52 THEORY OF EVOLUTION 



even when, as in this case, there is a perfectly 

 graded series from one end to the other, by 

 testing out individual!}^ enough of the flies to 

 show that one-fourth of them never produce 

 any descendants but ebonies, one-fourth never 

 any but sooties, and one-half of them give rise 

 to both ebony and sooty. 



JNIexdel's Second Disco^^ry — Ixdepexdext 



assortmex't 



Besides his discovery that there are pairs of 

 characters that disjoin, as it were, in the germ 

 cells of the hybrid (law of segregation) Men- 

 del made a second discovery which also has 

 far-reaching consequences. The following case 

 illustrates Mendel's second law. 



If a pea that is yellow and round is crossed 

 to one that is green and wrinkled (fig. 21), all 

 of the offspring are yellow and romid. Inbred, 

 these give 9 yellow round, 3 green round, 3 

 yellow wrinkled, 1 green wrinkled. All the 

 yellows taken together are to the green as 3: 1. 

 All tlie round taken together are to the wrin- 

 kled as three to one ; but some of the yellows are 

 now wrinkled and some of the green are now 



