Acrobatidae Aplin, 1987

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 : 240-242

Acrobatidae Aplin (in Aplin and Archer), 1987

CONTENTS: Acrobates and Distoechurus (fig. 48).

STEM AGE: 29.0 Mya (95% HPD: 26.4–32.7 Mya).

CROWN AGE: 14.1 Mya (95% HPD: 10.3–18.1 Mya).

UNAMBIGUOUS CRANIODENTAL SYNAPOMORPHIES: Foramen for ramus lateralis of mandibular nerve present at anteromedial end of the glenoid fossa (char. 53: 0→1; ci = 1.000); hypotympanic sinus floor formed by petrosal and “entotympanic-like” ossification (char. 56: 0→2; ci = 0.667); ear canal largely occluded by bony disk (char. 62: 0→1; ci = 1.000); malleo-incudal and stapedio-incudal articulations fused, without sutures (char. 63: 0→1; ci = 1.000); stapedial footplate strongly convex (char. 66: 1→2; ci = 0.286); process of the exoccipital extending anterolaterally to the exit of the facial nerve (stylomastoid notch or foramen) present (char. 80: 0→1; ci = 0.500); squamosal and the pars canalicularis of the petrosal seamlessly fused, with no evidence of a suture between the two bones even in juveniles (char. 88: 0→1; ci = 1.000); diastema between C1 and upper incisor row absent (char. 110: 0→1; ci = 0.167); premolariform second upper premolar (P2) distinctly taller than premolariform P3 (char. 119: 2→0; ci = 0.118); M4 absent (char. 129: 0→1; ci = 0.667); and neomorphic labial cingulum present on M1–3 (char. 133: 0→1; ci = 0.200).

COMMENTS: The distinctiveness of Acrobates and Distoechurus relative to other “possums” was recognized by Aplin and Archer (1983), Archer (1984c), and Baverstock (1984). These two genera were formally referred to their own family, Acrobatidae , by Aplin (in Aplin and Archer, 1987), who identified a long list of distinctive features of the hard and soft tissues shared by Acrobates and Distoechurus . We also find the family Acrobatidae to be characterized by numerous craniodental synapomorphies, including several uniquely derived features of the auditory region and associated basicranial structures.

Aplin and Archer (1987: lix–lx) argued that acrobatids share a number of distinctive morphological synapomorphies with Tarsipes (the only known representative of the family Tarsipedidae ) and that the two families should be placed together in the superfamily Tarsipedoidea , of uncertain relationships to other phalangeridans. However, our molecular (figs. 27–29) and total-evidence (figs. 32, 33) analyses agree with other recent molecular studies in placing Acrobatidae in the superfamily Petauroidea , together with Tarsipedidae , Petauridae , and Pseudocheiridae . A close relationship between Acrobatidae and Tarsipedidae is not supported here or in recent molecular studies ( Phillips and Pratt, 2008; Meredith et al., 2009a, 2009 c, 2011; Mitchell et al., 2014; May-Collado et al., 2015; Duchêne et al., 2018; Álvarez-Carretero et al., 2021); instead, Acrobatidae is consistently found to be the first petauroid family to diverge, with Tarsipedidae sister to Petauridae + Pseudocheiridae . By implication, the putative synapomorphies shared by acrobatids and Tarsipes ( Aplin and Archer, 1987) either evolved convergently in those clades, or they are synapomorphies of Petauroidea that were secondarily reversed in the common ancestor of petaurids and pseudocheirids ( Phillips and Pratt, 2008: 602). Of these two possibilities, the second appears less likely given the more conventionally phalangeridan morphology of petaurids and pseudocheirids compared to acrobatids and Tarsipes ( Aplin and Archer, 1987; Aplin, 1987, 1990).

Fossil acrobatids have been found in late Oligocene to middle Miocene sites at Riversleigh World Heritage Area ( Brammall and Archer, 1999; Long et al., 2002; Archer and Hand, 2006; Black et al., 2012b; Fabian, 2012). Fabian (2012) concluded that these fossils include early members of both the Acrobates and the Distoechurus lineages, which, therefore, must have diverged prior to the late Oligocene, a temporal inference that is congruent with the results of several recent molecular studies ( Meredith et al., 2009a; Mitchell et al., 2014; Duchêne et al., 2018). By contrast, our dated total-evidence analysis ( fig. 33) supports a somewhat younger (Miocene) date for the Acrobates - Distoechurus divergence; if so, then the late Oligocene Riversleigh acrobatids should fall outside the crown clade. However, these potentially important fossils have yet to be formally described.

Also of interest are reports of a currently unnamed “possum” from middle Pleistocene deposits at Mount Etna in Queensland. Hocknull (2005; 2009) considered this mysterious fossil to be superfamily incertae sedis but noted that it has features characteristic of both burramyids and acrobatids. Several authors have concluded that burramyids retain more plesiomorphic craniodental features than other phalangeridans (e.g., Thomas, 1888; Bensley, 1903; Aplin and Archer, 1987; Wroe et al., 1998; but see Winge, 1941; Archer, 1976e), so it is possible that the Mount Etna taxon is a stem acrobatid retaining plesiomorphic similarities to burramyids. However, regardless of its true affinities, the Mount Etna “possum” considerably postdates the inferred origin of the acrobatid crown clade.