46 MR. E. P. ALLIS ON POSTORBITAL ARTICULATION OF 



develops late in Lepidosteus, and he considers its articulation with 

 the metapteiygoid to be a secondary acquisition, and Luther 

 holds the same opinion. The ventral surface of the process, its 

 aiiterior edge, and the adjacent portions of its dorsal surface are 

 covered and supported by a process of the parasphenoid which 

 Avould seem to correspond to an anterior portion of the ascending 

 process of the parasphenoid of Amia, and it is apparently this 

 process of the bone that alone gives articulation to the meta- 

 pterygoid in Osteoglossum, as explained below. The process of 

 Lepidosteus lies ventral to the ramus ophthalmicus profundus and 

 the vena jugularis, and if it be the homologue of the floor of the 

 orbital opening of the myodome of Amia, as I concluded, it would 

 seem as if it must lie, morphologically, between the profundus 

 and trigeminus nerves, for its homologue in Amia supports the 

 parasphenoidal leg of the alisphenoid, which actually lies between 

 those two nerves. In its relations to these nerves this process 

 thus differs radically from the postorbital process, which always 

 lies, so far as I know, posterior to all branches of the nervus 

 trigeminus, and usually between the foramen for that nerve and 

 the secondary, or definitive, foramen for the nervus facialis. The 

 processus metapterygoideus also lies morphologically between the 

 trigeminus and facialis nerves, while the processus basalis, which 

 articulates with the processus basipterygoideus, should have, 

 morphologically, the same relations to those nerves that the 

 processus basipterygoideus has. 



In Osteoglossum a laterally projecting process of the para- 

 sphenoid articulates with a semicylindrical groove on the dorsal 

 edge of an antero-dorsally projecting portion of the metapterygoid 

 (Bridge, 1895), vs^hich, as it has exactly the position of the pro- 

 cessus basalis of Aviia, is evident!}'' that process, and when the 

 palatoquadrate swings outward and upward, under the action of 

 the levator arcus palatini, the groove on the metapterygoid has a 

 sliding motion on the process. Posterior to this process, there is 

 a short dorsally projecting process of the metapterygoid -which 

 lies against the external surface of the hyomandibula, and hence 

 evidently represents the processus metapterygoideus. The 

 laterally projecting process of the parasphenoid has its origin 

 from the base of a pretrigeminus portion of the ascending 

 process of that bone and corresponds to that part of the ascending 

 process of Lepidosteits that underlies and supports the basiptery- 

 goid process of that fish, and hence is, functionally, a processus 

 basipterygoideus. It corresponds also to that anterior portion of 

 the ascending process of the parasphenoid of Amia that underlies 

 the floor of the orbital opening of the myodome of the latter fish, 

 and the pretrigeminus portion of the ascending process of the 

 parasphenoid of Osteoglossum corresponds to one of the two legs 

 of the alisphenoid bone of Amia, but which one is uncertain, for 

 the relations of the nerves and the vena jugularis to it are not 

 given. 



The conditions in Heptanchus aiid Hexanchus I have not 



