ottom to the base board of the
instrument by a screw (Fig. 156). From the top rises a pin, P, with a
transverse hole through it to accommodate the pin K E, which has a
triangular section, and stands on one edge.

[Illustration: FIG. 155.--An aneroid barometer.]

Returning to Fig. 155, we see that P projects through S, a powerful
spring of sheet-steel. To this is attached a long arm, C, the free end
of which moves a link rotating, through the pin E, a spindle mounted in
a frame, D. The spindle moves arm F. This pulls on a very minute chain
wound round the pointer spindle B, in opposition to a hairspring, H S. B
is mounted on arm H, which is quite independent of the rest of the
aneroid.

[Illustration: FIG. 156. FIG. 157. The vacuum chamber of an aneroid
barometer extended and compressed.]

The vacuum chamber is exhausted during manufacture and sealed. It would
naturally assume the shape of Fig. 157, but the spring S, acting against
the atmospheric pressure, pulls it out. As the pressure varies, so does
the spring rise or sink; and the slightest movement is transmitted
through the multiplying arms C, E, F, to the pointer.

A good aneroid is so delicate that it will register the difference in
pressure caused by raising it from the floor to the table, where it has
a couple of feet less of air-column resting upon it. An aneroid is
therefore a valuable help to mountaineers for determining their altitude
above sea-level.

BAROMETERS AND WEATHER.

We may now return to the consideration of forecasting the weather by
movements of the barometer. The first thing to keep in mind is, that the
instrument is essentially a _weight_ recorder. How is weather connected
with atmospheric weight?

In England the warm south-west wind generally brings wet weather, the
north and east winds fine weather; the reason for this being that the
first reaches us after passing over the Atlantic and picking up a
quantity of moisture, while the second and third have come overland and
deposited their moisture before reaching us.

A sinking of the barometer heralds the approach of heated air--that is,
moist air--which on meeting colder air sheds its moisture. So when the
mercury falls we expect rain. On the other hand, when the "glass" rises,
we know that colder air is coming, and as colder air comes from a dry
quarter we anticipate fine weather. It does not follow that the same
conditions are found in all parts of the world. In regions which ha