The transport of thiamine across the intestine has been characterized in rats but has not been adequately studied in humans. To determine the kinetics of thiamine intestinal transport directly in humans, mucosal tissues were obtained during routine endoscopy from normal-appearing sites at the second portion of the duodenum. With 3H-dextran as the marker of adherent volume, the uptake of 14C-thiamine hydrochloride by the excised mucosa was measured in vitro. By this method thiamine uptake was linear with tissue weight and with incubation time up to 5 min. Results showed that at low thiamine concentrations (0.2 to 2.0 microM), uptake was saturable whereas at high concentrations (5 to 50 microM), uptake was linear with thiamine concentrations. Pyrithiamine, anoxia, N-ethylmaleimide, and replacement of sodium chloride by mannitol reduced the uptake of 0.5 microM thiamine by 42%, 37%, 32% and 35%, respectively (p less than 0.05) but had no effect on the uptake of 20 microM thiamine. These data suggest that, as in the rat, the intestinal transport of thiamine in humans proceeds by a coexistent dual system. At physiologic concentrations, thiamine is transported primarily by an energy-requiring, sodium-dependent active process, whereas at higher pharmacologic concentrations thiamine uptake is predominantly a passive process.