222 AGRICULTURE OF MAINE. 



paper is now in press having the title *The Fitting of Logarith- 

 mic Curves by the Method of Moments." 



When the tables spoken of above are completed and published 

 in bulletin form it will be possible for any farmer who keeps a 

 record of the milk production of his heifers at their first lactation 

 to predict, with an average error of rather less than two per cent, 

 what the production of the same cow will be when she is seven 

 years old. Furthermore, it will be possible for a dairyman to 

 give each one of his cows an absolute rating in comparison with 

 advanced registr}' animals of the same breed at any given age. 

 If he will keep a milk record, he can with the help of these 

 tables say whether or not a particular cow is better or worse, 

 and by what proportion, than the average of advanced registry 

 cows of the same age. 



In addition to this work on the relation of milk flow to age 

 a comprehensive study has been made on variation in milk flow 

 and butter fat percentage in Ayrshire cattle. For this purpose 

 we have made use of records kindly furnished us by the Scottish 

 Milk Records Society giving exact data on the milk production 

 of large numbers of pure-bred Ayrshire cows. This study has 

 been completed and is now being written up for publication and 

 will be issued sometime during the coming year. 



2. THE STUDY OF INBREEDING IN DAIRY CATTLE. 



The work on this line during the last year has consisted in 

 the first place of further extending the theoretical consideration 

 of the problems of inbreeding, in preparation for final analysis 

 of the results of our pedigree studies on the Jersey and Hol- 

 stein breeds. We have now got the theoretical mathematical 

 foundation laid for these studies and are ready to go ahead with 

 the actual analysis in the two breeds of cattle mentioned. Dur- 

 ing the year a paper has been published which gives a method 

 of further measuring the degree of inbreeding in a particular 

 case through the medium of relationship coefficients. We have 

 now reached the point in the analysis where it is possible to say 

 in any given pedigree exactly what proportion of the observed 

 inbreeding results from the relationship of the sire and the dam 

 and what proportion results from the relationship from earlier 

 ancestral generations. Some of the results which come from 

 this study are very interesting and at first sight somewhat para- 



