Double series cross-post here.
It should be no surprise to some of you that oviraptorosaurs have one of the strangest jaw articulations among dinosaurs. The truth is, they have one of the most unusual jaw articulations among diapsids, or even amniotes. This includes the inverted, modified jaw of modern mammals, or the angled, widening jaws of many squamates (including mosasaurs).
Most oviraptorosaurs, excluding the “toothed” ones (which form a paraphyletic grade at the base of Oviraptorosauria) possess a unique jaw articulation. And by unique, I mean that there are virtually NO diapsids (or amniotes, for that matter) with a jaw articulation like them.
The typical jaw articulation for any amniote consists of a condylar “ball” on the cranium, almost always based on the quadrate (even in modern mammals, this articulation while separated from the jaw itself persists in the inner ear, forming the contact between the malleus and the incus bones; the actual jaw articulation shifts to the braincase and the dentary, and develops an “inverted” structure, a socket-in-ball as it were). The mandibular side of this articulation is almost always a socket, with a set of bony walls forming a “cage” for the ball of the cranium to fit into.
In typical animals, the socket is not fully developed: there is sometimes only a wall on two, maybe three sides, leaving one open. This articulation is contained by ligaments and a soft-tissue sac known as a bursa. Movement in this joint is restricted both by the elasticity of this ligament system and the sac, but also the presence of the walls and the shape of the articulation. When this articulation is long, or wide, it is generally assumed this indicates freedom of movement in a particular direction:
For example, a typical archosaurian jaw has generally only two distinct walls: An anterior/mesial wall and a lateral/labial wall. The posterior/distal margin generally does not possess a barrier, while occassionally the medial/lingual one will (but always shorter than the lateral/labial one.
When these walls are shallow, or nonexistent, the socket is also very shallow, and this usually indicates a broad range of movement. This is seen, for example, in testudinine jaws, where a broad anterior wall is not restrictive of movement as much as the complete lack of any others, and the articulation is broad and elongated. In most birds, an anterior and lateral wall exist, but there is virtually no boundary to the medial or posterior extent, and these articulations indicated some freedom of movement that is observable. Humans, like other living mammals, have a jaw with almost no boundaries around the socket, and have a broad range of movement.
Movement in the mandible is described using a few technical terms:
Orthal — Indicates movement restricted to the sagittal plane. Most carnivores use this form of jaw action predominantly.
Transverse — Indicates movement to the left and the right. This usually shows up in most mammals, and is often required for precise bilateral occlusion (occlusion between the upper and lower jaws fitting perfectly, but only on one side at a time), and you see this in carnivoran mammals, such as the cat.
Propalinal — Indicates movement to the front and back. This usually occurs in mammals with highly complex, cuspidate molars (such as insectivorans, multituberculates, etc.)
The jaw articulation is directly related to movement in many cases: A wide articulation generally allows transverse movement, permitting the articulations to slide to the side; this usually involves rotating one articulation to the rear, while the other slides to the side, and indicates similar ability for the jaw muscles on one side of the skull to act independent of the other side.
An elongated jaw articulation usually indicates propalinal movement, although the limits are hard to infer, unlike the general process in transnversal movement. This is due to propalinal movement being generally defined by muscular elasticity, or limits external to the articulation. Obviously, some propalinal animals have limits in their jaw articulations, but these are very few. In these animals, such as placodonts or turtles, it exists only on the anterior margin (or partially to the lateral side of the socket); the other extents are open, confined only by the ligaments and bursa. In dicynodont synapsids, and in a group of oviraptorosaurs, this jaw articulation is, amazingly, biconvex; that is, both the cranial and mandibular halves of the articulation are condylar.
The biconvex jaw articulation presents no bony limits in dicynodonts or oviraptorosaurs, save for a median, intercondylar ridge, which probably restricted the articulation to a propalinal/orthal system, preventing transversal movement. This all permitted a broad range of movement, depending on the shapes of the mandibles:
Here, the position of bony stops in the cranium prevent the mandible from being positioned any further cranial (above, left) because the coronoid eminence cannot move through the articulation between the ectopterygoid & jugal; the jaw cannot move any further posteriorly because the jaw articulation is at its limit. These represent the absolute limits of movement possible, although it is an assumption to say that these represent the natural limits of movement: It is almost certainly incorrect, and the natural limits would be mitigated more due to extension of the cranio-mandibular musculature (the m. adductor mandibulae and m. pterygoideus groups) than strictly by the contact between various bones.
These issues pose many questions for the mechanical qualities of the jaws of oviraptorosaurs, and while recently some work (still in press ) has utilized many of these assumptions, there is more to be said on the mechanics of the jaw itself, rather than properties of the bite and exerted forces.
 Sakamoto M. 2010. Jaw biomechanics and the evolution of biting performance in theropod dinosaurs. Proceedings of the Royal Society of London, B (Published online before print June 9, 2010, doi: 10.1098/rspb.2010.0794).