214 DB. W. G. EIDEWOOD ON THE CRANIAL OSTEOLOGY OP 



osteans not genetically related, whenever, for some reason or 

 other, a reduction in the size of the mouth takes place. 



Of the other characters mentioned in the summary on p. 201 

 as distinctive of the Mormyroid skull, these are all characters of 

 specialization, and throw no light on the question of affinity 

 with the Albulidse. The presence of a distinct symplectic in 

 Albula, and a distinct entopterygoid, the presence of teeth on 

 the vomerine, palatine, and pterygoid bones, the freedom from 

 fusion of the palatine with the vomer, and of the right with 

 the left premaxilla, the presence of a surmaxilla, the absence 

 of reduction of the subopercular and of the hypobranchial 

 skeleton, the possession of as many as 15 branchiostegal rays 

 instead of 4-8, the existence of a separate glossohyal, the 

 normal character of the urohyal, and the possession by the 

 post-temporal of an opisthotic limb attached to a distinct 

 opisthotic bone, all these are characters that one would naturally 

 ■expect to see in any Malacopterygian fish of more archaic con- 

 stitution than a Mormyroid. 



On the other hand, the large lateral cranial foramen, covered 

 hj a thin scale-like supratemporal, so constant in the Mor- 

 myridse, and the large paired tendon-bones that project down 

 from the side of the second basibranchial are not represented 

 in the Albulidse; while the roofing of the posterior temporal 

 fossse and the presence of subtemporal fossae (lying ventral to 

 the posterior part of the facet for the head of the hyomandibular), 

 characters which Albula possesses in common with the Elopidse 

 and Cyprinidse, are not to be found in the Mormyridse. The 

 inflation of the hinder part of the base of the cranium of Albula 

 is also not recognizable in the Mormyridse, although, curiously 

 enough, it occurs in Notopterus. 



Concerning the genera JSfotopterus and Hyodon there is but 

 little to be said, except that the latter possesses a greater 

 proportion of primitive characters. Of the forms described in 

 the present paper, there can be little doubt that the Mormyridse 

 are the most specialized, and ^yo<^o« the least specialized; but 

 the close study of the skulls of these fishes does not lend support 

 to a view of relationship recently expressed by Boulenger. On 

 p. 116 of his book ' Les Poissons du Bassin du Congo,' 1901, 

 he writes : — " Les Notopterides me semblent occuper vis-a-vis 

 des Hyodontides une position analogue a celle qu'occupent les 



