OENOTHERA GRANDIFLORA AIT. 207 



merit. The leaves are dark green. The stature is almost the same 

 as in the species, although at the end it is 2—3 dm. lower. Our 

 climate, which is hardly favorable for the Alabama species, is still 

 less so for this mutant. Not rarely the spikes miscarry, and bare 

 anthers are of quite common occurrence. Especially in 1915 I found, 

 during the whole summer, scarcely enough pollen for selffer+ilization 

 and some few crosses. The flowers are somewhat smaller and the 

 petals less broad than in the parent species, and the fruits are thinner 

 and more cylindrical. These differences are small, however, and 

 probably a result of the insufficient nourishment by the narrow 

 leaves. This latter character is always sharp and clear, and no inter- 

 mediates have been observed. From two self-fertilized mutants of 

 1914 I cultivated a second generation, and from one of them I 

 derived in 1916 the third one. They were uniform lots and strikingly 

 different from the original species. They embraced in 1915 in the 

 first instance 60 specimens, all of which flowered, and in the second 

 about 100 seedlings, which were thrown away as soon as their 

 uniformity was beyond doubt. The third generation in 1916 consisted 

 of 61 plants, almost all of which flowered and resembled their 

 parent. 



I crossed 0. lorea with 0. grandiflora in 1915, but could not find 

 pollen for the reciprocal cross. In June 1916 I had among 59 indi- 

 viduals 35 grandiflora, 15 ochracea, and 9 lorea, giving about 60, 

 25, and 15 per cent. The figure for ochracea is too low, since some 

 seedlings were yellow and died in the seedpan, but it coincides 

 sufficiently with the coefficient of mutation from the parent species 

 as determined in the summer (15—20 per cent in table I). That for 

 lorea is more reliable, since no losses could interfere here. It must 

 be considered as due to the combination of all the mutated pollen 

 grains of grandiflora with lorea egg cells. It points to a high amount 

 of mutated sexual cells, but my cultures were too small and too 

 few to justify a further discussion of this interesting point. 



I have also crossed the two mutants with one another. The results 

 were as follows in June 1916: 



grandiflora ochracea lorea Total 



O. ochracea x lorea 22 8 30 



O. lorea x ochracea 31 23 1 55 



Total 53 31 1 85 



Percentage 62 37 1 



The results of the reciprocal crosses may be assumed to mean 



