1881.] Himalayas, Tibet, and Afghanistan. 93 



Arvicolce with rooted molars in the adult state. By both Baird and Coues 

 Evotomys and Hypudceus are classed as a distinct genus. The true Arvicola 

 of North America are divided into four subgenera. As none of these are 

 Himalayan, it is unnecessary to specify them more fully. One of these 

 sections {Herniotomy s of Baird, Myonomes of Coues), like the European 

 Agrieola and the Himalayan Neodon, has 3 salient angles on the inside of 

 the second upper molar ; but it has a 7-spaced first lower molar with but 

 4 outer and 5 inner angles, instead of a 9-spaced tooth with 5 outer and 

 6 inner angles. 



We may now return to the Himalayan and Tibetan forms. None, so 

 far as I know, have rooted molars. I have not been able to examine all 

 the species ; but as there are no roots to the molars of A. sikimensis, A. 

 wynnei, A. blythi, A. blanfordi, or A. stracheyi, I think it improbable that 

 the allied forms should exhibit so remarkable a character. All but two 

 A. sikimensis and A. melanogaster, have the first lower molar, as in 

 Blasius's section Paludicola (and as in Hypudceus or Evotomys), with 7 

 prismatic spaces, and normally with 4 outer and 5 inner angles, a fifth 

 more or less rudimentary outer angle in front being present in some forms. 

 The second lower and the second upper molar have also the same structure 

 as the corresponding teeth in European forms of Paludicola — the circum- 

 stance that in some forms there are but 3 double spaces in the second lower 

 molar instead of 5 simple prisms being, in fact, of no structural importance 

 since the difference depends on whether the angles on opposite sides of the 

 tooth correspond or alternate ; and there is every possible gradation between 

 one condition and the other. The Himalayan species, with 7-spaced an- 

 terior lower molars, however, are divided into two groups by the structure 

 of the hinder upper molar. In A. stoliczkanus, A. stracheyi, A. roylei, A. 

 blanfordi, and A. wynnei this tooth terminates in a narrow elongate lobe, 

 produced backwards in the direction of the row of teeth, and posteriorly 

 destitute of salient angles ; whilst in A. blythi and in A. tnandarinus (?) 

 the last upper molar terminates, as in A. amphibius and its allies, in a 

 crescent or transverse lobe, with a well-marked internal salient angle. 

 The latter two species appear to belong to the European section Paludi* 

 cola ; for the former I venture to propose a new section, Alticola, of which 

 A. stoliczhanus may be considered the type. In this form and in the 

 nearly allied A. stracheyi the structure of the last tooth is so peculiar that 

 they certainly deserve distinction. There are two small external angles 

 anteriorly and two posteriorly, the two pairs being separated by a deep 

 sinus, and there are but two interior angles, both, however, much stronger 

 than those on the outer side of the tooth. These two typical forms of 

 Alticola, moreover, are well distinguished from the others by the absence 

 of any claw to the thumb. Of the other three forms A. ivynnei is distin- 

 13 



