"Thus the question of house plumbing is closely connected with that of the sewerage systems of cities and towns, and these again with one another throughout the entire state and country. The question of sewage disposal is too broad to be considered as a municipal problem simply. It concerns the condition of the water courses throughout the state, and cities and towns bordering upon these streams and rivers can only be properly treated together as a whole in one broad, comprehensive scheme. Again, the investigations and experiments to which I have referred, in determining the effect of siphonage, back pressure, evaporation, capillary action and flushing action in plumbing, should be made by the state for all of its cities and towns together. In this way much more exhaustive and satisfactory researches and conclusions can be obtained with the same money outlay than by leaving each city or town to investigate the matter for and by itself.

"In our second diagram it will be observed that all the plumbing fixtures have been constructed on the principle of the flush tank. That is, their outlets have been made as large in their clear waterway as the waste pipes serving them and, where possible, their discharge has been made automatic. This automatic discharge is always possible, and, for good results, necessary, with kitchen and pantry sinks, of which one is shown a little to the right of the center of the picture.

"It will also be observed that only three traps are required on each story to serve six fixtures. There are four strong reasons for this, one of which is that it saves both expense and complication. A second and still more important reason is that this arrangement protects the seal of the water-closet from siphonage. The small trap is, in this case, a common pot-trap of great width but comparatively small depth. This renders it anti-siphonic. A slight modification of the form of this trap will render it also self-scouring. But its seal is only two inches deep, while the seal of the water-closet is four inches. Accordingly, the pot-trap, being rendered by its form anti-siphonic, permits air to pass through its water seal under siphoning action without destroying that seal. Thus it serves as a back vent pipe for the water-closet seal. For the latter, being twice as deep, cannot be broken so long as a shallower trap seal connects with it. This is a law of plumbing hydraulics which we can easily understand without further explanation, and it is strange that it has not been made use of in plumbing before this by the practical plumber.

"A third and equally important reason for placing the traps at or near the floor level instead of close to the fixture, as required by the law, is that it protects their seals from the effects of back pressure because the pipe above the trap is long enough to form a water column of length sufficient to resist the atmospheric pressure produced by this action.

"Finally, a fourth important reason is that the traps receive a much better scour when arranged in this manner than under the usual arrangement. The basin trap is scoured by the entire discharge of the bath tub.

"The reason why cesspools in traps should be avoided is because they give rise to that form of fermentation which is called putrefactive decomposition, and although it may seem to be rather a trifling thing to debar them in house traps, yet when we reflect that a dozen such small cesspools in a single house must be multiplied by thousands or hundreds of thousands, when considered from the standpoint of the purification and proper management of the public sewers, it becomes a very important consideration indeed. For putrefaction generates the anaerobic or dangerous classes of bacteria which work without oxygen and which are hostile to the friendly or aerobic bacteria which thrive best in large volumes of fresh air. The modern principles of sewer construction require the sewage to be carried through the sewers and deposited upon the irrigation or filtration fields in its fresh state before putrefaction begins. In this state it leaves the sewer air innocuous and also forms a better fertilizer, whereas in its putrid state it becomes an element of danger in both places, and is even destructive to fish life, when it is carried directly into the ocean as at Boston. In its fresh state sewage forms a useful food for fish.

"The recent researches of Laws, Andrewes and others show, first, that the number of germs of all kinds in sewer air is much smaller than in the air of the streets above them, and that this is due to the fact that the germs come in contact with the sewage and damp walls of the sewer and drain pipes, from which they cannot under normal conditions again escape. Second, that the bacteria found in sewer air are not of the same kind as those found in the sewage itself, but are of the same kind as those found in the outer air above the sewers, showing that the bacteria come, therefore, not from the sewage, but from the outside air above them. Third, that disease germs are unable to live long in sewage, where myriads of bacteria of decomposition, hostile to them, but friendly to man, abound; and that as a matter of fact disease germs have not been found in the air of sewers in the very careful experiments so far made.

"In corroboration and explanation of these conclusions the experiments of Carmichael, Wernich, Miquel, Naegeli, Pumpelly and Smyth and others show that germs cannot detach themselves from the surface of water at rest at normal temperature, nor from the damp surfaces of sewers, showing that the water seal of a trap forms an effective barrier against their passage into a house.

"Hence sewers can be constructed and ventilated in such a manner that the air within them becomes innocuous, and it is then evident that a simple system of house drainage without a main or intercepting trap and without back-venting can be made perfectly safe."