1 84 COMPARATIVE PHYSIOLOGY 



winged (referred to henceforth as " longs " for the sake of 

 brevity) flies of the F.i thus differ from the parental longs 

 in producing gametes some of which are characterised by the 

 possession of a material something — let us call it 2igene, with- 

 out discussing its nature — which leads to the production of the 

 vestigial condition of the wings. If we denote the gene which 

 determines the long-winged condition by the symbol V and 

 the gene which determines the vestigial condition by the 

 symbol v, we may refer to the F.i longs as Yv, to denote that 

 they form gametes bearing both V and v. By analogy the 

 parental long and vestigial types, which since they breed true 

 may be regarded as forming one type of gamete only, may be 

 denoted by the symbols VV and vv. Let us proceed to examine 

 the progeny (F.2) of the F.i longs mated inter se. One quarter 

 are vestigial indistinguishable from the original vestigial, 

 breeding true to type w^hen mated with their like or with 

 the original vestigial type. They may therefore be denoted 

 by the symbol vv as before. The remaining three-quarters 

 are longs. They do not all behave in the same way on crossing. 

 If they are individually crossed back to the original vestigial 

 stock, one-third of the F.2 longs produce, like the original 

 longs, only long- winged offspring ; and individuals which 

 behave in the back-cross in this way, when mated inter se 

 breed true to type ; they may therefore be denoted by the 

 symbol VV. The remainder when back-crossed to vestigial 

 give offspring half of which are longs and half vestigials. These 

 impure F.2 longs when mated inter se behave like the F.i longs, 

 giving a 3 : I ratio of long to vestigial. Thus they may again 

 be denoted by the symbol V^^. The constitution of the F.2 

 is therefore i VV : 2 Vz; : i z;?^. Now if we make a very simple 

 assumption about the distribution of the genes V and v in 

 the formation of the gametes the quantitative relations of all 

 these crosses fall into line. Let us suppose that on the average 

 the gametes produced by an individual consist of equal numbers 

 containing the gene derived from one or the other parent. The 

 F.I long receives from its vestigial parent the gene v, and from 

 its long parent the gene V : one-half of the gametes it produces 

 carry V and the other half v. V may fertilise V or v. Similarly 



