120 PROVISIONAL METHODS FOR ANALYSIS OF FOODS. 



8. DETERMINATION OF VEGETABLE COLORS. 



A great many tests for vegetable colors are given, depending largely on color reac- 

 tions with different reagents, but these must be used with very great discrimination, 

 as they depend very largely on a fine judgment of shades of colors which many 

 eyes are not able to distinguish. 



A great deal of work has been done on detection of vegetable colors, a but only in 

 a very few cases are the reactions specific enough to be decisive. 



9. DETECTION OF TjRMERic. b 



Extract the color with alcohol. Dip a piece of filter paper into this tincture and 

 dry at 100 C. 



Then moisten in a weak solution of boric acid to which a few drops of hydrochloric 

 acid have been added. On drying this, a cherry red color will be developed in the 

 presence of turmeric which is characteristic. 



10. DETECTION OF CARAMEL. 

 AMTHOR TEST. C 



Ten cubic centimeters of the solution to be tested are put into a high, narrow glass 

 with perpendicular sides, as, for example, a small bottle; add from 30 to 50 cc of 

 paraldehyde, depending on the intensity of the coloring, and enough absolute alco- 

 hol to make the solutions mix. In the presence of caramel a brownish yellow to 

 dark-brown precipitate will collect in the bottom of the glass. Decant the liquor, 

 wash once with absolute alcohol, dissolve in small amount of hot water, and filter. 

 The color of this will give some idea as to the amount of caramel present. 



It is not allowable to concentrate a solution by evaporation on a steam bath, as 

 caramel may be formed; if it is necessary to concentrate it must be done over sul- 

 phuric acid or at diminished pressure. 



In order to further identify the color it is poured into a freshly prepared solution 

 of phenylhydrazin (2 parts phenylhydrazin-hydrochlorid, 3 parts sodium acetate, 

 and 20 parts of water). The presence of a considerable quantity of caramel gives a 

 dark-brown precipitate in the cold, which is hastened by heating a little. 



In the case of a very small amount it takes some hours for it to collect. 



11. DETECTION OF COCHINEAL. 



Cochineal is used to a certain extent as a coloring matter in foods, and a very satis- 

 factory test for it is that given in Girard and Dupre. d Dissolve the food product in 

 water, filtering if necessary. Acidulate with hydrochloric acid and extract with 

 amyl alcohol, which becomes colored more or less yellow or orange, depending on 

 the quantity of cochineal present. Separate the amyl alcohol and wash until neu- 

 tral. Then separate into two portions; to the first add drop by drop a very dilute 

 solution of uranium acetate, shaking thoroughly after each addition. In the pres- 

 ence of cochineal a characteristic emerald-green color is produced. 6 



To the secor'l portion add a drop or so of ammonia, and in presence of cochineal 

 a violet coloration results. This, however, is not so sensitive to very small amounts 

 as the first tests, and many fruit colors give tests hardly to be distinguished. 



Cochineal carmine is liable to contain tin, as it is often a tin lake, although alum 

 is also used. It is also liable to adulteration with lead compounds. 



Girard and Dupre 1 , Analyse des Matures Alimentaires, etc., 580-581, also 169; A. W. Blythe, Foods, 

 their Comp. and Anal., p. 91-109; Allen Com. Org. Anal., Vol. Ill, Pt. I; E. Brucher, Fals. Subst 

 Alim., p. 162; W. Lenz, Ztschr. anal. Chem., 1885, 24, 285. 



i> Allen, Com. Org. Anal., Vol. Ill, Pt. I, p. 359; U.S.Dept.of Agr., Div.of Chem., Bui. 51, p. 131. 



c Ztsch. anal. Chem., 1885, 24, 30; Borgemann, Anal, des Weines., p. 98. 



d Analyse des Matieres Alimentaires, etc., p. 580. 



The writer has tested this reaction on a number of amyl alcohol extracts from fruits, and in no 

 case was there any chance of mistake in the reaction. Most fruits give a brown color, while black- 

 berries and currants give a bluish color. 



