272 TYPES OF ANIMAL LIFE 



have grouped in the same order with it — the order 

 Edentata. 



Of its five groups (i) sloths ; (2) ant-eaters ; (3) arma- 

 dillos; (4) pangolins and (5) aard-varks, the fifth and 

 last stands wide apart from the other four, and shows no 

 sign of real relationship with any one of them. The 

 aard-vark is placed with the other edentates rather 

 because naturalists do not know in what other order to 

 put it, and shrink from erecting it into an order by itself. 



The pangolins also show little affinity to the first 

 three groups. We have seen how even an animal, of so 

 fundamentally distinct a nature as the echidna, may 

 nevertheless present us with ant-eater-like characters, 

 which cannot have been due to inheritance, but must have 

 arisen independently. The ant-eater-like character of 

 the pangolins, then, may also have arisen independently. 

 Nevertheless, the very singular fact, that the bones of the 

 neck have a tendency to diverge from the number almost 

 universal in the class of mammals, seems to indicate a 

 possibly deep-seated affinity betw^een them and the 

 sloths which present that abnormal character in a yet 

 more marked degree. Nevertheless, we can only speak 

 of it as a " possibly deep-seated affinity," because, as 

 before said, we find similar abnormality in the manatee, 

 which can hardly be supposed to have any exceptional 

 affinity with either the sloth or the pangolins. 



The armadillos do show some evident marks of 

 affinity to the ant-eaters, especially in the structure of 

 the lumbar legion of their backbone. Nevertheless, in 

 their external defensive armature, they differ widely from 

 them, and from every other kind of beast whatsoever. 

 Moreover, the most ancient form of the group yet known 

 to us — the glyptodons — were even more exceptional 

 in structure than those which exist to-day. 



