72 



Krakatao in 1886 consisted mainly of ferns, the island was before the 

 eruption totally covered with a Phanerogamic vegetation. But this 

 is not all proven for the basalts of Zwarte Hoek, nor is it, consid- 

 ering the vegetation in such-like localities in the same kind of 

 climate, probable, And it is just as little probable that ferns prepare 

 the locality for Phanerogams. By the mechanical and chemical action 

 of the roots and also by retaining decaying vegetable matter and 

 fruits or seeds which otherwise might be washed away, a kremno- 

 phytic vegetation can, of course, prepare the way for a subsequent 

 vegetation which puts higher claims on the fertility of the soil. But 

 this is only a mechanical-chemical-oecological succession, not a trans- 

 ition from ferns to Phanerogams. In Krakatao the kremnophytic 

 vegetation could for the greater part consist of ferns because of 

 the island receiving in the dry monsoon comparatively much rain; 

 in other parts of the Dutch Indies (the much drier island Madura *) 

 f.i.) the kremnophytic vegetation consists mainly of other plants. 

 In Krakatao f/?e/ - e has not been a succession Cryptogams-Phanerogams 

 as such, but only a survival and victory of such species present as 

 were momentarily the most fit to live under the existing conditions. 

 As long as the soil is but little weathered and unfertile it bears a 

 kremnophytic flora, which as the fertility increases, is sooner or later 

 superseded by other, more exacting plants. The speed of succession 

 depends mainly on the conditions offered by the soil; the sooner 

 these are modified, the sooner the vegetation changes- The lava- 

 streams which Verbeek in 1880 -) saw in the no more existing 

 northern part of Krakatao must have been at least 200 years old and 

 yet they bore only a very scanty vegetation. Of the rocks visited by 

 Treub in 1886 Penzig said 11 years afterwards that the aspect 

 of the vegetation for the greater part was unchanged. The general 

 aspect of the vegetation of steep road- and riversides in the sur- 

 roundings of Buitenzorg has, in the course of at least 20 years, not 

 perceptibly changed. Rather more than a century ago Daendels 

 built his renowned road from West- to East-Java. Wherever this road 

 has been cut through a hill-side the vertical parts of its banks, if not 

 bare, are still covered with kremnophytes. In localities more favourable 

 to plant-growth the vegetation may doubtless change much more 

 rapidly, but nowhere Ferns, as such, precede the Phanerogams. 



It is not at all proven that the present forest and grass-jungle on 



') Madura is a comparatively small island lying north of the eastern part of |ava- 

 '*) See p- 16, footnote 3. 



