Chap. I] THE FORMATIONS 163 



in millions by the slightest breath of wind and i4rcatly facilitate its 

 general distribution, whilst its creeping and deeply penetratinf roots 

 increase the difficulty of eradication when once this grass, so tenacious 

 Df life, has established itself I have reasons for believing that while 

 :he land was in its original condition the alang-grass was restricted to 

 sundry sterile, arid, waterless tracts of the hot zone, and was chiefly 

 ;onfincd to heavy, hard, easily dried clay soil, with an iron-pan, but that 

 It the present time, wherever we meet this grass on a fertile light soil and 

 HI mountain-slopes at above 2,000 feet, this state of aftairs is first brought 

 ibout by the hand of man. ... In Northern Sumatra, especially in the 

 Satta country that has been devastated by war, grassy wastes have conse- 

 luently come into existence which cover everything far and wide with 

 hideous uniformity and overrun plain, mountain, and valley with their 

 diitish-grecn mantled' 



It is probable that, unless man should again intervene, the alang- 

 tcppe will, in the course of time, again give way to the forest, for 

 imatic conditions arc in every way more suited to forest than to grass- 

 ind. In a decided forest climate, for example along the Brazilian coast, 

 Drest quickly succeeds to devastated forest, although it is of a more 

 erophilous character than before, being the so-called ' capoeira.' 

 Desert, the third leading t}-pc of climatic formation, originates when, 

 n account of too great drought or cold, climatic conditions are hostile 

 3 all vegetation ; the types of both woodland and grassland then become 

 :unted and their differences become obliterated, for the straggle between 

 lem ceases. The soil is then monopolized by such woody or herbaceous 

 lants as can still contend successfully against the inclemency of the 

 imate. Transition forms between desert on the one hand, and wood- 

 uid or gra.ssland on the other, are termed semi-deserts. 

 In the woodland and in the grassland such plants alone can thrive as 

 re at their oecological optimum in respect to all external factors, other- 

 ise they would perish in the struggle with stronger competitors. In 

 eserts this is no longer a necessary condition, as the struggle between 

 le plants ceases. Woodlands and grasslands are closed formations, at 

 ast in an oecological sense ; more components cannot be admitted 

 to them and numerous seedlings are continually perishing in the general 

 Miflict. The desert, on the contrary, is oecologlcally an open formation, 

 lost seeds do not germinate in it, and seedlings frequently succumb to 

 le inclemency of the climate. Others prolong their miserable existence, 

 lan)' plants die and their places are not reoccupied. There are always 

 any vacant spaces to be filled in the desert". 



' Junghuhn, op. cit.. Bd. I, p. 153. 



Very sparsely stocked s/a/io/is in climatic districts suitable for woodland or grassland 

 ist not be confounded with climalic deserts. See under Edaphic Formations, p. 176. 



M 3 



