224 KNUT F.EGRI 



flora had been subject to a very strong forest selection; only those biotypes 

 able to survive under the conditions of dense forest could persist unless they 

 found refugium in an "atypical" habitat. Such habitats were found above the 

 (dense) forest Umit in the mountains or in places covered by some non-climax 

 vegetation. That would mean bogs, beaches (of the sea, lakes or rivers), 

 screes, or fissures, cracks, or shelves in steep rocks. This again means that 

 species with ecologically special requirements (chamaephytes, halo- or 

 hydrophytes, etc.) would have a certain chance of survival, and so would plants 

 that could grow in the upper, open mountain forests. On the other hand, the 

 ordinary open-ground species, above all those of the meadows, would have a 

 very difficult time, although we should not underrate the effect of grazing by 

 the spontaneous (deer) fauna. To what extent spontaneous forest fires or 

 catastrophic gales (Sernander, 1936) influenced the picture, remains con- 

 jectural. 



What happened when land was originally cleared for agricultural purposes 

 is a phenomenon whose nature cannot only be inferred but also confirmed, 

 on the whole, by information from pollen analyses. The effects may be 

 summarized in the following statements: 



1 . Certain plants were directly used for different purposes and their number 

 reduced by selective utilization. 



2. Certain plants occur in soil which was preferred for agriculture, for 

 which reason they were more or less completely exterminated together 

 with the corresponding vegetation types. 



3. On the soil thus vacated there were possibilities for the estabhshment of 

 new vegetation types composed of (a) archaeophytes — many of them 

 previously suppressed or restricted by and in the chmax vegetation, 

 (b) cultivated plants, and (c) immigrant weeds and ruderals. 



The difficulty here consists in distinguishing between Groups 3 (a) and (c). 

 To cite an example, pollen evidence shows unmistakenly that Plantago 

 ianceolata expanded enormously with the introduction of agriculture, but are 

 the very few P. Ianceolata grains observed before the land-clearing phase due 

 to long-distance dispersal and or contamination, or do they mean that some 

 few P. Ianceolata specimens existed even before agricultural land was cleared ? 

 And what about the many subspecific taxa within this species? A similar 

 question may be raised for P. major, but whereas it is easy to imagine a 

 possible refugium for this species on beaches (the beach-form may be second- 

 ary!), it is more difficult to visualize where P. Ianceolata could have taken 

 refuge, except on those shelves in rock walls where I am afraid we shall feel 

 obliged to put up a major part of our flora if we do not accept the possibilities 

 of its anthropogeneous origin. Where else can we place a plant like Digitalis 

 purpurea ? 



It is very difficult to give any estimate of the number of species in these 



