'9i7l 



De VRIES— OENOTHERA 



21 



nervis, originated through the loss of the typical red color, but 

 without change in the brittleness of the stem. These new mutants 

 will have to be dealt with in another article. 



TABLE II 



Cross 



Culture 



O. blandina 



a 



XLamarckiana. . . . 



s\ . . . . 



X nanella . 

 O. LamarckianaX blandina 



1914 



1915 



1915 

 1914 



Percentage of germs 



94 



97 



99 

 90 



Reciprocal 



95 



The most interesting question in this situation, however, is 

 that concerning the seeds of the first generation of the crosses 

 between O. Lamarckiana and 0. blandina. I made the crosses in 



TABLE III 



Cross 



Percentage op germs 



O. blandina X rubrinervis 

 u Xerythrina. , 



u X deserens . . 



Reciprocal 



97 



100 



1914, cultivated the hybrids in 1915, and tried their self-fertilized 

 seeds during the winter. From each of the 3 crosses I had 5 vigo- 

 rous specimens of the laeta type, and an equal number of the 

 velutina type. I shall deal with the laeta first (table IV). 



TABLE IV 



Self-fertilized seeds of laeta specimens of the first generations 



Cross 



■— 1 % _ , 



O. LamarckianaX blandina. . . . 

 O. blandina XLamarckiana 



" ■ XnaneUa 



Percentage of germs 



97 



96 



95 



95 



92 



36 



43 



4i 



39 





The 3 crosses may be taken as instances of one type and com- 

 bined on this ground. We see that a splitting occurs, n indi- 

 viduals having the high percentages of the one patent, and 4 others 



/ 



