318 LEROY POWERS 



otropy produced the relations noted by the above authors between the four 

 series of genes F/, Ss, Rr, and Lclc with the exception of the Ff and Ss re- 

 lation, because all the associations noted are those expected on the basis of 

 linkage. If pleiotropy were involved, such relations would be coincidental, 

 which for all these gene series is highly improbable. However, as pointed out 

 by Powers, Locke, and Garrett (1950) some of the genes of the Ff and Ss 

 series must be identical, as percentage of flowers that set fruit has an effect 

 on period from first bloom to first fruit set. The Lclc and Ww series of genes, 

 differentiating number of locules and weight per locule, respectively, were 

 independent as regards linkage and pleiotropy. In these studies pleiotropy 

 was not of major importance. 



Phenotypic and genie dominance furnish some information concerning 

 the interallelic and intraallelic interactions of the genes. That genie domi- 

 nance is dependent upon the genotypic milieu was pointed out by Fisher 

 (1931) and many others (Dobzhansky, 1941). Hence both interallelic and 

 intraallelic interactions as measured by end products are second order inter- 

 actions, genes X genes X the environment. 



Any of the interactions of genes noted as affecting any of the component 

 characters dealt with in the study by Powers, Locke, and Garrett (1950) 

 were interactions of genes differentiating yield of ripe fruit per plant. With 

 this fact in mind, it is interesting to note the interactions of the genes differ- 

 entiating the component characters. The intraallelic and interallelic interac- 

 tions of the Ff gene series were such that genie dominance was intermediate. 

 The intraallelic and interallelic interactions of the Bb series of genes were 

 such that one of the six dominant genes shortened the period from seeding to 

 first bloom as much as all six, which shows that both dominance and epistasis 

 were complete. For the 5^ series and Rr series of genes, genie dominance was 

 complete. Also, the effects of the gene pairs were cumulative. 



Had the dominant genes of the Ss series entered the cross from one parent 

 and the dominant genes from the Rr series entered the cross from the other 

 parent, the Fi hybrid would have shown heterosis for earliness of maturity. 

 Porter would then represent an inbred line which by recombination of 

 genes retained the earliness of maturity of the Fi hybrid. Genie dominance 

 was partial for genes (LciLc^) tending to produce fewer locules per fruit and 

 for the (Lcs) tending to produce more locules per fruit. A series of genes such 

 as Lci and Lc2, some entering the cross from one parent and some from the 

 other, would produce an Fi hybrid showing heterosis for fewer locules per 

 fruit. On the other hand a series of genes such as (Lcs), some entering the 

 cross from one parent and some from the other, would produce heterosis for 

 more locules per fruit. 



Finally, for the Wiv series of genes, genie dominance was partial for in- 

 creased weight per locule and the effects of the gene pairs were cumulative. 

 As regards this character, both parents did contribute genes for increased 



