560 



THE AMERICAN NATURALIST [Vol. LV 



ing and minus-acting growth factors gives a very true, though 

 somewhat formal conception of the general situation in all organ- 

 ized beings; — the interplay of growth-promoting and growth- 

 inhibiting factors may be thought of, in a figurative sense, as 

 forming, within the limits of fluctuating variation, a sort of 

 elastic "mold" into which any organism, whether plant or 

 animal, develops, and which gives it its wonderful specificity of 

 form and size. 



But on the basis of this conception of factors, acting, some 

 in a positive and some in a negative direction, the combined 

 action of the negative group exactly balancing the combined ac- 

 tion of the positive group, and jointly determining the mean 

 size or the average condition with respect to any blending char- 

 acter which may be under consideration, it becomes unnecessary 

 to assume the absence of dominance. I have been teaching my 

 students for the past six years that the postulation of lack of 

 dominance which has always been made the basis of the multiple- 

 factor interpretation of inheritance of size or of other blending 

 characters is wholly unnecessary and that those who have dis- 

 cussed this type of inheritance have been led to place an alto- 

 gether unnatural and unwarranted stress on the occasional oc- 

 currence of incomplete dominance in other cases. 



But whether any or all of the size-factors are dominant or 

 not materially affects the amount of change which they effect 

 in the value of the F2 standard deviation, and must correspond- 

 ingly change the estimate of the number of factors involved when 

 that estimate is based on the value of these Fo standard devia- 

 tions. As a simple example, I may cite the hypothetical illus- 

 tration given in my 1914 paper (p. 129), referred to above: 



Thus, if a plant possessing a partial inhibitor or reducer of intei'- 



internode length, all the other genes being the same in the two cases, 

 the height of the plants Avould be intermediate between the heights 

 of the parents, with variability due alone to fluctuation, as it is in the 

 homozygous parents. The would show increased variability, and 



were domina.nt, than if dominance xcere absent.'' 



To illustrate this fact further, let us assume that the six 

 size-modifying factors which differentiate two mates are 

 AABBGC, acting in the plus-direction, and DDEEFF, acting 



5 Not italicized in the original. 



