11 



crowding might be expected to have some such effect as that 

 observed. So far as known, all other conditions under which these 

 plants grew, were conducive to vigorous development. At any 

 rate this culture was in all other regards on an equal footing with 

 the other cultures which were being grown during the same 

 period, and which gave normal development of the several 

 biotypes. 



To test the suggestion that crowding might be respon- 

 sible for the failure of dominance, a second sowing was made 

 on February 25, 1911, each seed being sown separately and 

 spaced in such a manner that the seed-pan contained only 120 

 plants. On April 7 these were potted and their growth has been 

 continuously healthy and vigorous. They have not been subjected 

 at any time to the least injurious crowding, yet they show the 

 same tardy development of the tenuis characters as observed 

 before. On April 22 only three of the 120 plants could be distin- 

 guished from B. bp. simi lex, though nearly all were sufficiently 

 advanced that if it had been a normal family representing the same 

 hybrid combination, almost a complete separation of the alter- 

 native types might have been made. The final census of the plants 

 derived from this second sowing, made on the same basis and 

 with the same care as in the case of the first sowing, showed 

 85 B. bp. tenuis and 35 B. bp. simplex, - - a ratio of 2.43 : 1. 

 The conclusion is reached therefore that the low grade of the 

 tenuis characters in this family is inherent, and not a direct 

 effect of unfavorable conditions of the environment. 



This result appears to furnish an illustration of the phenom- 

 enon known as variable potency" *), but whether the gene A 

 for the tenuis characters is really different in family 09275 from 

 that in the other families must remain for the present an open 

 question. It is conceivable that the relative inefficiency of A in 

 this family is due to the operation of some other factor which 

 acts as a partial inhibitor, so that although the actual character 

 of A remains unchanged, it must meet a greater resistance, and 

 therefore produces a less effect. To avoid a decision of the 

 question whether the gene has less power or whether it must 

 meet a greater resistance, I use the expression relative 



!) Davenport, C. B., Heredity and Mendel's law. Proc. Washington 

 Acad. Sci. 9 : 179-187, 31 Jy 1907. 



