Nitrite loss from the stomach was studied using dogs equipped with Thomas cannulas for direct access to the stomach lumen. Solutions containing sodium nitrite and non-absorbable volume marker (polyethylene glycol, PEG) were infused into the stomach, and samples were taken over 60 min to determine the concentration of 'total nitrite' (including NO2-, HNO2 and other species in equilibrium with NO2-) and rate of dilution of the stomach contents as a function of time. Changes in stomach volume were also measured. Nitrite loss was found to be very rapid, with total nitrite concentrations declining to less than half the initial levels in 10 min. The decay in total nitrite concentrations was due predominantly to gastric absorption, with small additional contributions from dilution of the stomach contents (inferred from PEG concentrations) and chemical reactions (from in vitro kinetic data). Results for initial nitrite concentrations varying over a range of 0.15-4.5 mM showed absorption to be first order in total nitrite. The permeability-area product for nitrite absorption (PA) was about 0.6 l/h, and was unaffected by the addition of 1 mM SCN- or Cl-. All of these results are consistent with nitrite absorption in the form of NO2- or HNO2. Buffering the infusate with HCO3- to increase luminal pH from approximately 2 to 7 caused a three-fold reduction in the apparent value of PA. When pentagastrin was used to stimulate acid secretion, nitrite absorption was only half as fast as when acid secretion was inhibited with cimetidine, or when no drug was given. This effect could not be explained by variations in luminal pH, and suggests that acid secretion either decreases PA or is accompanied by active secretion of nitrite. Based on these data, a mathematical model was developed to stimulate the physical and chemical factors governing nitrite concentrations in the stomach.