Metabolism of Squamate Reptiles: Allometric and Ecological Relationships

We used multiple regression analysis to evaluate the relationship between metabolic rate and three independent variables-mass, temperature, and standard or resting state-for squamate reptiles. For comparisons among adults of different species, mass raised to the .80 power explains 88% of the variation in metabolic rate. (The .80 mass exponent is significantly greater than the .75 predicted by theoretical considerations.) A further 8% of the variation in metabolic rate is explained by body temperature and whether the lizard is in a standard or resting metabolic state. Residuals were used to determine whether metabolic rates varied as a function of phylogenetic relationship or ecological grouping. Familial associations explained 16% of the variation in metabolic rate for varanids, lacertids, iguanids, colubrids, scincids, xantusiids, gekkonids, and boids. More variation (45%) was explained when lizards were partitioned into four ecological categories: day-active predators, hervibores, reclusive predators, and fossorial predators. A single equation relating metabolic rate to mass is thus inappropriate to estimate the metabolism of squamates. For intraspecific comparisons, the mass exponents of the relationship between metabolic rate and mass are significantly lower than .80 for 25 of 28 data sets. Estimating the metabolic rates of juvenile squamates from equations based on comparisons among species is thus invalid. Moreover, there is significant variability among mass exponents among the 14 species that met the statistical requirements for analysis of covariance, and a common mass exponent cannot be assumed for intraspecific comparisons.