Food-induced thermogenesis is generally reported to be higher in the morning, although contrasting results exist because of differences in experimental settings related to the preceding fasting, exercise, sleeping and dieting. To definitively answer to this issue, we compared the calorimetric and metabolic responses to identical meals consumed at 0800 hours and at 2000 hours by healthy volunteers, after standardized diet, physical activity, duration of fast and resting. Twenty subjects (age range 20-35 years, body mass index=19-26 kg m(-)(2)) were enrolled to a randomized cross-over trial. They randomly received the same standard meal in the morning and, 7 days after, in the evening, or vice versa. A 30-min basal calorimetry was performed; a further 60-min calorimetry was done 120-min after the beginning of the meal. Blood samples were drawn every 30-min for 180-min. General linear models, adjusted for period and carry-over, were used to evaluate the 'morning effect', that is, the difference of morning delta (after-meal minus fasting values) minus evening delta (after-meal minus fasting values) of the variables. Fasting resting metabolic rate (RMR) did not change from morning to evening; after-meal RMR values were significantly higher after the morning meal (1916; 95% confidence interval (CI)=1792, 2041 vs 1756; 1648, 1863 kcal; P<0.001). RMR was significantly increased after the morning meal (90.5; 95% CI=40.4, 140.6 kcal; P<0.001), whereas differences in areas-under-the-curve for glucose (-1800; -2564,-1036 mg dl(-1) × h, P<0.001), log-insulin (-0.19; -0.30,-0.07 μU ml(-1) × h; P=0.001) and fatty free acid concentrations (-16.1;-30.0,-2.09 mmol l(-1) × h; P=0.024) were significantly lower. Delayed and larger increases in glucose and insulin concentrations were found after the evening meals. The same meal consumed in the evening determined a lower RMR, and increased glycemic/insulinemic responses, suggesting circadian variations in the energy expenditure and metabolic pattern of healthy individuals. The timing of meals should probably be considered when nutritional recommendations are given.