116 
Penzic already distinguished the formations of beachplants, 
the Barringtonia-formation and that of the grassy wilds. The 
latter consisted mainly of the well-known jungle-grasses, glagah 
- (Saccharum spontaneum L.) and alang-alang (Imperata arundi- 
nacea Cyr.), together with ferns, various orchids and small 
clumps of trees. 
When in 1906 Ernst, in company with Backer, Campsetn and 
Putiz, arrived on the island this process of the formation of 
plant-communities had again made further progress; also the 
number of plants had vastly increased. Whereas Trevs found 
in all 26 species of vascular-plants, and Prnzic mentions 65 
in his list, that of Ernst comprising the plants met with in 
three explorations is enlarged to 108 forms. The majority of 
these plants, just like what was perceived in the Pxrnzic-explo- 
ration, again belonged to the common beach-plants whose seeds 
are conveyed by ocean-currents. 
The list made out by Ernst also contains under plants found 
near the heach, a number of names of species generally not 
met with in those places, but after some time when both for 
mations, the beach-community and the Barringtonia-belt shall 
have expanded, those others will be pushed more inwards or 
else become extinct. I myself also found plants in the same 
case, but partly others than those referred to by Ernsr. The 
additions to the flora observed during the Ernst-exploration 
consisted mainly of the increase in the number of Phanerogams. 
Contrarily the Cryptogams, among them the ferns, had not 
greatly increased. The grassy wilds still occurred, but the iso- 
— lated clusters of trees were already more numerous than at the 
moment of Penzie’s visit. In the ravines were already found a— 
oe few forest-trees but the ascent of the mountain being a failure, 
they did not know what things looked like higher-up. 
oe ‘Two years later Mr. Backer visited the island again, nor did 
e he reach the top, but got about halfway-up. He already pointed 
out: that i in contrast to what is observed in Java the forest was 
| by the predominance of a small number of spe- 
no cause. for. wonder, when one considers that: — 
a3 irst, and ae Bee 3 
