2022 in review

It was a relatively quiet year on the temnospondyl research front – I think we may be seeing the real effects of the pandemic accumulating now, as people have started running out of leftover projects. By my count, there were just 17 papers either focusing entirely on temnospondyls or with a substantial temnospondyl component, and four of those were published just in the past two weeks! Nonetheless, there was some very exciting work this year, including a disproportionate amount of metoposaurid studies; this seems to be in a trend in recent years, driven almost entirely by teams working on the Polish material, which is a real testament to Krasiejów. I think there is some exciting stuff coming up the pipeline in 2023 (not from me), and I am looking forward to hopefully a more productive year for temnos!

---

Put it on the map

Otero et al. (Journal of South American Earth Sciences) described new non-marine Triassic records from the Atacama Desert of Chile. This includes the first occurrence of temnospondyls from Chile, which has been historically depauperate in Triassic terrestrial records. The small cranial fragment (estimated skull length of just 9 cm) represents most of the right lateral margin of the skull with the tooth row and some palatal bones. It's interesting that although the authors only referred it to Temnospondyli indet., they use the outline of the Late Triassic Argentinean chigutisaurid Pelorocephalus to show what part of the skull the fragment comes from, thereby suggesting that perhaps they believe it to be specifically a brachyopoid of some sorts. Their reticence to formally refer it as such may stem from the purported Middle Triassic age – this is a poorly sampled interval across South America's non-marine deposits (and arguably the least well-sampled globally in the Triassic record of temnospondyls).

Figure 3 from Otero et al., showing the photograph (A) and interpretation (B) of the new cranial fragment; its position within the reconstruction of the chigutisaurid Pelorocephalus (C, D); and the counterpart mold (E) with a tooth (F). Scale bar is 10 mm.

Figure 1 from Wu et al., showing the isolated tooth in multiple anatomical views (A–D); a CT cross-section of the tooth to show infolding (E); and close-up photographs of the external striations (F–G).

Figure 1 from Shi et al., comparing the newly reported Chinese material (D–E) with other capitosauroids (Eryosuchus [A], Cyclotosaurus [B], and Mastodonsaurus [C]). Scale bars equal to 5 cm.

Wu et al. (Vertebrata PalAsiatica) reported a large, isolated tooth from the Tongchuan Formation exposures of northwestern China; the horizon is Middle Triassic in age and thus represents the youngest occurrence of temnospondyls on the North China block.

Shi et al. (Vertebrata PalAsiatica) reported a new Late Triassic non-marine locality from northern China that consists primarily of temnospondyl remains. Among the specimens are a partial interclavicle (shown at right), a partial rib, and various stereospondylous intercentra. The authors compared it favorably to Mastodonsaurus (there is some discrepancy here in their referral to Capitosauroidea but a subsequent remark that this expands the range of Mastodonsaurus). Mastodonsaurus is known only from Europe, but it's worth noting that Cyclotosaurus has been reported from Europe and also occurs in Thailand. However, it's unclear whether the similarity used to argue for fine-scale taxonomy is somewhat exaggerated by the absence of substantive comparative material from what is now eastern Eurasia; the interclavicle is mostly compared to just three of the many capitosaur genera.

Figure 5 from Werneburg et al. showing the holotype in dorsal view (natural cast) with regular photography (A) and x-ray tomography (B).

Figure 10 from Werneburg et al. showing the reconstruction of the skull of Chemnitzion richteri.

Werneburg et al. (PalZ) described a new zatracheid, represented by an essentially complete skeleton, from the early Permian Chemnitz Fossil Lagerstätte in Germany, Chemnitzion richteri. This deposit records the dying moments of a terrestrial environment being buried in ashfall and the subsequent pyroclastic flow of a volcanic eruption, thus preserving a remarkable diversity of organisms (including plants and invertebrates) in 3D natural molds and casts. Despite a decent number of specimens, zatracheids remain relatively rare in the fossil record – both Acanthostomatops vorax (Germany) and Dascyeps bucklandi (England) are only known from a single site, which may reflect the lower frequency of dryland habitats recorded in the Permo-Carboniferous of Europe compared to North America. The postcranial record of the clade is also only known from Acanthostomatops, and the nearly complete skeleton of C. richteri provides further details on the variation within zatracheids.

Figure 4 from Gee & Sidor. Photographs of the partial interclavicle (A–B) and the natural mold (C). Scale bars equal to 1 cm.

One of my postdoc projects with Chris Sidor (Gee & Sidor; Journal of Vertebrate Paleontology) focused on describing some new (and also not so new) material from the Middle Triassic of Antarctica. The informal upper Fremouw Formation has mostly produced remains of very large capitosaurs, which seems to be the result of a high-energy depositional setting that's filtering out anything small. We reported some additional very large lower jaws, some of which approach 1 m in total length, but the real gem of this paper is the partial plagiosaurid interclavicle. This clade is extremely rare in the southern hemisphere for some reason, despite being very abundant in the northern hemisphere. It continues

---

Old friends

Figure 1 from Schoch & Sues, showing photographs (6, 8) and interpretive drawings (7, 9) of juvenile and adult specimens of Parioxys in dorsal view.

Figure 4 from Schoch & Sues, showing reconstructions of the skull roof of juvenile (1) and adult (2) individuals of Parioxys.

Schoch & Sues (Journal of Paleontology) provided a long-awaited reassessment of the enigmatic early Permian temnospondyl Parioxys ferricolus. Despite being known from an appreciable sample size of specimens from Texas, the anatomy and taxonomy of this taxon have long been confusing because the original descriptions were limited to what are now grainy photographs and stylized reconstructions. The original descriptions from nearly 70 years ago were influenced by the  descriptor's (Y. Shawki Moustafa) hypothesis that Parioxys was closely related to Eryops in how comparisons and reconstructions were made, but other workers have long-suspected that the taxon might belong to a different clade. Schoch & Sues' redescription and reassessment provided more compelling evidence for dissorophid affinities, specifically with the cacopines, based on features like a transverse nuchal ridge on the postparietals, a foreshortened posterior skull table, and modified features of the palate.

Figure 9 from Gee & Kufner, showing a photograph (A) and interpretive drawing (B) of a referred specimen of Buettnererpeton bakeri. Scale bar equal to 5 cm.

One of my long overdue projects  that started a few months before the pandemic with Aaron Kufner (Gee & Kufner; PeerJ) was a redescription of 'Metoposaurus' bakeri, the third and most taxonomically ambiguous of the North American metoposaurids. The deeply convoluted history of metoposaurids and their barely differentiated anatomy has led this taxon in particular to be bounced around between genera. We provided what is undoubtedly the longest description I've ever done (which says a lot), included hundreds of photographs, and determined that it should instead be placed in a new genus instead, which we named Buettnererpeton to restore E.C. Case's original honoring of a longtime museum preparator (originally honored in the now defunct Buettneria).

Canada isn't particularly well-known for its temnospondyl fossils, but one of the most iconic Permo-Carboniferous taxa, Dendrerpeton, is well-represented at the famed Joggins Fossil Cliffs locality in Nova Scotia. Having been explored since the mid-19th century, many of the temnospondyl fossils from Joggins have been subjected to repeated taxonomic revision (many junior synonyms), and the oft-flattened specimens do not lend themselves easily to study. As a result, the number of valid genera, and therein the assignment of species, is poorly resolved. However, a robust idea of the taxon is important because Dendrerpeton frequently serves as the outgroup in temnospondyl phylogenetic analyses.
   Arbez et al. (Papers in Palaeontology) CT-scanned and redescribed the cranial anatomy of one of the most complete and least distorted specimens. There remains a paucity of CT data for temnospondyls, and this study provides some of the first CT data on the internal cranial anatomy for the entire clade. Arbez et al. also argued for the synonymy of Dendrysekos with Dendrerpeton (so this specimen would be Dendrerpeton helogenes) and demonstrated the relatively poor support for many basal nodes in temnospondyl phylogeny.

Figure 3 from Arbez et al., showing the dorsal view of the CT-rendered skull. Scale bar equal to 1 cm.

Figure 2 (in part) from Schoch & Mujal, comparing the skull of Trematolestes throughout ontogeny, with increasingly mature specimens to the right. Colors focus specifically on the postfrontal (green) and postorbital (orange) as examples of bones that underwent pronounced changes.

Fresh off the press from last week is a revision of the Middle Triassic trematosaur Trematolestes hagdorni by Schoch & Mujal (Neues Jahrbuch für Geologie und Palaöntologie). When originally described in 2006 by Schoch, the taxon was represented by a number of essentially complete specimens, but these were interpreted as belonging to immature specimens. New material described in this study greatly expands the ontogenetic range on both sides, from highly immature individuals to demonstrably mature adults. The ontogeny of many trematosaurs (and arguably most stereospondyls) remains very poorly known due to a lack of variably sized specimens, and the new material of Trematolestes represents the most completely known ontogeny among Trematosauria.

In the same vein of poorly known early stages of stereospondyl ontogeny, Witzmann & Schoch (Journal of Paleontology) redescribed the holotype and only specimen of Platycepsion wilkinsoni, a diminutive brachyopid that has long been recognized as an immature individual. However, previous descriptions were greatly hindered by limitations on photography (these flat, low-contrast specimens are not easy to work with), and the comparative framework for such immature stereospondyls was very poor. The framework is not substantially improved, but we have a great deal more information on temnospondyl ontogeny in general (and better cameras). Witzmann & Schoch's redescription clarifies some previous interpretations, highlights several informative features, like five pairs of ossified ceratobranchials. The anatomy of this specimen, while being the "same" as was observed by previous authors in the sense that no additional preparation was undertaken, can now be more fully contextualized taxonomically and ontogenetically in the modern framework. The pattern of a well-ossified skull but poorly ossified postcranial skeleton, along with the presence of external gills, links the later diverging stereospondyls with the better-known Paleozoic temnospondyls and provides evidence for a conserved ontogenetic trajectory (head first, everything else later).

---

Under the microscope

Figure 1 from Surmik et al., showing the pathological vertebra (the larger asymmetrical-looking one) and the articulated non-pathological cervical vertebra behind it in various views.

Figure 3 from Surmik et al., showing the histological sections of the pathological vertebra. A, oblique subtransverse section; B, coronal section. C–G represent close-ups of different parts of the section and show the pathologic bone contrasted against the "normal" tissue.

Another hot off the press paper, Surmik et al. (BMC Ecology and Evolution) report the oldest unequivocal occurrence of osteosarcoma (bone tumor) in the vertebra of an "amphibian" (non-amniote). Given the scarcity of pathologies in the fossil record in general, it likely comes as little surprise that this osteosarcoma was identified from the extensive sample of Metoposaurus krasiejowensis from the Late Triassic Krasiejów locality in Poland. Through CT scanning and histology, the authors provided a detailed description of the tissue type and organization, which permitted their diagnosis.
   One of the interesting discussion points raised by the authors is the scarcity of identified occurrences of cancer in fossil "amphibians" given their sample size; indeed, I have never seen such a malformed intercentrum out of perhaps >1,000 that I have seen in North American collections. Modern amphibians have a relatively low rate of reported bone cancer (most occurrences of cancer are reported from the skin), and the authors speculated that temnospondyls may have been similarly resistant to cancer through a variety of developmental and genetic mechanisms.

Figure 3 from Kalita et al., showing the black-and-white images of thin sections across the ontogenetic range of interclavicles of Metoposaurus krasiejowensis. Black represents bone, white represents empty space, and cyan represents taphonomic/artificial cracks. Specimens decrease in maturity to the right.

One of the distinctive features of stereospondyls is their massively enlarged pectoral elements, which form large, plate-like structures. It has often been assumed that such bones would serve as ballast to help these animals sink to the bottom of water bodies; increasing bone weight and/or density is a common feature among many aquatic animals. However, general size is not necessarily reflective of weight (although in fossilized form, it certain does) – bone compactness is the real metric that can be used to assess how a bone contributes to buoyancy. Kalita et al. (Journal of Anatomy) compared the compactness of the clavicle and interclavicle of Cyclotosaurus intermedius and Metoposaurus krasiejowensis and found a high degree of compactness in both taxa. However, differences in microanatomical structure hint at different lifestyles (Metoposaurus more benthic, Cyclotosaurus more active swimming), corroborating other research suggesting interspecific niche partitioning. At least among the several sampled specimens of Metoposaurus, there was no indication of intraspecific niche partitioning, indicating a benthic lifestyle was adopted quite early. 

Figure 6 from Weryński & Kędzierski, showing high-resolution SEM images of cross-sections of metoposaurid teeth, near the base (A) and near the tip (B).

Weryński & Kędzierski (Geological Quarterly) used both histology and SEM to examine the external and internal microstructure of teeth of Metoposaurus krasiejowensis (yes, it's another year filled with research on Polish metoposaurids). In addition to the typical labyrinthodont infolding that has been known for over 150 years in temnospondyls, the authors also identified directional porosity in the canals making up this structure that they interpret as an adaptation for counterbalancing stress forces during biting. Additionally, they identified what they interpret as growth marks within the teeth, equivalent to four seasonal cycles, which adds to the data used to assess the perceived biological response to any local climatic periodicity.

Figure 1 from Bowler et al., showing a schematic of a typical Platyhystrix "hyperelongate spine' (A), the transition zone where a prominent lateral tubercle occurs (B–C), and a parasagittal section spanning this transition zone (D). Scale bars equal to 5 mm.

One of the most iconic temnospondyls is the sail-backed Platyhystrix, known primarily from its greatly elongated neural spines, which are covered in a distinct tubercular texturing. However, it has been questioned whether these are in fact elongated spines, like what is seen in contemporaneous synapsids like Dimetrodon, or whether there might be a dermal contribution, particularly because dissorophids, the group to which Platyhystrix belongs, are characterized by osteoderms associated with their vertebrae.
   Bowler et al. (Journal of Vertebrate Paleontology), in yet another 11th-hour publication, histologically sectioned Platyhystrix material to tackle this question, and found compelling evidence for a dermal-endochondral co-ossification, indicating that the peculiar sail represent is not formed by hyperelongate spines but instead by a "dorsal blade" capping the spine that is likely homologous to the osteoderms of other dissorophids that remain entirely distinct from the underlying spine. This has a number of implications for phylogenetic hypotheses, homology, and homoplasy, but more work on temnospondyl osteoderms is needed to better contextualize this mode of ossification (dermal-endochondral co-ossification is also seen in turtles, for example, which provide little context).

---

Estimating Eryops

Studies seeking to infer the locomotory style of early tetrapods have often turned to salamanders because they're sort of the closest analogue among living tetrapods (four limbs, normal-sized tail, etc.). However, a lot of this has historically been based on what can be politely termed a "best guess" – historical workers could easily conjecture but lacked the tools needed to robustly test these hypotheses. No way to reanimate a skeleton after all. The technology we have now enables us to test a lot of these hypotheses, which is what Herbst et al. (Integrative & Comparative Biology) did in modeling the locomotory style of the famous Permian temnospondyl Eryops.
   Using a newly developed multi-joint pose viability model that allowed them to test entire limb configurations (rather than single joints), the authors tested whether Eryops' skeleton would permit a sprawling hindlimb gait like that observed in the modern-day fire salamander (a stereotypically-built salamander) and found that such a gait was indeed possible in Eryops. This, of course, does not mean that it definitively moved in this way, but it does provide a rigorous test of the plausibility of longstanding analogous comparisons.

Figure 4 from Herbst et al., showing a comparison of the observed live locomotion of the fire salamander hindlimb at different stages of locomotion (A–E) in comparison with the modeled motion of the hindlimb of Eryops using three different knee spacings: tight (G–J), intermediate (K–N), and large (O–R).

Figure 1 from Hart et al., showing skeletons and convex hull reconstructions of the seven sampled taxa. In order from top to bottom: two species of giant salamanders (Andrias), the tiger salamander, the California newt, the salwater crocodile, Paracyclotosaurus, and Eryops.

In the same vein as locomotion, general estimates of body size are another paleobiological attribute that wasn't historically feasible to robustly test. However, like locomotion, body size is quite important for general inferences of ecology, especially for temnospondyls where size is often one of the most variable features compared to amniotes, which, for example, can have highly variable dentition. Hart et al. (Palaeontology) tested a series of body mass estimation models on Eryops and the Triassic capitosaur Paracyclotosaurus, both of which can have the entire skeleton reconstructed and then used for modeling. By applying the same models to living animals whose mass is definitively known, they were able to ground-truth which models have the highest accuracy, while balancing that with computational feasibility and technological complexity. After comparing models, they concluded that Paracyclotosaurus likely had a body mass of between 159 and 365 kg, and Eryops likely had a body mass between 102 and 222 kg (1 kg = 2.2 lbs for Americans), which is, as one article put it, the same size as a pygmy hippo (conveniently, the title omitted 'pygmy').

---

The problem with growing up

It was a light year for phylogenetic work, but one of the more intriguing studies of the year was by Schoch (Journal of Systematic Palaeontology), who examined the phylogeny of amphibamiform dissorophoids known specifically from a range of ontogenetic classes. Small-bodied temnospondyls have often been problematic because of uncertainty over their ontogenetic maturity, and therein, their taxonomy (e.g., relatively mature but small-bodied adult or relatively immature and small-bodied larva). This problem has been particularly acute for dissorophoids, which have a mix of legitimately small-bodied clades (micromelerpetids, amphibamiforms) and larger clades for which small immature specimens are unknown (dissorophids, trematopids).
  Through both morphological comparison and phylogenetic analysis, Schoch provided another test of the longstanding question of "what, if anything, is a branchiosaurid" and what the ontogeny of clades for which early stages are unknown, look like. Based on the results, Schoch argued that branchiosaurids are a legitimate clade (not too controversial these days) and that larval olsoniforms would likely look more like micromelerpetids or early diverging amphibamiforms than like branchiosaurids. The latter result is much more interesting as there is longstanding contention over interpretation of certain branchiosaurid-like individuals as larval olsoniforms, as Schoch's results suggest that those are actually branchiosaurids and that larval olsoniforms remain entirely unknown.

Figure 4 from Schoch, showing morphological changes in the skull of three amphibamiforms, the amphibamid Platyrhinops lyelli, the amphibamid Amphibamus grandiceps, and the branchiosaurid Branchiosaurus salamandroides.

---

Reference list

• Arbez, T., Atkins, J.B. and Maddin, H.C., 2022. Cranial anatomy and systematics of Dendrerpeton cf. helogenes (Tetrapoda, Temnospondyli) from the Pennsylvanian of Joggins, revisited through micro‐CT scanning. Papers in Palaeontology, 8(2):e1421. DOI: 10.1002/spp2.1421
• Bowler, N., Sumida, S.S. and Huttenlocker, A.K. 2022. Histological evidence for dermal-endochondral co-ossification of the dorsal blades in the late Paleozoic amphibian Platyhystrix rugosus (Temnospondyli: Dissorophidae). Journal of Vertebrate Paleontology, e2144338. DOI: 10.1080/02724634.2022.2144338
• Gee, B.M. and Kufner, A.M. 2022. Revision of the Late Triassic metoposaurid “Metoposaurus” bakeri (Amphibia: Temnospondyli) from Texas, USA and a phylogenetic analysis of the Metoposauridae. PeerJ, 10:e14065. DOI: 10.7717/peerj.14065
• Gee, B.M. and Sidor, C.A., 2022. Cold capitosaurs and polar plagiosaurs: new temnospondyl records from the upper Fremouw Formation (Middle Triassic) of Antarctica. Journal of Vertebrate Paleontology, 41(4):e1998086. DOI: 10.1080/02724634.2021.1998086
• Hart, L.J., Campione, N.E. and McCurry, M.R. 2022. On the estimation of body mass in temnospondyls: a case study using the large‐bodied Eryops and Paracyclotosaurus. Palaeontology, 65(6):e12629. DOI: 10.1111/pala.12629
• Herbst, E.C., Manafzadeh, A.R. and Hutchinson, J.R. 2022. Multi-joint analysis of pose viability supports the possibility of salamander-like hindlimb configurations in the Permian tetrapod Eryops megacephalus. Integrative and Comparative Biology, 62(2):139–151. DOI: 10.1093/icb/icac083
• Kalita, S., Teschner, E.M., Sander, P.M. and Konietzko‐Meier, D. 2022. To be or not to be heavier: The role of dermal bones in the buoyancy of the Late Triassic temnospondyl amphibian Metoposaurus krasiejowensis. Journal of Anatomy, 241(6):1459–1476. DOI: 10.1111/joa.13755
• Otero, R.A., Rubilar-Rogers, D., Soto-Acuña, S., Vargas, M.A., Rojas, G.M., Ugalde, R., Rojas, O., Rojas, J. and Novas, F.E. 2022. New records of continental vertebrates from the Triassic of the Atacama Desert, northern Chile. Journal of South American Earth Sciences, 121:104155. DOI: 10.1016/j.jsames.2022.104155
• Schoch, R.R., 2022. Phylogeny of the amphibamiform temnospondyls: the relationship of taxa known by adults, larvae and neotenes. Journal of Systematic Palaeontology, 20(1):2113831. DOI: 10.1080/14772019.2022.2113831
• Schoch, R.R. and Mujal, E. 2022. Ontogeny and adult osteology of the Middle Triassic temnospondyl Trematolestes hagdorni. Neues Jahrbuch für Geologie und Paläontologie-Abhandlungen, 306(3):265–286. DOI: 10.1127/njgpa/2022/1106
• Schoch, R.R. and Sues, H.-D. 2022. The dissorophoid temnospondyl Parioxys ferricolus from the early Permian (Cisuralian) of Texas. Journal of Paleontology, 96(4):1–11. DOI: 10.1017/jpa.2022.10
  
• Shi, Y.-T., Chen, J.-Y. and Liu, J. 2022. A new Late Triassic tetrapod locality from China. Vertebrata PalAsiatica 9 pp. DOI: 10.19615/j.cnki.2096-9899.220818
  
• Surmik, D., Słowiak-Morkovina, J., Szczygielski, T., Kamaszewski, M., Kalita, S., Teschner, E.M., Dróżdż, D., Duda, P., Rothschild, B.M. and Konietzko-Meier, D. 2022. An insight into cancer palaeobiology: does the Mesozoic neoplasm support tissue organization field theory of tumorigenesis?. BMC Ecology and Evolution, 22(1):1–13. DOI: 10.1186/s12862-022-02098-3
  
• Werneburg, R., Witzmann, F., Schneider, J.W. and Rößler, R. 2022. A new basal zatracheid temnospondyl from the early Permian Chemnitz Fossil Lagerstätte, central-east Germany. PalZ, 1-24 pp. DOI: 10.1007/s12542-022-00624-8
• Weryński, Ł. and Kędzierski, M., 2022. Microstructural characteristics and seasonal growth patterns observed in Metoposaurus krasiejowensis teeth. Geological Quarterly, 66(3):1–11. DOI: 10.7306/gq.1658
• Witzmann, F. and Schoch, R.R. 2022. The larval brachyopid Platycepsion wilkinsoni from the Triassic of New South Wales provides insight into the stereospondyl life cycle. Journal of Paleontology, 96(6):1–14. DOI: 10.1017/jpa.2022.57
• Wu, R., Tu, L. and Han, F.-L. 2022. A Temnospondyl tooth from the Middle Triassic of the Ordos Basin, Shaanxi Province. Vertebrata PalAsiatica, 60(1):54–58. DOI: 10.19615/j.cnki.2096-9899.210810