Studies on variation and selection. 



177 



such genes as happen to influence the same character, as this way 

 of writing the formulae might lead some people to think that these 

 genes somehow were of the same kind. 



In the practice of animal-breeding, it is found every day, that 

 nearly all characters of any importance, and therefore subjected to 

 selection, are influenced by several genetic factors for which the 

 population is not pure. It can not be too often emphasized that it 

 is only for this reason, that such characters can be modified by 

 selection. The greater the number of offspring in each generation, 

 the more strictly inbred the family is, and the more severe the 

 selection, the quicker will its effect be felt. The quicker also will 

 the limit be reached, to which the character modified can be bred. 



Which is the limit, to which any given character, or any organ, 

 can be brought ? Obviously, there is, for a great many things, a 

 physiological limit. This is clearly the case in such instances as 

 that of the proportionally already too small head of certain tumber- 

 pigeons. But it must not be thought, that in many cases we will 

 come anywhere near this physiological limit. For in all cases, there 

 is a limit, given by the most favourable conditions, together with 

 the, for the development of the organ or character, most favourable 

 combination of genes realisable. Genetic developmental factors, which 

 tend to modify the quality in question in the desired way, can be 

 combined only in so far, as by crossing they can be brought into 

 the material, and this can further be bred pure for all such genes. 

 On the other hand, genetic developmental factors, which tend to an 

 influence in the opposite direction, can be eliminated in as far as the 

 material is already impure for them, or can be crossed to organisms, 

 lacking these genes. 



It will be clear, that the limits thus imposed upon different 

 qualities must necessarily often fall short of the physiological limits. 

 Thus, for example, there are clearly no physiological limits which 

 restrict the range of colours in Mus norvegicus, and keep it far nar- 

 rower than that in Mus rattus, or in Mus musculus; only, in the 

 brown rat all individuals are still pure for many genes which are 

 absent from some individuals of the other species. The range of 

 colours may become as extensive as in the other two forms, if these 

 genes happen to get lost in some individuals, either spontaneously, 

 as such genes have been found to get lost occasionally, or by mating 

 to individuals of the other species. No amount of selection alone 

 ■could do it. 



Induktive Ahsiammungs- und Vererbungslehre. XI. 12 



