ANNUAL MOLT AND EGG PRODUCTION 75 



affecting fecundity and upon annual egg production, it is desirable to know 

 whether it affects the duration of molt. 



Records are available on 936 birds giving date of hatching and length of 

 molt period at the close of the pullet laying year. Of this number, 7 birds 

 or 0.75 per cent exhibited no complete cessation of egg production at the end 

 of the year as is usually the case during complete molt. Unfortunately, com- 

 plete records of feather change were not kept on these individuals. This 

 group was made up only of hens showing no cessation of egg production for 

 a period longer than seven days at the end of their pullet laying year. There- 

 fore, all individuals ceasing to lay for eight or more days at the end of the 

 pullet year were classed as molting birds and used in the correlation studies. 

 In case of two periods of 30 days or more of non-production separated by a 

 period of laying, the longer period was considered as associated with complete 

 molt. 



The calculation of the simple coefficient of correlation between duration of 

 molt period and hatching date for the 929 molting birds gave the following 

 constants: 



Number of birds _. 929 



Mean length of molt period — days 75.60 



Molt period standard deviation — days ±30.38 



Mean hatching date (Apr. 16) 4.17 



Hatching date standard deviation ±2.21 



Coefficient of correlation - —.0943 ± .0219 



Correlation ratio __ .1 190 



The class interval used for molt period was 10 days and for hatching date 

 7 days. An average period of 75 days duration without egg production char- 

 acterized the molt period. The magnitude of the standard deviation in molt 

 shows that there was wide variability in the population with respect to dur- 

 ation of molt. The actual range was from 8 to 197 days. 



The coefficient of correlation between duration of molt and hatchin,g date 

 was — .0943 ± .0219. This constant has a very small absolute magnitude and, 

 judged by criteria already referred to in this report, has no significance. 



Regression of molt duration upon hatching date by the X^ test gave a 

 probability of .6695 of being linear. It seems justifiable to assume, therefore, 

 that the simple coefiicient of correlation was a true measure of association 

 between length of molt and hatching date. 



The correlation ratio of .1190 for molt duration to hatching date did not 

 differ significantly from the correlation coefficient of .0943 according to 

 Blakeman's test. The two constants squared differed by only .0053 ± .0032, 

 which is but 1.66 times its pro-bable error. 



Inherited Characteristics 



2. Age at First Egg versus Length of Molt Period. 



Age at first egg offers a definite criterion for selecting heavy layers because 

 early laying or early sexual maturity is correlated with high intensity and 

 with high persistency (Hays and Sanborn, 1926 and 1927 a). Information on 



