﻿XX INTRODUCTION. 



although there are no paired appendages in the former genus, we 

 are inclined to think that another noteworthy point of resemblance 

 occurs in the appendicular skeleton, the rows of plates in the 

 paddles of the Asterolepidas being an extreme modification of the 

 arrangement observed in the azygous fin-membranes of the Cepha- 

 laspidoe, and markedly different from the actinotrichian development 

 by which the fin-rays of ordinary fishes arise. Even the support of 

 the anterior border of the dorsal fin oiPterichthys is not a true spine, 

 but merely a longitudinally bent (perhaps primitively double) scale. 



DIPKOI. 



Concerning the evolution of the Dipnoi, palaeontology as yet 

 affords no information. So long ago as the Devonian period, there 

 were members of the subclass agreeing precisely with the existing 

 Ceratodus in the development of the fins and the axial skeleton of 

 the trunk. At that remote period, too, the chief part of the 

 dentition had assumed the form of great plates upon the splenial 

 bones and the palate ; and the principal difference between such a 

 type as Phaneropleuron and the existing genus just mentioned 

 seems to consist in the comparative fewness of the cranial roof- 

 bones in the latter and the absence of membrane-bones on the 

 margin of the jaw. The typical Dipnoi of the Devonian period had, 

 indeed, already become more specialized than any known in later 

 times ; Dipierus exhibiting differentiated dorsal fins and a hetero- 

 cercal tail. 



The latter fact is of all the more interest when the tendency of 

 modern research in regard to the Coecosteus-]ike fishes is taken 

 into consideration. According to existing diagnoses, these fishes 

 must be assigned either to the Dipnoi or to the Teleostomi ; and 

 the extremely specialized character of their paired fins, so far as 

 known, proves that, wherever they be placed, they occupy a com- 

 paratively high position. If they are Teleostomi, they pertain to 

 the Actinopterygian order, and hence ought to exhibit a well- 

 developed hyomandibular bone. At least, in every undoubted 

 Actinopterygian Teleostome possessing ossifications equal in extent 

 to those of Coccostens and its allies, the hyomandibular bone is both 

 large and considerably ossified. In the extinct group now under 

 discussion, however, such a bone is not exhibited even by the most 

 exquisitely preserved specimens. On the other hand, all appear- 

 ances in the crania of Dinichihys and allied genera from the Waverlv 



