Table I. 



The deficiencies seen to be consistently present in all these 

 pedigrees in the number of rhomboidea and simplex were pro- 

 bably due at least in part to an error of classification, those rlwm- 

 boideas having^ the greatest elongation of the terminal segment 

 of the lobes having doubtless permitted their classification as 

 minus-fluctuations of heteris, and the sharpest-lobed simplex as 

 minus-fluctuations of tenuis. The greatest surprise was occasioned 

 by a ratio of about 22 : 1 in the form of the capsules, as it was 

 thought probable that the difference between the two forms 

 would be found to be dependent upon the presence and absence 

 of a single gene, the expected ratio on this assumption being 3:1. 



At the close of the second generation the cultures were 

 necessarily discontinued, and could not be resumed until in the 

 autumn of 1910. The appearance of a paper by Nilsson-Ehle *), 

 showing that certain characters of wheat and oats are independ- 

 ently determined by two or more distinct units or genes, gave 

 the suggestion that the capsule-character of B. bursa-pastoris 

 might be determined in like manner by two genes, the absence 

 of both of which produces the Heegeri-type of capsule, although the 

 observed ratio 22 : 1 is a bad approximation to the expected ratio 



J ) Nilsson-Ehle, H., Kreuzungsuntersuchungen an Hafer und Weizen. 

 pp. 122. 1909. Lund: Hakan Ohlssons Buchdruckerei. 





