Sunday, May 31, 2015

Won't somebody please think of the champsosaurs?

Every so often, you may stumble across a crocodile-like thing lurking at the edge of your typical "two-page spread of all kinds of prehistoric critters" found in popular dinosaur books. If the book is of any quality, it includes a legend identifying the denizens, and the crocodile-like thing will be tagged "Champsosaurus" (if there isn't an error in the legend, of course). Unless you're intrigued by the name, or something else tickles your fancy, you'll probably continue along. There are a lot of crocodiles and crocodile-like things to keep track of, after all. What makes this one unusual?

Champsosaurus, despite looking like an unarmored version of the long-snouted fish-eating crocodilians known as gharials, was actually not particularly closely related to crocodiles. It was instead a member of one of the odder clades of reptiles, Choristodera. The choristoderes are now extinct, although this can hardly be attributed to lack of effort: choristoderes are notorious for phasing in and out of the fossil record. This is probably due in part to another of their interesting meta-characteristics: put simply, the choristoderes had an evolutionary habit of copyright infringement, either looking like something that came before or something that would come later. Many choristoderes spent decades labeled as members of other groups. Some of them looked more or less like lizards: Cteniogenys of the Jurassic and Cretaceous of North America and Europe, and Lazarrusuchus of the Paleocene to Miocene of Europe, are examples of this route. Some of them looked like nothosaurs, the paddle-less cousins of plesiosaurs: Hyphalosaurus of the Early Cretaceous of China is the best-known exponent of this route. Monjourosuchus of the Early Cretaceous of east Asia was a bit like a lizard and a bit like a tuatara, with a hint of salamander. Champsosaurus, as noted, went with an "unarmored gharial" look. Simoedosaurus of the Paleocene and Eocene of Europe and North America was similar, but shorter-snouted.

Champsosaurus laramiensis on display at the Science Museum of Minnesota in a relatively neutral floating posture. The Science Museum is one of the few places where you can see multiple champsosaur skeletons on display.

At this time, there are more than a dozen genera of choristoderes spread from the Triassic to the Miocene, from east Asia to Europe to western North America. They accomplished the difficult trick of surviving the end-Cretaceous extinction, although given they apparently possessed knowledge of extradimensional travel they probably cheated somehow. Choristoderes were generally small, on the order of a meter or two long for most species; Champsosaurus itself was the "champ", with the common Paleocene species C. gigas reaching more than three meters. Most are thought to have had piscivorous and/or insectivorous, depending on a species' size and aquatic capabilities. What they came from is uncertain, as is when they disappeared. Evans and Klembara (2005), in their description of Lazarussuchus dvoraki from the Early Miocene of the Czech Republic, speculated that, based on the climate, similar lizard-like choristoderes could have persisted into the Late Pliocene of Europe, when cooling became more pronounced. This is only a few million years ago, meaning we just missed out. (Or perhaps they're still out there—they become so much easier to understand if you assume extradimensional travel. All that hopping between time periods exposed them to strange radiation that caused mutations—heck, we even have a two-headed Hyphalosaurus!)

Champsosaurus itself was a common reptile of the Late Cretaceous and Paleocene of western North America, with several species named. Although choristoderes as a whole are rather under-represented in the literature, Champsosaurus itself has been the beneficiary of several lengthy publications (Brown 1905; Erickson 1972, 1985; Gao and Brinkman 2005; Parks 1927, 1933; Russell 1956), so its bony anatomy is pretty well understood. The skull features a very long and narrow snout with numerous conical teeth and the nostril opening at the tip, facing forward. An additional "shagreen" of teeth coats the palate. Posterior to the long snout are the orbits, which face up, anteriorly, and laterally, and permitted some degree of binocular vision (Erickson 1985). The posterior part of the skull is made up of several bony arches and flares out quite broadly, making it clear that you are not looking at just another fish-eating crocodile. The torso is broad laterally and shallow vertically, with thick short ribs and a remarkable "cuirass" of gastralia ("belly ribs"). Champsosaur vertebrae have long been noted as common and easily recognizable fossils, usually consisting of just the centrum (the main body of the vertebra without the various processes) with characteristic flat surfaces on both sides where it would articulate with other vertebrae. The shoulder and pelvic bones are not large, but are solidly built. The hind limbs are longer than the fore limbs, and both the hands and the feet have five long fingers or toes. Skin impressions from several areas of the body show that instead of scutes or large scales, the surface featured sub-millimeter pustule- and rhomboid-shaped scales (Erickson 1985).

A larger Champsosaurus at the Science Museum Note the thin snout and flared posterior arches of the skull, and the broad torso.

Although Champsosaurus has often been compared to the gharial, there are several notable differences. The most obvious are the lack of scutes and the flared skull arches, but the most significant function difference has to do with the placement of the nostrils. Crocodilian nostrils face up, allowing crocs to cruise at the surface of water bodies. Champsosaur nostrils face forward, which doesn't work quite as well when you're cruising at the surface. Forward-facing nostrils at the tip of a long snout do, however, make a good snorkel. Therefore, it is more probable that champsosaurs lived much of their lives submerged, only poking the tips of their snouts to breath as necessary, as discussed in Erickson (1972, 1985). A submerged lifestyle also implies that champsosaurs detected prey and predators mostly by vision, with little use of scent (if you can only smell when your nose is out of the water, but you spend all your time under the water, smell can only help so much), and that they would have preferred water bodies that allowed them to be submerged. Hunting may have been done via short quick thrusts and lunges from lying in wait, using the hind limbs for propulsion.

Another SMM champsosaur in snorkeling aspect, this one permitting a partial view of the gastralia.

Champsosaurus is well-represented in Campanian, Maastrichtian, and Paleocene rocks of western North America, including such favorites as the Judith River Formation, Dinosaur Park Formation, and Hell Creek Formation. It just barely made into the Eocene, though; it shows up in the Willwood Formation and then goes "pfft". It's interesting that champsosaurs did not make it into the Green River Formation, deposited in long-lived Eocene lakes in areas where champsosaurs once were common. The Paleocene Wannagan Creek lake fauna, which hosted C. gigas, also hosted the large near-crocodile Borealosuchus (ex Leidyosuchus) formidabilis and the house-pet size alligator Wannaganosuchus, as well as turtles, lizards, and snakes (Erickson 1999). A few million years later, the Green River Formation hosted another species of Borealosuchus (B. wilsoni), another small alligator (Tsoabichi greenriverensis), and turtles, lizards, and snakes (Grande 2013), but no champsosaurs.


Brown, B. 1905. The osteology of Champsosaurus Cope. American Museum of Natural History Memoirs 9(1):1-26.

Erickson, B. R. 1972. The lepidosaurian reptile Champsosaurus in North America. Science Museum of Minnesota, Monograph 1, Paleontology: 1-91.

Erickson, B. R. 1985. Aspects of some anatomical structures of Champsosaurus (Reptilia: Eosuchia). Journal of Paleontology 5(2):111–127.

Erickson, B. R. 1999. Fossil Lake Wannagan (Paleocene: Tiffanian), Billings County, North Dakota. North Dakota Geological Survey, Bismark, North Dakota. Miscellaneous Series 87.

Evans, S. E., and J. Klembara. 2005. A choristoderan reptile (Reptilia: Diapsida) from the Lower Miocene of northwest Bohemia (Czech Republic). Journal of Vertebrate Paleontology 25(1):171–184.

Gao, K., and D. B. Brinkman. 2005. Choristoderes from the Park and its vicinity. Pages 221–234 in P. J. Currie and E. B. Koppelhus, editors. Dinosaur Provincial Park: a spectacular ancient ecosystem revealed. Indiana University Press, Bloomington and Indianapolis, Indiana.

Grande, L. 2013. The lost world of Fossil Lake. The University of Chicago Press, Chicago, Illinois.

Parks, W. A. 1927. Champsosaurus albertensis, a new species of rhynchocephalian from the Edmonton Formation of Alberta. University of Toronto, Geological Series 23:1-48.

Parks, W. A. 1933. New species of Champsosaurus from the Belly River Formation of Alberta, Canada. Transactions of the Royal Society of Canada 27:121–137.

Russell, L. S. 1956. The Cretaceous reptile Champsosaurus natator Parks. National Museum of Canada Bulletin 145:1–25.


  1. Good article. You may not be aware but at choristoderes nest with Doswellia as sisters to parasuchia and proterochampsidae, all descending from BPI 2871, a small proterosuchid without an antorbital fenestra and Elaphrosuchus, a larger proterosuchid with an antorbital fenestra.

  2. I must confess that I'm fairly agnostic about where choristoderes fit in, coming at them as I do more than halfway through their history and via one of the more specialized examples. The main issue, of course, is that there are spotty fossil records, and then there is the choristodere record: about two dozen species over 200 million years, concentrated in the Early Cretaceous and Campanian–Paleocene, and with several distinct lineages.