RODENTIA ARVICOLINAE ARVICOLA PIXETORUM. 547 



has one anterior, one interior, and two exterior angles, the first and last lateral heing both 

 exterior, or the reverse of the first molar. The posterior upper molar has hardly any lateral 

 triangles, the indentations of opposite sides being opposite and meeting, instead of alternating. 

 There are, then, an anterior and a middle transverse triangle, with a posterior V-shaped loop, 

 the outer leg of which connects with the adjacent inner salient angle. This V is usually 

 slightly concave on either side, so as to make a Y. There are thus three inner salient angles, 

 and two external, besides the V; the upper part of the external leg of this may sometimes carry 

 a small additional external salient angle. 



In the lower jaw the anterior molar is about as long as the two succeeding ; the middle 

 longer than the last, which is narrower than the preceding. The anterior molar has one 

 posterior triangle, one external, and two internal ; the anterior lobe has a deep re-entrant acute 

 angle on either side, cutting off salient angles, and the sides of the anterior terminal loop are 

 slightly indented, so as to be concave. There are thus an anterior loop, five salient angles on 

 the inside, and four on the outside. The dentine of the two anterior on either side communicates 

 with that of the loop. 



The middle lower molar has an anterior and a posterior triangle, and a lateral on either side, 

 making four. The posterior has the two lateral thrown into one transverse, making three 

 transverse triangles. 



With a limited number of specimens before me, when I first took up the examination of this 

 species, I felt constrained to establish several species on what I considered good characters. A 

 revision of the subject, however, with much more material at my command, has shown so many 

 connecting links between the species, as established, that I can scarcely do less than throw them 

 all into one. The size of the ear and of the body varies in different specimens from the same 

 locality. The colors vary from rather bright chestnut to a darker tint, almost brown. This 

 seems to be more usually the case with specimens from Illinois, although some are very similar 

 to South Carolina skins. As a general rule, specimens from the low lands of the southern coast 

 are smaller than those in the mountains or to the north. A specimen in alcohol, from Long 

 Island, (Sag Harbor,) the locality of the original Arvicola scalopsoides, of Audubon and Bachman, 

 exhibits no difference of apparent importance. 



In one skin from Society Hill, the middle of the back is of a dark rufous brown ; the 

 sides bright yellowish chestnut as a broad stripe sharply defined against the color of the back ; 

 the belly is hoary whitish ash. The tail is bicolor. The contrast of color between the back 

 and sides, I have observed in no other specimen. 



It is quite possible that there may be more than one species included in the specimens enumerated, 

 and it is more than likely that those from Illinois may eventually prove distinct. 1 Their most 

 striking characters are the longer and looser fur, and its darker or more purplish color and gloss. 

 I find no permanent characters, however, by which to separate the Long Island scalopsoides 

 from the southern pinetorum. 



1 Should this Illinois species really be distinct, it may very appropriately be called Arvicola kennicottii. It cannot, with 

 propriety, bear the name of scalopsoidea. 



