Monday, September 25, 2017

Very Specific Strange Reptiles

Azendohsaurus madagaskarensis by Matt Celeskey.
The term "strange reptiles" could apply to just about every animal I've ever written about on this blog, so you'll forgive me for not loving the name Allokotosauria, an up-and-coming group that was formalized in 2015. The name really says nothing about its members, the similarly newly-minted Azendohsauridae and the longstanding Trilophosauridae. These are archosauromorphs that sit well outside of the Ornithodira-Crurotarsi divide, and are instead related to such eclectic animals as rhynchosaurs and protorosaurs. As I suspect my readers have at least heard of Trilophosaurus, I'll start this essay by discussing Azendohsaurus.

Azendohsaurus laaroussii
Azendohsaurus laaroussii was initially described in 1972 by Dutuit as a Moroccan ornithischian dinosaur—it had those leaf-shaped teeth, after all, but it was quickly moved into the “Prosauropoda” by subsequent studies. There's not much to the holotype--a fragment of dentary and a partial maxilla, both with teeth--but that was enough (in 1972) to warrant a dinosaurian identity. Several decades later, disarticulated postcranial remains discovered at the holotype site suggested that either Azendohsaurus was not a dinosaur at all or that taphonomic processes had mixed the holotype skull—still possibly a dinosaur—and the postcranial remains, which were definitely NOT from a dinosaur. Unambiguously associated cranial and postcranial material would be necessary to determine the true identity of Azendohsaurus.

Well, don’tcha know those associated remains were found in the 1990's in Madagascar of all places, during a field survey. John Flynn and colleagues (2010) discovered a monotypic bonebed of an animal that matched up quite well with the Moroccan holotype skull. The authors named and described their new animal as Azendohsaurus madagaskarensis and determined that it was not a dinosaur—far from it, in fact. Among its obviously non-dinosaurian features include nasal bones that do not contact the premaxillae, making for an undivided naris; an incomplete lower temporal bar; and a multitude of teeth on the vomers, palatines, and pterygoids (and that's just the skull). The authors eventually found it to be outside of Archosauriformes while noting its many features convergent with prosauropods. While a description of the postcranial material was outside the scope of Flynn et al. (2010), that paper did come a few years later.

The skull of Adendohsaurus madagaskarensis
Nesbitt et al. (2015) describes the completely-known skeleton of Azendohsaurus madagaskarensis in an impressive monograph for the Bulletin of the American Museum of Natural History. It paints a picture of a large-bodied quadruped with a long neck and surprisingly robust pectoral girdle. The animal’s fingers are all of similar size and shape as opposed to, say, Trilophosaurus with its fingers of increasing length toward the outer edge of the hand (except the short pinkie). The feet are of similar construction, with large claws on each toe. A phylogenetic analysis placed Azendohsaurus as a sister taxon to a group containing Trilophosaurus, Spinosuchus, and Teratepeton (but read on). India’s suddenly-not-that-weird Pamelaria, described in 2003 as a new kind of protorosaur, turns out to be an outgroup to this dichotomy.

It is this large clade—Pamelaria plus those other guys—that constitutes the Allokotosauria ("strange reptiles"). Azendohsauridae only included one genus (Azendohsaurus) and two species (laaroussii & madagaskarensis) until just recently. 

While I was waiting in the Seattle airport to board my plane to Calgary for SVP this year, a new azendohsaurid was described—Shringasaurus indicus. Another Indian form, Shringasaurus is known from another large monotypic bonebed featuring animals of many different ages. In general, it looks similar to Azendohsaurus with one big caveat: each of its frontal bones bears a robust, forward-curving horn that Sengupta et al. (2017) compares to ceratopsids. You can actually line up frontals from young to old animals and watch the horns grow. Surprisingly, one pair of frontals in the bonebed, while the same size and shape of the horned frontals, completely lacks horns. The authors cite this as evidence of sexual dimorphism. I’m not sure you can make that argument with one data point, but it’s an interesting idea. The authors’ phylogenetic analysis placed Shringasaurus as a sister taxon to Azendohsaurus and Pamelaria moves from being the most basal allokotosaurian to the most basal azendohsaurid (not a big leap).

A full skeletal reconstruction of Azendohsaurus madagaskarensis.
Generally speaking, azendohsaurids were large-bodied quadrupeds with long necks and small heads equipped with spatulate teeth. They look like the group of animals that sauropodomorphs evolved from in some alternate universe. They were around during the Middle-Late Triassic, so they may have lived alongside their dinosaurian counterparts for a little while.

The reconstructed skeletal and selected elements of Shringasaurus indicus.
Geographically speaking, azendohsaurids were, before this year, limited to Gondwannaland: Madagascar, India, and Morocco. Madagascar and India were still joined together during the Triassic, but Morocco was a fair distance away. Sterling Nesbitt gave a talk at SVP this year about Malerisaurus, an old taxon that was originally classified as an "Eosuchian" (and later as a chimaera) that his team recognized as a basal azendohsaurid that’s from the southwestern United States. It was a great talk; there’s a lot of material assigned to Malerisaurus and it apparently survived for a very long time. I got the feeling it was the Vancleavea of allokotosaurs. I eagerly await the paper.

Skeletal reconstruction of Trilophosaurus buettneri

And then there’s the Trilophosauridae. Like the Azendohsauridae, it's generally thought to include only one genus split into two species: Trilophosaurus buettneri and T. jacobsi (but read on). Surprisingly, despite being known since 1928, Trilophosaurus wasn’t fully described until 1945, by Gregory. In 2008, Spielmann et al. penned an extensive monograph for the New Mexico Museum of Natural History and Science Bulletin. They have lots of material to work with—Trilophosaurus is apparently quite common. Two other animals have been flip-flopping around the Trilophosauridae for awhile now—Teraterpeton hrynewichorum and Spinosuchus caseanus. Nesbitt et al. (2015) make the case that Trilophosaurus jacobsi is a junior synonym of Spinosuchus caseanus. While Teraterpeton is fragmentary and its assignment to Trilophosauridae tenuous, Spinosuchus is more likely to be a trilophosaurid (Spielmann et al. 2009).

Trilophosaurus in the treetops by Matt Celeskey
Trilophosaurus itself was an iguana-sized animal that would have looked a lot like an iguana in life and probably spent a lot of time in trees…like an iguana. Its namesake comes from the distinctive teeth that fill its mouth. The teeth are narrow front-to-back but quite wide and only appear in the maxillae (and dentary). Along the width of each tooth are three “peaks” and two “valleys,” giving the animal its name (“three-crested lizard”). Allokotosaurs (azendohsaurids + trilophosaurids) share several features in the forearm and skull, but there was some question (in Nesbitt et al. 2015) about whether Pamelaria is an outgroup to this pairing or not: 

“Although Pamelaria dolichotrachela is identified as the nearest outgroup  of Trilophosauridae + Azendohsauridae, curiously a number of derived character states occur in Azendohsauridae and Pamelaria dolichotrachela to the exclusion of trilophosaurids, apparently homoplastically.”

This ambiguity is seemingly resolved by Sengupta et al. (2017) who find Pamelaria to be the basalmost azendohsaurid, rather than the basalmost allokotosaur.

Pamelaria: weird in 2003, not as weird in 2015.
Azendohsaurs are especially fascinating to me, evolving in the sauropodomorph direction long before sauropodomorphs existed. They were probably the largest herbivores in their ecosystems and could go after food too high for their neighbors. Trilophosaurs are also strange in that they were strongly arboreal--I can't think of any other Middle to Late Triassic reptiles (besides basal pterosaurs) that were spending a lot of time in the trees apart from drepanosaurs--and even they're not a sure thing. Both groups disappeared certainly prior to the Early Jurassic. I imagine azendohsaurs were outcompeted by their dinosaur mimics, the sauropodomorphs, and trilophosaurs may have simply been an unlucky casualty of the Triassic-Jurassic extinction event that cleared the way for dinosaurs to take over.
Shringasaurus indicus by Gabriel Ugueto
Special thanks to Matt Celeskey and Gabriel Ugueto for letting me use their wonderful pieces for this essay!


  1. Awesome article. Big, big fan of the Triassic hellasaur articles but also Triassic in general. Would love to see something on Phytosaurs or perhaps some of the croc-line archosaurs.

    1. Glad you like them! Phytosaurs are definitely on the to-do list, but they will require quite a bit of research.