﻿VI INTRODUCTION. 



himself to impart a life-like aspect to tlie head in his restored 

 figure of the Dipnoan and Crossopterygian genera Dipterus and 

 Diplopterus l . In the interpretation of fins, again, close comparison 

 with existing fishes led to some noteworthy fundamental errors, 

 such as the restoration of the dorsal fin of Coccosteus 2 , as if it per- 

 tained to the most modern specialized type ; and many other cases 

 might he cited of an essentially similar character. Quite in modern 

 times, indeed, the reiterated association of the Cephalaspidae, Astero- 

 lepidae, and Coccosteidae with recent Sturgeons by Owen 3 ; the still 

 more elaborate comparison of the Coccosteidae with existing 

 Siluroids by Huxley 4 ; and the quite recent adhesion to this Silu- 

 roid theory by dewberry 5 — all must now be regarded as resulting 

 from too narrow a conception of the limits within which certain 

 minor skeletal characters may occur. The ascertained facts of 

 embryology and the well-established broad principles of palaeonto- 

 logy are now at the disposal of the investigator ; and it is hoped 

 that a detailed review of the whole subject, such as is attempted for 

 the first time' in the present volume, may tend towards a more 

 philosophical understanding of the early representatives of the class 

 under consideration. 



The first essential fact requiring special emphasis, at the outset, 

 seems to be, that although the Palaeozoic fishes certainly belong to 

 the most generalized great divisions of their class, a large proportion 

 of the known types are extremely specialized members of these 

 divisions. This is clearly indicated by the characters of the fins in 

 many forms. Just as in the existing fauna, the most striking 

 examples of extreme specialization are comprised within the dominant 

 higher groups of Actinopterygian Teleostomi, so in the Palaeozoic 

 fauna the same instances of development occur almost exclusively 

 in the then dominant orders of the Ostracodermi, Elasmobranchii, 

 and Crossopterygian Teleostomi. If in the one case specialization 

 proceeds sometimes almost exactly in the same manner as it does in 

 the other, everything seems to point to the conclusion that this is 



1 L. Agassiz, ' Poissons Fossiles du Vieux Grres Eouge,' (1844), pi. E. 



2 L. Agassiz, ibid. pi. vi. fig. 3. 



3 E. Owen, ' Paleontology,' ed. 2 (1861), p. 139 ; and ' Anatomy of Verte- 

 brates,' vol. i. (1866), p. 12. 



4 T. H. Huxley, " Preliminary Essay upon the Systematic Arrangement of 

 the Fishes of the Devonian Epoch " (Mem. Geol. Surv. dec. x. 1861), p. 29. 



5 J. S. Newberry, " Tbe Palaeozoic Fishes of North America " (Mon. U. S. 

 Geol. Surv. no. xvi. 1889), p. 141, et passim. 



