62 B f-.'ii mmm^ 0) mm :^ ?.'i 



Gametic Eepolsion. 



The case in which more sufScient data have been obtained up to the present 

 is that of the uormal-yollow ropulsion. Relating to the reduplication under 

 consideration, over 70 families involving more thaa 20,000 individuals were 

 reared, certain families of them bning carried up to F^ generation. The hetero- 

 zygotes (AaBb) were, in the course of experiments, also crossed with the double 

 recessives (aabb) reciprocally. The gametic series thus revealed were 



? 1:1 



The recombination of these gametic forms will result in the zygotic series 



A-B- A-b- a-B- a-b- 

 2:1:1:0 

 So far as the experiments have been carried on up to F^ in the straight 

 direction, the case is hardly distinguishable from an ordinary case of complete 

 repulsion that occurs alike on both sexes. In the present case, however, F. A-B- 

 form is not even in its zygotic constitution but is composed of various biotypes 

 which may give the offspring in various combinations of F3 forms — i.e. the result- 

 ing F3 families can be grouped into four classes as follows : 



Class (a) involving 3 zygotic forms, A-B-, A-b-, a-B- ; 

 „ (b) „2 „ , A-B-, A-b- ; 



„ (c) „2 „ , A-B-, a-E- ; 



„ (d) „ only A-B- form. 



Fo A-b- and a-B- individuals are each composed of two gi'oups, one breed- 

 ing true to their parents, while the other producing aabb animals in addition to 

 those of the parental type. 



All this is beyond explanation if the ordinary complete repulsion be supposed 

 as the case, but it can easily be accounted for if the gametic series above given 

 was assumed. More positive evidence for such dimorphism of gametic series 

 has been actually obtained from the crosses between the heterozygote and the 

 double recessive. 



