The effect of meal composition and energy content on the thermic effect of food (TEF) was investigated in sixteen adult, non-obese female subjects. Each subject consumed four different test meals, each meal on a different day. Meals were of high-carbohydrate-low-fat (HCLF) with 0.70, 0.19 and 0.11 of the energy content from carbohydrate, fat and protein respectively, and low-carbohydrate-high-fat (LCHF) with 0.24, 0.65 and 0.11 of the energy content from carbohydrate, fat and protein respectively. The energy contents of the test meals for each composition were 2520 k J (600 kcal) and 5040 k J (1200 kcal). The basal metabolic rate (BMR) and the postprandial metabolic rate (PP-MR) were measured by open-circuit indirect calorimetry using the Douglas bag technique while the subjects were in the supine position. The mean BMR value was 3.63 (SE 0.07) kJ/min (087 kcal/min (SE 0.017)). The 5 h-TEF value for the 2520 k J (600 kcal) HCLF meal was 228 (SE 11.8) k J (54 kcal (SE 2%)) and for the LCHF meal was 228 (SE 9.6) k J (54 kcal (SE 2.3)). The corresponding values for the 5040 k J (1200 kcal) meals were 356 (SE 20.4) k J (85 kcal (SE 4.9)) and 340 (SE 15%) k J (81 kcal (SE 3.8)). There was no significant (P = 049) effect of meal composition on TEF, but the energy content of the meals had a significant (P < 0,001) effect on TEF. In all subjects and for all meals, PP-MR had not returned to premeal level 5 h after a meal, indicating that the TEF values measured underestimate total TEF. The present study suggests that TEF is significantly influenced by the energy content of a meal but not by meal composition.