Estimates of initial splanchnic uptake of ingested glucose and the concomitant suppression of endogenous glucose production were obtained in man by validated tracer techniques for non--steady-state turnover measurement. Nine normal volunteers (18--44 yr old) fasted overnight received intravenous infusions of tracer (3-3H-glucose or 1-14C-glucose) and a low (45 +/- 1 g) or high (96 +/- 5 g) oral load of glucose labeled with an alternative tracer (1-14C-glucose or 2-2H-glucose). A two-compartment model was used to derive rates of peripheral appearance (Ra) of glucose from all sources (total) and the Ra of ingested glucose. Ra (total glucose) and Ra (ingested glucose) were integrated from the first appearance of ingested glucose until the basal Ra (total glucose) or 116 +/- 6 (SEM) mg/min was reattained. The total amount of glucose reaching the systemic pool in this time was 95 +/- 4 g and 46 +/- 3 g with high and low doses, respectively. Of these quantities 86 +/- 4 g and 40 +/- 3 g originated in the oral glucose, representing 90% +/- 4% of the administered glucose. The remainder (11% +/- 2% of the total) represented endogenous production, suppressed by 66% +/- 6% relative to basal. Sequestration of ingested glucose and subsequent release did not take place during the study since identical results were obtained with ingested 1-14C-glucose or 2-3H-glucose. The latter label would have been lost if the glucose had entered the hexose--phosphate pool. Thus, in normal man approximately 90% of an ingested glucose load is absorbed and passes through the liver to appear in the systemic pool.