208 
Journal o f Agricultural Research v 0 i. xxviii, No. 3 
It is apparent that there is a correlation between egg production and sex ratio. 
Between classes 1 and 2 the difference with its probable error is 5.45 ±2.61, 
between classes 2 and 3 it is 12.46 ±2.77, between classes 3 and 4 it is 0.39 ±2.15> 
between classes 4 and 5 it is 6.96 ±1.21, and between classes 5 and 6 it is 5.45 ± 
1.52. The differences are significant in every case except two between classes 
1 and 2 and between classes 3 and 4. Between classes 1 and 6 the difference 
with its probable error is 30.71 ±1.84, and between classes 2 and 5 the difference 
with its probable error is 19.81 ±2.13. The correlation between antecedent 
egg production and sex ratio is high and is significant, since r= —0.704 ±0.037. 
In other words, with the birds used in this study and under the conditions of 
this study, the conclusion is justified that as egg production increases the sex 
ratio decreases. 
Since it has been demonstrated that a correlation exists between antecedent 
egg production and sex ratio and since egg production is affected, to some extent 
at least, by seasonal variations, it becomes necessary to inquire concerning the 
possibility of a significant correlation between seasonal conditions and sex ratio. 
In this connection it should be pointed out that of the 45 birds, 3 commenced 
laying in August, 7 in September, 11 in October, 12 in November, 9 in December, 
and 3 in January. In view of the widely different times of the year at which the 
birds commenced laying, it seems logical to assume that if season affects the 
sex ratio birds commencing to lay early in the season as compared with birds 
commencing to lay late in the season should give different sex ratios for the first 
families of 10 or more. In order to test the validity of this assumption the sex 
ratios of the first families of 10 or more, .obtained as nearly as possible from the 
time laying commenced, and according to the month laying commenced, were 
tabulated and averages determined. On this basis, the following average sex 
ratios are as follows: August, 59.17; September, 60.18; October, 61.17; Novem¬ 
ber,'60.91; December, 56.67; January 58.50. It is apparent, therefore, that 
the season of the year has not affected the sex ratios of the first families of 10 
or more. 
On the other hand, if season as such affects the sex ratio, then it would be 
expected that in the antecedent production classes shown in Table VI, each class 
for which the sex ratio was determined should contain families of 10 or more 
produced at approximately the same season of the year. In class 1, for instance, 
antecedent egg production 0-20, in which R c? = 62.91 ± 1.44, it should be expected 
that the families of 10 or more would have been produced at approximately the 
same season. But it is found upon examination that the following number of 
families occurred in the months designated: September, 5; October, 6; Novem¬ 
ber, 7; December, 4; January, 3; February, 1. In class*2, antecedent egg pro¬ 
duction 21-40, in which case R<? = 57.46±1.88, the following number of families 
occurred in the respective months: December, 3; January, 6; February, 9; 
March, 5. These results show quite clearly that season of itself apparently does 
not directly affect the sex ratio, otherwise irrespective of antecedent production, 
families occurring at approximately the same season of the year should give sex 
ratios of approximately the same value. 
The possible effect of season on sex ratio can be tested in still another way. 
On the assumption that season affects sex ratio there should be a significant 
correlation between the sex ratio of the families and the month in which they 
occur. The value of this correlation has been determined as follows: r = —0.165 ± 
0.076. This correlation is of such low value as not to be significant. Since the 
correlation between antecedent egg production and sex ratio is — 0.704±0.037, 
and since the correlation between season and sex ratio is —0.165 ±0.076, it may 
be concluded that season does not affect the sex ratio unless indirectly through 
affecting egg production. 
