of the work done by the contractor during the month just past. This involves the measurement of the number of feet of pipe laid and the number of cubic yards of excavation made both of earth and rock, with the number of manholes, lampholes and other appurtenances. The method of computing the excavation depends on the specifications, but the most reasonable method is for the specifications to prescribe the width of the trench that will be paid for, in terms of the diameter of the pipe and of the depth of the pipe invert. There should be room on the outside of the bell for the workman's hands, with a small margin for alignment of the pipe, a width of 12 inches more than the outside diameter of the bells being a reasonable amount. The outside diameter of the bell for a 12-inch pipe is 17 inches, so that the trench width for a 12-inch pipe should be 29 inches, the width to be used in estimating. A 6-inch pipe would in the same way be estimated as 211 inches and a 24-inch pipe as 43 inches. These are minimum widths and would probably be narrower than the trench would actually be dug. This would be particularly applicable if sheeting were used, since sheeting as ordinarily driven requires for one row 6 inches additional on each side. Strictly, then, two feet should be added where sheeting is to be used, and one foot where the trench is stable without it. The specifications might therefore well differentiate except for the opportunity of collusion with the contractor, who by sticking up an occasional brace in the trench could secure a measurement of an additional foot, the engineer being willing. It would introduce, without collusion, an uncertainty, the contractor being able to claim the additional foot whenever he decided sheeting was necessary and placed, even if it were placed only for the purpose of getting the extra measurement. It is better then, in ground where sheeting will probably be used, to make the additional width two feet, but where the ground is stable to reduce this to one foot. This will hold to depths of about 10 feet. Below this two rows of sheeting would have to be driven, so that the upper sheeting would require another foot of width, adding three feet to the diameter of the outside of the bell. It is probably fair also, since the cost of excavation increases with the

depth, to assume this width to continue to the bottom of the second row of sheeting, even though the actual width of the trench is reduced. In rock, where no sheeting is needed, it is simplest and at the same time is perfectly fair to assign some width for all sizes of pipe up to 12 inches, the width varying with the depth of the trench. In sedimentary rock this can properly be made 3 feet for depths up to 8 or 10 feet. For. greater depths, that is 8 to 16 feet, 4 feet and over 16 feet 6 feet are fair widths to be used in computation, irrespective of the actual width of excavation, provided the specifications have been so drawn. For sizes 12 to 24 inches, 1 feet should be added to the above. With the widths thus fixed by the specifications, the computation of earth work consists of multiplying this width by the length of trench excavated and by the depth. The latter is usually taken from the profile.

Records of Y branches is another important part of the surveying to be done during construction, and accuracy here is most important. There are two methods in use. One is to measure carefully from the nearest manhole up grade, afterwards giving the proper station number to the Y with an R or L to indicate which way it looks. The objection to this is that usually the Y's are laid before the manhole is built, and the mason, in building up the manhole, may bring the center of the cover, presumably the center of the manhole, a foot or more out. Measurements, therefore, taken from the assumed center before building, and the cover center after building will not agree, and will make Y's hard to find. A stake may be driven at the assumed center from which measurements are made, or the offset spikes may be used to locate the Y's, but the center of the cover, or the stake driven, must have the correct station number determined or recorded before the stationing of the Y's is made.

Another method of locating the Y's is by reference to side lines of houses which are built near the Y. The record book would then show a sketch as in Fig. 182 and to recover the Y's the side line of the house shown is produced by eye to the middle of the street or to a point about over the sewer pipe, and then the

proper distance measured. For greater certainty, strips of wood such as edgings from a saw mill, or pieces of lath or pieces of telegraph wire, are often left vertically in the trench at the Y so that subsequent excavation may first find the upper end a few inches below the ground surface and then follow down to the Y with perfect certainty of finding it.

CHAPTER XVII.

TRENCHING.

ORDINARILY the methods used in laying pipe interest the engineer only in so far as the safety of the pipe line and the watertightness of the joints are concerned. Since, however, it is sometimes required that the engineer act as contractor and immediately supervise the work, some reference to this part of construction may be made.

The variations in methods of trenching depend on the character of the soil, on the depth of the trench, and on the amount and depth of the ground water. The center line of the trench being laid out, the side lines are marked with a pick, making grooves in the surface, and the laborers are strung out to open up. The width of the trench is determined by the diameter of the pipe, and by the size of the sheeting, if used. The full outside width of an 8-inch sewer pipe at the hubs is 12 inches, and since room must be left outside the hubs for making joints and for correcting the alignment of the trench, a trench two feet wide is the least width to be opened. If sheeting has to be used, assuming 2-inch sheeting with 4-inch rangers, another foot is added, making a 3-foot opening, the narrowest where sheeting is used.

Some contractors, in order to minimize the danger of banks caving, open the trench about 4 feet wide on top, narrowing to 18 inches at the bottom of an 8-foot trench, thus adding about 6 cubic feet of excavation per linear foot, or 22 cubic yards per 100 foot length, an additional cost of about $10.00. The sheeting for 100 feet may be estimated to cost about $50.00, so that the additional excavation is apparently justified, if the sloping banks allow sheeting to be discarded. But practically any trench that will stand with side slopes at such an angle will stand with ver

tical sides and if sheeting is needed with the vertical trench it will also be needed with the sloping sides. It is better, then, to have the trench sides always truly vertical, and then, if bracing becomes necessary, it can be put in.

In rock trenches, the width depends on the character of the rock, on the depth of trench, and on the manner of excavation. If the rock is granite, or igneous formation, without seams, blasting will remove large irregular masses, and the width on top will be nearly equal to the depth of the trench. In sedimentary rock, the strata may be kept broken off so that the width is but little more than that of an earth trench. In deep trenches, however, a batter is gradually acquired, a trench 10 feet deep having a top width of about 4 feet. If blasting is freely resorted to, with deep holes and large charges, the width becomes greater than with shallow holes and small charges, though it is possible without any blasting, in soft sedimentary rock, to carry down a trench, with picks, bull points, wedges, and hammers, and have a trench of about the same width from top to bottom. If the soil is dry clay, or dry clay loam, a trench can be carried down without any sheeting, but à rain storm may flood the trench, soften the clay, and cause the banks to fall in. A dry gravel, or sand without any clay mixture, will need tight sheeting to hold up the banks. Wet clay, or sand, will also need tight sheeting, and in the latter cases a considerable pressure is exerted. A wet trench will always need sheeting, which, in running sand, should be tongue-and-grooved, or provided with splines.

In placing the sheeting, the trench is first excavated to a depth of about three feet, or through the top soil and into the waterbearing strata. Then the two rangers, usually 16 feet long, their size dependent on the estimated pressure, and varying from 4 by 4-inch to 10 by 12-inch, are laid along the trench, one on each side. Between each of these and the side of the trench are placed three pieces of sheeting plank, vertical, one at each end and one in the middle; the rangers are crowded back against the three planks and cross struts, or braces, wedged in tightly and driven into place. Then the trench sides are lined with vertical plank driven down