92 Negative Correlation In Oenothera Ilj/hrids 



in the spotted group was perceptibly darker green and the leaves 

 slightly broader and more crinkled than in the unspotted group. 

 As the rosettes grew older, a small number were noted, which were 

 differentiated from the rest in being rather coarsely crinkled, with the 

 obvate to oblong-dbvate blades abruptly contracted, with a slight 

 undulation, to a ti'iangular- winged petiole. The lea\es <if' these had 

 more numerous and more conspicucius red spots than their spotted sibs. 

 That this gioup grades into the more common type of spotted ])lants, is 

 shown by the fact that I separated only 6 out of the 9 plants which 

 subse(juent development pro veil to belong to the same natural group. 

 Fifty plants were set in the field, including 30 spotted and 20 unspotted 

 plants, the members of each of these groups being taken at random, 

 with the exception of one of the crinkled-leafed, strongly spotted plants 

 just described, which was discovered before the reservations were made 

 for the field cultures. Aside from this one individual, therefore, the 

 distribution of the different types in the field should accurately repre- 

 sent, within the limits of the piobable error, the composition of the 

 entire progeny. The adult plants were all of one type with respect to 

 general habit of branching and the long capsules (2'5 — 27 cm.), being 

 in these regards like my selfed 0. ruhricalyx. Among them were three 

 classes of individuals sharply distinct from one another, namely, 



{a) 9 [27]^ with very dark, dull-red stems, and buds wholly free 

 from red pigment ; 



(b) 5 [17] with brilliant red stems, and Ldinairkiinm-Wki- buds, 

 i.e., with pink cones and green hypanthia; and 



(c) 36 [79] with greenish stems and rubrical yx-hiuh, i.e., brilliant 

 red hypanthia and cones. 



It would have been desirable to make an actual determination of 

 the relative amounts of anthocyan in the stems of groups (ff) and (b), 

 but for lack of time this determination was not made. It was my 

 impression, however, that the dull blackish-red stems of the former 

 contained more anthocyan than the brilliant stems of the latter. If 

 this impression is correct, the three grades of pigmentation of the buds 

 are completely associated with three grades of pigmentation in the 

 stems, but the bud-series and the stem-series run in opposite directions, 



' In brackets are given the number of each type which should liave been present if tlie 

 entire progeny had been grown to maturity. Tliese are the significant numbers, as tlie 

 proportionality among the observed numbers is distorted by the fact tliat a relatively larger 

 number of unspotted rosettes were grown to maturity, than of spotted ones. 



