Six subjects were infused with [U-13C]glucose (0.03-0.05 mg . kg-1 . min-1) starting 8-9 h after a meal, and the production of glucose, the recycling of glucose (the Cori cycle), the dilution of glucose by unlabeled carbon into the hepatic lactate-pyruvate pool, and gluconeogenesis were determined in these fasted volunteers by use of mass isotopomer analysis and equations previously described [J. A. Tayek and J. Katz. Am. J. Physiol. 272 (Endocrinol. Metab. 35): E476-E484, 1997]. A primed continuous 11-h infusion was started at 6:00 AM, and the above parameters were calculated after 3 h (for the 12-h fast) and at the end of the infusion (for the 20-h fast). Another group of five subjects was fasted for 40 h, and the above parameters were calculated as before. At 12, 20, and 40 h of fasting, respectively, blood glucose was 93 +/- 2, 83 +/- 2, and 71 +/- 2 (SE) mg/dl; glucose production was 2.3, 1.8, and 1.77 mg . kg-1 . min-1; the recycling of labeled carbon was 8, 15, and 15%, and that of glucose molecules (Cori cycle) was 18, 35, and 36%; the contribution of gluconeogenesis to glucose production was 41, 71, and 92% or 0.96, 1.29, and 1.64 mg . kg-1 . min-1; and the contribution of other sources to glucose production was 1.37, 0.53, and 0.15 mg . kg-1 . min-1. The recycling of glucose is important in prolonged fasting for the maintenance of plasma glucose concentration. We demonstrate here that gluconeogenesis can be easily measured and that it accounts for approximately 90% of glucose production after a 40-h fast.