Chap. I] THE FORMATIONS 171 



fatal level. As the juniper has absolutely no power of emitting suckers from its 

 roots or even supplcmentarj' shoots from the base of the stem, there results a 

 low tabular little tree, whose dense umbrella-shaped crown attains a diameter 

 of 3-4 meters, and whose central cylindrical stem at an age of 300-400 years 

 may have a diameter of more than thirty centimeters. The height of the whole 

 plant is, on the average, about i meter, hut may occasionally reach nearly 

 2 meters. ... If the little stem, which is in great request as firewood, 

 should remain standing long enough, there comes, sooner or later, a moment 

 when the adherence of the roots to the soil can no longer resist the increasing 

 pressure upon the crown b)' the wind ; the little tree falls over and is obliquely 

 held up by the half of its crown that is now turned downwards, whilst its 

 jpper half quickly dies off and disappears along the critical line. 



The line above which all twigs perish is defined bj' the average height of 

 he snow-covering at the commencement of the thaw. ... In April, 1S89, 

 I was able to satisfj- m^-self that the living juniper branches reached close up to 

 he surface of the melting snow, or that at the most thev projected a few 

 ■entimeters above it. I have convinced mj'self that the juniper in Russian 

 -apland can withstand the winter only if for several months it is completely 

 overed with snow'. . . . The birch also assumes the form of tabular or 

 :losely cropped bushes, which are quite characteristic of the tundra landscape, 

 n accordance with the wide distribution of this tree beyond the limit of forests V 



Finally, as an extreme case, Kihlman mentions the formation of mats, ' that 

 Merely reach the height of the surrounding felt of lichens or undershrubs, but 

 ■hich occasionally attain quite considerable dimensions in the horizontal plane. 



. . The shallow-rooted spruce assumes this form most successfully (Fig. 92) ; 

 long the tundra belt near Orlova I saw mat-like spruces from one to five 

 iieters long, the thin sterile twigs of which had crept about in the felt of lichens 

 nd evidently all sprang from one seedling plant. ... In all these mats one 

 nds that the onr-ycar-old shoots, in so far as they piojict aboiw tlie level of the 

 iirroiiiidiiig tiiftg of moss and lichens, are dried up and defoliated^.' 



As will be shown in a later chapter'', the conditions iii regard to the 

 novements of the air also determine the vertical limit of forest vegetation. 



It appears to be advisable to group together s/inibs and dzvarf-trccs as 

 ntsIn<.<ood. 



The existence of brushwood also is determined by the amount of water 

 1 the soil, and the season when the water is renewed is a matter of in- 

 ifference. The amount of water necessary for the well-being of brush- 

 "ood is less than that for tree-growth ; when it increases, tree-growth 

 ppears. Like forest, brushwood thrives better in damp than in dry 

 ir, and better in calm than in agitated air ; in both respects, however, 



is more accommodating than is tree-vegetation. 



A good tvoodlaiid climate is, thai, composed of the folloiving elements : — 



' Kihlman, op. cit., p. 71. - Id. p. 73. 



' Id. p. 68. * In Part III. Sect. IV. 



