76 MASS. EXPERIMENT STATION BULLETIN 264 



the relationship between age at first egg and duration of molt period would 

 be of value to the poultry breeder, A total of 929 birds was used in these 

 studies. Class intervals of 10 were used for classifying the birds with respect 

 to age at first egg and duration of molt in the correlation table. The follow- 

 ing constants were obtained: 



Numiber of birds 929 



Mean length of molt period — days 75.60 



Molt period standard deviation — days ±30.38 



Mean age at first egg — days 191.50 



Age standard deviation— days - ±21.66 



Coefficient of correlation +.2039 ± .0212 



Correlation ra tio 2829 



Age at first egg ranged from 140 to 319 days in the above population. The 

 magnitude of its standard deviation also indicated marked variabilitj', which 

 was brought about both by inheritance and environment. 



The coefficient of correlation was of such magnitude as to be insignificant 

 by the criteria used in this report. The regression of molt duration upon age 

 at first egg was not linear. The regression line by the X^ method showed a 

 probability of only .0020 of being linear. The correlation coefficient, therefore, 

 fails to measure the correlation desired, and the correlation ratio must be 

 emtployed. 



The correlation ratio of molt duration to age at first egg amounted to 

 .2829 which is beJow the minimum set for significance. In other words, the 

 constant has an absolute value too small to be significant in selection. This 

 constant diifered from the correlation coefficient by 4.63 times its probable 

 error, and is therefore the true measure of association desired. 



3. Intensity versus Length of Molt Perw<l. 



A population of 935 with records of winter intensity and length of molt 

 period was available for study. There were seven individuals in this group 

 showing no molt period. Intensity was measured' by the mean size of winter 

 clutch as suggested by Hays and Sanborn (1927 a). The seven non-molting 

 birds had a mean clutch size of 5.02, while the 928 molting birds showed a 

 mean clutch size of 3.4. Here is a suggestion that non-molting birds tend to 

 be intense layers. The class interval for intensity was 1.0 and that for molt 

 10 days. The following constants were calculated: 



Number of birds ™. 928 



Mean length of molt period — days 75.53 



MoJt period standard deviation— days ±30.31 



Mean clutch size 3.40 



Clutch size standard deviation ±1.61 



Coefficient of correlation —.2399 -f .0209 



Correlation ratio „ 3339 



The mean intensity of these birds stood at a high level, but the variability 

 for intensity was extreme as shown by its standard deviation. The actual 

 range in clutch size extended from 1 to 15.9. 



