Testing the aerobic model for the evolution of endothermy: implications of using present correlations to infer past evolution.

The evolution of endothermy is one of the most puzzling events in vertebrate evolution, for which several hypotheses have been proposed. The most accepted model is the aerobic model, which assumes the existence of a genetic correlation between resting metabolic rate (RMR) and maximum aerobic capacity (whose standard measure is maximum metabolic rate, MMR). This model posits that directional selection acted on maximum aerobic capacity and resting metabolic rate increased as a correlated response, in turn increasing body temperature. To test this hypothesis we implemented a simple two-trait quantitative genetic model in which RMR and MMR are initially independent of each other and subject to stabilizing selection to two separate optima. We show mutations that arise that affect both traits can lead to the evolution of a genetic correlation between the traits without any significant shifting of the two trait means. Thus, the presence of a genetic correlation between RMR and MMR in living animals provides no support in and of itself for the past elevation of metabolic rate via selection on aerobic capacity. This result calls into question the testability of the hypothesis that RMR increased as a correlated response to directional selection on MMR, in turn increasing body temperature, using quantitative genetics. Given the difficulty in studying ancient physiological processes, we suggest that approaches such as this model are a valuable alternative for analyzing possible mechanisms of endothermy evolution.