II] GEOGRAPHICAL DISTRIBUTION 33 



parent types. This, however, has not been proved to 

 be the case. Darwin in speaking of Forbes' Essay in 

 a letter to de CandoUe in 1863 says that he differs 

 from most of his contemporaries 'in thinking that 

 the vast continental extensions of Forbes, Heer, and 

 others are not only advanced without sufficient evi- 

 dence, but are opposed to much weighty evidence'(i2). 

 The alternative view is to regard Arbutus and its 

 compatriots as post- Glacial arrivals and not as sur- 

 vivals from a widely spread Tertiary flora. 



A recently published account of the New Flora 

 of the volcanic island of Krakatau furnishes an 

 instructive and remarkable demonstration of the 

 facility with which a- completely sterilised island, 

 separated by several miles of ocean from neigh- 

 bouring lands, may be restocked with vegetation (24). 

 In 1883 the island of Krakatau, then densely covered 

 with a luxuriant tropical vegetation, was partially 

 destroyed by a series of exceptionally violent volcanic 

 explosions. After this catastrophe only a third of 

 the island was left : the surface was deeply covered 

 by pumice and volcanic ash and no vestige of life 

 remained. In 1906 a party of botanists who spent 

 a few hours on Krakatau collected 137 species of 

 plants : the vegetation was in places so dense that 

 it was with the greatest difficulty they penetrated 

 beyond the coastal belt, and some of the trees had 

 reached a height of 50 feet. The seeds and fruits 



s. 3 



