Agreodontia Beck et al., 2014

Beck, Robin M. D., Voss, Robert S. & Jansa, Sharon A., 2022, Craniodental Morphology And Phylogeny Of Marsupials, Bulletin of the American Museum of Natural History 2022 (457), pp. 1-353 : 206-207

Agreodontia Beck et al., 2014

CONTENTS: Dasyuromorphia , Notoryctemorphia , and Peramelemorphia .

STEM AGE: 48.0 Mya (95% HPD: 44.3–50.9 Mya).

CROWN AGE: 45.7 Mya (95% HPD: 42.3–49.2 Mya).

UNAMBIGUOUS CRANIODENTAL SYNAPOMORPHIES: Median parietal suture at least partially fused in subadults (char. 25: 0→1; ci = 0.143).

COMMENTS: Beck et al. (2014) gave the name Agreodontia to a clade comprising the Australian orders Dasyuromorphia , Peramelemorphia , and Notoryctemorphia , which they defined as the most inclusive clade including Perameles nasuta , Notoryctes typhlops , and Dasyurus maculatus , but excluding Phalanger orientalis ( Beck et al., 2014: 132) . Monophyly of Agreodontia has been consistently supported by analyses of nuclear and combined nuclear and mitochondrial sequence data (e.g., Amrine-Madsen et al., 2003b; Phillips et al., 2006; Beck, 2008a; Meredith et al., 2008b, 2009 c, 2011; Mitchell et al., 2014; Duchêne et al., 2018; Álvarez-Carretero et al., 2021), but it is supported only by mitochondrial sequence data when base composition is corrected for ( Nilsson et al., 2004; Phillips et al., 2006). A single, uncontradicted retrotransposon insertion also supports agreodontian monophyly, but this does not represent statistically significant support ( Nilsson et al., 2010; Gallus et al., 2015a). Recent morphological analyses vary as to whether they recover Agreodontia or not (e.g., Horovitz and Sánchez-Villagra, 2003; Beck et al., 2008 a, 2014; Horovitz et al., 2008, 2009; Carneiro and Oliveira, 2017a; Carneiro et al., 2018; Carneiro, 2019), but this clade has been recovered by most total-evidence analyses ( Beck et al., 2008 a, 2014, 2016; Beck, 2012, 2017a; Maga and Beck, 2017) with the exception of that of Asher et al. (2004).

Our results conform to this general pattern of inconsistent support: whereas our nuclear (fig. 27), combined nuclear and mitochondrial (fig. 29), and total evidence (figs. 32, 33) analyses support monophyly of Agreodontia, our mitochondrial (fig. 28) and morphological (figs. 30, 31) analyses do not. Only partial fusion of the median parietal suture before adulthood optimizes as an unambiguous craniodental synapomorphy of this clade in our dated total-evidence analysis, but this transformation is reversed within Dasyuromorphia .

Putative peramelemorphians have been reported from the earliest Eocene Tingamarra Local Fauna ( Godthelp et al., 1992; Archer et al., 1993a; Muirhead, 1994; Archer et al., 1999; Long et al., 2002). Given the published early Eocene (54.6 Mya) radiometric date for Tingamarra ( Godthelp et al., 1992), this material would represent the oldest known record of the Agreodontian crown clade and would markedly predate our estimate for the most recent common ancestor of Agreodontia (45.7 Mya; 95% HPD: 42.3–49.2 Mya). However, examination of these Tingamarran specimens by R.M.D.B. revealed some similarities to bunodont, nonperamelemorphian metatherians from the Palaeogene of South America and Australia (e.g., Chulpasia , Rosendolops ; see Archer et al., 1993a; Crochet and Sigé, 1993; Goin and Candela, 1996; Sigé et al., 2009), so we are not convinced that they represent peramelemorphians.

A major gap in the Australian fossil record after the early Eocene ( Archer et al., 1999; Long et al., 2002; Woodhead et al., 2014) means that the oldest definitive agreodontians are from the late Oligocene, which is much younger than our estimate for the age of the most recent common ancestor of Agreodontia. Specifically, representatives of Peramelemorphia and Dasyuromorphia are known from multiple late Oligocene sites in central Australia and at Riversleigh World Heritage Area ( Long et al., 2002; Wroe, 2003; Archer et al., 2006; Archer and Hand, 2006; Warburton and Travouillon, 2016; Kealy and Beck, 2017; Eldridge et al., 2019). A single partial upper molar of an alleged thylacinid (NTM P2815- 10; Murray and Megirian, 2006b) and a single upper molar of a probable notoryctemorphian (NTM P2815-6; Murray and Megirian, 2006b; Beck et al., 2014: 151, 2016: 166) are known from the Pwerte Marnte Marnte Local Fauna in the Northern Territory, which is probably slightly older than the central Australian and Riversleigh sites ( Megirian et al., 2010), but which is still late Oligocene.