The allowable error in closure depends on many things, the chief of which is financial. If it will cost more to locate and correct the error than the value of the land saved, it will not pay to do so. In other words, far more time (consequently, money) may be spent upon a survey of coal land worth $2,500 an A. than upon land costing but $5 an A. In ordinary rolling country, such as prevails in the coal fields of the eastern states, with instruments in good adjustment and using ordinary precautions, the error in closure should not be greater than 1 ft. in 3,000 ft. to I ft. in 5,000 ft., and trained corps will do better. In bituminous mines, which are commonly in flat coal, underground surveys may easily be closed within 1 ft. in 10,000 ft. to 1 ft. in 20,000 ft. The higher accuracy obtainable underground is due chiefly to the fact that mine temperatures are extremely uniform so that corrections for expansion or contraction of the tape are unnecessary. Likewise, the tape is stretched on the ground and errors due to sag are thus eliminated.

LEVELING

DESCRIPTION OF INSTRUMENTS

In leveling, but two instruments are used, the level and a leveling rod. The level consists of a telescope to which is fitted, on the under side, a long level tube. The telescope rests in a Y at each end of a revolving bar, which is attached to a tripod head very similar to that used for a transit. The telescope is similar to the telescope of a transit.

The leveling rod is merely a straight bar of wood, 6 ft. or more in length, divided into feet and tenths of a foot. A target divided into four equal parts by two lines, one parallel with the staff, and the other at right angles to it, and painted red and white, so as to make it prominent at a distance, slides on the rod and is provided with a clamp screw. The center of the target is cut out and a vernier, graduated decimally, is set in, which enables the rodman to read as close as robo ft. If a long rod is required, it is made of two sliding bars, which, when closed, are similar to a single rod, as described above. When used at points where it is necessary to shove the target to a greater height than 6 or 6 ft., the target is clamped at the highest graduation on the front of the rod, and the rod is extended by pushing up the back part, which carries the target with it. The readings, in this case, are made either from the vernier on a graduated side, or a vernier on the back. The rodman must always hold his rod perfectly plumb or perpendicular.

LEVEL ADJUSTMENTS

The proper care and adjustment of the level is of great importance. A very slight error in adjustment will completely destroy the utility of any work done. 1. To Adjust the Line of Collimation.-Set the tripod firmly, remove the Y pins from the clips, so as to allow the telescope to turn freely, clamp the instrument to the tripod head, and, by the leveling and tangent screws, bring either of the wires upon a clearly marked edge of some object, distant from 100 ft. to 500 ft. Then with the hand, carefully turn the telescope half way around, so that the same wire is compared with the object assumed. Should it be found above or below, bring it half way back by moving the capstan-headed screws at right angles to it, remembering, always, the inverting property of the eyepiece; now bring the wire again upon the object, and repeat the first operation until it will reverse correctly. Proceed in the same manner with the other wire until the adjustment is completed. Should both wires be much out, it will be well to bring them nearly correct before either is entirely adjusted. 2. To Adjust the Level Bubble.-Clamp the instrument over either pair of leveling screws, and bring the bubble into the center of the tube. Now turn the telescope in the wyes, so as to bring the level tube on either side of the center of the bar. Should the bubble run to the end, it shows that the vertical plane, passing through the center of the bubble, is not parallel to that drawn through the axis of the telescope rings. To rectify the error, bring it by estimation half way back, with the capstan-headed screws, which are set in either side of the level holder, placed usually at the object end of the tube. Again bring the level tube over the center of the bar, and adjust the bubble in the center, turn the level to either side, and, if necessary, repeat the correction until the bubble will keep its position, when the tube is turned in. or more to either side of the center of the bar. The necessity for this operation arises from the fact that when the telescope is reversed, end for end, in the wyes in the other and principal adjustment of the bubble, it is not easy to place the

level tube in the same vertical plane, and, therefore, it is almost impossible to effect the adjustment without a lateral correction.

Having now, in a great measure, removed the preparatory difficulties, it is possible to proceed to make the level tube parallel with the bearings of the Y rings. To do this, bring the bubble into the center with the leveling screws, and then, without jarring the instrument, take the telescope out of the wyes and reverse it end for end. Should the bubble run to either end, lower that end, or, what is equivalent, raise the other by turning the small adjusting nuts, on one end of the level, until, by estimation, half the correction is made; again bring the bubble into the center and repeat the whole operation, until the reversion can be made without causing any change in the bubble. It is well to test the lateral adjustment, and make such correction as may be necessary in that, before the horizontal adjustment is entirely completed.

3. To Adjust the Wyes.-To adjust the wyes, or, more precisely, to bring the level into a position at right angles to the vertical axis, so that the bubble will remain in the center during an entire revolution of the instrument, bring the level tube directly over the center of the bar, and clamp the telescope firmly in the wyes. Place it, as before, over two of the leveling screws, unclamp the socket, level the bubble, and turn the instrument half way around, so that the level bar may occupy the same position with respect to the leveling screws beneath. Should the bubble run to either end, bring it half way back by the Y nuts on either end of the bar; now move the telescope over the other set of leveling screws, bring the bubble again into the center, and proceed precisely as just described, changing to each pair of screws, successively, until the adjustment is very nearly perfected, when it may be completed over a single pair.

The object of this approximate adjustment is to bring the upper parallel plate of the tripod head into a position as nearly horizontal as possible, in order that no essential error may arise, in case the level, when reversed, is not brought precisely to its former situation. When the level has been thus completely adjusted, if the instrument is properly made and the sockets are well fitted to one another and the tripod head, the bubble will reverse over each pair of screws in any position. Should the engineer be unable to make it perform correctly, he should examine the outside socket carefully, to see that it sets securely in the main socket, and also notice that the clamp does not bear upon the ring that it encircles. When these are correct, and the error is still manifested, it will probably be in the imperfection of the interior spindle. After the adjustments of the level have been effected and the bubble remains in the center in any position of the socket, the engineer should carefully turn the telescope in the wyes, and sighting upon the end of the level, which has the horizontal adjustment along each side of the wye, make the tube as nearly vertical as possible. When this has been secured, he may observe, through the telescope, the vertical edge of a building, noticing if the vertical hair is parallel to it; if not, he should loosen two of the cross-wire screws at right angles to each other, and with the hand on these, turn the ring inside, until the hair is made vertical; the line of collimation must then be corrected again, and the adjustments of the level will be complete.

USING THE LEVEL

When the instrument is being used, its legs must be set firmly into the ground, and neither the hands nor person of the operator be allowed to touch them. The bubble should then be brought over each pair of leveling screws successively, and the instrument leveled in each position, any correction being made in the adjustments that may appear necessary. Care should be taken to bring the wires precisely in focus, and the object distinctly in view, so that all errors of parallax may be avoided.

An error of parallax is seen when the eye of an observer is moved to either side of the center of the eyepiece of a telescope, in which the foci of the object and eyeglasses are not brought precisely upon the cross-wires and object; in such a case, the wires will appear to move over the surface and the observation will be liable to inaccuracy. In all instances, the wires and object should be brought into view so perfectly that the spider lines will appear to be fastened to the surface, and will remain in that position however the eye is moved.

If the socket of the instrument becomes so firmly set in the tripod head as to be difficult of removal in the ordinary way, the engineer should place the palm of the hand under the Y nuts at each end of the bar, and give a sudden upward shock to the bar, taking care, also, to hold his hands so as to grasp it the moment it is free.

FIELD WORK

If the survey has been carefully made and vertical angles taken at every sight, leveling will be necessary only in cases where extreme accuracy in regard to vertical heights is necessary. In most cases of practical work at collieries, particularly in determining thickness of strata, general rise or fall of an inside road, etc., the elevations calculated by the use of the vertical angle will be close enough, but there are frequently instances when leveling must be done, to insure success in certain work. In this connection, it is well to state that if the transit telescope is supplied with a long level tube, and it is, as a whole, in first-class adjustment, levels can be successfully run with it if the transitman uses due care. Having his instrument in proper adjustment and his notebook ruled, the levelman is ready to proceed with the work.

"

The rodman holds the rod on the starting point, the elevation of which is either known or assumed. The levelman sets up his instrument somewhere in the direction in which he is going, but not necessarily, or usually, in the precise line. He then sights to the rod and notes the reading as a backsight or + (plus) sight, entering it in the proper column of his notebook, and adding it to the elevation of the starting point as the "height of instrument.' The rodman then goes ahead about the same distance, sets his rod on some welldefined and solid point, and the levelman sights again to the target, which the rodman moves up or down the rod until it is exactly bisected by the horizontal cross-hair in the telescope, as he did when giving the backsight. This reading is noted as a foresight or - (minus) sight. The foresight subtracted from the height of instrument gives the elevation of the second station. The rodman holds this latter point, and the levelman goes ahead any convenient distance, backsights to the rod, and proceeds as before. In this case, it is assumed that levels are only being taken between regular stations or two extreme points.

If a number of points in close proximity to each other are to be taken, the rodman, after giving the backsight, holds his rod at each point desired. The readings of any number in convenient sighting distance are taken and recorded as foresights, and any descriptive notes are made in the column of remarks. These are each subtracted from the height of instrument, and the elevation found is noted in column headed Elevation. After all the intermediate points are taken, the rodman goes ahead to some well-defined point, which is called a turning point (T. P.) in the notes. The elevation of this is found and recorded. The rodman remains at this point until the levelman goes ahead, sets up and takes a backsight. This backsight reading, added to the elevation of the turning point, gives a new height of instrument from which to subtract new foresights, and thus obtain the elevation of the next set of points sighted to. When running levels over a long line, the levelman should set frequent bench marks (B. M.). These are any permanent well-defined marks that can be readily found and identified at any future time. By leveling to them he has secured the elevation of points from which to start any subsequent levels that may be necessary. A good bench mark can always be made on the side or root of a large tree or stump by chopping it away so as to leave a wedgeshaped projection with the point up. A nail should be driven in the highest point of this, to mark where the rod was held, and the tree or stump blazed

above the bench mark. In this blaze, the number of the bench mark, which should, of course, correspond with the number in the notebook, should be cut or painted. In the mines, prominent frogs or castings in the main roads, if permanent, make good bench marks.

In underground leveling, extreme care must be observed to record the algebraic signs of the readings, which show whether the level rod was beld in its usual position, indicated by a + sign or the absence of any sign, or upside down, indicated by the --sign.

Proof of Calculations.-The calculations are proved by adding together the backsights and also the foresights taken to turning points and last station. Their difference equals the difference of level between the starting point and last station. Thus:

Trigonometric leveling determines the difference in elevation between two points from the measurement of the distance between the points, and from the vertical angle between them. Although generally less accurate than leveling with a Y level, it is much more rapid and is especially adapted for pre

FIG. 1

liminary work in a hilly country, or for the leveling of mine slopes and pitching rooms where the Y level cannot be used with any advantage or accuracy. By reading the angles and by checking the measurements, a very high degree of accuracy can be obtained in trigonometric leveling.

Case 1.-Assume the elevation of A, Fig. 1, to be 100 ft. above tide. With the transit set up over A and properly leveled, sight to a point C on a rod so that BC equals AD. Measure the

vertical angle Z and the inclined distance DC, then the difference in the elevation between A and B equals BC=CDX sin Z, and the elevation of B equals 100+BC.

Case 2.-Assume the elevation of station A, Fig. 2, in the roof of a mine to be 100 ft. above tide. Then, with the transit set up directly under A and properly leveled sight to a point C upon the plumb-line suspended from the station B, measure the vertical angle X, inclined distance DC, and roof distance BC. From this, the distance CY = DC X sin X. The elevation of B is then found as follows: The elevation of B elevation of A-AD+(DCX sin X) +BC.

FIG. 2

There are many modifications of this simple method, but from these diagrams the most complex modifications can be worked out. TRIGONOMETRIC LEVEL NOTES

CONNECTING OUTSIDE AND INSIDE WORK THROUGH SHAFTS AND SLOPES

SURVEYING SHAFTS

As the dip of the bed increases, it becomes more difficult to make a connection and the chances of accuracy diminish. In the survey of a pitching plane, one station is located, with respect to the adjacent ones, by multiplying the distance by the cosine of the vertical angle. The greatest angular accuracy for a given distance is where the vertical angle is 0°. As the pitch or vertical angle of sight increases the cosine diminishes until, at a vertical sight, distance Xcos. vertical angle =0°.

In the case of an adit level, or a slope of less than 45°, there is no difficulty beyond the want of absolute rigidity in setting up the transit, and the danger of moving it in going about it. The difficulty increases more rapidly than does the pitch, and as the distance X cos. vertical angle diminishes, though the distance is fixed, the chances of error increase. When the slope reaches 60°, there is an impracticability in running a line down a slope, as the line of collimation of the telescope strikes the graduated limb of the instrument. A person can use a prismatic eyepiece and see up the slope; but cannot look down. As it is assumed that it is unnecessary to use an additional telescope, the line must be run by intermediates. To do this, the transit should be set up at the bottom of the slope where the longest sight up the same can be secured and a backsight taken on a station of the underground work; or a backsight should be set for the occasion (both stations will afterwards be connected with the work below). With the prismatic eyepiece, a sight should be taken up the slope on a line that will give the longest sight and, at the same time, afford a good intermediate place to set up the transit, as, on a pitch of 60° or more, it is absolutely necessary that the legs of the transit should be set solidly (in holes in the floor, or between the sills of the track) so that they will not be moved by subsequent walking about it. By this method, all the sights will be taken from one side alone, and the tripod legs can be shortened to make the sight possible without building a standing place-if the man is short.

Call this station A; at the foot of the slope locate B, where the transit can be readily set up, and as far up the slope as possible (this distance must be at least 100 ft.), and in a continuation of AB, locate C. Set up at B and take foresight to C; locate D under the same conditions that governed the placing of B, and, in a continuation of the line BD, place E. Set up at D with foresight at E, and locate F and G as before. The survey is carried by the intermediates B, D, F, etc., to the top, by a series of foresights to C, E, G, etc.

The term shaft in American coal-mining practice is applied only to vertical openings, though in metal mining, both in the United States and abroad, it is also applied to highly inclined slopes. For such shafts, most of the methods given in the textbooks are worthless, as they are for transit work and the distance Xcos. vertical angle in rare cases may be as great as 20 ft., while the distance varies from 100 to 1,500 ft. Again, to sight down a shaft necessitates the erection of a temporary (and therefore more or less unsteady) support for the tripod of the transit, and the chances of variation in its position as the different sights are made are so great that it is difficult to say when a movement has not taken place that will vitiate the work.

In sighting up a shaft of greater depth than 100 ft., there is annoyance-if not danger-from dripping water or the fall of more solid substances. In a wet shaft the object glass is instantly covered with water, and a sight is impossible. Also, it is necessary to stand upon a platform, and it is hard to tell when this is perfectly rigid. From all these considerations the methods with a transit are never used by engineers in the anthracite regions, and the connections are made as follows:

When the bottom of the shaft can be reached by an adit or a slope in a roundabout route of such length as to render errors in measurement of distance of great importance, the angles are carried by a transit with as long sights as possible, and no distances are measured, from a point on the surface in the shaft to a point vertically below it in the mine. Sometimes the guide of the cage is taken when it has been recently set, as the guides are plumbed into position; but the better way is to suspend an iron plummet by a copper wire: sink the former in a barrel of water or bucket of oil so as to lessen the tendency to swing on account of the pull upon the bob and wires from the air-currents, or falling drops in a wet shaft. The top of the barrel should be