207 



It will !> seen f ru in this table that in an\ .-trakdit American whisky !M) per cent 

 i.i the ( .!,, riim matter is soluble in the amyl alcohol and ethyl alcohol layer, while 

 in an imitation whisky 1 I JMT cent is soluble in the upper layer. This method gives 

 a much more complete separation of the coloring matter taken from wood and from 

 caramel than either the water-insoluble method or the ether-soluble method, and 

 [eems to be the iii<i.-t reliable and satisfactory test that we now have for the detection 

 ol added coloring matter and its estimation. 



Further, if a whisky contains a certain amount of caramel this method will give a 

 partial separation, and the percentage of caramel added can be approximate!) e-ii- 

 mated. The present provisional method, known as the '-' Crampton and Simons test." 

 for caramel depending on the insolubility of caramel in ether, is not nearly so satis- 

 factory as the method here presented, as the separation of the coloring matters is 

 much less complete. The ether-soluble method, as given in Bulletin 107, Ke\i>-d, 

 lul, i.- i-iimbersoine, and calls for unnecessary special apparatus. Accurate 

 practical re-nils ha\e always been obtained by the following procedure: 



i : HI K-OOl UBXI OO] OI MI rinu. 



' CC of the sample just to dryness mi the water bath; wash into a 50 

 << lla.-k with _' . ,,i alcohol and dilute ! mark with water. Transfer -'> cc with a 

 pipette (. a -eparatory funnel and add 50 << of ether. Shake at intervals for half 

 an hour, let -ettle, draw off the aojueoii- layr, and make up to -^ cc with water. 

 Mix thi- latter and compare with the _'." .. ,,t the solution which were not treated 

 with ether Kxpre-.- the amount of color removed on the |>errentai:e basis as ether- 



soluble color. 



Thi.- modification -imply eliminates the social Hpunwell apparatus, and the 

 method i- but little u.-ed in the Bureau ! < hemi.-try, but the change is presented 

 as essential in appUing the meth'Nl. \Ye do, how-\er, u.-e the method of determin- 

 ing' the color insoluble in water, caramel, ,,f OOU ;-erfectl\ .-..luble ami tin- 

 coloring matter of whi-kv beiipj praeiically in-oluble in water. Thi- u r i\e> a method 

 of separation which i- \er\ .-.it i.-fai-t>r\ , the ppM-ediire outlined in this laboratory 



U-INSOl I HI I . ..| in: MM lll> 



i .-ample ju-t to drynetw. Take up with < old water, u-ii 



.Miately l"i CC, and tiller, washing "with water until nearly _'"> cc of filtrate i.- 



obtained. Add about _' .dcoh-.l. and coni|>lete the volume to the 



mark by tlie addition of water. Mix thoroughly and compare in a colorimeter 



with the color of the original sample, -tatin-j r-ult- as percentage of color insoluble 



in water obtained by .-ubtraetiiu; the pen entage soluble color, reading from 100. 



The following table compare- the re-ulis obtained by the determination of water- 

 in- .bible color with the ether-soluble color on a number of -trai^ht whiskies and on 

 spirits colored with caramel: 



' '<ni'/> nt, I, rnlnr nnil ft fur-soluble color on different type* o 



This table shows that results obtained by the ether-soluble method do not show 

 enough difference between the straight whiskies and those artificially colored. That 

 i- to - 1\ . iu ether the difference in solubility between whisky color and caramel is 



