Sex Ratios in Pigeons. 
471 
number of deaths in proportion to the total number of individuals of 
that age, and this number rapidly decreases in our flock with advanced 
age as the birds die off, are killed, or are otherwise disposed of. Since 
therefore it would be very difficult or practically impossible for us 
to calculate the total population and age composition of our flock 
for the period under discussion, it is impossible to give any curve which 
represents accurately the death rate in our pigeons. 
The point of interest which appears from the statistics (see Fig. 1) 
presented is that no such definite demarkations exist for separating 
Groups B, C and D from one another as marks Group A off from 
Group B. These groups were in fact established before the tabulation 
of deaths was made, on the basis of the life cycle of the birds. Twenty- 
eight days was taken as marking approximately the termination of 
the nestling stage, since at that age most squabs have left the nest. 
Six months was chosen arbitrarily for the beginning of the adult 
stage as representing approximately the average age at which pigeons 
are sexually mature and may begin breeding if conditions are favor¬ 
able. It is true that a few early hatched birds may breed the same 
season, but the great majority do not begin until the following year. 
With the above classification as a basis we may now examine the 
question of a differential mortality. In order to test the statement 
quoted from Darwin (p. 465) tabulations have been made of cases 
in which the two nestlings were known to be of opposite sex and one 
or both died before reaching adult life (period D). A total of 172 
such cases is recorded: in 88 of these the male died first and was 
survived for a longer or shorter period by his female nestmate; in 
84 cases the female died first and was survived by the male. Since 
these figures relate only to bisexual broods, the number of males and 
females hatched was equal, namely, 172 of each.* The excess of 
four male deaths over the female deaths is so slight as probably not 
to be significant, so we may say that the death rate for the two sexes 
was also essentially equal. 
*The excess of males hatched (in the ratio of 105 : 100; cf. p. 465) must be accounted for by an 
excess of male unisexual broods, or by a greater number of males among broods containing but a 
single squab. 
