The relationship between the palaearctic and nearctic faunas 297 



swept barren land of the arctic region works in favour of flightless forms and species 

 (the dominance of the constantly brachypterous subgenus Cryobius, of the Carabid 

 genus Pterostichus, on the Siberian and Alaskan tundras is a good illustration); but 

 also, that the Bering Isthmus was available as a trafficable bridge at any period of 

 existence for animals enduring an arctic climate, and consequently that flightless 

 arctic forms were at less disadvantage crossing it than the more fastidious forms 

 of the forest regions. It is interesting to observe in this connection that no species 

 of Conifer trees (provided Picea glauca Moench. is specificially distinct from 

 abies L. and obovata Ledeb.) occurs on both sides of the Bering Strait (Hustich, 



1953)- 



Many circumpolar animal species seem to be uniformous, from a taxonomical 

 point of view, within their entire vast area, though this may be largely due to 

 imperfect study of sufficient series from different localities. In others, two or more 

 geographically separated forms, usually termed "subspecies", have been described. 

 One instance of a multiformous complex is the Reindeer {Rangifer tarandus L.; 

 fig. 53), in which the relationship between the different subspecies seems rather 

 obscure. 



The usual trend among circumpolar animals is that subspeciation, or whatever 

 it may be termed, has the character of a sliding change of certain morphological 

 properties which thus can be expressed as dines (in Huxley's sense) running from 

 east to west. A good instance is provided by the Moose or European Elk {Alces 

 alces L.; fig. 54), as investigated by Peterson (1952). Here, with respect to the 

 shape of the palate bones, a double cline is formed, running in both directions 

 from the E. Siberian— Alaskan area (diagr. 10). 



The result is, in this case and in the character mentioned, that the forms at 

 the periphery of the species' area, in Europe and eastern North America, are converg- 

 ing. A similar instance is provided by Davenport (1941) who, in his monograph of 

 the butterfly genus Coenonympha, the "Ringlets" or "Heaths", analyses the distri- 

 bution of the circumpolar, extremely multiformous C. tullia O. F. Miill. (fig. 55). 

 He found two points of general interest: (a) that northwestern North America and 

 northeastern Siberia are inhabited by the same form (sbsp. mixturata Alph.); and 

 (b) that the Nearctic form occurring farthest east (sbsp. inornata Edw.) is strikingly 

 similar to the forms of western Europe, especially to the Fennoscandian form 

 (sbsp. suevica Hemm.). Davenport is probably right in his conclusion that both 

 of these, geographically speaking, peripheral subspecies are similar, not because of 

 direct historical connection, but because they have remained primitive and like 

 the original ancestor of the species, whereas other forms situated closer to the 

 centre and origin of its area (probably in eastern Asia), but near the climatic 



