Whether Beaks

One of the more interesting topics that has me a-giddy is the relation of the overlying rhamphotheca or keratinous horny beak to the underlying bone and the correlates that allow one to infer the former from the latter. Much of this work is in prep, but not by me, and some has been mentioned (again, I invoke the Witmer Lab — purveyors of all things snouty and correlated).

Partial mandible of the loggerhead sea turtle, Caretta caretta, shown (Left) without the preserved rhamphotheca in left lateral and dorsal views, (Middle) with the preserved rhamphotheca, and (Right) without the rhamphotheca but with an outline showing its relation and shape to the underlying bony features.

There have been many arguments for inferring the presence and extent and shape of the beak in fossil vertebrates, focusing primarily on dinosaurs but extending to pterosaurs and such, where the presence of a hooked rostrum, or a mandibular beak longer than the upper would be possible, and so forth. In dinosaurs, Greg Paul has asserted that dromaeosaurs like Velociraptor mongoliensis sported a beak on the premaxilla and opposing mandibular region, extensively illustrated such as in Predatory Dinosaurs of the World (1988) and elsewhere, based partially on his argument that they are degenerate forms of “protobirds,” which derived from less-toothed, more-beaked ancestors.

It has been variously argued that a rugose surficial texture (as on the “faces” of tyrannosaurs or abelisaurids) results in facial keratinous sheets, analogous to extensions of the avian rhamphotheca, while in other cases it has been a highly foraminated (or nutriated) bone surface, often resulting in a “hole-y” appearance (pitting and such-like). Additionally, recent work has been used to imply that ceratopsian dinosaurs, much like turtles, had a nearly entire upper cranium encases in forms of keratinous covering, extending to the horns and various features around the frill  (Hieronymus et al., 2009), although the types of tissues involved varies based on the surificial texture, degree of foramination, and microstructure and histology. Jack Horner has (in extension to a jointly-presented talk at the 2002 SVP conference) championed the idea that not just features but the entire upper skull was covered in a keratin sheet, largely due to the impression of blood vessels in the frill surface and facial region. These are typically not associated with a rugose surface, or highly-foraminated structure, but are comparable to the form of the supraorbital horns.

Dinosaurs with facial skin and keratin reconstructed after Horner et al. for Triceratops horridus (above) and Velociraptor mongoliensis after Paul (below).

Extending this to other facially-keratinous animals such as turtles and in some cases birds and several squamates becomes important for understanding even basic correlates to basic tissue reconstruction. I think, based on much of this, that no dinosaur with a toothed jaw margin has a comparative keratinous region on the same extent; there may be, as in ornithischians, a toothed upper and a beaked lower portion of the jaw, but much of this is speculative on the basis of hypothesis, and hasn’t been rigorously tested.

There are certainly some interesting things to say about the form and extent of keratin when there are no absolute direct analogues to the underlying bone in both turtles and birds, and this has me especially piqued due to my desire to reconstruct the cranial soft anatomy of oviraptorosaurs, oviraptorids in particular. The image at top represents part of a project in which I describe the various shapes of rhamphotheca and compare them to the underlying bone and its features in both turtles and birds. The ultrastructure alone is explored here, but the infra- and microstructure are just as important: the posterior portion of the mandibular symphysis in Caretta caretta above is more highly and deeply invaginated by nutrient foramina than elsewhere on the rostrum, while the arrangement of the foramina increases towards the margins (tomia) and posterior end, which the rhamphotheca overlies and surrounds. Further elements of this illustration include a medial view with rhamphotheca and outline to demonstrate the wrapped beak. Similarly, such features of the foramina exist on the upper skull, where the posterior palate is just as foraminated. When I show the medial view, the correlation of this will be apparent.

The problem of Paul’s illustration, however, goes beyond placing a beak without clear analogues to its presence. Rather, the difficulty lies largely in Paul’s reconstruction (followed above) where the ligamentous “lip” is divided on each side of the rostrum.

Jaws, lips and beaks in a lizard (left) and Velociraptor (middle and right). Showing the ligamentous band (blue) superimposed over the upper and lower jaws in both (left and middle) and divided by a rostral beak as proposed by Paul (right). The third image is considered improbable.

The ligament band, a thickened tissue that wraps around the oral margin on both upper and lower jaws, is a loose band attached by small muscles to the skin, not the bone (unlike in mammals, where the lips are formed by a large array of small muscles around the oral cavity and which are anchored both to bone and skin). In snakes, for example, this band is extremely elastic and allows extreme displacement of the mandibular symphysis when the snake swallows food larger than its own head, but otherwise in crocs and lizards it is much thinner and is externally indicated by the ring of lip, or labial scales. In no living reptile that I am familiar with does this band become divided (and for this I seriously have no reference, although I doubt the tissue becomes divided regardless); similarly, when a beak is present, there is neither a gum tissue present (the beak corresponds to a labial and a lingual plate in birds at least, and the lingual, or palatal plate in the upper jaw excludes gum tissue, as it excludes teeth), nor a ligament band. Thus, when placing a beak on the rostrum, this band would become divided, and no longer be loose.

Similarly, it is thus unlikely that it would be loose enough to depress below the jaw margin enough to allow the upper teeth to slide between it and the jawbone itself, when the jaw closes (as above), and even more problematic when dealing with taxa with such long and protrusive teeth as to make such a band impossible (for example, Dilophosaurus wetherilli, here). Such a bizarre improbability would result is a very, very divergent jaw anatomy than experienced so far and — unlike any living animal — permit an animal to have both gummy, lipped jaws and a beak.

I tend to think then that if a beak were to exist, it would need to supplant gum and ligament, and could not do so rostrally; when present, either such a ligament could not be used to “hide” teeth as in lips, or would be modified or lost. In some toothed animals with rostrally edentulous jaws (e.g., many ornithischians, therizinosauroids, some fossil birds), the teeth become fairly small and unlikely to fit over the lower jaw. While this raises the possibility of a “cheek” I will pretend I haven’t heard of this idea so I can have space for a follow-up post or two.

Hieronymus, T. L., Witmer, L. M., Tanke, D. H. & Currie, P. J. 2009. The facial integument of centrosaurine ceratopsids: Morphological and histological correlates of novel skin structures. The Anatomical Record 292:1370-1396.

This entry was posted in Biology, Paleobiology, Reconstruction and tagged . Bookmark the permalink.

2 Responses to Whether Beaks

  1. Pingback: On the Nature of Being a Pterosaur | The Bite Stuff

  2. Pingback: The Skimmer, Exploded | The Bite Stuff

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s