262 



MUTATIONS OF OENOTHERA SUAVEOLENS DESF. 



the characters of the first very clearly, whereas the other was the 

 same as the hybrid between Oe. Lamarckiana and Oe. suaveolens. 

 I sowed the seeds of the four lots in February and counted the 

 young plants in June, when the differences were clear and sharp. 

 I found: 



Specimens % lata 



No. 1, base of spike 109 46 



No. 1 , top of spike 113 46 



No. 2, base of spike 118 41 



No. 3, top of spike 114 40 



I cultivated some lata and some suaveolens-Wke individuals until 

 September. The latter maintained their resemblance to Oe. Lamar- 

 ckiana x suaveolens in all respects, as was to be expected, whereas 

 the first developed the characters of the mutant Oe. suaveolens 

 mut. lata, described above. Their stature was low, exceeding one 

 meter only in September, their leaves were broad and dark green, 

 with rounded tops, being intermediate between the leaves of the 

 two parent forms. The stigmata were thick and unequally developed 

 in the same manner as is so characteristic of Oe. Lamarckiana mut. 

 lata but in a lesser degree, and in some instances with the same 

 broadening of the upper part of the style. The anthers were mostly 

 dry, without fertile pollen, but in many flowers I could find a suf- 

 ficient supply of this powder, almost as in the mutant. 



No lutescens appeared in these cultures. 



The figures given show no difference between the base and the 

 top of the spike, but a clear, though small one between the two 

 sets of parents. This indicates a dependency on the individual 

 vigor of the plants. Leaving aside these minor points, the main 

 result indicates a complete analogy with the cross between Oe. 

 Lamarckiana mut. lata and Oe. biennis. 



Summarizing the figures of these experiments, we get the following 

 table: 



Crosses between Oe. suaveolens and Oe. Lamarckiana. 

 A. First generation 



