148 CHIMyliROID FISHES AND THEIR DEVELOPMENT. 



bizarre features of Myriacanthus, Chimaeropsis, and Squaloraja; no highly special- 

 ized plates and spines in the head region, no spine-shaped frontal clasping organ, 

 no presymphyseal element, and no second pair of "vomerine" plates. Among 

 recent forms, Callorhjmchus, a Cretaceous genus, has probably retained in most 

 regards the striking characters of its Mesozoic kindred. And it is not to be 

 wondered at, therefore, that its developmental features appear more conservative 

 than in other genera. On one side of this early genus we may place Chima^ra, 

 which, as we have seen, is in many ways a highly modified form; and on the 

 other side would be arranged Harriotta and Rhinochimsera, similar to one another 

 outwardlv, but (on the evidence of dental characters) long separated from a common 

 ancestor. 



It yet remains to consider the probable relationships of the earlier forms. It 

 is clear, first of all, that in the Jurassic epoch there existed three distinct t3'pes of 

 Chimseroids. One, as we have noted, is that of Ischyodus and its allies, from 

 which unquestionably all recent Chimseroids are descended. The second, Squal- 

 oraja, represents an aberrant and terminal group; it is to its kindred as is Pristi- 

 ophorus to sharks. On the other hand, one must admit that it shows certain 

 characters* which all}' it to the stock from which Ischyodus-like forms must have 

 arisen. The third Jurassic type, represented by Myriacanthus and Chimaeropsis, 

 is the most difficult to interpret. From present data it can hardly have pictured 

 the ancestral line of modern Chimaeroids, for from what we already know of the 

 elaborate dermal plates of the head and its "trachyacanthid" spines, we infer that 

 it was already too highly specialized to have had the evolutional vigor to give rise 

 to forms in which shagreen-like conditions again occur, for such a series would 

 present an analogy not as close, c. g. , to the descending line of the sturgeons as 

 to the line of the Cestracionts, in which the modern form is related only collaterally 

 to the elaborately spined and heavily plated genera of the late Palaeozoic. Espe- 

 cially puzzling are the dental characters of Myriacanthids; for how are to be 

 interpreted the symphj'seal chisel-shaped element and the anterior pair of subnasal 

 plates? One might readily suggest that the former element was developed on the 

 copula of the mandibular arch — a suggestion which bears with it a greater shade of 

 probability when we consider the size and importance of the mandibular copula 

 as recently described b}- both Schauinsland and the younger Fiirbringer. And 

 following a similar line of speculation we might maintain that the "vomerine" 

 plates were developed on the phar^mgobranchial element of the jaw arch, just as 

 the palatine plates were developed on the next lower (epibranchial) element. In 

 support of this hypothesis we may note that, as in Chimaera a phar\'ngobranchial 

 element is present in the hyoid arch, a similar serial element appears also to have 

 been present in the mandibular arch ((/'. figs. 1 10 and 1 1 1). A second hypothesis — 

 hypothesis may be a little too dignified a term — is that the "vomerine" and "pre- 

 vomerine" plates of Myriacanthus represent the palatine plates of premandibular 



*E. g., number and disposition of dental plates, clasping organs, integumental defenses. 



