76 TECHNICAL BULLETIN T 



possibly be more pronounced in the l)roody population. The actual correla- 

 tion between the presence of broodiness and winter production above the mean 

 is siiown by the following table: — 



Winter Production I Broody Non-Broodv 



Coefficient of correlation — .02^4 = .0143 



The above coefficient is so small as to iie of no significance and it is less 

 than three times the magnitude of its probable error. The deduction nuist be 

 made from this study that broodiness and winter egg production are entirely 

 independent e\en though broody birds do lay at a sligiitly higher rate in 

 winter when they are laying. 



25. Correlation Between the Presence of Broodiness and Winter Production 

 above the Mean of Broodies and yon-hroodies combined — Pullet Year (Un- 

 improved Flock 1916). 



Coefficient of correlation — .3759 it .0318 



The above constant shows that broodiness l)ears a rather intimate correhi- 

 tion to high winter production in the 1916 flock. Such an assumption is based 

 on the conclusion that the individuals laying more eggs in winter than the 

 average of the flock (46.87 eggs) are high producers. Even though a small per- 

 centage of the 1916 flock reduced this winter record by being broody before 

 March first, broody birds appeared to carry intensity to a sufficient extent to 

 enable them to lay more eggs for the period than did the non-broody l>irds. 

 It is rather striking that the total population (section 24) should not exhibit 

 a con-stant similar to that for the 1916 flock. No doubt changes in early ma- 

 turity and winter pause have operated to modify winter production to a 

 greater extent than any possible lowering of intensity by the elimination of 

 broodiness has been responsible for. 



