INFLUENCE OF ISOLATION ON THE FORMATION OF SPECIES 295 



change of food and other conditions, such as the conscious!}^ or 

 unconsciously exercised processes of selection. It follows from this that, 

 in the gradual diffusion of snails all over the island, similar localities 

 would almost never be colonized by exactly similar immigrants, but 

 by individuals containing a different combination of the existing 

 variations, so that in the course of time different conda id forms vrould 

 be evolved through amixia in relatively isolated localities. 



But everything would be different in the diffusion of a new 

 species of snail in a region which was already fully or at least 

 abundantly occupied by snail-species. Let us leave out of account 

 altogether the first factor in variation, the changed climate, and we 

 see that a species in such circumstances would have no cause for 

 variation, because it would find no area unoccupied outside of the 

 sphere to which it was best adapted ; it would therefore not be 

 impelled to adapt itself to any other, and in most cases could not do 

 so, because in each it would have to compete with another species 

 superior to it because already adapted. 



The case would be the same if an island were suddenly peoj^led 

 with the whole snail-fauna of a neighbouring continent, with which 

 a land connexion had arisen. If the island had previously been free 

 from snails, all the species of the mainland would be able to exist 

 there in so far as they were able to find suitable conditions of life, 

 but each species would speedily take complete possession of the area 

 peculiarly suited to it, so that none of their fellow migrants would 

 be impelled, or would even find it possible, to adapt themselves to 

 new conditions and thus to become variable and split up into varieties. 

 If Ireland were at present free from snails, and if a land connexion 

 between it and England came about, then the snail-fauna of England 

 would probably migrate quite unvaried to Ireland, and in point of 

 fact the snail-fauna of the two islands, which were formerly con- 

 nected, is almost the same. For the same reason the fauna of 

 England, as far as terrestrial snails are concerned, is almost the same 

 as that of Germany. 



On the other hand, it may be almost regarded as a law that an 

 individual migrant to virgin territory must become variable. This 

 could not be better illustrated than by the geographical distribution 

 of terrestrial snails, which emphasizes the fact that a striking wealth 

 of endemic species is to be found on all oceanic islands. Moreover, the 

 fact that the number of these endemic species is greater in proportion 

 to the distance of the island from the continent, indicates that the 

 variability sets in more intensively and lasts longer in proportion to 

 the small number of species which become immigrants in the island, 



