The utility of the doubly labeled water method for the determination of energy expenditure and water output was investigated in humans. Approximately 10 g of 18O and 0.5 g of 2H as water was orally administered to four healthy adults. Total body water was determined from the isotope dilution, and the ensuing 18O and 2H disappearance rates from body water were determined for 13 days by mass spectrometric isotope ratio analysis of the urinary water. During this period, subjects were maintained on a measured diet to determine energy and water intake. The energy expenditure from the doubly labeled water method differed from dietary intake plus change in body composition by an average of 2%, with a coefficient of variation of 6%. The water outputs determined by the two methods differed by 1%, with a coefficient of variation of 7%. The doubly labeled water method is noninvasive, and the subjects could maintain their daily activities without restriction.