No. 501] GENETIC AL STUDIES OX 2ENOTHERA 549 



or nauella). Critics have pointed out that this behavior 

 indicates that Lcnnarcliana is heterozygous or hybrid in 

 character since it must form at least two different types 

 of male gametes. De Vries apparently believes that the 

 "twin hybrids" in my crosses show that the grandiflora 

 parent is in a condition similar to that of Lamarckiana 

 and that the "twin hybrids" are due to the mutations of 

 f/rtuuli flora. My interpretation of the behavior is quite 

 the opposite, for, as will be shown, the evidence indicates 

 that the biennis parent, with respect to the characters con- 

 cerned, is heterozygous and that the race of grandiflora 

 is stable. If this is true the evidence does not indicate 

 that the race grandiflora B exhibits with respect to these 

 characters the habit of mutation as claimed by De Vries 

 (12, p. 30). 



Among some 300 plants of grandiflora grown from 

 wild seed and 200 more grown in isolated lines none have 

 presented reel-colored papilla? over green portions of the 

 stem. All green-stemmed forms of biennis have proved 

 perfectly true to this character. One of the best known 

 types of green-stemmed biennis is the Dutch plant exten- 

 sively grown by De Vries and Stomps, and this, as far as 

 I know, is constant. Furthermore, all green-stemmed F x 

 hybrids have in later generations proved constant to this 

 form of coloration. There is thus much evidence that 

 the absence of red in papilla? over green portions of the 

 stem constitutes a homozygous condition. The type 

 biennis I), as stated before, can not be distinguished in 

 other respects from wild plants which lack the red colora- 

 tion in their papilla*, and it seems probable that this as- 

 heterozygous with respect to the character of their stem 



Although I can not as yet present experimental proof 

 that the red coloration of papilla? is a character dominant 

 to its absence, we should expect this to be the case because 

 anthocyan coloration is obviously a character in addition 

 to that of the green and because its inheritance appears 



