MURID^—ARVICOLISOE— EVOTOMYS. 133 



mys, the molars have only two parallel roots apiece, one directly behind the 

 other, and both broad ; and the rooting is simply the closing-up of the ends 

 of the roots from failure of the pulps that in other Arvicolince are supplied 

 indefinitely, causing the roots to persist open. On the other hand, in Murince, 

 the roots of the molars are distinct diverging prongs, closed from the first; 

 there are at least three such prongs (two external and one internal) in Ameri- 

 can Hesperomys or Sigmodont Murines, and even more in the Old World Mus, 

 each perforating the alveolus separately. In Evotomys, there are but two 

 such perforations of the alveolus, and these even are almost confluent. 



From Arvicola, the next most signal difference of Evotomys is seen in 

 the construction of the bony palate. In Arvicola (e. g. amphibius or riparius), 

 the palate behind has a little pit, or fossa, on each side opposite the last molars, 

 and the whole space between them is depressed ; and this depression is fissured, 

 or excavated, by the advance from behind of the inter-pterygoid cavity, which 

 either ends at the palate with a single curve, or with a double curve from the 

 development of a little azygos process on the middle line of the posterior 

 margin of the palate. "Thus," to use Baird's words, "there is a step from 

 the plane of the bony palate to the bottom of the fossa, and another thence 

 to the base of the skull or body of the sphenoid"; and the sides of the palate 

 behind run out continuously into the pterygoids. Now, in Evotomys, all this 

 depressed or fossate part of the palate is done away with ; the palate ends 

 by an abrupt transverse edge, as a straight shelf, opposite the middle molar 

 (or rather opposite the space between the middle and last molar), leaving the 

 excavation of the base of the skull apparent in the whole space between the 

 last molars, and breaking all connection with the pterygoids. This construc- 

 tion of palate, so unusual in Arvicolince, is, however, again found, with no 

 appreciable difference, in the Lemmings; but, singularly enough, the genus 

 Synaptomys, which repeats Evotomys in external form and Myodes in denti- 

 tion, has the palate constructed as in Arvicola. The curious interrelation of 

 Myodes, Synaptomys, and Evotomys is sufficiently interesting without consid- 

 ering the murine affinities of the latter ; but, while we are on the subject of 

 the palate, we may here allude to some of its conditions in murine forms. In 

 Mus decumanus, the palate behind has no step downward or depressed fossate 

 part, as Arvicola has, ending as a straight, sharp, transverse shelf, as in 

 Evotomys; but then it reaches as a continuous plane far behind the last molar, 

 and runs directly into the pterygoids on either hand, the median excavation 



