Abstract
With the data of Harris and Benedict and those of the authors as basis for comparison of accuracy of prediction, the studies revealed no significant difference as between the linear and surface-area formulation. The results were practically identical for the Mayo Foundation data and for those given by Harris and Benedict. The slight differences were preponderantly in favor of the surface-area formulation. In the light of the variation inherent in the data, and within the range of variability of build represented in the normal series, the effect of shape or build of the body on the metabolism, independently of surface area, could not be ascertained statistically. The hypothesis that when other factors are equal, the normal metabolism is proportional to surface area, when this is measured by the formula of DuBois, is consistent with all the observations made.