Results
The regression of maximum lifespan on adult female body mass was highly significant (R2 = 0.59, F = 40.4, d.f. = 1,28, P < 0.0001). The residuals of this analysis were fairly close to a Gaussian distribution (Shapiro-Wilk statistic = 0.98, P = 0.84), and were used as an estimate of mass-independent lifespan. The positive association of maximum lifespan and body mass was strong and significant (r = 0.768, t = 6.4, d.f. = 28, P < 0.0001).
Maximum lifespan and especially body mass were strongly associated with the phylogenetic pattern (respectively; \(\rho\) = 0.453 and 0. 799, [credible intervals] CI0.95 = 0.150-0.932 and 0.382-0.992, N = 30). Mass-predicted lifespan also had a moderate to strong association with phylogeny (\(\rho\) = 0.470, CI0.95 = 0.094-0.897, N = 30). Mass-independent lifespan had only a small association with phylogeny (\(\rho\) = 0.140, CI0.95 = 0.038-0.774, N = 30). Adult telomere length and TROC had trivial to small associations with the phylogeny (respectively;\(\rho\) = 0.070 and 0.059, CI0.95 = 0.015-0.480 and 0.017-0.563, N = 30).
Adult telomere length was not significantly correlated with maximum lifespan, mass-predicted lifespan, or mass-independent lifespan (Figure 2a). Once the phylogenetic pattern was taken into account statistically, adult telomere length exhibited trivial to moderate correlations with different estimates of lifespan (adjusted maximum lifespan, r = -0.119, CI0.95 = -0.650-0.577; adjusted mass-predicted lifespan, r = -0.317, CI0.95 = -0.807-0.457; adjusted mass-independent lifespan, r = 0.309, CI0.95 = -0.536-0.768; all d.f. = 27, P > 0.20). It is noteworthy that the adults of large species tended to have shorter telomeres, but at a moderate correlation at best (unadjusted r = -0.249, t = 1.4, N = 30, P = 0.18; phylogeny adjusted r = -0.234, CI0.95 = -0.608-0.224, d.f. = 27, P > 0.20).
TROC was significantly positively correlated with lifespan (r = 0.497, t = 3.0, d.f. = 28, P < 0.01), and this pattern was similar when the phylogenetic pattern was taken into account (adjusted r = 0.546, CI0.95 = -0.323-0.863, d.f. = 27, P > 0.20) (Figure 2b). Once phylogeny was taken into account, the correlation of TROC and species body mass was trivial to small (adjusted r = -0.126, CI0.95 = -0.529-0.317, d.f. = 27, P > 0.20). TROC exhibited a small to medium positive association with mass-predicted lifespan that was not significant (r = 0.288, t = 1.6, d.f. = 28 P = 0.12), and with phylogenetic “adjustment” this association remained similar (adjusted r = 0.354, CI0.95= -0.649-0.721, d.f. = 27, P > 0.20). TROC and mass-independent lifespan exhibited a significant positive correlation (r = 0.432, t = 2.5, d.f. = 28, P = 0.02), and the effect size of this association was strengthened when the phylogenetic pattern was taken into account statistically (adjusted r = 0.582, CI0.95 = -0.394-0.866, d.f. = 27, P > 0.20).