EXPERIMENT STATION BULLETINS. 



385 



TABLE IV.— Concluded. 

 The Butter Scores. 



A very marked instance of scores which are not all in agreement is shown 

 in the case of lots I, III and V, all of which were made from the same lot of 

 cream, I being churned sweet after pasteurizing, III, sweet without pasteur- 

 izing, and V pasteurized and then ripened with a pure starter. In the first 

 February score it will be noted that V was given a very low score as com- 

 pared with I, while in the second score the numerical values are equal al- 

 though the flavor of these two samples was very different, I being quite 

 light and mild, while V was rather high in flavor. In the one case, V was scored 

 down for this high flavor which to the second scorer did not seem objection- 

 able, although admittedly present. These butters were more than a year 

 old at this time. 



Score. 

 Foster. . 

 Rabild.. 



In regard to the flavors which characterize the different butters the agree- 

 ment is very much closer, although even here some surprising differences 

 are noted. This is due, in part, no doubt, to the sequence in which the butter 

 happened to come at the time of scoring; any particularly good or bad butter 

 influencing to some extent the score of the butter which immediately followed 

 it. From our own experience and the contradictory evidence obtained at 

 49 



