a rectangle 8 X 17 feet in the clear of all timbers. This allowed ample room for two hoistingcages (elevators), and also a space for a ladder-way and for the air-pipes, etc. The ladders were generally omitted by the contractors. Shaft 24, located on the east side of the Harlem River, was excavated 8 X 10 feet in the clear of all timbers, as it was practically only used for one heading. It was provided with one cage and with a ladder. Shaft 25, located on the west bank of the Harlem River, was excavated 16 X 33 feet in the clear of all timbers, as it was to contain two wells, 12 feet 3 inches in diameter, one forming part of the aqueduct and the other serving as a pump-well for the bailing-buckets used in emptying the siphon under the Harlem River. A detailed description of this important shaft will be found on page 164. Shaft 26 (see page 121), which forms the overflow for the aqueduct on Manhattan Island, was excavated 16 × 16 feet in the clear of all timbers.

To facilitate the work of the engineers in giving line, the shafts, with the exceptions of Nos. 25 and 26, were located with the greater dimension of their cross-section parallel with the centre-line of the tunnel, and were placed to one side of the centre-line of the aqueduct, so that the wires suspended by the engineers for giving the alignment in the tunnel would just clear the timbers of the shaft. A space 15 inches wide was reserved for the wires, to avoid the necessity of stopping the cages while line was being given. The manner in which the shafts were located on the centre-line is shown in Fig. 50, page 175.

With very few exceptions the shafts had to be timbered from top to bottom, in order to sustain the sides of the excavation and to prevent stones from falling down. Frames of 12" X 12" hemlock were placed 3 to 6 feet apart, and supported either by "hitches " (steps) cut in the rock or by being suspended by iron bolts from the frames above. The "lagging" on the outside of the frames consisted of 3" X 12" planks. The space between the "lagging" and the rock was filled by cord-wood or loose stones carefully packed.

In some cases, especially in Shaft No. 2, the timbering consisted entirely of heavy frames, one placed above the other.

In good rock the shafts were sunk 50-100 feet before being timbered. In these cases the timbering was done from the bottom upwards, the lowest frame being well supported by hitches." Plate 55 illustrates how the shafts were timbered. The weekly progress in sinking the shafts is given in the table on page 298.

With few exceptions, the shafts used during the construction of the work were lined with masonry after the aqueduct had been constructed, and made permanent places for entering the tunnel for inspection or repairs. They serve also as ventilators and as openings where the air can escape when the conduit is being filled, except on the lower part of the work, where they are subjected to water-pressure and had consequently to be closed by manhole-covers.

66

The manner in which the shafts not subjected to water-pressure were built up is shown in Plate 57. At the bottom of the shaft, a working-chamber having an elliptical cross-section (6 x 12 feet) was built just above the conduit, in the roof of which an opening 6 feet in diameter was left. At a height of 10-20 feet above the conduit the cross-section of the working-chamber was gradually reduced until at a height of 20-35

feet it was circular, the inner diameter being 6 feet. This cross-section was continued to the top.

The elliptical working-chamber was lined with 24 inches of brickwork, backed with rubble masonry to the sides of the shaft excavation. The circular part of the shaft was lined with 16 inches of brickwork, the space between the brickwork and the sides of the shaft excavation being packed with dry filling. About every 50 feet a brick arch, 24 inches thick, was built across the shaft excavation (an opening being left for the manhole-shaft) to carry the weight of the dry filling. A cast-iron ladder fastened to the brickwork was placed in each shaft, landings being provided at intervals of about 35 feet to enable the workmen to pass each other. Over eleven of the shafts masonry head-houses were constructed (see Plate 57), each being provided with a large sheave placed directly over the shaft for a hoisting-rope. The shafts "under pressure" were finished as shown on Plate 58.

Shaft 11C, at the south end of the siphon under Gould's Swamp, was built up so as to have a circular section 14 feet 3 inches in diameter, in order to provide sufficient room for lowering a pumping-apparatus into the tunnel for emptying the siphon should it become necessary. The shaft is closed with a special iron cover.

Two portable hoisting-plants were purchased and stored in brick engine-houses near the aqueduct, one at the Yonkers blowoff gate-house and the other at the Pocantico gate-house. Each plant consists of:

Ist. A twenty-horse-power portable boiler.

2d. A hoisting-engine with the necessary rope and two iron buckets.

3d. A water-wagon.

4th. A truck large enough to carry 1 tons of coal. It is provided with a chest containing necessary working tools, etc. The hoisting-plants are to be used whenever repairs may be required, and also to furnish the power for lowering and raising buckets for inspection,

etc.

Shaft 26 (see Plate 59) near Tenth Avenue is the overflow-shaft for Manhattan Island. It is built up from the top of the tunnel to elevation 118.0 as a well, 12 feet 3 inches in diameter, lined with 16 inches of brickwork which is "backed" to the solid rock with rubble masonry. A cast-iron ladder, fastened to the brickwork, extends from the top to the bottom of the shaft. At elevation 118.0 the well is finished by a granite coping and covered by a strong brick vault, to which access is obtained from the street by means of two 5-foot manholes. Two sets of granite stones are built opposite each other in the walls of the vault, on the side of the well towards the Harlem River. In each set of stones is cut a groove (9 inches deep and 20 inches wide), into which a A third casting is placed half-way between the two just mentioned. The castings have grooves for two sets of

casting is fastened.

6" × 12" stop-planks, between which a puddling of clay, 6 inches thick, is rammed. An overflow-dam, the height of which can be varied by means of the stop-planks, is thus formed. In front of this dam a small basin is constructed from which two 48-inch cast-iron pipes lead the overflowing water to the Harlem River. The aqueduct on Manhattan Island is thus protected from being exposed to any undue pressure.

At elevation 136 the vault has an iron flooring, made of "I" beams and cast-iron plates. Tunnel. The alignment and profile of the aqueduct tunnel are given on Plate 54. The depth of the tunnel below the surface of the ground varies from 50-500 feet. The tunnel under the Harlem River is about 300 feet below high tide. Where the aqueduct is less than fifty feet below the surface, the excavation was generally made in open cut.

The tunnels between the different shafts were excavated by driving first a "top heading" (upper half of tunnel) and removing then the "bench" (lower half of tunnel). The only exception to this method was a stretch of 2215 lineal feet on section 2, which was excavated by means of a bottom heading.

The heading and bench were usually excavated simultaneously, the latter being kept to within 50-75 feet of the former. The drilling was performed by means of percussion-drills operated by compressed air, supplied from compressors located on the surface, near the shafts. Dynamite, forcite powder, and other similar explosives were used for blasting, the firing being done by electricity. Some details of the manner and cost of excavating the tunnels are given in Chapter VII.

The material excavated (“muck") was removed in cars, which were hauled on tracks by mules, or, for short distances, shoved by hand to the shaft, where they were hoisted on cages" (elevators) to the surface. Each car held about a cubic yard of broken stone. The muck was dumped near the shaft, the good stone being used subsequently for the rubble masonry in the tunnel and shafts. Plate 56 shows a shaft head-house, cage and cars.

In the heading the muck was generally removed in wheelbarrows and dumped from planks, supported by beams ("runners") placed across the tunnel, into the cars at the foot of the bench.

In some cases, after the tunnel had advanced a considerable distance from the shafts, the headings were driven alone to a meeting, the work on the bench being temporarily stopped. This was done with the twofold object of improving the ventilation in the tunnel (which was generally abominable) and of avoiding excavating the bench until the stones. could be used for the rubble masonry. In these cases, an incline, about 400 feet long, was excavated from the heading to the bench, the tracks being laid in the heading as the excavation advanced. Stones to be used for rubble masonry were occasionally piled up on both sides of the tunnel. The tunnels were illuminated by electric lights, gasoline torches, miners, lamps, and, in laying the masonry in the top of the arches, by candles. In the early part of the work, gasoline torches, attached to the side of the tunnel, were used extensively. Each torch had a flat tin reservoir, which held about 3 gallons of gasoline or naphtha, and had to be refilled every five hours. From the bottom of the reservoir a small gas-pipe extended downwards for about 2 feet and then horizontally for 10 inches, terminating in a burner. Three or four of these torches were used in the heading and six to eight for the masons.

The air in the tunnel became so vitiated by these lights that the Aqueduct Commissioners, after an investigation by experts, forbade their use in the tunnel. At times the smoke from these lamps made it impossible to see a light at a distance of 25 feet. In 1886 electric lights were introduced in all the tunnels, but as the lights were generally placed 500