Or it will be still better to set the station-staves equally distant from the instrument (suppose about 16 or 20 perches each) at an angle of about 60° or so as to form nearly an equilateral triangle therewith, and level the 2 vanes (A and B fig. 130) as before, which will be then both in the same horizontal level, whether the instrument be right adjusted or not, because one will be as much above or below the true level of the instrument, as the other, being at the same distance from it; then remove the instrument as near as may be to one of them, suppose A, and raise or lower the vane A to the exact level of the visual ray in the instrument, noting precisely how much it is moved, and have the other vane B moved just as much in order to bring them again to a level, allowing for the correction of the apparent level if it be a sensible quantity, than adjust the instrument to the level of the vane at B.

To adjust the rafter level (fig. 131.) which may be 10, 12 or 14 feet in the span AB; set it on a plank or hard

ground nearly level, and mark where the plumb lines cuts the beam mn, suppose at c, then invert the position by setting the foot A in the place of B, and B in that of A marking where the line now cuts as at e; the middle point between c and e will be the true levelling mark.

To continue a level course with this instrument, set the foot A to the starting place, and move B upward or downward toward D or E, till the point B be deter, mined and marked for a level with A, then carry the instrument forward in the direction of C till the foot A rests at B, whence the point C is levelled as before, &c. Sights may be placed at r and s and the instrument adjusted to them, as before, by reversing them in the direction of some distant object.

After the instrument is duly adjusted, you may proceed to use it. Let the example be this annexed (fig. 132.) where A every where represents the level, and B the station-staves; and suppose the route be made from a to e; first plant the instrument between the staves a and b at A direct the level to a B, bring the bubble to the middle of the divisions, and instruct your assistant so to place the vane, that the hair in the telescope cuts the middle of the vane; then in a book divided into two columns, the one intitled Back Sights, the other Fore Sights, enter the feet, inches, and parts cut by the upper edge of the vane at a B, in the column intitled Back Sights.

Then look towards the other staff 6 B, bring the bubble to the middle of the divisions, and direct your assistant to place the vane so, that the hair cuts the middle of the vane; then enter the feet, inches, and parts cut by the upper edge of the vane in the column of Fore Sights.

Now, plant the instrument at A', still keeping the staff Bb exactly in the same place, and carry the staff aB forwards to the place cB; now look back to the (staff Bb, and enter the numbers cut by the vane there, under the title Back Sights; hen look forwards to cB. and enter the observation under the title Fore Sights. Do the like when the instrument is planted at A, A &c. always taking care to keep the staff in the same place when you look'd at it for a Fore Sight, till you have also taken with it a Back Sight.

Having finished your level, add up the column of Back Sights into one sum, and the column of Fore Sights also into one sum; and the difference between these sums is the ascent or descent required. And if the sum of the Fore Sights be greater than the sum of the Back Sights, then e is lower than a ; but if the sum of the Pore Sights, be less than the sum of the Back Sights, e is higher than a. For example, let the numbers be as in the following table;

1. And if the distances thus taken are short, the curvature of the earth may be rejected. For if the distance from the instrument be every where about 100 yards, all the curvatures in a mile's work will be less than half an inch.

2. If the distances from the instrument to the hindmost staff, be every where equal to the distance from the instrument to the corresponding staff; the curvature of the earth, and the minute errors of the instrument will both be destroyed. Hence it will be much best to set the instrument as equally distant from both staves as may be.

3. If the distances of the instrument from the staves, be very unequal and very long, the curvatures must be accounted for, and the distances, in order thereto, must be measurd.