SCIENCE AND PRACTICE OF DAIRYING. 



approximately estimated if the addition is so considerable that the per- 

 centage of ash in the milk is increased above its ordinary amount. 



Small quantities of benzoic acid are most easily and most certainly 

 detected by the following test (Meissl): 250 to 500 c.c. of milk are 

 rendered alkaline by the addition of a few drops of lime or baryta water, 

 evaporated down to about a fourth of its volume, stirred into a paste 

 with gypsum powder, pumice-stone powder, or sand, and then dried on the 

 water-bath. If condensed milk is to be investigated, 100 to 150 grams of 

 the milk may be treated directly with gypsum and a few drops of baryta- 

 water. The dry mass is then powdered, moistened with dilute sulphuric 

 acid, treated four times in the cold with about twice its volume of a 

 50-per-cent alcohol solution, which easily dissolves benzoic acid, and which 

 has little or no action on fat. The alcohol washings, which show an acid 

 reaction, and which contain in addition to benzoic acid, milk-sugar and 

 inorganic salts, are then mixed, neutralized with baryta - water, and 

 evaporated down to a small volume. This residue is rendered acid with 

 dilute sulphuric acid, and finally is shaken up with small quantities of 

 ether. On diluting the ether, benzoic acid is left behind in an almost 

 pure condition. If not pure, it only contains traces of fat or ash con- 

 stituents. For quantitative determination it is dried at 60 C. in the 

 desiccator, weighed, the benzoic acid is sublimed, and the residue is again 

 weighed. Sublimation is best effected on the water-bath, and is best 

 carried on in such a way that the small basin containing the substance 

 is covered with another basin of similar size, or with a watch-glass. The 

 sublimate on the little basin lying on the top may be used for qualitative 

 test, while the lower basin is heated uncovered for some time until all the 

 volatile substances are expelled. The qualitative reaction for benzoic 

 acid, which is the most striking, is its reaction with neutral iron chloride; 

 the substance dissolved in water must, however, be treated with a few 

 drops of sodium acetate. 



Boiled milk may be detected from unboiled milk, in addition to the 

 flavour test, by the ozone reaction, which unboiled milk gives but boiled 

 milk does not. Unboiled milk colours guaiacum tincture blue, boiled milk 

 does not. Potassium iodide starch-paper with oil of turpentine is quickly 

 coloured blue by unboiled milk. Boiled milk does not exhibit this reaction, 

 or at any rate no more quickly than the mixture itself becomes blue. The 

 detection of starch in milk offers no difficulty. If starch has been added 

 to cold milk, it settles on the milk being left standing, and can be easily 

 collected in the bottom of the vessel. In order to detect the presence of 

 boiled starch in milk, a large quantity of an iodine solution is necessary, 

 since a considerable quantity of iodine is required to saturate the albumin- 

 oids before the iodine reaction is exhibited. 



