A. St Clair Capokn 273 



dark, translucent colour, which, however, was not purple, it nevertheless 

 gave rise to an F^ in which two flushed purples appeared (both beardless). 

 It is of course not at all unlikely that here the foreign pollen came 

 from another kind of non-coloured altogether, namely, one possessing 

 an inhibitory factor; but in a case like this, mth only 8 F3 plants to 

 study and the roguing as far back as the F^ generation, it is not safe to 

 attempt deductions. In No. 146, on the other hand, the roguing took 

 jjlace later, — in a single F., grain to judge by the bearded nature of 

 the F„ ear and the F^ of 22 bearded + 1 beardless flushed. Hence if 

 the purjjle plant be not a stray from some other row, and the rogue 

 pollen fi'om a non-coloured form, the F-, segregation may possibly be 

 correct, though I doubt it. 



Bearded like the parent the last F^ row. No. 34, came from one of 

 about half a dozen F^, ears with such shrivelled, sprouted grains that it 

 was not thought they would grow, and they were consequently planted 

 speculatively with the colour undetermined. In any case, however, 

 five plants are scarcely enough on which to decide the dominance or 

 otherwise of purple in this particular row. But although extraneous 

 pollination does not appear to have taken place, it is not unlikely that 

 the purple individual came out of No. 35 which possessed but a single 

 plant : mistakes are easily made in harvesting two adjacent rows such 

 as these with extremely few and scattered plants in each. 



To sum up this group of doubtfuls : of the six, Nos. 104, 35, 

 o, and 160 can be totally rejected on account either of insufficient 

 numbers or of hopeless roguing at an early stage, while only the 

 remaining two, in which the degree of doubt is slightly less, though 

 still very strong, may qualify for a few further words in the following 

 discussion of the results. 



At the outset let it be understood that no attempt is being made to 

 formulate a theory to fit the facts. Tentative suggestions and inter- 

 pretations may have to be adopted to explain some of them and also to 

 aid in the readier appreciation of certain aspects of the problem ; but 

 for all that, the contradictory nature of the figures makes them so 

 thoroughly baffling that at present it must be honestly confessed no 

 scheme which will embrace them as a whole can be advanced. 



In some general features the colour inheritance has points in common 



with that of aleurone colour in maize. East and Hayes\ in dealing 



with a case involving purple, red, and non-coloured forms, write that 



" the only difficulty in alining the results obtained with the ordinary 



' East and Hayes, 1911, p. 83. 



