400 MESSRS. HANCOCK AND ATTHEY 



formed expansion is much produced, while the anterior angle 

 is only slightly produced. The frontal portion (the pre- 

 sphenoid) is rounded, inclining to conical at the extremity, 

 and fits in between the divergent bones that support the dental 

 plates. The lozenge-formed expansion lies partly behind 

 these bones ; and the elongated posterior extension (the basi- 

 sphenoid) is continued for a considerable distance further back, 

 in the large species for nearly five inches. It is therefore pretty 

 clear that in Dipterm, in which the sphenoid reaches only a 

 short way behind the extremities of the palato-pterygoids, the 

 head is proportionately short in comparison with that of 

 Ctenodus, in which it must be much elongated. 



We have in our possession numerous sphenoids, belonging 

 to five or six species, three of which demonstrably are those 

 of C. tuberculatus, C. obliquus, and C. elegans, respectively. 

 They are all verj^ similar in character, varying only a little in 

 the proportions of the parts. The largest are seven or eight 

 inches long ; the smallest, that of C. elegans, is only half an 

 inch in length ; the usual size is five or six inches. The basi- 

 sphenoid at its junction with the lozenge-formed expansion is 

 usually thick and nearly circular ; elsewhere it is flattened. 



In Dipterm, too, the vertebrae are ossified ; but there is 

 nothing to show that this is the case in Ctenodus. Indeed the 

 total absence of any appearance of vertebras in the specimen of 

 C. elegans before referred to is a pretty good proof that in this 

 genus the central axis of the skeleton was cartilaginous. 



The above distinctive features will perhaps be considered 

 sufficient to warrant the generic separation of these two forms, 

 notwithstanding their evidently close relationship — and this 

 without referring to the minute structure of the dental plates, 

 which exhibits nevertheless some diversity in character. 



In proof of the relationship of the two genera we have only to 

 look to the general form of the oral armature, and to the manner 

 in which the dental plates are placed in the mouth. We have 

 already noticed the similarity of the palato-pterygoid bones to 

 which the upper dental plates are attached, and have pointed 

 out that, while in the one genus the bones are distinct, they 



