48 



Some degree of association exists between low production and late hatching 

 but there is lack of consistency. A striking and consistent degree of relation- 

 ship is seen in table 2 between winter egg record and age at first egg. Winter 

 egg record and loO-day weight also show considerable dependence. There is 

 some evidence that the low producers are heavier at first egg than the high 

 producers. The average daily gain in weight increases as we advance down 

 the table to the heavy winter layers. The general deduction seems warranted 

 from table 2 that low winter egg records depend in part upon late hatching, 

 too great an age at first egg, light 150-day weight, and slow rate of gain in body 

 weight between 150 days old and age at first egg. In order to determine 

 specifically how important these various relations are it is necessary to resort 

 to the coefficient of correlation. 



Age at First Egg versus Winter Production. 



The degree of correlation between age at first egg and annual production in 

 Rhode Island Reds for a period of years was found to be — .4380zb.0134 (Hays 

 and Bennett 1923). The fact that winter egg yield is definitely ended March 

 first while annual egg record does not terminate until 364 days after a pullet 

 lays her first egg makes the correlation more intimate between age at first egg 

 and winter record than between age and annual production. 



The correlation coefficient between age at first egg and winter production 

 has been calculated on 803 pullets hatched in 1923 without regard to the differ- 

 ence in hatching date. Constants calculated from this study follow : 



Number of birds .... 



Mean age at first egg 



Age at first egg standard deviation 



Mean winter production . 



Winter production standard deviation 



Coefficient of correlation 



Mean age at first egg is 210.96 days. Standard deviation of age is 28.62 

 which exhibits the wide range in age at first egg. The mean winter production 

 is 44.46 eggs with a standard deviation of 23.04. Again winter fecundity 

 shows its extreme variability as would any trait dependent upon so many hered- 

 itary factors and environmental influences. 



A significant negative coefficient of correlation of .6061 ±.0151 appears. 

 Thus in six cases out of ten in these pullets hatched over a period of seventy 

 days, there is definite association between early age at first egg and high winter 

 egg record. In other words, the length of time that a pullet has opportunity 

 to lay previous to March first should be given very weighty consideration in 

 breeding for winter fecundity. 



To secure an exact figure on the degree of correlation between age at first 

 egg and winter fecundity it would be necessary to make hatching date con- 

 stant by studying only those pullets hatched at the same date. Such a study, 

 we believe, would reduce tlie number of individuals to such an extent that the 

 mathematical error of calculation would be inordinately large. Below are 

 presented the constants calculated on the 154 birds in the first three hatches. 

 The hatching date range is thus reduced to fourteen days. Constants are as 

 follows : 



Number of birds 



Mean age at first egg 



Age at first egg standard deviation 



Mean winter production 



Winter production standard deviation 



Coefficient of correlation . 



154 



211.58 



±39.80 



56.90 



±27.61 



-.6413±.0320 



When the range in hatching date is reduced from 70 days to 14 days the co- 

 efficient of correlation between age at first egg and winter production increases 

 from -.6061±.0151 to -.6413±.0320. This fact clearly proves that hatch- 



