ITS NEAREST EXISTING RELATIONS. 55 



When fully developed, the horns are of a more or 

 less conical form usually curved backwards at their 

 ends, and often grow to a great length (three or even 

 four feet), but they are constantly worn away at the 

 ends and sides by being rubbed against trees or 

 stones, and are continually growing at the base. 

 Their length and shape are, therefore, subject to 

 considerable variation, even in the same individual 

 at different times, and so cannot be depended upon 

 for the distinction of species, as some naturalists 

 have imagined. Though not normally shed, they 

 are occasionally torn off at the base,* in which case 

 a new horn will grow in its place, although, if the 

 matrix, or portion of the skin to which it is attached, 

 is much injured, it may assume a more or less irreg- 

 ular shape. 



As regards the dentition, the incisors are varia- 

 ble, generally reduced in number, and often quite 

 rudimentary and entirely disappearing at a very 

 early age.f The canines in existing species are 



* This happened in 1870 to the male Indian rhinoceros 

 still living in the gardens of the Zoological Society of Lon- 

 don, in an effort to raise with its horn a strong transverse iron 

 bar at the lower part of the railings of the inclosure in which 

 it was confined. 



t It is difficult to see what advantage the great African 

 two-horned rhinoceroses can find in the complete absence of 

 their front teeth, but this is one of those numerous cases in 

 which we must be content to acknowledge our ignorance and 

 wait for the explanation. 

 5 



