of the work, the contractor devoted his entire attention to the work of excavation. Nearly all the excavation east of Fifth Avenue was done before any of the lining was placed. At a number of points west of Fifth Avenue and at a few points to the east the nature of the rock was such that the two operations had to be done simultaneously.

Single-Tunnel Method.—For an average distance of 350 ft. west from the First Avenue Shafts there were four single tunnels. The rock was sound and comparatively dry. A top heading of the full size of the tunnel and about 8 ft. high was first driven. It was drilled by four drills mounted on two columns, and was blasted in the ordinary way. The bench was about 13 ft. high. Tripod drills, standing on the bench, drilled the usual holes, but, owing to the lack of head-room, steels long enough to reach the bottom of the bench could not be used. Tripod drills were set as low as possible at the foot of the bench and drilled lifting holes. These holes were inclined downward from 10° to 15° to the horizontal, and were spaced to converge at the location of the drainage ditches. The heading was usually driven from 10 to 20 ft. in advance of the bench. At this distance a large part of the muck from the heading was shot backward over the bench. In the single tunnels the muck was loaded by hand.

Twin-Tunnel Methods.—From the end of the single-track tunnel westward to Fifth Avenue on 33d Street, and to Madison Avenue on 32d Street, with some exceptions, each pair of tunnels was excavated for the entire width at one operation. Three different methods of work were extensively used. They were the double-heading method, the center-heading method, and the full-sized-heading method, and these differed only in the manner of drilling and blasting. The bench was usually within 10 or 15 ft. of the face of the heading, and was drilled and fired in the same way as in the single tunnels. After the installation of the permanent plant, most of the muck was handled by steam shovels.

In the double-heading method, shown on Plate LVII, the top headings for each tunnel of the pair were driven separately, leaving a short rock core-wall between them. The headings were drilled from columns in the manner described for the single tunnels. The temporary rock dividing wall between the headings was drilled by a tripod drill on the bench of one of the headings, and was fired with the bench.

In the center-heading method, also shown on Plate LVII, only one heading was driven. It was rectangular in shape, about 8 ft. high and 14 ft. wide. It was located on the center line between the tunnels. In general, the face was from 6 to 12 ft., or the length of one or two rounds, in advance of the remainder of the face at the top. The center heading was drilled by four drills mounted on two columns. By turning these drills to the side, they were used for holes at right angles to the line of the tunnels, by which the remainder of the face of the heading was blasted. By turning the drills downward, the bench holes under the center heading were also drilled. The center heading explored the rock in advance of the full-width heading, and gave a good idea as to the care needed in firing.

For the full-width-heading method, Fig. 2, ten drills were mounted on five columns set abreast across the face. Holes were drilled to form a cut near the center line between the tunnels. The remainder of the holes were located so that they would draw into the cut. The bench was frequently drilled from the same set-up of columns by turning the drills downward. In sound rock this method proved to be the most rapid of any.

Practically all trimming was left until immediately before the concreting. It was then taken up as a separate operation, but proved to be costly and tedious, and a hindrance to the placing of the lining.

Materials Encountered.—All the rock encountered was the familiar Hudson schist, but it varied widely in its mineral constituents and in its physical characteristics. In many places where the rock surface was penetrated, a fine sand was found that was probably quicksand. The material above the rock in the open-cut sections was mostly sand.

Fig. 2.—Method of Excavating with Full-Width Heading Cross-Town Tunnels, Manhattan.

The concurrence of the watercourse, shown on General Viele's map of Manhattan Island (Plate IX[D]), with the points where difficulties in the construction of the tunnels were encountered has been noted in a previous paper.

In all cases where the course of this ancient stream was crossed (except at its final intersection of 33d Street), the rock was found to be very soft and disintegrated, a large quantity of water was encountered, and heavy timbering was required. The construction at these localities will be taken up later. In addition, disintegrated rock, but of a less troublesome character, was invariably met under the depressions in the rock surface developed by the borings from the streets and test holes from the tunnels. Many of these places required timbering, and no timbering was elsewhere necessary except at the portals. These coincident conditions were especially marked in 32d Street, which for a long distance closely adjoins the course of the former creek.

Disposal of Spoil.—The materials excavated from the tunnels were dumped at the 35th Street pier on barges furnished by the Railroad Company under another contract, and were towed to points near the Bayonne peninsula where the spoil was used principally in the construction of the Greenville Freight Yards and the line across the Hackensack Meadows to the tunnels. Details of this work will be given in a subsequent paper. After December, 1907, when the excavation was about 85% completed, the contractor furnished the barges and effected the complete disposal of the spoil.

Difficulties of Excavation.—As stated in a previous paper, the excavation of the Twin Tunnel in 33d Street was continued westward to the west line of Fifth Avenue on the original grade. At that point the contractor started three drifts in the three-track section. The relation of the drifts to each other and to the cross-section are shown by Fig. 3. The center heading was driven a little in advance of those on the sides. At a distance of 65 ft. west of Fifth Avenue the rock surface was broken through in the top of the heading, and a very fine sand was encountered. For some distance east of this point the rock was badly disintegrated, and the heading required timbering. Through the soft material, tight lagging was placed on the sides and roof of the heading, and the face was protected by breast boards. There was a moderate flow of water through the cracks, and, in spite of every effort, some of the fine sand was constantly carried into the heading.

In one or two instances considerable ground was lost at the face. On the evening of