1904.] 



Testing Milk and Cream. 



37i 



have made arrangements with the committee of the National 

 Physical Laboratory for the examination and verification of the 

 pipettes, measuring glasses, and test bottles used. It has been 

 pointed out that there is some liability to error in reading the 

 percentage of fat, depending on the point which is taken in 

 the curved surface of the fat in the neck of the test bottle. If 

 the lowest point of the curved surface is read, the amount 

 of butter-fat is underestimated ; if the upper surface is read, 

 the estimate is too great ; the error involved in this latter 

 case is, however, at least twice as large as that caused by 

 reading the lowest point of the curve. Thus, if the estimate 

 from the lower reading is 3*30 per cent, and that from the 

 upper reading 3*40 per cent, the true reading would be about 

 3*33 per cent It has been found by experiment, however > 

 that in consequence of the shortness of time during which the 

 centrifugal motion is maintained in the Babcock apparatus the 

 total quantity of fat is not extracted ; and, as a rough com- 

 pensation for this, the reading of the top edge of the fat may 

 be taken. 



With the tubes in use in the Lister-Gerber methods the case 

 is somewhat different. The extraction of fat is, it is believed, 

 more nearly complete, and the necessity for any allowance is 

 in consequence less. Thus the bottom surface may be taken, 

 though some experienced observers allow for the small under- 

 estimate thereby caused by reading a point slightly above 

 the lowest point of the curve. For these reasons, therefore, 

 it is considered that for the Babcock test the upper edge should 

 be read, while for Lister-Gerber tests the lowest point of the 

 curve may safely be taken. 



Some investigations, have recently been carried out by the 

 United States Department of Agriculture in connection with 

 the above point, particularly as regards testing of cream. The 

 system in vogue in some districts of America of paying for 

 cream delivered at a central creamery on the basis of butter-fat, 

 as ascertained by the Babcock method, renders the accurate 

 manipulation of the test a matter of vital importance. Many 

 of the creameries are stated to be gradually departing from the 

 rules for operating this test, as experience seems to show that 

 they obtain less butter from 100 lb. of butter-fat from cream 



