12 



South-Sumatra. By these watercourses seeds are, for a part at least, 

 conveyed to the sea which may carry them yet further. It has never 

 been proven that after having reached the sea all these seeds soon sink 

 to the bottom or lose their germinative power *). In the tidal creeks 

 of West-Java the water is often locality tinged brown by millions of 

 floating spores of the typical mangrove-fern Acrosticfwm aureum 

 L. When the tide is low these spores are carried off to the sea 

 and it is not probable that all of them soon perish there. It is a 

 well-known fact that light seeds readily adhere to driftwood, to 

 pumice or to other floating objects. In this way spores and seeds 

 might have been carried over. In the Westmonsoon when the rivers 

 discharging their water in the Sunda-straits are often much swollen 

 by continuous heavy rains, they not seldom carry eradicated trees 

 far into the sea. Seeds, fruits or epiphytes adhering to the emerging 

 parts of these trees can in this way reach Krakatao within the time 

 of ti single day -). Ernst 3 ) found in 1 906 two fungi growing on 

 a trunk washed ashore; he supposes that the mycelium, which had 

 formed new reproductive bodies, had survived the transport, but this 

 opinion, however plausible, is not corroborated by proofs. 



Schimper already expressed the opinion that in the tropics 

 winds take but a small part in the dispersal of seeds of littoral plants. 

 The arguments alleged by Ernst '') against this opinion are far 

 from conclusive. The assertion of Ernst: ,,Von ausserordentlicher 

 Bedeutung fur die Enstehung der neuen Krakatao-flora ist, wie 

 schon durch den ersten Besuch Treub's in iiberraschender Weise 



gezeigt worden ist der Wind", is unsufficiently accounted for. 



I he suggestions Ireub made on this point of the problem prove 

 nothing <jf all. Rikli ') gives a striking instance how easily one may 

 be mistaken when considering winds as a means of conveyance over 

 long distances: ,,Ein kleiner Vorfall aus dem Jahre 1903 zeigt, wie 

 bei der Beurteilung der Verbreitung von Blattern durch den Wind 

 gro'sste Vorsicht geboten ist. Bei der Ueberschreitung der vorderen 

 Furka zwischen dem Pommat (Italien) und dem Val Bosco (Tessin) 

 fanden wir in einer Hohe von 2300 m. Buchenblatter. Da die obersten 



') Birger ( Beiliefte zum Botanischen Centralblatt XXI (1907), p. 263, seq.) has 

 shown that many seeds of European non-halophytes after having been submerged 

 during several days in a salt solution of 3,4% had retained their germinative power. 

 I do not know of any experiment made on this point with the seeds of the plants 

 found on Krakatao. 



2 ) Ernst, Neue Flora Vulkaninsel Krakatau (1907), p. 56. 



3 ) Ernst, Neue Flora Vulkaninsel Krakatau (1907), p. 58. 

 _') Ernst, Neue Flora Vulkaninsel Krakdtan (1907), p. 62. 



'; Rikli, in Fortschritte Naturwiss- Forschung III (1911), p. 259. 



