WATER VOLE. 317 



Arvicola musignam, De Selys, Rev. Zool. 1839. 



,, viouticola, De Selys, Rev. Zool. 1838. 



Rat iVcau, Buffon, Hist. Nat. VII. p. 368, t. xliii. 



Water Rat, Penn. Brit. Zool. I. p. 118. Sii.vw, Gen. Zool. II. 



p. 73, t. cxxix. 

 Var. /S, deep black above and beneath. 

 Ai-vicolaater, MacGillivray, Trans. Wern. Soc. VI. p. 424. 



In the former edition of this work we conformed to 

 the prevailing opinion that the Arvicoke could not witli 

 propriety be placed with the Mice, but that tliey con- 

 stituted an aberrant group of the Beavers, and acting 

 under this impression, the patronymic name of Ai'vicoUdoi 

 was made to give way to that of Castoiidce. But the able 

 researches of our friend Mr. Waterhouse liave shown us 

 that this conclusion, although sanctioned by the opinion 

 of former writers, as well as by a certain general external 

 resemblance between the animals of the two groups, is 

 directly at variance with their osteological characters^ and 

 these, as clearly made out by Mr. Waterhouse, constitute 

 the only sure basis for the classification of this difficult 

 order of Mammals. The views which that gentleman 

 entertains respecting the position of the Arvicohe will be 

 best given in his own words. Alluding to the genera 

 Ondatra^ Arvicola, and Lemmus, represented respectively 

 by the American Muskwash, the Voles, and the Lem- 

 mings, he says : " The animals comprising these groups 

 have all the essential characters of the Muridce, but dilFer 

 in having rootless molars, and in the form of the lower 

 jaw. They have, moreover, some peculiarities in the 

 structure of the cranium, which have been pointed out. 

 Here all the characters alluded to are combined with 

 three true molars, the normal number in the Muridd', 

 and may be conveniently used to define the Arvico/idce 

 as a sub-family of that group. In my paper on the 

 ArvicolidcE I had placed in that section, besides the 



