Tests of Cream Separators. 695 



tor, although this undoubtedly is usually the most important 



factor. 



Hajward, in some tests made in factories in Pennsylvania,* 



found a very wide variation in the efficiency of separators, and tests 



reported from different experiment stations have also shown large 



variations in the efficiency of separators of the same make. An 



effort was, therefore, made to determine so far as possible what 



these variations are in actual creamery practice, both in regard to 



the machines themselves and the skill of the persons operating 

 them. 



Accordingly, during the summer, occasion has been taken to visit 

 a considerable number of factories in the State located within con- 

 venient distances of Ithaca. In all, 22 factories and three private 

 dairies have been visited and 30 machines of four different makes 

 and nine different sizes and styles have been tested. 



These tests were made on various dates between the middle of 

 July and early in October, at a time when the flow of milk had 

 largely shrunken and when the cows were harassed by flies and 

 insects in bare pastures ; an unfavorable season of the year for pro- 

 ducing milk of easy separation. The results of these tests are shown 

 in Tables YII to X, inclusive. 



The different machines represented were the Alexandra Jumbo, 

 DeLaval, Sharpies and United States. JS^o factories using the 

 Danish Weston or Reid's Improved Danish were found in the terri- 

 tory visited, and it was not found possible in the limited time at our 

 disposal to visit the localities in the State where these machines are 

 used, but it is hoped that we may be able to do so at some future time. 



In these tables we have included not only the average speed of 

 bowl and average temperature of milk but the range of both, as it 

 has been noticed that uniformity, particularly of speed, is an im- 

 portant factor in clean separation. In all the tests the machines 

 were entirely in the hands of the factory operators and were run by 

 them in their ordinary manner. 



Equal portions of the skim milk were taken from the skim milk 

 outlet at intervals of 10 or 15 minutes, according to the amount of 

 milk separated, and from these mixed together a sample was drawn 

 for analysis. The determinations of fat in the skim milk were 

 made in skim milk Babcock bottles in all tests that were made pre- 

 vious to September 14th. In the tests made on and after that date 



* Report of the Peunaylyania Agricultural Experiment Station, 1894, p. 33. 



