It is controversial whether food intake in humans is under day-to-day regulation to maintain constant body glycogen stores. In eight white males with a mean (+/- SD) age of 30 +/- 4 y, body weight of 82 +/- 20 kg, and percentage body fat of 22 +/- 5%, exercise and diets were used to produce either high (HG) or low glycogen (LG) stores in a randomized crossover design. After each treatment a vastus lateralis muscle biopsy was obtained. Subsequent ad libitum food intake was measured with an automated food-selection system during 2 d in a respiratory chamber. Despite a 46 +/- 21% difference in muscle glycogen between the two treatments, ad libitum 2-d food intakes (energy, weight, or macronutrients) were similar between treatments (HG: 23.80 +/- 4.67 MJ/d; LG: 21.20 +/- 6.73 MJ/d). However, energy intake on the second day of ad libitum feeding was negatively correlated with carbohydrate balance on the first day, adjusted for the effect of total energy intake and treatment. Adjusted carbohydrate balance on day 1 only explained 9% of the variance in energy intake on day 2. The 24-h respiratory quotient on the first day after treatment was higher after the HG than after the LG treatment: 0.94 +/- 0.04 and 0.88 +/- 0.07 (P < 0.001). The findings suggest that 1) body glycogen stores play at most a minor role in shortterm food intake regulation, and 2) in the short term, imbalances in glycogen stores are corrected by adjustments of macronutrient oxidation rates.