Extended Data Fig. 7: The influence of temperature and depth.

(a) Ultrametric phylogeny of the Ophiuroidea with depth and sea temperature where each sample was found (n = 2699) displayed as coloured rings (27 equal-sized categories, root=265 my). (b-e) Influence of site substitution rate variation aggregated at 3 organisational scales. (b) Phylogenetic Generalized Least Squares (PGLS) analysis showing a weak relationship between sample root-to-tip site substitution path length (RTTPL) from RAxML phylogram and temperature (df=2039, p = 1.4e-4, adjusted R² = 0.018). (c) Regression showing a non-significant relationship between mean RTTPL per biome and mean age of the tip branch on our ultrametric chronogram per biome (df=35, p = 0.6), with point labels indicating A=abyssal, B=bathyal and S=shelf biomes. (d) A 2-factor multiple regression of biome means (df=34, linear regression t-tests, adjusted R² = 0.62) showing a positive relationship of tip age and temperature, but a negative one with RTTPL. (e) Mean sample values aggregated into depth layers showing that the range of mean tip ages is far greater than for RTTPL or the smoothing rate applied to the tip branches.