i:?s lock: studies in plant brkkdino, 



Jn order to test the accuracy of the sampling, and at the 

 same time to obtain a longer series of numbers, a further 

 generation wa-s raised. 



F3. 



78 white and 126 yellow grains (not specially selected) 

 were taken for this purpose from cob No. 5, Table 31 — a cob 

 which showed a rather distinct deficiency of yellow grains 

 (46*6 per cent). In this generation a pure white dent variety 

 was used as the pollen parent, as it was desired at this stage 

 to eliminate the red pericarp character. 



Expt. 58. — From the white grains 59 plants produced fruit. 

 4-4 of these bore exclusively white grains. The remaining 

 15 cobs showed white grains with the following exceptions : — 



Table 32. 

 I oob -liiuvod 2 blue grains 



6 cobs showed each 1 yellow grain 

 6 <obs showed each 2 yellow grains 

 I oob showed 3 yellow grains 

 I cob showed (5 yellow grains 



Total \ cllow 



2 



6 



12 



3 



6 



27 



Ml the plants on which these yellow grains appeared grew 

 close together near one end of a row, and it seems clear that 

 some pollen hearing the yellow character had escaped from 

 a plant in the nexl row. The purity of the reoessive form is 

 thus well established in this generation. 



Expt. 59. With regard to the offspring of yellow (/rains, 

 a aingle Cob was gathered from each of 96 plants. One of 

 these ooba showed 32 blue grains in addition to 253 yellow 

 and 205 white, due doubtless to an accidental previous orosa 



by blue not visible as " \cnia." 



I ie remaining e<>b> without exception showed yellow and 



white grains only m the proportion of approximately 50 per 

 oent. oi each in every oob. The aotual percentages are given 

 in Table 33. 



