FOOD-COLORING SUBSTANCES. 19 



the residue dissolved iii water, and the constituent colors identified 

 in portions of this solution. A portion of the slightly acidified solu- 

 tion may be warmed, and a few decigrams of urea and then one or 

 two drops of sodium nitrite solution added. Indigo Carmme is con- 

 verted into the pale yellow Isatinsulphonate, while the other dyes 

 are but little affected. The Isatin compound is not ordinarily pres- 

 ent m sufficient concentration to tmt the solution, but it differs from 

 Tartrazin also in being much less readily extracted by amyl alcohol 

 from strong acid solution (less than one-half from four-normal acid). 

 Amaranth is much more quickly attacked hy most reducing agents 

 than Tartrazin, and may be destroyed in mixtures containing Tar- 

 trazin by treating the neutral or faintly acid solution at room tem- 

 perature with sodium hydrosulphite. (In the presence of sodium 

 carbonate the reduction of Tartrazm takes place still, more slowly.) 

 The reagent should be added very carefully, either in dilute solution 

 or as the powder, sufficient time being allowed after each addition for 

 the reduction to take place. When the color shows that the Ama- 

 ranth has been completely destroyed, the mixture should be shaken 

 at once with air to oxidize the slight excess of hydrosulphite before it 

 can react further on the Tartrazin. To separate the Indigo Car- 

 mine another portion of the neutral or faintly acid solution may be 

 heated to boiling and hydrosulphite added very carefully, a few par- 

 ticles at a time, until all the dyes are reduced. On shakmg with air 

 the Indigo Carmine is ciuickiy re-formed. 



The fourth-normal acid solution (or the salt solution) containing 

 the Ponceau and Naphthol yellow S is treated with enough hydro- 

 chloric acid to make it about twice normal, and is shaken out a few 

 times with washed ethyl acetate.^ The yellow is removed from the 

 combined ethyl acetate extracts by shakhig out with water. It must 

 always be remembered that Naphthol yellow S is almost colorless in 

 strongly acid solutions, and its absence in v/ashings, etc., must never 

 be assumed until they liave been made alkaline. The Ponceau 3 R 

 is finally separated from the acid solution by shakuig the solution with 

 amyl alcohol, then washing out the dye from this extract with a few 

 small portions of water. 



^Vllen (with mixtures containing Orange I) the washings of the 

 ethyl acetate, which should contain only Naphthol yellow S, become 

 redder with alkalies, they should be combined, made fourth-normal 

 with liydrochloric acid, and the contaminating orange removed by 

 extracting with amyl alcohol gasoline mixture (1:1). Or they may 

 bo tr(!atcd with one-fifth their volume of concentrated hydrochloric 

 acid and the dyes extracted by shaking once with amyl alcohol. 

 From this solution the yellow may bo removed by washing with 

 several portions of 5 per cent salt solution. 



' UhC'I Instead of amyl acetate m suggested by LDoml^, U. S. Dopt. Agr., Tlnr. Cliom. Bui. No. I(i2, p. !>7. 



