103 



,,without trouble, because the glagah (Saccharum) is not so high and 

 ,,by far not so dense. Between the canes (i.e. Saccharum} a terres- 

 trial orchid with long reed-like stems and white flowers is fre- 

 quent". This terrestrial orchid was Arundina speciosa BL, recorded 

 already on p. 31 among the kremnophytes. Localities where such 

 kremnophytes abound, are unfit for a luxuriant growth of Saccharum^}. 

 In the parts of Krakatao investigated by P e n z ig in 1897 this grass 

 apparently was locally quite common, as may be concluded from 

 3 of the photo's of the interior, reproduced in his paper 2 ). On 

 each of these Saccharum is recognizable by its inflorescences. 



The third grass named by Penzig, Neyraudia, is a robust 

 strongly tillering, arundinaceous plant, reaching a height of 2 5 m. 

 and deceivingly resembling a Phragmites. With the latter genus it 

 has often been confounded, but besides by its villous glumae it can 

 easily be distinguished by its massive culms. Old specimens often 

 form clumps of some square meters' surface but I did never see in 

 Java these clumps blend into a dense vegetation of very consider- 

 able extent. As to its habitat this species stands between the two 

 former; it thrives quite well in a rather unfertile soil, it is often 

 found in stony localities '), frequently even in great numbers though 

 never covering large areas. But such unfavorable habitats as Pen- 

 nisetum is contented with, narrow, rather undeep crevices in rocks 

 for instance , seem, however, to be unfit for the growth of Neyraudia, 



Penzig keeps silent about the individual frequency of the 

 three species. He only says that between the grasses which formed 

 in the valleys a nearly uninterrupted jungle, other plants sporadically 

 occurred, so that the most suitable name for the vegetation in 

 question was savanna or grass-steppe. In the rather humid parts with 

 a deep soil Saccharum has probably predominated, forming there a 

 rather dense, locally perhaps very dense vegetation, whilst in the rocky, 

 less moist localities, besides a thin growth of low -Sacc/ian/m-plants, 

 Neyraudia, Pennisetum and other plants than grasses will have been 

 found. Penzig says that, when he penetrated into the grass- 

 jungles of the interior, numerous twining plants with their tough stems 



') But it should not be forgotten that, within a small space, considerable local 

 differences may occur. Ravines may quite well bear a kremnophytic vegetation on the 

 almost vertical parts of their walls and a SaccAart/m-jungle on the more inclined parts 

 and near their bottom- One should also bear in mind that localities unfavorable for a 

 luxuriant growth of Saccharum, but adjacent to favorable localities, may be constantly 

 populated from the latter by means of the creeping rhizomes- In judging of a vegetation 

 in the tropics one should always have a keen eye to its surroundings- 



2 ) Ann. |ard. Bot- Buitenzorg XVIII (1902), pp. 101, 103, 104. 



3 ) See also p. 53, footnote 2. 



