COUES ON INSECTIVOROUS MAMMALS. 643 



All are highly aquatic, comparing with the species of Sorex and Bla- 

 rina as a Mink or Otter does with most Mustelidce, or as a Muskrat with 

 ordinary Arvicolw. 



To the type-species is to be added the Sorex imlustris of Richardson, 

 and possibly another species. 



II. Genus SOREX, Linn. 



(a) 



Sorex, Linn. (Type S. vulgaris, fide Nathusius, Arch. f. Naturg., 1838, 45.) 



Sorex of Authors referring to S. vulgaris (not of those applying the name to Crocidura 



aranea). 

 Corsira, Gray, Proc. Zool. Soc, 1837, . . . (Type . . .) 

 Otisorex, De Kay, Zool. N. Y., 1842, . . . (Type 0. plaiyrhinus.) 

 Hydrosorex, Duvbrnoy, 1836, nee 1834. 

 Amphisorex, Duvernoy, 1834, nee 1836. 

 Hydrogale, FoMEJ., . . . 1848. (TjT^e S. fimiripes Bach. Genus based in error.) 



Paradoxodon, Wagner, . . . 1855. (Separate subgenus.) 



(c) " 

 So7-iculus, Wagner, . . . 1855. (Separate subgenus.) 



id) 

 Microsorex, Baird, MSS., 1861. (Separate subgenus. Type S. lioyi Bd.) 



(e) 

 Notiosorex, Baird, MSS., 1861. (Separate subgenus. Type S. crawfordi Bd. n. sp. infr^.) 



In the present uncertainty which obtains respecting the limitation of 

 the original Linufean genus, I do not attempt to give a diagnosis cov- 

 ering all its modifications. Its value is very differently rated by authors, 

 some confining the term to *S^. vulgaris and its strict congeners, while 

 others, even among the latest systematists, make the genus coextensive 

 with the family, as i'eters and Mivart* have done. I think that the 

 genus may be most naturally defined to exclude Grossojms and Croci- 

 dura with their respective allies, and Blarina, making subgeneric sec- 

 tions according to the numerical composition of the teeth, though some 

 of the forms included in the foregoing synonymy may be worthy of full 

 generic rank. It will tend to prevent confusion to hold S. vulgaris as 

 the type. 



To the genus Sorex, as thus rigidly restricted, belong nearly all the 



* According to the latter author, the constants of the teeth of Soricidw are as fol- 

 lows : — 



. — 1-1 — 3-3 



i. izi'^--izi'i"»-in'-:5=3; 



the variables being confined to the upper incisors and upper premolars, which may be 

 either 



2-0 3-3 2-^ 3-3 1-1 '^-2 1-1 



pm. — -; or i. '— ^, P»»- . — -} or t. , pm. ; or i."- — , jjm. — 



8bh 



