To determine the limits to oxidation of exogenous glucose by skeletal muscle, the effects of euglycaemia (plasma glucose 5 mM, ET) and hyperglycaemia (plasma glucose 10 mM, HT) on fuel substrate kinetics were evaluated in 12 trained subjects cycling at 70% of maximal oxygen uptake (VO2,max) for 2 h. During exercise, subjects ingested water labelled with traces of U-C-14-glucose so that the rates of plasma glucose oxidation (R(ox)) could be determined from plasma C-14-glucose and expired (CO2)-C-14 radioactivities, and respiratory gas exchange. Simultaneously, 2-H-3-glucose was infused at a constant rate to estimate rates of endogenous glucose turnover (R(a), while unlabelled glucose (25% dextrose) was infused to maintain plasma glucose concentration at either 5 or 10 mM. During ET, endogenous liver glucose R(a) (total R(a) minus the rate of infusion) declined from 22.4 +/- 4.9 to 6.5 +/- 1.4 mu mol/min per kg fat-free mass [FFM] (P < 0.05) and during HT it was completely sup- pressed. In contrast, R(ox) increased to 152 +/- 21 and 61 +/- 10 mu mol/min per kg FFM at the end of HT and ET respectively (P < 0.05). HT (i, e., plasma glucose 10 mM) and hyperinsulinaemia (24.5 +/- 0.9 mu U/ml) also increased total carbohydrate oxidation from 203 +/- 7 (ET) to 310 +/- 3 mu mol/min per kg FFM (P < 0.0001) and suppressed fat oxidation from 51 +/- 3 (ET) to 18 +/- 2 mu mol/min per kg FFM (P < 0.0001). As the rates of oxidation at more physiological euglycaemic concentrations of glucose were limited to 92 +/- 9 mu mol/l min per kg FFM, and were similar to those reported when carbohydrate is ingested, the results of the current study suggest that the concentrations of glucose and insulin normally present during prolonged, intense exercise may limit the rate of muscle glucose uptake and oxidation.