POULTRY WORK. 200, 



statistics. The complete analysis of the statistics is an 

 extremely laborious task and will take considerable time for its 

 completion. It is gratifying to state, however, that the first part 

 of this work has been finished and is now in the hands of the 

 U. S. Government printer awaiting publication. This first 

 portion of the discussion of the records of egg production deals 

 with the annual records. That is to say, that unit of this part 

 of the investigation is the total production of the birds during 

 the first laying year. A brief summary of some of the more 

 significant results of the study of annual egg production which 

 have a practical bearing on the future policy of the Station in 

 its poultry work are included here. It is to be understood that 

 this summary is a direct compilation from the extended and 

 detailed paper by Doctors Raymond Pearl and F. M. Surface, 

 the biologists of the Station, setting forth the results which are 

 to be published as a bulletin by the Bureau of Animal Industry. 

 In such a summary the evidence on which apparently dogmatic 

 statements are made obviously cannot be given. 



The chief reason or indeed necessity for making a detailed 

 study of the annual egg production during the course of the 

 breeding experiment obviously lies in the following considera- 

 tions : The initial purpose of the experiment was to find a way 

 to breed poultry so that an increase in egg production might be 

 obtained. In order to get at this knowledge a specific method 

 of breeding was tried during a period of 9 years. The only 

 way in which it is possible to get any light at all as to whether 

 this particular method of breeding is the best or even a good 

 one to attain the desired end is to find out precisely what hap- 

 pened in egg production in each year covered by the experiment. 

 If the method of breeding is a good one for its desired purpose 

 then it would be expected that, barring accidents, in each suc- 

 cessive year of the experiment there ought to be an increase in 

 the average production of the flock. As a result of unavoid- 

 able accidents it might happen in any such experiment that in 

 a particular year the average production would drop below 

 that for the preceding year. Consequently in order to form any 

 just estimate of the value of the method of breeding it is neces- 

 sary to consider the results over the whole period of the investi- 

 gation together. That is to say, the general trend of the egg 

 production over the whole period must be taken account of. and 



