356 BOTANICAL GAZETTE 



[NOVEMBER 



will be discussed later in connection with the nature of the 

 hermaphrodites. 



CORRENS (6, p. 17), with undoubted justification, maintained 

 that the germ cells of monoecious, hermaphrodite, and dioecious 

 species possess the tendency to develop into individuals having 

 the distribution of sex organs characteristic of the particular 

 genotype to which they belong; but when he likens the association 

 of organs of both sexes in the same individual to the mosaic of red 

 and white colors in striped flowers, and of pigmented and white 

 spots in the coats of spotted animals, his justification becomes less 

 obvious. Both striped flowers and spotted pelages are known 

 from many investigations to be due to the presence or absence of 

 a definite Mendelian gene, a so-called " spotting factor" or "pat- 

 tern factor." 



One of the chief aims in the arrangement of my cultures for 

 1910 was to test the possible existence of such a mosaic or "pat- 

 tern factor," H, as a proximate cause of hermaphroditism in 

 Lychnis, and the most striking result secured is the decisive manner 

 in which such a possibility is denied. The hermaphrodite character 

 is not only incapable of reaching expression in the female 5 (as 

 might be expected, since the female is homozygous), but it is also 

 as a rule not transmitted through the egg cell to the male 

 offspring. The males in the progeny of any cross agree in their 

 sexual type with the male parent of that cross, regardless of its 

 antedecent history. All the assumptions and implications involved 

 in the first section of table I, in which an independent gene H was 

 postulated, may therefore be rejected. 



si refer here only to the normal functional hermaphroditism with which this 

 paper deals, and not the pseudo-hermaphroditism which results when females of 

 Lychnis dioica are attacked by the smut, Ustilago violacea, as reported by STRAS- 

 BURGER (35). I was inclined (SHULL 26) to consider STRASBURGER'S interpretation 

 of the effect of Ustilago a mistake, as it appeared to me more probable that the infected 

 plants were males whose heterozygous nature easily permitted the development of 

 the female characters. STRASBURGER (36) has shown conclusively, however, by two 

 series of facts, that his original observations and interpretation were correct. These 

 facts are (a) that diseased plants are not always completely infected, in which case 

 the uninfected branches of the pseudo-hermaphrodites have normal female flowers; 

 and (b) that male plants may be infected also, but such infection does not in this case 

 result in the development of the female organs. 



