DOCTERS VAN Leeuwen : GaMs of Krakatau and Sebesy. 289 



distance. Of thèse phenomena I hâve already treated in my article just 

 mentioned to which ! may refer the reader. At that time I was not aware of 

 an article by Felt '), on the distribution of the gallmidges, in which some 

 important observations are communicated regarding the manner of the distri- 

 bution of thèse créatures. He informs us e. g. that the gahmidge-galls may 

 be spread by birds, such as the galls formed by Cincticonia, which are eaten 

 by birds. \n galls with a thick-walled gall-cavity the inmates might perhaps 

 pass through the intestinal canal of birds without being digested. Still this 

 manner of distribution cannot be of great importance according to Felt, 

 who lays more stress on the spreading by the wind, when either the 

 créatures are simply blown along, or they actively aid the wind by flying 

 along with it. 



In my previous article 1 pointed to the possibility of the gallmidges 

 having corne over by floating or drifting from coast to coast inside their 

 galls. But it had not yet been demonstrated that such galls could stand 

 such a prolonged transport without péril. During my last stay at Verlaten 

 Eiland I was in a position to test this point, albeit for only one species of 

 gall. I was dwelling in a tent at the edge of the saltwater lake, whose 

 salinity does not differ much from that of seawater. In the vicinity of this 

 place the gall-midge gall on Clerodendron inerme was very prévalent. 

 I gathered a number of twigs studded with galls, tied them together and 

 then let them float about in the lake for a week fastened by a rope to a 

 po!e in the lake. A part of them was regurlarly above water, another part 

 was regularly submerged, and was tossed about, though not so briskly as 

 in a billowy sea. After this week ail the galls were still normal, they had 

 remained green and the gall-creatures inside the galls were very much 

 alive. I think 1 may assume that also other species may keep alive for a 

 shorter or a longer time in the sea, when enclosed in their galls. Ernst 2) 

 in his well-known study of the new flora of Krakatau has already discussed 

 the point that, wind and weather permitting, the distance from Java and 

 Sumatra to thèse islands may be traversed within a few days by things 

 carried by a current in the sea. As to the distribution of other kinds of 

 gall-creatures we are still in many respects in the dark. 



I already pointed out in my earlier article that thèse galls ought not 

 to be used as proofs of the antiquity of a flora, the number of varieties of 

 galls that hâve crossed to thèse islands being so great. In ail I hâve already 

 found on Krakatau and Sebesy together as many as 82 forms of galls. Itis 

 therefore certainly not possible to speak of accidentai circumstances hère. 



Both on Krakatau and on Sebesy the number of gall-mite galls is 

 prépondérant; this is a corollary not only of the degree of ease with which 



') E. F. Felt. Distribution of Gall-Midges. Proceedings of the Nation. Acad. of 

 Science. Vol. 111. 1917. p. 349. 



2) A. Ernst. Die neue Flora der Vulkaninsel Krakatau. Vierteljahrschr. der 

 Naturw. Gcsellsch. in Ziirich. Jalirg. 52. 1907. 



