CONTENTS: † Nimbacinus and Thylacinus (fig. 40).

STEM AGE: 31.2 Mya (95% HPD: 26.6–36.5 Mya).

CROWN AGE: 17.1 Mya (95% HPD: 11.6–24.0 Mya).

UNAMBIGUOUS CRANIODENTAL SYNAPOMORPHIES: Presphenoid exposed in roof of nasopharyngeal fossa above posterior palate (char. 43: 1→0; ci = 0.091) and p3 distinctly taller than p2 (char. 156: 0→2; ci = 0.118).

COMMENTS: Thylacinidae —defined by Kealy and Beck (2017: table 1) as the most inclusive clade including Thylacinus cynocephalus, but excluding Dasyurus viverrinus and Myrmecobius fasciatus —comprises only Thylacinus and † Nimbacinus in our dated total-evidence analysis (fig. 33). As already discussed (see Dasyuromorphia above), † Badjcinus was originally described as a thylacinid by Muirhead and Wroe (1998) and is placed within Thylacinidae in our undated totalevidence anlaysis (fig. 32) but not in our dated total-evidence analysis (fig. 33). † Mutpuracinus archibaldi is another dasyuromorphian that was originally described as a thylacinid (Murray and Megirian, 2000, 2006a), but in our dated totalevidence analysis it is recovered in a clade with another fossil dasyuromorphian, † Barinya wangala, and the only extant myrmecobiid, Myrmecobius; this clade is, in turn, sister to Dasyuridae (fig. 33). † Barinya itself was originally described as the oldest and most plesiomorphic known dasyurid (Wroe, 1999), but this inference is not supported here. Again, our results are broadly congruent with those of Kealy and Beck (2017), which likewise did not support thylacinid affinities for † Mutpuracinus nor dasyurid affinities for † Barinya (see also Eldridge et al., 2019; Rovinsky et al., 2019).

The oldest putative thylacinid remains are a partial upper molar (NTM P2815-10) from the late Oligocene Pwerte Marnte Marnte Local Fauna in the Northern Territory, and a partial lower molar (QM F16809) from the late Oligocene (Faunal Zone A) D-site at Riversleigh that was originally described as a paratype of the thylacinid † Nimbacinus dicksoni by Muirhead and Archer (1990), but which Murray and Megirian (2000: 159) concluded “represents a different taxon, probably another genus,” Wroe and Musser (2001: 502) considered to be of “uncertain” taxonomic status, and Rovinsky et al. (2019) formally classified as Thylacinidae incertae sedis. Neither specimen has been included in published phylogenetic analyses, so even if they really are thylacinids, their relationship to the two thylacinid terminals included here, † Nimbacinus and Thylacinus, is unclear. † Ngamalacinus timmulvaneyi from the early Miocene (Faunal Zone B) sites at Riversleigh (Muirhead, 1997) has likewise not been included here due to its incompleteness, but it was consistently recovered as a thylacinid in the total-evidence analyses of Kealy and Beck (2017), so we consider it to be the oldest definitive thylacinid currently known.

Our estimate for the divergence between Thylacinus and † Nimbacinus is latest Oligocene to middle Miocene, congruent with the age of the oldest known representative of the genus Thylacinus, T. † macknessi, which is from middle Miocene (Faunal Zone C) sites at Riversleigh (Muirhead, 1992; Muirhead and Gillespie, 1995; Yates, 2014, 2015a; Rovinsky et al., 2019).