556 



THE AMERICAN NATURALIST [Vol. LV 



SHORTER ARTICLES AND DISCUSSION 

 ESTIMATING THE NUMBER OF GENETIC FACTORS 

 CONCERNED IN BLENDING INHERITANCE 



In Science for July 29, 1921, Dr. W. E. Castle develops a 

 "method of estimating; the number of g-enetic factors concerned 

 in eases of blending- inheritance." which appears so simple, 

 so attractively cogent, and so usable, that it is to be feared that 

 the erroneous assumptions upon which it is based may be given 

 less consideration than they merit. 



About seven years ago,^ I took some pains to point out certain 

 fallacies which had crept into genetieal literature through the 

 tacit (or express) assumption that the several factors affecting 

 the size of an organism or its parts, or the intensity of develop- 

 ment of any character, are similar to each other in kind and 

 equal in effectiveness. It has seemed to me that since that time 

 there has been marked improvement in the literature dealing^ 

 with this particular phase of genetical phenomena — whether in 

 response to my paper or through independent following out of 

 the simple logic of the case, it matters not. It comes as a dis- 

 tinct shock, therefore, to see the sudden reversion in Dr. Castle's 

 paper to a supposedly outgrown and abandoned conception. 



Dr. Castle has probably forgotten the bearino- of ray paper, 

 though he commended it very higlily in a letter written at the 

 time of its publication; for so far as 1 can now recall lie lias 

 never referred, to it in any of his fre<iiicm i»;ii»crs. puhlislicd 

 since that time, on subjects irivolvingr the niuh lpic factor liypoth- 

 esis as an explanation of blended inheritance and llu- nioiliii- 

 cation of a so-called "unit-character." This omission lias been 

 the more interesting because my paper even gave the precise 

 interpretation of his hooded-rat case, which has been very lately 

 espoused by him.^ 



my paper said : 



Attomi)t3 to dciteriniiie liow mniiy plural determiners for any 



ql^MMtitati^o <-l,aract."r nn- irn,.U.d in a particular cross are as yet 



pivniM-in-.-. ^nrl, ;,itMMpi- .nv l,;-.,] the imi»-overi liyi>otliesis 



1 ■' 1).,- ''.rr,. i;, , - I.,. h -n, , r.n.a hurm-pa^toris," Zeitschr. 



2 Cattle, W. K., " rioWald Kat^ a.ul tlu. Thc.ry of Genes." Pror. Nation. 

 Acad. Sci. [U. S. Amer.\, 5 : 126-i;U.), 1 April, 1919. 



