Table 1.
Model selection based on Akaike Information Criterion (AIC) with clade (teleost or elasmobranch) as a categorical predictor variable
| Model | AIC | ΔAIC |
|---|---|---|
| log10(RMR)~log10(mass)+temp + clade + temp*clade | 37.33 | 0 |
| log10(RMR)~log10(mass)+temp | 38.64 | 1.31 |
| log10(RMR)~log10(mass)+temp+clade+log10(mass)*clade+temp*clade | 38.66 | 1.33 |
| log10(RMR)~log10(mass)+temp+clade | 40.60 | 3.27 |
| log10(RMR)~log10(mass)+temp+clade+log10(mass)*clade | 41.11 | 3.78 |
Regressions were performed with the phylogenetic generalized least squares (PGLS) method. The best model with lowest AIC is denoted by bold.