THE INTERACTION OF GENES: THE EFFECTS 



159 



is dependent on the interaction between different loci. Thus in Droso- 

 phila melanogaster there are at least 30 loci whose primary expression 

 is an effect on the eye colour. Similarly in maize well over 50 genes 

 have been described which affea the formation or distribution of 





P1 



F1 

 F2 



F3 



H 

 F5 



O to 



a. 



61 



60 

 80 



UU 

 85 



41 

 90 



Fig. 76. Inheritance by Polynneric Genes. — The table shows the results of a cross 

 between varieties of Nicctiana longiflora differing in corolla length. The corollas 

 are grouped in classes whose mid-values are shown in the upper row of figures; 

 the table gives the number of corollas falling into each class. The varieties are 

 presumed to differ in several factors affecting corolla length. In the F1 these 

 factors would be mainly heterozygous and the corolla length intermediate. In 

 later generations plants are obtained which are homozygous for some of the 

 factors and which therefore have more extreme corollas, either short or long. 

 Owing to the number of factors involved, the pure parental types are not 

 recovered till the F5.^ 



The increased variability of the F2 and later generations as compared with 

 the F1 can only be explained as a result of the segregation of factors, and is usually 

 accepted as sufficient evidence that the character in question is inherited through 

 Mendelian factors even if these cannot be separately identified. 



chlorophyll'^ and in Primula sinensis there are 11 genes for flower 

 colour.^ They are called multiple factors. 



The detailed interaction between such genes has not always been 

 worked out; the magnitude of such a task is obvious in cases like that 

 mentioned in maize. In the simplest case the genes which affect the 

 same character may interact in a purely quantitative way, either adding 

 to or subtracting from the total effect. Genes of this kind are known as 

 polymeric genes, and their presence may often be deduced from breed- 

 ing results, since they give rise to an intermediate Fi, but an F2 in 



^ Data from East 19 10. 

 - Cf. Eyster 1934. 



^ de Winton and Haldane 1932. 



