brought
about mainly by variations in temperature, that we owe all its
currents, and it is upon these winds that the features we term climate
in largest measure depend. Every movement of the winds is not only
brought about by changes in the relative weight of the air at certain
points, but the winds themselves, owing to the momentum which the air
attains by them, serve to bring about alterations in the quantity of
air over different parts of the earth, which are marked most
distinctly by barometric variations. These changes are exceedingly
complicated; a full account of them would demand the space of this
volume. A few of the facts, however, should be presented here. In the
first place, we note that each day there is normally a range in the
pressure which causes the barometer to be at the lowest at about four
o'clock in the morning and four o'clock in the afternoon, and highest
at about ten o'clock in those divisions of the day. This change is
supposed to be due to the fact that the motes of dust in the
atmosphere in the night, becoming cooled, condense the water vapour
upon their surfaces, thus diminishing the volume of the air. When the
sun rises the water evaporated by the heat returns from these little
storehouses into the body of the atmosphere. Again in the evening the
condensation sets in; at the same time the air tends to drift in from
the region to the westward, where the sun is still high, toward the
field where the barometer has been thus lowered; the current gradually
attains a certain volume, and so brings about the rise of the
barometer about ten o'clock at night.

In the winter time, particularly on the well-detached continent of
North America, we find a prevailing high barometer in the interior of
the country and a corresponding low state of pressure on the Atlantic
Ocean. In the summer season these conditions are on the whole
reversed.

Under the tropics, in the doldrum belt, there is a zone of low
barometer connected to the ascending currents which take place along
that line. This is a continuous manifestation of the same action which
gives a large area of a disklike form in the centre or eye of the
hurricane and in the middle portion of the tornado's whirl. In
general, it may be said that the weight of the air is greatest in the
regions from which it is blowing toward the points of upward escape,
and least in and about those places where the superincumbent air is
rising through a temporary or perm