408 PROP. H. G. 3EELEY ON PSEPHOPHORTTS POLYGONrS. 



the armour of a ganoid fish. There is no trace of the plates having 

 been ankylosed together ; for although they for the most part still 

 retain their natural positions, they have separated sufficiently to 

 show that the original connexion was maintained by a fibrous or 

 coriaceous investiture. On the dorsal surface the plates, except in 

 the median keel, nearly all show a beautiful radiating sculptured 

 ornament, which certainly recalls that seen in some of the larger 

 Armadilloes, rather than the condition in any reptile. On the under 

 surface the plates are perfect^ smooth. As arranged, the transverse 

 measurement of the plates appears to be greater than the longitu- 

 dinal measurement, five plates occurring transversely in one place 

 in a distance of twelve centimetres, and six or seven occupying the 

 same distance longitudinally. 



It is difficult to say which should be regarded as the anterior 

 part of the fossil, as the remains are so fragmentary that they give 

 no indication whatever of its complete form or size ; biiit I incline 

 to believe, from the analogy of the curved ridge in SjpJiargis, that it 

 is an anterior fragment from the left side. The plates are remark- 

 able for their thickness, which sometimes amounts to nearly a centi- 

 metre, though most of them are thinner. Towards one corner of the 

 slab are a few plates, much thinner, partly covered with matrix, 

 which look as though they might have belonged to an under or ven- 

 tral armature ; but having regard to the state of preservation of the 

 fossil, it would be unsafe to overlook the probability that they may 

 bo a portion of the carapace displaced and inverted, or of its margin, 

 which would naturally have an inverted position. But Psepho^horus, 

 unlike /SpZiar^iV, may have also possessed a ventral shield of thin plates. 

 A question may arise as to whether the fragmentary bones on the other 

 side of the slab, which is about six centimetres thick, are portions 

 of the skeleton of the same animal ; but all probabilities seem to me 

 to lean in that direction ; for this fossil was evidently stranded, much 

 as porpoises and other animals are thrown up on sandy shores at 

 the preseut day, and the skeleton, becoming knocked to pieces, has 

 been scattered. The resemblance of the carapace to that of a mammal 

 is certainly sufficiently close to have justified any one in so regarding 

 it; and the large size of the armour-plates, as well as their sculp- 

 tured surfaces, would support this resemblance; but the elevated 

 rounded keel, increasing in height as it passes onwards, with the 

 somewhat flattened sides of the shield, are more in harmony with 

 iSjpJiargis ; and if skeletons of the covering of Sphargis had been pre- 

 pared in our Museums, it is certain that the same kind of irregular 

 arrangement of the plates would have been recorded. There is little, 

 except the relatively large size of the plates in the fossil and their 

 perfect ossification, to distinguish them from the comparatively small 

 elements which make up the bony skeleton of the covering of the 

 leathery turtle. 



So far as I am aware, the dermal shield of Sphargis, though often 

 figured, has never had its osteological characters described ; for the 

 thin epidermic covering conceals bones which constitute a shield 

 wliich is only comparable to a tessellated pavement, corrugated by 



