30 



again shaking a gradual change is observed, and the alcoholic layer becomes dis- 

 tinctly yellow and the bisulphid solution of the fats quite colorless, or retaining only 

 the pale yellow color due to the natural coloring matter of the butter. 



3. Test for saffron, curcuma, carotin, rocou (Chevallier and Baudrimont). 



Treat with warm alcohol and test extract with subacetate of lead, an orange color 

 indicates the presence of saffron; treat with alkali and a brown color indicates cur- 

 cuma, and a green color, carotin; treat with concentrated sulphuric acid and a red- 

 brown color followed by blue indicates rocou. 



4. Test for egg yellow. 



Three hundred grams of fat are melted at 50 C., 150 cc of a 2 per cent solution of 

 sodium chlorid added, and the mixture kept at 50 C. for two hours. The mixture 

 is then well cooled and the aqueous solution separated and repeatedly passed through 

 a filter until as clear as possible. A perfectly clear and colorless filtrate, however, 

 indicates the absence of egg yolk. On adding an equal bulk of concentrated hydro- 

 chloric acid to a filtrate which is persistently turbid, a clear solution is obtained 

 again, becoming turbid on heating. Should egg yolk be present a further test is to 

 acidify 10 cc of the salt solution with 1 cc of a 1 per cent sulphuric-acid solution, 

 heat to boiling for a short time, cool, and shake with 2 cc of ether. The latter dis- 

 solves the coloring matters of the egg yolk should any be present, and the ethereal 

 layer is colored faint yellow. Consult an extended article on this subject by Dr. G. 

 Fendler. & 



C. BUTTER, FATS, ETC. (RANCID). 



If butter has become rancid the acid water from the petroleum ether extract as 

 obtained when testing fresh butter should be examined for reduced coal-tar dyes by 

 the method given below under milk (sour). 



D. MILK (FRESH), 

 a. LEACH'S METHOD. c 



The author uses the generic name of "anilin orange" as applied to all the coal- 

 tar dyes of the classes of diazo compounds of anilin, thus including single or mixed 

 dyes under this head. About 150 cc of the milk are curdled by the aid of heat and 

 acetic acid, preferably in a porcelain casserole over a Bunsen flame. By the aid of a 

 stirring rod the curd can nearly always be gathered into one mass, which is much the 

 easiest method of separation, the whey being simply poured off. If, however, the 

 curd is too finely divided in the whey the separation is effected by straining through 

 a sieve or colander. All of the annatto or of the anilin orange present in the milk 

 treated would be found in the curd and part of the caramel. The curd pressed free 

 from adhering liquid is picked apart, if necessary, and shaken with ether in a corked 

 flask, in which it is allowed to soak for several hours or until all the fat has been 

 extracted, and with it the annatto. If the milk is uncolored or has been colored 

 with annatto, on pouring off the ether the curd should be left perfectly white. If, on 

 the other hand, anilin orange or caramel has been used, after pouring off the ether 

 the curd will be colored more or less deeply, depending on the amount of color 

 employed. In other words, of the three colors, annatto, caramel, and anilin orange, 

 the annatto only is extracted by the ether. If caramel has been used, the curd will 

 have a brown color at this stage; if anilin orange, the color of the curd will be a 

 more or less bright orange. 



Diet, des alterations et falsification des substances alimentaires, etc., 7th ed., p. 236. 

 &Zts. Nahr. Genussm., 1903, 971. 

 cj. Chem. Soc., 1900, 207. 



