Behavioral and ecological factors account for variation in the mass-independent energy expenditures of endotherms.

A persistent controversy has concerned the identification of the factors that influence the quantitative variation in the physiological characters of species, an example of which is the basal rate of metabolism of endotherms. The most important factor accounting for its variation is body mass as long as the range in mass is appreciable. But mass never accounts for all of the variation and none if species have the same mass. Most of the residual variation around the mass curve is associated with behavioral characters, ecological factors, and phylogeny, i.e., history. These agents influence energy expenditure by different means and at different stages in the life history of species. Phylogeny describes the historic origin, evolution, and distribution of character states in contemporary species. However, the level of energy expenditure is quantitatively determined by the collective of realized states in combination with conditions in the environment. Therefore, two stages determine energy expenditure: (1) the evolution of character states and (2) their impact in conjunction with conditions on the environment. Behavioral characters and ecological factors, when coupled with log10 mass, usually account for >94 % of the variation in the log10 basal rates of birds and mammals, a capacity not found in phylogenetic analyses. The difficulty of determining a direct impact of phylogeny on physiological characters results from its correlation with behavioral characters. When appropriate, the passerine/non-passerine dichotomy in birds and the sub/infraclass dichotomy in mammals combine with behavioral characters, ecological factors, and log10 mass to increase r (2) to account for 96-99 % of the variation in log10 basal rate. This occurs because dichotomies incorporate factors other than those already in the analyses. The clearest demonstration of the direct impact of character states is the equality of energy expenditure in species convergent with species from other clades without any evidence of the clade to which the species belong. A multifactorial approach depends on the inclusion of life history characteristics of species, but this analysis applies only to continuous, quantitative relationships, which are often modified by discontinuous agents. Distinctive character states distinguish species and therefore must be included in any attempt to account for differences in their level of energy expenditure. No analysis that ignores life history characteristics can account for the quantitative variation in energy expenditure beyond that associated with body mass.