586 SCIENCE PROGRESS 



that in the egg which produces a female this male-determining 

 X-element meets a dominant female-determiner represented by 

 the X-chromosome of the egg-nucleus ; (2) that after maturation 

 there are two kinds of eggs — half containing the female-deter- 

 miner (X,) and half the male-determiner (X 2 ); (3) that the 

 former kind are fertilised only by the X-class of spermatozoa, 

 the latter kind only by the Y-class (" selective fertilisation "). 1 

 These assumptions are in essential agreement with an earlier 

 hypothesis of Castle, based on more general grounds. 2 They 

 form the first of two alternative interpretations offered (not 

 advocated) by the writer in 1906 3 as possible solutions of 

 the problem. This interpretation seems to me, however, 

 improbable in itself; and when we look further afield the 

 improbability becomes still greater. For example, in the 

 bee, where females arise from fertilised eggs and males from 

 unfertilised, the following assumptions must apparently be 

 made : (1) the fertilised egg contains, as usual, a dominant 

 female-determiner (Xj) and a recessive male-determiner (X 2 ) ; 

 (2) in maturation of the egg the female-determiner is elimi- 

 nated, leaving the mature egg with only the male-deter- 

 miner ; (3) in fertilisation a dominant female-determiner is 

 introduced by the spermatozoon. But this is an utter absurdity ; 

 for, by the assumption, the mature unfertilised egg from which 

 the male (and hence the spermatozoon) arises has eliminated 

 the female-determiner (!). 4 A multitude of difficulties akin to 

 this arise under any form of the assumption that specifically 

 distinct and opposing male and female " tendencies" or " deter- 



1 Such a selective fertilisation was suggested on other grounds, independently 

 and almost simultaneously, by several writers. Beard, in the work already cited 

 (1902), postulated this as the primitive mode of fertilisation in the original types 

 having fourfold gametes, but did not apply it to existing Metazoa. It was sug- 

 gested by McClung in the same year as a possibility in insects having an accessory 

 chromosone (op. at.). The same assumption formed an essential part of Castle's 

 earlier Mendelian theory of sex (1903) since abandoned. 



3 Bull. Mus. Comp. Zool. 1903, 27. 



3 Jonrn. Exp. Zool. 3. 



4 This difficulty (first pointed out by Castle) might be met by the further 

 assumption, which has in fact been made by Beard and others, that there may be 

 two kinds of unfertilised eggs, male and female, of which only the female are 

 capable of fertilisation. There is, however, no evidence of this in the bee ; while 

 in Hydatina the results of Maupas, which have been confirmed by Whitney 

 {Journ. Exp. Zool. 1909), show with at least probability that the egg, which if 

 unfertilised produces a male, produces a female if fertilised. 



