| 
ee ee 
xi,¢,4 ° Brown, Merrill and Yates: Volcano Island 205 
distributed. In view of the fact that such a small proportion of 
the species that have invaded Volcano Island have become com-. 
mon and wide spread, it is not surprising that no phanerogams 
became prominent on Krakatau during the first three years 
after the eruption, when we consider the fact that Krakatau is 
much farther from the mainland than Volcano Island and so 
was invaded by a much smaller number of species. © 
In 1887, or fourteen years after the eruption of Krakatau, the 
island was visited by Penzig,?> who found that the vegetation 
of the inland consisted of a kind of grass-steppe in which the 
grasses sometimes reached the height of a man and in several 
places formed a thick jungle. Trees were very scarce. Small 
grasses, ferns, and a few seed plants grow on the hills and ridges. 
The vegetation of the rock surfaces consisted largely of ferns 
and showed little change from the conditions observed by Treub 
in 1886. 
The prominence of grasses and scarcity of trees is similar to 
_ the condition observed on Volcano Island. Grasses were much 
slower in becoming prominent on Krakatau than on Volcano 
Island; but it may be that they would have been much more 
prominent on Krakatau in-the early stages, if their seeds had 
been transported to that island. 
The essential differences between the revegetation of Volcano 
Island and Krakatau seem to be connected with the fact that 
Krakatau is situated in salt water and, therefore, has developed 
a strand formation which is lacking on Taal; while Taal being: 
much nearer a large land mass has been invaded by many more 
species than Krakatau. These points will be considered later. 
z ENVIRONMENTAL CONDITIONS 
The slowness of the revegetation of Volcano Island is prob- 
ably not due, to any great extent, to the aérial environment, 
as the surrounding country supports a luxuriant vegetation and 
the indications are that it was originally covered by a tall dip- 
terocarp forest. The unfavorable factors are apparently con- 
nected with the condition of the substratum. The most evident 
of these are erosion and lack of weathering.of the soil particles. 
Most of the steeper slopes are composed of soft loose material 
which is very readily eroded. An extreme case of erosion is 
seen in Plate XVI, fig. 2, which shows the outer slopes of the 
* Penzig, O., Die Forstschritte der Flora des Krakatau, Ann. Jard. Bot. 
Buitenzorg II 3 (1902) 92-113. 
