reduction ot roughly 20.000 km.-, or nearly one third of the whole present woodland. 

 Here the low t e m p e r a t u r e (tiie main temperature of the 4 summer months) dra\\s 

 the limit on tlie heights, w^hile, in the interior of Asia, in my opinion, the drought 

 limits the expansion of the forest in the lowland. It is therefore, as already mentioned, 

 especially in lower regions that the forest in the interior of Asia is apt to be dried up, 

 and it is also in the same regions, where, according to Krassnow's statement, it has 

 the greatest difficulty in invading burnt areas again. And as the forest, which has like- 

 wise been reported by Krassnow too, is more apt to be regenerated at higher altitu- 

 des, I consider this as a natural consequence of the greater amount of moisture, and the 



Fig. 42. From the south side of the Sayansk mountains at Tshernoretska. In the Ijackground 

 dry trunks of spruce; the sound trees are larch. 



forest is, accordingly, given much better and more natural conditions of existence in 

 these tracts. It is also really here, in the higher regions of the Sayansk, that I found 

 the densest coniferous forest of the finest growth, as mentioned above. 



Similar climatic changes of cyclical as well as of quite irregular nature, both in 

 point of temperatures and conditions of moisture, even within, geologically -speaking, 

 comparatively short periods, are, as in known, far from being any isolated pheno- 



70 



