SHULL REVERSIBLE SEX-MUTANTS 361 



by males not derived from a hermaphrodite family. These u 

 males appeared in hermaphrodite families comprising a total of 

 5467 individuals, thus possibly indicating a somewhat greater 

 coefficient of mutability than that reported for the production 

 of hermaphrodites from normal males. It seems therefore that the 

 modification of the gene M (or /) into a hermaphrodite gene H, 

 and the reversal of this modification so that a normal male gene 

 is again produced from a hermaphrodite gene, occurs with somewhat 

 unequal facility, but the difference is not great enough to warrant 

 the belief that mutation in the one direction is caused by the 

 appearance of a new, independent organ, while its reversal is due 

 to the disappearance of that organ. It seems to me more probable 

 that these reversible mutations are due to reversible modifications 

 of an element or organ continuously in existence, and not to the 

 production of a new element or the dropping out of an old one. 



The change from a male to a hermaphrodite condition and the 

 reverse are processes both striking and sudden. Perhaps they are 

 as fundamental mutations as those observed among the oenotheras. 

 The interpretation given here of the process of mutation in the 

 sex character of Lychnis seems to be available for other mutations 

 as well. The sudden acquirement of new functions by a gene 

 already in existence is different from the conception presented by 

 DE VRIES in Die Mutationstheorie, to account for the origin of 

 the Oenothera mutants, and is in accord with SPILLMAN'S "teleone 

 hypothesis." SPILLMAN (30) is inclined to. attribute the remarkable 

 mutations in Oenothera to irregularities of mitosis, but in these 

 sex mutants of Lychnis, abrupt genotypic modifications have 

 taken place which can hardly be assigned to such irregular mitoses. 



One puzzling feature of the inheritance of sex in Lychnis is 

 the fact that self-fertilized hermaphrodites produce similar ratios 

 of females and hermaphrodites as are produced when unrelated 

 females are fertilized by sperms from hermaphrodites. Since it is 

 obvious that the two types of offspring are due to the heterozygous 

 character of the male, we are led to the conclusion that even though 

 the hermaphrodite individual is heterozygous in respect to sex, 

 its egg cells 7 are of a single type like those of the normal female 



7 Perhaps I should say "its successful egg cells." 



