Peg Teeth

I’m in the midst of drafting (for the third time) an upcoming post about Drepanosaurus and Avicranium but I doubt they’ll be done by the end of the month. However, I wanted to get something up on this darn blog because I'd like to maintain the illusion of being loyal to my seven or eight readers. So I’m going to briefly discuss something strange about ceratopsians that nobody ever seems to comment on: the weird peg teeth of basal neoceratopsians.

What teeth am I talking about? Well here’s a specimen of Yinlong (IVPP V14530) that shows them pretty nicely.

Yinlong has some big ones.

These peg teeth are widely distributed among basal ceratopsians and are surprisingly consistent between taxa. There are always at least two, and never more than four, peg teeth per premaxilla, they differ substantially from the cheek teeth, and they oppose the predentary bone. In many of these basal taxa, the rostral bone originates ahead of the tip of the predentary, which makes me wonder what purpose the rostral is even serving in these basal forms. Even assuming “cheek-like” structures covering the cheek teeth, the peg teeth would have been visible in the living animal even if the mouth was closed given that they oppose the predentary.

The morphology of these peg teeth is interesting. The crowns are broadest at the base and narrow towards the tip—a sort of asymmetrical teardrop shape. In many peg-teeth-bearing taxa, the crowns are equipped with serrations.

Interestingly, while peg teeth are present in Yinlong and many other basal ceratopsians, they are absent in Psittacosaurus and Mosaiceratops.

Auroraceratops has some big premaxillary teeth.

However, among non-psittacosaurid basal neoceratopsians, they are present in a number of taxa: Aquilops (that's it at the top of the post), Archaeoceratops, Auroraceratops, Liaoceratops, Yamaceratops, and Protoceratops andrewsi but not P. hellenikorhinus. There might be others I'm forgetting but those are the papers I have in front of me and I'm trying to bang this post out quickly. Peg teeth are absent in leptoceratopsids (that I know of), Zuniceratops, Turanoceratops, and Ceratopsidae.

Among other ornithischians, the only other group I know of that has dentition that even vaguely resembles those peg teeth are…heterodontosaurs. While best known for their impressive caniniform teeth, heterodontosaurs also have small premaxillary teeth that oppose the predentary. They are present in Echinodon, Tianyulong, Abrictosaurus, Heterodontosaurus, and Lycorhinus. In most of these animals, the premaxillary teeth are morphologically quite different from those of basal ceratopsians. 

Sereno's reconstruction of Echinodon's teeth.

Rather than having a teardrop shape, their teeth are more or less straight and taper to the tip. In Heterodontosaurus and Lycorhinus, the premaxillary teeth become progressively larger posteriorly. The largest of these (pm3) is a proper fang, slightly recurved, that would have overlapped the lateral edges of the predentary. In Heterodontosaurus, who’s been rigorously monographed by Norman et al. (2011), the two anterior premaxillary teeth are not serrated but the largest “caniniform” tooth bears “small square-edged serrations” on its distal edge.

In Echinodon, however, Sereno (2012) writes that:

There are three premaxillary teeth in middle and posterior portions of the premaxilla, preceded by an edentulous margin…the crowns are slightly swollen, the mesial side of the crown base more bulbous than the distal side, and have smooth surfaces without denticles or serrations. The crowns are gently recurved with apices set slightly distal to the center of the crown base.

Indeed, in his reconstruction of Echinodon, above, there are superficial similarities to the premaxillary teeth of basal ceratopsians.

I don’t think this similarity is necessarily evidence of a close relationship between heterodontosaurs and ceratopsians (although that idea has been floated before). However, functionally speaking, there may be some overlap here.

Heterodontosaurus' teeth are quite different from Echinodon.

Sereno writes that the premaxillary teeth of heterodontosaurs do actually show wear on their lingual side from tooth-to-bill contact. As for the large caniniform teeth of Heterodontosaurus and Lycorhinus, Sereno assigns their function to puncturing, cropping and rooting: 

“Crown breakage in the premaxillary series suggests at least occasional contact with hard materials, as may occur in the course of agnostic or rooting behaviors.”

Could their premaxillary teeth be used to acquire squirmier food? Barrett (2000) suggested that heterodontosaurs may have been facultative omnivores. This position was supported by Butler et al. (2008), who noted that the caniniforms develop early in ontogeny and so were not strictly for display. Porro (2011) also suggested frequent omnivory. Norman et al. (2011), in their Heterodontosaurus monograph, considered both it's caniniform teeth and grasping, raptorial forearms as evidence for catching and consuming small prey items, raiding nests and/or opportunistic scavenging.

Peccary from Wikipedia, by Tim Vickers.

However, Thulborn (1974) and Norman et al. (2011) also compare heterodontosaurs to peccaries, which are modern herbivores with canines. Peccaries use their canines to settle arguments and as defense against predators “…and may represent the closest living analogs to heterodontosaurids.” Sereno finds that heterodontosaurs were mostly likely predominantly, if not exclusively, herbivorous.

The analogy with heterodontosaurs may be helpful in figuring out the function of the premaxillary teeth of basal ceratopsians. They share many similarities:

This juvenile Liaoceratops has 'em.

1. Size and number of premaxillary teeth is variable between taxa;

2. Premaxillary teeth turn up early in ontogeny (Hone et al. 2014, Xu et al. 2002);

3. Premaxillary teeth oppose the predentary.

From these three similarities, we can make three inferences:

1. Unlike the cheek teeth of ceratopsians or the carnivorous dentition of theropods, the variation seen in basal ceratopsian premaxillary teeth implies they were not consistently used for any single purpose, like gathering or processing food, because in that case their morphology would be more uniform.

2. Their appearance early in ontogeny, coupled with difficulties in demonstrating sexual dimorphism in ceratopsians, shows that the premaxillary teeth were not primarily for display.

3. The premaxillary teeth must have been at least somewhat important for cropping vegetation until the embiggening of the rostral overtook their role.

But, like peccaries and heterodontosaurs, I think it's likely that basal ceratopsians used their peg teeth to settle arguments and nip attackers. Note that, in Protoceratops, the rostral is enormous and the premaxillary teeth have been reduced to toothpicks—a far cry from the proportionately larger teeth of Aquilops, Liaoceratops and Auroraceratops. Given their apparent loss in P. hellenikorhinus (Lambert et al. 2001), I wonder if their presence is variable in P. andrewsi. Maybe at a certain point their presence became as variable as our wisdom teeth. But certainly, in Protoceratops, the rostral was doing their job just fine.

But wait, you say, what about those other marginocephalians, the pachycephalosaurs? What do they have in the premaxillary dentition department? Unfortunately, pachycephalosaurs are not famous for their beautifully complete skulls or skeletons, but Stegoceras validum does have three small premaxillary teeth (per side) that oppose the predentary. The crowns are roughly teardrop-shaped have large denticles. Like heterodontosaurs, the wear patterns are suggestive of tooth-to-bill motion. Prenocephale apparently has similar wear facets on its premaxillary teeth. It would be interesting to find out how variable premaxillary tooth count and morphology are in pachycephalosaurs, but that part of the skull is so rarely preserved. For what it's worth, Dracorex (and I assume Pachycephalosaurus) does not have any premaxillary teeth.

Stegoceras skull: diagram at left and closeup of the premaxillary teeth on the right.

It seems that six premaxillary teeth is the plesiomorphic condition for ornithischians, as seen in Lesothosaurus, Scuttelosaurus, Isberrysaura, and even Thescelosaurus. If heterodontosaurs are the basalmost ornithischians, then they reduced their premaxillary tooth count independently of marginocephalians, but it is interesting to note that basal ornithopod Haya has five rather pointy, unserrated premaxillary teeth and Jeholosaurus has six. 

The nicely-preserved skull of Haya. Note the elongated premaxillary teeth.

Given that marginocephalians and ornithopods are usually found to be sister groups, it's interesting that basal members of both groups are quite different regarding their premaxillary teeth, both in number and morphology.

(not that I'm doubting the validity of a monophyletic Cerapoda, mind you.)

So there you have it. Basal ceratopsians have weird peg teeth in the premaxillae, their function might be similar to that of heterodontosaurs, and they differ both from their closest relatives (pachycephalosaurs) and from basal members of their sister group (Ornithopoda).

Oh, and Happy Halloween! Don't let the vampiritic anurognathids bite!