Bone and fat as a function of body weight in adult mammals

Three independent data sets, for both bone and fat weight, in adult mammals, expressed as a function of body weight, were submitted to linear regression analysis of the log-log transformed data. For land mammals generally, weighing up to 6.6 metric tons, the slope of the best-fit regression line for skeletal weight is 1.073 +/- 0.021. This regression line underestimates skeletal weight in the elephant by about 40%. For cetaceans, varying in body weight from about 0.1 to over 100 metric tons, the slope of the best-fit regression line for skeletal weight is 1.133 +/- 0.044. Since the slopes for these two groups of mammals are not statistically different, and since cetaceans are normally shielded from gravity, due to buoyancy, it is suggested that the slope (1.073) in land mammals may not be an adaption to gravity. After pooling the data from the three data sets for fat, the resultant regression has a slope of 1.146 +/- 0.026. It is argued, on theoretical grounds, that slopes greater than 1.2-1.3 will not be found for the log-log regression of any major tissue on body weight, taken over the whole mammalian weight range.