TESTING 



241 



than in a fourteen-day composite sample. However, it 

 should be noted that these are the results of only two series 

 of experiments. It may also appear that there is a loss 

 of fat to the farmer by the composite method of sampling 

 milk. On the other hand, it should be noted that all the 

 tests in both tables are within the limit of error in reading. 

 The most interesting item to be noticed in these tables is 

 the close comparison of the different methods of composite 

 sampling. 



Kent ^ compared the daily and the composite testing 

 for a period of thirteen days. There were fifty-two pa- 

 trons. Three samples were taken as follows from the 

 milk delivered by each patron : a daily sample, a composite 

 sample for which the daily portions were obtained with a 

 small dipper, and a composite sample for which the daily 

 portions were secured with a Scoville sampling tube. 

 The average daily test was 4.01 per cent, the dipper 



' Kent, F. L., Testing Milk and Cream, Oregon Agri. Exp. 

 Sta., Bui. 70, 1902. 



