owing to the warm air of the sewer, but a heavy storm or frozen mud may easily do so.

Since the proportion of air in a small sewer to the discharge into the same is much less than in the case of a large combined sewer, and consequently the effect of a given discharge is a greater compression of, and pressure transmitted by, the air in the smaller sewer, the sewers of the separate system need ventilation or safety-vents even more than do those of the combined. In case there are storm-water inlets to which ventilation-pipes from house-sewers may be led this method may be adopted; but ventilation through untrapped house-connections is probably more efficient. This extra ventilation is very often--perhaps in the majority of casesneglected, but such omission is undoubtedly attended with danger.

For house-sewers, ventilating manhole-heads and untrapped house-drains; for combined sewers, these with the addition of untrapped street-inlets; and for storm-sewers, manholes and inlets-these, with flap-doors on steep grades, seem to the author the best methods so far devised for ventilation; and without ventilation any system will almost surely become a nuisance and a danger. The aim should be to secure by whatever method the greatest possible number and freedom of communications between the sewer and the outer air; and there is little doubt but that when this is realized the sewer air becomes so diluted and the organic matter floating in it so oxidized as to render it less dangerous and objectionable than the air of a crowded church or theatre. When this is not true the sewers are probably in great need of cleaning and flushing. (See Appendix No. 1 for data on this subject.)

CHAPTER VI.

COLLECTING THE DATA.

ART. 30. DATA REQUIRED.

ANY plans made before the full and complete data are at hand may be shown by further information to be inadvisable, while their very existence may create a prejudice against the substitution of more efficacious ones. Therefore, although the development of the plans may suggest the desirability of further data the necessity for which was unforeseen, as much as seems necessary in this line and that of surveys should be done preliminary to any designing.

The first necessity will be for a map of the district under consideration. This will usually include the city or town and all land over which it may spread in the future; also all adjacent areas which shed their water into or across the surface of this territory. This map should show all streets, lanes, etc.; all parks or other areas permanently devoted to vegetation; all rivers, creeks, ponds, or other bodies of water—in fact all natural and artificial divisions of the area embraced by the corporate limits. It usually happens that this much can be found already mapped for other purposes; but unless it is known that the measurements from which such map was prepared were accurately taken a sufficient number of check measurements should be made to establish its accuracy or the reverse. On the point of accuracy a question may arise as to how exactly the measurements should be taken. If these

should involve an error of no more than .2% they would be sufficiently accurate for the work in hand. For, as sewer grades are ordinarily run from manhole to manhole, and these are about 300 feet apart, an error of .2% would mean that of .6 foot in that distance, which on a grade of .5% (a fairly steep one) would involve an error in grade of .003 foot, which is much less than the least which could be expected in the construction of the sewer.

It will be advisable to obtain also the location of all streetrailroads, and of all gas- and water-pipes, their distance from the curb or side lines of the street and the depth of the pipes being noted. Also the location, grade, size, and condition of any existing sewers and appurtenances should be ascertained, by actual inspection if possible.

The data for computing the extent of tributary drainageareas will ordinarily need to be collected in their entirety, as it is seldom that such information exists in a serviceable form. The topographical surveys which have been made of several of the States, however, may be used to great advantage in this connection. The data desired includes the boundaries of the watersheds whose run-off does not reach a confined channel before entering the limits of the territory to be sewered. (Such water as passes through this territory in the form of streams rather than flowing over the ground does not affect the problem, unless these streams are to be walled in, in which case each one will form a problem by itself.) Also the slope of the ground and the character of the soil as to permeability should be ascertained, the location and extent of rock at or near the surface, of woods and of orchards. Care should be taken to note and locate any slightly worn channels along which storm-water ordinarily flows to the nearest creek or rivulet across territory not yet built up, as these, if they cross into the sewer district, indicate the points at which the storm-water must be intercepted.

Such levels must be taken as are necessary for the plotting of profiles of each street, alley, or any other surface under which a sewer is to run, including a profile across the bed of each stream crossed, with the elevation of high- and low-water marks; also the elevation of the body of water into which either the crude or purified sewage is to be discharged, the elevation during drought and flood as well as the ordinary elevation being ascertained. The depths must be obtained of all cellars whose bottoms are not evidently above the grade of the proposed sewer, unless all sewers are to be placed at a fixed minimum depth, which is to be increased only by the demands of the necessary sewer grades and not by the depth of any cellar or basement (see Art. 37). Also if grades have been adopted for any street, but not yet carried into effect, these as well as the existing surfaces should be obtained.

If a disposal-ground is to be used for filtration or irrigation a careful levelling of its entire surface must be made, and test-pits sunk to ascertain the character of the material to a depth of 5 to 8 feet.

If it is considered desirable to discharge the crude sewage into a given body of fresh or salt water careful search should be made for the point best suited for the outlet; also in case of a river whether the dilution afforded in time of drought will be sufficient to prevent a nuisance. For this purpose the action of currents, tides, and prevailing winds should be investigated. Gaugings of the discharge of streams should be made, and inquiry as to whether and at what points further down the river the water is used for a public supply. It is well also to have analyses made of river-water taken at intervals below the proposed outlet for use in possible suits against the city for nuisance; this whether or not the sewage is to be treated.

The engineer should in person pass through every street in the district to be sewered, noting the character of each, the

location of the business and factory districts, the general character of the pavements and yards, and the average size of lot occupied by each residence. He must also ascertain as nearly as may be the present population and its past rate of increase; the probable direction and extent of the future growth of the business part of the city, as well as of the city as a whole. He should obtain the figures, if they exist, of water-consumption in this and neighboring cities; also all possible data concerning the rainfall.

A considerable amount of other information will in many instances be desirable, called for by the peculiarities of each case. Many items, such as cost of materials and labor (for use in the estimate), will suggest themselves as they are needed.

ART. 31.

SURVEYING AND PLOTTING.

Since extreme accuracy is not necessary in the transit survey, the use of the ordinary stadia methods will be found advantageous for either check or original surveys. Stadiahairs in the level, for use in running street-profiles, will be found to expedite this work, and will permit reducing the number in the level party to two. The adjustment of the stadia-hairs should be frequently checked.

The tributary drainage-areas will not need to be surveyed in great detail. If the natural features are boldly accentuated it may be sufficient to locate by a transit-line the limiting summits and ridges, both main ridges and spurs. If the country is gently rolling or generally flat contour surveys should be made of the whole drainage-area, or at least of any portion of it the disposal of whose run-off may offer difficulties.

Of such undeveloped areas as may be reached by the city in its future growth and which will be embraced in drainageareas for which sewers are to be at once designed accurate