Tracer experiments have been carried out using gorging and nibbling mice to study several related aspects of carbohydrate metabolism: 1) inhibition of gluconeogenesis shortly after animals ingest a glucose-rich meal; 2) the extent to which dietary glucose carbon is recycled by way of 3C compounds after dietary glucose is absorbed; and 3) recycling of glucose by exchange between free and a hypothetical, "bound" glucose pool. Fasted, gorging mice were allowed to eat 120 mg [U-14C,6-T]glucose (58% glucose diet) in 4 min. Plasma glucose-C specific activity rapidly reached that of the dietary glucose-C. Superficially, this suggested nearly complete inhibition of hepatic gluconeogenesis. However, plasma [6-T]glucose, glycogen-[14C, 3H]glucose analyses, and [14C]-glycerol conversion to glucose showed that hepatic gluconeogenesis continued during alimentary hyperglycemia. Half of liver glycogen seemed to be formed from a hepatic G-6-P pool that was never labeled. Indirect kinetic evidence of a large, bound exchangeable glucose pool was presented. Since no direct evidence of such a pool has been obtained, the possibility is raised that a serious artifact of the tracer technique exists or else some unconventional model of carbohydrate metabolism is required to explain our data.