ANIMAL PRODUCTION". 675 



lish a definite size and shape for each variety of poultry, it is suggested that in 

 judging poultry the measurements of the body should be taken into considera- 

 tion. A list of the most desirable measurements is given. 



The language of domestic fowls, B. Carpentek (Country Life [London], 

 28 {1910), Xo. 714, pp. 368, 369).— A study of the psychical life of fowls, with a 

 list of 23 different notes or cries of fowls and their probable meaning. 



A biometrical study of egg production in the domestic fowl. — II, Seasonal 

 distribiition of egg production, R. Pearl and F. M. Surface ( U. S. Dept. Agr., 

 Bur. Aniiii. Indus. Bui. 110. pt. 2, pp. 81-170, figs. 30). — A continuation of a 

 previous study of variation in the total annual egg production at the Maine 

 Station (E. S. R., 21, p. 271). 



The purpose of the present paper is to analyze the seasonal distribution, the 

 calendar month being taken as a provisional unit as it is not more open to 

 criticism than any other time unit. The same data were used as in Part I, and 

 aside from the following exceptions the biometrical methods used were in 

 general the same. The constants for the monthly egg production were calcu- 

 lated directly from the ungrouped frequency distribution as given in the ap- 

 pendix, and the raw second moment instead of Sheppard's correction was used 

 for calculating the standard deviation. Each month's production for the given 

 year was weighted with the number of birds which made the record. The 

 months in which the egg production was abnormal were not included. A brief 

 discussion of seasonal distribution has been previously noted (E. S. R., 21, 

 p. 273). 



Among other results obtained are the following: 



" The month of maximum productivity varied in the experiments furnishing 

 the present data with the method of housing. In a closed, warmed house the 

 month of maximum production was April ; in a curtain-front house it was 

 March. 



" The greatest relative variability in egg production is at the beginning of 

 the laying year (month of November). The month of lowest variability, both 

 absolute and relative, is April. 



" The laying year may be divided into four natural periods or cycles with 

 reference to egg production. The first of these periods (roughly November 1 

 to March 1) is the winter period, wherein egg production is essentially a non- 

 natural (i. e., forced or stimulated) process. The second or spring period 

 (roughly March 1 to June 1) is the natural laying period of the domestic fowl 

 in its normal reproductive cycle. The third (roughly June 1 to September 1) 

 and fourth (roughly September 1 to October 31) periods are not sharply sepa- 

 rated from one another. The summer egg production represents in part a 

 natural continuance of the normal breeding season (rearing of a second brood 

 by the wild Gall us) and in part a stimulated process. This period is terminated 

 by the molt, which is the characteristic feature of the fourth period." 



" So far as there was any change whatever in variability in monthly egg 

 production during the period when selective breeding was practiced, this change 

 was not in the direction of a reduction as a result of the selection, but, on the 

 contrary, there was an actual increase in variability in all but one month of the 

 year, and here the platted variability line did not sensibly deviate from the 

 horizontal. 



" The present statistics show no bad effect on egg production in the winter 

 months (November to March) of keeping birds in large and crowded fiocks (up 

 to the limits included in the present study). On the other hand, overcrowding 

 tends distinctly to lower summer (and to a smaller extent spring) egg produc- 

 tion. It is chiefly as a result of this effect on summer production that the mean 

 annual production is lower in the large flocks. 



