510 bulletin: ml^seltm of comparative zoology. 



tundra-living animals coming in from the west. If we assume that 

 the Labrador Lemming represents the original American stock, we 

 may further postulate that it became wiped out west of Hudson Bay, 

 either by imfavorable conditions at the south of the great ice-sheet 

 in glacial times, or more probably, by a subsequent in^•asion from the 

 Old ^Yo^ld of the more highly specialized type which now prevails, 

 associated with Lemmus of somewhat similar habits. The common 

 ancestry of the two types must therefore be sought still farther back 

 in time. Such an invasion by Old World genera was possible in late 

 Pleistocene times when "a broad land bridge with Asia existed, and 

 the continental shore line extended far north into the present Arctic 

 Ocean" (Osborn). The possible former existence in Europe of a 

 lemming with teeth like those of the li\-ing Labrador species, is sug- 

 gested by the figures of Hensel (1855, pi. 25, fig. 12), drawn from 

 Pleistocene specimens from Quedlinburg, Germany; he admits that 

 they are not wholly accurate, yet his description corroborates the 

 general features indicated. Sanford (1870) too has figured the lower 

 jaw of a lemming from the cave-deposits of Somerset, England, 

 which if correctly draA\Ti, resembles that of the Labrador species 

 in lacking the antero-internal accessory lobe of enamel in the last 

 molar. 



L ntil twenty years ago, it was customary to refer the Collared 

 Lemmings of both Old and Xew Worlds to the single species, torquatus, 

 described by Pallas in 1778, from northwestern Asia. The same 

 author, however, pointed out the color-characters distinguishing the 

 Labrador species, which he named hudsonius; yet it was not until 

 1897 that the distinctness of this species was generally recognized. 

 Three years later, in 1900, Dr. C. Hart ^Slerriam described as new 

 species the Collared Lemmings of Alaska, Unalaska, and the west side 

 of Hudson Bay. The status of these forms has since remained 

 uncertain, and recent writers have usually grouped them as sub- 

 species of hudsonius. The excellent series of Dicrostonyx from Arctic 

 America in the Yluseum collection has induced me to undertake a 

 revision of the American forms in an attempt to define more clearly 

 their relationships and distribution. I have had for examination, 

 in addition to the series in the M. C. Z., nearly all the specimens 

 available in the United States; yet there is still a great lack of material, 

 particularly of summer skins, from the archipelago north of Hudson 

 Bay, and from the Alaskan Peninsula. The Old World species are 

 almost unrepresented in American museums, so that it has been 

 impossible to consider them in any revisionary way. 



