currant- and black-currant bushes were seen, strangely enough, to cover nearly exclusi- 

 vely the mountain sides, Uiousands of bushes forming, as it were, one continuous garden. 

 About Ujuk there occur small quantities of Devonian sandstone forced into the 

 eruptives, and farther down the Soyote Steppe the Devonian sandstone is nearly sole 

 prevailing again. Towards Ujuk the mountains become lower and more rounded; the 

 primeval foi-est — the moist taiga — gradually retreats again, and the wood becomes 

 more open and lighter, with an admixture of larches and various foliage trees bearing an 

 unmistakable evidence of a drier climate. In this region it is very interesting to study the 



Fii>. (io. From the Bti ken 



i ui jioroj^. The wood on the ri.^ht side h;)s been 



ravaged tjy forest-fire. 



steppe and the taiga fighting for the upper hand. Gradually the larch-forest with its 

 attending flora becomes nearly sole prevailing on the drier and warmer southern slopes, 

 while tlie taiga proper is now only to be found on the cooler and moister northern 

 slopes. But here, too, it is also at first by degrees mixed up with and later on altogether 

 replaced by the larch, whereby tlie last remains of the moist taiga, which were to bo 

 found in the tracts between Sebi and Ujuk, have disappeared. Here we find ourselves 

 once more in a transition zone, which, as to floristic conditions, is to be referred to the 

 wooded steppes. But soon it also becomes too dry for the larch: it begins thinning, at 

 first on declivities with a southern aspect, and is here slowly but surely forced to 

 yield for the benefit of a more xerophilous vegetation pressing forward, so as to 

 constitute a completely woodless and pure steppe scenery here. Thus, the soutliern 



92 



