Boiling Point Thermometer Tables.-Table III., compiled by Mr Francis Galton, F.R.S., will facilitate the computation of heights by boiling point thermometer.*

For example from Table III.

At station A, boiling point = 195°.1, tabular No. = 9,040

Approximate difference of height = 8,153 ft.

To correct for temperature of air

At station A, temperature

= 65° F.

= 73° F.

Mean temperature (65° + 73°)=69° F.

From Table III., Part II., the multiplier for 69° is 1.082, and corrected difference of height = 8153 x 1.082 = 8,821 ft.

Instruments: Aneroid Barometer. The most useful barometer for surveying purposes is the aneroid. A 5 in. aneroid, with a range of about 4,000 or 5,000 ft., is about the best size. Fig. 143 shows a 5 in. aneroid by Stanley. When reading the instrument, hold it horizontally, and tap it several times. As the aneroid is mechanically compensated for temperature, the correction for temperature of the instrument itself (correction for temperature of mercury) is not required. The correction for temperature of the air must, however, always be made, as also the

Fig. 143.

Aneroid Barometer.

other corrections when any great degree of accuracy is aimed at. As a rule, when several trips are made between two stations, it will be found that the mean of the "uphill" journeys is more correct than the mean of the "downhill" journeys. The reason of this is that the pressure of the atmosphere puts a strain on the spring inside the vacuum box of the aneroid, and in going "uphill" the pressure of the atmosphere decreases, and this strain

Given by permission of the Royal Geographical Society, from "Hints to Travellers."

is relieved. In going "downhill" the reverse is the case, and the instrument is apt to be slower in responding to the change of atmospheric pressure. The instrument should always be held in the same position when taking the readings, i.e., either always horizontally or always vertically, as a mere change of position usually affects the readings.

An aneroid should be compared with a standard mercurial barometer, and have its index error ascertained. Most aneroids will not agree with a standard mercurial barometer at low pressures, even if correctly graduated in the first instance. The index error is fairly constant in good instruments, but will always be considerably increased when low pressures have been the rule for any length of time.

Aneroids should be treated with almost as much care as a chronometer, and should not be dangled about or shaken in the pockets. As the index error is apt to change, for absolute heights the results should be compared with the portable boiling point thermometer (Fig. 147).

Above 8,000 ft. elevation the aneroid is not reliable.

Mountain Barometers.-These may be either cistern or syphon barometers. The best cistern barometer is Fortin's, as improved by Guyot. This is shown in Fig. 144, and consists of a column of mercury in a glass tube, its lower end being in a cistern of mercury. The tube is covered by a brass case terminated at the top in a ring c for suspending the instrument. At the bottom is a flange G to which the cistern is attached. At D is a vernier from which the height of the mercury is read. The zero point is a small ivory point at H, the mercury in the cistern being raised or lowered by the screw K until its surface is in contact with this ivory point. At F is the attached thermometer which gives the temperature of the mercury.

The syphon barometer is shown in Fig. 145. In this form the cistern is dispensed with, and the tube is turned up at its lower end as shown. A small hole at s admits the air. The height of the mercurial column is equal to the difference of the heights of the mercury in the two branches of the tube. The whole is protected by a brass case, and is fitted with verniers, thermometers, &c., as in Fortin's barometer. The best syphon barometer is GayLussac's, as improved by Bunten.

Fig. 146.--Captain George's Mercurial Cistern Barometer.

Fig. 146 shows Captain George's mercurial barometer (cistern barometer), as made by Cary, London.

Boiling Point Thermometer.— This instrument is shown in Fig. 147, and consists of a thermometer AA graduated from 180° to 215°, a spirit lamp в which fits into the bottom of a brass tube c that supports the boiler D, and a telescopic. tube EEE, which fits tightly on to the top of the boiler. The thermometer is passed down the tube E from the top until within a short distance of the water, which it should never touch, and is supported in that position by an indiarubber washer F. The steam passes from the boiler up the tube E and escapes by the hole G. The whole of the apparatus fits into a cylindrical tin case 6 in. long and 2 in. diameter, with the exception of the thermometer, which is packed in a brass tube lined with indiarubber, and having a pad of cotton wool at each end.

In using the boiling point thermometer, wait until the mercury becomes stationary before reading, and at the same time take the temperature of the air in the shade with an ordinary thermometer. As the lamp is very often too small, in purchasing see that the lamp will hold a proper supply of spirit. A screen of some sort is usually required on the windward side. of the thermometer, and at low temperatures is indispensable. The heat of the lamp will otherwise be carried off too rapidly for the water to boil properly.

Levelling by Angular Measurements or Trigonometrical Levelling. The simplest case for short distances in which curvature and refraction are neglected is as

shown in Fig. 148. In this case, if the distance PC and the angle BPC are measured, we have

BC= PC tan BPC

C

(1)

As the best conditioned triangle has the angles at P and в each equal to 45°, the distance PC should be selected as nearly as possible equal to the height BC. In most cases it will happen that the distance PC cannot conveniently be directly measured as in Fig. 149. In this case measure the distance PD to any convenient point D in the same vertical plane as P and B. Also measure the angles of elevation CPB, CPD, and the angle PDB at D.

Fig. 148. Trigonometrical Levelling.

Then in the triangle PBD, having given the side PD and the angles DPB, PDB, calculate the side PB, and we get

PBD is of course equal to 180° - (BPD + BDP). Then from PB and the angle BPC calculate BC, and we get—

If the height of в above D is required, we have

DE=PD sin CPD.

Then DE FC and BF BC

FC. Otherwise measure the angle of elevation FDB at D, and in the triangle PBD calculate BD, then we have

BF BD sin BDF.

When a distance PD in the same vertical plane as P and B cannot be measured, run out a line PD in any convenient direction as shown in Fig. 150, and measure PD. Measure also the angle of elevation BPC and the horizontal angles CPD, cdp. Then in the triangle CPD we have