A. S. PARKES 
as an indication of the lesser viability of the male foetus, a con- 
clusion which is supported by the fact that observations on the 
sex incidence of abortion have usually shown a large excess of 
males. It might be supposed, therefore, that the earlier in 
pregnancy, the higher the proportion of males among live 
foetuses. This conclusion, however, has not always been borne 
out by examination of the sex-ratio of live human foetuses 
removed for therapeutic purposes. It is to be hoped that con- 
clusive figures on this point will be forthcoming from the mass 
abortion programmes being carried out in Japan and other 
places. Evidence from mammals, however, has in some cases 
shown a steadily decreasing ratio of males during pregnancy. 
The domestic pig is very suitable for studies of this kind and 
Crew estimated the sex-ratio at conception in pigs at about 150. 
Recent work on the golden hamster also suggests a very high 
sex-ratio at conception. 
Another way of looking at this problem is to consider the 
sex-ratio of living births under conditions where prenatal mor- 
tality is known to be high, as with older mothers. The sex-ratio 
of live births during a recent quinquennium was much higher 
with younger mothers than with older ones. This fact probably 
explains the effect of the two world wars, which substantially 
lowered the age of marriage, on the sex-ratio. In doing so, it 
explains neatly the old idea that nature increases the sex-ratio 
after wars to replace the lost males. 
The extent to which the prenatal sex-ratio is affected by 
differential mortality will of course depend on the amount of 
mortality as well as its sex incidence. Overt abortions are said 
to amount to about 20 per cent of all pregnancies, but loss in 
the early stages is probably much greater. Total prenatal mor- 
tality is therefore heavy and almost certainly falls differentially 
on males. It is thus very likely that there is a large excess of 
males at conception; estimates of the sex-ratio in man at this 
time vary between 120 and 150. 
This probability of a large excess of males at conception raises 
some interesting social and biological problems. First, as pre- 
natal and postnatal mortality decreases still further, the excess 
se 
