Although it has been known for years that the kidneys possess the enzymatic capability to synthesize glucose from gluconeogenic precursors it has generally been felt that they do not contribute to whole body glucose production (GP) after an overnight fast. A recent paper by Cersosimo et al. ( 1 ) challenges this concept and suggests that the renal contribution to GP may be much more significant than previously thought. We recently reported data (2) that agree with those of Cersosimo et al. (1) ( Table I). In both studies (1, 2) catheterization procedures were used to measure net renal balances of labeled and unlabeled glucose. Renal glucose uptake was calculated by dividing the uptake of [3H ]glucose by the arterial [3H]glucose specific activity (SA). Renal GP was calculated as the difference between renal glucose uptake and net renal glucose balance. While the kidneys were taking up glucose in a net sense, they were producing glucose at rates equal to 24 (1 ) and 13% (2) of tracer-determined whole body GP. A similar finding was made in the rat (3). Net renal glucose balance is similar in the dog and human ( 1, 2, 4), but it remains to be determined whether humans, like dogs, exhibit renal GP after a brief fast.Indirect support for a renal contribution to tracer-determined GP comes from an earlier study in which hepatic glucose balance was assessed (2). for -2.6 ,gmol/kg/min of glucose. Given its magnitude, in all likelihood that source was the kidney. The question arises, however, as to whether renal GP calculated as described quantitatively represents "real" GP. There are two technical concerns which must be considered. First, the approach assumes that within the kidney the site of uptake of [3H]glucose precedes the site ofcold glucose entry. Ifthis is not the case then renal uptake would be underestimated. Since the glucose SA drops only slightly (-5%) across the kidney this assumption is of little consequence. Second, it is unclear whether the reduction in the glucose SA which occurs as blood traverses the kidney is the result of the addition of unlabeled glucose or the loss oflabeled glucose in an exchange process. As discussed below this is a more difficult problem.Renal glucose kinetics are unique in that -20% ofthe load of glucose presented to the kidney is filtered at the glomerulus and quantitatively taken into the proximal tubular cells (PTC). Since this cell contains gluconeogenic enzymes (5), has a source of unlabeled glucose-6-phosphate (gluconeogenic precursor uptake), and has a limited but existent glycolytic capacity (5), it is possible that a labeled glucose molecule could exchange with an unlabeled molecule creating "apparent" glucose production. From the perspective of the tracer method it does not matter whether production is "real" or "apparent,"and it must be concluded that the kidneys are responsible for a portion, (albeit small 13-24%) of tracer-determined GP even after a brief fast. This implies that tracer-determined GP cannot simply be taken to be liver derived, and data must be inte...