Heat tolerance vs taxon age
Heat tolerance was not related to the age of species estimated in Dicket al. (2013) (Fig. 3). Linear regressions between age and T50 and between age and TCrit were not significant (p =0.10 for both), suggesting no gradual trends. Likewise, species with warmer (Miocene) origins did not have systematically higher heat tolerance than species that emerged in the cooler recent (Pleistocene) past (two-sample t-test. T50: t= –1.5, p =0.19. TCrit: t= –1.5, p =0.25). Standardizing T50 and TCrit to sea level values did not change these results. We also tested for trends of heat tolerance over longer timescales, using family age estimates from Wikström et al. (2001), who presented estimates for 29 of the families (representing 96 of the species) in the current study. As was the case for the data in Fig 3, there were no trends in heat tolerance with family ages (Fig. S6).