Dec. 6, 191S Inheritance of Length of Pod in Certain Crosses 



417 



grown under more adverse conditions than were their second-generation 

 parents. 



Table IX compares the parental and progeny pod lengths of families 

 not known to throw short-podded. The averages of the progenies are 

 here less than the parental averages by 3.5 mm. (See above.) 



Table IX. — Comparison of the pod length of the parents and progeny of families nqt 

 known to throw 'short-podded. Third generation. Parents probably or certainly 

 homozygous for E 



F2 plants not certainly known to be homozygous for E. 



t> A black plant throwing velvet. 



To sum up, investigation of the third generation gives evidence that 

 E is completely dominant; that its multiplying value is 1.5 (one family 

 being an exception); that the genetic range of pod length was fully 

 developed in the second generation; that the minor factors show zero 

 dominance. 



FOURTH GENERATION 



The frequency arrays of fourth-generation families are given in Table 

 X. By this time it was, of course, known which second-generation 

 plants were Ee, and only two Ee families were grown. It was not pos- 

 sible to select directly for long-podded plants homozygous for E, as 

 selection could only be made after growing the progeny. If the minor 

 factors show zero dominance, selection for specially long pods should be 

 speedily efficacious. Among other desirable characters, extra length of 

 pod was sought for. Hence, the chances were that most selected third- 

 generation plants would be the homozygotes in their families with regard 

 to minor factors. 



