104 REDUCTION OF VARIABILITY. 



seen by subjecting all the members of a family of albino rats 

 to a suitable test, namely to a cross with pigmented animals 

 lacking the gene in question, such as blacks, or pink-eyed 

 creams. A gene may influence two different developmental pro- 

 cesses in the life of the same individual, and if one of the two 

 processes is so changed that it does not any more come under 

 the influence of our gene, the fact that the gene itself is not 

 latent or dormant, is seen from the fact that its action on the 

 other process remains unimpaired. 



If we should ever be called upon to give as complete a list as 

 possible of the genes for which a given family of unpigmented 

 rats is impure, we should have to make suitable test-matings, 

 to decide whether or not the individuals were pure for such 

 genes as did not happen to show their presence. 



We propose throughout this bjook to use the term Total 

 Potential Variahility for the number of the genes in respect to 

 which an individual or any group of individuals is not pure, 

 homozygous. In doing this we place ourselves upon the stand- 

 point, that in inheritable variation we are concerned with the 

 influence of the genes exclusively, and that the individual 

 genes are qualitatively stable (Law of Johannsen) so that 

 variabilis, and potential variabiUty becomes synonymous 

 with genotypic impurity. 



In practice, it is obviously impossible to determine the exact 

 number of genes for which an individual is not pure, or for 

 which not all the individuals of a group are pure, and therefore 

 to put down the exact number, expressing the total potential 

 variabihty. But even so, this number is a definite one, even if 

 we do not know its magnitude, something we can work with 

 in certain ways. We can investigate which processes make the 

 number larger, and which things reduce it, and in what 

 measure they do reduce it. 



In our investigation of the methods of evolution, we see at 

 once that we are evers^where concerned with two mutually 

 opposed tendencies, principles. Darwin called these variation 

 and heredity. 



