardized meal given at the conclusion of the deactivation phase of the glucose clamp. These results demonstrate that the deactivation of insulin action in the periphery, liver, and pancreas lags behind the disappearance of insulin from the plasma. The mechanisms responsible for this lag in in vivo deactivation are not known for certain, but may include slower clearance of insulin form tissue compartments than form the plasma, the necessity for the target tissues to generate specific deactivation signals, or a slow rate of decay of saturable steps in the cellular activation process.