216 PHYSIOLOGIC GENETICS 



SUMMARY 



The two properties of primary interest in the genetics of any quantitative character 

 are the degree of genetic determination and the heritability. 



The degree of genetic determination is the proportion of the total phenotypic 

 variance that is attributable to genetic differences between individuals. It is estimated 

 from a comparison of the variances of a genetically uniform group, such as an F x 

 of two inbred lines, and a genetically heterogeneous group, such as an F 2 or a random 

 bred strain. The number of animals that must be measured in order to attain a given 

 degree of precision in the estimate can be determined from formulae and graphs 

 presented. 



The heritability is the proportion of the total phenotypic variance that is attribu- 

 table to additive genetic variance. It is estimated from the degree of resemblance 

 between relatives. The regression of offspring on parents is usually the best method for 

 laboratory mammals. The design of the experiment to give maximal precision is 

 discussed. In general it is best to have as many parents as possible at the cost of fewer 

 offspring per parent. The method of computing the regression when the families 

 contain unequal numbers of offspring is explained. 



The correlation of half sibs is another method of estimating the heritability. 

 The most efficient design is very inconvenient with laboratory mammals, and as a 

 compromise it is suggested that as many males as possible should each be mated to 

 3 or 4 females and between 5 and 10 offspring from each female should be measured. 

 The structure of the sib analysis with full and half sibs is described ; and the method 

 of computing the components of variance, and from them the correlations when 

 there are unequal subclass numbers, is explained. 



The genetic information that can be obtained from artificial selection and from 

 inbreeding or crossing inbred lines is briefly explained. Finally, the estimation of 

 the number of genes contributing to the variation is discussed. Only in very restricted 

 circumstances does the estimate have any useful meaning. 



