192 Davis. 



closel}' the conditions of the pollen parents but the coloration appeared 

 to be uniformly red as in franciscana. This is the first instance in 

 my experience with hybrids of Oenothera where a character of one 

 parent has appeared as dominant in the Mendelian sense. It is note- 

 worthy that this should be a character of color. Gates ('12) has re- 

 ported dominance of red in crosses between 0. ruhriealijx and certain 

 other Oenotheras with a segregation in the Fä generation, and Herebeht- 

 NiLSSON ('12) found similar behavior of anthocyan coloration in his 

 extensive studies on forms of Lamarckiana and its derivatives. I shall 

 follow with interest the behavior in the Fo of the red coloration of 

 the sepals and papillae in the hybrids between biennis and franciscana. 

 The reciprocals of biennis and franciscana differed strikingly from 

 those of biennis and muricata in being conspicuously vigorous, in pro- 

 ducing abundant pollen, in maturing a large yield of seed, and in having 

 a long flowering season. There was no evidence of diminished vigor 

 and the prospects of obtaining a large progeny in later generations 

 seems good. The cross may then prove to be exceptionally favorable 

 in comparison with the results reported by DE Vries for hybrids of 

 biennis and muricata. The hybrids of biennis and franciscana exhibit 

 strong patroclinous tendencies in the moiphology of the rosettes, in 

 the pubescence of the stem, sepals and capsules, in the habit, foliage 

 and bracts of mature plants, in the form of the sepal tips, and in the 

 length of the capsules. Here is excellent material for examination 

 through the progeny of the selfed reciprocals, double and sesquirecip- 

 rocals, and in the iterative hybrids: they will fdiin the subjects of 

 future studies. 



(3) Reciprocal Crosses between Ocnotheiui hiemii.s Linuaeus and 

 O. f/randi/iora Solander. 



My observations on these reciprocal hybrids are based on cultures 

 grown in 1912 of crosses between the Dutch biennis of the same stock 

 employed in the studies previously described in this papei' and a race 

 of grandiflora from wild seed collected at Dixie Lauding, Alabama. 

 The grandiflora, known as grandiflora D in my cultures, has been de- 

 scribed in an earlier paper (Davis '11, p. 205 — 207) chiefly in relation 

 to certain American Oenotheras at that time designated in my cultures 

 as biennis A and biennis B. These two forms, biennis A and B, are 

 essentially the same and very different from the Dutch biennis 

 (0. biennis Linnaeus): they must eventually be given another name. 



