Sarcopterygii: Overview 

Orbiting the Gnathostomes

One of the really extraordinary things which has happened in paleontology in the last few years is the progress made toward filling what is possibly the biggest hole in the whole vertebrate story: the origin and basal condition of the gnathostomes.  Progress on the jawless side of the great divide between "Agnatha" and Gnathostomata has been slow.  In fact, as discussed in the section on thelodonts, much progress has been in reverse.  That is, we are unlearning things we thought we knew but turned out to be wrong.  

By contrast, those who have been working backward from the gnathostome side have been moving by leaps and bounds, finding all manner of strange things in the process.  Among these are weird new synapomorphies of the Gnathostomata.  It turns out, very probably, that basal gnathostome characteristics included:

a) fin spines on both dorsal and lateral fins (once thought to be unique to acanthodians); 

b) parasymphysial tooth whorls -- buzz-saw blades of multiple teeth at the front of the lower jaw; and

c) eye-stalks.

Obviously, the first fish with jaws -- bristling with spines, bulging, possibly mobile eyes, and a radial saw blade of teeth projecting from the front of the mouth -- looked very different from the mental picture we may have had only a decade ago.  In fact it is hard to imagine what line of jawless fishes could possibly have produced such monstrosities.  However, monstrous or not, it is also increasingly clear (if it was ever in doubt) that the gnathostomes are monophyletic and that one Early Devonian jawed fish was very much like another in many details.  Here, we focus on the gnathostome orbit.  It really has nothing much to do with sarcopterygians in particular.  But, there is no really good place to take this up elsewhere; and the discovery of an eyestalk in the basal sarcopterygian Psarolepis is one of the more important pieces of data from the last decade.  

Orbits are where we put eyeballs.  However, fossilized eyeballs are, to say the least, infrequently found.  What can be worked out is the pattern of muscles which moved the eyes, the extrinsic eye muscles.  These leave their mark various ways, some details of which are discussed below.  First, we need to get a handle on the basic pattern, more or less as presented in any comparative anatomy text.

That pattern is shown in the first image.  The orbit of gnathostomes sits in a comfortable circular nest just in front of the otoccipital half of the braincase, where it begins to rise from the depths of the pituitary (the sella turcica), past the trigeminal notch, towards the distant peaks of the tall otic capsules.  See, generally, images and discussion of the Braincase.  The orbit is supported internally and sometimes below by the  sphenethmoid part of the braincase and normally connects to some element of the nasal cavity anteriorly.  

The basic pattern of external eye muscles is a diamond shape.  The anterior sides of the diamond are formed by the superior and inferior oblique muscles, while the posterior sides are the superior and inferior rectus muscles.  These are supplemented by the horizontal medial rectus, and the vertical (and confusingly named) lateral rectus.  Primitively, the orbit has two large, separate, and more-or-less centrally located holes.  These foramina are the passages for the optic nerve (Cranial nerve II) and the eyestalk.  In all gnathostomes, all but two of the extrinsic eye muscles are controlled by the oculomotor nerve (Cranial Nerve III).  The exceptions are the superior oblique, which is enervated by the trochlear nerve (Cranial nerve IV), and the lateral rectus, which is enervated by the abducens (Cranial Nerve VI).  Perhaps significantly, these three cranial nerves (III, IV and VI) are completely dedicated to controlling the extrinsic eye muscles and do nothing else [2].  The second figure shows how this looks in a shark -- after an unfortunate encounter with a table saw.

The third figure shows an idealized fossil, a compromise between various sources.  Note how the muscle origins can generally be located by reference to depressed areas, myodomes, which accommodated the proximal muscle mass [1].  The eyestalk is supported by the basisphenoid pillar (also called the postorbital pila), a process which rises from near the base of the basipterygoid process.  It is frequently reinforced by a lateral buttress connecting the pillar back to the postorbital bar.  Above the buttress is the oculomotor myodome, with the foramen for the oculomotor (III) nerve.  Below it is a depressed region containing the pituitary vein and the jugular canal.  Somewhere in the region between them is the foramen for the Abducens (VI) nerve, and the region is said to be the abducens myodome.  However, the oculomotor nerve has a complex trajectory, with multiple points of contact, and the exact topology cannot be determined in fossils.  At least in the coelacanth Latimeria, a branch of the oculomotor  enters the orbit anteriorly to support the inferior oblique muscle.  The trochlear (IV) myodome and foramen are located dorsally, sometimes quite far from any other foramina.  This makes it a good landmark in a variety of forms.

CONTINUED ON NEXT PAGE

[1] Distally -- away from the myodome -- the muscle fibers spread out over the surface of the eyeball.

[2] Yes, there are exceptions -- interesting ones at that.  In Latimeria, the abducens also ennervates the basicranial muscle which flexes the joint between anterior and posterior braincase.  I also understand that the abducens controls the retractor bulbi which, when present, retracts the entire eyeball via the eyestalk.