142 



CHIM^ROID FISHES AND THEIR DEVELOPMENT. 





_r 



\ 



spine, figs. 137 and 137 A, resem- 

 bling closely the dorsal fin-spine of 

 many sharks (no second dorsal 

 spine or even a dorsal fin is known 

 in Squaloraja, a condition which 

 suggests that this form may have 

 been bottom-living and that the 

 dorsal fin may have become shifted 

 into the region of the tail). The 

 antero-pelvic claspers are shown 

 by the presence of neighboring 

 radial cartilages to be reasonably 

 deduced from such elements, and 

 the short, wide, shagreen-coated 

 mixipterygia are 

 "~\ v also shark-like in 



"\ pattern. Their 



derivation from 

 radial cartilages 

 ; is also indicated. 

 ,.x' On the other 



hand, Squaloraja 

 gives no positive ground for 

 the belief that the fine rings 

 in its vertebral column are the 

 homologues of selachian 

 centra. For in this Liassic 

 form they are nearly as nu- 

 merous as in the living genera, 

 and the best evidence that they 

 are derived from metameral centra is that 

 the rings become slightly reduced both in 

 number and in diameter in the region just 

 behind the occiput.* 



138. Squaloraja polyspondyla. Details and partial restoration. 

 After specimen P 2276 in British Museum, figured by Smith Wood- 

 ward (1886). 



The narrowing of the snout is indicated in specimen No. 1 147 in Harvard Museum and in an undescribed specimen in the Museum of Science and Arts, Edinburgh. 

 Fin outlines hypothetical. Details of dermal tubercles are shown in A and ft. In f the ventral occipital region is figured after the above-noted specimen of 

 Harvard Museum. Here the condylar region is admirably preserved ; behind it centra appear at the right, neural arches at the left. And " ring " vertebrae 

 apparently grade into metameral centra, ar, anterior radials ; mif, miripterygium ; uvl, antero-ventral clasper ; oc, occipital condyle ; r, anterior "ring 

 vertebrae" ; cf, tract of enlarged dermal denticles. 



.ar 





mix 



Fig 



*Since the foregoing was written additional light has been thrown upon the question of metameral segmentation 

 in the column of Squaloraja; in the Harvard specimen already referred to, a coarse segmentation, which suggests 

 outwardly cyclospondylous vertebrae, is well shown in the postoccipital region, fig. 138 c. It is not certain, however, 

 that these coarse segments are serially homologous with the fine rings in other parts of the column ; it is possible, as 

 embryology indicates, that they belong to the outer chordal sheath. 



