n6 BOTANICAL GAZETTE [FEBRUARY 



cases only one sex is heterozygous and the other is homozygous; 

 and this is the conception now generally entertained by those who 

 adopt a Mendelian explanation of sex. According to this view, if 

 the male is heterozygous, as has just been decided to be the case in 

 Lychnis, the female must be homozygous. As there are two kinds of 

 homozygotes, namely, "positive" and "negative" (SHULL 9), there 

 remain two possibilities regarding the nature of the female; it 

 may be either a positive homozygote or a negative homozygote. 

 While either of these assumptions will explain about equally well the 

 facts brought to light in Bryonia, neither will fit all those observed 

 in Lychnis without encountering important difficulties. 



Making first the supposition that the female is a positive homo- 

 zygote, as suggested by CASTLE (3), all conditions found in Bryonia 

 will be satisfied if it be also assumed that B. alba is a homozygous 

 monoecist (a modified female condition in this case) in which the 

 absence of the monoecious character is dominant over its presence. 

 The sterility of the F T hybrids in Bryonia unfortunately makes it 

 impossible to test the correctness of these assumptions. It is less 

 easy to make a positive homozygous condition of the female fit the 

 results found in Lychnis as described in this paper. This can be 

 done, however, by assuming: (a) that all egg cells of both females 

 and hermaphrodites carry the gene for the female sex, and not 

 that for hermaphroditism; and (b) that all the sperms of the her- 

 maphrodites carry a gene for the hermaphrodite modification, regard- 

 less whether they possess the gene for the female sex (the "x 

 element") or not. The first of these assumptions seems necessary 

 from the fact that the results are identical in each case, whether 

 a female or a hermaphrodite is used as the female parent; and 

 the second from the fact that females and hermaphrodites result 

 from pollinating a normal female by pollen from a hermaphrodite. 

 The second proposition might be replaced by one involving spurious 

 allelomorphism. It is expected that a second generation will demon- 

 strate the correctness or incorrectness of these several hypotheses. 



The alternative assumption, namely, that the female is a negative 

 homozygote, will just as simply represent the conditions of the F r 

 generation. If the monoecious character in Bryonia be considered a 

 modified male condition similar to the hermaphrodite character of 



