Tales told by a toddler tyrannosaur

Cover of the issue of the Journal of Vertebrate Paleontology in which our Tarbosaurus research is the Featured Article. It's open access and freely downloadable!

Really young animals are always interesting to scientists, because they often shake up long-held notions based on the better-known attributes of adults. Youngsters are doubly interesting to paleontologists because they’re so rare. So, when the beautifully preserved, almost complete skull of a young Tarbosaurus came into the lab, we knew we had an important specimen. It was brought to the WitmerLab in early 2007 by Takanobu Tsuihiji, who had joined us as a postdoctoral fellow in 2006. Taka had participated in the joint expedition to the Gobi Desert of Mongolia led by the Hayashibara Museum of Natural Sciences and the Mongolian Paleontological Center. The first of our studies on this specimen just appeared as the open-access Featured Article in the May 2011 issue of The Journal of Vertebrate Paleontology. Others, such as Brian Switek and Jaime Headden, have also blogged about this article, and so I’ll try to focus on more of the back story from an inside perspective. We provide more resources on our university site, as well as a photo album on our WitmerLab Facebook page.

The block preserving the young Tarbosaurus skeleton prior to the skull being removed. Courtesy of the Hayashibara Museum of Natural Sciences and WitmerLab at Ohio University.

A treasure from the Gobi
The discovery itself is the stuff of dreams: an almost complete, articulated dinosaur skeleton. The only parts missing were the end of the tail and the neck. The skeleton was discovered in 2006 in the famous Bugin Tsav locality of the Nemegt Formation of the western Gobi Desert. The skull was removed first while the rest of the skeleton is being prepared. The right side of the specimen was exposed to weathering and so is missing some elements, but the left side was still buried and is gorgeous. Thanks to bilateral symmetry, we have excellent information on every bone of the skull, except for a small bone of the mandible (the articular…sadly a personal favorite). Although we tried to avoid the hyperbolic use of superlatives that plague dinosaur news releases, we felt completely justified in regarding this as the most complete, youngest skull of a tyrannosaurid.

A family of Tarbosaurus, rendered in Maya by Ryan Ridgely. The toddler is in the foreground, the teenager is sulking moodily in the background, and the adult nuzzles (or prepares to devour) her young at right.

Looking back and looking forward
In April 2007, Taka and I took the skull over to O’Bleness Memorial Hospital here in Athens, Ohio, for CT scanning. Taka presented our initial findings at the SVP annual meeting in October 2007. The scan data are beautiful, but our recent article just hints at how informative these data are proving to be. Not surprisingly given our recent NSF grants, we’ve already generated a virtual brain endocast, inner ear, and related elements, so we’ll be able to put neuroanatomical ontogeny in Tarbosaurus into the context established by the Witmer & Ridgely (2009) article. The skull looks flattened, but it’s more “folded” than crushed. We can basically take it apart bone-by-bone (as we did with our 3D Alligator), mirror missing elements, repair broken elements, and assemble a complete skull. The intent here isn’t to tease but rather to assure you that this little guy has more stories to tell. We did a fair amount of this “second project” while Taka was still a postdoc in the lab, but we sort of shelved it because it made sense to do the logically-prior, basic descriptive article first. The whole enterprise took a break when Taka moved back to Tokyo for his postdoc at the National Museum, but shortly thereafter he kicked the descriptive article back into high gear.

Clicking the above image will take you to the Ohio University news release, of which this image was a part. The OU release circulated widely.

Everybody loves a baby
I’m not going to recap all the findings from the article. It’s a JVP article, and as such, it’s a “journeyman” work, reporting on bony doodads of interest to professionals. Please check out the article itself, as well as our attempt to provide some supplemental information. The latter did double-duty as a means of providing information for the media, because, as a Featured Article, the SVP’s Media Liaison Committee issued a press release. To be honest, I initially had little enthusiasm for doing what would be my fourth media thing since last August, the previous three being terror birds, dinosaur limb cartilage, and just last month, olfactory evolution in birds and other dinosaurs. But this was a major article for Taka (as senior author and architect of the project) and the Hayashibara Museum (which has fallen on hard times)…plus how can you not want to talk about a specimen like this?

Skull of the juvenile Tarbosaurus with the rock matrix digitally removed to reveal the delicate nature of the skull's construction.

Ontogenetic niche partitioning
That fancy phrase is what ecologists use to refer to the phenomenon of individuals within a species shifting their dietary or feeding strategies as they get older, sometimes as means of exploiting differing resources and sometimes as means of decreasing competition among age classes. One thing that the juvenile Tarbosaurus skull shows is that it was a very different beast than its parents. Its skull and teeth were almost delicate in comparison to the heavily buttressed and reinforced skulls of the adults. At the age of two- or three-years-old, our little guy simply couldn’t have engaged in the bone-crushing, vigorous mode of predation that the adults employed. Still, we have no reason to believe that it was anything other than an active little predator, out on its own terrorizing the countryside in its own way. It must have used its speed and agility when young, becoming a more power-based grappler as it aged. The point is that the prey community might not have regarded Tarbosaurus as a single kind of predator, but rather almost like a bunch of different predators…a single species filling the niches of multiple species in other ecosystems. I’m just kinda riffing here, but it’s an intriguing idea, and we’re not the first to suggest it.

The Cleveland tyrannosaur skull (CMNH 7541): the holotype of Nanotyrannus or a juvenile T. rex? Volume rendering of the skull based on CT-scan data.

Whatever, dude…what’s all this tell us about Nanotyrannus?
Bony doodads and ecological riffing only go so far with some folks. They want to know how finds like this impact “controversial” specimens. A number of tyrannosaur specimens have been variably regarded as juveniles of known species or juveniles/adults of other species. Chief among them are the Cleveland skull and the Burpee Museum specimen (“Jane”), which some people think are juvenile T. rex and others think are a separate beast, Nanotyrannus. We presented the significance of the new Tarbosaurus juvenile for that debate (and stay tuned for a forthcoming post here devoted to the 2010 Witmer & Ridgely article on the Cleveland skull). Let’s start with a classic debate, loss of tooth positions. For the Cleveland or Burpee skulls to be young T. rex, they would have had to have lost tooth positions as they aged because adult rex has far fewer teeth. The new Tarbosaurus has a tiny skull only half the length of the Cleveland skull but already has the same number of teeth as adult Tarbosaurus. Apparently no loss of tooth positions in Tarbosaurus. Also, both the Cleveland and Burpee skulls have an unusual opening in a cheek bone (quadratojugal), but adult T. rex lacks it. Is this a novel characteristic of Nanotyrannus or something that may have changed as the animal aged? Again, the new juvenile shows that there was no change in Tarbosaurus as it grew, since both the juvenile and adults lack the opening, just as in adult T. rex. Maybe T. rex did change in these attributes during growth. We certainly found that Tarbosaurus changed a lot with regard to the feeding apparatus (from delicate to monstrous). But we now know that T. rex’s closest relative, Tarbosaurus, didn’t change in these two “marquis” attributes that have become emblematic of the debate.

Comparison of the skulls of Raptorex (from Sereno et al. 2009) and the juvenile Tarbosaurus. Similar but not identical. Note that the Tarbosaurus skull is not restored and so presents somewhat of a side and top view simultaneously.

And what about Raptorex?
For many, a first glance at the little Tarbosaurus skull elicits a response something like, “Hey, that looks a lot like Raptorex.” Raptorex was announced by Paul Sereno’s team in 2009 as “a new, small-bodied, basal tyrannosauroid from Lower Cretaceous rocks in northeastern China.” According to Jim Clark’s accompanying piece in Science, it turns out that the provenance is a little shakier that anyone would like, having been sold on the market with basically unknown locality data before finding its way to scientists. The fact that it was originally thought to be “a baby Tarbosaurus” puts an interesting spin on the resemblances to our unquestionable “baby” Tarbosaurus. We re-ran the Sereno et al. phylogenetic analysis, including our little Tarbosaurus, and it came out right next to Raptorex. Given that neither Taka nor I nor any of the authors on our JVP article had studied the Raptorex fossils firsthand, we weren’t in a position to question the anatomy or provenance assessment. If Raptorex is legit, then, as we say in the article, our juvenile Tarbosaurus would be an excellent example of ontogeny recapitulating phylogeny. Our conclusion was that our team had no objective basis to dispute the legitimacy of Raptorex and didn’t do so. That said, some skepticism remains appropriate given the uncertain provenance. The take home message here is that distinguishing juvenile attributes (our little Tarbosaurus) from primitive attributes (Raptorex) can be really hard, and that specimens with unquestioned identity (again our little Tarbosaurus) can help us interpret more controversial ones (the Raptorex fossils, the Cleveland and Burpee tyrannosaurs, etc.).

–Larry Witmer