134 CARNEGIE INSTITUTION OF WASHINGTON. 



THE INHERITANCE OF GERMINAL PECULIARITIES. 

 SEX. 



Studies in inheritance in sex in Lychnis, so long conducted by Dr. 

 Shull, were continued by him. This yesLT has to a large extent con- 

 firmed the anomalous results previously secured (Year Book No. 13, 

 p. 119) without, however, giving a clue to their explanation. The 

 narrow-leafed males have again produced in most crosses only male 

 offspring, and one narrow-leafed hermaphrodite which was crossed 

 with 31 different females has produced a total progeny of 1,287 her- 

 maphrodites, 2 females, and 1 male. It appears, however, that not aU 

 narrow-leafed males are alike in regard to the proportion of males 

 which they are capable of producing. Of 17 families from crosses in 

 which narrow-leafed males were used, 10 consisted of 517 males and 

 no females, 4 contained 160 males and 5 females, and 3 had 95 males 

 and 41 females. In this last group the females were also all broad- 

 leafed individuals. There has still been not a single narrow-leafed 

 female, and the little dwarf female which was classed as a narrow- 

 leafed female in 1913 has yielded only broad-leafed progenies when 

 crossed with narrow-leafed males. 



WTiile hermaphrodites of Lychnis have usually yielded a mixture 

 of hermaphrodites and females and less frequently males and females, 

 Dr. Shull found in a white-flowered German strain of Lychnis {Melan- 

 drium alburn Garcke) a single hermaphrodite mutant which, in 1913, 

 produced only female offspring in three different families for which it 

 served as the male parent. In 1915 this result has been exactly 

 duplicated in 8 progenies from a second hermaphrodite mutant which 

 occurred in the German strain in 1913. These 8 families have included 

 187 females and not a single hermaphrodite nor male. 



INHERITANCE RATIOS OF BURSA. 



During the current year Dr. Shull has continued his prolonged 

 experiments on the breeding of Bursa and has been able to show: 

 (a) that there is no significant correlation between the number of seeds 

 in the capsule and the ratio of hursa-pastoris to heegeri plants in the 

 progenies produced from those seeds, thus indicating that if selective 

 elimination is involved in the production of defective ratios such elimi- 

 nation is independent of the causes which affect the number o' ovules 

 which develop into seeds; (6) that the plants which are homozygous 

 with respect to the long, sharp lobes characteristic of tenuis develop 

 this character at an earlier age than do those which are heterozygous 

 for this character; (c) that there is a rather strong correlation between 

 the time of flowering and the extent of development of the leaf-charac- 

 ters in certain biotypes from western America, the lobation of the 

 leaves being less well developed in precocious plants than in those 

 which develop more tardilj-^ — the same fact was apparent in a similar, 

 though not identical, biotype from Holland; (d) that there is a close 

 agreement between the number of recessives and of homozj^gous 



