92 SCIENCE PROGRESS 



due to altered environment or other circumstances usually 

 amount to only a few per cent. ; but if they are consistent in a 

 large population they must be taken into account in any theory 

 of sex-determination. There is a vast number of papers of this 

 kind, suggesting that a great variety of external circumstances 

 may affect the percentage of the sexes among the offspring born ; 

 among these we may choose a few of the more recent as 

 examples of the kind of result obtained. Pearl 1 shows that of 

 over 200,000 births in Buenos Ayres, the proportion of males 

 is significantly greater when the parents are of different racial 

 stocks than when they are of the same. The difference ranges 

 from about one to about five per cent, but is always on the same 

 side. Punnett 2 finds that in London the proportion of males 

 is lowest in the poorest portion, highest in the wealthiest and 

 intermediate in the intermediate portion. The males per 100 

 females were respectively 99/5, 102*2 and 1007; but he points 

 out that these differences are probably wholly explicable on 

 the grounds of differential infant mortality, birth-rate and 

 probably marriage-rate. Heape, 3 from statistics of over 17,000 

 greyhounds, concludes that whilst males are always considerably 

 in excess (averaging 118*5 t0 I0 ° females), the proportion is 

 noticeably higher in the season during which fewest pups are 

 born. In a later note in the same volume {loc. cit. p. 201) Heape 

 gives some curious figures with regard to canaries, showing that 

 in one aviary (out of 200 birds hatched) the ratio of males was 

 about 77 to 100 females, while in another (out of 68 birds) the 

 males were in the ratio of 353 to 100 females. Evidence is 

 given that these differences are not ascribable to mortality ; 

 Heape supposes that in both cases the proportion of the sexes 

 is due to a selective action of conditions on the ova which are 

 matured. He assumes that ova bear either maleness or female- 

 ness and that some forms of environment favour the matura- 

 tion of one kind, some of the other. The same explanation is 

 applicable to other cases in which the proportion appears to be 

 influenced by external circumstances. 



It appears therefore that the idea that the proportion of 

 the sexes may be influenced by conditions acting on the parents 

 is not inconsistent with the hypothesis that the germ-cells bear 



1 Biol. Bulletin, xv. 1908, p. 194. 



2 Proc. Camb. Phil. Soc. xii. 1903, p. 262. 



3 Ibid. xiv. 1907, p. 121. 



