TESTING 



247 



Scoville 1 also found that there is very little difference 

 in the amount of fat in the cream when sampled in- 

 dividually or compositely. The following table shows the 

 results of his work. Each composite sample was composed 

 of portions taken from two to six cans of cream : 



TABLE XXXIII A COMPARISON OF INDIVIDUAL AND COM- 

 POSITE SAMPLING OP CREAM 



This table shows a difference of 13.79 pounds of fat 

 during the twenty days or a little over .4 per cent of the 

 entire amount of fat in favor of the patrons. This is 

 the opposite of what Lee and Hepburn 2 found, for, 

 according to Lee's readings, the composite fell below the 

 daily test .27 per cent and, according to Hepburn, the 

 composite samples showed .16 per cent less fat than in the 

 daily sample. 



The composite sampling of cream seems to be as 

 accurate as testing the daily or individual samples, and it 

 is evidently as satisfactory as the composite method of 

 sampling whole milk. Again, the time-saving feature of 

 composite testing should be emphasized. It is true that 

 in many creameries it would not be possible to keep good 

 daily records without a daily test of each patron's cream, 



1 Scoville, R. I., Some Factors Affecting the Over-run in 

 Creameries, A Thesis in Cornell Univ. Library, p. 21, 1916. 



2 Lee, Carl E., and Hepburn, N. W., Comparison of Methods 

 of Sampling Cream for Testing, 111. Agri. Exp. Sta., Bui. 153, 

 p. 574, 1912. 



