MUTANT RACES DERIVED FROM OEN. SEMIGIGAS 



683 



and they are grouped in the following tables according to their 

 external marks and to the condition of their offspring. 



After self-fertilization in 1923 some of the mutants of Oe. semigigas 

 produced a sufficient supply of seeds, while others gave only a small 

 harvest. Only the first category could be used for the present purpose, 

 and cultures of about 60 individuals were derived from them, where- 

 ver possible, in 1924. Almost all of these plants have flowered, with 

 the exception of a small part of specimens of the Lamarckiana type, 

 when this was indubitably clear in the stage of rosettes of root 

 leaves. 



The different constituents of the cultures have been counted 

 in this stage and afterwards during the summer, especially shortly 

 before flowering and in the flowering and fruiting condition. The 

 general results of these separate countings will be given in the tables. 

 These deal with representative members of five of the seven main 

 classes and with some of their secondary forms. In the categories 

 of lata and pallescens no plants have given a useful amount of seeds, 

 but the tables may suffice to warrant the general conclusions. 



Oenothera Lamarckiana semigigas Mut. scintillans. 



From the cross of Oe. semigigas with Oe. (biennis x Lamarckiana) 

 velutina studied by Boedijn and myself (1924, b, p. 261) I had two, 

 and from that with Oe. tardescens, three plants of Oe. scintillans 

 (1924 b, p. 224), all of them of a very pure type. They proved to 

 be sufficiently fertile after self-fertilization and about sixty speci- 

 mens of the offspring of each were cultivated and counted during 

 the flowering period. The two first-named parent plants had 15 

 chromosomes, but for the three others the number has not been 



Table 1 

 Oenothera semigigas mut. scintillans. Second generation. 



