Tunnelling in clay is frequently attended with formidable difficulties which render it very expensive. It is, when tough, a difficult material to remove, blasting being of no use, and spades and pickaxes being almost inapplicable. Lecount states that in such cases hatchets may be used to advantage, but that cross-cut saws answer best. The Primrose Hill tunnel, on the London and North-Western railway, passing through the London clay, is an example of the difficulties of such a material. The engineers, warned by the failure of the tunnel attempted some years before through the same material at Highgate, on what is now called the Archway road, which fell in owing to the insufficiency of the brick lining, adopted unusual precautions in the first instance, excavating only nine feet in advance of the brickwork, and supporting the clay by very strong timbering until the arching was complete. Owing however to the extraordinary mobility of the moist clay, the pressure upon the brickwork was so great as to squeeze the mortar from the joints, and to bring the inner edges of the bricks in contact. The evil was augmented by the form of the bricks, which, according to the custom in and near London, were made with hollow surfaces, and were consequently unfit to bear great pressure, because their sirs only could come in contact. The result was, as stated in Lecount s' History of the Railway, connecting London and Birmingham,' pp. 32, 33, that " by degrees the bricks were Finding to duet, and the dimensions of the tunnel insensibly, but irresistibly, contracting." This difficulty was overcome by the use of very hard bricks laid in Roman cement, which, by setting hard before the external pressure became so great as to force the bricks into actual contact, enabled the whole surface of the brick, instead of its edges only, to resist the pressure. The thickness of the brickwork was also increased, so that in most parts of the tunnel it amounts to twenty-ecven inches. The occurrence of a similar material in the line of the Fareham tunnel, on the Gosport branch railway, occasioned great expense, and produced a slip of the imperincumbent earth which carried away about forty yards in length of the brick arching, although it was of the unusual thickness of three feet.
Tunnels formed through chalk are often impeded by faults or cavities filled with wet gravel or sand, which poor a flood of semifluid matter into the excavation as soon as they are cut into. The irruption of such loose materials,as well as of water alone, has in many cases occasion( d difficultbe almost insurmountable. In the Watford tunnel. on the London and North-Western railway, which posses through the upper chalk formation, where it is covered with a thick irregular bed of gravel, such breaks occasioned great inconvenience. The chalk had occasional fissures, sometimes as ranch as one hundred feet deep, filled with clean gravel, " which," observes Lecount (' History,' &c., p. 114), " when worked into, rushed down with such violence, as to plough the walls of the tunnel as if bullets had been allot against it." Such an accident, occurring at the foot of one of the working shafts, overwhelmed ten men who were at work in the tunnel, and led to the construction of the large ventilating shaft near the centre of the tunnel, which occupies the site of the cavity. Loose sand is perhaps the most diffioult soil that can be met with in tunnelling, but it has been in several instances successfully passed through. In the tunnel on the Leicester and Swannington railway, one of the earliest railway tunnels, a loose dry running sand was encountered for a distance of five hundred yards, in which it was necessary to make a wooden tunnel to support the soil while the brickwork was executed. When water occurs with the loose soil the difficulty is still greater. This was the case In part of the Wapping tunnel at Liverpool, a portion of which fell in, to a depth of thirty feet from the surface. The Kilsby tunnel, before alluded to, is a more 'striking example of this kind of difficulty. Mr. Stephenson, the engineer, conquered this obstacle by sinking shafts, a little boucle tho line of the tunnel, for pumping out the water, and so draining the sand until it was sufficiently dry for tunnelling. These shafts were
sunk through the quicksand by means of wooden tubbing, and from them headings were driven to collect the water and to conduct it to the pumps. Steam-engines were erected to work the pumps, which were used incessantly for nine months before the sand was dry enough to allow the work to proceed : during a great part of that time two thousand gallons water were removed per minute. When the working of the tunnel was recommenced, headings were driven from the pumping shafts to the bottom of the working shafts, by means of which the tunnel was freed from water. The quicksand extends over about four hundred and fifty yard,' of the length of the tunnel, and Its bottom dips to about six feet below the arch. On the occasion of an imiption of water in another part of the tunnel, in which it was desirable to complete the arching of a portion already executed before it was possible to get rid of the water, that object was effected by floating the men and materials to the spot upon a raft. Water has been met with in large quantities in several other tunnels. It flowed so freely from fissures in the freestone rock through which the Box tunnel is driven, that in November, 1837, the steam-engine employed in pumping proved insufficient, and the water filled ono division of the tunnel, and rose to the height of fifty-six feet in the shaft, thereby suspending the work until the following July, when the water was over come by means of a second engine, of fifty-burse power. After another irruption in the same tunnel, the water was pumped out at the rate of thirty-two thousand hogsheads a day. The progress of the 31enstham and several other tunnels was stopped for a time in like manner. In no case however has the irruption of water or the badness of the ground proved so serious a difficulty as iu the Thames tunnel.
Short tunnels are occasionally excavated from the ends only, but those of considerable length are usually formed by sinking vertical shafts, about nine feet in diameter, down to the level of tho tunnel, and excavating in each direction from the bottom of those shafts, until the several parties of workmen meet in the intermediate portions. By this means the work can proceed at any required number of points or faces, so as to bring the execution of the tunnel, whatever may be its length, within a moderato period of time. The accurate junction of these detached workings is provided for in the following manner : In netting out the tunnel, the engineer plants a transit-instrument in an observatory erected in the line of the tunnel, supporting it on a pier insulated from the building to prevent vibration. If a road happen to pass near the observatory, the ground should be excavated round the pier to a depth of from six to ten feet, according to the traffic, for the same purpose. A distant mark should then be selected in the lino of the tunnel, and a fixed point placed as an adjusting spot for the line of direction, which point should be at a considerable distance. Inter mediate marks for the working and ventilating shafta may then be set out correctly; and an these shafts are slink, the points determined by the trauma-instrument are carried downwards by carefully suspended plummets, which should be of iron, and let down in buckets of water, or, which is better, in cups of mercury, to chock vibration. When the shafts are cleared out at the bottom, other transit-inatruments may be placed in them, the plumb-line and transit being kept as far apart as possible. The intersection of the vertical hairs in the transit with the plumb-line will then enable the engineer properly to set out the work. By these means junctions are effected between the several workings, or shifts, with surprising accuracy. In a length of 1620 feet between two shafts of the Box tunnel, which has a slope of 1 in 100, the junction of the two shifts was perfect in point of level, and did not deviate more than an inch and a quarter in any place at the sides. Even in curved tunnels, although the difficulty us increased, great exactness is attainable. In those on the Glasgow and Greenock railway at Bishopton, the deviation from perfect correctness nowhere exceeded two indica.