219 



plants that manage to live on them at the start, are 'accordingly, as a rule, 

 those hard starvelings, accustomed to be content with a very modest spot 

 on the earth, which can make a shift to subsist on what little there is of 

 food, and capable of defying the unfavourable conditions of soil and climate; 

 these are generally the littoral and the mangrove-plants. The latter require 

 mostly for their growth an oozy muddy soil, whereas the genuine beach plants 

 will put up even with comparatively bare coral-rocks. As a general rule the 

 first plants are conveyed to these islands by ocean-currents. Among these 

 plants also belongs the coco-palm. In the opinion of Mr. Docters van 

 Lee U wen the study of Krakatao has established it as a fact that the coco- 

 nut is capable of spreading independently of man. Therefore, on finding a 

 coco-palm sprouting in the driftwood at only a few yards' distance from the 

 sea, he fixed this fact by means of a photo. After certain pioneers borne by 

 the sea have been washed ashore on the islands, these are visited by birds, 

 which may carry all sorts of seeds attached to their feathers or in their 

 intestines. Moreover the wind also brings seeds and spores. 



I cannot agree with the somewhat fantastical description given 

 here of the character and manner of development of the flora on 

 very young coral-islands. I pointed out already (See p. 63) that such 

 islands consist, so to say, entirely of a young beach, hence contain 

 in their soil much NaCl, lack in their centre a basin of sweet 

 water (stored rainwater) and therefore can bear, at least in the 

 east-monsoon, only a halophilous vegetation '). There is no reason to 

 suppose that their lack of fertility is such as to prevent forest from 

 developing on them. Tropical forests do not require a fertile soil 

 though individual species may do; in many regions of Java it may 

 be observed that forests thrive quite well in unfertile localities, if 

 only they can dispose of a sufficiently c/ecp soil and a sufficient 

 amount of water. But young coral islands do not possess a deep soil, 

 neither at least in the east-monsoon in that soil a great water- 

 supply, hence they can bear no forests but only scattered trees or 

 not even these. As coral-islands grow older and increase in size, the 

 coral weathers to some depth and they may gradually store rain-water 

 in their centre. Such old coral-islands are very often densely wooded, 

 provided the trees are spared by man, which is frequently not the case. 



That young coral-islands need not at all be unfertile is abundantly 

 shown by the vigorous habit of the herbaceous and shrubby halo- 

 phytes often copiously growing on them. These are by no means the 

 ,,hard starvelings" Mr. Docters van Leeuwen speaks so pathe- 

 tically of, but often quite robust individuals, needing no man's pity. 



The edaphic differences between the habitats of mangrove-plants 

 and ,,genuine beachplants" have not well been grasped by Mr. 



*) In the west-monsoon one often finds on young beaches seedlings of non-halo- 

 philous species, but these die off the next east-monsoon or already earlier. 



