Why Don’t Tyrannosaurs Have All Bananas For Teeth?

Shortly after my last post, a little discussion ensues, and I realize there is a little more I should say here on the subject.

In a message sent to the Dinosaur Mailing List (the original of which can be found here [when the link is available!]), I wrote:

Tyrannosaurus rex has been called “heterodont[“] by some authors, an allusion not only to the ziphodont crowns of the typical jaw margins and the D-shaped premax teeth, but also to the lower-crowned, broad teeth in the rear of the jaws. These latter teeth show less recurvature and blunter apices than other teeth in the jaws excepting the “incisiforms” of the premaxillae, and were suggested by some authors of relating to bone-crushing. Similar teeth, although much more incrassate, are found in living crocodilians, some of which lack the morphology but others (such as Alligator mississippiensis) showing an extreme form of it. This has been linked to their increased durophagous behavior over, say, more “flesh-eating” Crocodilus niloticus.

[Modified from the original for formatting. I have also removed a reference to the person this was being responded to, but it offers little to the discussion at hand. Sorry, Dan!]

But I have never heard of a primary bone-crushing manner of dispatching prey, nor with the brevity of such exapted teeth in the rear only portions of the jaw do I think that Tyrannosaurus rex could have employed a crushing-bite method so easily. Because of their position in crocodilians and blunt-toothed lizards (Varanus exanthematicus as well, also durophagous), specialized care in processing is usually done to select foods (turtles in the crocs, gastropods and even eggs in the lizards) and in a precise manner (they are “handled” into position). I can see this behavior as an extension of carcass-stripping and marrow-scavenging, but not as a primary dispatching method.

Despite this, tyrannosaurids all still seem pretty head-dominated as adults, so much so that their jaws WERE the primary tool of killing, as they pretty much lacked any other. My off the cuff impression is that of jaw-based grappling and then disabling through progressively more grievous wounds. Horner can even have his say here as a tyrannosaur can spend a day or two (as wolves can do) following previously injured prey in order to ensure it keeps the eventual kill. None of this requires “slash-and-burn” as inferred for Allosaurus/Carnotaurus. However, I am glad to be wrong on this, as I also suspect Tyrannosaurus rex may have occasionally used the leopard strategy: A precise crushing bite to the skull[.]

There’s a lot to say here in fleshing this out. First, let me show you something:

Skulls of species of Varanus, showing variation in the jaw anatomy that are essentially dietary in nature. A, Varanus exanthematicus (Bosc's monitor); B, Varanus gouldii (sand monitor, goanna). Bars beneath skulls represent jaw partitions, with labels: a, recurved, compressed teeth; b, blunt, incrassate teeth; c, position of the coronoid apex; q, position of the quadrate nadir; r, end of the retroarticular process. Scale bar (top left) represents 1 cm.

Crushing Teeth

Bosc’s Monitor (Varanus exanthematicus) is a consumer of snails, predominately and almost exclusively, although it also consumes various other non-vertebrate (and generally soft) foods. The massive mandible and anteriorly placed jaw adductor muscles (as noted by the high and forwardly-positioned coronoid process of the jaw, help provide crushing strength to the rear teeth, which are more globular, thicker, wider, and lower than the teeth anterior to them. These features help enforce the “durophagous” jaw morphology of various amniotes. Then there’s a more “typical,” carnophagous Varanus (the sand monitor [or Gould’s goanna], Varanus gouldii). It bears a more slender skull; angles of muscles (including the adductor complex) act on the jaw in a more longitudinal rather than vertical manner, so vertical reinforcement of the skull (as in Varanus exanthematicus) is lacking. The sand monitor is a generalist, consuming a variety of vertebrate and invertebrate prey, is an active hunter, and also consumes eggs, but it specializes on vertebrates. Such teeth are precisely developed to allow the tearing of flesh, rather than the crushing of shells.

Skull of Crocodilus niloticus (top) and Crocodilus anthropophagus (bottom). From Brochu et al., 2010. Note the modern croc's posterior teeth are lower and blunter. This is further exaggerated in Alligator mississippiensis.

Typical carnophagy involves the use of a recurved tooth slicing along the “stratum” of flesh, both being able to pierce the muscle and tissues surrounding it, but then to cut through as the animal pulls its jaw backward. This is usually aided by the presence of a “cutting edge,” formed from either a carina (or “keel”) or by denticulation or serration of the edge, or both. Otherwise, the tooth merely punctures, then pulls the entire bulk with the animal, and this is precisely aided by the lack of above characteristics, although it exterts far greater force onto the skull because of the weight of the animal being “pulled” instead of the resistance of the tissue itself. Durophagy, on the other hand, requires no puncturing ability, but instead uses compressive forces to exert the bulk of the jaw’s bite strength to crack the defenses of a shell or bone. This is useful when predating on turtles, snails (see the sand monitor, above), crustaceans, molluscs, or even birds’ eggs. Teeth adapted for durophagy are always blunt, and about as wide as they are long at the base, and are “low-crowned.” This prevents them from breaking under the enormous pressure. Enamel is usually extremely thick in such teeth. Human molars are an example of this morphology a little closer to home, dear reader.

Jack Horner has used the presence of incrassate (or “swollen”) teeth in tyrannosaurs (specifically Tyrannosaurus rex) to argue for the idea that tyrannosaurs from the Hell Creek were predominately scavengers, likened to hyenas, and consumed the kills of other animals rather than hunted its own, or merely chanced upon casual deaths. This has been attacked in the literature, but that’s not the point of this discussion.

A Tyrannosaurus rex maxilla (MOR 1125), showing three distinct "morphologies" of teeth.

Tyrannosaurus rex exhibits generally three distinct dental “morphologies,” all of them essentially the same tooth form (a compressed cone) with modifications: The premaxillae and the first few crowns of the maxillae have “incisiform” or D-shaped teeth, with the mesial carina rotated around the crown and which are shorter, rounded at the tip, and resemble mammalian incisors (hence the term “incisiform”). Most of the teeth of the maxilla and dentary are the “ziphodont” type, flattened side to side and recurved somewhat, but they are still rather thick. This is in comparison to, say, Albertosaurus sarcophagus or Gorgosaurus libratus, both of which have much narrower crowns and are generally not blunted and a little more recurved. This leads to the term “incrassate,” meaning swollen, and has likened their appearance to bananas.

The third type, as noted by Horner et al. recently, is a blunter, more “plantain-like” than “banana-like” tooth in the rear of the maxilla and dentary. I suggested in the last post and the quoted section above that these were similar to various durophagous animals (see the crocodilian and lizard skulls above for a direct comparison). There are differences, of course, but not by much. Incidentally, the Tyrannosaurus skull above is the same specimen from my last post which produced the teeth the figure is based on. It comes from here (which is the website of the Museum of the Rockies, where Horner works). Smith (as reminded to me by Jerry Harris), has attempted to quantify variation in teeth of some theropod dinosaurs as part of a massive project to morphometrically evaluate “heterodonty” ( a question I raised previously).

Tyrannosaur teeth are often “banana-y” a feature that helps resist bending during compression, and is very useful during high-pressure, low-impact biting, and is extremely prevalent in durophages. But tyrannosaurs lack truly incrassate teeth save at the very rear of the jaws, usually restricted to smaller, last few teeth, rather than a third or half of the dental series (as in the lizards shown above). Precision crushing should occur in the most “blunt” region of the dental series, for maximal resistence to damage of the jaws and teeth, and yet this region is very small. Unlike Horner et al., I tend to be less willing to ascribe this morphology (advanced they say at only very advanced age) to specialized adult-only diet. Instead, I suggest the increasingly incrassate teeth (progressing further posteriorly) is a natural remnant of progressively stronger biting of durous material (e.g., dense, bone-laden skin as in crocs and ankylosaurs, the distal long-bones of the limbs of most dinosaurs, and the skulls of various dinosaurs and crocs). In fact, I find a strong favorability to this in the precision crush-bite of leopards, which are prone to biting the skulls of prey to cause nearly instant death. This may then imply that the high agility of the tyrannosaur foot (I’ll get into this later, but needless to say, Tom Holtz and Eric Snively’s work will feature prominently) and agile turning capability coupled with low CoG (Don Henderson this time) allow tyrannosaurs to be very maneuverable and very likely enabled precision grappling with the jaws. (And before anyone says, I lack the expertise to even attempt to write a paper on this!)

So maybe that’s why tyrannosaurs do not have all bananas for teeth. Sometimes, a “ziphodont” tooth is just fine for slashing meat, and doesn’t need to be for crushing bone.

Tyrannosaur Scavenging, a Short Perspective

Jack Horner has attempted to use the concept of the “scavenging tyrannosaur” as a bogeyman for concepts of other paleontologists as sticklers for “sexy” ideas, especially the fast ultra-predator that was envisioned in Jurassic Park (of which Horner was a, if not the only, paleo consultant — Bakker argues he was also one, in his preface to Raptor Red, but I digress). In this concept, formulated in a paper and later in a book with Don Lessem, and further in various presentations and television programs, he has derided the so-called predatory features and advanced various other factors that favored scavenging as the sole means through which Tyrannosaurus rex, the only multi-ton theropod dinosaur in the latest Maastrichtian, would gain sustenance. This was supported by Ruxton and Houston on an ecological comparison (you, dear reader, should mind that the Hell Creek was very different from any modern biota if only due to the sheer volume of the animals involved and their effect on relatively smaller bits of vegetation, as well as that it is a preservational facies, and thus does not represent the entire extent of the range to which any preserved animal was knowable).

On both ecological and morphological grounds, this ideology of tyrannosaur-as-sole-scavenger has been challenged by Farlow and Holtz in assessing dietary regimes in theropods and later specifically by Holtz, with the general conclusion that there are no true pure-scavengers, and that tyrannosaurs would have been opportunistic feeders (using a modern analogy, more like lions today than hyenas, although the latter also hunt!), capable and unlikely to pass a meal-on-the-hoof (as it were).

Now, Horner et al. have pulled a 180° turnabout from Horner’s original position, but they do this in a surprisingly disingenuous manner:

This census suggests that Tyrannosaurus was not strictly a predator, but instead more of an opportunistic feeder, possibly selecting similar food choices under circumstances comparable to that of hyenas in extant ecosystems, a trend unrecognized in earlier census studies.

See, Horner et al. do an egregious thing here: They establish a strawman argument, setting up a proxy position (that of an obligate predator) and then using the census data they collected to argue against it — amazingly coming to the same conclusion as Farlow and Holtz & Holtz. The argument never existed. No one has ever argued that Tyrannosaurus rex was “strictly a predator,” or if they did, it was off the cuff: formal publications assessing ecologies in the Hell Creek (including some by Bakker for his earlier work) have always implied some amount of opportunism (sometimes scavenging) on top of active predation. A second egregious thing, that Horner has directly argued against this position, is not mentioned at this point (the work Horner used to establish the “poor predator” argument is not even cited).

(I’d like to note that it was David Marjanović who used the “strawman” appellation on the DML in response to this post.)

I suspect rather that this work intends to help shore up previous conclusions that Horner’s “scavenger” features are just to be indicative of the opportunism hypothesis, despite using comparisons to carrion-finders and scavengers (such as his oft-mentioned turkey vulture, Cathartes aura) in the past. Hyenas are now used, as classic scavengers, although they are predators as well. Horner et al. use associations collected over ten years at one major network of sites in eastern Montana, extrapolate them across the ecology of the entire formation (exposed also in the Dakotas and Wyoming), and use the association of specimens from the western extent of the formation to the rest of it, and reason that this allows tyrannosaurs to be opportunistic feeders.

Brochu, C. A., Njau, J., Blumenschine, R.J. & Densmore, L. D. 2010. A new horned crocodile from the Plio-Pleistocene hominid sites at Olduvai Gorge, Tanzania. PLoS ONE 5(2):e9333. doi:10.1371/journal.pone.0009333
Farlow, J. O. & Holtz, T. R., Jr. 2002. The fossil record of predation in dinosaurs. pg.251-266 in Kowalewski & Kelley (eds.) The Fossil Record of Predation. The Paleontological Society Papers 8.
Holtz, T. R., Jr 2008. A critical reappraisal of the obligate scavenging hypothesis for Tyrannosaurus rex and other tyrant dinosaurs. pg.371-396 in Larson & Carpenter (eds.) Tyrannosaurus Rex. Indiana University Press, Bloomington.
Horner, J. R. 1994. Steak knives, beady eyes, and tiny little arms (a portrait of T. rex as a scavenger). pg.157-164 in Rosenberg and Wolberg (eds.) Dino Fest. Paleontological Society Special Publication 7, University of Tenessee, Knoxville.
Horner, J. R. & Lessem, D. 1994. The Complete T. rex: How Stunning New Discoveries Are Changing Our Understanding of the World’s Most Famous Dinosaur. Simon and Schuster, New York City.
Horner, J. R., Goodwin, M. B. & Myhrvold, N. 2011. Dinosaur census reveals abundant Tyrannosaurus and rare ontogenetic stages in the Upper Cretaceous Hell Creek Formation (Maastrichtian), Montana, USA. PLoS One 6(2):e16574 doi:10.1371/journal.pone.0016574
Ruxton, G. D. & Houston, D. C. 2003. Could Tyrannosaurus rex have been a scavenger rather than a predator? An energetics approach. Proceedings of the Royal Society of London, Biology 270:731–733.
Smith, J. B. 2005. Heterodonty in Tyrannosaurus rex: implications for the taxonomic and systematic utility of theropod dinosaurs. Journal of Vertebrate Paleontology 25(4):865-887.
Wilson, J. B. 1996. The myth of constant predator: prey ratios. Oecologia 106:272–276.