The skull of Nimbacinus dicksoni (Thylacinidae : Marsupialia)

The exceptionally well preserved skull and mandible of the Miocene thylacinid Nimbacinus dicksoni is described. Phylogenetic analysis supports the contention that, within the family, the dentition of N. dicksoni is unspecialised, less derived than the recent Thylacinus cynocephalus for at least 12 features. However, relatively few cranial specialisations evident in T. cynocephalus clearly distinguish it from N. dicksoni. These two taxa share at least three derived cranial features not present in the most generalised thylacinid known from significant cranial material, the late Oligocene Badjcinus turnbulli. On the other hand, where comparison is possible, even the most specialised thylacinid, T. cynocephalus, is plesiomorphic for at least 10 cranial features common to modern dasyurids and five present in the Miocene dasyurid, Barinya wangala. Two character states found in thylacinids are more derived than in B. wangala. Relative to the remaining dasyuromorphian family, Myrmecobiidae, represented by the monotypic Myrmecobius fasciatus, thylacinids are derived for five cranial features and plesiomorphic for five. It appears that despite considerable anatomical diversity among the dentitia of thylacinids and the presence of many highly specialised dental features in some species, the crania of thylacinids have remained remarkably conservative. Even with respect to dentitia, in terms of overall similarity, the Miocene Thylacinus macknessiand late Oligocene material referred to Thylacinus does not differ greatly from the recently extinct T. cynocephalus. It now also seems probable that T. macknessi was also very similar to T. cynocephalus with respect to cranial anatomy. Numerical parsimony analysis incorporating this new material produced moderate bootstrap and Bremer support for a monophyletic Thylacinidae. In this same treatment strict consensus placed Myrmecobius fasciatus as the sister taxon to Thylacinidae–Dasyuridae, but bootstrap and Bremer support was lacking. Both of these results are contra those of the most recent attempt to resolve dasyuromorphian relationships using numerical parsimony and anatomical data. In the present analysis, the early Eocene Australian taxon, Djarthia murgonensis, fell outside a clade inclusive of all other Australian taxa and was monophyletic with the borhyaeniod, Mayulestes ferox. This latter relationship is based on limited material, poorly supported and considered highly unlikely, but it does strengthen the argument that formal placement of D. murgonensis beyond the level of Marsupialia incertae sedisis unwarranted at present.