410 H. I . OSIJORX RESTORATION OF KLKIMIANTS AND MASTODONS 



is almost entirely conjectural since the top of the skull has not yet been 

 discovered, and it is not known, therefore, whether the animal had the 

 characteristic peaked cranium of the true mammoth type, or the flattened 

 cranium of the African type, or the bulbous cranium of the Indian type. 



The hairy mammoth is by far the most probable restoration in the 

 extinct series because it is based, first, on the complete skeleton, second, 

 on the data furnished by the frozen Siberian mammoth, third, and most 

 important, by the extraordinary likeness which prevails in all the numer- 

 ous drawin«:s, engravings, and sculptures of E. prinngenin^ by the artists 

 of Upi)er Paleolithic times. 



In preparing these models we were at oine struck by the highly dis- 

 tinctive differences in the contour not only of the forehead but of the 

 backbone. The L. cydotls, for example, while of diminutive size and 

 with rounded ears, has the distinctive backbone profile of the African 

 elephant, which is hollow betv\-een the shoulders and the hips. The back- 

 bone of the Indian elephant is uniformly arched upward : that of the 

 mammoth rapidly falls away toward the hind quarters, and a similar 

 character is doubtfully attributed to the imperial mammoth. 



The extraordinan' dome over the head of the woolly mammoth, sepa- 

 rated by a deep valley from the dome over the back, is probably due to an 

 accumulation of hair and wool and possibly to the presence of a storage 

 reservoir of adipose tissue, because we know that this rounded form does 

 not coincide at all M'ith the peaked, flattened forehead of the skull within. 

 For purposes of casting, the hair, which nearly touches the ground be- 

 neath the neck and belly of the mammoth and constitutes a uniform 

 fringe around the lower part of the limbs, is reduced. 



