INHERITANCE 187 



the range of variability in F.2 should extend to or beyond the 

 limits of the two parental ranges. This has been shown to 

 be true in cases w^orked out by East and Jones. 



We have next to inquire how the transmission of one gene 

 reacts upon that of another ; and what results occur when two 

 or more pairs of genes are involved in a cross. 



Independent Assortment.— In the wild form of the banana fly, 

 Drosophila, which has been the material of a considerable volume 

 of research by Morgan and his school, the body is gray and 

 the wings extend beyond the tip of the abdomen. Two true- 

 breeding mutants have appeared in Morgan's cultures respec- 

 tively distinguished by the shade of body colour known as ehony 

 and by a vestigial condition of the wings. Both are recessives 

 to the wild condition. On crossing an ebony fly with long- 

 wings with a gray fly with vestigial wings, all the F.i are of the 

 gray-long type ; and the F.2 the four combinations : gray- 

 long, ebony-long, gray- vestigial, and ebony-vestigial in the 

 proportions 9:3:3:1. On the assumption that the pair 

 of genes responsible for the ebony and gray characters, on the 

 one hand, and the long and vestigial characters on the other 

 are transmitted quite independently, there is a 3 : i chance 

 of any individual having either dominant character in the F.2. 

 The probability of an individual having both dominant 

 characters is (|)2, that of it having one dominant but not the 

 other and vice versa is | x J, and that of having neither dominant 

 character (J)^. This gives the 9:3:3:1 ratio and proves that 

 the assumption is correct. This is further borne out by the fact 

 that identical results follow the mating of an individual of ebony 

 colour and vestigial wings (double recessive) with the wild type. 



Linkage. — This independent assortment of separate pairs 

 of genes is very com^mon in all organisms investigated. If, 

 how^ever, separate pairs of genes always segregated in this 

 way, we should be compelled to postulate an indefinite num.ber 

 of structural units to provide for the material basis of inheritance. 

 As a matter of fact, independent assortment is not a universal 

 rule. Association of genes belonging to different allelo- 

 morpliic {i.e. segregating) pairs in the process of transmission 

 in contradistinction to the independent assortment illustrated 



