I. [nheritance in Shlrley Poppy 



71 



But sinco tho prcsent systeni of weight iiijr niakos and C the saine, wc havo 



wliere Sc i« tlio moaii S.D. df all the offsprinsf arrays of eapsules = o-c\/(l — p"), if p be 

 the coefficient of hoinotypic correlatioii. Thus wu liave 



R = rjp (vii). 



(vii) wouid probably bu a good result to work from if we wanted to find the 

 parental heredity R from r and p in matorial where neither pareiits nor offspiing 

 were selected, and where there was iio niarked chancfc of environmunt between the 

 two generations. Pos.sibly the inflnence of selection of offspring and of their 

 environnient, since tliey affect both /• and p, niay bo le.ss marked in (vii) than in 

 rac/a-p, the rcgres.sion given by the first nicthod, for r and p (.see Table VI.) may 

 tend to rise and fall together, and thus the third method may in some cases give 

 US better results than the first. 



TABLE V. 



Parental I nheritance. First Method. 



(7) Results for Parental Inheritance. We have seen that the iirst method 

 ought to eliminate the eftect of a selected parentage, but that it woulil not be 

 uninfluenced by a selection of offspring. There can hardly be a doubt that the 

 Order of Table V. is practically that of the stringency of environment for the five 

 crops, — the most starveling crop being the Highgate oae and the most flourishing 

 the Kiddermiuster. Or, it would seeni that the more luxuriant the crop, the less 

 intense is the strength of heredity. The mean value of the slope of the regression 

 line is not far from the ^ originally given for parental heredity in human stature 

 by Galton, but it is considerably less than the value (about •45) recently obtained 

 for about sixteen cases in man*. We have already pointed out that the restriction 

 of the variability of the general population to be found in an array of offspring 

 due to a single parentage might still be maintained, if different parents had 

 received different treatments. With the assistance of Marie A. Lewenz this 

 point was investigated. She Struck the means of all the eapsules on each of the 

 907 Enfield I plants. These plants were then grouped into familics and thcir 



* Unpublisbeil Fauiily Measuremeuts ou upwaids of 1000 famiües in the posaession of K. l'earsou. 



