568 
GENETICS: HARRIS, BLAKESLEE AND KIRKPATRICK 
the information which the physiologist desires. A special coefficient 
due to Pearson and Harris^ shows the correlation between the annual 
total and the deviation of the monthly record from the value which it 
should have if variation in monthly production were directly propor- 
tional to variation in the annual production. Figure 2 gives the results 
for 1914-1915 — those for the preceding year being similar. 
The coefficients show that in both years the winter months Novem- 
ber, December, January and February and the following autumn 
months, September and October, show an increase over their theoretical 
quota of eggs when the annual egg production rises above the normal. 
The spring and summer months, March, April, May, June and July, 
show a lower relative contribution to the annual total than is theoreti- 
cally to be expected when this total varies in the direction of an increase 
above the mean annual egg production of the flock as a whole. Thus 
in the diagram the coefficients in both years increase from November 
to December or January, then drop to the first negative value of the 
year in March and reach their numerically negative largest value in April, 
after which they rise and become positive in sign but numerically very 
low in August and attain their maximum positive value in October. 
A knowledge of the correlation between the egg records of the indi- 
vidual months is essential to a full understanding of the physiology of 
egg production in the fowl. We have worked out sets of correlations, 
110 coefficients in all, for the production of five of the individual months 
and the production of each of the other months of the contest year. 
The months selected are November of the pullet year and the follow- 
ing October and three intervening months, January, April and August. 
The fact that these 110 coefficients are without exception positive 
in sign seems to us a result of very material biological significance. It 
indicates that if abnormally high laying at one period tends, as the 
result of nutritional or other physiological factors, to result in abnor- 
mally low production during a subsequent period, the reduction is not 
sufficient to outweigh the influence of the initial differentiation of the 
birds in their capacity for egg production suggested above. 
Two laws are evident in these inter-mensual correlations. These 
are to some degree mutually obscurant and must be considered in 
their mutual relations, but for the sake of simplicity may be stated 
categorically. 
1. The correlation between the egg production of the individual 
months tends to become smaller as the records upon which the correla- 
tions are based become more widely separated in time. 
2. There is a more intimate correlation between the egg production 
