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.
A test for atmospheric air in carbonic acid gas and which is quite simple and sufficient in most cases, may be made thus: Pour a sample of the carbonated water in a glass, whereby a strong effervescence takes place; if the water appears bright and transparent, and large, transparent, single distinguishable gas bubbles eliminate on the sides of the glass, the carbonic acid gas may be considered sufficiently pure; if, on the other hand, the water appears more or less milky, and numerous small, single but difficult distinguishable gas bubbles arise for some time, before the water becomes bright and transparent, then it contains atmospheric air. The gas bubbles of such a water arise more from the bulk of the fluid than on the sides of the glass as in the former case, and the effervescence ceases generally when the water finally becomes clear; while water impregnated with pure carbonic acid gas will continue to sparkle and do so gradually less. In water charged with carbonic acid gas and air, it is the atmospheric air, divided in infinitesmal molecules, which as soon as the pressure is relieved escapes in extremely fine bubbles, and these on account of their enormous number and diminutiveness make the water appear milky.
Mineral waters, on account of their salt ingredients; saccharine beverages on account of their saccharine matter; and wines, when poured out in a glass, also appear milky at first, but brighten very soon, and show in regard to their purity or presence of atmospheric air the same signs in gas bubbles or sparkling as mentioned before.
To test the quality of carbonic acid gas of a carbonate, and to arrange the purification accordingly, Dr. Hager gives the following directions:
Put in a beaker about half an ounce of the powdered carbonate, saturate with warm water and gradually pour on it diluted sulphuric acid. Warm the beaker with its contents slightly over a light and examine the eliminating carbonic acid gas by the smell. Then add some more of the carbonate and diluted sulphuric acid and cover the beaker with a sheet of white filtering paper that has been saturated with a solution of sugar of lead. If after several hours a brownish coloration is visible, an impure-ness of the carbonate by pyrites is to be considered established. If the smell was agreeable a purification with water and solution of soda, or with marble chips in addition, may be sufficient. If the smell is disagreeable, bituminous, or reminding of decaying animalic matters, then it is necessary to add a solution of permanganate of potassium to washers, or filter the carbonic acid gas through animal charcoal. If that brownish coloration has occurred it proves the presence of sulphuretted hydrogen, which makes the addition of a mixed solution of sulphate of iron and bicarbonate of soda to the water in purifiers absolutely necessary. A pure white color of a carbonate is not always a proof of its purity.