The Trematopids of Pittsburgh

Fresh off a collections visit to the Carnegie Museum in Pittsburgh, it seems only fitting to blog about what I think is one of the greatest strengths of their collection - the Paleozoic tetrapods! The Carnegie owns one of the largest collections of Paleozoic tetrapods from the American southwest (especially New Mexico) and from the more geographically proximal Dunkard Basin (Ohio, Pennsylvania, West Virginia). Much of this was both collected and studied by emeritus curator David Berman and collections manager Amy Henrici, as well as the late Peter Vaughn, a longtime professor at UCLA who taught Dave (among many other paleontologists). Among the stash of early tetrapods are important holotypes of three trematopids, the large dissorophoids with peculiarly long nostrils that I've done some work on myself, so it seems only fitting to do a blog post highlighting a little bit more about these weird animals. 

Ecolsonia cutlerensis​

Ecolsonia is interesting if only because its taxonomy is about as controversial as its gets for a large dissorophoid. The holotype, a partial skull roof was originally a UCLA specimen that was then transferred to the Carnegie when the UCLA collections folded (its Carnegie number rarely appears in publications but is CM 47801) and was originally described by Vaughn in 1969 (collection in 1953). It comes from the Cutler Formation exposures of New Mexico and was named for American paleontologist E.C. Olson (whether this is a very creative or a very un-creative name can be debated). 

Vaughn originally placed it as a trematopid and noted many similarities with dissorophids, but his interpretation was badly biased by a number of things. For one, the small-bodied amphibamiforms ("amphibamids" historically) were lumped together with dissorophids. For another, only one other definitive trematopid was known at the time, Acheloma - "Trematops," recognized then as a distinct trematopid, is now a synonym of Acheloma, and "Trematopsis," a purported third trematopid, is actually a synonym of the dissorophid Cacops aspidephorus. And lastly, Vaughn often cited Longiscitula, a purported dissorophid with trematopid-like features (such as a long nostril), which is in fact a badly squashed and distorted skull of the dissorophid Dissorophus, which looks very different from most trematopids (Milner, 2003). 

Now it's important to note that Vaughn wasn't wrong - dissorophids and trematopids are closely related - they collectively form Olsoniformes - and even though amphibamiforms are not dissorophids, they are the sister group to Olsoniformes. It's sort of like getting the right answer on one of those grade-school math problems but doing it in some weird way that your teacher didn't expect or teach you to do. But realistically, the holotype of Ecolsonia is one of the most incomplete of any trematopid (probably only that of Mattauschia laticeps is more incomplete), and it lacks most of the traditional features used to identify trematopids, such as a definitive exclusion of the palatine from the interpterygoid vacuities or a medial inflection of the prearticular. The nostril is long, but it also seems that long nostrils show up several times outside of trematopids (see Nanobamus, for example).

Skull of the most complete specimen of Ecolsonia from Berman et al. (1985).

At least some of this was ameliorated by the recovery of much more complete material of Ecolsonia that was described by Berman et al. (1985). The complete skull on the left is one of the nicest trematopid specimens out there and is only slightly squashed. It also comes with a hearty serving of postcrania, which is relatively rare for most trematopid species. However, the new specimens also muddled the story further, as Berman et al. reclassified Ecolsonia ​as a dissorophid, an interpretation that again was biased by the relative lack of comparative information available at the time. For example, they cited the structure of the otic notch as being more similar to dissorophids, but the main trematopid comparisons were "Trematops" and Acheloma, which for a time was considered to not even have an otic notch. It's also worth pointing out that the otic notch of Acheloma is pretty unusual among olsoniforms in being very constricted by a massively expanded quadratojugal (a close parallel can be seen in the large dissorophid Anakamacops). A dorsal process that meets the tabular to close the otic notch was also a feature not known in trematopids at the time. Features shared with Acheloma/Trematops, such as a supinator process were interpreted as either primitive for temnospondyls at large or convergence (the long nostril of "Longiscitula" again comes into play here).

Articulated postcrania from the anterior trunk region of a specimen of Ecolsonia. 'os' refers to osteoderms, and the numbers refer to the different types. os(1) are the osteoderms covering the neck region; os(2) are the osteoderms covering the top of the trunk; and os (3) are the osteoderms covering the underside of the trunk (which may be scales; figure from Berman et al., 1985).

The position of Ecolsonia within Dissorophoidea continued to remain unresolved into the 21st century until the development of computer-assisted phylogenetics and the recovery of more trematopid and dissorophid material that revealed that Ecolsonia is probably a trematopid, albeit an unusual one. Of the trematopids, Ecolsonia is the most heavily armoured one, with a series of four osteoderms that would have probably covered the neck and an extensive dorsal and ventral covering totally unlike anything seen in any other olsoniform. Unlike in dissorophids, whose osteoderms are confined to the backbone, it appears that Ecolsonia had ornamented osteoderms covering the entire trunk region, which hypothetically at least would have conferred more of a traditional "armour defence" compared to dissorophids in which the narrow osteoderm covering left a lot of the body uncovered. Probably they helped to stabilize the trunk region a bit as well. 

---

​Anconastes vesperus

Anconastes ("valley inhabitant")  is the second trematopid known from New Mexico, being described shortly after the new material of Ecolsonia by Berman et al. (1987). It comes from a lower stratigraphic horizon in the latest Carboniferous of El Cobre Canyon and has always been regarded as an unequivocal trematopid; arguably it was one of the first taxa to begin to stabilize a broad concept of Trematopidae beyond the clearly derived long-snouted Acheloma and the more perplexing Actiobates, represented only by a relatively small and totally pancake-d holotype. One of the features that probably made a close association to trematopids more identifiable than was possible for Ecolsonia is the apparent constriction of the otic notch in which the ventral margin is nearly horizontal. Having seen the specimen, this characterization is at best a bit generous; some crushing and distortion has obviously occurred that might exaggerate this, and Anconastes differs in many ways from the much larger and long-snouted Acheloma​. Anconastes also benefited from the recognition of Actiobates as a trematopid, as those taxa are represented by smaller, more comparably sized specimens in contrast to the large Ecolsonia and share a lot of features that may at least vary somewhat ontogenetically in trematopids (also ongoing projects for me). 

Like Ecolsonia, Anconastes also comes with a hearty serving of postcrania that includes a dense covering of osteoderms, although they are much smaller than in Ecolsonia. It's interesting to note that it's only the New Mexico trematopids that seem to have these dorsal ossifications...

​Fedexia striegeli

Fedexia is the most recent of the Carnegie trematopids to be described, and it has arguably the weirdest name of the three on account of its provenance. As the story goes, a student from the University of Pittsburgh named Adam Striegel was on a geology field trip back in 2004 near the Pittsburgh Airport where some construction had produced a fresh roadcut of the kind that makes sedimentary geologists salivate. The student picked up a chunk of fairly nondescript rock on which he thought he saw a generic fern fossil and eventually tossed it away. But he ended up picking it up again and showing it to his professor, who pointed out that it was definitely not a fern but rather a skull (the teeth were confused for fern fronds). They ended up showing the fossil to Dave Berman, who immediately recognized it as a trematopid. The name? Derived from the FedEx company on whose land the specimen was collected. You can read a little bit more on the press coverage from the Pittsburgh Post-Gazette here. The genus and species were named in 2010 by a team led by Dave and Amy.

A cast of the holotype skull (background) and a life-sized reconstruction, in Pittsburgh Steelers' colours to boot (foreground) of Fedexia. On display at the Carnegie Museum; image copyright to Bryan Gee.

Scientifically speaking, Fedexia is also a really important taxon for understanding trematopid evolution. It's the second-oldest trematopid in North America and one of the few specimens outside of the Permo-Carboniferous hotspots of New Mexico, Oklahoma, and Texas. As a result, it is rather morphologically distinct from other trematopids, with a number of unusual features such as a long prefrontal process along the top of the nostril, an inverted-heart-shaped nostril (you have to get a little creative to really see the heart), and a smoothly arcing skull roof that has a relatively tall profile. There's no doubt that Fedexia is on the more primitive side of trematopids, but most analyses don't sample all of the known trematopids (for a number of reasons), so it's hard to figure out where it fits in the grand scheme of things (an ongoing project for me). Hopefully more exploration in the eastern half of North America will produce some more Fedexia​-looking trematopids to help resolve some of these relationships.

Trematopids are a pretty cool group that's really growing on me; they don't get quite as much attention as some of the other Paleozoic temnospondyls, probably because quite a lot of them are only known from a single specimen, but whatever that giant nostril was for is a neat anatomical feature (that maybe we will never really understand). Imagine these little guys running around chomping on other tetrapods - somebody should put that in a BBC documentary! Stay tuned for some more trematopid research coming out in 2020!

If you visit the Carnegie (highly recommended), the cast and reconstruction of Fedexia are unfortunately the only trematopids on display, but you can find more trematopids on display at the Field Museum in Chicago!

---

Refs

• Berman DS, Reisz R, Eberth DA. 1985. Ecolsonia cutlerensis, an Early Permian dissorophid amphibian from the Cutler Formation of north-central New Mexico. New Mexico Bureau of Mines and Mineral Resources Circular 191: 1-31. [PDF]
• Berman DS, Reisz RR, Eberth DA. 1987. A new genus and species of trematopid amphibian from the Late Pennsylvanian of north-central New Mexico. Journal of Vertebrate Paleontology 7(3): 252-269. doi: 10.1080/02724634.1987.10011659
• Berman DS, Henrici AC, Brezinski DK, Kollar AD. 2010. A new trematopid amphibian (Temnospondyli: Dissorophoidea) from the Upper Pennsylvanian of western Pennsylvania: earliest record of terrestrial vertebrates responding to a warmer, drier climate. Annals of Carnegie Museum 78(4): 289-319. doi: 10.2992/007.078.0401
• Milner AR. 2003. Longiscitula houghae DeMar, 1966 (Amphibia: Temnospondyli), a junior synonym of Dissorophus multicinctus Cope, 1895. Journal of Vertebrate Paleontology 23(4): 941-944. doi: 10.1671/18