78 



The two instances selected by Treub in corroboration of his 

 argument were badly chosen. The group Juan Fernandez 1 ) consists, 

 besides of some small islands, of two larger ones, Mas-A-Tierra and 

 Mas-A-Fuera, respectively 75 and 59 km-, in size. Mas-A-Tierra, the 

 highest top of which reaches about 1 000 m., has a mean annual tem- 

 perature -) of 1 5 C and a considerable rainfall, averaging 1004 mm. 

 a year, distributed as follows: 



Jan. Febr. March April May June July Aug. Sept. Oct. Nov. Dec. 

 31 40 60 98 144 147 160 124 72 58 43 27. 



Rainfall and temperature are sufficient for a rich forest-growth. 

 The island, famous for its splendidly coloured flowers, is partly covered 

 with rain-forest, though the most valuable indigenous trees have been 

 exterminated. Epiphytical ferns (Polypodiaceae and Hymenoptiyl/aceae) 

 are plentiful in the rain-forest. Up till now 42 species of ferns y ) have 

 been found in Mas-A-Tierra, 51 in the entire group. There is only 

 one Phanerogamic epiphyte, but numerous trees and shrubs, leading 

 as a rule a terrestrial life, occur as accidental epiphytes in very 

 moist localities. The undergrowth in the forest consists for by far 

 the greater part of ferns and cell-cryptogams. About one third of 

 the flora is endemic. 



It is by no means proven that this abundance of ferns and scarcity 

 of Phanerogams in the undergrowth of the forest is to be ascribed 

 so/f/v to seeds of Phanerogams not being introduced by cause of the 

 isolation of the island. In every not too young rain-forest in Java one may 

 observe that introduced plants either are entirely deficient or occur 

 only in glades, clearings and along broad roads receiving a somewhat 

 large quantity of light, but that they are unable to penetrate into 

 the forest, though at a distance of a few meters numerous fructiferous 

 specimens of these introduced plants may be found. The forest near 

 Tjibodas on Mount Gede strikingly illustrates this fact. Along the 

 path to the cataracts many introduced plants occur along the forest- 

 borders; in the forest proper adjacent to it not a single one. 



The conditions of growth presented by Ascension ( 90 km'-'.) 

 differ greatly from those on Juan Fernandez. From the description 

 given by Treub of the soil of Ascension may be concluded that 



l ) Cf. Schimper, Pflrmzengeographie 2. Aufl. (1008) pp. '142, 520; Encyclopaedia 

 Kritannica, 11tli Ed. XV (1911). p. ">29; Winkler Prins, Geillustreerde Encyclopaedic 

 X (1018), p. t>t>l. 



'*) Data on the climate received by courtesy of Dr. Visser, subdirector of the 

 Royal Magn. Meteor. Observ. Batavia- 



3 ) Carl Christensen and Carl Skottsberg, The Pteridophyta of the 

 |uan Fernandez iblands, p. 45. 



