I enjoyed reading the preprint and was impressed by the breadth of your analyses across so many eukaryote groups. I agree with your discussion that internal branches could hold some of the signal of rate changes that are most dramatic in the morphological data, but could also be prone to error. For example, looking at the rate changes from lizard to snakes in Fig. 2 it seems that this is where much of the largest rate shifts occur (as expected), and your analyses miss this. But I also do see the difficulty of including this, however I would love to see some of the preliminary anlayses you referred to on this. Maybe there is a way to run a subset of analyses including only the branches that retain confident estimates?

The eukaryote tree in Figure 1 has a few issues that should be fixed, though I’m aware they don’t affect your results. Archosaurs are drawn as sister to mammals rather than within reptiles. The relationship should be mammals sister to archosaurs plus squamates. “Fishes” is also not a monophyletic group: sharks should be sister to ray-finned fishes plus tetrapods. Maybe this grouping was chosen so the fish data sets could be shown together, but I don’t think it’s justified. I would also double-check the placement of Crustacea and Insecta. Given these issues, it would be worth confirming that all analyses are restricted to monophyletic groups. Finally, the figure is a little hard to read where some data sets span higher clades (e.g., Placentalia). Moving or highlighting those labels as differnt would make it clearer that they aren’t collapsed tips.