1896.] ZOOLOGICAL BXPBDIXION TO MADASABOAB. 979 



uromys, Hypogeomys — Brachytarsoviys stands somewhat apart from 

 the others and requires further investigation — belong to the so- 

 called Cricetine group of Muriform (" J\IuridsB," auct.) Eodents, 

 of which they are the lowest of existing forms, having affinities 

 with some of the least specialized of the family DipodidsB, as deJSned 

 by Winge, viz. to SmintJius and Zapus. 



The African and Asiatic Rhisomyes, usually considered as 

 belonging to the Spalacidas, but which the last-named author 

 places amongst the lowest Muridas, alongside with the tertiary 

 Cricetodon and Eomys, are nearly related to the Malagasy group of 

 llodents by means of the Abyssinian Tachyoryctcs (JRliizomys) and 

 the Malagasy Brachyuromys, the former being but a very specialized 

 fossorial form of the more generalized Brachyuromys ramiroMtra. 

 The molars are almost identical in both, only but slightly more 

 hypselodont in Tacliyoryctes. If we divest the Taohyoryctes skull 

 of its fossorial characters and of the consequences of the more 

 hypselodont molars, we obtain a Brachyuroinys skull. Likewise 

 the skulls of the young Tacliyorydes bear much greater resemblance 

 to Brnchyuromy.i tlian the adult. There is further a great corre- 

 spondence in external characters if we disregard the smaller ears 

 and eyes of Tacliyoryctes and its fossorial claws. 



As to the aflinities of the Malagasy Eodents with the lower 

 Bipodidfe, they are revealed by the skull as well as by the confor- 

 mation of the molars. The infraorbital foramen is large through- 

 out and especially in Brachyuromys, though on the whole showing 

 the form characteristic for the Muridro', the posterior part of 

 the zygomatic arch is bent dowuv\'ards, the malar bone strongly 

 developed and approaching the lachrymal more than in any other 

 Murida), the size and shape of the incisive foramina nearly approach- 

 ing what obtains in the Dipodidfe, &c. With regard to the teeth, 

 the group of Malagasy Eodents, together with the Abyssinian 

 Tachyoryctcs, differ in a very important condition from the more 

 specialized MuriusB, and even from the Cricetine Eodents, in 

 having their molars of almost equal size and form ; the two 

 anterior molars especially are very much like each other. This 

 likewise is a character in which they approach the lower Eodents, 

 especially the Dipodidfe ; in the pattern of the molars there is 

 equally a strong resemblance of them all with Dipodidfe (Sminthus, 

 Alactaga, Zapus) ; in this respect the mosaic pavement-hke tri- 

 turating surface, both in the Malagasy Gymnuromys and the 

 Nearctic Zapus, is especially noteworthy. 



The relation of the Madagascar Eodents to Crketus, which is 

 considered to be the type of the group, is viewed by me as 



^ Tho raioceno Pacicidus, from tlio .Tolin-Bivy bods in N. America, is con- 

 eidered by Scott to stand in most respects in an intermediate position bolwcon 

 Protopiychus (whicli Scott supposes to bo tlie ancestral form of tlio Dipodida;) 

 and tbe Dipodida?, altliougb it lias lost all the premolars, and the lower portion 

 of the infraorbital foramen forms, as in the Muridae, a distinct notch for the 

 passage of the nerve. (" Proiopiychus hatchari, a new Rodent from the Uinta 

 Eocene," Proo, Ac. Nat. Sc, Philadelphia, 1895, p. 269.) 



