86 THE NEOG^EIC REALM. [CHAP. 



associated with nesodons, astrapotheres, and homalodontotheres, 

 which have been assigned by Dr Ameghino to genera distinct 

 from those of the Santa Cruz beds. But from my own observations 

 on a series of remains of these animals in the La Plata Museum 

 obtained in association with those of Pyrotherium, they appear to 

 be genetically inseparable from their Santa Crucian represen- 

 tatives. 



Now represented only by the living Indian and African 

 elephants the Proboscidea were formerly a some- 

 deans b SC1 what extensive subordinal group of the ungulates, 



easily characterised by their peculiar molar teeth 

 and tusks (the latter of which may be present either in the upper 

 or lower jaw alone, or in both) ; their five-toed feet, in which the 

 bones of the ankle and wrist are arranged on the linear plan, and 

 the astragalus has a flat upper surface; and the presence of a 

 trunk. While true elephants (Elephas) are totally unknown in 

 South America, the genus Mastodon, distinguished by the low 

 crowns and simpler structure of the molar teeth, is represented by 

 two species in the Pampean of Buenos Aires, and is also stated to 

 occur in the Monte Hermoso beds ; but (assuming the distinct- 

 ness of Pyrotheriuni) the sub-order is unknown in the older deposits. 

 Although at the present day Neogaea contains no existing fami- 

 lies of either carnivores or ungulates absolutely pecu- 



Rodents. .. ....... 3 * 



liar to it, the case is widely different with regard to 

 the rodents, or gnawing mammals. Existing rodents are divided 

 into seventeen families, arranged under four sectional groups ; 

 nine out of these seventeen families occurring in the Neogaeic 

 realm, among which four are absolutely peculiar to it. A fifth 

 (Octodontida) is mainly South American and West Indian although 

 possessing a few representatives in Africa south of the Sahara, and 

 a sixth (Hystricida) has two genera which are practically only 

 South American. The significance of these facts will be apparent 

 when it is stated that of the other zoological regions of the globe 

 only two have any families of the order peculiar to them, and 

 neither of these has more than two such families. In the Ethio- 

 pian region, for instance, the only peculiar family is that of the 

 African flying-squirrels (Anomalurid(E\ represented by two genera ; 

 while the western division of the Holarctic region has the sewel- 



