382 



THE I '/ 1' 1:1' . I A' XATFRALls'J 



[Vol. LII 



seem that this latter explanation is the one which should 

 be used. 



On the first hypothesis the F, individuals are, there- 

 fore, BbGrgYYPp forming eight kinds of gametes. From 

 this assumption then we would expect 48 black pubes- 

 cent : 9 gray pubescent : 3 gray smooth : 4 yellow smooth. 

 The ratio of pubescent to non-pubescent would be 57 : 7. 

 The observed numbers in the second generation were 

 378 pubescent : 42 smooth. We would expect 374.06 

 pubescent : 45.94 smooth. The observed numbers from 

 the three third generation families gave 266 pubescent : 

 42 smooth, while we would expect 274.31 : 33.69. Consid- 

 ering the two groups together we have 644 pubescent : 84 

 smooth, while we would expect 648,375 : 79.625. We see 

 that the observed facts agree very well with the theory. 



While it is not intended to go into details regarding 

 the F, generation, it may be said that a number of ob- 

 served facts tend to substantiate this hypothesis. For 

 example, we should have some families segregating in 

 the third generation giving 15 pubescent : 1 smooth. 

 This we find to be so. Again we would, according to 

 theory, expect to find some F 3 families segregating into 

 blacks and grays where the grays would all be smooth. 

 This we find also. We would also expect some gray oats 

 to segregate into 9 pubescent : 7 smooth. This combi- 

 nation has also been found. 



Regarding the base, it might be well to state that in 

 studying the segregation of this series into the third and 

 even into the fourth generation as yet no yellow oat has 

 been found exhibiting the wild type of base. These re- 

 sults do not agree with those obtained by Surface al- 

 though in general we might expect them to be similar 

 since the yellow Sixty Day oat and the Kherson type are 

 classed by some as the same variety of oat. Yet when 

 we know that it is possible to obtain different strains out 

 of a variety, particularly so far as the inheritance is 

 concerned, it is not surprising that these results should 

 not agree. Let us illustrate this by some results we have 



