688 



MUTANT RACES DERIVED FROM OENOTHERA 



the time of flowering (de Vries 1924 b, p. 229; de Vries and Boedijn 

 1924 b, pp. 261, 266, fig. 2). It reaches half the height of the parent 

 species, is hardly branched and has long, narrow, conical flower- 

 buds with a reddish tinge. It is rich in pollen and in seeds. In the 

 table given by Boedijn and myself (1924 b, p. 261) five specimens 

 with 15 chromosomes and of a pure type have been mentioned. I 

 have self-fertilized them in 1923 and got in the next summer the 

 results shown in table 6. 



Table 6. 



Oenothera semigigas mut. pulla. Second generation. 



I have made a control experiment with my race of this name, 

 which originated directly from Oe. Lamarckiana, and which is to 

 be described elsewhere 1 ), and found as an average from the offspring 

 of two parent plants, 34 percent Lamarckiana, 40 percent pulla, 

 and 26 percent of mutants. The constitution is thus sufficiently 

 proven to be the same in both cases. The amount of barren grains 

 was 50 percent for the race and 38 percent (27 to 48 percent) as an 

 average for the five specimens referred to. The mutants of 1924 

 were 2 albida, 5 oblonga, 7 auricula, 7 cana, 4 pallescens, 1 spathulata 

 and 5 nanella. Evidently they showed the same range as the ordinary 

 mutants from Oe. Lamarckiana and no other types occurred. Like 

 the primary forms of the six other main clasess, the pulla, which 

 is the only trisomic mutant in the central group, is a dimorphic 

 form, giving one-half of empty seeds and reproducing from the other 

 half, partly its own type, and partly that of the parent species. 



Mutants of the type Sesquiplex. 



In a previous article I have proposed this name for those mutants 

 which have about one-half of empty seeds, but are constant and 

 uniform in their living progeny (de Vries 1923 a). Their generical 



*) Boedijn, K. Mehrfache Chromosom-verdoppeliingen, Zeitschr. f. Bot. 

 XVIII, 1925, S 161. 



