246 THE BOOK OF BUTTER 



These tables show that there is very little difference in 

 the test of the cream, whether thoroughly agitated, 

 partially mixed or not stirred at all. The latter table 

 also shows that there is very little difference in the test 

 of the cream when the sampling is made with a McKay 

 sampling tube, which is supposed to obtain very nearly 

 an aliquot sample, and the little, easily handled, inex- 

 pensive sampling dipper. 



The question of the advisability of composite sampling 

 of cream is still unsettled. Hunziker ^ thinks that it 

 is not satisfactory largely from the standpoint of cream- 

 ery records, because the amount of fat in the cream should 

 be known every day. From the viewpoint of accuracy, 

 this method of sampling cannot be criticized if made prop- 

 erly. The sample jar or bottle must be air-tight. 

 Ground glass stoppers are absolutely essential. Lee 

 and Hepburn ' have probably done the most extensive 

 work on the composite sampling of cream. Their com- 

 parisons of daily and composite tests were on the cream 

 of seventy-seven patrons, and their work covered a period 

 of one year. Their results show a marked tendency for 

 variation between composite and individual sampling 

 for short periods, but that the distribution of this varia- 

 tion is such as to occasion very small amounts of differ- 

 ence when figured on seasonal periods. Their figures 

 indicate that composite samples tend to test slightly 

 higher than individual ones in summer, and lower in 

 winter. The yearly average shows practically no differ- 

 ence between the composite and individual samples. 



' Hunziker, O. F., Testing Cream for Butter Fat, Purdue 

 Univ. Agr. Exp. Sta., Bui. 145, p. 552, 1910. 



'' Lee, C. B., and Hepburn, N. W., Comparison of Methods of 

 Sampling Cream for Testing, III. Agri. Exp. Sta., Bui. 153, pp. 547 

 and 574, 1912. 



