WATER ANALYSES AND THEIR INTERPRETATION. As an aid to the study of the quality of water recourse is had to an inspection of the drainage area and all possible sources of pollution, and to physical. chemical, and biological examina tions of samples of the water. The physical ex aminations involve observations on temperature, color, odor•, and turbidity. Color is generally determined by comparing the samples of water with fixed color standards based on chemical so lutions of known composition and regularly vary ing strength. The most commonly accepted measure of turbidity is the depth at which some bright object will disappear from view if gradu ally lowered in water. High temperatures give rise to troublesome organic growths. As a rule. however, temperatures arc quite beyond control, except in choosing between different sources of supply and in locating the depths of the pipes through which water is drawn from deep lakes and reservoirs. Deep temperature readings are taken by means of the thermophone. Absence of color, odor, and turbidity, combined with low temperatures, are most desirable qualities, but all of them together are of vastly less im portance than freedom from organic impurities. Unfortunately. no method of analysis has been discovered which will do more than indicate the probable safety or danger of a given water. Chemical analyses point to past contamination, and give some evidence of its nearness or dis tance in point of time and whether it was of animal or vegetable origin. Bacterial analyses are of comparatively avail except to show the numbers of bacteria present, and to tlu•ow' some light on their prob able origin. The differentiation of the typhoid germ, the chief object of fear in America, from the harmless water bacteria is a long. tedious, and at best a very uncertain operation. The more direct microscopical (-xaminations of forms of life above the bacteria, although of much future promise, is limited chiefly to determining the absence or occurrence of those growths that valise bad tastes and oilors. But when such growths are detected the most that can be done, ordinarily, is to shut off the water in question, take steps for its purification, or exclude light, when the latter is an essential to troublesome visitors. As an aid to detecting organic im purities, chemical determinations of albuminoid and free ammonia, nitrates, and chlorine are made. The first four substances indicate
past contamination. Their relative amounts, in the order named, throw some light upon the degree of natural purification, as measured chew l', that has taken place. The chlorine, if above the local normal for unpolluted water, in dicates that the organic matter came from sew age. High nitrates and chlorine combined, espe cially if accompanied by large numbers of bac teria, render water very suspicious. If, in addi tion, Co/i communis, a kind of bacteria found in large quantities in the human intestine, are present. the evidence against the water is often held to be conclusive. It may be accepted as quite so if it is learned that sewage is being discharged into the stream or lake above the point of inlet. See DISEASE, GERSI THEORY or. No determinations of the mineral contents of water are made in the case of surface supplies unless they come from limestone regions and ex cessive hardness is feared. Underground waters are so likely to contain the sulphates and car bonates that cause hardness, or else other troublesome mineral salts, particularly iron com pounds, that it may be well to test for at least those substances when examining well water. Lead poisoning is not due to lead normally in the water, but to lead taken up by the passage of certain water through, and more particularly by its standing in. lead pipes or cisterns. Very soft waters are especially liable to attack lead pipe. Chemieal standards for water are very mis leading. The best chemists and engineers interpret each set of analyses in the light of all other known facts and attempt to lay down no arbi trary standards. A minimum of 100 to 500 bac teria per cubic centimeter has been set by vari ous authorities; but this, also, depends on all the other local conditions. Hardness is chiefly a question of more or less soap in connection with water used for household purposes, and of scale formation in steam boilers. 11 excessive, the supply may be rejected, or a softening plant may be supplied. The substitution of pure mountain for contaminated Passaic River water at Newark in 1S9?, and of filtered for unfiltered water, at Lawrence. Mass.. in 1893, and at Al bany. N. Y., in 1899. are among the best Ameri can examples of the decline in typhoid mortality following upon improved water supplies.