The amount of butter- milk, (this was determined by sub- 

 tracting from the weight of cream the weight of the drained 

 butter, this is the most accurate method, for it is impossible to 

 draw off all the butter-milk and weigh it, as a portion will be 

 left in the butter and must be worked or washed out,) the % of 

 fat in the butter-milk, the % of water, fat, and casein in the 

 butter, the amount of butter, its quality as determined by ex- 

 perts who knew nothing of the identity of the samples, the 

 efflciency of the systems and the pounds of milk required per 

 pound of butter. 



The committee of experts who examined the butter con- 

 sisted of C. H. Waterhouse President of the New England 

 Creamers Association, and H. W. Sadd of Connecticut, 



The butter was marked in the following scale for perfect 

 butter: Flavor, 50, Grain, 25, Color, 15, Salt 10, 



Table II gives the full marking of each sample. 



Total, 100 89 85 86 95 86 



The efficiency of the various sytems is based upon the per 

 cent of the total fat in the milk which is found in the butter, 

 (See rule 8 of "Rules of Test") this was taken because it is the 

 only true test. If one system gives a cream which churns less 

 thoroughly than the cream given by another system, it is just 

 as much to the discredit of the former as though the fat was 

 left in the skim-milk, consequently if we are to arrive at a know- 

 ledge of the value of any given system we must follow the origi- 

 nal fat in the milk through to the lump of butter, and find the 

 per cent of this fat that is recovered. 



Of the fat not recovered in the butter, the greater the part 

 of it that is left in the skim-milk, and the less the part left ia 

 the butter-milk the better, because of the fact that skim-milk is 

 in better shape to feed to calves than butter.milk. First get as 



7 



