THE HYRACOIDEA. 367 



to which the names oiLoxodon [JE. Afrlcanus) and Euelephas 

 {E. IncUcus) proposed by the late Dr. Falconer may be very 

 properly aj^plied. The oldest rocks in which their remains 

 occur are of Miocene age. Fossil remains of elephants occur 

 not only in the Old World, but also in both North and South 

 America. 



III. The Hyracoidea. — The genus Hyrax, which is the 

 sole member of this group, was referred by Pallas to the Ro- 

 dents ; and by Cuvier, who demonstrated that it could not be a 

 Rodent, it was placed among the Ungulata, in the immediate 

 neighborhood oi H/mioceros, without any better evidence than 

 that afforded by the characters of the molar teeth. Prof. 

 Brandt, of St. Petersburg, in an elaborate memoir just pub- 

 lished, arrives at the conclusion that it is a " gliriform Ungu- 

 late," intermediate, in a certain sense, between the Hodents 

 and the Ungidataj but, still, more Ungulate than Rodent. It 

 appears to me to be neither Ungulate nor Rodent, but the 

 type of a distinct order, in many respects intermediate be- 

 tween the JJiigidata, on the one hand, and Rodentla and In- 

 sect Ivor a^ on the other. 



The small. Rabbit-like, animals comprised in the genus 

 Hyrax are plantigrade, and provided w4th four visible toes in 

 front and three behind. The nails are not hoof-like, but 

 nearly flat, except the innermost of the hind-foot, which is 

 peculiarly curved. The body is covered with fur, and the 

 muffle, or snout, is split, as in the Rodents. There is a pen- 

 dulous penis, but no scrotum ; and there are four inguinal and 

 two axillary teats. 



There are from twenty-nine to thirty-one dorso-lumbar 

 vertebrae, which is the greatest number known in any terres- 

 trial mammal. Twenty-one or twenty-two of these are dorsal. 

 No mammal, except Choloepus, the two-toed Sloth, possesses 

 so large a number of dorsal vertebrae as this. The transverse 

 processes of the last lumbar vertebra articulate with the 

 sacrum, as is the case in many Ungulate Mammals. In the 

 skull, the post-orbital processes, which are chiefly furnished by 

 the parietal and the jugal, nearly meet. Part of the articular 

 facet for the mandible is formed by the jugal, which extends 

 forward until it comes into contact with the lachrymal bone. 

 The base of the external pterygoid process is perforated by a 

 canal, as in Perissodactyla and JLemuridce. There are large 

 pre- and post-tympanic processes, and the post-tympanic is 

 much shorter than the par-occipital process. The premaxillce 

 are large, and unite extensively with the nasal bones ; the 



