i9i6] DeVRIES— DIMORPHIC MUTANTS 271 



cm. The spikes were densely flowered, the flowers a Httle smaller, 

 the fruits thin and long. The production of pollen was insufficient 

 in many flowers, but this may have been the effect of the ndi- 

 viduals being transplanted from their boxes to the bed in June, 

 which is relatively late in the season. The seed developed badly 

 and contained only a small percentage of normal grains. 



I have sown the seeds of 8 of the 9 specimens with the parental 

 type, and of two of the type of 0. biennis Chicago. The first split 

 into two types, the second only repeated the form of the parent. 

 The splitting percentages were 11, 12, 13, 15, 15, 17, 18, and 25, 

 with a mean of 16 per cent, but the germination had been very 

 poor, giving only 444 seedhngs for the 8 boxes. The progeny of 

 the two specimens of the Chicago type was uniform with '252 and 

 60 seedlings. I counted them in May and June, and left one group 

 of each type to flower. The group from the saligna type contained 

 two flowering saligna, identical with those of the second generation ; 

 that of the atavists 60 flowering plants of the uniform type of 

 0. biennis Chicago. From these facts we may conclude that in 

 this race all of the specimens of the parental type give a dimorphic 

 progeny, while the offspring of the plants with an atavistic type 

 remains uniformly so. Resuming the cultures, we get the follow- 

 ing pedigree : 



1913 Mutant Mutant 



1914 Second generation 50 per cent saligna Chicago 



1915 Third generation 16 per cent saligna Chicago Chicago 



The behavior is exactly the same as in the dimorphic races 

 issued from 0. Lamarckiana. 



Crosses of Oenothera Lamarckiana mut. cana. — As indicated 

 on p. 258 I chose in 19 13 the second generation of a cana mutant 

 which arose from 0. lata (no. 3) for a series of crosses. This cul- 

 ture was the most vigorous one of all my annual cana families, and 

 its percentage figures seemed to be more normal than in the other 

 cases. The crosses were made in both directions with the pure 

 strains of my species and races described in Gruppenweise Art- 

 bildung, and the seeds were sown in 19 14. This first generation 



