The Harris Benedict equations (HBE) were derived from indirect calorimetric data obtained in 239 normal subjects. Using these data and additional data published by Benedict, which were obtained from subjects spanning a wider age range (n = 98), the present study evaluated the relationship between measured resting energy expenditure and age, sex, and predicted body cell mass (BCM). When the additional subjects from the subsequently published series are included, the regression equations, standard error of the estimate, and 95% confidence limits are similar to the original equations. The HBE estimate resting energy expenditure of a normal subject with a precision of 14%. Resting energy expenditure is directly related to the size of the BCM and is independent of age and sex. The variables of height, weight, age, and sex in the HBE reflect the relationship between body weight and the BCM. Indirect calorimetry and body composition measurements were performed in both normally nourished and malnourished patients (n = 74) to assess the accuracy of the HBE in malnourished patients. Malnutrition is associated with an increase in resting oxygen consumption (VO2) which becomes apparent only when VO2 is expressed as a function of the BCM. There is no difference in resting VO2 between the sexes when expressed as a function of BCM. A regression equation was derived from the Harris Benedict data to predict resting VO2 from age, height, weight, and sex. Predicted VO2 was not significantly different from measured VO2 for the normally nourished patients (n = 33) whereas in the malnourished (n = 41) predicted VO2 underestimated the measured value. The HBE accurately predict resting energy expenditure in normally nourished individuals with a precision of +/- 14%, but are unreliable in the malnourished patient.