The ordinary mercurial barometer in use to-day is, essentially, nothing more than the glass tube and vessel of Torricelli’s famous experiment. A simple form of the mercurial barometer is shown in Fig. 5. It consists of a glass tube about one-quarter of an inch in inside diameter and about 36 inches long. This tube, closed at the top and open at the bottom, is filled with mercury, the lower, open end dipping into a cup of mercury known as the cistern. The space above the mercury is a vacuum. The mercury extends inside the tube to a height corresponding to the weight or pressure of the air, the vertical height of the top of the mercury column above the level of the mercury in the cistern, in inches and hundredths of an inch, being the barometer reading. At sea level the normal barometer reading is about 30 inches. There is an opening near the top of the cistern, at the back of the instrument, through which the air gains access to the mercury and holds up the mercury column. It will readily be seen that, as the mercury in the tube rises, the level of the mercury in the cistern falls, and vice versa, so that there is a varying relation between the two levels. In order to have the reading accurate, it is necessary that the surface of the mercury in the cistern should be just at the zero of the barometer scale when a reading is made. To accomplish this, the bottom of the cistern consists of a buckskin bag which may be raised or lowered by means of a thumb-screw, seen at the lower end of the instrument. The level of the mercury may thus be changed and adjusted to the top of a black line, marked on the outside of the cistern, and which indicates the zero of the scale. Before making a reading, the surface of the mercury in the cistern must be raised or lowered until it just reaches this black line. Then the top

of the mercury column will give the pressure of the air. The reading is made on an aluminum scale at the top of the wooden back on which the tube is mounted, this scale being graduated both on the English and on the metric system. This barometer may be hung against the wall of a room.

The aneroid barometer (Greek: without fluid), although less desirable in many ways than the mercurial, is nevertheless a useful instrument for rough observations. The aneroid is not good for careful scientific work, because its readings are apt to be rather inaccurate. To be of much value in indicating exact pressures, it should frequently be compared with and adjusted to a mercurial barometer. An ordinary aneroid barometer is shown in Fig. 6.

Fig. 6.

In this instrument the changes in atmospheric pressure are measured by their effects in altering the shape of a small metallic box, known as the vacuum chamber. The upper and lower surfaces of this box are made of thin circular sheets of corrugated German silver, soldered together around their outer edges, thus forming a short cylinder. From this the air is exhausted, and it is then hermetically sealed. A strong steel spring, inside or outside of the vacuum chamber, holds apart the corrugated surfaces, which tend to collapse, owing to the pressure of the external air upon them. An increase or decrease in the air pressure is accompanied by an approach, or a drawing apart, of the surfaces of the chamber. These slight movements are magnified by means of levers, a chain, and a spindle, and are made to turn an index hand or pointer on the face of the instrument. The outer margin of the face, underneath the

glass, is graduated into inches and hundredths, and the pressure may thus be read at once.

As the tension of the steel spring varies with the temperature, aneroids are usually compensated for temperature by having one of the levers made of two different metals, e.g., brass and iron, soldered together, or else by leaving a small quantity of air in the vacuum chamber. This air, when heated, expands, and thus tends to compensate for the weaker action of the spring, due to the higher temperature. At best, however, this compensation is but imperfect, and this fact, together with the friction of the different parts, the changes in the spring with age, and the need of frequent adjustments, makes aneroids rather inaccurate. They may be adjusted to mercurial barometers by means of a small screw, whose head may be found on the lower surface of the instrument. The words fair, stormy, etc., which frequently appear on the face of aneroid barometers, are of little use in foretelling weather changes, as no definite pressures always occur with the same weather conditions. The instrument should be tapped lightly a few times with the finger before a reading is made. The second pointer, which is often found in aneroids, is set by the observer on the position marked by the index hand when he makes his reading. The difference between the pressure marked by this set pointer and that shown by the index hand at the next observation is the measure of the change of pressure in the interval.

Another column must now be added to the record book (preferably between the columns devoted to temperature and wind) to receive the “Pressure in Inches and Hundredths.”

Is the pressure constant (i.e., are the readings always the same) or does it vary? If it varies, is there any apparent system in the variations? Is there a tendency to a daily maximum? To a daily minimum? If so, about what time do these occur, respectively? What is the average variation (in inches