Dear Gael,

Thank you very much for your post to this discussion and for raising the point of the effect of including the Late Cretaceous fossil of Mesopachymerus in our study.

First of all, it is important to note that we were not confident to include the supposedly palm bruchid fossil (Poinar, 2005) in our calibrations because our sampling was particularly poor for the Bruchinae, including only two species of what is considered a single genus. Besides, we failed to amplify reliable ribosomal sequences from dried Sagrinae specimens, the likely sister group to the bruchids. These two facts together make it impossible to recognise what breadth of the tree is actually attributable to the bruchid lineage, let alone placing any fossil on the corresponding node or branch.

Most Canadian amber is from the Upper Cretaceous, especifically from the Campanian (83.5 ± 0.7 to 70.6 ± 0.6), although there are claims that some could actually belong into the Tertiary. Assuming that Mesopachymerus is truly Cretaceous, its age could be as "young" as 70 myo, and not necessarily reach the lowest bound of the Campanian period. With the cautioning resulting from the above mentioned limitations for that part of the tree, our bruchid lineage (pink branch) reaches well within this particular age interval (57-76). Therefore I don't understand why it is assumed that including this evidence would "strongly affect" our conclusions. This evidence is arguably consistent with our results already. Marking that particular node at 85 myo (the oldest possible for Canadian amber), would still make the Chrysomelidae 93.5 myo in our tree, way younger than the origin of angiosperms. The only circumstances that this particular fossil evidence would challenge our age estimates would be those making the origin of the Bruchidae, its separation from its sister lineage, very young, e.g. half the width of the linearised tree or younger. But this still remains to be shown...