from the use of lead pipes and cisterns at Glasgow, since it has been supplied with Loch Katrine water, which is exceedingly soft, it appears probable that the ill-effects from the use of lead in this way have been exaggerated. Galvanized iron cisterns are fast taking the place of leaden ones. They are very durable, and of course far cheaper than lead. Stone or even brickwork lined with cement are sometimes used at or below the ground level for the storage of water, and are open to no objections so far as the material is concerned. Stoneware cisterns are now made, and are admirably suited for cottages, for use in basement floors, etc. Slate cisterns are not unfrequently used for upper stories, as well as ground floors. Of course, slate in itself is an excellent material for such a purpose, but slate cisterns, unfortunately, are very apt to leak after a time, and the joints are then filled in with red lead from the inside of the cistern-a practice which is, of course, very objectionable. The use of wooden receptables, such as tubs,

butts, etc., ought to be discouraged, if only because they are difficult to be kept cleansed. A self-cleansing tank is sold by the Sanitary Engineering and Ventilating Company. The bottom, instead of being flat, is made to slope from all sides towards the center, where the waste pipe is fixed. On lifting up, by means of a lever, that part of the waste pipe which stands up in the cistern, and which is fitted accurately into the commencement of the pipe at the bottom of the cistern, so as to make a water-tight joint, the water runs out of the cistern, and on account of the sloping bottom washes all the sediment away with it. generally supplied to the the pipes through a tap known as the "ball valve." To it is attached, by means of a metal bar, a hollow copper sphere or ball, which floats on the water as it rises in the cistern, and when it has risen to a certain height turns off the tap. It is because these taps are liable to get out of order, that a waste or overflow pipe This waste or overflow

is necessary.

The water is

cistern from

pipe should, in all cases, without any exception, discharge freely, as over an area, etc., so that you can see the water coming out at it. All receptacles of water should be well covered, in order that dust may be kept out of them. Nevertheless, ventilation space between the water and the cover, by means of holes provided with a grating, at the sides, is advisable.

Of course, for drinking water, we ought to choose a source of supply that is unpolluted. As Mr. Simon has said, "It ought to be an absolute condition for a public water supply that it should be uncontaminable by drainage." We ought not, then, to take confessedly impure waters and try to purify them, so as to make them fit to drink. On the other hand, it is obviously unnecessary to use very pure water, except where there is a superabundance of it, for washing the streets, flushing the sewers, and supply. ing the water closets, and so it may be advisable in some places to have a double supply of water, one of pure water for drinking and cooking, derived, for in

stance, from artesian wells, and the other of an inferior character for other uses. This has been lately proposed for London, and whatever may be said against it on the score of expense, I think most people will agree that it will be very desirable to have water to drink which has not been first polluted with sewage and then filtered. The advantage of this plan, too, was perfectly well recognized by the ancient Romans. Frontinus tells us that it pleased the Emperor (as he puts it) to order that the water supplied by certain aqueducts should be furnished to the people for drinking purposes, while that supplied by some others, from its being occasionally turbid and of inferior quality, was to be used for "viler purposes."

As, however, we do not, as a matter of fact, in the majority of instances, imitate the ancient Romans, either in this particular or in bringing pure water from a distance to supply the towns, but use the nearest water that we can get, whether good, indifferent, or bad, it is of

course necessary for us to do all that we can to purify it before use. This is done on a large scale by filtration through layers of sand and gravel, after the coarser suspended matters have been allowed to deposit themselves in a settling tank. I shall not describe this method of filtration in detail here, as it is a little beside the scope of these lectures, but, as the principle on which it acts is the same as that upon which the success of most forms of domestic filter depend, I may say a few words about it once for all. The experiments made by Dr. Frankland for the Rivers Pollution Commissioners showed that when foul water was passed through layers of porous soil, or sand and gravel, the amount of organic matter in it was reduced, if two conditions were fulfilled; these are that the filtration be downwards and in termittent. It was found that if the fil tration were upwards or continuous no such purification occurred after a time. The explanation of these facts is simple. The filtering material acts in two ways.