LIFE IN TERTIARY TIMES 307 



descended from Megatherium, to whose diet they have remained 

 faithful, there is no reason why their dentition should have 

 been modified. Ant-eaters have the same way of walking, 

 and their pectoral mammae indicate that they are descended 

 from tree-climbing animals, and the structure of the repro- 

 ductive apparatus makes it clear that these tree-climbers 

 were Sloths. But they have changed their diet : they live on 

 Insects, and their extraordinarily long vermiform tongues are 

 all they need to take hold of their food, which does not have 

 to be masticated. Lack of usage can explain the total dis- 

 appearance of the teeth, the form of the tongue, and the 

 elongation of the head and jaws. 



In the skin of Mylodon, related to Megatherium, and 

 which has not long disappeared from South America, there 

 were numerous ossicles. These ossicles formed a complete 

 carapace in Glyptodon, whose back was hemispherical and 

 nearly two metres in diameter. The head still resembled 

 that of the Megatherium, but the feet rested flat on the ground. 

 It is quite likely that the Armadillos of to-day, which can be 

 traced back to the Tertiary (Eutatus, with a carapace formed 

 entirely of mobile strips, and Dasypus), are related to them 

 in some degree ; but in their case the jaws are elongated, and 

 this coincides with a multiplication of teeth, which in the 

 great Armadillos are twenty-six in the upper half-jaw and 

 twenty-four in the lower, a total of one hundred in all. 



Orcteropius and the Pangolins of the Old World seem to 

 form a special group in which we observe the same abortion of 

 the teeth. They date back to the Miocene. 



And here we encounter a new difficulty. Among the 

 fossils of the Eocene beds of Montmartre, Cuvier found a bifid 

 phalange with a nail which he attributed to a very large 

 Pangolin, the only animal except the Mole that possessed 

 a similar character. This hypothetical Edentate received from 

 Lartet the name of Macrotherium, and some time after a head 

 attributed to a kind of Horse was named Chalicotherium. Sub- 

 sequently, however, an unexpected discovery in the beds of 

 Sansans proved conclusively to Filhol that Macrotherium and 

 Chalicotherium were one and the same animal. Related forms 

 were dug up in the Eocene deposits of North America, and 

 completely reconstructed by Professor Holland (Moropus, 

 Eomoropus, and Pematherium). They had the general 



