Here flying squirrel mimic Vilevolodon
nests with the Jurassic squirrel-like multituberculate, Shenshou. According to Luo et al., tiny middle ear bones fail to develop here (Fig. 2). Rather these bones remain as post-dentary jaw bones. True, but in the scheme of things, balanced against a larger suite of synapomorphies, this is an atavism (= reversal). Jaw and ankle trait misinterpretations (see below) caused Meng et al. 2017 to consider Vilevolodon a pre-mammal. Rather it is a derived mammal with an odd throw-back ear region for reasons currently unknown.
Where is the jaw joint?
Most basal mammals (and many pre-mammals) have a large dentary bone that includes a 1. large coronoid process, 2. a large articular process and 3. a large retro process posteroventrally. Vilevolodon (Fig. 1) lacks #2, but originally the huge retro process was considered the articular process. Tiny posterior jaw bones were purported to arise from a medio-ventral trough (Figs 2, 3). In all other mammals of this grade (like Morganucodon and Yanoconodon, Fig. 2) the two articular surfaces are next to each other. That’s not the case in the Vilevolodon where the two purported articulations are far apart.
No basal mammals or pre-mammals
have the same tooth count with giant incisors seen in Vilevolodon (Fig. 2. In the large reptile tree (LRT, 1269) Vilevolodon has the opportunity to nest anywhere in the Tetrapoda, and it nests within Rodentia. In many rodents the jaws are free to move on a sliding jaw joint supported by a complex sling of jaw muscles. That seems to be the case here, as well.
Many rodents with sliding jaws
have flat-topped molars and pre-molars for grinding seeds and other plant materials. These taxa have a sliding jaw joint. That’s not the case with Vilevolodon, where the teeth occlude precisely, any sort of jaw joint is absent and the articular surface angles the other way.
(Fig. 4) is similar to modern rats and mice in having flattened teeth and sliding jaw joint. Similar to, yet distinct from those of Vilevolodon where the molars must occlude. Everything else in the mandible has to be engineered so that becomes the end point (Fig. 5).
Then there is the ankle issue.
Meng et al. interpreted a primitive ankle joint (astragalus and calcaneum side-by-side) for Vilevolodon. Reexamination indicates a fairly typical rodent ankle here (Fig. 6; astragalus on top of the calcaneum) simply lacking a large tuber for the calcaneum. We saw a similar error in Maiopatagium earlier here.
I present a possible solution to the procoracoid/coracoid issue, almost following the reconstruction of Meng et al. with modification. In Vilevolodon the procoracoid and coracoid are loose (Fig. 7) and reconstructed like the same bones in a juvenile platypus, many nodes distant. No other related taxa back to prototheres have a procoracoid and coracoid. These few traits appear to be atavisms or reversals because they do not tip the balance enough to nest Vilevolodon in or around prototheres. On that note, no prototheres have a suite of rodent-like traits to match those of rodents.
Reversals like this are real problems for paleontologists.
Without the rest of the skeleton to tip the scores toward rodents, the presence of a procoracoid and coracoid would have been protothere indicators, which is what traditional paleontologists have thought for decades. Don’t pull a Larry Martin! One, two or a dozen traits do not nest taxa. Only a suite of 200+ traits can be trusted to correctly nest taxa, overcoming the phylogenetic problems of convergence, reversal and ‘eyeballing’ phylogeny based on a short list of cherry-picked traits.
Whenever Maiopatagium and Vilevolodon
are someday tested with squirrels and porcupines by other workers, then we’ll see if those two nest basal to Theria or within Rodentia. At present, taxon exclusion, the use of suprageneric taxa and some misinterpretations appear to be problems with prior studies.
We’re all learning as we go.
I better understand things today than yesterday, then sharing those new thoughts with you.
Added six hours later:
The extant aye-aye (genus: Daubentonia), the closest living relative of Vilevolodon, seems to provide insight into the derived/atavistic rodent pectoral girdle and ankle described above. The precise drawing of Daubentonia (Fig. 8) seems to show a procoracoid and coracoid medial to the humerus, as shown above. It also shows an astragalus side-by-side with the calcaneum, except that the tibia and fibula are rotated medially (= dorsal pes rotated laterally), so the astragalus, relative to the tibia, still sits atop the calcaneum. Such solutions must be acknowledged when they appear in nature— or is this just artistic liberty?
Luo Z-X, Meng Q-J, Grossnickle DM, Neander AI, Zhang Y-G and Ji Q 2017. New evidence for mammaliaform ear evolution and feeding adaptation in a Jurassic ecosystem. doi:101.1038/nature 23483\
Meng Q-J, Grossnickle DM, Liu D, Zhang Y-G, Neander AI, Ji Q and Luo Z-X 2017.
New gliding mammaliaforms from the Jurassic. Nature (advance online publication)