1 T*» LOCK : STUDIES IN PLANT BREEDING 



inferred from analogy. It is clear that in the case of these 

 definite characters of the endosperm the difference cannot be 

 simply impressed upon the grains by the form of the pericarp. 

 and if this is so it is quite possible that the case of the indent 

 of a similar kind. Tschermak's conception of a corre- 

 lation between a vegetative and asexual character must there- 

 fore hold good, at least provisionally.* 



In bringing this paper to a conclusion, I desire to take the 

 opportunity of once more expressing my indebtedness to 

 Mr. Bateson for a great deal of valuable advice. 



APPENDIX I— EXPEK1MENTS WITH KIDNEY 



BEANS. 



Experiments with kidney beans, in which attention was 

 paid especially to the colour characters of the testa, have been 

 carried out on a large scale by Tschermak, who obtained ' 

 results of great interest and complexity. In some of the 

 most typical of Tschermak's experiments the plants obtained 

 in Pj from self-fertilization showed a great variety of types 

 of colour in the testas. The main groups of these showed 



i illy the Mendelian ratio of 3 : I, but sometimes a ratio 

 of I : i was Been, a result not yel fully explained. Within the 

 main colour groups, sub-groups, differing in the particular 

 -had'- of colour or in the presence or absence of mottling, were 



found to exhibit similar numerical ratios. 



ult- of a -miliar nature have heeii obtained by Kmmer- 

 ■on, who observed also cases of irregularity in dominance, 

 which may possibly he comparable with those shown by the 

 blue aleUTOne layer in maize. 



• C.p., Annul* II., p. J27. 



