66 FOSSIL REPTILIA OF THE LONDON CLAY. 



On a general review of these numerous and rich accessions to our previously 

 scanty evidence of extinct Serpents, I may sum up by stating that the generic character 

 of Palaop/iis is chiefly manifested in the length of the neural spine, in the pointed aliform 

 productions of the back part of the neurapophyses, in the uniform convexity of the 

 diapophysial tubercles, and the minor transverse production of the zygapophyses. 



The Palaop/iis toliapicus is distinguished by its longer vertebrae in proportion to 

 their breadth, by its sessile diapophyses, and by the carinate character of the lower 

 part of the centrum in the vertebrae of the abdomen. 



The Palaophis Typluzus is distinguished by its shorter and broader vertebrae, by its 

 pedunculate diapophysis, and by the anterior and posterior hypapophyses of the 

 vertebrae of the abdomen ; its neural arch is narrower, and its sides not longitudinally 

 ridged. 



The Palaophis porcatm is characterised by the longitudinal ridges connecting the 

 anterior with the posterior zygapophyses, by its broader and squarer neural arch ; but 

 it has the two haemal spines below like the other large species from Bracklesham. 



Pal^eophis (?) longus, Owen. Tab. XIV, figs. 35, 36, 37, 45, 46. 



Vertebrae of a serpent agreeing in character with those of the London clay at Sheppy, 

 but smaller, have been obtained by Mr. Colchester, from the sand of the Eocene 

 formation underlying the Red Crag at Kyson or Kingston in Suffolk. In these, as in 

 most of the trunk vertebrae of Palaophis Typhaus, the hypapophysis is a small sub- 

 compressed tubercle at the under and back part of the body of the vertebra ; but there 

 is no repetition of a smaller process at the fore part ; and no ridge is continued 

 'backward from the hypapophysis, as in the Palcsop/iis toliapicus. The tubercle for the 

 rib is single ; in Naja it is almost divided into two, the upper being convex, the lower 

 moiety concave ; in the Python the upper half of the tubercle is convex, and the lower 

 half concave, but the two facets are not marked off. In the fossil serpent from 

 Kingston, as in the PalceopJiis from Sheppy and Bracklesham, the costal tubercle is 

 simply convex. The chief characteristic of the Ophidian vertebrae from Kingston is 

 the length and slenderness of their bodies, in which respect they exceed those of the 

 Palaophis toliapicus, and resemble some of the existing tree-snakes (Dendropfns) with 

 elongated vertebrae. The origin of the neural spine is limited to the posterior half of 

 the arch (fig. 36) ; but the mutilation of the neural arch in the specimens I have yet 

 had the opportunity of examining, prevents a prosecution of the comparison with any 

 adequate advantage. 



