MUTANT RACES DEVIRED FROM OENOTHERA 

 LAMARCKIANA SEMIGIGAS. 



Introduction. 



The progeny of Oenothera Lamarckiana semigigas splits up into 

 numerous mutant types, among which the prominent ones are seen 

 to repeat the external features of the old mutated races of Oe. 

 Lamarckiana more or less clearly. Besides these, deviating types are 

 observed, but as a rule the deviations are small and do not consti- 

 tute really new characters. They are to be considered as the result 

 of different combinations of the same elementary changes. 



The question arises, whether these mutants are also like the older 

 races in their hereditary characters. If this is so, they may be con- 

 sidered as belonging to the same general types and as identical with 

 them in the special cases. The semigigas would then have to be 

 considered simply as due to a sevenfold mutation or to the simul- 

 taneous production of the seven main derivatives lately distinguished 

 by Boedijn and myself for 0. Lamarckiana, and of their secondary 

 forms (1923, See also Opera VII, p. 496). 



This problem could be solved by means of the cultivation of the 

 second generation of the mutants described in previous articles. 

 These were derived from three crosses, made in 1922, on a biennial 

 specimen of Oe. (lata x Lamarckiana) semigigas. For one of these 

 the pollen of Oe. (biennis x Lamarckiana) velutina was used and for 

 the offspring of this cross the external features of each type have 

 been described and the chromosomes counted for every individual 

 bij Boedijn and myself (1924 a, b, p. 258). For the two others I have 

 chosen the pollen of Oe. Lamarckiana mut. tardescens and of Oe. 

 Lamarckiana mut. blandina. Tardescens has pure velutina pollen and 

 the offspring behave like those of the first-named cross. In the 

 pollen of Oe. blandina (synonym: Oe. Lamarckiana mut. velutina) 

 the characteristic lethal factor of the velutina type is absent, and 

 therefore the corresponding seeds, which are ordinarily barren, are 

 viable here, producing a large part of the progeny in consequence 

 of their preferential fertilization (de Vries 1924 a). For these two 

 crosses the chromosomes of the mutants have not been counted, 



