396 



SCIENCE 



[N. S. Vol. XXIX. No. 740 



ness is not, then, the Mendelian allelomorph 

 to femaleness, but a differential factor be- 

 tween male and female is allelomorphic to 

 absence of that factor. Presence of that fac- 

 tor means femaleness, absence of it means 

 maleness. This differential factor is in- 

 herited as a Mendelian character dominant 

 over its absence. 



Such a statement will, I believe, bring into 

 harmony the seemingly discordant results of 

 Wilson, of Correns, and of Bateson and his 

 associates. For CorrenS' urges the view, as I 

 did in 1903, that the male and female sex- 

 characters as such are inherited. He be- 

 lieves further that the female organism is a 

 homozygous recessive (?2) and the male a 

 heterozygous dominant (<??), for he finds that 

 the egg-cells of Bryonia dioica all transmit a 

 female sex-tendency, whereas the pollen cells 

 transmit, half of them the female tendency, 

 half the male tendency. His facts are un- 

 questionable. I question only the supposed 

 recessive nature of the female sex-character. 

 Wilson cautions us against the view that 

 sex as such is inherited, believing that the dif- 

 ference between the two sexes is in reality a 

 quantitative one. He finds the female char- 

 acterized by the possession of two X-chromo- 

 somes, the male by one, and regards a second 

 x-chromosome as the differential factor be- 

 tween male and female. In the view that the 

 essential difference between the sexes is a 

 quantitative one, Wilson makes general an 

 assumption made earlier by Morgan* for a 

 particular case. 



This suggestion seems to me very helpful. 

 Among other things, it clears up fully the 

 long mysterious matter of sex-determination 

 in the honey bee, of which I gave in 1903 an 

 interpretation since proved to be wrong. But 

 though we regard the distinction between 

 male and female as quantitative, we must not 

 forget that it is discontinuous. The female 

 is the male condition plus a distinct unit- 

 character Mendelian in heredity. 



We must also not follow Professor Wilson 

 too closely in his assumption " that a single 



'Science, Vol. 21, 1905; Am. Naturalist, Vol. 

 41, p. 715, November, 1907. 



X-element in itself causes or determines the 

 male tendency, while two such elements in as- 

 sociation create, or at least set free, the fe- 

 male tendency." For we shall presently see 

 reasons for believing that in certain cases one 

 X-element may determine the female tendency, 

 while no X-element may determine the male 

 tendency. But in both categories of cases 

 alike the essential difference between male and 

 female would seem to be one X-element, which 

 the female possesses over and above the male. 

 We may leave open the question whether or 

 not the " X-element " of Wilson is the actual 

 material basis of this differential Mendelian 

 unit-character of sex. The X-element at least 

 behaves in cell-division as- we must suppose 

 that the basis of a Mendelian character would 

 behave, and it will be convenient in what fol- 

 lows to treat it as actually representing such 

 a basis. 



Wilson's hypothesis will account satisfac- 

 torily for the experimental results of Cor- 

 rens, for it necessitates the production in 

 gametogenesis of eggS' all alike in sexual 

 tendency, bearing X, but it calls for the pro- 

 duction of spermatozoa of two different sorts, 

 half of them bearing X, half of them without 

 X. Eggs fertilized by the former should pro- 

 duce females (XX), those fertilized by the 

 latter should produce males (X). Correns's 

 observations accord with this interpretation. 

 But the Wilson hypothesis breaks down if 

 we attempt to extend it to the cases discov- 

 ered' by Bateson and his associates. For in 

 these it is evident that the eggS', not the 

 spermatozoa, are dimorphic in sex tendency, 

 whereas the spermatozoa are all alike. We 

 can not reconcile such a condition with the 

 hypothesis that XX produces a female, X a 

 male. But the condition in question does 

 harmonize with the assumption, X^^a 

 female, no-X = a male, and this condition, no 

 less than that described by Correns for 

 Bryonia, agrees with the more general as- 

 sumption that the female possesses one more 

 X-element than the male. 



The cases to which reference has been made 

 in which the female produces eggs with dif- 

 ferent sex-tendencies, but spermatozoa all with 



