By now, anyone who thinks mammals were small, shrew like critters in the Mesozoic is deluded. While definitely far less anatomically varied than nowadays, mammals spawned several lineages, from aquatic critters to wolverine sized predators. Some even competed with dinosaurs in several ecological niches, and one of said niches was of small herbivores.
In recent years, several genera that appear to have been ungulate-like have been recovered from the Maastrichtian, and some date as early as the Campanian. The most iconic are the indian Kharmerungulatum and the north american Protungulatum, the latter hailing from Hell Creek and likely having co-existed with Tyrannosaurus rex. These mammals have large, rectangular molars with extensive grooves and jagged cusps that imply an almost certainly herbivorous/omnivorous diet, something that only one other group of Mesozoic mammals, the multituberculates, were previously known to have. Most importantly, they are very similar to the teeth of other ungulates, such as the numerous Palaeocenic “condylarth” clades, which they may or may not be related to.
Where in the tree of life do these mammals fit?
Normally described as “condylarths”, that it an almost misleading classification, as Condylartha is now considered polyphyletic, and most of the caldes within it are now thought to not be especially closely related to each other. However, it’s interesting that Cretaceous ungulates might in fact be closely related, or even ancestral, to major Cenozoic ungulate clades.
Protungulatum is classified as a sister genus to Cetartiodactyla in J. R. Wible, G. W. Rougier, M. J. Novacek and R. J. Asher’s 2007 description. If so, this would mean that Laurasitheria was both existent in the KT event and that was already very diverse, with all modern branches already set. That the earliest Protungulatum fossils date to the Campanian extend this laurasitherian radiation probably as back in time as 80 million years. On the other hand, J. David Archibald, Yue Zhang, Tony Harper and Richard L. Cifelli’s 2011 description renders this dubious at best, and indeed according to them it is probably a non-placental eutherian.
Kharmerungulatum is even less clear, but there have been some suggestions that it is a basal meridiungulate, a logical conclusion based on it’s gondwannan location. It may also be the sister taxa to the mysterious Palaeogene Tingamarra, implying that a radiation of basal meridiungulates or at least basal ungulate-like mammals once occured in eastern Gondwanna.
With the appearence of these mammals, it seems clear that competition with dinosaurs was even less of an issue than previously thought, and given the relative rarity of small ornithopods, it might imply that these mammals arose to fill an ecological void (I should mention that new non-hadrosauroid ornithopods have shown up in the late Cretaceous fossil reccord, though on average they seem to be medium sized animals). Most importantly, it also might mean that modern placental clades were already beginning to get established, or that ungulate-like forms predate the appearence of true placentals.
That Protungulatum survived the KT event seems also to suggest that the previous size limits for it’s survivors are likely incorrect.