into which all discharge sewage and from which all obtain their water supplies, would be less with partial sewage purification than if this were made complete; and at this writing (1918) the general tendency is towards the prevention of nuisance only; but on the other hand, persistent and intelligent efforts are being made to actually obtain a general enforcement of such treatment of all sewage. This means that the number of sewage purification plants is being greatly increased, but that the efficiency demanded has been placed much lower than it was some years ago.

The difficulty of setting a general standard to be met by all sewage-disposal plants lies not only in the various requirements to be met, but also in the varying characteristics of both the sewage and the stream which receives it. Where the stream is small, a much higher degree of purification is necessary to prevent a nuisance than where it is large. Moreover, the amount of free oxygen in the stream is an important consideration. The scientific method would be to ascertain the amount of free oxygen in the diluting stream passing the effluent outlet per second, and to permit no more unoxidized organic matter to reach such stream per second than can be fully oxidized by or of this amount of oxygen (since the intermingling of sewage and stream probably will not be complete, and a part of the free oxygen must be left in the water). So far as all organic matter except bacteria is concerned, the above standard would also insure a safe potable water if time and opportunity for complete intermingling and oxidation be afforded.

In examining a stream for sewage pollution, it should be remembered that the presence of chlorine in excess of the local normal is generally an indication of sewage pollution; nitrates indicate the amount of organic matter rendered innocuous; and albuminoid ammonia is taken as an index of the polluting organic matters still present. The presence of numerous bacteria is not necessarily indicative of sewage pollution, but B. coli are generally assumed to be (although small numbers may have been voided by animals other than man). The character of an effluent should not be judged by its appearance alone, by its

chemical or bacteriological analyses, but by the three combined; since it may be clear, but contain many pathogenic bacteria or dissolved matter which may be precipitated or putrefy and create a nuisance; also a turbid effluent may contain only mineral matters or such organic ones as are harmless and will undergo no change but oxidation.

ART. 70. DISPOSAL BY DILUTION

There are undoubtedly conditions under which disposal by dilution is much less objectionable than any other available method. And in considering this, it must be borne in mind that the liquid must ultimately be discharged into some stream or body of water; the question being therefore to what extent, if at all, it must be purified or modified before being so discharged. Under what conditions and to what extent a water receiving sewage will purify itself is a question which has received less attention than have methods of treating sewage, although it is much the most common method of disposal. The self-purification may be considered as to the organic matter and as to the bacteria. (Sand and other mineral suspended solids cause shoaling near the mouth of a sewer, unless this be in a swift current. But the treatment of this problem has already been considered; catch-basins or occasional dredging being the most common solutions.)

Considering first the organic matter, the agents of purification are sedimentation and dilution, accompanied and followed by liquefaction and oxidation; and the agency of animal and vegetable life. Any considerable amount of sedimentation is objectionable, especially if concentrated in a limited area; because the matter deposited undergoes putrefaction, and the products of this are disagreeable and render the water above unsuitable for a public supply if they do not even create a nuisance. If the deposit is thin, however, the products of putrefaction, both liquid and gaseous, may be so small in quantity that they are rapidly oxidized by the water above. This will depend upon the relative amounts of putrefaction products and of oxygen available per day or hour in the water above.

By dilution, or intermingling of the sewage with large quantities of water, the ratio of available oxygen in the water to organic matter may be made sufficient to cause the rapid oxidation of all the nitrogenous matter before sedimentation occurs. Such dilution prevents putrefaction; but total reduction to nitrates by oxidation is in general slower than by combined anaerobic and aerobic action. It is the opinion of many experts that still water purifies itself more rapidly than flowing, because of the liquefaction taking place in sediment, which is deposited more abundantly in still water. The chief advantage possessed by running water is that the constant delivery of fresh water insures a constant mixture and completeness of diffusion not secured by discharge at one point in quiet water. But, as Mr. X. H. Goodnough has said in a report to the Charles River Dam Committee: "The sewage discharged into a pond or stream may be objectionable or not, in the neighborhood of the outlet, depending upon the location of the outlet with reference to the stream or pond, and the conditions in the neighborhood. Observations upon the discharge of sewage into water at many places show that there is much advantage in discharging it at several outlets, since the sewage then mingles much more rapidly with the water, and is subjected more quickly to those actions which tend to remove its effect." Experience at places at which sewage is discharged into a pond or slowly moving stream indicates that sewage discharged into such bodies of water has a less noticeable effect upon their waters than an equal quantity of sewage has upon a rapidly moving stream of equivalent volume."

The above does not refer to stagnant water, since there must be fresh quantities of oxygen continually available in the water of dilution. This may be supplied in a large body of water by absorption of oxygen from the air into the surface layers of water, combined with a continual vertical circulation of water due to differences in temperature or to winds. But these causes are less reliable than a constant but slow translation, as by stream flow or tidal currents. Whatever the condition, the organic matter must come in contact with sufficient oxygen to permit

mineralization before putrefaction products reach the surface of the stream, if a nuisance is to be avoided.

No matter how large the volume of diluting water, unless the current at the immediate outlet be sufficiently swift to effect rapid and thorough diffusion and mixing, the discharge of large volumes of sewage at one point will cause a local nuisance which might be entirely avoided by providing several outlets some distance apart. A large volume of sewage discharged from a single outlet into a stream or lake can frequently be traced by its color for a long distance, only slowly mixing at its edges with the purer water.

While rapid diffusion and intimate intermingling are necessary, the degree of rapidity depends partly upon the putrescibility of the sewage; a corollary to which is that the amount of sewage which a given volume of flow will receive without nuisance is similarly dependent. Certain processes of sewage treatment produce, as their chief effect, reduced or delayed putrescibility; the most important being the sprinkling or trickling filter.

From what has been said, it is apparent that the amount of flow of a diluting stream required to inoffensively dispose of a given volume of sewage depends upon the strength and putrescibility of the sewage, the available oxygen in the water, conditions favoring sedimentation or rapid intermingling, diversity of outlets and other conditions. The limits are placed by most authorities at between 1500 gallons and 4cco gallons of diluting water per day per capita contributing sewage. (The proportion is sometimes stated in terms of cubic feet or gallons of sewage, but since the amount of impurity is not increased by greater per capita consumption or waste of water, the former method seems preferable. Investigations now being made by the U. S. Public Health Service promise to develop a more scientific method of estimating required dilution. See page 332.) This means that this amount of water must not only be flowing past the sewer outlet, but must be mixed with the sewage. It is quite possible to have a local nuisance created, in the form of nauseous gases and floating matter, by failure to effect rapid mixing, but to have a rapid reduction to inoffensive and harmless conditions after the

mixing has taken place. This reduction by oxidation is a function of time rather than of distance traveled by the stream, and this furnishes an additional advantage for discharge into slowly moving streams, in that the effect of the unoxidized sewage does not extend so far. Two to three hours after thorough intermingling are frequently sufficient for the reduction of much of the nitrogenous matter into nitrates.

It should be remembered that the water of dilution has been considered in the above discussion to be unpolluted; and that the same water, swinging back and forth with the tide past a sewer outlet, will soon become grossly polluted. The actual dilution will be closely indicated by multiplying the actual cross-sectional area of the channel by the distance separating the positions occupied by a given float at two successive ebb-tides, as compared with the sewage discharged in the same time. There will, however, be a somewhat greater reduction of the sewage than is indicated by such a calculation, since the water will continually absorb fresh oxygen from the air above and sedimentation will continue to remove sewage matter, and more complete intermingling will assist in oxidation and diminish the possible nuisance by mere physical dilution. Thus the same volume of water, moving back and forth with the tide and receiving no fresh pollution, will continually improve in character.

Sedimentation is most active when clay, sand or other heavy matter is carried in the sewage; this, in sinking, carrying with it other finer and lighter matter, including bacteria. A rough bottom, shallow water and high velocity, each and especially all combined, interfere with sedimentation, but assist in intermingling. It is found that sedimentation is much more rapid in salt water than in fresh; consequently more attention should be paid to the location of outlets in the former, to insure their discharging into rapid currents or in small quantities through numerous outlets. Mr. H. W. Clark, Chemist to the Massachusetts State Board of Health, found that " temperatures and other conditions being equal, salt water apparently holds less oxygen in solution than fresh water. This being so, it is evident that, volume for volume, fresh water can receive the greater