PCMs and trait correlative analyses
According to PGLS, SVL and LAD are strongly related with variables of PC2 (dominated by precipitation) + PC3 (Mean diurnal range and Precipitation seasonality) + PC4 and PC5 (Min temperature of coldest month). Specifically, our results indicated that the evolution of the climate niche is associated with higher precipitation levels and Min temperature of coldest month, both of which conditions are principally presented in the distribution in E. A and E. B clades, which suggested SVL and LAD may be controlled by Mean diurnal range, precipitation and extreme temperature. In the same vein, HW and SL are controlled by PC1+PC3+PC4+PC5 (P < 0.05), which means temperature contributes most in these traits. PC1+PC2+PC3+PC5 best fitted in ED (P < 0.05) and TW controlled by PC1+PC2+PC3+PC4 (P < 0.05). Interestingly, UEW significant to PC1 (P < 0.05) but selected model by AICc is PC1+PC2+PC3+PC4 (P > 0.05). For other best fitted models, please see Table 6.