the level to the correct staff reading at B will not appreciably affect the staff reading at A, as the distance da is very much less than the distance dB. In any case the second adjustment of the level to the proper staff reading at в will probably not affect the reading at A. The adjustment will therefore, as a rule, be effected at the second trial. When finally adjusted see that both staff readings at A and B are correct when the bubble is at the centre of its run. The adjusting of the bubble by the screws at e, Fig. 104, sometimes moves the level and alters the staff readings, and it is advisable therefore to try both staff readings after moving the screws at e.

3. To make the Telescope and Spirit Level together Perpendicular to the Vertical Axis, or to make the Instrument "traverse," i.e., so that the bubble remains in the centre of its run in every position of the telescope.

Bring the telescope over one pair of screws and level it up. Now turn it through 180° so as to reverse the ends of the telescope. If the bubble remains in the centre of its run the adjustment is correct; if not, correct half the deviation by means of the screws at t, t, Fig. 104, and half by means of the levelling screws s, s. Repeat the operation until the bubble remains in the centre of its run, when the telescope is reversed end for end. The instrument when levelled over both pairs of screws, or in the three screw instrument over one pair and then over the third screw, will then remain in the centre of its run in every position of the telescope. It is most generally useful to have this adjustment correct when a great many intermediate sights have to be taken, as in cross sectioning; but it is very liable to get out, and seldom remains for long correct. It does not, however, appreciably affect the accuracy of the work; it merely saves time in levelling up the instrument for each sight taken.

As to centring of object glass and eyepiece, horizontality of cross hairs, these adjustments are to be made as described for the theodolite, pages 209, 210.

Focussing the object glass and eyepiece, or adjustment for parallax, comes under the head of temporary adjustments, and are to be made as described for the theodolite, page 203.

Adjustment of the Y Level.-The Y level may be adjusted by twisting the telescope half round on its axis in the Y's, and noting whether the intersection of the cross hairs still continues

to coincide with the same object. If not, half the deviation is corrected by the diaphragm screws. This makes the collimation line coincide with the axis of the Y's, with which the axis of the telescope is supposed to coincide.

To adjust the bubble, level up the instrument and turn it end for end by lifting out the telescope and replacing it. If the bubble deviates from the centre of its run, half the deviation is corrected by the screws attaching the level to the telescope and half the deviation by the levelling screws. This makes the bubble parallel to the Y's, and therefore parallel to the collimation line if the collimation line has been previously adjusted to be parallel to the Y's. Any inequality in the size of the rings on the telescope on which it rests in the Y's will, however, affect the adjustment. See Precise Levelling "pivot correction," pages 192, 193, 195.

It is, however, much more accurate to adjust the Y level by the "peg methods" already described. The Y level is made to "traverse" as already described.

See also instructions for precise spirit levelling, pages 194 to 201, for adjustments of Y level.

CHAPTER V.

RAILWAY SURVEYS AND SETTING OUT.*

Home and Foreign Work. As railway surveys at home are conducted under very different conditions from similar surveys abroad, this chapter is confined to home surveys for railways and to the setting out and incidental work prior to and during construction, reserving for Chapter XI. railway surveys abroad, in jungle, dense forest, and unmapped country.

Parliamentary Surveys. The preliminary or first survey for a railway is called the Parliamentary survey, and is so called because its object is the preparation of the plans, sections, and estimates for submission to Parliament in order to get its sanction to the proposed railway.

Ordnance Survey Maps.-On the Ordnance Survey maps of the United Kingdom it is possible to lay down a line between any two points with a very fair approximation to the best location. The 6 in. Ordnance maps are usually used for this purpose, as contours are given on them, while the 25 in. maps have no contours; the 6 in. scale also conforms to the Parliamentary Standing Orders as to scale of deposited plans.

Special Local Considerations. In home work special local considerations to a very great extent determine the route of the line, which has often to be laid out through expensive cutting to avoid some local landowner's property, &c. &c., while detours have to be made to bring the line within easy reach of towns or villages on the route, or to avoid these when opposition is to be feared, and so on. These matters are usually to be considered in consultation with the promoters of the scheme and the local agent.

Ruling Gradient and Minimum Radius of Curve.The next point is to fix upon the ruling gradient and minimum radius of curve. For these, see pages 230, 231, 232.

*

Practically all this chapter also applies to Road Surveys and Setting Out. For surveys abroad see Chapter XI.

Laying down Line on 6 in. Ordnance Map.-It being supposed that these points have been duly considered and determined, the best procedure is to lay down the line on the 6 in. Ordnance sheets in pencil, as nearly as may be judged from the contours. The bench marks and levels along public roads which are given in the map are also of assistance in doing this.

In laying down the line on the map in this manner, we first fix upon the position and level of the starting point and a gradient, then by scaling the proper distance between adjacent contours we arrive at the position of the line. A good way is to calculate the distance required to rise 50 or 100 ft. on the proposed gradient. For instance, if the gradient is 1 in 50, on this we can rise 100 ft. in a distance of 5,000 ft. Take now 5,000 ft. in the compasses and fit them in between successive 100 ft. contours, marking the points where each contour is intersected. Theoretically the line should follow these points, but this will seldom be possible, and the best that can be done is to put in the curves so as to follow this line as nearly as possible and with as little cutting or embankment as may be. In the hilly parts of the country the contours are only given 100 ft. apart vertically, in flatter parts 50 ft. apart, and in the easiest country every 25 ft. When 25 or 50 ft. contours are given, the distance required to rise or fall 25 or 50 ft., as the case may be, is to be calculated, and the compasses stepped between successive contours, as above described.

Fixing Gradients.-The best gradient, when the ground permits of a flatter gradient than the ruling or maximum gradient, can only be found by trial, and as a rule will not be definitely fixed until the longitudinal section is plotted. For location purposes the gradient must be judged from the configuration of the ground as nearly as can be by estimation with the eye. This is therefore to a great extent a matter of practical experience and outdoor practice. In the first instance, however, the gradients to be tried must be found from a careful study of the map and the levels on it.

Reconnaissance and Sketching in Location on Map. -Having now the line thus laid down on the map, get the sheets pasted together, and cut into continuous rolls 15 or 18 in. wide, and as long as possible, or mounted so as to fold up like a pocket map. Taking this along, now walk carefully over the line and sketch in best location that will fit the ground as nearly as can be

judged, taking the pencil line already laid down as a basis. It is advisable to have a pocket scale and a pair of pencil bows and compasses; a hand level will often be useful on very steep ground for getting approximate differences of level. In locating a gradient up or down the steep hillside of a valley, a good plan is to first fix upon the valley or summit level, then calculate the levels of the gradient proposed to be run, and pencil them on the map at every 10 chains, or at fences approximately 10 chains apart. These figures may be pencilled on along the line already laid down on the map from the contours. The points on the ground where the surface level corresponds with these levels may be found by levelling from the nearest contour or other known level with the hand level or by judging with the eye. These points may then be fixed and marked on the map by measuring or stepping distances from fence corners, buildings, &c., shown on the map. A line joining these points is the line the railway should theoretically follow, and the straights and curves may be drawn on indoors so as to follow this line as nearly as possible.

Plotting Sketched Location.-Having thus gone carefully over the ground and sketched in the proper location by hand, taking care to use the compasses or pencil bows in the field so as to avoid sketching in curves of less than the minimum radius, next draw in the curves and straights to coincide with the line sketched in the field as nearly as possible, and mark on the chainage at every 10 chains along the line. It is also a good plan to scale the chainage of each fence crossed and figure it in pencil on the map, as this will save the leveller a considerable amount of time in taking the levels for the section.

Levelling the Longitudinal Section. The next operation is to take levels over the line thus located. The leveller must be supplied with a map, with the line marked on it, and he takes levels at the points where the slope of the ground changes along the line, booking the chainage of each point where a level is taken. As a rule these levels will be taken at the points where the line intersects fences, and he will locate the exact position of the line on the ground by measuring the distance scaled off the map along the fences from the nearest corner or by directing the staff-holder to step so many paces from it. Points intermediate of fences will be fixed in like manner by measuring or pacing along the line from