254 RIKARD STERNER 



Several forest trees belong to the continental element. It would not seem 

 rash to suppose that the western distribution-limits in Europe of these species 

 are partly caused by the climate. In this connection it seems especially natural 

 to think of the decreased season of rest, which should be able to have an unfavour- 

 able influence on the formation of wood (compare Dengler 1904 and 1Q12). It is 

 natural that many species in the ground flora of these forests should within not 

 too great areas cease to exist at the same time as the forest trees reach their 

 boundaries (compare Hock 1895 and »i896»). 



A universal feature of these continental species is, further, that they are found 

 only in soil of a more or less continental character, or in water with a compara- 

 tively great content of electrolytes (» nutritive waters»). As has already been 

 pointed out the species may be supposed to be in most cases confined to such 

 localities. As these naturally become more and more rare as the climate assumes 

 a more and more maritime character, this may be one of the causes of the 

 western distribution-limits of the species. 



Even species belonging to less continental vegetation-types, such as forest 

 communities and hydrophilous communities, may thus be supposed to be more 

 or less dependent on continental geographical conditions, thanks to certain eco- 

 logical qualities. Naturally, however, they are not dependent in the same degree 

 as the steppe species. The sorts of localities where they are wont to occur, 

 are much more widely represented in maritime regions than those of the steppe 

 species. The former have also in several cases a wide and even distribution 

 rather far out towards western Europe. 



Eor certain continental species, numerously distributed in the forest regions 

 of East Europe, however, their western distribution-limits are often so placed 

 that a connection with changes in the climate and the nature of the soil in a 

 maritime direction can hardly be discerned. The mode of occurrence of these 

 species within their distribution-area is not, as is the case in regard to the 

 steppe species, of the kind that suitable localities might be expected to be lacking 

 outside the limits. It is perhaps not too bold to seek causes of the limits, 

 regarding such species chiefly in other conditions than those mentioned. — 

 Besides, a detailed examination of the distribution-boundaries of almost all spe- 

 cies would probably show that there are many factors to be considered in deter 

 mining the situation of boundaries. The cause is not a certain climatic factor, nor 

 is it to be found solely, perhaps not at all in the cHmate. We must consider 

 other causes connected with the ecology of species, in the first place the nature 

 of the soil or causes of a totally difl"erent nature, such as causes connected 

 with migrational history and rivalry with other species. 



