Rethinking the metabolic allometry of ants

I re-examined the scaling of metabolic rate versus body mass in eight species of ant (static allometry) to illustrate one of the ways by which contemporary concepts of metabolic allometry have been negatively impacted by the widespread use of a standardized procedure for analyzing bivariate data. The procedure in question is the one promoted by Julian Huxley in his monograph on Problems of Relative Growth, and entails back-transforming the equation for a straight line fitted to logarithmic transformations of the original observations to form a two-parameter power function on the arithmetic scale. The scaling exponents for power equations fitted to data by Huxley’s protocol were used by the original authors to evaluate predictions from a theoretical model for the evolutionary optimization of body size in animals. However, my analyses—which were based on a different philosophy and protocol for modeling bivariate data—show that observations for each of the species can be described equally well by two or more different equations with different functional form. Metabolic rate is generally higher in large individuals than in small ones, but little more can be said about metabolic allometry in ants because a clear distinction cannot be made between alternative statistical models that require very different interpretations for biological importance. Conclusions from any investigation that relies exclusively on Huxley’s analytical protocol will be based on incomplete and potentially misleading evidence.