At any rate it is doubtful whether the test of the number of micro-organisms should determine the question whether a water is or is not safe to drink. Dr. Koch’s gelatine peptone test has enabled the analyst to recognise the absence or presence of microphytes; but, as was stated at a recent meeting of the Society of Medical Officers of Health, a sample of river water which might be marked “very good” by this test would develop an enormous number of colonies if kept for a few days, even in a “sterilised flask” protected from aerial infection. Prof. G. Bischof says, in fact, that a sample of New River water kept for six days in the above manner compares unfavourably as regards the number of “colonies” with a sample taken from the company’s main and polluted with one per cent. of sewage, or with a sample of Thames water taken at London Bridge. It seems certain too that the water stored on board ship must develop an enormous number of “colonies”; but no special amount of disease is attributable to them, and it would seem to follow that, unless the number of microphytes can be shown to indicate, or to be a measure of, pollution, Koch’s test is of little utility except as a guide to waterworks’ engineers, by pointing out that the filters want cleaning. In the laboratory the test is no doubt of considerable value; but in analysing water it must be applied with discrimination, and waters of a totally different character should not be compared by the number of organisms. For instance, the water from Loch Katrine might contain large numbers of micro-organisms, and yet be perfectly safe as compared with a water in which few microphytes could be found, but which had been accidentally polluted by some of those pathogenous germs which undoubtedly exist, and which produce disease when they find a suitable environment. Not until we are able to discriminate between the harmless and the disease-producing microphytes, shall we be able to test a water supply and declare it practically pure. The foregoing paragraphs will suffice to show what a very unsatisfactory state our present knowledge of water is in. The only useful fact to be deduced from all the argument is that every household should filter its own drinking-water and take care that the filters are always kept clean and in good working order. There is one simple test for the purity of water, introduced by Dr. Hager in 1871, consisting of a tannin solution, directions for which will be found in the Housekeeper’s section. It remains to notice the chief kinds of filter. Filtration is destined to perform three distinct functions, at least where the water is required for domestic use; these are (1) to remove suspended impurities; (2) to remove a portion of the impurities in solution, and (3) to destroy and remove low organic bodies.