ANNUAL MOLT AND EGG PRODUCTION 77 



A negative coefficient of correlation of .2399 ± .0209 is not significant, for 

 the amount of variability in length of molt period associated with intensity 

 was equal to the .square of .2399 or .0576. 



The correlation ratio of length of molt to intensity was .3339 and differed 

 perceptibly from the simple correlation coefficient. (Correlation ratio) ^ — 

 r^ = .0539 ± .0097. This difference was 5.56 times its probable error and is 

 certainly significant. By this test the relationship between intensity and 

 length of molt period was non-linear, and the correlation ratio of .3339, there- 

 fore, expresses the true relationship and shows that intensity and length of 

 molt period were significantly correlated. 



The regression line for length of molt period on intensity, by the X^ method, 

 showed no probability of being a straight line, which further confirms non- 

 linear relation.ship. 



4. Duration of Winter Pause versus Length of Molt Period. 



The duration of winter pause is probably governed very largely by en- 

 vironmental influences. The possibility exists also that length of molt period 

 may be influenced by simUar reactions to environment. Records are available 

 on 926 birds with respect to ■winter pause and length of molt. In this popu- 

 lation 393 individuals showed no cessation of production as great as four 

 days during the winter season and were, therefore, classed as non-pause 

 birds. These non-pause birds showed a mean molt period of 71.76 days. 

 In correlation studies 533 pause birds were classified with respect to length 

 of pause and duration of molt, using class intervals of 10 days in each case. 

 The constants obtained are as follo^'s: 



Number of birds - 533 



Mean length of molt period — days — 78.32 



Molt period standard deviation — days.._ ±31.15 



Mean winter pause — days 23.73 



Winter pause standard deviation — days ±15.62 



Coefficient of correlation ™ - -... +.2432 ± .0275 



The mean length of the molt j>eriod for the winter pause birds was 78.32 

 days compared with 71.76 days for the non-pause birds. It seems probable, 

 therefore, that the winter pause group was made up of 'birds having greater 

 susceptibility to adverse environmental influences insofar as they affect egg 

 production both in winter and during the molting period. 



The winter pause period of non-production had an average duration of 

 23.73 days, with an extremely high variability as shown by the standard devi- 

 ation. Its actual range extended from 4 to 113 days. 



The coefficient of correlation between winter pause duration and length of 

 molt period was +.2432 ± .0275. This constant is below the minimimi of .30 

 adopted as significant. From the standpoint of egg production, pullets that 

 exhibited winter pause probably were undesirably susceptible to adverse en- 

 vironmental influences compared with non-pause birds. 



The regression of duration of molt upon length of winter pause was strict- 

 ly linear. By the X- method the probability of linearity stood at .7468. The 

 simple correlation coefficient is therefore a true measure of association. 



