544 bulletin: museum of comparative zoology. 



possesses a very complex flora. This flora, according to Fernald 

 (1911), contains representatives of four distinct centers of dispersal: — 



1. Boreal types, which gained access to Newfoundland probably 

 somewhere from the north. 



2. Western or Canadian types not included in the above, many of 

 these being species which, although not found today in any of the 

 country east of the Canadian Rockies, appear again in Newfoundland. 



3. Southwestern types, representing the remnant of a coastal 

 plain flora which flourished on an off-shore barrier beach or chain of 

 islands extending along the edge of the continental shelf. 



4. Endemic plants or species unknown on the iVmerican continent, 

 but finding their nearest relatives in identical or related species of the 

 Irish or neighboring coasts. 



Bangs (1913, p. 509, 510) has shown that the evidence afforded by 

 the mammalian life of Newfoundland bears out the existence of at 

 least two of these centers of dispersal, viz. — a boreal center to the 

 north and coastal plain center to the south. In view of the fact that 

 many of the typical Labrador mammals, viz. — ^ Moose, Marmot, 

 Porcupine, Squirrels, certain Mice, Shrews, etc., are absent from New- 

 foundland, it would appear that this region has not been recently 

 connected with the mainland. The mammalian life of Newfound- 

 land consists of a flux of two elements : — a southern coastal plain 

 fauna, represented by the Vole, Muskrat, etc., and a northern tundra 

 fauna, represented by the Caribou, Hare, etc. 



An analysis of the bird fauna lends some support to the views out- 

 lined above, but there is no evidence among any of the vertebrates 

 of the existence of an Atlantic landbridge between Newfoundland and 

 the Irish coast. Most of the evidence afforded by the birds is of a 

 negative nature, and yet the absence in Newfoundland of many 

 typical Labrador birds strangely suggests a barrier between New- 

 foundland and the mainland. The natm^e of this barrier is well 

 expressed by Fernald (1918, p. 238): — 



" In explaining ^ the migration to Newfoundland of a large element 

 from the Atlantic coastal plain of the United States it has been nec- 

 essary to reconstruct the Tertiary continental shelf, which is now 

 depressed as a shallow bench off the east Atlantic coast of America; 

 and -from the botanical and zoological evidence, as well as from re- 

 cently published geological evidence,^ it now seems perfectly settled 



1 Rhodora, 13, 135-162, 1911. 



* Barrell, Amer. journ. sci. ser. 4, 40, 1-22, 1915, 



