Sunday, July 14, 2013

Deinodon's Identity Revisited

My recent illustration of a specimen traditionally assigned to Gorgosaurus was labelled Deinodon, usually considered a likely synonym. But is it really? Deinodon is, in fact, much more likely to have been Daspletosaurus all along. Read on for the nitty-gritty stratigraphy!

I've discussed the issue of the validity of the classic tyrannosaurid (er, deinodontid) genus Deinodon horridus on this blog several times before, most notably here. To sum up, De. horridus has, since Russell's 1970 revision of Canadian tyrannosaurs or before, usually been considered a nomen dubium for several reasons.

Number 1: It is diagnosed by large, D-shaped teeth (now known to be characteristic of all advanced tyrannosaur species). Since all tyrannosaurs have teeth of this kind, it cannot be compared to species based on teeth with associated skeletons based on anatomy.

Number 2: The holotype teeth are from the Judith River Formation of Montana, and equivalent beds across the Canadian border (the Belly River Group, consisting of the Foremost, Oldman, and Dinosaur Park formations) contain at least two or three distinct species of tyrannosaur: Daspletosaurus torosus (Oldman), Gorgosaurus libratus (Dinosaur Park), and Daspletosaurus sp. (aka Daspletosaurus "chicagotyrannus"). So, De. horridus cannot be assigned to any particular species based on stratigraphy, a fact first pointed out by Russell in his influential 1970 paper.

Previous workers had often suggested that De. horridus was probably a synonym of G. libratus, back when that was the only named Judithian-age tyrannosaur. Russell, in the same 1970 paper, pointed out that his new genus Daspletosaurus torosus, which had been found in 1921 but previously thought to be a new species of Gorgosaurus, meant that there were two Judithian tyrannosaurs and that neither could be confidently identified as the true identity of Deinodon horridus.

I had previously discussed current opinions about the classification of D. horridus, noting that most authors since Russell have said it is a nomen dubium probably synonymous with Gorgosaurus libratus, and therefore labelled my retro-style Gorgosaurus restoration as Deinodon.

One issue that's come up as I've read more about this situation is the problem of stratigraphy. Were all these species really contemporaries? Gorgosaurus libratus is only present in the lower to middle portions of the Dinosaur Park Formation, and only overlaps in time and space with Daspletosaurus "chicagotyrannus", never Da. torosus, which is present in the older Oldman Formation.

Across the border, the Judith River Formation is in dire need of divvying up. As illustrated in the stratigraphic chart of Sullivan and Lucas (2006, fig. 5), the JRF spans the entire length of the Judithian age, equivalent with the entire Belly River Group. Without detailed information on the stratigraphic level of the original De. horridus lectotype specimens, trying to match the tooth taxon with the skeletal taxa may be futile.

However, it seems that the traditional identification of De. horridus with G. libratus may be the least likely hypothesis. In recent years, for the first time in a long while, we have good skeletal remains of stem-avians coming out of the Judith River Formation. Can their stratigraphy tell us anything about the correspondence of US and Candian Judithian-age species?

Let's consider Medusaceratops. Described in 2010, this ceratopsid was originally believed to belong to Albertaceratops, described in 2007 from the Oldman formation. Further research was used to distinguish these two genera, though there may still be representatives of Albertaceratops among the specimens of the Judith River bone bed where Medusa was found. This year, another new genus, Judiceratops, was reported from the Judith River. In his description, Longrich notes that the fossils were found just across the border from outcrops of the Oldman and foremost formations, making Judiceratops older than other ceratopsid fauna, rather than a representative of a unique fauna endemic to Montana. So far, it looks like the Judith River beds that yield stem-avians are decidedly older than the Dinosaur Park beds, roughly equivalent to the Oldman Formation.

What about hadrosaurs? Index fossils help out here as well. Aside from Albertaceratops, the only named dinosaur species definitely present in both the Judith River Formation and the Belly River group is Brachylophosaurus canadensis, which is also known from, you guessed it, the Oldman Formation across the border.

It looks like, with what limited data we have right now, the best guess for a Canadian match for any given Judith River dinosaur fossil is probably the Oldman, not the Dinosaur Park. And there's only one tyrannosaur known to be present in the Oldman: Daspletosaurus torosus. If I had to make an educated guess about the identity of Deinodon, my money would be on Daspletosaurus.

It would be ironic, to say the least, if Russell's paper famously sinking the name Deinodon because it was known from no coeval body fossils was the same one in which he inadvertently named the very first skeletal remains of De. horridus! But what we really need is some way to test this, and see if daspletosaur and gorgosaur teeth can be distinguished from each other, as well as to look back through Leidy's notes for hints as to the stratigraphy of the horridus lectotypes. Until then, Deinodon still remains a fascinating nomen dubium amongst a sea of unnamed Judithian tyrannosaur species.

Oh, and there's one more thing... the mysterious, unnamed tyrannosaurine (deinodontine?) known as "Sir Wlliam". This is another tyrannosaur from the Judith River Formation, distinguished from Da. torosus based on its lower tooth count (13 vs. ~18), making possibly more closely related to Tyrannosaurus. As reported, this seems to have come from the upper Judith River formation (it was originally misidentified as having come from the Hell Creek formation), and may be transitional between Da. torosus and more advanced tyrannosaurines. So, apparently not all of the stem-avians from the Judith River correspond with the Oldman, but we'll need to wait until Sir William is published to learn more.

Part of the problem here may be the hidden diversity of Judithian tyrannosaurs. There is a huge amount of overturn in herbivorous stem-avian fauna during the ~10 million year duration of the age, and it seems slightly odd that there are only three recognized species of large carnivores spanning the whole thing. There are probably a few more species of tyrannosaurine and albertosaurine that will one day be recognized, which makes finding out more about the exact stratigraphy of De. horridus even more critical to solving the mystery of its identity.

And don't even get me started on Trachodon...


Longrich, N. R. (2013). Judiceratops tigris, a New Horned Dinosaur from the Middle Campanian Judith River Formation of Montana. Bulletin of the Peabody Museum of Natural History54(1), 51-65.

Russell, D. A. (1970). Tyrannosaurs from the Late Cretaceous of western Canada (p. 34). Queen's Printer.

Ryan, M. J., Russell, A. P., & Hartman, S. (2010). A new chasmosaurine ceratopsid from the Judith River Formation, Montana. New perspectives on horned dinosaurs. Edited by MJ Ryan, BJ Chinnery-Allgeier, and DA Eberth. Indiana University Press, Bloomington, 181-188.

Sullivan, R. M., & Lucas, S. G. (2006). The Kirtlandian land-vertebrate “age”–faunal composition, temporal position and biostratigraphic correlation in the nonmarine Upper Cretaceous of western North America. New Mexico Museum of Natural History and Science Bulletin35, 7-29.


  1. This comment has been removed by the author.

  2. why have you restored it all tripoddy? Has the pendulum on posture swung back? [edited for typo]

    1. No, but there's nothing in tyrannosaur anatomy that kept them locked into a strictly horizontal posture 24/7! As Tom Holtz once said, "animals move." ;)

  3. Interesting idea, especially as the serrationless nature of Aublysodon suggests it is Daspletosaurus too (or rather that the latter is the former), and Aublysodon and Deinodon were both described from the same collection of teeth. I hadn't heard of "Chicagotyrannus" before. Is that from Bakker's paper in Dinosaurs Past and Present?

    1. Actually i'm not sure of the original source... on his website Ford attributes it to Bakker & Currie 1987, but then states "I asked CURRIE about this and he doesn’t know what BAKKER is talking about, in fact when I asked him it was the first he heard about it!"

    2. After some searching, there is no "Bakker and Currie, 1987" paper. The specimen (FMNH PR308) has been known for some time of course, recently being moved from Gorgosaurus/Albertosaurus to Daspletosaurus, and often being called the stretch-snouted Daspletosaurus. The earliest use of the name seems to be from Tim Williams on the DML in 1997 ( ), which Naish later questions and Williams responds that he heard of it privately via a DML member ( ) and had it confirmed by Olshevsky. I wouldn't be surprised if Bakker credited Currie with coauthorship without his knowledge. Maybe Bakker wrote an obscure 1987 magazine article doing this, or maybe Ford is confusing unpublished knowledge of a name worked on by Bakker and supposedly Currie with the 1987 nomen nudum "Clevelanotyrannus". The latter was an early name for Nanotyrannus published in the bibliography of Currie (1987). Supporting this idea is that Olshevsky never lists "Chicagotyrannus". I'll ask Ford and see what I come up with.

  4. I quite like the premise that Aublysodon teeth are juvenile premaxillary teeth from other tyrannosaurids although the lack of serrations does bother me. I cannot see why juvenile premaxillary teeth would necessarily be denticle-less and some Aublysodon teeth are relatively large.

    1. The problem with Aublysodon is that like Troodon, they've been described from many different formations and time periods. In general they may be from juvenile tyrannosaurs, but as Mickey notes the holotype teeth of Aublysodon mirandus (the only ones that count when it comes to the name) are probably synonymous with Daspletosaurus torosus.

    2. Well, I only say Aublysodon mirandus is probably Daspletosaurus. It's your Dinosaur Park Fm = Oldman Fm idea that would make it D. torosus.

      As for Mark's comment- Juvenile tyrannosaurs resemble basal coelurosaurs more, and most coelurosaurs have serrationless anterior teeth. We also know this to be true for juvenile Daspletosaurus (RTMP 94.143.1) and Tyrannosaurus (LACM 28471). Note at least some supposedly large unserrated teeth (e.g. Aublysodon OMNH 10131 of Lehman and Carpenter, 1990- now identified as Bistahieversor) turned out to have serrations after all.

  5. CAN we get you started on Trachodon? I mean, as long as you're on a nomen dubium kick...

    Even as a 90's child, I grew up with Trachodon, a name whose absence I feel almost as keenly as that of Brontosaurus (my parents only got me outdated books, okay?).

  6. The "D" shaped teeth were given the name Aublysodon (and are small not large), so they aren't part of the type material of Deinodon. And the type material of Deinodon is really, really scrappy, incomplete, non diagnostic teeth.