New publication: Redescription of Anaschisma (Temnospondyli: Metoposauridae) from the Late Triassic of Wyoming and the phylogeny of the Metoposauridae (Gee, Parker & Marsh, 2019; Journal of Systematic Palaeontology)

​Title: Redescription of Anaschisma (Temnospondyli: Metoposauridae) from the Late Triassic of Wyoming and the phylogeny of the Metoposauridae
Authors: B.M. Gee, W.G. Parker, A.D. Marsh
Journal: Journal of Systematic Palaeontology

Link to paper - sorry, not open access...email / DM me if you want a copy of the PDF!

Summary for non-scientists
Metoposaurids are my favourite clade of temnospondyls and lived in freshwater environments throughout the Late Triassic, primarily in what was the northern hemisphere then (and still mostly the northern hemisphere now): North America, western Europe, northern Africa, India, and Madagascar. They are extremely abundant fossils such that some metoposaurids are represented by dozens of complete skulls. Much of what we know comes from what we term as 'mass death assemblages' that preserve a large number of individuals. The classic interpretation of the formation of these deposits was a catastrophic drying up of a pond in which metoposaurids live, but other ideas (e.g., transport into a more concentrated deposit) generally hold more traction. Because we have these types of deposits, most research focuses on just those localities, even though metoposaurids can be found in many more places in reduced abundance. One of the areas that is often overlooked is the more northern parts of the US, such as Wyoming. 
   There are a number of metoposaurids from Wyoming, but most of them are considered nomina dubia or junior synonyms of other taxa. One of these is Anaschisma, represented by two species, A. brachygnatha and A. browni. Both come from the same locality and in fact were found on top of each other in the same horizon. This led to much debate in the literature about whether one was just a juvenile of the other. The putative absence of otic notches was also a feature that some considered to indicate close affinities with the much smaller Apachesaurus (which I blogged about a few weeks back). In this study, we re-examined the holotypes of the two Anaschisma species and determined that (1) they are not separate species; (2) the absence of otic notches is just because one of the framing elements is damaged; and (3) that the Wyoming metoposaurids are essentially indistinguishable from the much better-known Koskinonodon perfectus, which is best known from Texas, Arizona, and New Mexico. This isn't particularly surprising because the other Wyoming metoposaurids were previously synonymized with K. perfectus, but it has an important nomenclatural outcome, which is that K. perfectus (named in 1922) becomes a junior synonym of A. browni (described in 1905). This means that for the second time in the 21st century (see Mueller, 2007 for the previous time), the proper name of the most extensively studied North American metoposaurid is changed once again. The Wyoming metoposaurids may be some of the oldest metoposaurids, possibly indicating that A. browni was a relatively long-lived and successful taxon. We also include a lot of discussion points about broader systematic and evolutionary patterns among metoposaurids (e.g., are there too many genera right now).

---

FAQ

​What's the premise?
Way back in 2016 when I was cooking up my first pair of publications on metoposaurids from Petrified Forest with Bill Parker, we wondered how this poorly known metopo from the Late Triassic of Wyoming, Anaschisma, fit into everything. So lest you think that I turn all of my projects into papers in <1 year, this is not one of them! We got a hand from Mike Eklund, who does all kinds of fancy photography, who went to the Field Museum and got some photographs for us (I also visited separately and checked these out while I was supposed to be doing PhD stuff). One of the main problems with a lot of these historic specimens (Anaschisma was named in 1905) is that they did a lot of "cosmetic" reconstruction with plaster, putty, etc. to fill in gaps and oftentimes, entirely reconstruct portions based entirely on their idea of what it should look like. They would even paint and texture it to look like the actual preserved fossil, so it can be pretty hard to tell what's real and what's not. 

A brief chronology
Like most other metoposaurids, Anaschisma has really bounced around quite a lot taxonomically, being synonymized, split back out, repeat many times. A brief history is as follows:

• 1905: Anaschisma named for two skulls (plus other stuff) from the Popo Agie Formation in Wyoming by Branson. Two species are named: A. brachygnatha and A. browni.
• 1908: Moodie suggests that the two species of Anaschisma are just individuals of the same species and of different maturity, hence some of the morphological differences.
• 1929: Branson & Mehl review the North American metoposaurids; they maintain the two species of Anaschisma.
• 1947: Romer suggests that all of the metoposaurids from the Popo Agie (3 genera, 4 species) were all the same genus and might even be the same genus as the better-known Buettneria from the southwestern U.S.
• 1956: Colbert & Imbrie agree with Moodie that the two Anaschisma species are the same as each other and the other Popo Agie metoposaurids; these are collectively called Eupelor browni.
• 1965: Chowdhury agrees with Colbert & Imbrie but puts Eupelor (and every other metoposaurid genus) within Metoposaurus.
• 1980: Gregory suggests (contra previous workers) that the shallow otic notches of Anaschisma are a real biological feature shared with a new diminutive metoposaurid from New Mexico (this would go on to become Apachesaurus​).
• 1993: Hunt's (1993) revision of the Metoposauridae declares Anaschisma to be a nomen dubium because it is not diagnostic. Other Popo Agie metoposaurids are considered junior synonyms of Buettneria perfecta ​from Texas.

So what did you do here?
The classic bread-and-butter anatomical approach of paleontology. Redescribed and figured both Anaschisma browni and A. brachygnatha ​(note: both are pretty reconstructed). Ran a phylogeny. Attempted to diagnose them.

That's....a phylogeny?
It sure is! Looks pretty rough, eh? You get out what you put in, and when you only start with around a dozen parsimony-informative characters, you're probably not going to get much out. We basically got no resolution (metopos are highlighted by the red boxes). Is this a bad result? Bad science? I would argue no, this is exactly what I hypothesized was going to happen. "Negative results" (and you can't even really get "negative results" in a phylogenetic analysis unless you pre-bias your analysis) are still valuable! Getting the same results is still valuable! Just because something is not novel does not mean it shouldn't be published. That was actually a comment we got during review - this isn't novel, so just take it out (good luck to them in trying to convince an editor in a systematics journal to take out a phylogeny). In fact, no one's sampled every recognized metoposaurid in a phylogeny since Hunt's (1993) review, which was done the really old-fashioned way before high-throughput computation was available. A redescription of the Indian metoposaurids came out last year with a phylogeny that I'm a little lukewarm about methodologically (Chakravorti & Sengupta, 2018), but either way, the phylogeny of metoposaurids need more testing. Basically no one includes all metoposaurids in any given analysis. This is that test (well, we cut out Apachesaurus ​b/c of all the work showing it's a juvenile), and it backs up what is often noted qualitatively, which is that metoposaurids all look the same and are very hard to differentiate, even for experts.

What do we get out of this exercise? It's really hard to differentiate North American and European metoposaurids in particular, which is a point that I discuss in the paper as well. Might they really belong to the same genus...? Hmmm... At least the Moroccan ones came out as a cluster; they're probably the easiest ones to differentiate from the whole pack. The other thing we got out of this, which I did not see coming, was that Anaschisma and Koskinonodon, the latter being much better-known from Arizona and New Mexico, cannot be distinguished. Maybe the nostrils are a little bigger in Anaschisma? Not enough to warrant taxonomic separation if you ask us. So we synonymized them! Koskinonodon has been used more often, but this little rule of "which came first" says that the first name is the one with precedent, so Anaschisma (1905), which pre-dates Koskinonodon (1929), is the winner of this synonymy. 

101 problems with metoposaurid taxonomy
It's not secret that metoposaurid systematics are extremely unstable. With this latest paper, the best-known North American metoposaurid has now switched names twice in the 21st century, going from Buettneria (the original designation from 1922) to Koskinonodon (2007) to Anaschisma. This is emblematic of broader splitting and lumping (and repeat a million times) of the clade, which exceeded 20 genera at some point and is now down to just six. The practice of ascribing essentially any perceptible difference to taxonomy (i.e. at least warranting a new species) isn't exclusive to metoposaurids, but it's made very apparent because of how similar they all are to each other compared to equally ranked clades (family-level) of other temnospondyls or tetrapods.

​The taxonomic studies and acts accompanying the various changes to metoposaurids are emblematic of the inherent subjectivity in paleontological taxonomy. With modern species, we have many more avenues through which to evaluate whether two populations of animals are two species or just geographically separated populations of the same species. Put them together and see what happens, to name one example. Of course, in the fossil record, we don't have this option (amongst other things like molecular or soft tissue data). As a result, there is often a tendency to associate extinct tetrapods with the geographic ranges in which they occur and to use this to determine species differentiation. Of course, this underestimates their true geographic range because of gaps in the fossil record; if rocks from a certain time period are not preserved, then animals from that time period that did live there can't be preserved there. For example, in North America, our best records of metoposaurids are from the southwest, but they also occur along the eastern seaboard and then into northern Africa and western Europe. Simply because we don't find them in the middle of the U.S. doesn't mean that they didn't live there - we just don't have records from that time period out that way. Metoposaurid classification has long been plagued by these ideas that they must be stratigraphically, geographically, temporally, etc. restricted. A good example of this is Colbert & Imbrie's (1956) classification scheme (below); they really bought into this subspecies idea (very rarely used in vertebrate paleontology) and used it to split taxa by which geologic formation they occur in. Like I said above, they all look basically the same, so are there actually this many genera or species...? More on that in the future.

Main conclusions?

1. Look at historic specimens.
2. Partially reconstructed specimen =/= undiagnostic.
3. Anaschisma = Koskinonodon and replaces it as the valid name because it came first.
4. I just turned my Twitter handle, Pokemon Go IGN, and countless other internet usernames and profiles into a junior synonym.

Do you ever actually work on your dissertation?
​On occasion.

---

Refs

• ​Branson E.B, Mehl MG. 1929. Triassic amphibians from the Rocky Mountain region. The University of Missouri Studies, 4, 154–253.
• Branson EB. 1905. Structure and relationships of American Labyrinthodontidae. The Journal of Geology, 13, 568–610.
• Chakravorti S, Sengupta DP. 2018. Taxonomy, morphometry and morphospace of cranial bones of Panthasaurus gen. nov. maleriensis from the Late Triassic of India. Journal of Iberian Geology. DOI: 10.1007/s41513-018-0083-1.
• Chowdhury TR. 1965. A new metoposaurid amphibian from the Upper Triassic Maleri Formation of Central India. Philosophical Transactions of the Royal Society of London B: Biological Sciences, 250, 1–52.
• Colbert E, Imbrie J. 1956. Triassic metoposaurid amphibians. Bulletin of the American Museum of Natural History, 110, 399–452.
• Gregory JT. 1980. The otic notch of metoposaurid labyrinthodonts. Pp. 125–135 in L. L. Jacobs (ed) Aspects of Vertebrate History. Museum of Northern Arizona Press, Flagstaff.
• Hunt AP. 1993. Revision of the Metoposauridae (Amphibia: Temnospondyli) and description of a new genus from western North America. Museum of Northern Arizona Bulletin, 59, 67–97.
• Moodie RL. 1908. The lateral line system in extinct Amphibia. Journal of Morphology, 19, 511–540.
• Mueller BD. 2007. Koskinonodon Branson and Mehl, 1929, a replacement name for the preoccupied temnospondyl Buettneria Case, 1922. Journal of Vertebrate Paleontology, 27, 225–225.
• Romer AS. 1947. Review of the Labyrinthodontia. Bulletin of the Museum of Comparative Zoology, 99, 1–368.