290 Noji-Isolablc Races. 



tricotylous specimen and had been raised from seed ob- 

 tained in excliange; it flowered in 1893 in complete iso- 

 lation and produced 6% yellow seedlings amongst its off- 

 spring. I planted out about 60 of the green plants which 

 grew to healthy individuals in 1899. I left the pollina- 

 tion to insects but saved the seeds of each plant separately 

 and then counted the number of yellow seedlings for each 

 in a lot of 300.' 



Y. S. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 22 24 27 30 



Ex. 27 6 1 1 1 1 1 2 3 1 1 2 4 3 1 1 



Or 27 63 4 5 6 31001 



"V. S." signifies the yellow seedlings in each lot of 

 seed and ''Ex." the number of plants which exhibited 

 this proportion amongst their seedlings. In the lower 

 line from 2% onwards these are arranged in groups of 

 3 in order to emphasize the general result. The num- 

 bers of this last row are plotted in Fig. 55. This curve 

 is similar to the curves of monstrosities wdiich have been 

 subjected to a selection extending over many years. ^ 

 Even as these, it consists of a half curve and of a bilateral 

 curve. It indicates therefore the selection of a latent 

 character which in this particular case started with the 

 choice of a tricotylous plant which happened to be a 

 partictilarly suitable one.^ 



The observations and experiments which we have 

 described or passed in review show that variegated plants 

 constitute a group of forms which in spite of being se- 

 lected for years or even for a century have manifested 

 no further improvement in the quality and constancy of 



^ See the second section and Sur les coiirhes golfouiennes dcs 

 monstniositcs. Bull. sc. de la France et de la Bclgiquc, published by 

 A. GiARD, Vol. XXVII, April 1896, p. 396. 



^ See Uchcr einc Methode, Zivangsdrchungen aufziisuchcn, Ber. 

 d. d. hot. Ges., Vol. XII, 1894, p. 25. 



