234 Reviews — Palceozoic Fishes. 



whole, Prof. Cope is inclined to accept the conclusion arrived at in 

 the British Museum Catalogue of Fossil Fishes ; namely, that the 

 Arthrodira are extremely specialized Dipnoi. "The nuchal portiou 

 [of the head-shield of MacropetaJicli'.hijs] with its lateral nuchal 

 elements is represented by the cartilaginous mass which extends 

 posterior to the median occipital bone in Ceratodns, in which this 

 region has very much the form of the nuchal shield in Macro- 

 petnlichthys, although it is relatively shorter. The chordal groove 

 with its descending laminae resembles much the produced occipital 

 bone of Lepidosiren. The parasphenoid in both Lepidosiren and 

 Ceratodns is produced posteriorly abnormally, and it is only necessary 

 to imagine this part to be reduced to its normal length to have the 

 conditions found in Macropetalichthys. The broad parasphenoid and 

 vomer remind one of that of Cte)wdus. As I have shown that 

 Macropetaiichthys is allied to Dinichthi/s, we can add in favour of 

 the supposition of affinity to the Dipnoi the peculiar dentition of 

 tliat genus. The ectetramerous structure of the dorsal fin shown 

 by Von Koenen and Traquair to exist in Coccosteus, and shown to 

 be probably present in Dinichthys by Newberry, are in favour of the 

 Dipnoan theory." 



Prof. Cope evidently continues in the belief that Pterichthys and 

 its allies have no connexion whatever with the Arthrodira ; and 

 tlie researches of Dr. Eohon quoted above come as a welcome con- 

 firmation of this view. Dr. Eohon points out that the histological 

 structure of the shield of Pterichthys is very different from that of 

 Coccosteus, and closely similar to that of Pteraspis, Tremataspis, and 

 their allies. This result also confirms the arrangement of the early 

 fishes in question in the second pai't of the British Museum Catalogue. 



Prof. Cope's remarks on the limbs of Holonema and Megalichthys, 

 however, are far from satisfactory. In the first place, we venture 

 to re-affirm that the so-called dorsal shield of Holonema is really the 

 ventral shield turned the wrong way forwards ; and the genus 

 belongs to the Arthrodira, not to the Ostracodermi. The limb 

 referred by Prof. Cope to Holonema is the distal segment of the arm 

 of Bothriolepis, originally named Steriacanthtis bj^ Leidy. It is stated 

 that " the spine differs from tliat of both Bothriolepis and Pterichthys 

 in being without complete segmentation ; " on the other hand, we 

 may remark, the distal half of the arm of Bothriolepis is nearly 

 always incompletely segmented. With regard to the paired fins of 

 Megalichthys, which are said to " approach those of the Arthrodira 

 very distinctly," we venture to assert, from a knowledge of other 

 Osteolepidaj, that the apparent simplicity of the arrangement of the 

 cartilages in Prof. Cope's specimens is due to imperfect preservation, 

 while the paired limbs of the Arthrodira are far too imperfectly 

 known to admit of comparison. 



The final result of Prof. Cope's researches in the Crossopterygian 

 ganoids is of great interest, and briefly summarized in an amended 

 classification, which we propose to consider elsewhere on a future 

 occasion. The Professor is at last converted to the belief that the 

 PalteoniscidEe and Platysomidce are closely related to the primitive 



