Our third requirement calls for a trap on the main drain between the sewer, cesspool or flush tank, and the fresh air pipe. A trap is practically a suitable bend or dip in the drain, which retains a sufficient quantity of water to prevent the passage of sewer gas.

The opinions of experts as to the advisability of trapping the main drain are divided, some considering the trap necessary, while others claim it should be omitted.

The objections urged against the use of traps are as follows:

1. They impede the ventilation of the pub lic sewers.

2. They form an obstruction to the flow of the sewage in the house drain, and are, therefore, the cause of accumulations of foul matter in the drain, which by its d< composition will generate noxious gases; also

3. Foul matters will lodge in the trap.

While the first objection does not strictly belong to the subject of this paper I will say that it is accepted by most authorities that house drains and soil pipes should not be used as ventilators for the street sewers. In exceptional cases - such as, for instance, where an entirely new sewerage system is built, designed and constructed according to uniform plans, and where not only the construction of sewers, but also the house plumbing is under constant supervision of the engineer and designer of the system* - the trap (and consequently the special fresh air pipe) may, perhaps, be left out. But I believe that a proper ventilation of sewers can be effectually carried out without ventilating through the houses.†

In regard to the second and third objections, I would say that obstructions do not frequently occur if the drain is carefully laid, with sufficient and continuous fall to insure a cleansing velocity of the flow. If such an inclination cannot be given to the drain, proper flushing appliances should be used, and these will by daily or more frequent washings, insure the removal of all matters liable to lodge in the trap. Another most necessary precaution to prevent accumulations in the trap, where the fall is very slight, may be found in the use of a proper grease trap, about which I shall speak hereafter.

* For instance, at Memphis, Tenn., and at Hamburg, Dantzic, Frankfort-on-Main, Berlin, Breslau, and other places in Germany.

† See Mr. Edward S. Philbrick's articles on "Ventilation of Sewers," in the Sanitary Engineer, Vol. I. See also Sanitary Engineer, Vol. V. Number 12, page 246.

No amount of care in laying the drain will prevent its obstruction through carelessly introduced articles; these will mostly lodge in the trap. A cleaning hole should therefore be provided with the trap, and is rarely omitted in good work, or else a Y branch, closed with a trap screw, should be inserted just a little above the trap.

In Vol. III. of the "Sanitary Engineer" will be found a discussion of the advisability of trapping the main drain. My own opinion, as stated in a communication to that journal, is as follows:

"If we could have ideal sewers, house drains and soil pipes, it might, perhaps, be possible to dispense with such a trap altogether. But since all sewers may have temporary stoppages from some cause, since house drains may settle or leak, and joints of soil pipes crack, thus allowing sewage matter to undergo putrefaction and enter the interior of houses, I would in all cases advise the use of a safeguard, consisting in a disconnecting trap and a well ventilated soil pipe. This latter arrangement is a conditio sine qua non, and rather than have a trap without ventilation I would advise to have none at all......I would always condemn as unsafe a system of house drainage in which the public sewers are ventilated through the houses.....

The work of ventilating public sewers should, in my opinion, be done by the same public authorities who devise the sewer system, and not by the householders."

Leaving aside, however, the case of a house drain connecting with a public sewer, it seems quite evident that, in the case of a house discharging its sewage into a cesspool, an effective barrier should be imposed to the gases constantly generated in that receiver of all foulness from the household; and equally so in the case of a flush tank which temporarily holds a large amount of fæcal and other refuse matter, which sometimes undergoes decomposition.

The principle of disconnecting each house from the street sewer was first advocated in England, and its importance becomes most apparent in the case of an epidemic, as by the use of a trap each house will be isolated, while if all houses have an open connection with a sewer, this and the house drains may become the channels for spreading the disease from one house to another. It has been said by those not in favor of such disconnec tion, that the air of the house drain, the soil pipe and the branch wastes is much worse than that of most city sewers, and that consequently no harm could be done by allowing the sewer to breathe through the pipes in the house. Such statement may be true in regard to the sewers of some cities; in others, sewers, especially if built long ago, are extremely foul. But it seems to me that just where the air of drains and pipes is foul, it needs a strong dilution and purification by abundant fresh air, which an opening to the outside atmosphere can furnish, but never a direct connection with a sewer.

An open connection of the house drain with a sewer or cesspool is necessarily based upon the condition that every joint in the house is perfectly tight, and every trap perfectly trustworthy. As plumbing is done in most houses these conditions are only seldom fulfilled. But even where in new work such a standard of design and workmanship has been reached, the work may not remain so forever. It is, therefore, advisable to use a trap on the main drain as a safeguard, but in addition to this to insist upon occasional inspections. These become a necessity in the case of large buildings, such as hotels, schools, large factories, jails and almshouses.

Incidentally, it should be mentioned that a trap on the drain performs a most useful office during repairs or alterations of the plumbing work in keeping from the interior of the building the gases from the sewer.

Much, of course, depends upon a proper kind of trap for such disconnection. The old so-called "cess pool trap" is, next to the pan closet and the D-trap, the worst device ever proposed in connection with house drainage. As usually constructed it is of very large size, with square corners, and soon accumulates filth, becoming in a short time in reality a cesspool.

The common running trap, which is manufactured in earthenware as well as in iron is the simplest and at the same time the best of all forms. It should preferably have a vertical drop of a few inches from the drain to the water line in the trap in order to expel any solids that would tend to lodge in it. The running trap is often provided with a cleaning and inspection hole at the house side of the water seal, which serves as a fresh air inlet, when the trap is placed in a manhole outside of the house. In other instances a rain leader is inserted into the opening of the trap, which thus receives abundant flushing at each rain fall. The running trap is sometimes located on the line of the iron drain, just inside of the foundation wall, so as to be at all times easily accessible. A trap, in iron, with a cleaning hole and a cover is then used. Care should be taken to close the cover perfectly air-tight.

In all cases the trap should be so located as not to be liable to freeze in cold climates or exposed localities.

In England various "disconnecting traps" have been used, such as Moles-worth's trap, Prof. Reynolds' and Dr. Buchanan's disconnectors, Hellyer's Triple-Dip Trap, Pott's Edinburgh "air-chambered sewer trap," Stiffs "interceptor " sewer trap, Weaver's disconnecting trap, Mansergh's, Buchan's, Banner's, Stidder's, Bavin's traps, the "Eureka" sewer air trap, and many others. All of these may have certain merits, but nothing could be better nor cheaper than the common running trap with fresh air pipe used almost exclusively in American plumbing.

For those exceptional localities where undue pressure in the sewer, from wind blowing into the outlet of the sewer, or from sudden changes of temperature (when exhaust steam is allowed to enter a sewer), or from heavy accumulations of surface waters gorging the sewer, or from the action of the tide in tide-locked sewers, frequently forces the seal of the trap, two running traps with a proper vent pipe between them have been recommended. I have myself, for some time, advocated such an arrangement, which, after further experience, I think complicated and unnecessary. It would require either a pipe extended through the roof, between the two traps, or else an open shaft (a manhole) between them, and besides this, in every case, a fresh-air pipe entering the drain above the upper trap.