III.] RODENTS. 89 



Monte Hermoso. Of the former Plexochairus differs from Hydro- 

 chorus only by the somewhat simpler structure of the last molar, 

 Hydrochcerus itself occurring in the Monte Hermoso beds. Other 

 forms from the Parana stage are Eucardiodon and Cardiotherium, 

 apparently more nearly allied to Cavia. Here it should be 

 mentioned that certain European Oligocene genera (Issiodoromys 

 and Nesocerodori) have the molars so like those of the cavies that 

 they are regarded by Dr Schlosser as nearly allied' to the family. 

 By Prof. Zittel they are, however, included in the extinct family 

 Theridomyidce, which is classed with the dormice and certain 

 other families in a group regarded as intermediate between the 

 Sciuromorpha and Hystricomorpha. If, as would seem probable, 

 they are really allied to the latter, they are of the utmost import- 

 ance as indicating a connection between the middle Tertiary 

 rodent fauna of Europe with that of South America 1 . 



Nearly allied to the cavies are the agutis (Dasyprocta) and 

 pacas (C&logenys), collectively constituting the family Dasy- 

 proctidcE, and differing from the former in that the folds of enamel 

 merely form notches on the sides of the crowns of the cheek-teeth. 

 In a fossil state the family seems to be only known by remains of 

 the existing genera from the Brazilian caves. The third peculiar 

 family (Dinomyidce) is known merely by a single specimen from 

 Peru. The only other family exclusively confined to the realm is 

 that of the Lagostomatidcz (Chinchillidce], which includes not only 

 the true chinchillas (Eriomys) and Cuvier's chinchilla (Lagidium), 

 but likewise the viscacha (Lagostomus] of the Argentine pampas. 

 All the members of this family have long bushy tails, elongated 

 hind limbs, and the cheek-teeth divided by complete transverse 

 folds of enamel into thin plates. Lagostomus occurs fossil not 

 only in the Pampean, but likewise in the older Tertiaries of 

 Parana; while it may be doubted whether Pliolagostomus and 

 Prolagostomus of the Santa Cruz beds are really entitled to generic 

 distinction. Other allied forms from the latter deposits are 



although their lowest portion is older than the Monte Hermoso stage. (See 

 Ameghino, Bull. Ac. Cordoba, Vol. xiil. pp. 260, 261, 1894.) They may be 

 partly made up of the remains of pre-existing beds. 



1 Dr Schlosser writes me to the effect that he is fully assured these forms 

 are the ancestors of the Caviidce. 



