HOW CHARACTERS BEHAVE 1 45 



when the mixture is bred. This fact was long a great stum- 

 blingblock to breeders, involving the business of improvement 

 in unfortunate and, as we now know, unnecessary mystery. 



Mendel's law of hybrids. 1 This so-called Mendel's law, 

 named for its first discoverer (I say first, for it was lost till 

 rediscovered), attempts to predict what will be the real char- 

 acter of the offspring of mixed or hybrid parents when the 

 characters of the mixture will not blend. 



What really happens in such a case is this : The hybrid 

 offspring, instead of possessing a new character which is a 

 kind of mean or blend between the different characters of the 

 two parents, will contain them both; and when these hybrids 

 are bred together, their offspring will be not of one but of 

 three distinct kinds, namely, a group that is like the one origi- 

 nal and pure parent, another group that is like the other 

 original and pure parent, and a larger group that is hybrid 

 like its immediate parents. 



For example, let x and y represent any two nonblending 

 characters in separate individuals. What will happen when 

 they are bred together, and when their hybrid offspring are 

 afterwards bred among themselves ? 



The problem stands thus : One parent produces both x and y 

 characters. The other parent also produces both x and y char- 

 acters. What are the combinations that will take place ? Mani- 

 festly these combinations will follow the law of chance. In one 

 case out of four the two x's will unite, making pure xs (x 2 ) ; 

 in one case out of four also the two j/'s will unite, making pure 

 ys (y 2 ) ; and in the two other cases the x and the y will unite, 

 making again xy offspring in numbers equal to both the others ; 

 that is, the total result of breeding together a lot of hybrid indi- 

 viduals with mixed characters x and y will be in the proportion 



1 Mendel, an Austrian monk, carried on experiments in his garden that 

 brought out the principle here stated, but all of which was lost and lay un- 

 known for many years. For a more extended account, see H Principles of 

 Breeding," p. 513. 



