SEGREGATION. 



15 



germ cells of the female Barred Plymouth Rock we have the two ele- 

 ments F, which causes the individual to be a female, and B, which gives 

 the barring on the feathers. In ordinary cell division both of these 

 elements divide; but in the reduction division F and B unite into 

 one body. This body then separates as the cell divides, so that one 

 of the daughter cells contains F and the other B. Such a female 

 therefore produces two Idnds of eggs, one containing the determiner 

 for the female sex, the other that for barred feathers. 



Several other cases are known in which Mendelian pairs are formed 

 of determiners for characters that are apparently unrelated. Such 

 a case occurs in the purple primrose previously mentioned. We may 

 explain the phenomena presented by this hybrid as follows: Let us 

 assume that the original wild species from which the red and the 

 white races in question are descended had purple flowers and that 

 this purple color was due to two functions of the same cell organ. 

 One of these functions, which we may designate as R, had to do with 

 the production of red coloring matter, while the other, which we may 

 designate as P, changed red into purple, somewhat after the manner 

 in which an alkali changes litmus from red to blue.'^ The determiners 

 R and P were functions of the same cell organ, probably a chromo- 

 some. Since this body had two functions we may represent it by 

 the symbol R-P, the hyphen indicating that the two functions 

 belong to the same cell organ. Since there is a pair of these bodies 

 in each cell, the complete status of these determiners in the body 

 cells would be represented by R-P R-P. 



Now, let us suppose that in one section of the species the deter- 

 miner R becomes latent or is lost. Our formula then becomes 

 r-P r-P, or simply PP in this race, which would, of course, have 

 white flowers, since the determiner for red is absent. In another 

 section of the species the determiner P vanishes, leaving R-p R-p 

 or simply RR. Now, when we cross these two races we bring the 

 determiner r-P into the same cell with R-p. Here we have the red 

 coloring matter produced by one determiner and converted into 

 purple by the other. This would account for the purple color of the 

 hybrid, as well as the red of one variety and the white of the other. 

 Other cases of Mendelian pairs of this nature will be mentioned 

 later. 



In this purple hybrid we may consider that we have one character 

 pair consisting of R from the red parent and r from the white parent, 

 so that this pair consists of red and absence of red, while along 

 with it we have another pair consisting of P from the white parent 

 and p from the red parent, so that this pair consists of the presence 

 of P and the absence of P. But the fact that P and R can not be 

 transmitted together indicates that R and p pertain to the same 



«See article by Shull in American Naturalist, July, 1909, 

 81599°— Bui. 165—11 2 



