THE INVESTIGATION AND TESTING OF BUTTER. 199 



directly has been in my experience unsuccessful, as in washing the fat free 

 residue with water, sometimes more, sometimes less, of the nitrogenous 

 bodies is apt to be found in the solution. Naturally the uncertainty which 

 belongs to the number denoting the quantity of protein bodies will influ- 

 ence the number calculated by difference, which represents the quantity of 

 non-nitrogenous soluble organic bodies. 



If butter has to be tested for its percentage of preservatives, with the 

 exception of salt, or for the determination of foreign solids which have 

 been added for the purpose of increasing its weight, the following process 

 may be adopted if it be not desired to detect the presence of foreign fats. 



10 to 40 grams of the butter to be tested are melted in double or 

 three times the quantity of warm distilled water, a little alcohol is then 

 added, and the mass is stirred very slowly for about fifteen minutes at a 

 temperature just above the melting point of fat. It is then allowed to 

 stand still for some time, and the liquid lying below the fat, as well as the 

 residue, is submitted to chemical and microscopical investigation. Since it 

 is unnecessary to add to the special precautions to be taken in detecting 

 different substances, and since adulteration of butter with potato meal, 

 gypsum, water, glass, &c., only occurs very rarely, there is no necessity 

 to describe the methods for the detection of all possible and impossible 

 adulterants. If the butter be not adulterated, the liquid below the fat 

 becomes clear, or almost perfectly so, when it is warmed with soda lye, 

 added in slight excess. 



For the detection of the usual colouring agents, Hilger recommends the 

 following process: About one-half of the liquid which has been filtered 

 from the sediment lying at the bottom under the fat, and which has been 

 obtained by the method above described, is evaporated down to a fourth 

 part of its volume, and is then divided into three like portions, a, b, and c- 

 Portion (a) is decomposed with hydrochloric acid. If this be followed by 

 a yellow coloration it indicates the presence of binitrocresol or binitro- 

 naphthol. Portion (b) is decomposed with ammonia for the detection of any 

 turmeric colouring matter. Portion (c) is finally heated with some sugar 

 and hydrochloric acid. The appearance of a red colour points to the pre- 

 sence of saffron. The remaining half of the original solution is evaporated 

 to dryness, and the residue treated with concentrated sulphuric acid. If 

 annatto be present a blue colour is produced. For the detection of colours 

 derived from carrots or marigolds no reliable tests are known. Genuine 

 saffron should not colour petroleum ether, as has been asserted. 



The method for the determination of foreign fat in the butter has 

 been already described. 



