272 Preponderance of Males and Females in Fish 



(B) Sex-chromosome formula %ZW, (^ZZ. 



In this case, a very different result is produced. There is no limiting 

 value to the proportion of somatic {ZZ) females, and the process 

 continues until all the females are somatic, the TF-chromosome being 

 entirely eliminated from the stock. The limiting sex-ratio is therefore 

 the same as the number of ZZ organisms transformed into females in 

 any generation. I.e. if this proportion is less than 50 °j^, the pre- 

 ponderance of females will become converted into a preponderance of 

 males by the action of the modifying factor itself If the modifying 

 factor does not cease to operate until after all the females have become 

 ZZ, the swung-back generation will be all males, and therefore incapable 

 of further existence without the importation of individuals from other 

 stocks. Even after a very few (two or three) generations the swung- 

 back preponderance of males may be considerable, and greater than 

 the previous preponderance of females. No non-viable zygotes will be 

 produced. 



(II) If the external factor is of such a kind that it transforms 

 zygotes of female constitution into somatic males, the results given 

 above must simply be transposed, those for the XX, XY formula now 

 applying to organisms of the ZW, ZZ type, and vice versa. 



The figures recollected by Mr Boulenger make it clear that, if a 

 modifying factor of the kind here postulated was at work in Girardinus, 

 then the sex-chromosome formula for this species must be of the $ XX, 

 ^XY type. 



It may be noted that there exists another possible explanation for a 

 swing-back of sex-ratio ; viz. the acclimatization or adaptation of the 

 organism to the modifying factor, which would then cease to exert its 

 effect. 



It is proposed to carry out some experiments with a view to testing 

 these hypotheses. Meanwhile this note is published with the purpose 

 of drawing the attention of those engaged in genetic research to certain 

 theoretical expectations with regard to sex-ratios. 



I must thank Mr E. G. Boulenger for supplying me with unpublished 

 data with regard to Girardinus, and Mr J. B. S. Haldane for his help 

 with the mathematical calculations involved. 



To sum up in general terms: If, in any given species, (1) there 

 exists a chromosome-difference between the two sexes which plays a 

 part in the determination of sex ; and (2) this determination by means 

 of chromosomes can be modified by the action of other factors; then 

 (3) we shall find organisms with the chromosome-constitution of one 



