E. V. Wulff —160— Historical Plant Geography 



points on this subject, often quite contradictory, have been expressed. 

 Despite their great interest, we cannot go into them here. The chief 

 cause of the many disagreements was the chance nature of the findings 

 of plant remains and the lack of precision in determining the numerical 

 ratios of the various species in the deposits, this leading to contradic- 

 tory interpretations. Fortunately, we now possess a more perfected and 

 precise method of determining the species composition of former floras 

 and the numerical ratios of the component species. This is the method 

 of pollen-statistics, which involves a study of the fossilized pollen of 

 plants found in peat deposits. 



Method of Pollen-Statistics: — Fossilized pollen grains in post-glacial 

 deposits have been known since the end of the nineteenth century. 

 The first work on a calculation of species ratios based on pollen-statis- 

 tics was published by Weber in 1896, and the first methodological 

 principles of pollen analysis were worked out by Lagereieim in 1905 

 and 1909. But a detailed elaboration of the method of pollen-statistics 

 was made only in 191 6 by von Post and his associates, the method 

 being at first applied chiefly in Sweden. It has now become widely 

 adopted, and, as a result, we already have considerable data and maps 

 of the former distribution of forests and the species composing these 

 forests for a number of European countries. At the same time, the 

 method itself has been further elaborated and made more precise. 



The method of pollen-statistics is based on the fact that the pollen 

 grains of different species of plants may be distinguished from one 

 another by their specific structure. The structure of the outer coat of 

 the pollen grain is specific for each genus and often for each species, 

 constituting a good character for taxonomic classification. Moreover, 

 the pollen of many species is well preserved in peat deposits, which 

 makes it possible by studying such fossilized pollen not only to deter- 

 mine the genera and species comprised in a forest at the time the pollen 

 was produced, but also, by calculating the number of pollen grains for 

 each species, to determine the relative proportions of the various species 

 in the given forest. 



The many investigations of this kind that have been made have al- 

 ready elucidated quite precisely for a number of countries the com- 

 position of the vegetation at different stages during the post-glacial 

 period. As an example, we may take Rxidolf's (1931) conclusions, 

 based on a summary of all available data, whereby he divides the 

 history of the forests of central Europe into the following four periods: 



1. Birch-Pine Stage {Pre-Boreal Period). Throughout all of central 

 Europe the chief trees are the pine (Pinus), the birch {Betula), and the 

 willow (Salix). The predominance of the birch, and also of the willow, 

 is greater toward the west and north. Other forest species are sporadi- 

 cally, though in places quite widely, distributed. The spruce (Picea) 

 for some time is dominant in central Russia, and it is also widely dis- 

 tributed in the Carpathians and eastern Alps. At times, instead of the 

 birch, the pine becomes predominant. 



2. Hazel Stage (Boreal Period). The hazel (Corylus) frequently 

 occupies a dominant position, particularly in the west and in moun- 

 tainous districts of the east. Oak woods begin to become widely dis- 



