100 



unfertile localities, frequently on steep slopes but also in crevices of 

 rocks and such-like stony places. In Java, where it is often cultivated 

 for ornamental purposes but is only rarely met with in a wild state, 

 it never grows gregariously but always in scattered specimens. The 

 specimens found in 1906 and 1908 on Krakatao, in localities which 

 bore a dense vegetation as well as in those which bore only a 

 meagre one, grew in the same manner and there is no reason to 

 suppose that in 1897 the reverse would have occurred. On the mode 

 of dispersion of this plant see p. 48. 



In quite another way Saccharum spontaneum L- (Glagah, Kaso) 

 behaves, a most variable species, generally 3 5 m. high, with large 

 (30 100 cm. long) terminal flower-panicles, which during and after the 

 anthesis (in Java especially from February till August; in other months 

 flowering specimens are comparatively scarce) are very conspicuous 1 ) 

 by the long violet or white hairs clothing the bases of the spikelets. 

 This rather deeply rooting, heliophilous grass is spread from West- to 

 East-Java, from the plains up to an altitude of 1800 m. It is not a 

 kremnophyte; though it does not require a fertile soil, it does not 

 grow well in a very poor one. Wherever it occurs in great numbers 

 the soil must be considered sufficiently fertile for divers agricultural 

 plants. It needs rather much water to thrive well 2 ); in regions where 

 the east-monsoon is strong and the soil dries out to a great depth 

 it is restricted to depressions of the ground and to water-sides, but 

 where the east-monsoon, as a rule, is interrupted by repeated rains it 

 also grows in other localities, especially in glades, on margins of 

 forests, in thickets and young secondary forests, on ravine-sides and 

 very often in old alang-fields. The occurrence of this grass in num- 

 berless specimens in the pumice- and alang-fields of Krakatao proves 

 not only that in this island the somewhat deeper layers of the soil as 

 a rule do not dry out very much, but also that the soil is not very 

 poor. Because of the rhizome ramifying very much, creeping far and 

 at short distances emitting new culms, old specimens of this grass 

 finally form large and very dense bushes. In localities where the 

 plant thrives particularly well these bushes may blend into vast jungles 

 which can be penetrated into only with the aid of a chopping-knife. 



!) See Ernst, Neue Flora Vulkaninsel Krakatau (1907), tab. VI, fig. 10. 



2) Docters van Leeuwen | Handelinqen Eerste Ned.-lnd. Natuurw. Congres 

 (1920), p. 49; Ann. |ard. Bot. Buitenzorg XXXI (1921), p. 120] wrongly calls this plant a 

 grass of dry soils. )ust the reverse is true. One should not judge of the humidity of 

 a soil by casting a look on its surface. See also what the excellent observer lunghuhn 

 (Reisen durch lava, p. 210, 343) writes on the Saccharuni-'\unq\e on Mt. Galunggung 

 and on the occurrence of Saccharum at the lake of Ngebel 



