10 



MASS. EXPERIMENT STATION BULLETIN 443 



the number of days concerned; spring rate, number of eggs laid in March, April' 

 and May, divided by 92; summer rate, number of eggs laid in June, July, and 

 August, divided by 92; and fall rate, number of eggs laid in September and 

 October, divided by 61. The net rates of laying in each of the seasons was cal- 

 culated after deducting all pauses of eight or more days. Gross annual rate 

 represents the production during the first 365 days after first egg, divided by 

 365. The net annual rate was calculated after the deduction of all pauses of 

 eight or more days. 



Table 4. — Correlation Between Mothers and Daughters in Rate 

 OF Laying and in Production by Periods. 



*Not significant. 



Table 4 presents the simple correlation between mothers and daughters with 

 respect to rate of laying. In cases where regression was non-linear by Blakeman's 

 test, the correlation ratio is given. In all cases where regression was linear there 

 was a small but significant correlation between mothers and daughters. When 

 regression was non-linear, the magnitude of the correlation ratio suggests a 

 significant association in mothers and daughters except for gross summer rate. 

 In gross annual rate and net annual rate there was a higher correlation between 

 mothers and daughters than was observed for any seasonal periods. 



When gross egg production was studied bj' periods, it was found that in all 

 cases the regression of daughters on mothers was nbn-linear. The association 

 between mothers and daughters was closest during the winter and spring periods. 

 The data do not suggest the possibility of accurate prediction of the character of 

 daughters that may be produced by selected mothers. High production is most 

 difficult to maintain in summer and fall, and there is lower correlation between 

 the production of mothers and daughters in these two periods than was observed 

 m winter or spring. 



