196 SCIENCE AND PRACTICE OF DAIRYING. 



entirely absent, OK are only partly present and in very small 

 quantities. The proportion of the quantity of volatile to the non- 

 volatile fatty acids found in the fat tested is correspondingly deter- 

 mined by one or other of the methods. If this proportion were 

 invariable in butter-fat, it would be possible to detect the smallest 

 possible quantities of foreign fats in butter. Since, however, it 

 varies within comparatively wide limits, the case with regard to the 

 testing of butter is almost the same as with regard to the testing 

 of milk. Adulteration in small quantities is as difficult to detect in 

 this case as in the former. No doubt, under very special unfavour- 

 able circumstances, such as very rarely occur, butter may contain 

 20 to 25 per cent of margarine adulteration, and the adulteration 

 cannot be proved by investigation. On this account, in addition to 

 the determination of the quantity of volatile or non-volatile fatty 

 acids, a number of other tests for butter-fat have been applied. 

 Thus, for example, there is the determination of the specific gravity 

 of the fat at the boiling temperature of water with a margarimeter, 

 since it has been observed that most of the different kinds of fat 

 show a lower specific gravity than butter-fat. This method of 

 testing is, however, only valuable in the cases in which the mar- 

 garimeter shows a lower specific gravity in the fat investigated 

 than that of butter-fat, since various vegetable fats, such as earth- 

 nut oil, sesame oil, and poppy oil, have the same, or even a higher, 

 specific gravity, than pure butter-fat. It has further been recom- 

 mended to determine the coefficient of the fracture of the fat at 

 a certain temperature, by means of a refractometer, since it has 

 been found that pure butter-fat has a less high fracture coefficient 

 than most of the other kinds of fat. The fat should also be 

 tested in polarized light by means of a 75 linear enlargement, owing 

 to the fact that the fat from melted margarine, on cooling, assumes 

 a kind of crystalline structure, and exhibits characteristics in 

 polarization, which butter-fat does not show, even although it has 

 not been somewhat equally melted and again cooled. It is not 

 possible to refer to the many different proposals for the detection 

 of adulteration which have been made in addition to those above- 

 mentioned. 



In the testing of butter for the detection of substances which are not 

 fat, the centrifugal butter-tester of Lefeldt is useful, as it renders the 

 investigation easier, and points quickly to the discovery of suspicious 

 butter. Up till about 1870, it was not possible to distinguish butter-fat 



