these sources will be considered under the head of Houseconnections (Art. 82).

ART. 29. METHODS OF VENTILATION.

It is evident that the danger from sewer-air may be avoided, or at least lessened, in two ways: by preventing the creation of gases, and by preventing the sewer-air from reaching human beings in dangerous quantities or under dangerous conditions. No method has yet been found for perfectly accomplishing either of these aims in practice, but both may be partially attained.

Aside from illuminating-gas most of the objectionable gases are given off by putrefaction, and the prevention of this in the sewers is therefore most necessary. This is best accomplished by the removal of all sewage to the outlet before putrefaction can begin; and here is seen the advantage of daily flushing, cleaning the upper laterals of deposits before they reach this dangerous stage. The use of disinfectants in sewage for this purpose is seldom advisable, both on account. of the enormous cost and practical difficulties of applying them and because the various and changing characters of sewage in different cities and from hour to hour may introduce such matter as will combine with any given disinfectant to produce deposits and gases fully as injurious as those due to sewage alone. The transporting of germs by sewer-air is probably reduced by reducing putrefaction, although there is very little definitely known on this point, it being uncertain even whether disease-germs are carried by sewer-air at all.

To prevent air from the sewer from entering houses two general methods are in use: placing a barrier in the houseconnection, and removing the sewer-air through other outlets. The former is one of the aims of the plumber and is usually attempted by the use of traps. The latter has been aimed at

by the use of many ventilating devices, in few or none of which has positive action been successfully obtained. A combination of these two methods gives reasonably good results in most cases, a partial obstruction to the air being placed in the house-connection or its branches in the shape of water-sealed traps, and the power of the air to force its way through these being lessened by ventilation.

If the sewer were a tight conduit with no inlets or outlets except through the house-connections and the main outlet the sewer-air must remain constantly unchanged and stagnant, or must find exit and entrance through these house-connections. The first condition is impossible, for the amount of sewage varies from hour to hour and must displace and in turn be displaced by air driven to and derived from some outside source. In case of a sudden discharge of sewage into such a sewer the air will be driven through the only outlets-the house-connections-unsealing the main traps, and the secondary ones also unless these be amply vented. A strong wind blowing up the sewer from the outlet may produce the same result. In addition to other ventilation of both sewer and soil-pipe it is therefore advisable to thoroughly vent all housetraps.

Attempts have been made to constantly remove the air from sewers by either sucking out the foul air or forcing in fresh; that is, by producing a current through the sewer to a given outlet by either the vacuum or plenum process. Both have proved failures as well as very expensive. In no experimental case has the effect been felt more than 1000 feet from the fans or other apparatus, not only on account of the great amount of air in the sewer-mains and laterals to be moved, but because the traps in the house-connections were unsealed by the pressure and air admitted from or forced into the buildings, according to the system employed.

The Metropolitan Board of Works, London, concluded,

after exhaustive study of the question, "that the method of ventilation adopted in mines, where there are only two openings to be dealt with (an inlet for the air at one end and an outlet for it at the other), is inapplicable to sewers. This characteristic of a sewerage system renders impracticable all methods of ventilation depending upon one or two ventilators to each line of sewers: such as connecting the sewer-end with a chimney, which would afford little more ventilation than an untrapped soil-pipe at the same point or a special ventilatingmanhole.

Many expedients for ventilation have been devised and tried-among them connecting the sewers to street-lamps, where a suction is caused and the gas burned by a constant flame; placing in the crown of brick sewers small perforated pipes connected with "uptake-shafts," expected to cause a continuous removal of the gases; leading pipes from the sewer to special flues constructed in houses, within the body of the walls, adjacent to the chimney, or upon the outside of the house and running up above all windows; leaving the main house-drains untrapped and extending them above the roofs; placing flap-doors in the sewers, opening downward for the sewage, but closed to air, which can escape through openings just above such flaps; placing in the street centre at intervals along the sewer manholes or other ventilating-shafts with perforated covers; connecting the sewers by untrapped pipes with street-inlets at the curb line. In connection with these charcoal and other deodorizers are sometimes placed at the air-outlets. (See "General Conclusions, Metropolitan Board of Works," London.)

There seems to be evidence in favor of the conclusion that the greatest danger exists in the house-connections themselves and not in the sewers, although the latter should be prevented from contributing to this danger. Of many analyses of sewerair made not one to the author's knowledge has shown a

greater impurity than that in a crowded city street, whether CO,, oxygen, or bacteria be taken as the basis of comparison. Equally positive proof goes to show that the average houseconnection or the adjacent soil near open joints in the same does give rise to dangerous gases. (It is probable that the upper ends of branch sewers, if not flushed well and often, are open to the same charge.) However, a rush of comparatively pure air from the sewer forced through the traps of a foul house-connection is as objectionable as though it itself were polluted, since it forces into the building the impure air existing in such connection. The vents on all traps should hence be of such capacity and so placed as to give full and immediate passage to all the air necessary to prevent forcing or siphoning of traps.

This fact, that the house-connections themselves are fully as foul as, if not more so than, the sewers should be more generally recognized and better provision made for ventilating them. This is reasonably well done by placing a vent-shaft just above the main trap, continuing the soil-pipe above the roof and venting each trap throughout the house. But a still better circulation of air is obtained by omitting the main trap altogether and permitting the air from the sewer to pass through the house-connection unobstructed. The danger of this air passing the traps on house-fixtures is no greater than that of the soil-pipe air doing the same, and in the majority of cases the sewer air is the less dangerous. Such construction is also of great assistance in ventilating the sewer. If only an occasional house-connection be left untrapped, however, the odors from this may be objectionable, the sewer air being but little diluted by the infrequent openings. But the author knows of no city which makes this method compulsory in all connections where it is not perfectly satisfactory. (See also page 344; and Appendix No. 1.)

The use of street-lamps as outlets may be advantageous, but the electric light has rendered argument for and against this plan obsolete. The use of hollow electric-light poles has recently been introduced in Columbus, O., with what success it is too early yet (1899) to state. The use of flap-doors in the sewers presupposes a regular flow of air in a fixed direction through the sewer, which investigation has found does not ordinarily exist; this, however, may be advantageous on steep grades, where there is a tendency for the air to rise past intermediate ventilating-points to the highest ones. Ventilation through manholes and other ventilating-shafts most, if not all, engineers recommend, although many do not consider these sufficient.

In

The use of storm-water inlets for this purpose is much opposed by many, who contend that the sewer-air should not be discharged so near to passers-by upon the sidewalk. fact this same argument is used by a few against ventilation through manholes in the centre of the street. It is probable that the danger from this cause is very slight, if it exists at all, since it is dependent, not upon the gases, which are enormously diluted upon reaching the outer air, but upon the presence of disease-germs in the exhalations, which is not proven. Moreover, the average catch-basin, even if just cleaned (as this cleaning is ordinarily done), is more offensive than any rightly designed sewer is at all likely to become; and it is extremely doubtful if, in connection with its odors, any contribution of air from the sewer could be detected. For these reasons it seems to the author desirable to connect the sewer with the street-inlets by ventilating-pipes and to place manholes with perforated heads at intervals. Since the latter are apt to be sealed in winter by ice and snow, and in summer by mud, the additional ventilation through the streetinlets would seem to be advisable, particularly if the sewer be not ventilated through the house-drains. A small amount of snow will not ordinarily stop the openings in a manhole-cover,