To clarify further the independent relationships of body composition parameters to energy expenditure, resting metabolic rate (RMR) and postprandial thermogenesis were studied in four groups who were matched for absolute fat mass (study 1) and relative fatness (study 2). In study 1, five lean [group A, 15.4 +/- 0.6% (+/- SE) body fat] and five obese men (group B, 25.0 +/- 0.9% fat) were matched on body fat mass (13.0 +/- 0.9 vs. 14.4 +/- 0.8 kg, respectively). Fat-free mass (FFM) and total weight were greater for group A than B. RMR was measured for 3 h in the fasted state and after a 720-kcal mixed meal. RMR was greater for group A than B (1.38 +/- 0.08 vs. 1.14 +/- 0.04 kcal/min, P less than 0.05). The thermic effect of food, calculated as 3 h postprandial minus fasting RMR, was greater for group A than B (65 +/- 6 vs. 23 +/- 9 kcal/3 h; P less than 0.05). In study 2, two groups (n = 6 men/group) were matched for percent body fat (33 +/- 1% fat for both) but differed in lean, fat, and total weights: 50.8 +/- 3.1 kg FFM for the lighter (group C) vs. 68.0 +/- 2.8 kg FFM for the heavier (group D) group, P less than 0.05. RMR was lower for group C than D (1.17 +/- 0.06 vs. 1.33 +/- 0.04 kcal/min, P less than 0.05), but the thermic effect of food was not significantly different (31 +/- 3 vs. 20 +/- 6 kcal/3 h).(ABSTRACT TRUNCATED AT 250 WORDS)