105 



in sweet water; Stenoch/aena palustris fiec/o 7 . thrives as well where the 

 water is very slightly brackish as where it is sweet ') If these 4 ferns 

 really have been found near each other, the two first must have 

 been absent in the locality where the 2 others occurred and con- 

 versely. One should always bear in mind that the bottom of a 

 steep-sided ravine may be oecologically very different from its 

 walls; the walls being frequently deficient in humus and very unfer- 

 tile, consequently fit only for kremnophytes; the bottom more humid, 

 often much richer in humus, hence as a rule much more fertile and 

 bearing a quite different vegetation 2 ). In vertical ravine-sides by some 

 cause or other some parts may be dug out, the bottom of which 

 not seldom becomes more or less horizontal and therefore less liable 

 to erosion. This bottom, when not too young, usually bears a non-kremno- 

 phytic flora, quite different from that of the adjacent vertical parts. 

 In the surroundings of Buitenzorg this phenomenon is very conspicuous. 



After having ascended the rocks Penzig and his companions 

 found several Phanerogams between grasses and in fern-associations. 

 Here also the vegetation gave the general impression of a savanna. 

 Grasses either small or large predominated; between them other 

 herbs and shrubs occurred in scattered specimens. The vegetation 

 varied but little, its components were few, a common occurrence 

 with young vegetations in the tropics. 



Two of the grasses of the somewhat higher parts of the mount- 

 ain are recorded by Penzig under a specific name, viz. Imperata 

 cy/indrica P.B. (I. arundinacea Cyr.) and Pogonatherum paniceum Hack. 

 (P. crinitum Trin). These two grasses also differ very much as to habitat. 

 Pogonatherum paniceum is one of the commonest kremnophytes of 

 Java: from the plains up to an altitude of 1700 m. it is spread through- 

 out the island, except in regions where the east-monsoon, as a rule, 

 is extremely long and dry. It grows exclusively on little weathered or 

 much washed-off localities or in habitats oecologically identical with 

 these: on steeps banks, on rocks and old walls, on big stones in rivers, 

 frequently in great numbers. Banks of rivers and of hollow roads are 

 often for a great length densely clothed with it; rarily it occurs in loca- 

 lities which are level or almost so. When the fruits are ripe, the then 

 fragile axis of the inflorescence disarticulates into small fragments 

 to each of which one fructiferous spikelet remains attached. These 



L ) See p. 54, No. 2. 



2) See also C. A- Backer en Dr. F. van Slooten, (avaansche Theeonkruiden 

 (1924), pp. 14 seq. On Krakatao the lower parts of the ravines were shortly after 

 the eruption locally saliferous and consequently very fit to bear halophytes. Cf. p. 50, 

 footnote 3- 



