88 EXPERIMENT STATION RECORD. 



"In regard to the tests as a -svhole, the jiercentages of fat are cousiderably higher 

 than those found in the machines nsed at the station, and it will he seen also that 

 inmost of the different kinds of machines there is quite a large variatiou between 

 the highest and lowest percentage of fat, in every case amounting to 100 per cent 

 and in most cases to considerably more. 



"In the case of all of the various makes, except the Jumbo, at least one of the 

 machines tested did what is called 'practically clean skimming; ' that is, the per- 

 centage of fat in the skim milk was 0.1 per cent or less. In the case of the machines 

 where a greater percentage was left in the skim milk, in many cases it was evidently 

 due to carelessness of the operator, but in other cases it seemed to be some inherent 

 cxuality of the machine. It would seem, therefore, that since it is ijossible that 

 machines of the various makes that will do perfect work can be made that it is due 

 the operator to demand from the manufacturer a guarantee of such perfect work." 



Results of tests at various stations. — The data are summarized for the 

 tests of separators which have been reported by the Vermont, Penn- 

 sylvania, Wisconsin, Iowa, and Xew York Cornell stations, and the 

 following- average is given of the percentage of fat found in the skim 

 milk from the different separators and in different series of trials. 



Fat i)i skim milk from different separators. 



liind of machine. 



Accumulator 



Alexandra Jumbo 



Columbia 



Danish Weston. . . 



De Laval 



Sharpies 



United States . . . . 

 Victoria 



"The results of all of these trials show that it is possible to separate the cream 

 from milk with a loss of not more than 0.1 per cent of fat iu the skim milk. 



" That in all probability there is nearly as much difference in efficiency of separa- 

 tion between different machines of the same make as there is between the different 

 makes themselves." 



Tests of dairy apparatus, J. L. Hills ( Vermont 8ta. Rpt. 1894, 

 pj). 151-160). — In connection with the dairy school tests were made of 

 the relative efficiency, power con.sumption, and steam consuuiption of the 

 De Laval Alpha Turbine, De Laval Acme, Jumbo, United States No. 3, 

 United States No. 5, De Laval Alpha No. 2, and De Laval Alpha No. 3 

 separators. There were 4 hand and 5 i)ower separators. The data for 

 these trials are tabulated, together with the losses in churning. The 

 average percentages of fat iu the skim milk were as follows: 



Per cent. 



De Laval Alpha Turbine 0. 05 



De Laval Acme * 10 



Jumbo 21 



United .States No. 1 B 08 



Reid Improved Danish 07 



United States No. 3 07 " 



