The Return(?) of Stygivenator
New fossils suggest both Nanotyrannus and Stygivenator shared the Hell Creek with Tyrannosaurus rex.
Nanotyrannus, or Stygivenator?
NOW that the long debate over Nanotyrannus-vs-T. rex seems to have finally been concluded with a definitive win in favor of Nanotyrannus,
I figured it would be worth reopening yet another debate that people have assumed was settled long ago— the validity of Stygivenator molnari, a small tyrannosaur from the Hell Creek Formation.
Some years ago I visited the Black Hills Institute and got to see a cast of the Dueling Dinosaurs, which at the time was still in search of a home. Up until this point I was pretty convinced that Nanotyrannus was simply a juvenile T. rex, which was the conventional wisdom of the day. But I took one look at the skull and thought my god, that’s not a T. rex.
It was abundantly clear that the long, low snout was too different to turn into a T. rex; this wasn’t the same species, and not even in the same part of the tyrannosaur evolutionary tree— it was not a tyrannosaurine (the subfamily that contains T. rex), and maybe not even a member of the family Tyrannosauridae. It reminded me of the long-snouted Alioramus from Mongolia more than anything… including the Nanotyrannus holotype.
So moments after I’m having my mind blown— Nanotyrannus is real???— I’m getting another shock. Is this animal even Nanotyrannus at all?? And I’m trying to figure out just what the hell I’m looking at, because at this point I simply have no idea.
I vaguely remembered, lodged away somewhere in the back of my brain, another tyrannosaur specimen that had an oddly long, low snout. This was a partial skull collected by the Natural History Museum of Los Angeles, from the Hell Creek Formation near Jordan, Montana, consisting of part of the snout, part of the lower jaws, and part of the skull roof. It was pretty scrappy and had long ago been written off as just a beat-up juvenile T. rex.
It was for a while named the “Jordan Theropod”, then described as a new species of Aublysodon, Aublysodon molnari, by Gregory Paul in Predatory Dinosaurs of the World. At some point after this, Olshevsky gave it a new genus Stygivenator, the hunter from the river Styx, which is probably the most metal name you can imagine. About the only way to make it more metal is to add an umlaut or two, as Stygïvenatör.
You will notice the two- Dueling and the Stygivenator type- share a relatively long, low snout.
Whereas the holotype of Nanotyrannus looks rather different. Note the much shorter, blunter snout, more classically tyrannosaurid in its proportions:
Nanotyrannus lancensis holotype
Dueling Dinosaurs tyrannosaur
And again, Dueling Dinosaurs for comparison.
I have a very hard time believing that these are the same animal.
A number of features potentially differentiate the Dueling Dinosaurs skull from N. lancensis, but appear to be shared with Stygivenator:
• the antorbital fossa has a V-shaped anterior margin
• the antorbital fossa is long and low
• the anterior end of the maxilla is much longer than tall
• the snout is upturned, so the tip of the maxilla lies more dorsally
• the lower jaw is very slender, and its tip is strongly upturned
• teeth are straighter, projecting more ventrally, versus more recurved in Nanotyrannus
• very narrow, slit-like naris
Other features differentiate the Dueling Dinosaurs specimen from the Nanotyrannus type, but can’t be observed in the Stygivenator type:
• premaxilla much shallower
• posterodorsal ramus of maxilla straplike (versus triangular and tapering in Nanotyrannus)
• skull much shallower posteriorly
• weak ventral projection of the jugal boss
• very broad jugal postorbital process
• very broad dorsal wing of the quadratojugal
In my opinion, more than enough features exist to show that
(a) the Dueling Dinosaurs specimen is not a Nanotyrannus lancensis,
(b) the specimen is probably the same species as Stygivenator molnari and
(c) Stygivenator molnari is a valid species
The Stygivenator type is obviously not the prettiest specimen in the world, and I’m unsure how many of the features it shows are autapomorphies, but it has a unique combination of characters that differentiate it from Tyrannosaurus, Nanotyrannus lancensis, or anything else previously named, and many dinosaurs have been named on less complete material. And while it’s a bit arbitrary where you draw the line between “new species” and “new genus” I’d say Stygivenator is probably more distinct from Nanotyrannus than, say, Tarbosaurus is from Tyrannosaurus, and so it merits its own genus. I could tell the difference at a glance, whereas I find Tarbosaurus and Tyrannosaurus frustratingly difficult to tell apart. These two lineages have probably been separate for millions of years.
This then brings us to the question of Nanotyrannus lethaea, the famous Jane specimen. What is this animal?
It seems to be intermediate (sort of) between Nanotyrannus lancensis and Stygivenator molnari, with a unique character combination.
Jane, holotype of “Nanotyrannus” lethaea
Jane resembles S. molnari (type + Dueling specimens) in having:
• small premaxilla
• the upturned tip of the maxilla
• a skull that is relatively low posteriorly
• the strap-shaped posterodorsal process of the maxilla
• weak ventral projection of the jugal
• robust postorbital process of the jugal
• broad dorsal end of the quadratojugal (not shown in the photo above, but preserved)
Jane differs from Stygivenator molnari in having:
• premaxilla not quite as small as Stygivenator molnari
• an antorbital fossa that is taller, and more U-shaped
• a shorter, blunter anterior end of the maxilla
• a more robust, weakly upturned dentary
But it differs from N. lancensis in having:
• a longer, more triangular anterior end of the maxilla
• a longer, lower antorbital fossa
• a longer, more slender dentary
• weaker ventral projection of the jugal
• broader postorbital process of the jugal
• broader dorsal end of the quadratojugal
• overall shallower back of the skull; anteroposteriorly wider lateral temporal fenestrae
Many of the features that make Stygivenator so unusual- the V-shaped antorbital fossa, the strongly upturned dentary- are clearly derived (they must be, as no other tyrannosaurids seem to show them; autapomorphies are by default derived). “Nanotyrannus” lethaea exhibits some of these derived features, but not all, and so Jane seems to be overall more primitive, and less specialized than the Dueling Dinosaurs specimen.
Now, this is admittedly a pretty quick-and-dirty take on nanotyrannosaur taxonomy, but I think it’s striking just how many characters one can find with even a fairly cursory study of the problem.
In light of this, I think the appropriate name for Jane is not Nanotyrannus lethaea, but it is probably more appropriately named Stygivenator lethaea. I suppose it is possible that Jane is a Stygivenator molnari and the differences are just ontogenetic and/or juvenile but I doubt it; the differences in snout shape seem far too extreme. Also, even though the Dueling Dinosaurs specimen is smaller than Jane, it had basically stopped growing. It wasn’t going to get bigger and turn into something that looked like Jane. Jane hadn’t stopped growing completely but its growth was decelerating: it’s unlikely it looked radically different when it became a few years older; it was basically mature. Both animals show the adult morphology.
Nanotyrannus seems to be primitive in some ways (deep maxilla with U-shaped antorbital fossa) but in other ways it looks derived: the oddly reduced dorsal wing of the quadrate, the reduced lateral temporal fenestra, the small squamosal, the strong ventral projection of the jugal, the overall widening of the skull.
The Nanotyrannus type skull has undergone some crushing, and the preservation isn’t fantastic so maybe some of this could be due to that, but the CT scans suggest this morphology is mostly real and I’m inclined to trust Bakker and Currie’s reconstruction of the skull. I concede I have not had time to sit down with these specimens in a while, and hands-on study or CT scans would doubtless reveal details and nuances I’m missing from the photographs and pictures… but the differences seem pretty extreme. I don’t think it can all be down to crushing and poor preservation of the Nanotyrannus type. My hunch is that a closer study of these animals (and better material) would reveal additional differences that my quick survey has missed.
So while Nanotyrannus is in many ways more primitive than Stygivenator, it seems it can’t be directly ancestral or purely plesiomorphic: it has also picked up derived features. It’s on a separate branch and the two seem to be evolving in parallel in different ways: Nanotyrannus with its short snout and weirdly short, broad temporal region; Stygivenator with its more normal temporal arcade and its freakishly long, narrow greyhound snout.
There is presumably niche partitioning in how Nanotyrannus and Stygivenator hunted. Nanotyrannus is probably better at taking down proportionately large prey (remember, this animal is the size of a polar bear- by our standards it’s a big game hunter, even if it’s not as big as T. rex, it’s probably taking out 1,000 pound herbivores). Stygivenator maybe hunted somewhat smaller prey.
Hell Creek Tyrannosaur Diversity
This leaves us in a rather extraordinary situation- we may have as many as five different species of tyrannosaur in the Hell Creek:
Tyrannosaurus rex
Tyrannosaurus imperator(?)
Nanotyrannus lancensis (known only by holotype)
Stygivenator molnari (incl. holotype; Dueling Dinosaurs; Frenchman maxilla?)
Stygivenator lethaea(?) (incl. holotype; Zuri)
These species are probably not all there at the same time; the Hell Creek spans about 1 million years and there would have been faunal turnover between the bottom and the top (this is what we see in Triceratops, with T. prorsus replacing T. horridus).
But even so there’s probably as many as three distinct genera- Tyrannosaurus, Nanotyrannus, and Stygivenator- simultaneously inhabiting the Hell Creek. I would tend to assume the highly derived S. molnari is at the top of the section, and there is a maxilla from the Frenchman with the weird “V” shaped antorbital fossa, which seems to fit that… but paleontology is rarely as neat and orderly as we’d like. So who knows.
If you find this improbable, well… so do I.
For a long time it seemed unclear that there was more than one species in the Hell Creek, now there’s five?
But again, my working framework before I saw the Dueling Dinosaurs specimen was “it’s all T. rex” not “there are three genera of tyrannosaur in the Hell Creek”. But I found the fossils did not accommodate the framework I walked in the door with. Currently, I am struggling to accomodate the diversity we see in the anatomy in just three species.
Where Next
Anyway, to sort this out, we’ll probably need:
• some kind of specimen-level phylogenetic analysis, including all nanotyrannosaurs with cranial material, and including the characters outlined here (and any other potentially informative characters)
• also to try to tie these things into section, and figure out if there’s any sort of stratigraphic separation,
• and of course we’ll just need more specimens (and more study, and better description, of the stuff we do have). Another Nanotyrannus lancensis would be helpful, but more material of the braincase of Stygivenator would be useful as well.
• a better understanding of tyrannosaur ontogeny (which has been impossible until now because of the assumption that these things underwent some kind of radical metamorphosis) to understand which types of variation are taxonomically informative.
I could have (and possibly should have) tried to write this up as a paper, I suppose; but honestly trying to get anything through peer review these days is exhausting, especially in a field as politically charged as tyrannosaurs. I figured it would be better to just throw these observations out there.
What it Means
What’s my point here?
Well, for many years, we had a rigid orthodoxy— there was one and only one tyrannosaurid in the Hell Creek, Tyrannosaurus rex. Anyone who debated this was attacked, derided, their professionalism and the quality of their work questioned. A rigid orthodoxy was imposed and those who doubted it were treated as heretics. Evidence that did not fit this paradigm was either explained away, or ignored entirely.
This often seemed to be about being right, not about getting it right— and there’s a big difference.
Getting it right means going wherever the evidence takes you— sucking up your pride and admitting you had it wrong (as I did with Nanotyrannus, although I concede this was a lot easier for me, as I hadn’t published a lot on tyrannosaurs, I had less skin in the game).
Being right means winning the argument, which means you tend to ignore or dismiss evidence that doesn’t support your side. In short, the Nanotyrannus debate often seemed more driven by ego than by figuring out the truth; it was often driven by politics and personal reputation and not science.
And yet it seems to me that we have yet to untangle the knot of tyrannosaur systematics, and I’m worried we’re just going to replace one rigid orthodoxy with another.
How many T. rex species are there? I honestly don’t know.
Probably more than one. I don’t think the case for naming two is compelling at this point. But if it’s not, we need to just sit down, study, describe and figure the material in detail, and try to sort it out, instead of just attacking the people we disagree with and dismissing their work. There is way more variation in T. rex than I’d expect to see within a species, but if there are two (or god forbid, three) species, then what diagnoses them? And what specimens belong to which species? More study, description, fossils are needed.
How many genera of nanotyrannosaur are there? Probably two— Nanotyrannus and Stygivenator. I concede that could be incorrect, but it seems like a lot more work needs to be done before Stygivenator can be dismissed.
So, five tyrannosaurs in the Hell Creek? Maybe. My hunch is that we might eventually find six.
Three genera: Tyrannosaurus, Nanotyrannus, and Stygivenator, with different species in the upper and lower part of the Hell Creek. This is of course speculative, which is often treated as a dirty word in paleontology. But another word for speculation is “hypothesis” and what makes these hypotheses scientific is they are testable against new observations, by finding new fossils, and additional observations of the fossils we have. And consider: we don’t even have that many tyrannosaur fossils, a few dozen, to work with. If this relatively meager sample produces a minimum of three and possibly as many as five species, what might a more complete sample— hundreds of skeletons— reveal?
Given the radical shift we’ve seen in tyrannosaur taxonomy in the past few years, we should be cautious in drawing conclusions, and open to the idea that there’s still a lot more to find out. There’s a lot more to be discovered about these animals, and we know far, far less about dinosaur diversity than we think. A little intellectual humility would go a long way here.
The whole point of paleontology and science more generally is that we don’t know everything. Which is a wonderful thing. Because if we did know everything, we could stop looking for dinosaurs, stop digging them up, and studying them. The fact that we don’t know everything, and don’t have all the answers, is the only reason we can do paleontology.