Levels. The longitudinal section of the surface must be levelled from end to end of the tunnel and bench marks fixed at

Fig. 201.—Alignment of Dunmail Rise Tunnel, Thirlmere Aqueduct.

convenient points close to each shaft. The levels of the bench marks must be very carefully checked before proceeding to transfer the levels underground.

Transferring Levels down Shafts. To get the levels down each shaft the best plan is to accurately determine the level of a point marked on one of the slides or guides in which the cage travels, by levelling from the bench mark established at that shaft. Another point is now marked on the guide near the bottom of the shaft and the distance between these two marks is then measured with a steel tape. When the depth of the shaft exceeds 100 ft., so that this distance cannot be measured at one operation, it must be

* Minutes Proceedings Inst. C. E., vol. cxvi.

measured in tape lengths by descending in the cage 66 ft. or 100 ft. at a time, temporary platforms being fixed in the shaft to enable the assistant to hold the end of the tape. When wire rope guides are used for the cage it becomes necessary to make the measurements on the side of the shaft itself, in which case care must be taken that the tape is held vertical at each operation. The whole measurement must be carefully repeated and the results should practically coincide if the measurements are made with sufficient

care.

Underground Bench Marks. Having thus obtained the level of the mark at the bottom of the shaft, the staff is held on it, the level being set up near the bottom of the shaft, and a permanent bench mark is established.

BM

Fig. 202. Underground Bench Mark.

Iron wedges or nails driven into the side of the tunnel at a convenient height form good bench marks, and should be marked by a broad arrow in white or red paint so as to be easily found. When the centre line is fixed by pegs surrounded by brick piers as described on page 298, these may also be used as bench marks. In the case of iron-lined tunnels the flanges of the segments form convenient points for the bench marks, as shown in Fig. 202.

Checking Underground Levels and Centre Line through between Shafts. As the tunnelling proceeds new bench marks are conveniently fixed near to the working face of the tunnel from time to time, and the levels are thus carried forward until the faces of the headings meet, when the levels may be checked through from each shaft and the error ascertained. As headings are usually driven forward a certain distance ahead of each face, any small discrepancy between the levels from each shaft may be adjusted by putting in a "junction gradient" between the already executed portions of the tunnel. This also applies to the centre line, any error in which is rectified by putting in an extremely flat curve, or when the error is inconsiderable by putting in a short straight to join up the two lines.

Setting Ribs and Profiles. In the case of brick-lined tunnels when there is an invert and side walls there are profiles

erected to guide the bricklayers in laying the brickwork. These have to be carefully set to the correct level by the engineer, and are adjusted to his direction by the foreman in charge. The operation of setting these profiles simply consists in levelling from the nearest bench mark and directing them to be raised or lowered until they are at the correct level, but it requires considerable care, as any mistake leads to very serious results. In the case of a new "break up," as the commencement of the widening out of the heading at any point to full tunnel section is called, a mistake is specially liable to be made as there is no adjacent finished work to check from. In setting profiles a level should always be taken on the corresponding part of the last executed portion of tunnel as a check.

When the tunnel is in good rock, invert and side walls are sometimes dispensed with, and in that case it is the ribs of the centring of the arch which are to be set. The ribs are supported on wedges by means of which they are raised or lowered to the desired level. Both springings of the ribs are to be levelled, and the level of the crown of the rib should also be taken.

As regards setting ribs and profiles to correct line, the centre line should be ranged out with the theodolite and points marked on the centre line opposite each rib or profile to be set. The latter are then adjusted to line by measuring the proper half width from the marks with a steel tape. After each alteration of the rib or profile for level it should be tested for line and vice versa.

Setting Segments of Iron-lined Tunnels. In the case of iron-lined tunnels each ring is bolted in succession to the last executed portion, and is therefore to a certain extent fixed by what has gone before. Any adjustment required is made by packing in between the rings with wood packing. Thus if the last ring is slightly too low or too high, or an alteration in the gradient occurs, the rectification is effected by packing in the bottom or in the top. Similarly if the last ring is slightly off the centre line, or if there is a curve in the centre line, the adjustment is effected by packing in the left side or in the right side as the case may be.

In the case of iron-lined tunnels, therefore, the engineer may test the last executed ring of the tunnel for centre line and level, and order packings accordingly. As regards testing for centre line, the best plan is to string a line across the tunnel, fasten

ing it to the bolts of the segments, and then range in a plumb bob hung from this line, as shown in Fig. 203. The steel tape is then stretched across a horizontal diameter and the measurements to the plumb line will indicate whether or not the position of the iron lining is correct.

As regards the levels, the crown and invert of the last rings may be levelled and packings ordered accordingly.

Taking Level of "Crown" of Tunnel.-In taking the level of the "crown" of a tunnel (the highest point of the soffit),

String

Tape Line

Plumb-bob

Fig. 203.

Checking Segments of Iron-lined Tunnels.

hold the staff upside down on the "crown," and book with a plus sign in the level book.

Alpine Tunnels:* Mont Cenis Tunnel.-The first long tunnel through the Alps was the Mont Cenis Tunnel. It was driven in a straight line from end to end with a gradient of 1 in 43 on the French side, and gradients of 1 in 2,000 to 1 in 1,000 on the Italian side, rising on the French side to attain the higher level of the Italian end, and on the Italian side only to get a fall for the drainage. The length of the straight tunnel was 7.6 miles, and this was increased to 7.97 miles by two curves subsequently made to connect with the approaches at each end. The curve

* Minutes Proc. Inst. C.E., vol. xcv., "Alpine Engineering," by L. F. Vernon-Harcourt, M. A., M. Inst. C.E.

at the French end was 464 yds. long, radius 24.8 chains, on a gradient of 1 in 43. The curve at the Italian end was 891 yds. long, part of it being the same radius and on a gradient of 1 in 333. The object of driving the heading right through on the straight from end to end was of course to ensure accuracy in the alignment. The exact line of the straight tunnel was fixed by very careful triangulation and marked by permanent marks on the surface. The line was ranged into the headings from observatories erected on the line on the further side of the valleys opposite each end of the tunnel in the same manner as described on page 286.

The work of driving was commenced at both ends towards the end of the year 1857, and occupied thirteen years and one month. When the headings met the tunnel proved to be 45 ft. longer than calculated, while the heading on the French side was 1 ft. too high in level, probably owing to the miscalculation of length. The direction of the line was, however, quite correct, proving the care with which the triangulation and the alignment of the headings was effected. The average daily progress was 2.57 lin. yds., and the total cost of the tunnel £3,000,000, or about £224 per lineal yard. The tunnel was opened for traffic towards the end of 1871, or nearly fourteen years after its commencement.

St Gothard Tunnel. This tunnel was driven in a straight line from end to end, and constructed for a double line throughout. The gradients were 1 in 172 on the north side and 1 in 500 to I in 2,000 on the south side, the gradient on the south side being simply to ensure drainage. The length of the straight tunnel was 9.26 miles, but with curves afterwards put in at the ends to connect with the approaches the finished length of the tunnel is 9.31 miles. The driving of the tunnel was commenced in September 1872, and dynamite was used for blasting, whereas gunpowder only was used in the Mont Cenis Tunnel. The headings met on the 29th February 1880, and the length driven from the northern end was 631 yds. more than from the southern end. The driving of the headings occupied seven years and five months, or little more than half the time of the Mont Cenis Tunnel. The average daily progress was 6.01 yds. as compared with 2.57 yds. in the Mont Cenis. When the headings met the error in the line was 13 in. and the error in level 2 in. The length of the tunnel was 25 ft. less than calculated. The greatest depth of the tunnel