OENOTHERA GRANDIFLORA AIT. 217 



the cross 0. Lamar ckiana x biennis, which gives uniform hybrids 

 of a type in which the characters of the male parent largely dominate; 

 but the results are very different, as we shall presently see. 



The two parents of the cross have both a large supply of good 

 seeds. The character of 0. Lamarckiana to produce at least one- 

 half of empty grains is not present in either of them. There is no 

 reason to expect this phenomenon among their crossed seeds, there- 

 fore, and as a matter of fact I counted 85 good germs in 100 seeds 

 from this cross; whereas the reciprocal cross, which produces the 

 laeta and velutina as we have seen, had 79 germs in 100 seeds. The 

 figures do not essentially differ 1 ). 



I made the cross in 1914, and in 1915 had a set of 54 plants, 

 among which 45, or 83 per cent, resembled their pollen parent in 

 almost all respects, whereas 9, or 17 per cent, repeated the marks 

 described for the mut. ochracea. All of the latter and the larger 

 part of the former flowered in August. In the biennis type the leaves 

 were narrower (3.0 x 11 cm.), with reddish midveins; whereas 

 the ochracea had the ordinary broad leaves (3.5 x 1 1 cm.) and white 

 veins. The stature of the biennis exceeded that of the ochracea in 

 July by 10—20 cm. (60 cm. as compared with 40—50 cm.). During 

 August these differences gradually increased and the spikes with 

 ripening fruits were compact in the one and loose in the other type, 

 corresponding to those of the pollen parent and of the mutant. The 

 hybrids of the biennis type became at the end very stout, reaching 

 almost twice the height of the ochracea plants. In 1916 I cultivated 

 a second generation from each of the two types, embracing 70 and 

 51 specimens, most of which flowered. The offspring of the biennis 

 plants were a uniform lot, exactly repeating the characters of their 

 parent; those of the ochracea were dimorphic. Some plants made 

 first a rosette of large radical leaves and from this produced a stout 

 stem, whereas the others did not produce a rosette, but at once 

 grew up, causing the stems to be thin and weak. It should be men- 

 tioned that the initial rosettes are a character of biennis, whereas 

 0. grandiflora normally produces stems without this preliminary 

 step. In the first generation the differences had been the same in 

 this respect, the biennis plants having preparatory rosettes and 

 stout stems, but the ochracea lacking these characters. Thus we 

 see that this mark returned in part of the specimens of the second 



J ) My determinations gave, as a mean from 7 countings of lots of 200 seeds 

 each, 75—76 per cent of seeds with normal germs for the cross 0. Lamarckiana 

 X biennis (see 4, p. 268, Opera VII, p. 136). 



