'410 PROF. H. G. SEELEY ON PSEPHOPHORUS POLYGONUS. 



plan upon which the representative of the carapace itself is con- 

 structed ; and I believe that we have here an indication of a primary 

 division of the Chelonia to which palseontological discoveries may 

 hereafter give more importance than can at present be claimed for it. 

 From the point of view of evolution, it may fairly be anticipated 

 that the pavement-shielded type of Chelonian preceded that in which 

 the dorsal shield is formed of symmetrical bones. Perhaps the most 

 remarkable character in the dermal skeleton of Spliargis is the fact 

 that the bones of the plastron, although only forming an outer ring 

 to the ventral surface, are already defined and homologous with those 

 of the ordinary Chelonian, in which the carapace is specialized. And 

 this leads me to draw attention to the fact that the dermal bones of 

 Sijliargis, which may be detected without difficulty on the dorsal 

 surface of any ordinary shield, are altogether invisible on the interior 

 side of the shield, which was in contact with the neural arches of 

 the dorsal vertebrae and the slender flattened ribs ; so that it is quite 

 evident that the ordinary carapace of a Chelonian is in no way re- 

 presented by the dermal skeleton of PsejjJiopJiorus or SpJiargis. Im- 

 pressed by the fact that the dorsal ribs of Crocodiles bear upon their 

 hinder margins plates of fibro-cartilage, which in the Hatteria also 

 exist and are sometimes converted into bone, and believing that 

 these plates are homologous with the epipleural elements or uncinate 

 processes which become greatly developed and blended with the ribs 

 of certain birds, I have been led to speculate on the probable occur- 

 rence of such cartilaginous elements upon the dorsal surfaces of ribs 

 of Chelonians, and to believe that the development and ossification of 

 such plates upon the Chelonian rib would lead to a transference of 

 the osseous matter from the superficial portion of the skin to its 

 deeper-seated layer, and hence to the growth of epipleural plates, 

 which would become the costal plates of the carapace. The granu- 

 lations on the shield of Trionyx, on such a view, indicate dermal 

 bones like those oi Spliargis — which were originally separate, but have 

 become blended with the bony elements which were subsequently 

 developed beneath them. The costal plates are said to be always 

 distinct from the ribs in the young turtle when first hatched. They 

 are certainly very small in the young of some of the Trionychidae, in 

 which, indeed, they remain throughout life but partly united with 

 the endoskeleton. It is the impossibility of logically accounting for 

 the development of the Chelonian carapace without the aid of this 

 hypothesis that leads me to attach more than ordinary importance, 

 in a classificational point of view, to the characters presented by 

 Psepliophorus, which are only otherwise paralleled by the small spe- 

 cimen which Miiller figures from the Zeugiodont Limestone of North 

 America, and by the living Sjylmrgis. 



The best-preserved vertebral fragment (PL XY. fig. 2), as already 

 mentioned, is Chelonian in its characters. A few scattered plates 

 occur intimately associated with the vertebraB. Altogether there are 

 fragments of five vertebrae preserved. None shows the whole of either 

 centrum or neural arch ; nor do they collectively throw much light 

 upon each other. It is difficult to affirm with certainty the region 



