Enigmatic Teraterpeton understood, at last, with better data

some photographic skull material has appeared online for Teraterpeton (Fig. 1). Not sure when these first appeared. Could have been years ago. I have not been searching until a day or two ago (see below).

First added
to the large reptile tree (LRT, 1371 taxa) on the basis of drawings by Sues 2003, the long rostrum and antorbital fenestra + the infilling of the lateral temporal fenestra of Teraterpeton  are traits that don’t go together anywhere else on anyone’s cladogram. Sues considered Teraterpeton an archosauromorph nesting with short-snouted Trilophosaurus (Figs. 1, 3).

In the LRT
Trilophosaurus is a rhynchocephalian lepidosaur nesting between derived sphenodontids, like short-snouted Sapheosaurus, and primitive rhynchosaurs, like short-snouted Mesosuchus. All related taxa have a diapsid-like temple architecture, even though the new clade Diapsida (Petrolacosaurus and kin) is restricted to members of the Archosauromorpha in the LRT. Lepidosaurs with a diapsid architecture have their own clade name: “Lepidosauriformes.” Details here and here.

The latest thinking identifies the large hole in the rostrum of Teraterpeton
that extends nearly to the orbit as a naris alone, not a combination of naris + antorbital fenestra. Here (Fig. 1) a broken strut-like bone lying atop the slender maxilla appears to have separated a naris from an antorbital fenestra in vivo. Even so, the present scoring for Teraterpeton with an antorbital fenestra without a fossa does not nest it with other taxa having an antorbital fenestra with or without a fossa.

Figure 1. Skulls of Teraterpeton and Trilophosaurus compared.

Figure 1. Skulls of Teraterpeton and Trilophosaurus compare well aft of the orbit, not so much below the orbit or in the rostrum.

This would not be the first time
an antorbital fenestra appeared in a lepidosaur. Pterosaurs and their ancestors, the fenestrasaurs, also have this trait by convergence with several other tetrapod taxa.

Comparing Trilophosaurus to Teraterpeton
(Fig. 1) is difficult until you get to the postorbital region of the skull. Then it’s a good match. Trilophosaurus has a reduced rostrum, a small naris, a robust maxilla and no hint of an antorbital fenestra. But like Teraterpeton alone, the lateral temporal fenestra found in all related taxa, is infilled with a large flange of the quadrate. Even so, the crappy character list for the LRT is able to nest these two taxa together.

Trilophosaurus and Teraterpeton nest with
the distinctively different Shringasaurus and Azendohsaurus in the LRT. The large variety in their morphologies hints at a huge variation yet to be found here. This is yet one more case where a list of traits may fail you, but a suite of several hundred traits will eliminate all other possibilities by a statistical process known as maximum parsimony. While the parsimony is minimal in this clade, it is still more than any other candidate taxa can offer from a list that has grown to over 1300.

A recent abstract on Teraterpeton
by Pritchard and Sues 2016 bears a review.

“Teraterpeton hrynewichorum, from the Upper Triassic (Carnian) Wolfville Formation of Nova Scotia, is one of the more unusual early archosauromorphs, with an elongate edentulous snout, transversely broadened and cusped teeth, and a closed lateral temporal fenestra. Initial phylogenetic analyses recovered this species as the sister taxon to Trilophosaurus spp. New material of Teraterpeton includes the first-known complete pelvic girdle and hind limbs and the proximal portion of the tail. These bones differ radically from those in Trilophosaurus, and present a striking mosaic of anatomical features for an early saurian.”

I agree completely, which makes solving this mystery so intriguing, and one perfectly suited to the wide gamut of the LRT.

The ilium has an elongate, dorsoventrally tall anterior process similar to that of hyperodapedontine rhynchosaurs

In all cladograms trilophosaurs are close to rhynchosaurs.

The pelvis has a well-developed thyroid fenestra, a feature shared by Tanystropheidae, Kuehneosauridae, and Lepidosauria. 

These taxa all nest within the Lepidosauriformes in the LRT. Mystery solved.

Figure 1. Azendohsaurus skull reconstructed with two premaxillary teeth, not four.

Figure 2. Azendohsaurus skull reconstructed with two premaxillary teeth, not four.

“The calcaneum is ventrally concave, as in Azendohsaurus”. 

Teteraterpeton and Trilophosaurus nest as sisters to Azendohsaurus (Fig. 2)  in the LRT.

The fifth metatarsal is proximodistally short, comparable to the condition in Tanystropheidae.” 

This condition is also found in lepidosaur tritosaur fenestrasaurs, including pterosaurs. Tanystropheidae nest as tritosaurs in the LRT. Mystery solved.

“Much as in the manus, the pedal unguals of Teraterpeton are transversely flattened and dorsoventrally deep.” 

The unguals of Trilophosaurus are also exceptionally deep and transversely flat.

“Phylogenetic analysis of 57 taxa of Permo-Triassic diapsids and 315 characters supports the placement of Teraterpeton as the sister-taxon of Trilophosaurus in a clade that also includes Azendohsauridae and, rather unexpectedly, Kuehneosauridae.”

Add taxa and the unexpected kuehneosaurs will drift to a more basal node.

“The mosaic condition in Teraterpeton underscores the importance of thorough taxon sampling for understanding the dynamics of character change in Triassic reptiles and the use of apomorphies in identifying fragmentary fossils.”

The term ‘mosaic’ is misleading. In the LRT there are no closer sisters to Teraterpeton than Trilophosaurus, and then, rather obviously, the similarities are immediately obvious only in the cheek region, distinct from all other taxa in the LRT.

Figure 2. Trilophosaurus has filled in the lateral temporal fenestra, reduced the orbit and increased the upper temporal fenestra, among other differences with Azendohsaurus.

Figure 3. Trilophosaurus has filled in the lateral temporal fenestra, reduced the orbit and increased the upper temporal fenestra, among other differences with Azendohsaurus.

Forcing Teraterpeton
back to long-snouted clades with an antorbital fenestra, like the Diandongosuchus clade, adds a minimum of 11 extra steps to the LRT.

Key to understanding
the lepidosaur nature of these taxa involves first understanding that the first dichotomy in the clade Reptilia separates the new Archosauromorpha from the new Lepidosauromorpha. Until someone else does this and it becomes consensus, we will continue to experience the confusion exhibited by Pritchard and Sues 2016 (above). This has been documented online for the last seven years.

Pritchard AC, Sues H-D 2016. Mosaic evolution of the early saurian post cranium revealed by the postcranial skeleton of Teraterpeton hrynewichorum (Archosauromorpha, Late Triassic). Abstract from the 2016 meeting of the Society of Vertebrate Paleontology.
Sues H-D 2003. An unusual new archosauromorph reptile from the Upper Triassic Wolfville Formation of Nova Scotia. Canadian. Journal of Earth Science 40(4): 635-649.

Thanks to
reader NP for bringing this taxon back to my attention.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.