120 MENDELISM chap. 



the appended scheme. The figures are obviously 



Purple long x Red round 



I 

 Purple long 



Purple long Purple round Red long Red round 



4831 390 393 1338 Actual numbers. 



4814 408 408 JJJ2 Expectation on 



7:1:1:7 basis. 



far from being in accordance with a normal 9:3:3:1 

 ratio. Nevertheless these numbers, which at first 

 sight appear so irregular, are capable of receiving a 

 simple interpretation. The 9:3:3:1 ratio pro- 

 duced in Fg by a plant which is heterozygous for 2 

 factors depends upon the formation by the F^ plant 

 of 4 types of gamete in equal numbers. If the 

 gametic series given by our F^ purple were of the 

 form BL \ Bl \ bL \ bl we should get a 9 : 3 : 3 : i 

 ratio in F^ of the four classes. But if we suppose 

 that the gametes of the two classes BL and bl are 

 seven times as numerous as the two classes Bl and 

 bL we get an explanation of the peculiar proportions 

 in which the four classes of plant appear in the F^ 

 generation. 



Let us suppose then that the gametes produced by 

 the Fj purples are of the nature 7 BL -. i Bl: i bL-.y bl. 

 Such a series of ova fertilised by a similar series of 

 pollen grains will give a generation of the following 

 composition : 



49 BBLL^i BBLl+7 BbLL + ^g BbLl 

 + 7 BBLl+7 BbLL-^ i BbLl 



+ I BbLl 

 + 49 BbLl 



177 purple long 



