DOWNS, C.J.**; HAYES, J.P.; TRACY, C.R.; University of Nevada, Reno; University of Nevada, Reno; University of Nevada, Reno: Temperature, Sampling Bias, and Scaling of BMR in Endotherms
The basal metabolic rate (B) for an organism can be scaled to body mass (M) using the allometric equation, B = cMb, where i is the normalizing constant and b is the scaling exponent. Whether b should be approximately 0.75 or 0.67 is contentiously debated, however metabolism certainly is not related solely to body mass, and different values of the scaling constant may reflect the effect of other (unspecified) variables. We fit a multiple regression to scale B, body temperature (T), and M for birds, for mammals, and for birds and mammals combined. Residuals for the regression of log(B) and log(M) were not normally distributed, but the addition of t as a predictor yielded normally distributed residuals. We regressed M and T onto B for all three data sets using the basic model B = cMbTg, where Tg is either 10T or 101/T. Both Tg variables contributed significantly to the explained variance in B, and both yielded normally distributed residuals for all three data sets. Regressions for all three data sets produced 0.67 as the scaling exponent for body mass, and taxonomic group was not significant when body temperature was included in the model, suggesting T can account for the differences between metabolic rates of mammals and birds. We also tested whether or not sampling biases in the body masses of mammals included in allometric studies of B could result in a scaling exponent of 0.67 for M. We randomly sampled mammals with known metabolic rates in proportion to the body size distribution of all extant mammals. This random sampling was repeated 500 times, and the range of scaling exponents for M never included 0.75, but always included 0.67. We conclude that (1) T is important when scaling B with M, and (2) that B ∝ M0.67.
