ISOLATION 193 



plants. The higher animals at any rate, are capable of exercising 

 choice, and when differences of colour, for instance, show them- 

 selves all those of one tint will sometimes shun those of another. 

 This has been the case with the cattle that have run wild in the 

 Falkland Islands. They have separated into herds of different 

 colours, which do not intermix. Yet who can doubt that if 

 members of the different herds were domesticated they would 

 prove fertile together ? Among animals, then, it is probable 

 that varieties that were not isolated by intersterility have main- 

 tained themselves and gained the rank of species. Among plants 

 there is no such predeliction, and, among them therefore, it is 

 more probable that intersterility has come into play. 



To this view it has been objected that there must be varying Varying 

 individuals before definite varieties are formed, and that we can mdlvldua s 

 hardly assume intersterility between such individuals and other 

 members of the species. Let us grant this for the sake of 

 argument. We have then only to assume that there was at the 

 outset some geographical isolation. A seed was carried, say, to 

 the other side of a broad estuary, or across a watershed, and so 

 the new variety had time to take definite form before insects 

 appeared bearing pollen from its kin of the main stock. When 

 once intersterility had been established, then geographical isola- 

 tion might cease without producing confusion. But it is quite 

 possible that sterility with the parent species may often have 

 arisen simultaneously with the favourable variation that charac- 

 terised the new variety. 



If it be granted that such intersterility has not uncommonly insect 

 protected new variations, then it can no longer be denied that ^ r j^[, s ese 

 the conditions have been favourable for the work of insects, conditions 

 Given isolation, thus assured, it has been possible for them, by 

 perpetually rewarding brilliancy, gradually to make the earth 

 gay with flowers. Each variation in the direction of increased 

 brightness or better adaptation in shape to insect wants was, and 

 is, likely 'to survive, from the fact of its bringing with it a better 

 chance of cross-fertilisation. The danger of the swamping of 

 new forms through the wandering of bees from one variety to 

 another has been obviated by the intersterility which so generally 



N 



