70 TECHNICAL BULLETIN 7 



16. Correlation Between Total Days Broody and December Rate — Pullet 

 Year. 



In this study the same group of 1945 birds both broody and non-broody tliat 

 were studied in relation of times broody to December rate (section I) is con- 

 sidered. It is important to know wiiich is tiie more im^jortant from the stand- 

 point of rate, the number of broody periods or the actual number of days 

 spent in broodiness calculated so as to avoid winter pause and fall molt. 

 Constants calculated on this group of birds follow: — 



Number of birds 191-5 



Mean total days broody 23.20 



Total days broody standard deviation . . . ±: 27.07 

 Mean December rate ...... 59.60 



December rate standard deviation . . . ±20.40 



Coefficient of correlation -f .0529 ± .0153 



The standard deviation in total days broody exceeds the mean total days 

 broody liecause of the large percentage of non-broody birds in the grouj) 

 studied. 



The coefficient of correlation agrees rather closely with the figure gi\eii in 

 section 1 where times broody and December rate are considered. E\identl} 

 broodiness may be measured either by periods or by total days. The degree 

 of correlation is slight, and it is really a false correlation because based upon 

 a mixed population — broody and non-broody. 



17. Correlation Between Total Days Broody and December Rate for Broody 

 Birds Alone — Ptdlet Year. 



The relation between degree of broodiness, as measured liy total days of 

 non-production associated with broodiness during the pullet year, and Decem- 

 ber rate may be determined by using only the birds that went broody the first 

 year. Such a determination was made for the same group of 949 birds that 

 was considered in section 2. The following are the constants: — 



Number of birds ....... 949 



Mean total days broody 42.84 



Total days broody standard deviation . . . ±27.38 

 Mean December rate ...... 6'1.24 



December rate standard deviation . . . ±20.11 



Coefficient of correlation —.0002 ± .0219 



Regression broodiness on rate .... — .0003 



Regression rate on broodiness ' . . . • — .0001 



Tliat degree of broodiness and December rate are entirely independent is 

 shown by the above coefficient of correlation which is practically zero. TJiis 

 is rather conclusive evidence that December intensity bears no relation to tlie 

 presence or absence of the broody trait. 



18. Correlation Between Total Days Broody and Winter Rate — Pullet Year. 



A total of 2221 birds studied in section 6 are included in this study to dis- 

 cover tlie deuree of dependence or independence between total days broody 

 and winter rate. The constants calculated follow: — 



