HYBRID VIGOR IN DROSOPHILA 493 



ment approaches what a})pears to be the likely loci number, the better this 

 method agrees with the heterosis which is observed when the chromosome is 

 made heterozygous. The evidence favors several to many gene pairs per chro- 

 mosome as necessary for the heterotic effects. 



Heterotic effects of parental genomes as shown by a series of Fi hybrids 

 were analyzed. For the individual the most significant contribution to varia- 

 tion was that due to a large number of unanalyzed causes. This component 

 contributed over half of the total variation. Differences due to races contrib- 

 uted 15 per cent, w'hile those due to race-experiment and age interactions, 13 

 per cent. The interaction term shows that genes in quantitative inheritance 

 are not stable in their effects. In one condition the genes could react as reces- 

 sives; in another as dominants; in a third, show epistacy. 



The dependence of yield on the interrelation of environment and genotype 

 is of even greater importance. The model customarily chosen to represent 

 genetic and environmental effects ordinarily considers the interactions of 

 these terms zero when in truth they may be quite large. The data on both 

 Drosophila and corn general and specific combining abilities of inbred lines 

 show these interactions to be of major importance. Further progress in the 

 utilization of heterosis appears to lie in the adjustment of the hybrid genotype 

 to the environment. 



