PALATOQUADRATE WITH NEUROCRANIUM IX CCELACANTHIDS. 45 



The mnscnlus levator arcus palatini of the adult Lepidosteus is 

 said by Luther (1913) to have its origiii on the sphenotic, and to 

 run from there antero-mesially to have its insertion on the 

 metapterygoid bone and adjacent portions of the palatoquadrate 

 cartilage. Luther says {I.e. p. 13) that: " Dabei umgreift der 

 Muskel das Metapterygoid vor dem Basipterygoid-Gelenk und 

 reicht audi an der Medialseite bis zu etwa |- der Breite des 

 Knochens hinauf." This would seem to mean that the muscle 

 had its insertion in considerable part on the internal surface of 

 the processus basalis, which would be so unusual for a levator 

 muscle that I have looked it up in transverse sections of an 

 80 mra. specimen of this fish. In this specimen the muscle, as it 

 approaches the hind end of the palatoquadrate is wrapped arovmd 

 (" umgreift ") the lateral edge of the processus basipterygoideus, 

 and when it reaches the hind end of the processus basalis, that 

 part of the muscle that lies ventral to the processus basiptery- 

 goideus is inserted on the lateral surface of the processus basalis, 

 this surface of the latter process being presented ventro-laterally. 

 The processus metapterygoideus lies against the ventral surface 

 of this part of the muscle, partly imbedded in it, and apparently 

 gives insertion to some of its fibres, but the larger part of the 

 fibres of the muscle continue onward and are inserted on the 

 external surface of the palatoquadrate anterior to the bases of the 

 two processes, none of them being inserted on its internal surface. 

 That part of the adductor muscle that Luther calls the musculus 

 adductor maiidibulse postorbitalis lies ventral, and hence external, 

 to the processus metapterygoideus, and certain of its fibres have 

 their origins on that process. The action of the levator muscle 

 is said by Luther {I.e. p. 61) to be to pull the palatoquadrate 

 almost directly laterally, the motion at the basipterygoid joint 

 accordingly being a sliding one, in a latero-mesial direction, and 

 this is in accord with the position of the long axis of this joint, 

 which is practically at right angles to that of the long axis of 

 the hyomandibulo-cranial articular joint. The nerve that inner- 

 vates the levator muscle penetrates it from its external surface, 

 and hence lies morphologically between the metapterygoid and 

 basal processes, as it does in the fishes above considered. 



The processus basipterygoideus of Lejndosteus is said by Parker 

 (1882, pp. 453 k 461) to be developed in part from the trabecula 

 and in part from the parachordal. Veit first (1907) concluded 

 that it is developed from the hind end of the trabecula and 

 corresponds to a part of the floor of the orbit of the Selachii, but 

 he later (1911) says that it arises in connection with the trabecula 

 and the floor of the trigemino-fa,cialis chamber. I first concluded 

 (Allis, 1909) that it is of trabecular origin and, later (1913), that 

 it is the homologue of the floor of the orbital opening of the 

 myodome of Amia ; but it is to be particularly noted that there 

 is, in Lepidosteus ^ no parasphenoidal leg of the alisphenoid I'elated 

 to the process, as there is in Amia, and that there is no membrane 

 that I can recognize replacing it. Veit says that the process 



