This section is from the book "A Treatise On Beverages or The Complete Practical Bottler", by Charles Herman Sulz. Also available from Amazon: A Treatise On Beverages.
There are sugar colors for carbonated beverages, and for liquors, spirits, tinctures, essences, wines, rum, beer, ale, cider, vinegar, jellies, etc. They appear under different names and different prices - but all are but one kind. If all colors are prepared with the addition of carbonate of soda or potash, there exists but one difference in solubility. The color thus prepared causes a turbidity in vinegar, but is or should be soluble in all other aqueous or alcoholic solutions without a sediment. A color prepared with carbonate of ammonium must be soluble in vinegar as well as in aqueous or alcoholic solution, without precipitate. The commercial value of sugar color depends foremost on its perfect solubility. If made from glucose free of dextrine and sulphate of calcium, it is as well soluble in water as in alcohol of 80 per cent. A difference in solubility in stronger or weaker alcohol does not exist for the honest manufacturer; he makes but one grade. The next important factor is the density or the degree of concentration of the sugar color, which is determined by comparison (see tests). A hydrometer is of no service, since a reduced color could be improved in consistency by the fraudulent addition of glucose-syrup. Sometimes it happens, however, that even the best commercial color is declared inferior, because of its faulty application. This is especially the case when essences, containing no water, consist to a large extent of ethers. The commercial color is herein almost insoluble, and precipitates in flakes. To color an essence, for instance a rum essence, one or two parts of the color must be previously dissolved in some 80 per cent, alcohol. Of this coloring tincture sufficient is added to the ethereal essences, while vigorously shaking, to obtain the desired shade. Even then some sediment will occur, which cannot be prevented in such a liquid. but it is easily removed by filtration.
It is always most essential to determine whether the coloring material is capable of passing into solution and is yielding a bright and clear beverage, or, on the other hand, if it is causing turbidity and deposits. The strength of a color is also to be tested. It makes a big difference whether one ounce or two or three of color have to be used to give a definite shade to a certain quantity of liquid. The concentration of a color is important, but to measure its concentration (density) with the saccharometer is no proof test as already mentioned in another place, since a high specific gravity may be imparted to a reduced color by the addition of glucose syrup, etc. To test its strength it is necessary to compare one kind of color with another. With this test the above test for solubility can be combined.
Take the same quantity of each color, about one drachm, measured accurately, and dissolve each drachm separately in one pint of well-purified and filtered water. Compare the shades by holding the bottles up to the light, and decide which is the strongest. The same test should bo made with diluted alcohol, also with beer, etc. The other decision to be reached is in regard to the solubility of the color. Cork the bottle under test tightly, and set aside for a week or so. If on turning the bottle up-side down no precipitate and no turbidity is visible, and the liquid remains clear and brilliant, it may then be presumed that the article is fit for employment in carbonades, alcoholic solutions and beer, while on the other hand, if it occasions the slightest turbidity in the liquid, or a precipitate has been thrown down, the color may be condemned as imperfect and unfit for the bottlers' and brewers' purpose.
A well-made sugar coloring should have no sweet taste or a faintly bil-ter one, and be perfectly miscible in aqueous alcoholic liquors, carbonated beverages, and also in beer.
Another test should be made by dissolving thirty grains of tartaric acid and thirty grains of the coloring in one pint of water. If within an hour or so the liquid is clear and bright, and no precipitate visible at the bottom of the bottle, the coloring can be used with security.