CROCODILIA. 127 



notwithstanding the crocodilian characters of the small orbits, the long facial plates of 

 the prefrontal and lachrymal, the wide supratemporal apertures, the non-expansion of 

 the fore part of the palatines, and the non-appearance of the vomer on the palate, with 

 other minor marks of the like affinity. For all these characters arise out of secondary 

 modifications, and are presented in different degrees in the different species of 

 Crocodile, and are rather of a specific than a generic value. They determine the 

 judgment by the extent of their concurrence rather than by their individual intrinsic 

 worth, and for that reason, therefore, the exposed position of the lower canine in the 

 lateral groove of the upper jaw inclined the balance in favour of a reference of the 

 previously-described fossil to the true Crocodiles. One cannot, indeed, attach any real 

 generic importance to the modification of the upper jaw in relation to the lower canines. 

 In three examples, however, in the collection of the Marchioness of Hastings, the 

 crocodilian modification of this character is repeated, as it is shown in PI. 1 B, fig. 1 ; 

 and we have to choose, therefore, between the conclusion that Mr. Wood's specimen 

 (Pi. 1 C, fig. 2) presents an accidental variety in this respect, or to view the fossae in 

 the upper jaw as indicative of not only a different species but a distinct genus from the 

 Crocodilus Hastingsim. I should be glad to have more evidence on this point, and 

 especially the opportunity of comparing the posterior nostrils, the orbits, the supra- 

 temporal apertures, and the occipital part of the skull of a specimen from Hordle, 

 repeating the alligatorial character of the fossse in the upper jaw for the lower canines. 

 I am disposed to regard this character, notwithstanding its constancy in the living 

 species of Alhgator, as a mere variety in the Hordle fossil ; but pending the acquisition 

 of further evidence, it seems best to record this fossil under the title proposed for it by 

 the able geologist by whom it was discovered. 



Crocodilus Hastingsi^. Plate 1 1). 



VertebrcB referable to the Crocodilus Hastingsi^. 



The fossil crocodilian vertebrae obtained from the Eocene sand at Hordle, notwith- 

 standing the comparatively limited extent of the researches in that interesting 

 formation, are at least as abundant as those which have been discovered at Sheppy, 

 but they do not, as at that locality, indicate two distinct species ; all that have, hitherto, 

 been found belong to one and the same kind of Crocodile, and from their robust 

 proportions, would seem to have come from a species with a short and broad muzzle, 

 like that of the Crocodile or Alligator, the fossil skulls of which have been described. 



Perhaps the most perfect fossil reptilian vertebra that has hitherto been discovered 

 is the one figured, of the natural size, in PL 1 A figs. 1, 2, and 3. It is the fifth 

 cervical vertebra. As compared with that of the Crocodilus toliapicus (PI. 3 A, figs. 

 1, 2), which it resembles in size, the hyjiapophysis, hy (fig. 2, PI. I B), is much more 

 compressed, and the under part of the centrum is more extensively and deeply exca- 



