17 



.", cc "i am\l alcohol are added. The tubes are thoroughly shaken and the liquids 

 allowed tn ,-eparate. If in any case an emulsion is formed, afew drops of ethyl alcohol 

 UK- added. Similar tests are made on the remaining portions of the sample without 

 tin- addition of acid or alkali. 



(6) Ten cubic centimeters of the sample are made alkaline with baryta water, 

 shaken with an equal volume of amyl alcohol and, after observing the colors of the 

 two layers as directed, acetic acid is added to a filtered portion of the amyl alcohol 



(c) To 50 cc of the wine in a beaker an equal amount of water and a few cubic centi- 

 meters of dilute hydrochloric acid are added. In this is placed a piece of white, 

 fat-free WIM.I (loth, about 10 cm square, and the solution is boiled from five to ten 

 minutes. The < loth is removed, washed in a stream of water, and after noting the 

 color the wo<>l is tr-at i-d with a 2 per cent ammonia solution. 



(d) l"\\ e nibic cent i meters of concentrated nitric acid are added to an equal amount 



of the wine in a test tube. 



(e) To 10 cc of the Cample in a test tube 5 cc of a neutral or slightly alkaline mix- 

 >f a 10 per cent potas.-ium alum solution and a 10 per cent sodium carbonate 



.-olution are added, the mixture is shaken and the precipitate formed is separated 

 by filteriii'.;. In like manner the -uine is te-ted witli 5 cc of a mixture of lOpercent 

 aluminum acetate solution made alkaline with 10 per cent sodium carbonate solution. 

 A port ion of the .-ample i- al-o treated \\ithalOperceiitsoliitionofmercuricchlorid. 



f the sample in a te-t tube are added 3 cc of a 10 per cent solution of 



I- i i hakeii and filtered. It has been found difficult at 



time- to match the color of the precipitate closely with the color chart and in such 



cases the coli.r ha.- 1 ,ate.l in general terms such as blue-gray or brown-gray. 



About <)(>"> u'ram of pulverized yellow oxid of mercury is added to 20 cc of 



the -ample and the mixture heated to boiling and poured through -a double filter. 



After lir-i noting the color of the filtrate a few drops of hydrochloric acid are added. 



(h) To each of two 10 re portions of the sample in test tubes a few drops of 10 per 



t ferric chlorid and of ferrous sulphate are added, respectively. 



Application of the solubility tests described above has shown the natural color in 

 \\ine- t,, be insoluble in ether under all conditions and in amyl alcohol when the wine 

 i- previously made alkaline with ammonia. A second dye was obtained with one 

 wine, but its dull appearance and its reaction with ammonia afforded a ready means 

 of di-tin-uishinu' the dye from those of coal-tar origin and from vegetable dyes of the 

 lichen group. Hydrochloric, .sulphuric, and acetic acids brighten or intensify the 

 color of the original wine, ammonium and sodium hydroxids darken the solution, and 

 ammonia alum produces no change. Tests with chalk steeped in albumen a gave 

 u n. -a ti.- factory results. 



In general the follo\\in_' com lusions may be drawn from the results obtained by 



the.-e te-t.-: 



1. Nitric acid darken- the original solution of red wines, but produces practically 

 no change in white u 



The lead precipitates vary from a pale yellow in white wines to a 



deep blue-gray in red wines, but a violet or red color is never found in the precipitate 

 from a genuine wine. 



3. Some highly colored red wines give a slight coloration upon the addition of 

 hydrochloric acid to the filtrate from the yellow oxid of mercury. 



1 Ferric salts give a precipitate, ferrous salts do not. 



The following results were obtained by II. V. Frost on a sample of red wine labeled 

 "Vino \ ecchio del ( hianti, Toscano, Italia." No evidence as to the authenticity of 

 the \\ine was secured, although it was known that the sample was imported from 

 Italy. The designations as to color correspond to those in Mulliken's chart. 



"I". S. Dept. Agr., Bureau of Chemistry, Cir. 25, p. 17. 

 73t>73 r.ull. 11'2 01) 2 



