122 HUGO SJORS 



the Russian Arctic (Lavrenko and Sochava, 1954; Sotchava, 1954; Tik- 

 homirov, 1960). 



For the Alpine barrens, a subdivision into three regions or belts is now 

 unanimously accepted in Scandinavia (e.g. Du Rietz, 1928, 1930, 1942a and b, 

 1950; Nordhagen. 1936, 1943). Each of the Low Alpine, Middle Alpine, and 

 High Alpine belts has a phytogeographical significance about equal to one of 

 the zones of the forested country. Similar belts are expected to occur in other 

 northern mountains, and also in high mountains at somewhat lower latitude, 

 e.g. the Alps (see a comparison by Du Rietz. 1930). A detailed comparison of 

 the Arctic and Alpine zonations must be left to the botanists of the future. 



CONCLUSIONS 



Ecological conclusion 



On comparing the two maps. Figs. 2 and 3, the general similarity is striking. 

 A closer analysis reveals two interesting features. 



1. All zones bend southward when near cold seas. The eflfects of Hudson 

 Bay (incl. James Bay), the Atlantic off Labrador and Newfoundland, and the 

 White Sea are evident. The Woodland-tundra in these sectors (and even the 

 Sub-Arctic near James Bay) is extraordinarily narrow due to this depression. 

 The Baltic, although cool in spring, has an opposite effect, presumably 

 because of comparatively long-lasting autumnal warmth. 



2. When not influenced by the cold sea effect, the width of the zones and 

 sub-zones is dependent on slope. If the general slope is southern, they tend to 

 be narrow (Boreo-nemoral of Quebec, sub-zones of the Boreal in Sweden). 

 Great width is often related to a northward slope (Main Boreal of Canada 

 west of Quebec, Sub-Arctic of Labrador Peninsula, Russian Main Boreal and 

 Sub-Arctic). This, of course, is an effect of altitude, and shows that such an 

 effect is present not only in mountainous areas. Even moderate elevation has 

 considerable impact on vegetation, an impact that has not always been duly 

 appreciated. 



It would require too long a report to discuss the climatic factors by which 

 the zonations are conditioned. It should only be mentioned that at least for 

 the Boreal sub-zonation the length of the growth period appears to be highly 

 important. It is defined as duration of the period when mean temperature is 

 above +5.5" or 6". A lower temperature limit makes cool oceanic areas seem 

 more favourable than they are. Summer heat seems to be more important 

 near and above the tree-line. Winter cold is hardly of any general importance, 

 and there is little relation to the distribution of permafrost (Brown, 1960). 



Historical conclusion 



The present zonation of the northern vegetation is young. It was practically 

 non-existent as late as 9000 years ago; only 4000 and even 3000 years ago it 



