Julian S. Huxley 271 



ratio of 3 ? : 1 </•. As we have already seen, if the modifying factor were 

 now eliminated, the sex-ratio in the next generation would be 5 $ : 6{/'. 

 If on the other hand it were to continue its action, we should have the 

 following result : 



^1 constitution : 2ZX : 2XY. 



F^ sex-ratio : 2XX ? -MXF? : IXY^ = 3 ? : 1 c/*.. 



Fi gametes : Ova, 5X : 1 Y: Spermatozoa, IX : IF. 



F^ constitution : 5XX : 6XF : 1 FF (non- viable). 



50 7o of the X F's will again be converted into females, so that the 



V, sex-ratio will be 5XX ? -h 3ZF? : SZFc/ = 8 ? : 3 «/". 

 F^ gametes : Ova, 13X : 3F. Spermatozoa, IX : 1 F. 

 Fs constitution: 13XX : 16XF: 3FF. 



The swung-back sex-ratio, if the modifying factor were not to act on 

 this generation, would thus be 1S^:16(^, a preponderance of males 

 which is greater than if the modifying influence had ceased a generation 

 earlier. 



If the modifying factor were to continue its action, the sex-ratio of 

 F, would be 13XX ? + 8XF? :8XY^=21%: 16c^. This shows a less 

 preponderance of females than the first modified generation, but the 

 proportion of XF females is higher (^ instead of ^). 



Mr Haldane has kindly calculated for me the theoretical expecta- 

 tions for this progressive alteration of sex-ratio. The details will be 

 found in the appendix to this paper, but the general results may be 

 briefly summarized as follows : 



(I) External agency altering the sex-ratio by converting zygotes of 

 male chromosome-constitution into " somatic " females. 



(A) Sex-chromosome formula $ XX, cT X F. 



The decrease in the preponderance of females is slight,^ and finally 

 comes to a limiting value, which is itself determined by the existence 

 of a limiting value for the proportion of somatic (XF) to genetic (XX) 

 females. During the action of the modifying factor, the sex-ratio will 

 always show a preponderance of females. The preponderance of males 

 after a swing-back is always less than the previous preponderance of 

 females, and always slight ; it can never be greater than V2 . (/• : 1 ? . 

 A definite proportion of non-viable YY zygotes will always be produced, 

 the greatest possible number being slightly over | of the total number 

 of zygotes. 



