given for the trade winds.

In the southern latitudes there is a comparatively constant condition
of wind and weather, because the surface of the globe in these regions
is mostly water; but in the north, where most of the land surface is
located, we have a very different and a very complicated set of
conditions, as compared with the southern zones.

The freaks of wind and weather that we find prevailing upon the North
American continent are not so easily accounted for as the phenomena
heretofore discussed. In the northern part the land reaches far up
toward the north pole, while on the west lies the Pacific Ocean, which
merges into the Arctic Ocean at Bering Strait. The climate of the
western coast is affected by a warm ocean current that sets up as far
north as Alaska, while high ranges of mountains prevent the effects of
this warm current from being felt inland to any great extent; all of
which helps to complicate any theory that may be advanced regarding
changes of weather. Aside from the changes of temperature that are due
to the seasons, which are caused by the oscillating motion of the earth
between the limits of the Tropic of Cancer on the north and the Tropic
of Capricorn on the south, there are other changes constantly taking
place in all seasons of the year. While it is not difficult to account
for the change of seasons and the gradual change of temperature that
would naturally follow--owing to the difference of angle at which the
sun's rays strike the earth--it is more difficult to account for the
violent changes that occur several times during the progress of a
season, as well as the less violent ones that come every few days. In
fact, it rarely happens that the temperature is exactly the same on any
two successive days during the year. The diurnal changes are easily
accounted for by the rotation of the earth on its axis each day. But
there is another class of phenomena with which the "weather man" has to
struggle when he is making up a forecast of the weather from day to day.

In order that we may proceed intelligently, let us say a word about the
barometer. We speak of high and low barometer, and we make the
instrument with graduations marked for all kinds of weather, which
really mean but very little. The reading of a single barometer alone
will give us but a faint idea of what is really going to happen from day
to day. But if we have a series of barometers located at different
stations scattered all ov