Tetrapod Zoology

Tetrapod Zoology

Amphibians, reptiles, birds and mammals - living and extinct

The confusing diplospondylous tupilakosaurids


Sherri DeFauw's temnospondyl reconstructions from 1989 (DeFauw 1989). Some of them might be inaccurate (the trematosaurid, for example, should probably be longer-bodied and longer-tailed), but how many times has this sort of thing been done before? Not nearly enough.

Time for a quick look at another temnospondyl group. Today, we focus on the tupilakosaurids, a group of short-limbed, blunt-skulled, long-bodied Permo-Triassic temnos. Ossified ceratobranchials, poorly ossified limbs and long and flexible bodies all suggest that they were fully aquatic though – like some other aquatic temnospondyl groups – their bones lack lateral line sulci. Distinctive features of this group include a unique pterygoid incisure on the palate (a V-shaped notch adjacent to the braincase) and a cranial bone configuration that involves large prefrontals, small postfrontals and large postorbitals. None were very big, with total lengths of 20-40 cm.

Nielsen's reconstructions of different parts of the vertebral column of Tupilakosaurus heilmani (Nielsen 1954). What - neural arch straddling two centra? Yes, it's called diplospondyly.

Fans of palaeoart and of the history of ideas concerning bird origins will be interested to know that Eigil Nielsen named the type species of the group – Tupilakosaurus heilmani­ from the Lower Triassic of Greenland – after artist and author Gerhard Heilmann (Nielsen 1954), best known for his seminal The Origin of Birds. However, Nielsen incorrectly spelt Heilmann with only one ‘n’!

Nielsen (1954) was originally “convinced that [he was dealing with] the earliest Triassic ichthyosaurian so far known” thanks to the short, amphicoelous centra and tall neural arches of Tupilakosaurus. More specifically, the vertebrae of Tupilakosaurus are what we term diplospondylous: this is the weird condition (well, weird to us) where each vertebral ‘segment’ actually consists of two, equally-sized centra, with the neural arch straddling both. However, the discovery during preparation of skull roof elements revealed to Nielsen that he was dealing with a ‘labyrinthodont’ that he considered related to trematosaurs (despite this, much of his 1954 paper on Tupilakosaurus is devoted to a discussion of ichthyosaur origins; had he known of the short-faced, blunt-snouted appearance of the entire skull, he might not have taken the idea of ichthyosaur ancestry that seriously). The sediments that yield Tupilakosaurus are marine in origin and contain ammonoids, apparently showing that this was another temnospondyl clade that dwelt – for at least part of its history – in the sea.

Diplospondylous vertebrae in three tupilakosaurids: anterior is to the left. From Werneburg et al. (2007).

Dvinosauroids or plagiosauroids?

Tupilakosaurids share enlarged retroarticular processes, shallow (or absent) 'tympanic' notches, a distinctive palate where the pterygoids descend ventrally around the sides of the interpterygoid vacuities, and short and poorly ossified limbs with several other temnospondyl groups including the trimerorhachids we looked at previously (see links below) and the dvinosaurids.

Temnospondyl phylogeny from Yates & Warren (2000). You should be able to see that Tupilakosaurus is close to Dvinosaurus, and hence well away from Batrachosuchus and kin (the Brachyopidae, very bottom of diagram).

Shared similarities with trimerorhachids have led some authors to propose a sister-group relationship between trimerorhachids and tupilakosaurids (Milner 1990). However, tupilakosaurids are also linked to dvinosaurids due to their absence of both the lacrimal and supratemporal bones otherwise common to temnospondyls, their possession of a transverse toothrow on the vomers, and other characters (Yates & Warren 2000). Tupilakosaurids are thus part of Dvinosauroidea according to this view.

McHugh (2012), however, recovered the supposed tupilakosaurid Thabanchuia as part of Brachyopidae, and brachyopids are considered part of Trematosauria; Trematosauria is part of the great clade Stereospondyli. Without explaining the full and complex details of the temnospondyl tree, all I need do is tell you that this position places Thabanchuia very far away from dvinosauroids in McHugh's topology. Is it that Thabanchuia is not a tupilakosaurid, or is it that there really are strongly competing views on where tupilakosaurids might fit within temnospondyl phylogeny?

Warren et al. (2011) drew attention to the tupilakosaurid-like, diplospondylous vertebrae of the Tasmanian brachyopid Bothriceps australis. In their phylogenetic analysis, Warren et al. (2011) found dvinosauroids (dvinosaurians of their usage) to be nested within Stereospondyli, but they didn’t support any special relationship between tupilakosaurids and brachyopids. It would seem, therefore, that those distinctive diplospondylous, tupilakosaurid-style vertebrae evolved more than once. Indeed, diplospondyly is not present in all tupilakosaurids: Slaugenhopia from the Lower Permian and Kourerpeton (read on) both lack it, and it may be that it didn’t evolve in the group until the Late Permian (Werneburg et al. 2007, Warren et al. 2011).

Diplospondylous vetrebrae of the brachyopid Bothriceps, from Warren et al. (2011). (B) shows an enlargement of the right half of the neural spine. Scale bar = 10 mm.

Slaugenhopia, Thabanchuia and the ‘barber’s shop reptile’

Until recently, tupilakosaurids were only known from the Lower Triassic of Greenland and Russia. However, the presence of their probable sister-taxon (Dvinosauridae) in the Upper Permian means that their presence is expected in the Permian too. This was confirmed with the 2004 realisation that Slaugenhopia from the Lower Permian of Texas, originally named in 1962, was actually an overlooked tupilakosaurid (Milner & Sequeira 2004). Milner (1990) also reported Upper Permian tupilakosaurid vertebrae from South Africa but (as revealed by A. R. Milner’s hand-written notes on a reprint I own) these turned out to be dicynodont remains. An Upper Permian tupilakosaurid is also known from France (Werneburg et al. 2007). It’s the earliest known diplospondylous member of the group.

Partial vertebral column of a diplospondylous tupilakosaurid from the Upper Permian of France. From Werneburg et al. (2007).

Thabanchuia oomie from the Lower Triassic of South Africa was described as a good member of this clade (Warren 1998), meaning that tupilakosaurids were apparently present on Gondwana (though see above). Warren (1998) suggested that tupilakosaurids survived the end-Permian extinction event in Gondwana, and later re-invaded the north from this southern refugium. Diplospondylous vertebrae from the Lower Triassic of India, suggested as times to be from a tupilakosaurid, might actually be from a diplospondylous brachyopid.

Skull of the South African taxon Thabanchuia (total length c. 25 mm), from Milner & Sequeira (2004).

The Permian tupilakosaurid Kourerpeton was famously ‘discovered’ in the window of a barber’s shop in Arizona, later added to the collections of the University of Arizona, and formally ‘withdrawn’ from the collections in 1967 due to lack of provenance data (the barber was never 100% sure where it had come from and there have been recent suggestions that the specimen may not even have been American). It was later accessioned at the Manitoba Museum of Man and Nature in Winnipeg, and only then named and described: Kourerpeton, meaning ‘barber’s shop reptile’ (Olson & Lammers 1976). The latter authors gave Kourerpeton its own ‘family’, Kourerpetidae (which I think is etymologically incorrect: shouldn’t it have been Kourerpetontidae? I think the authors knew this – they used Kourerpetidae on p. 46 and p. 54 of the article, but Kourerpetontidae in the abstract and on p. 56). Kourerpeton lacks the pterygoid incisure present in other tupilakosaurids, so might be outside the clade that contains other taxa (Milner & Sequeira 2004).

And that… is the end of that; I look forward to much exciting, insightful, controversial tupilakosaurid-themed debate and discussion in the comments below. Many, many more temnospondyls to come.

For previous Tet Zoo articles on temnospondyls, see...

Refs - -

DeFauw, S. L. 1989. Temnospondyl amphibians: a new perspective on the last phases in the evolution of the Labyrinthodontia. Michigan Academician 21, 7-32

Milner, A. R. 1990. The radiations of temnospondyl amphibians. In Taylor, P. D. & Larwood, G. P. (eds) Major Evolutionary Radiations. Systematics Association Special Volume 42. Clarendon Pres (Oxford), pp. 321-349.

- . & Sequeira, S. E. K. 2004. Slaugenhopia texensis (Amphibia: Temnospondyli) from the Permian of Texas is a primitive tupilakosaurid. Journal of Vertebrate Paleontology, 24, 320-325.

McHugh, J. B. 2012. Temnospondyl ontogeny and phylogeny, a window into terrestrial ecosystems during the Permian-Triassic mass extinction. University of Iowa, dissertation (available at

Nielsen, E. 1954. Tupilakosaurus heilmani n. g. et n. sp. an interesting batrachomorph from the Triassic of East Greenland. Meddelelser om Grønland udgivne af Kommissionen for Videnskabelige Undersøgelser I Grønland 72 (8), 1-33.

Olson, E. C. & Lammers, G. E. 1976. A new brachyopoid amphibian. In Churcher, C. S. (ed) Athlon, Essays in Palaeontology in Honour of Loris Shano Russell. Royal Ontario Museum (Toronto), pp. 45-57.

Warren, A. A 1999. Karoo tupilakosaurid: a relict from Gondwana. Transactions of the Royal Society of Edinburgh 89, 145-160.

- ., Rozefelds, A. C. & Bull, S. 2011. Tupilakosaur-like vertebrae in Bothriceps australis, an Australian brachyopid stereospondyl. Journal of Vertebrate Paleontology 31, 738-753.

Werneburg, J., Steyer, S., Sommer, G., Gand, G., Schneider, J. W. & Vianey-Liaud, M. 2007. The earliest tupilakosaurid amphibian with diplospondylous vertebrae from the Late Permian of southern France. Journal of Vertebrate Paleontology 27, 26-30.

Yates, A. M. & Warren, A. A. 2000. The phylogeny of the ‘higher’ temnospondyls (Vertebrata: Choanata) and its implications for the monophyly and origins of the Stereospondyli. Zoological Journal of the Linnean Society 128, 77-121.

The views expressed are those of the author and are not necessarily those of Scientific American.

Share this Article:


You must sign in or register as a member to submit a comment.


Get All-Access Digital + Print >


Email this Article