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.
1. If, by adding a few drops of a solution of acetate of lead, a milky or cloudy appearance presents itself, it shows that the water is not capable of holding any lead in solution; but, on the contrary, if, upon the addition of five drops of a solution of bichromate of potash to another test-tube a dull or clouded appearance ensues, then it is certain that lead is present. The quantity in solution will be indicated by the degree of opaqueness produced; but, however small this may be, it may be taken for certain that such water is dangerous for use.
2. A solution of bichromate of potash, or iodide of potassium, added to water containing lead, will cause a precipitate if the lead is present in sufficient quantity. If the quantity of lead be too small to be detected by this means, the most certain way to detect its presence is, first, to examine what separates by exposure to the air, by dissolving it in warm acetic acid, and testing the solution with sulphuretted hydrogen; if this process fails to show the lead, the water should be concentrated to an eighth part and again tested. So says the National Bottlers' Gazette.
All these tests are dependent upon the appearance presented by the water after the addition of one or other of the test fluids. The best method of observing this appearance is by looking from above down into the tube, or, as in the tests for lead, carbonate, and sulphate of lime, looking sideways at the tube; not, however, holding it against the light, but against some dark object, when the cloudy appearance caused by the presence of the object sought for will, if it be present, be readily observed. In all cases, the test tubes should be nearly filled with the water to be tried.
4. If it cannot readily be bought, prepare a solution of sulphide of soda as follows: Thoroughly mix a small quantity of sulphur (about a teaspoonful) with twice its quantity of cooking soda; put the mixture in an iron spoon or ladle, and heat it over the coals until it is well melted and the flame of the sulphur has gone out. Scrape the black residue from the spoon, and add to it, in a small bottle, an ounce of water. Let the solution stand for several hours, until the insoluble parts have settled; then pour off the clear yellowish green liquid into another bottle. Have at hand a little hydrochloric acid (muriatic acid). Fill aptumbleror clear glass with the water to be tested; place it on a white surface in good light; add one drop of the sulphide of soda solution; stir the liquid, and if lead is present it will assume a brownish black color - the depth of color depending on the amount of lead. To ascertain whether the color is due to lead and not to iron (for the sulphide of iron is also black), add to the solution a single drop of hydrochloric acid, and stir it Do not add the acid until after the sulphide has been added. If the color disappears it is due to iron; if it grows paler, but does not disappear wholly, it is partly due to iron and partly to lead; and if the color does not change, lead is the cause of it. After the acid is added the liquid is apt to assume a slightly milky appearance from the separation of sulphur. Care must be exercised not to confuse this with an actual fading of the color.
Good water should contain less than one-tenth grain of lead per gallon (1 grain in 6,000,000 grains). The test gives a distinct reaction with less than this amount. But the exact quantity cannot be determined outside of the laboratory. Unless one is so particular to know the amount as to have the work done, it is best to reject a water that gives any coloration with the test, since it is safer to drink no lead at all.
The most serious as well as the most common metallic contamination of water is with lead. Although a water may contain but a very small quantity of lead salts, there is no doubt that its continued use will produce well-marked cases of chronic lead poisoning, and numerous instances are recorded in which disorders have been traced directly to their cause, and have ceased with their removal.