Background: Two recent studies published in the same issue of Nature reached conflicting conclusions regarding the phylogeny of early mammals: One places the clade containing haramiyidans and multituberculates within the Mammalia and the other separates haramiyidans from multituberculates and places the former outside of the Mammalia. These two contrasting results require that the minimally oldest divergence time of the Mammalia was within the Late Triassic or the Middle Jurassic, respectively. Morphological descriptions of the species named in the two papers were brief, and no comparisons between the newly named species were possible.
Principal findings: Here we present a detailed description of the dentary bone, teeth, occlusal and wear patterns of the haramiyidan Arboroharamiya and compare it with other haramiyidans and Megaconus. Using this new information, we suggest that tooth identifications and orientations of several previously described haramiyidan species are incorrect, and that previous interpretations of haramiyidan occlusal pattern are problematic. We propose that the published upper tooth orientation of Megaconus was problematic and question the number of upper molars, the length of dentition and mandible, and presence of the mandibular middle ear in Megaconus.
Conclusions: The additional morphological descriptions and comparisons presented here further support the view that Arboroharamiya, as a derived haramiyidan, shows similarity to multituberculates in tooth and mandible morphologies. Our comparison also suggests that Megaconus lacks many diagnostic features for the family Eleutherodontidae and that its close affinity with multituberculates cannot be ruled out. The detailed morphological data demonstrate that haramiyidans are more similar to multituberculates than to any other mammaliaforms.