46. In the survey of a long line of railway or canal, one of whose termini is in the vicinity of tide-water, the datum line is usually assumed at the level of mean high-water. In cases of surveys entirely inland, the datum line is taken at some convenient depth below the beginning point of the survey, and at such a distance that it shall be below the entire line on the surface. For such surveys, the system of notes described in the preceding section is insufficient.

47. As the survey progresses, fixed points of reference, called benches, are located in the vicinity of the line. Permanent objects are usually selected for benches; such as rocks, buildings, or trees, and at such distances from the line of the work as to be undisturbed by the subsequent construction.

48. Temporary benches, employed merely while changing the position of the levelling instrument, are called turning points. In either case, a well-defined point must be provided— one not easily disturbed by a blow, and, moreover, one upon which the rod can be held vertically.

NOTE.-The order of the surveys, on a line of road or canal, after the route has been determined by reconnoissance, is

1st. The Transit survey, establishing the centre line of the work.

2d. The section Level, or the measurement of the profile of the centre line.

3d. The cross-section work, or the measurements, with the level, across the centre line, to the full width of the road, for the purpose of determining the intersection of the slopes with the natural surface; and also, to afford data for the estimate of the amount of earth to be removed, or filled.

49. The following example will exhibit the method of re

to be thirty feet below the first bench. When the field-book is of the ordinary pocket size, the seven columns of notes will generally occupy two opposite pages; the first five being upon the left-hand page.

The bench having been selected and marked, its location

is described in the column of remarks.

The level is adjusted in some convenient place in the vicinity, and the reading of the rod is taken upon the bench. In the above example it is 1.637. As the bench is 30 feet above the

assumed datum line, the height of the instrument (or line of collimation) above this datum line is 31.637 feet.

The reading is recorded against Bench, in the column of + sights, and the "height of instrument" is recorded in its proper column, in the same line.

By referring to the above diagram it will be readily seen, that to obtain the height of the different points 0, 1, 1°°, &c., above the datum line, it is only necessary to take the readings of the rod, at these stations, and subtract them from 31.637. Such readings, therefore, are appropriately termed minus sights, and are recorded in the 4th column. As these readings are taken only to the nearest tenth of a foot, they are taken much more rapidly than the bench readings. The subtractions by which the surface heights are found, may be worked in the field or not, as the surveyor chooses. The unit of measurement, in the column of distances, is usually the engineer's chain of 100 feet. Readings are taken at intermediate points (as at 160 feet in the above example) when there are abrupt changes in the inclination of the surface.

50. When it becomes necessary to change the position of the level, such measures must be taken as will insure the exact "height of instrument," in the new position.

To effect this, a carefully-selected hard point is found (not necessarily on the exact line of the survey, but as far forward as convenience and accuracy will permit), and a reading of the rod is taken upon it, to thousandths.

If likely to be used for a single occasion only, it is called a “turning-point,” and marked T. P. in the distance column; otherwise it is called a Bench, and its location is described in the column of remarks.

A turning-point is taken between stations 3 and 4, in the

This is recorded in the sight column, and the height of the point is at once found (by subtracting this reading from Height of Instrument), and recorded in the column of "Heights."

The level is next carried forward to a new position, adjusted, and directed again upon the rod at the turning-point. The reading is taken to thousandths. This, when added to the height of the turning-point, evidently gives the height of instrument in its new position. It is recorded, therefore, as a sight. The survey is now continued by taking — sights at the various points along the line until it becomes again necessary to change the position of the level.

In the above example, the reading of the rod upon the turning-point, from the second position of the level, is 1.910. The height of the point upon which the rod stands is 20.224. The sum of these, or 22.134, is the "Height of Inst." for the second set of -sights. The successive subtractions of the readings from the Height of Instrument, give the surface heights as before.

The most extended section levels are but repetitions of this

process.

The rules for taking and recording field-notes in section levelling are as follows:

I. The "distances" recorded in the first column are the horizontal measurements, in chains, from the beginning of the survey to the points whose heights are to be determined. The heights are taken at each whole chain, and at such intermediate points as the irregularities of the surface require.

II. The first reading of the rod, after each setting of the level, is upon a bench or turning-point, and is a " + SIGHT:" all other readings are 66 SIGHTS."

III. The sight, added to the height of the point upon which the reading was taken, gives the "Height of Instrument."

!

IV. The

sights taken, at any position of the level, subtracted from the "Height of Instrument” for that position, give the corresponding "Surface Heights."

V. All the + SIGHT readings, and the last - sight of each set, being upon benches or turning-points, are taken to thousandths of a foot. The remaining" — sights" are taken to tenths only.

NOTE. It will be observed that when the column of "surface heights" is complete, the second, third, and fourth columns of the field-notes are no longer needed. The first and fifth columns, which together contain the horizontal and vertical measurements for the line of work, afford all the data necessary for mapping the profile and determining the grade-line.

The location of the benches should be so described in the column of "remarks," that any particular bench may be found at any time, by referring to the field-notes. The importance of this is apparent when it is remembered that the process of construction destroys or removes the stakes along the line of the survey, and that the question of the completion of the work can be determined only by reference to the benches. It is obvious, also, that they should be established somewhat off the line of the survey. The distance apart, of regularly established benches, should be governed by the above-mentioned uses of them.

Any turning-point may be profitably made a bench (when it can be made permanent), by carefully recording, so as to admit of its identification.

In conducting a section level through a rocky district, turning-points in abundance are found at hand, and cause no delay in their preparation, whereas a bench in the same section, requires marking and locating.

In levelling through flat and level sections of country,