R. C. PUNNKTT 79 



for E. As Table I shews the results from these three sources are all 

 in accordance and confirm the original view that the results are probably 

 the outcome of a gametic series of the form 7BL : IBI : IbL : 7bl. 

 That the reduplication is of the same form in both ^ and % gametes 

 is supported by some experiments made in pollinating red round- 

 pollened plants with F-^ plants ex BL x bl. 112 jjlants were raised in 

 this way and consisted of 



Purple long 50, Expectation being 1^9. 



,, round 7, „ „ 7. 



Red long 8, „ „ 7. 



round 47, „ „ ^P. 



The figures make it evident that the ^ gametic series is of the 

 form suggested, and as the F„ results can only be explained on the 

 supposition that this series of ^ gametes fertilizes a similar series of 

 % gametes there is every reason for supposing that the series is of the 

 same form in both. This will be assumed throughout the various 

 cases discussed. 



It is assumed therefore that the reduplication between B and L is 

 on the 7 : 1 basis inside the E. H. strain, and though in some families 

 the two forms Bl, bL are in excess while in others they are deficient 

 these will be regarded as chance results. Further there is some evi- 

 dence to shew that the nature of the reduplication may remain 

 unchanged when the E. H. strain is crossed with some other strain. 

 Eight such families are given in Table II and shew that the expec- 

 tation on the 7 : 1 basis is closely realised. These families also afford 

 evidence that the form of the colour-pollen reduplication is unaf- 

 fected by the presence of sterility in the families. 



There are however other cases in which the reduplication departs 

 considerably from the 7 : 1 basis. These are families derived from 

 plants which are also heterozygous for E. To these we shall return 

 later (p. 81). 



(/3) Nature of mating Bl x bL. 



Four families containing 419 plants were raised from this form of 

 mating and the numbers of the four classes were 



Purple long Purple roun<l Red long Red round 



226 9.5 97 1 



It has been pointed out elsewhere ((5), p. 297) that this result is 

 in harmony with the supposition that the gametic series was 



IBL : 7BI ; 7bL : Ibl. 



6-2 



