Abstract
Using the data of Brummelkamp for rodents, ungulates, fish, amphibia, that of Dubois for birds, and that of Hrdlicka and Spitzka for Macaca and Sciuridae, it is shown that the allometric formula, y = b x [alpha], is better fitted by minimizing [SIGMA] y2 (log y- log b -[alpha] log x)2 than [SIGMA] (log y-log b -[alpha] log x) 2 or [SIGMA] (y - b x [alpha] )2. On the basis of these computations, the hypotheses that a has an interspecific value of 5/9 or an intraspecific value of 0.26 and that log b is an integral multiple of [image] cannot be accepted.

This publication has 8 references indexed in Scilit: