Abstract

Macropodids are the most diverse group of marsupial herbivores ever to have evolved. They have been the subject of more phylogenetic studies than any other marsupial family, yet relationships of several key clades remain uncertain. Two important problem areas have been the position of the merrnine (Lagostrophus fasciatus) and the phylogenetic proximity of tree‐kangaroos and rock‐wallabies. Our osteological analysis revealed strong support for a plesiomorphic clade (Lagostrophinae subfam. nov.) containing Lagostrophus and Troposodon, which is likely to have originated in the early Miocene. The extinct short‐faced kangaroos (Sthenurinae) emerged in the middle Miocene as the sister lineage to a clade containing all other living kangaroos and wallabies (Macropodinae). New Guinea forest wallabies (Dorcopsini trib. nov.) are the most plesiomorphic macropodines; the other two main lineages include tree‐kangaroos and rock‐wallabies (Dendrolagini), and ‘true’ kangaroos and wallabies (Macropodini). These phylogenetic outcomes are broadly consistent with the results of recent molecular studies, although conflicts remain over the relative positions of some macropodins (e.g. Setonix, Onychogalea, and Wallabia). Given the presence of derived dendrolagins and macropodins in early Pliocene localities, it is probable that most macropodine genera originated in the late Miocene. Key functional–adaptive trajectories within the craniodental and locomotory systems of the dominant macropodid lineages represent varying responses to the spread of drier, open habitats following the Miocene Climatic Optimum.