CHELONIA. 43 



most dome-shaped tortoise. The flatter surface of the nodule was shghtly convex, 

 which I thought might arise from a layer of petrified clay adhering to the plastron. 

 A portion of the cranium was indicated at the produced angle of the nodule. To 

 ascertain whether this remarkable degree of convexity of the carapace, both length'wise 

 and transversely, was natural, I had the matrix carefully removed, with the permission 

 of the owner of the specimen, and the same was done on the opposite side, with a view 

 to expose the plastron. Instead of finding a plane plastron where it was expected, in 

 its natural horizontal position, it was found to have been crushed inwards, as repre- 

 sented in fig. 2, by the pressure of a hard petrified mass as big as a paving-stone, 

 which had been forced in upon this part of the body of the turtle whilst in a decom- 

 posing state ; and when finally lodged in the clay, the carapace and plastron, as they 

 became dislocated, had become more or less moulded upon it ; and thus was produced 

 the convexity which originally attracted my attention. In the breadth of the connate 

 rib, as compared with that of the costal plate, in the extent of the free extremity of 

 the rib, in the degree of concavity of the upper surface of the costal plate and 

 the curvature lengthwise, the distinctive characters of the Chel. crassicostata are well 

 show'n. The same characters are likewise presented by the parts of the plastron, as 

 in the breadth of the xiphisternals (<r.s), the curvature of the hyosternal (//*•), and the 

 form of the coracoid. The two scapulae, with the connate acromial clavicles, are 

 preserved, with the head of one of the humeri. A part of the basis cranii, showing the 

 broad diverging pterygoid, with their characteristically-channeled inferior surface, is 

 shown in fig. 2 ; these grooves are not so deep, however, as in Chel. lonc/iceps, but are 

 moFe like those in Chel. cuiieiceps. 



I beg to record my obligations to Mr. Bowerbank for the suggestion, and to 

 Mr. Bull for his ready response to it, to which I owe the opportunity of examining 

 this specimen of the thick-ribbed turtle of the Harwich Clio's. 



A fossil mandible of a Chelonian, in the collection of the Marchioness of Hastings 

 (figured, of the natural size, in PI. 33, figs. 1 and 2), most resembles that part in 

 the genus Chelone by its general form and proportions, and especially by the configura- 

 tion of the biting and grinding surface of the jaw- (fig. '2). The symphysis is confluent; 

 convex in both directions below ; longer than in the Chel. mydas and the Chel. breviceps 

 of Sheppy (PI. \7A, fig. 3,32) ; but not so long as in the turtles from Harwich (PI. 18, 

 fig. 1, and PI. 20, fig. 3) and Bracklesham, or as in the Chelone lonc/iceps of Sheppy. 

 The rami diverge more from each other than in the lower jaw of the Chel. convexa 

 (PI. 14, fig. 3) of Sheppy. 



A ridge, commencing at the fore part of the upper surface of the symphysis, passes 

 backwards, and divides the two ridges, diverging and circumscribing with the outer 

 sharp margins of the jaw an elliptical concave space on each side ; the space between 

 the diverging ridges is raised and rough : this part has been fractured. In the Trionyoc, 

 of which genus so many fine examples have been met with at Hordwell, the upper part 



