will be sufficient to show the method of keeping the book, and the system to be adopted throughout in taking the dimensions.

When there are a sufficient number of lines measured, proceed then to take the angles. It will be found in course of the survey that angles will not be required to every line or street, for when one or two of the principal lines are plotted, many others will fall into the work as proof lines; for example, Fig. 1.

Much information will be gained by a careful study of the plan. The dotted lines represent the chain lines; the north part called the block plan, marked A, having only the fronts of the buildings or streets; the south part, marked B, represents the plan when finished with all its details.

On the line a b, there are ten stations determined on in the first line. Commencing again at c, take the angle a cd, on which there are two stations, gh; commence again at e, take the angle a ef. Now, it is evident, if the two former angles and the lengths are measured correctly, that the line gf will be a proof line, and also the line ihk; the station at k will be a fixed point.

We now commence at ƒ and proceed to l, at which point it will be necessary to take the angle mlf, and that angle will prove the length of c d and n k.

Now measure by the chain the line 7m, passing through dn, that being the end of the second line; therefore the distance I to d must prove itself, as that point was before determined by the angle taken at c.

At n, on the line 7 m, leave a station, then will I m be a fixed line for a combination of the survey on the north part.

Now measure the line from n to k, on which leave stations at o and p; the lines n k and o g will be both proof lines.

So that in the great portion of this part of the survey three angles only are required, and every measured line afterwards proves itself.

The same principle is adopted on the south side, taking the longest and straightest strects to form a basis, and those lines

that are nearest to a right angle, as q to r, upon which numerous stations will be fixed.

Observe always to keep the work together as much as possible, so that it may be closed in, as shown by the lines e to s and s to t.

When the whole of the survey is completed, and the block plan plotted, proceed then with the details.

Prepare a book about the size of letter paper; copy off each block separately on an enlarged scale. With the tape, first measure the fronts of each house, passage, &c., separately, as thus (see Field-book): 24.3, 5.6, 20.0, and 29.3; add all these together, which should be equal to the measure of the whole length from 6 to c.

Then proceed with the depths of the houses and the back premises, taking diagonals from one angle to the other. A sketch of the premises having been previously made.

In many instances there are passages at the back, as at u v, which afford considerable assistance in measuring the details, particularly when straight, as at u; when they are crooked as at v, the prismatic compass will be found useful, as the angles must be taken by the needle.

In all cases take the supplement as well as the angle. (See Field-book.)

RAILROAD SURVEYING.

Problem 33.

Plate 26. To survey for a railroad.

When a survey has to be made for a railroad, it takes a long irregular tract of country from one important town to another, generally from 10 to 20 chains in width, in some cases more.

The course the intended line is proposed to be made is marked out by the engineer on the Ordnance or county maps.

The first and most important part in surveying for a line of railway is to determine and accurately pole out the base lines,

which should be as near as possible to the line described on the Ordnance map.

In chaining a base line the greatest attention and accuracy is required, as the only check to its length is when the levels are taken. Many bills in Parliament have been lost entirely through the inaccuracy of the plans and sections, involving not only the loss of a year, but the enormous amount of money required for the Parliamentary preparations.

Unfortunately these surveys are often got up in haste, and frequently under great opposition, all contributing to the many difficulties the surveyor has to contend against.

When a long line of country has to be surveyed a portion is allotted to each surveyor, as shown by the plans; each portion has to be connected at each end, requiring the greatest accuracy in fixing the base lines.

The most perfect method of connecting these lines, is by continuing them into each other's work, and fixing them by intersecting triangles.

Base lines that are connected by an angle taken with the theodolite are very ambiguous, from many causes; for instance, where the country is hilly there may be only a short distance from the instrument to the flag, and the line probably of considerable length, depending wholly on the angle taken by a sixinch radius; a very trifling error in reading the angle would throw that line greatly out of its real position. By the former method, if the lines are accurately poled out, the intersecting lines not only prove the correct position of each base, but they also serve to take up the details, and a great saving of time. Sometimes a base has been hung on by a fence only; such practice is highly censurable.

In a long line of railway it is impossible always to keep in the valley; a portion of the line will be rough and hilly; in such cases the theodolite is indispensable.

At the end of each base line the surveyor for that portion

should leave a distinguishing mark by cutting a triangular hole, and one or two other marks on his line, to enable the next surveyor to continue that line into his work, and adopting such tie lines to the two bases as in his judgment will be most desirable.

Again, supposing the surveyor of the middle portion should begin his base line before the first base line is finished; in that case leave a distinguishing mark at the beginning of his base, and other marks, as before, to point out the direction, as well as marks on the last fence line; then the surveyor of the first portion will pole back the second base and tie lines into his work, or such other lines as the case may require.

If every surveyor adopted this simple method, base lines could never be plotted wrong.

Problem 34.

Plate 26, Fig. 1, represents the portion of a line, the ending and beginning of two other portions.

Commence by poling out the base line from B top; at B is a distinguishing mark denoting the end of the first base, and at E another mark showing the direction of the line; from B continue that line to B1, where it intersects the second base line; at D B1 measure back from B' to E, and the tie line ED.

The second base may be said to commence at D, and ends at p; in chaining the line set up all the requisite stations or flags for the details; the line DF takes up the fence to the left, leaving flags for the cross fence from F to H, intersecting the base line at G, forming a large triangle and proof to the tie line ED; all the other fence lines within that triangle are proof lines as A C.

We now pass on from I to K, taking up the fence; and from K to O, intersecting the base line at S; the intermediate fence line proves the triangle I KS; from O proceed to L on the base line, which is a proof to I K besides the line TPQ;

finish the line OL to M; from M a short tie line, M Q, determines the position of the third base, which is proved by the line N R p.

Where fence lines are not determinately fixed, as c b, and without a box-sextant, take a short tie line with the chain, as at a b.

By adopting this system, the base lines cannot fail being correct, and every fence line becomes a proof, excepting in a very few instances.

Problem 35.

Fig. 2, Plate 26. To survey the same ground with the theodolite.

In this example the same base and fence lines are adopted as in the last example.

It is of no importance whether the angles are taken before or after the chaining.

When base lines are obtuse, it is therefore better to pole the line back and fix a flag, as at a, and measure the acute angle A B a.

Beside the angles taken at B and C, there would require at least six other angles for the fence lines, as at cdefg, after which all other fence lines will prove themselves.

Now by comparing the two plans it will be seen that there is but very little more chaining in the first example than in the second.

Considering the time in taking angles by the theodolite, the care required to preserve the instrument from accident, it will be found that the first example is preferable to the second in the saving of time; the angles taken by the chain are more to be relied on, supposing the country tolerably fair for chaining.

Where a ravine occurs, adopt any of the examples shown, Plate 14.