14 APPLICATION OF PRINCIPLES OF HEEEDITY TO BREEDING. 



ovules and two kinds of pollen. For in this cell division, which is 

 called the reduction division because in it the number of chromo- 

 somes is reduced by half, our determiners R and r may be assumed 

 to reside in separate chromosomes which unite to form a single 

 bivalent. While cell division is taking place this bivalent chromo- 

 some again separates, R passing into one daughter cell and r into 

 the other. If the determiners R and r are not simply the functions 

 of chromosomes themselves they at least pertain to bodies which 

 at some point in the hne of descent of the germ cells behave just as 

 we know the chromosomes do behave — that is, at some cell division 

 R and r unite into a pair, and when division occurs R goes one way 

 and r the other. Two determiners which thus behave tow^ard each 

 other are said to constitute a Mendehan pair. 



Most Mendelian pairs consist simply of the presence of a given 

 determiner on the one hand and the absence of that determiner on 

 the other. Furthermore, the determiner which represents the pres- 

 ence of a character is in nearly all cases dominant over the deter- 

 miner which represents the absence of that character. Dr. C. B. 

 Davenport, of the station for experimental evolution of the Car- 

 negie Institution, found that in poultry practically all the character 

 pairs known show this relation; i. e., presence of a character domi- 

 nant and absence of it recessive. We have seen, however, that there 

 are some exceptions to this rule, since the poll character is dominant 

 to horns and beardlessness in wheats is dominant to beards. The 

 difference between polled and horned cattle is the absence of horns 

 in one and their presence in the other. 



But cases are known in which this simple relation of presence and 

 absence of a character does not constitute the Mendelian pair. For 

 instance, if Barred Plymouth Rock females be mated with Indian 

 Game males all the female progeny of this moating will be black, while 

 all the males will be barred like the mother. Data accumulated by 

 the writer and an extended series of experiments performed by Mr. 

 H. D. Goodale " indicate that the female Barred Plymouth Rock pro- 

 duces two kinds of eggs. One of these kinds is destined to produce 

 females, and these female-producing eggs do not have the determiner 

 for barring in them. The other kind is destined to produce males, and 

 these eggs do have the barring factor. In this case it appears, there- 

 fore, that the determiner for femaleness and that for barring form a 

 Mendehan pair. Wilson has apparently shown that the determiner 

 for femaleness in many animals is a certain chromosome or group of 

 chromosomes that always act as a unit. If we assume that the deter- 

 miner for barring is another chromosome which unites with the sex 

 element to form a bivalent in the reduction division we have at once 

 an explanation of the behavior of the determiner for barring. In the 



165 



a See Science, June 25, 1909. 



