angular points as possible, by which the labour of removing the instrument backward and forward is considerably abridged. In choosing stations, avoid cattle or car-tracks, or other places where your station marks would be likely to be removed or obliterated,—a thing that frequently happens, to the great annoyance of the surveyor. The roads in this survey were sometimes determined from the lines employed to fill in the work, as well as the lines connecting the principal stations. Some of them were also surveyed by the method of traversinga term applied to all irregular surveying by the chain and theodolite. On starting from one of the gates leading into the Park, the instrument being set to zero, the telescope was directed to one of the most conspicuous stations, and after having turned it in the direction of the road to be surveyed, (the lower plate being clamped,) the angle was then read off, and the distance to the forward station chained. The instrument was then removed to the forward station, the upper plate being clamped, and a staff left at the starting point, the telescope was directed to the back station, and setting the upper plate free, the telescope was directed to the next forward station along the road, and the distance to the next forward station and reading on the theodolite registered in the field-book. Here it is necessary to observe, that the degrees shewn on the graduated limb of the theodolite are not the measure of the angle contained by these two lines, but rather the angle that this second line forms with that upon which the instrument was first set. P As the work progressed, other lines were selected, from which the forward bearings were measured. In pursuing this plan, greater accuracy is secured in plotting; as when all the angles are measured and plotted from one line, a trifling error committed in the direction of one line does not affect the next. When a great number of lines have reference to another given in position, the readiest and most correct way of plotting them is to use a circular protractor, made of pasteboard, with the centre cut out. Put the centre of the protractor at the angular point, and zero on the line whose position is given; then a ruler, having always its edge at the centre of the protractor, being applied to the several angles taken in the field, will point out the bearings of the respective portions of the road. Then by the parallel ruler, lines may be drawn parallel to these, on which the respective distances are to be laid off, from the scale of equal parts. This plan may be employed in plotting an estate which has been surveyed by measuring the sides and angles of its circumscribing polygon. But when great accuracy is required, a brass protractor with a vernier is better. When the road was considerably curved, offsets were measured to the curve from the several station lines, which were plotted in the usual way, and the extremities of the offsets joined by a regular curved line. When the road was straight, and of uniform width, offsets were taken only to one side; but where it varied in breadth, offsets were taken to both sides. A canal may be surveyed in the same way as a road; but when it is of unequal breadth, offsets must be taken to both sides, from station lines measured on each side of the canal. The triangles employed in the survey were all laid down from the lengths of their sides, not from the measure of their angles. Unfit as the sextant is to measure the angular distances between objects in planes inclined to the horizon, the use of the circumferenter is still more objectionable. The compass cannot take angles nearer than 15', independent of other errors arising from its centering and peculiar construction. Surveying by this magnetic instrument is founded on the supposition that the needle invariably preserves a parallel position, which is far from being the case, there being a variation by which it differs at different times of the day and year. The maximum of these variations takes place between noon and three o'clock in the afternoon, and the greatest diurnal variation generally takes place in the months of April, May, June, and July. In the year 1759, the diurnal variation of the needle was very carefully observed at London, and the greatest was found to be 13′ 21′′ in the month of June, and the least 6'.58 in the month of December. The results of recent discoveries disclose a very curious fact, namely, that all perpendicular objects, as houses, trees, &c. &c. have (at least in north latitude) a north pole at bottom, and a south pole at top, and that such objects exert a very sensible influence upon the needle; it may therefore be advisable, in delicate surveys, to keep at some distance from such objects, especially from houses where much iron is deposited |