PALATOQUADRATE WITH NEQROCRAXIUM IN CCELACAXTHIDS. 43 



processus basalis ; and that the nerve that innervates the levator, 

 the secondary afferent pseudobranchial artery, and the vena 

 jugu]aris all lie actually, or morphologically, between the two 

 processes. 



In Scomber (Allis, 1903) and the mail-cheeked fishes (Allis, 

 1909) these two processes (or flanges developed in relation to 

 them) have acquired contact v/ith the hj'omandibula and become 

 firmly attached to it, the processus basalis with its internal 

 surface and the processus metapterygoideus with its external 

 surface. Bub, notwithstanding this marked difference in the 

 direction of the two processes, and also in their relations to 

 the cranium and the hyomandibula, the levator arcus palatini, 

 the nerve that innervates it, and an artery that corresponds to 

 the secondary afferent pseudobranchial arteiy of Amia, still all 

 lie actually, or morpliologically {CoUus), between them, and the 

 adductor mandibulaj has its origin on the external suiface of the 

 processus metapterygoideus and not at all on the processus 

 basalis. The vena jugularis, also, still lies morphologically 

 between the two processes, for the dorso-mesial edge of theprocessus 

 basalis is connected, by tissue, with the lateral edge of the basis 

 cranii. 



In Amiurus the conditions appear, at fii'st sight, markedly 

 different from those in the fishes above referred to, but they axe, 

 nevertheless, quite certainly strictly homologous. In a 32 mm, 

 specimen of this fish. Kindred (1919) shows the pterygoquadrate 

 and hyomandibular cartilages completely fused with each other 

 and forming a V-shaped mass with the point of the V directed 

 ventrally. The h3^omandibula, has begun to ossify on the posterior 

 limb of the V", the metapterygoid on the anterior limb, and the 

 quadrate on the ventrally directed point. Nothing whatever is 

 shown in the hollow of the V, but it must certainly be spanned 

 by membrane, for it is in exactly this place that the metaptery- 

 goid bone first appears in Amia, as above stated, and in the adult 

 Amiurus the corresponding place is occupied by an anteriorly 

 projecting flange of the hyomandibula. which sutures anteriorly 

 with the hind edge of the metapterygoid and ventrally with the 

 dorsal edge of the quadrate. Along the dorsal edge of this 

 flange, and of corresponding length, there is a horizontal ridge 

 on the hyomandibula, and the levator arcus palatini is inserted in 

 part on the anterior end of the ridge and in poxt along the full 

 length of its dorsal surface (McMurrich, 1884). Certain of the 

 deeper fibx^es of the musculus adductor mandibulse have their 

 origins on the latero-ventral surface of this ridge, the superficial 

 portion of the muscle extending upward external to the ridge. 

 The anteriorly projecting flange of the hyomandibula of this fish 

 thus cori-esponds both topographically and functionally to the 

 posteriorly projecting flange on the hind edge of the metaptery- 

 goid process of the fishes just above considered, and is quite 

 certainly the homologue of that flange; and that it has become 

 a flange on the anterior edge of the hyomandibula, instead of on 



