This is a wonderful experiment/demonstration/simulation
and a wonderful fossil, matching a digitized 300 million-year-old Orobates skeleton to Orobates tracks.
This time the problem comes from basal tetrapod experts
One writes in the Dinosaur Mailing List, “As so much else in amniote phylogeny, it remains unclear whether the diadectomorphs are just outside Amniota or just inside. The question has never been tested in an analysis with enough taxa and enough characters; some matrices may have had close to enough of one, but definitely not of both, and there haven’t been many in total in the first place.”
Another comes from SmithsonianMag.com
“At first glance, the 300 million-year-old Orobates pabsti might look like a chunky lizard. In actuality, this animal from the Permian period is what experts know as a stem amniote—a vertebrate that’s part of the evolutionary lineage between amphibians, which reproduce in the water, and the last common ancestor of mammals and reptiles, which lay eggs on land.”
A third comes from the title of the Nature paper (Nyakatura et al. 2019)
“Reverse-engineering the locomotion of a stem amniote.” Clearly the authors have never heard of an amphibian-like reptile.
This is an example of wrong thinking.
The large reptile tree (LRT, 1378 taxa) nests Orobates deep inside the Reptilia (the clade Amniota is a junior synonym, see below). There are those who say the LRT needs more characters to… what? The LRT already lumps and splits virtually all of its taxa in complete resolution documenting a gradual accumulation of traits at every node. That is the yardstick by which all cladograms should be judged. How will adding characters suddenly shift taxa on the tree topology? Taxa nest where they do because they most closely resemble their sisters. And that clade most closely resembles their more distant sisters in a series that ultimately includes all tested taxa. All other possible nesting sites for taxa are provided in a wide gamut analysis.
Apparently basal tetrapod experts are hoping the LRT is somehow wrong. If so, which taxa are wrongly nested and where should they nest instead? They simply need to add the pertinent taxa listed in the LRT to their own analyses with their own character lists to find out for themselves and report the results. This is how science works. Anyone can repeat the experiment, but the experiment in this case, requires pertinent taxa, not more characters.
Case in point:
Remember how Hone and Benton 2007, 2009 deleted pertinent taxa in their purported quest to test two competing hypotheses on pterosaur origins? When they found out the fenestrasaurs were attracting pterosaurs, they deleted the fenestrasaurs and never did find out where pterosaurs originated.
Comment to Nature
“This is a wonderful experiment/demonstration/simulation. The problem comes from the systematics of Orobates. Testing a wide gamut of tetrapod taxa nests Orobates between Limnoscelis and Tseajaia + Tetraceratops. This clade nests between Saurorictus + Captorhinidae and Milleretta, all traditional amniotes. Phylogenetic analysis nests the amphibian-like reptiles Gephyrostegus (Late Carboniferous) and Silvanerpeton (Early Carboniferous) as the last common ancestors of all included amniote taxa. Three nodes of amphibian-like reptiles nest between these two and the Saurorictus + Captorhinidae clade. So Orobates nests well within the Amniota, now a junior synonym for the clade Reptilia. The wide gamut cladogram is found online here: http://reptileevolution.com/reptile-tree.htm”
Nyakatura JA, et al. (11 co-authors) 2019. Reverse-engineering the locomotion of a stem amniote. Nature.com PDF online.
Articles from the popular press: