SEGKEGATION". 



13 



therefore be represented by the symbol RR. This means that in 

 the cells of a plant of the red variety there are two determiners for 

 red. The corresponding determiners in the white variety may be 

 represented by the symbol rr, wliicli may be taken as representing 

 two bodies, neither of wliich performs the function necessary to the 

 production of red coloring matter, but wliich correspond in the white 

 variety to homologous bodies wliich do perform this function in the 

 red variet}^ The sjanbol of the hybrid would, of course, be Rr, in 

 which R represents the active" determiner derived from the red 

 variety and r the nonactive one from the white variety. In ordinary 

 growth, when a body cell has attained its maturity and divides into 

 two cells it is supposed that each character determiner present 

 divides, one part going into one of the new cells, the other into the 

 other. Thus, if a mature cell contains the determiners R and r, 

 then each of the new cells formed by its division likewise contains 

 both R and r. Thus every cell in the body of the individual may be 

 supposed to have both R and r in it. This is certainly true of those 

 cells which form the direct hne of descent from the original fertihzed 

 ovule to the new ovules and pollen grains produced by the individual. 

 The cells in this line of descent are called collectively the germ cells, 

 a term which we shall find convenient to use. 



If ovules and pollen cells were formed by ordinary cell division 

 such as that described above, it is clear that every ovule and every 

 pollen grain produced by the hybrid Rr would contain both R and r. 

 But the facts indicate that only half the ovules and half the pollen 

 grains contain R, while the other half contain r. There must be, 

 then, a cell division somewhere in the line of descent which differs 

 from the ordinary type of cell division, and there is unmistakable 

 cytological evidence that such is the case. Just before the forma- 

 tion of ovules and pollen grains (in fact, in next to the last division 

 of the germ cells) we find a cell division in which the chromosomes 

 do not divide in the usual manner. Instead they unite in pairs, 

 forming double, or bivalent, chromosomes. This union of chromo- 

 somes into pairs reduces the number of chromosomes to half what 

 it w^as before. Then, when the cell divides, these large chromosomes 

 divide, presumably into the two halves which united to form them. 

 If we call the large double chromosomes mother chromosomes and 

 the small ones into wliich they separate daughter chromosomes, 

 then in this cell division one of the daughter chromosomes passes 

 to one of the newly formed cells, wliile the other passes to the other 

 cell. Now, if these chromosomes either themselves are the bodies 

 whose functions are our ^^determiners," or if they contain smaller 

 bodies which are responsible for the determiners, we have at once 

 an explanation of the fact that our hybrid produces two kinds of 



165 



