287.—The Reading of the Staff.—The first position, which is afterwards termed the back reading, is taken at a distance behind the first forward position of the level. This is recorded exactly as it appears in the telescope, the height of the telescope being also noted in the levelling book, to be described. Thus in Fig. 112, S the first staff; L the first station for taking levels. The fore reading L to S′ reads to a higher part of the staff S′; L′ next level station back sight. L′S′ reads high on the staff S′; fore sight L′S″ reads low; back sight L″S″ again low, following the contour; fore sight L″S‴ low; thus giving data in the levelling book from which the contour can be plotted from the datum line, which is taken low to make all readings plus.
Fig. 112.—Practice of levelling.
Larger image
288.—The staff reading, as already described, is divided into feet, with two places of decimals. The safest method of taking this reading is to take the second decimal place first and then record it, then the first decimal, and finally the foot. In this manner no effort of memory is required, and the staff being sighted three times assures the certainty of the reading. The telescope should not be touched during the operation, so that the reading in this manner is only a cautious transfer.
289.—If two staves are used on fairly level ground, the second staff is now advanced 5 chains from the level to a measured station, the staff holder here sighting the line through the level to the back staff, and firmly treading down the staff plate if the land is soft or grass, or otherwise requires it, or an iron triangle is used. When time is given to hold the staff vertically by means of the staff level, the reading is taken in this position by the leveller as before, and this is recorded in the levelling book. The level is now moved forward 10 chains, that is, 5 chains ahead of the forward staff. The staff is carefully turned half round without pressure upon its standing place or plate to face the level as now placed, in which position it is then read off by the level as the back sight, the back staff now being moved 5 chains forward of the level, and so on alternately staff and level until the distance required to be levelled is completed, if there is no obstruction which causes another method of procedure to be adopted. A similar plan is pursued with a single staff; but care has to be taken in securing the right line of march, which will be by placing the staff in a sight line through the level with a fixed landmark instead of the back staff mentioned.
290.—The equal back and fore sights as far as practicable are insisted upon by all levellers, as by this means any inaccuracy in the level, if the run of the bubble is kept constantly true, is thereby compensated; but it is not always convenient, and when it is not the accuracy of the work must depend largely upon the qualities of the level. It is not necessary or convenient at all times to take the back and fore sight in a line—obstructions of woods, rivers, etc., may occur. In these cases very often what is quite equivalent may be done by taking equal angular back and fore sights from the apex of an equilateral triangle thus:—Say an obstruction occurs for the chain by a pond or wood, but that both points to which the levels are to be taken are visible at some lateral position. Levels may be taken from this place, and if the intermediate point of distance is equal from both stations there will be no instrumental error. Thus, suppose the direct level line east (90°), and that the two stations can be seen and the staves read at 150° and 210°; here, evidently, this is equivalent to a direct back and fore sight, the right angle to the level course being 180°—the one station is 150° = 180° - 30°, and the other 210° = 180° + 30°. If these equal angles can be even approximated with a fairly good level the error will be small. In this manner intermediate and extended points may often be conveniently taken by previous arrangement with a good staff holder. It is in this angular levelling that the greatest use of the compass is found to give the angles, to make entries of the work in the levelling book.
291.—In levelling hilly ground great loss of time would sometimes be incurred from taking equal back and fore sights; the best plan in this case is to make as much use as possible of the length of the staff in use. It is in hilly districts only that a staff longer than 14 feet is advantageous. With any staff in descending a hill only 5 feet of the staff can be used for the back sight, that is, a part of it equal to the height of the level, and sometimes 4 feet or less if there is grass, brambles, or other obstruction. Whereas for the foresight all the staff upwards of the height of the level, that is, about 9 feet in a 14-feet staff, can be used with certainty. The distance of setting up of the levels and staves must in this case entirely depend upon the length of the staff and other conditions present.
292.—For near reading of the staff on sharp inclines, reading to two places of decimals is not near enough, as errors may accumulate rapidly. It is in such cases that a fully divided staff is best. The divisions upon a near staff appear in the telescope much magnified; and three places of decimals may easily be taken by anyone used to reading a chain scale, particularly if a point diaphragm be used. Through valleys the level may be often checked at some point from hill to hill by a back sight: the contour must nevertheless be followed for the section. It is in these shorter unequal ranges and in distant sights that accuracy in the level is demanded; and it becomes interesting to know how nearly this may be depended upon for such readings.
293.—As already mentioned, a sensitive 14-inch level of Y construction, or a dumpy in perfect adjustment supported on the tribrach system, will work with a level tube divided to read 5 seconds in divisions 1/20 inch apart. There will be a little personal error in reading the bubble from difference of reflection, according to the direction of the light from the two ends of the bubble, as before discussed; but the bubble may be assumed to be read within less than half a division, that is, within 2½ seconds—say 2 seconds. A distinct staff may be read with a good glass within ·1 foot at one mile. A second of arc subtends ·025598 of a foot = approximately ·3 inch at a mile distance. Therefore a back reading at this distance can be taken within an inch or so of allowance for instrumental errors. A reading taken in this way at a mile distance would require a plus allowance for curvature of the earth of 8 inches, minus say 1 inch for refraction = 7 inches. From these data we can get a fair check level for hilly ground, possibly more accurate than by contour levelling for a distant station, even if we allow double the probable error, say ·1 foot for error of reading the staff at a mile distance.