96 INBEEEDING AIJD OUTBEEEDINO 



It will be seen from these figures that the proportion of 

 complete homozygotes and complete heterozygotes is in- 

 creased by linkage at the expense of the mixed class. The 

 proportion either of homozygous or of heterozygous fac- 

 tor pairs, however, is unaffected. It is evident then that 

 just as the reduction to homozygosity by self-fertilization 

 is independent of the number of factors involved, in the 

 same way it is independent of the way iu which these 

 factors are linked together: but in an experiment where 

 particular individuals are chosen as progenitors linkage 

 of factors reduces the chance that these will come from 

 the median classes of heterozygosity; hence, the rate at 

 which homozygosity is attained wUl vary more widely 

 between dfferent lines if the factors involved are par- 

 tially linked than if they are all iadependent. This merely 

 means that some lines will become uniform and lose the 

 stimulus of hybridization in, a fewer number of genera- 

 tions than will other lines and that this difference theoreti- 

 cally is increased by linkage. Bnt the hastening of the 

 attainment of homozygosity in some lines is balanced by 

 delay in other lines, so that on the average the curve 

 of inbreeding shown applies equally whether linkage of 

 factors is involved or not. 



If there were no other controlling factors the reduc- 

 tion in vigor resulting from inbreeding, in the majority 

 of cases, should approximate curve 1 in Fig. 24 on the 

 assumption that hybrid vigor or heterosis is associated 

 with heterozygosity. However, it should not be thought 

 that the amount of heterosis is perfectly correlated with 

 the number of heterozygous factors. Some have more 

 of an effect than others, and certain factors, when com- 

 bined together, may have a cumulative effect. Moreover, 



