K. S, FINLOW AND I. H. BUKKILL. 



89 



Table VIIL 



Shewing the extent of impurity, consisting of red plants, 

 ill plots which should have been pure green. 



This is equivalent to an amount of red impuritj' in the green 

 plots averaging I '5 per cent. 



In the F,, generation the green plants were surrounded by 

 pure and impure reds in the ratio of 3 : 1. The pure reds, the 

 number of which was presumably equal to about half that of the 

 impure ones, would, on crossing with the green plants, produce only 



