158 MK. A. C. HADDON ON THE EXTINCT 



the individual variation is not at all confined to absolute size ; it extends to the relative 

 proportion of divers parts of the bones, to processes or depressions upon thern, such 

 as are commonly held to be specifically characteristic, so that it is often utterly impos- 

 sible to predicate any definite limits of individual modification Nor is the 



variability of which wc speak wholly dependent on age or sex." 



To find that the same results attend the examination of these Tortoise-remains cannot 

 fail to suggest the possibility that when a sufficient series of specimens of any animal is 

 compared a similar amount of variability may be observed ; and if this consideration 

 should lead to nothing else, it should at least make us, in many cases, hesitate about 

 describing a new species of vertebrate from a single bone, or even from several 

 fragments. 



A second variety of coracoid has been added, which, since it is free, may be safely 

 referred to T. inepta; of this we have three examples, one, however, being frag- 

 mentary. They differ from the former type in both form and size (PI. XIII. fig. 6). 

 The absolute length is much greater than in T. triserrata ; the neck is more slender, 

 and the posterior edge straighter, as is also the internal margin. Even in these few 

 examples we see ample evidence of tendency to variation. The great difference of the 

 length of this bone in the two species is well seen by averaging the relative pro- 

 portions ; for taking the length of the coracoid to be n, the average breadth would be 

 about n — 30, and the least circumference n— 47. 



Humerus. — Six right and eight left belong to Dr. Gi'mther's group a (T. triserrata). 



Two right and four left of T. mepta (group c) ; one of the latter is of enormous size. 

 In one or two examples the hollow behind the head is extremely shallow, as in 

 Dr. Gi'mther's group b, thus showing that the latter is merely a variation ; indeed it is 

 often rather difficult to distinguish between humeri supposed to belong to the two kinds 

 of tortoises. 



2. triserrata. T. inepta. 

 millim. millim. 



Length of humerus from summit of head to middle of trochlea .... 175 250 



Narrowest circumference 70 112 



Longest diameter of head 43 50 



Shortest diameter of head 34 35 



Extreme breadth between condyles 71 83 



Ulna. — Three right examples apparently belonging to the same species. 



millim. 



Extreme length 114 



Narrowest circumference 44 



Radius. — Three specimens, only one of which is perfect. Dr. Gunther's types a 

 and c may be easily recognized. The one right, a, has a rather slender shaft. 



One well-marked right example of c; the other, left, is only a distal fragment; it 

 would, however, be somewhat larger. 



c. 

 millim. 



Length of radius 175 



Circumference in narrowest part 38 



Longitudinal diameter of semilunar facet 30 



Transverse diameter of semilunar facet 20 



