72 MR. E. R» ALSTON ON THE ORDER GLIRES. [Jail. 18, 



nostrils pointed above, sigmoid or linear. The ears are very generally 

 emarginate behind ; and the tail, when present, is cylindrical, hairy, 

 scaly, or subnaked. 



In the division of the hystricine Rodents into families much diver- 

 sity of opinion has prevailed. Mr. Waterhouse, laying too great 

 stress on dental characters, entirely separated the Dasyproctina from 

 the Caviina, and placed them between the Hystricina proper and the 

 Ecfiimyina*. Prof. Brandt reunited the two former in his family 

 Hemionychoides, equivalent to the Subungulata of Illigert. But the 

 group thus formed is so ill defined that Prof. Lilljeborg found it im- 

 possible to separate it from the Hystricidcsf. It seems to me that, 

 although Mr. Waterhouse was certainly misled in entirely separating 

 the Agoutis and Pacas from the Cavies and Capybara, they must still 

 be ranked as distinct but allied families, and that the same value 

 must be given to the curious form named Dinomys by Dr. Peters§. 

 Accordingly I would recognize six families of the Hystricomorpha. 



Of these the first, the Octodontidce, consists of three subfamilies ; 

 for here, I think, must be placed the Ctenodactylince, formerly asso- 

 ciated with the Jerboas, but of which the hystricine affinities have 

 been established by Dr. Peters ||. The other subfamilies are the Octo- 

 dontincB and Echinomyince of Mr. Waterhouse. Some of the genera 

 of the latter make a close approach to the next family, the Hystri- 

 cidcc, which in its turn is composed of two very distinct subfamilies, 

 Sphingurince and Hystricina; ; for I cannot follow Professor Lillje- 

 borg in relegating the former to the Octodontidce*^, principally on 

 account of their better-developed clavicles, which are probably an 

 adaptive peculiarity connected with their arboreal habits. Of the 

 remaining families the Chinchillidce form a small but very natural 

 group, connected in some characters with the Dinomyidce ; and the 

 latter, again, has close affinities with the nearly allied Dasyproctidce 

 and Caviidce. These latter families in many points, as in the mode 

 in which their incisors wear down, their emarginated palates, and 

 the large size of their basicranial foramina, show a striking ap- 

 proach to the next great group of Rodents. 



The second suborder, Glires duplicidentati, containing only 

 two families, is clearly less specialized than the first, and appears to be 

 a survivor, representing a comparatively early stage in the develop- 

 ment of the Rodent type. At birth, the upper jaw contains the 

 normal number of incisors; but only the two inner pairs are retained ; 

 and of these the second remain very small, and are placed directly be- 

 hind the large middle pair. In the mandible there is never more than 

 one pair. Another important proof of the inferior degree of speciali- 

 zation in the Glires duplicidentati is the fact that the enamel of the 

 incisors may be traced round to their posterior surface, though it is 

 here much thinner than in front**. Of cranial characters, it may be 



* Nat. Hist. Mamm. ii. p. 3G0. t Prod. Syst. Mamm. p. 92. 



\ Op. cit. p. 54. 



§ Festschrift. Gesellsch. nat. Freunde, pp. 227-234 (1873). 



Tr. Z.S. vii. pp. 397-409. 1 Op. cit. p. 51. 



*' f Cf. Owen, Comp. Anat. Vert. iii. p. 296. 



