1 62 Bulletin 66. 



uniform speed. The conditions of power and speed were as favor- 

 able for all of the machines as it was possible to make them. The 

 single turbine machine, the Sharpies' Standard Russian, received 

 the steam at a boiler pressure of from 60-80 pounds per square 

 inch which was reduced to 40 pounds before entering the turbine. 



The machines were run by the students in the Dairy Course, 

 under the direct supervision of Mr. Dewitt Goodrich, the instructor 

 in butter making, each student working in turn upon each ma- 

 chine. None of the tests here reported were made until after the 

 school had been in session for four weeks, and the students had 

 attained considerable skill in handling the separators. The school 

 closed on March 21st and the tests that were made after that date 

 were all made by one person, Mr. W. D. Saunders, a man thor- 

 oughly familiar with the handling of separators. Several runs 

 were made by him from each separator, with the exception of the 

 De Laval Alpha No. i which was returned immediately after the 

 .close of the school. It will be seen that the runs made by Mr. 

 Saunders agreed in their results very closely with those made 

 during the session of the school. 



The measure of the efficiency of a cream separator is the 

 amount of fat left in the skim milk, but all of the separators left 

 only small amounts and in order to make the tests as accurate as 

 possible the determination of fat in the skim milk was made by 

 the absolute gravimetric method by the Assistant Chemist of the 

 station, Mr. G. W. Cavanaugh. Where skim milk contains only 

 .1 or .2 per cent, of fat, it is so difficult to read the amount upon 

 the neck of the test bottle that we have entirely abandoned the 

 use of the Babcock Test in work of this sort where it is necessary 

 to distinguish slight differences. The Babcock Test is no less 

 useful in checking the work of separators, for the creamery man 

 should run his separator so that only a scarcely perceptible amount 

 of fat appears in the neck of the test bottle. It is a common 

 thing in testing skim milk with a Babcock tester, to find so small 

 an amount of fat that it will not reach clear across the neck of 

 the test bottle but will adhere to one side in the form of a small 

 drop. It is not uncommon to find such tests reported as ".05 per 

 cent.," ".01 per cent.," "mere trace," "globule size of a jnn 

 head, etc." We have found in a large number of such tests that 



