659 



are the product of consolidation of placoid scales. In the Elasmo- 

 branchii the skull is only represented by the chondrocranium, in the 

 skin covering the skull and body very small placoid scales are found. 

 By the union of these scales osseous plates are formed, which at first 

 vary much in shape, size and number, and are totally free from the 

 chondrocranium, but gradually become more enlarged and reach more 

 or less extensive qgnnection with the primitive skull. In the skull of 

 the Sturgeon, for instance, we find all the different stages represented. 

 In half grown specimens it is easily possible to separate the contin- 

 uous dermal plates from the chondrocranium. In the anterior region 

 of the skull the dermal bones are very small and numerous, but in 

 the middle and posterior region the scales have enlarged and it is 

 possible to homologize some of the bones with the elements of the 

 higher forms. The squamosals and prefrontals which can be distin- 

 guished have reached closer connection with the chondrocranium by bony 

 processes growing from the base of the bone into the cartilage. Later 

 in the series of fishes the dermal plates become more definite in 

 number, size and relations, and from this conditions the skull of the 

 Stegocephali must have originated. In the Stegocephali the chondro- 

 cranium was still greatly developed and most of the dermal bones 

 were free from it, but the number of elements is nearly absolutely 

 constant. 



That the Stegocephali did take their origin from a group of Fishes 

 is evident. The question to examine is: Which was this group? 



In the Devonian proceeding the Carboniferous, where the Stego- 

 cephali first appear, we have the following groups of fishes: the 

 Elasmobranchii — including the Holocephali — Ostracodermi, Dipnoi, 

 Crossopterygii and Chondrostei. The Elasmobranchii have no dermal 

 ossifications in the skull, they are therefere out of the question; the 

 Ostracodermi have of course nothing to do with the Stegocephali ; we 

 also can exclude the Chondrostei represented by Cheirolepsis. Only 

 the Dipnoi and Crossopterygii remain. 



The dentition of the Dipnoi is already so much specialized in 

 the Devonian, that it is impossible to derive the Stegocephali from 

 them. 



We have now to consider the Crossopterygians. The Crossopterygii 

 were established by Huxley in 1861, to contain the living Polypteridae 

 of Africa and the extinct Holoptychidae, Rhizodontidae, and Osteolepidae, 

 having lobate paired fins with an endoskeletal axis, more or less 

 fringed with dermal rays. The Crossopterygii are the most typical 

 fishes of the Devonian. There is no difficulty in homologizing the 



43* 



