38 ELEVENTH REPORT. 



as the (1) arctic-relict, (2) the pre-glacial alpine. (S) the endemic and (4) 

 the subalpine-lowland. This last element should in no wise be designated 

 alpine, extending as it does very generally out into the contiguous lowlands 

 as well as being lowland in point of origin. A fifth element of ruderal species. 

 e. g. Anaphalis margaritocea, Galeopsis tetrahit, Prunella vulgaris, Plantago 

 major, etc., which have followed man in his explorations and lumbering 

 operations may be added for sake of com])lete analysis. 



An enumeration shows upon Ktaadn and its lower slopes, mainly above 

 the two thousand foot contour, 221 species and varieties (2,7). Of these 

 no less than 111 or 46% are distinctly of the fourth element. It is at once 

 apparent that am* attempt to distinguish the pre-glacial alpine from the 

 arctic relict species is a priori hopeless, and further for our discussion such a 

 distinction would carry but little value. Nevertheless, all evidence seems 

 to point to the conclusion that the former must exist as a small minority 

 in both the alpine and arctic formation. 'We thus distinguish 118 species 

 as comprising the arctic-alpine element which embraces 54 per cent of the 

 total flora. Of the third category may be noted the single initial endemic 

 Carex Grahami, while Euphrasia Oakesii is an examjile of relict endemism. 

 having its only other record station upon Mt. AVashington (11). 



The glacial theory postulates a pre-glacial. continuous, arctic, circum- 

 polar land zone which under approximately identical climatic conditions 

 and unrestricted latitudinal migration would have possessed an almost uni- 

 form flora. "With the segregation of Northwestern Europe from North- 

 eastern North America and Northeastern Asia from Northwestern North 

 America as continental evolution progressed certain isolated and distinct 

 floristic regions came into existence as has been so clearly shown by Hooker. 

 It is from these segregated arctic floras which were driven southward in all 

 longitudes during glaciation that we must derive our arctic-relict forms. 



In order to point out the exact arctic regional alliances of these 118 species, 

 we may, following Hooker (9) divide the Arctic Realm, which includes a 

 circumpolar belt of 10° to 14° in breadth lying north of the arctic circle, 

 66^° north latitude, which forms its southern boundary, into five regions: 



(1) Arctic Europe, extending from Lapland eastward to the Gulf of Obi; 



(2) Arctic Asia, which begins with the Gulf of 01)i and extends to Bering 

 Straits: (3) Arctic Western America embracing the territory from Bering 

 Straits to the Mackenzie Basin; (4) Arctic Eastern America, from the 

 Mackenzie to Baffin's Bay; (5) Arctic Greenland. 



To show more clearly the lines of migration of our relict species and their 

 present diverse affinities an American Realm may be defined as extending 

 from the Arctic Realm southward to the United States and composed of two 

 regions: (1) Northwest America, comprising the band of country reaching 

 from Arctic Western America and traversed by the Selkirk's and Canadian 

 Rockies; (2) Northeastern America, the region lying between the southern 

 Canadian boundary and the Arctic Realm. It includes Laborador, New 

 Brunswick and New Foundland. 



A third realm, the White Mountain Region, located 161 miles to the south- 

 east, but only 60 miles due south, is introduced to demonstrate the striking 

 differences which may exist between the floras of two mountains within view 

 of each other. This region includes the Presidential Range of which 

 Mt. Washington is the principal peak. 



