Activity affects intraspecific body-size scaling of metabolic rate in ectothermic animals

Metabolic rate is commonly thought to scale with body mass (M) to the 3/4 power. However, the metabolic scaling exponent (b) may vary with activity state, as has been shown chiefly for interspecific relationships. Here I use a meta-analysis of literature data to test whether b changes with activity level within species of ectothermic animals. Data for 19 species show that b is usually higher during active exercise (mean ± 95% confidence limits = 0.918 ± 0.038) than during rest (0.768 ± 0.069). This significant upward shift in b to near 1 is consistent with the metabolic level boundaries hypothesis, which predicts that maximal metabolic rate during exercise should be chiefly influenced by volume-related muscular power production (scaling as M1). This dependence of b on activity level does not appear to be a simple temperature effect because body temperature in ectotherms changes very little during exercise.