1897.] MALAGASY GENTS BRACHYTJKOili'S. 717 



known of Muridse, as Cricetodon and Cricetus, would point to the 

 opposite assmnption " '. 



These two geaera are considered hy Winge to be amongst the 

 lowest of Miu"idse, on account of the greater agreement in size of 

 their m.l and m.2 ; m.l being more elongate in the Murince^. 

 To these views of Winge I have to object : — ■(!) The molars of some 

 of the species of Cricetodon show the transverse ridge in question. 

 Winge refers ^ to the figures of Cricetodon given by Gervais *, who 

 figured worn teeth. With regard to the Tertiary Muridae com- 

 prised under this generic denomination, the case stands thus : — 

 In the older forms, e. g. Cricetodon cadurcense, Schloss. % from the 

 Quercy, not only the middle transverse ridge, ending in 2, but 

 also 1 and 3 are present, as they are likewise present in Eutnys •* 

 (which does not seem to differ much from Cricetodon, both being 

 very similar to the less simple molars of raanj Hes^ieromyince) . 

 Besides, we find in Cricetodon cadurcense a primitive feature of 

 m. 1 ■which is almost identical with m. 2. In the more recent 

 species of Cricetodon, from the Middle Miocene of Steinhetm and 

 La Grive Saint-Alban, I observe a greater approach to Cricetus ; 

 m. 1 is larger and more complicated as compared with m. 2, and 

 the accessory crests of all the molars tend to disappear. 



(2) In the Malagasy Muridse the difference in size between m.l 

 and m.2 is even less, and their agreement in fonu greater than in 

 Cricetus, Cricetodon, and the Hesperomyina, which is one of my 

 reasons for regarding them as more primitive Muridae than the 

 above-named. And still \^e have found 1, 2, and 3 to be present in 

 molars of young Brachyuromys betsileoensis. In other Malagasy 

 Muridae, to be described on a future occasion {Nesomys, Gymnuromys, 

 Brachytarsomys), they are stiU more evident, and not in the least 

 limited to young stages '. 



(3) In the Tertiary such simple forms of molars as exhibited by 

 Mystromys, Cricetus, '■^ Hesper. expulsus," and " Hahrothrix lasiurus" 

 are not known. The latter two, apart from the pattern of their 

 molars, agi'ee in all the rest with the other Hesperomyince (having 

 more complicated molars) than with Cricetus. The latter genus is 

 apparently derived from some such form as Cricetodon by a greater 

 reduction of its molars, just as the HesperomyincB, with simpler 

 constructed molars, appear to be derived from those with more 

 complicated teeth. With reference to its teeth I have therefore 

 recently called Cricetus a terminal form amongst Muridse ^. 



Winge has, in this conjunction, urged the similarity of the m.2 

 of Cricetus with other Mammalia ". I know of no greater agree- 



' ' Gnavere fra Lagoa Santa,' p. 11 ; and footnote 6, ])• 151. 



2 " Om grseske Pattedyr," Meddel. Natm-h. Foren. i. Kjobenhavn, 1881 ; 

 ' Gnavere fra Lagoa Santa,' pp. 123, 124. 



3 " Om gr£eske Pattedyr," I. c. p. 22. •» Zool. Pal. frang. pi. 46. fig. 3. 

 = L. c. pi. xii. (viii.) figs. 28, 35. 6 Leidy. 



'' In the same way the Dipodine Zapus, in which m.l and m.2 are of 

 absolutely the same conformation and size, shows likewise 1, 2, and 3 well 

 developed. 



•^ P. Z. S. London, 1896, p. 980. 



9 Meddel. Naturh. Foren. 1881, p. 26 ; ' Gnavere fra Lagoa Santa,' p. 11. 



Proo. Zool. Soc— 1897, No. XL VII. 47 



