Daily human energy requirements calculated from separate components of energy expenditure are inaccurate and usually in poor agreement with measured energy intakes. Measurement of energy expenditure over periods of 24 h or longer is needed to determine more accurately rates of daily energy expenditure in humans. We provide a detailed description of a human respiratory chamber and methods used to determine rates of energy expenditure over 24-h periods in 177 subjects. The results show that: fat-free mass (FFM) as estimated by densitometry is the best available determinant of 24-h energy expenditures (24EE) and explains 81% of the variance observed between individuals (24EE [kcal/d] = 597 + 26.5 FFM); 24EE in an individual is very reproducible (coefficient of variation = 2.4%); and even when adjusted for differences in FFM, there is still considerable interperson variability of the daily energy expenditure. A large portion of the variability of 24EE among individuals, independent of differences in body size, was due to variability in the degree of spontaneous physical activity, i.e., "fidgeting," which accounted for 100-800 kcal/d in these subjects.