INFLUENCE OF ISOLATION ON THE FORMATION OF SPECIES 29 



In this way the immigrants to the Sandwich Islands must liave 

 adapted themselves in the course of time to their increasingly 

 specialized habitats, and in doing so have divided up into increasingly 

 numerous forms, varieties, and species, and indeed into several genera. 



But this alone is not sufficient to explain the facts. According 

 to Gulick's valuable researches there live on one little island of the 

 Sandwich group no fewer than 200 species of Achatinellidse, with 

 600-700 varieties! This remarkable splitting up of an immigrant 

 species is regarded by him as a result of the isolation of each 

 individual species and variety, and I do not doubt that this is correct 

 as far as a portion of these forms is concerned, and that isolation 

 plays a certain part in regard to them all. Gulick, who lived a lono- 

 time upon the island, attempts to prove that the habitats of all 

 these nearly related varieties and species are really isolated as far 

 as terrestrial snails are concerned ; that intermingling of the snails 

 of one valley with those of a neighbouring one is excluded, and that 

 the varieties of the species diverge more markedly from one another 

 in proportion as |^their habitats are distant. On the other hand, 

 species of different genera of Achatinellidse often live together on 

 the same area ; but they do not intermingle. 



Although Gulick's statements are worthy of all confidence, and 

 though his conclusions have great value as contributions to the 

 theory of evolution, I do not think that he has exhausted the problem 

 of the causes of this remarkable wealth of forms among the terrestrial 

 snails of oceanic islands. It is not that I doubt the relative and 

 temporary isolation of the snail-colonies at numerous localities in 

 the island of Oahu. But why have we not the same phenomenon 

 in Germany, in England or Ireland? Gulick anticipates this objection 

 by pointing out the peculiar habits of the Oahu snails. Many of 

 the species there are purely arboreal animals, living upon trees and 

 never leaving them, even during the breeding season, or in order 

 to deposit eggs, for they bring forth their young alive. Active 

 migration from forest to forest seems excluded by the fact that on 

 the crests of the mountains there is a less dense forest of different 

 kinds of trees, and dry sunny air, which could not be endured by 

 the species of Achat inella and Bulimella, which love the moist shades 

 of the tropical forests. Active migration over the open grass-land 

 at the mouths of the valleys is also excluded. 



It must be admitted that the isolation of these forest snails 

 in their valleys is for the time being very complete, and that inter- 

 mingling of two colonies which live in neighbouring valleys does 

 not occur by active migration, ivlthin the spctn of one or several 



