60 



is represented by the columella, thehomologue of the processus asceo dens 

 of amphibians, and this process arises from the processus pterygoideus 

 palatoquadrati and not from the body of the quadrate, this latter cartilage 

 lying posterior to the processus pterygoideus and wholly independent 

 of it. Whether the dorsal end of this independent quadrate cartilage 

 represents the processus oticus or the processus basalis is not said by 

 Gaupp, but it would seem as if it must represent the former process, 

 for the quadrate cartilage lies definitely lateral to the foramen faciale, 

 and the nervus facialis issues along its posterior margin. The quadrate 

 cartilage must, in that case, represent the ganoidean and teleostean 

 quadrate plus the post-trigeminus, or mandibular portion of the lateral 

 wall of the trigemino-facialis chamber of those fishes, and its separation, 

 in Lacerta, fi'om the ascending and pterygoid processes, both of which 

 I have assumed to belong to the premandibular arch (Allis, 1914 b), 

 is wholly natural. Because of this separation of these two parts of the 

 lateral wall of the chamber, a large open space is left, which may, in 

 certain reptiles, subsequently become closed by membrane (Fuchs, 1910), 

 and this membrane may then naturally be invaded and replaced, to 

 a greater or less extent, by certain of the adjacent dermal bones 

 (descending process of the frontal). 



In the crocodile (Shiino, 1914), the conditions here differ, in 

 principle, from those in Lacerta, only in that the palatoquadrate 

 remains entire, instead of being separated into two parts. The processus 

 oticus is largely developed, and its dorso-anterior portion is produced 

 into a stout processus orbitalis, while the processus ascendens is only 

 slightly developed ; these three processes forming the lateral wall of 

 the trigemino-facialis chamber. A floor to the chamber is apparently 

 wholly wanting, for Shiino says that there is no processus basiptery- 

 goideus. The processus baistrabecularis would seem to resemble 

 somewhat, in its relations to the ramus palatinus facialis and the 

 arteria carotis interna, the lateral wall of the myodome of Amia and 

 teleosts. The arteria temporo-lateralis traverses the trigemino-facialis 

 chamber and is apparently the homologue of the carotis externa of 

 ganoids and teleosts, and it is to be noted that while this artery is 

 still in the chamber it separates into two parts which later reunite; 

 this recalling the hyo-opercularis artery of ganoids and teleosts. The 

 arteria orbitalis gives off several ciliary branches, and one large branch 

 is said to break up, in the orbit, into an arterial net- work; this sug- 

 gesting the homology of some part of the artery with the arteria 



