METHODS OF BREEDING 579 



is inversely proportional to the number of factors concerned in the 

 selection, for the greater the number of factors the slower is the rate at 

 which the population approaches a uniformly homozygous condition. 

 Theoretically complete attainment of this condition is only reached after 

 an infinite number of generations, but practically the number of genera- 

 tions which is necessary to measure up to within 5 per cent, of the 

 possible limit is much smaller. It is, however, often so large that the 

 animal breeder would prefer to use some other method, if by so doing, 

 he could more quickly reach the desired standard of excellence and 

 stability of type. 



At this point, however, it should be mentioned that selection is often 

 made for characters which are recessive, or which give intermediates 

 when in the heterozygous condition. In such cases, of course, the 

 relation between phenotype and genotype is simpler and methods of 

 selection gain in effectiveness in consequence thereof. 



Phenotypic Selection Does Not Make Allowance for the Differences 

 Which May Exist Among the Genotypes of a Given Phenotype. 

 Simple examples of this proposition may be quoted without number. 

 In fowls for example there are dominant whites like the White Leghorn 

 and recessive whites like the White Plymouth Rock. The diverse progeny 

 which is obtained by mating these two breeds together has been described 

 in detail in a previous section. There is some evidence that a similar 

 condition may obtain in cattle with respect to white coat color. White 

 is, likewise, dominant in the horse, and may therefore conceal a large 

 number of latent factors. In the pig the same differences in behavior 

 with respect to white coat color have been noted. There is reason to 

 believe that the same kind of diversity in genetic constitution obtains 

 for economic characters, as for those not so strictly utilitarian. The 

 breeder who follows a method of phenotypic selection should not, therefore, 

 be surprised if crossing different strains results in a disappointing lack of 

 uniformity in his herd. It is not difficult to see that in differences of 

 genotype such as have been noted here, the breeder of best to best meets 

 one of his most perplexing problems. 



Phenotypic Selection Fails to Allow for Heterozygosis. In other 

 portions of this book the assumed effect of heterozygosis on vigor and 

 fertility has already been discussed at considerable length. If a hetero- 

 zygous condition ever can determine a more vigorous development than 

 the homozygous condition, then the breeding practice of the future will 

 sometimes be materially altered in order to take advantage of this fact. 

 But aside from this possible difficulty there is sometimes a very real diffi- 

 culty in the fact that selection has set as its standard a type absolutely 

 conditioned by a heterozygous genotype. The striking and ever-quoted 

 instance of this fact is the Blue Andalusian fowl, which no amount of 



