340 NILS HERIBERT-NILSSOX 



had been tried. Not albicans but the factor B, less important accor- 

 ding to Renner, is the one that has the greatest effect, as it is not only 

 responsible for the small flowers and short style-characters — since long 

 ago regarded to be species-separating for O. biennis; it is also respon- 

 sible for a whole group of characters in different parts of the plant. 

 Selffertilization is further made possible by this gene. Thus it seen 

 that the simple factor B has a very great morphological effect, while 

 the albicans typ.e is not to be defined morphologically. It is also 

 clearly seen that it is impossible to distinguish between characters of 

 greater or less value. 



If the following assumptions are made the constitution of 

 Oenothera Lamarckiana with regard to the laeta-veliitina characters 

 may be analyzed factorially in the same way as for other plants. Two 

 pleiotropic factors, L and V , are involved in the segregation, and these 

 syntheticize the Lamarckiana type. They show high repulsion, and 

 no positive homozygotes develop at the formation of the zygotes. The 

 dihybrids from the cross repeat the same process of segregation; the 

 very rare monohybrids (the mutants) are seemingly constant on 

 account of heterogamy. All the seemingly peculiar phenomena of va- 

 riation in 0. Lamarckiana become explained if these assumptions are 

 allowed all of which involve complications found to take place not 

 only in the plant in question but also in other plants. 



My lactorial theory explains just as good as the theory of com- 

 plex heterozygotes of Renner, why O. Lamarckiana is seemingly con- 

 stant, although the species is invariably a bastard, and why it always 

 gives by half sterile seeds. It explains in addition the segregation in 

 Fl in the cross O. Lamarckiana X O. biennis and, further, the cross 

 between the F^ types {velutina X laeta), which crosses may be de- 

 monstrated with the help of Renners theory but not analyzed. It 

 explains, finally, the immediate origin of laeta and velutina from 0. 

 Lamarckiana, a phenomenon already found by de Vries with regard 

 to the mutants simplex and blandina. The theory of Renner does not 

 explain this phenomenon, and it compelled de Vries to assume two 

 letal factors and a special process of mutation in addition. 



The theory of Renner and the strongly emphasized belief that 

 only two kinds of gametes are formed in O. Lamarckiana has led 

 LoTSY to the theory that this species is a nuclear chimaera developing 

 only the two kinds of gametes which once gave rise to the species. 

 According to my assumption four kinds of gametes are formed; the 

 origin of the simplex and blandina types demonstrates that this is so. 



