136 I The Process of Evolution 



more viable than either homozygote, so that physiological selection 

 would tend to establish a stable polymorphism. These nonvisual 

 selective forces, interacting with the selection pressures created by 

 visual predation, seem to be responsible, in large part, for the ob- 

 served pattern of variation in Cepaea. 



Several other factors may be of importance in some situations. One 

 is predation in which the selection pressure is a function of the rela- 

 tive frequency of the type of individual predated. Certain predators 

 may form search patterns that result in selection against the com- 

 monest type in the population, without regard for which type is 

 commonest. This sort of predation pattern could lead to a stable 

 polymorphism. Random processes, once considered to be the prime 

 factor in differentiating Cepaea populations, may have relatively 

 minor importance. Undoubtedly drift plays some role in the smaller 

 populations, and it may account for some patterns of variation 

 recorded by Lamotte in France. Recently established colonies may 

 not have achieved equilibrium with their environment, and their 

 composition may be strongly influenced by the genetic information 

 possessed by the snail or snails that established them. This influence 

 of the genetic endowment of the individuals involved in starting a 

 new colony is known as the "founder principle" and is discussed 

 further in Chap. 10. 



One of the arguments used to demonstrate that visual selection 

 does not play an important role in determining the characteristics of 

 Cepaea colonies is that, in mixed colonies of the two closely related 

 species C. nemoralis and C. hortensis, the phenotype frequencies of 

 the two snails were uncorrelated. Clarke has satisfactorily explained 

 this by showing that both species respond to visual selection pressure 

 but in difi^erent ways. It appears likely that the selective values of 

 various pattern and color genes differ in the different genotypes of 

 the two species. As a hypothetical example, a strongly banded pat- 

 tern might be at a selective disadvantage in a certain dry habitat. 

 In one species the genes for strong banding may also be involved 

 in producing individuals resistant to desiccation. In the other species 

 there may be no such system. Thus in the same habitat one species 

 may have a population with a large proportion of strongly banded 

 snails (because of heavy selection for resistance to desiccation), 

 whereas the other species may have no banded snails at all ( because 

 of the selection against banding). Although the genes controlling 

 the various factors seem to be homologous, they cannot be divorced 

 from their genetic environment. It will be recalled that in Chap. 6 

 genes were treated as if they were independent entities; results such 

 as these studies of Cepaea constantly remind us that this is an over- 

 simplification. 



I 



