CiiAP. I. THE SKELETON. 267 



The vertebra} of the chest and abdomen are, as in Birds, soldered together 

 into one inflexible and more or less convex arch, though there are still thin 

 cartilaginous cushions between them. That connection is chiefly effected by the 

 spinal processes, which grow continuously, without an intervening suture, into the 

 ossified shield formed in the corium all over the back, thus forming a kind of 

 framework for that superimposed roof. 



The ribs are fixed to the places where the vertebra? meet, but the vertebras 

 do not send out peculiar processes for their support. They are strongest in those 

 Turtles in which the ossifications of the corium are least extensive, namely, in 

 the Trionychida;, Sphargididaj, and Chelonioidae ; weaker in the Chelydroida) and 

 Emydoida) ; weakest, and indeed often disappearing entii'ely, in the land Turtles. 



The sternum consists of nine bones, four in pairs, one odd,'^ all of which are 

 true bones. Their relation to each other in size and connection varies greatly in 

 different families. AVhile in the land Turtles and Emydoidfe they form one solid, 

 continuous, broad shield, covering the whole chest and the abdominal region from 

 below, they are much less developed in some of the Cinosternoidie, (Ozotheca, for 

 instance,) and least in the Trionychidae, Chelonioidce, and Sphargididae. In all the 

 three latter families, the bones of the sternum are very narrow, meeting each other 

 by slender processes, leaving much room between them, which is filled out by the 

 corium, thus forming a flat, elastic sternum.^ The sternum is jointed to the ribs 

 by means of a bony bridge, which may be compared to the cartilaginous or bon}" 

 bridge of other vertebrates, while the so-called marginal bones are mere ossifica- 

 tions of the skin. 



The vertebra? of the tail are very movable, convex behind, concave before. 

 No spinal processes either above or below. 



The locomotion in Turtles is entirely restricted to the four legs. Tlie bones 

 which are subservient to locomotion, appear entirely peculiar to this order of 

 Reptiles, as far as their form and connection with each other, as well as their 

 position with reference to the other parts of the skeleton, are concerned. 



The shoulder apparatus no longer rests upon the ribs as in the other Verte- 

 brata, but lies in advance of the ribs, and is more or less withdrawn under them. 

 The whole construction of these Reptiles shows the intention to cover all soft parts 

 by a hard .shield. This being the case, there is no room for a movable appa- 

 ratus upon the ribs. As the shoulder apparatus with the humerus, so also is the 



' This odd bone is wanting in the full-grown generally, though not alwa}<, in land ami marsh 



CinostemoidoB. Turtles. In Sphargididsc the sternum is reduced 



^ It is for this reason, perhaps, that we do not to a bony ring, consisting of slender pieces, and the 



find, in these three families, the sternum of the males disc inclosed bj- it is mere corium. The odd bone 



scooped out, (to facilitate copulation,) as we find it seems to be wanting. 



