SELECTION. 181 



for the action of natural selection in the moulding of the type 

 of species. 



To a student of evolution it is apparent how domestic spe- 

 cies, dogs, poultry, sugar-beets, change under the action of 

 selection, and at first sight it seems logical to assume that 

 selection must gradually modify wild species in approximately 

 the same way. The essential thing, however, which distinguish- 

 es domestic species from wild ones, is the existence in wild 

 species of great blocks of individuals of one type, whereas in 

 domestic species, the group of individuals chosen to be the 

 parents of the next generation is always a very limited one, of 

 which each member is carefully chosen according to one stand- 

 ard. These few selected individuals are not left as a minority 

 in a mixed population, but they are themselves isolated in a 

 group in which they constitute the type, the ma ority. 



When a relatively small group of individuals of one species 

 splits off, such as is constantly happening, where seeds and 

 young animals wander into surroundings where somewhat 

 earlier the species had no representatives, the future of this 

 group as a possible new species depends upon several things. 

 One of these is the genotype of the inidividuals; if the group 

 is really representative of the old species, and consists wholly, 

 or almost wholly, of normal, average individuals, it will never 

 be able to develop a type of its own. Another important factor 

 is isolation. Given a sufficient potential variation, the new group 

 in getting pure, may become so for a somewhat different set of 

 genes than that of the average individual of the old species, but 

 this can only happen when for some reason, inter-crossing of its 

 individuals with members of the old species is relatively rare. 

 Geographic isolation is of an obvious kind, but isolation can 

 result from the very constitution of the group, as well as from 

 its situation. Any group, sufficiently isolated, must lose its 

 total potential variability, which means that it must eventually 

 become pure for its own type. And it is apparent, that at this 

 very point natural selection must make felt its influence. Else- 

 where we have calculated how great the chances are for the 



