time, or soon after the storm sets in, continues to fall
during its progress, and rises again, sooner or later, afterward. This is
the general rule.

On this subject Mr. Redfield's claim is this:

"EFFECT OF THE GALE'S ROTATION ON THE BAROMETER.--The extraordinary
fall of the mercury in the barometer, which takes place in gales or
tempests, has attracted attention since the earliest use of this
instrument by meteorologists. But I am not aware that the principal
cause of this depression had ever been pointed out, previously to my
first publication in this journal, in April, 1831, when I took the
occasion to notice this result as being obviously due to the
_centrifugal force_ of the revolving motion found in the body of the
storm.

"Since that period, inquiries have been continued by meteorologists
in regard to the periodical and other fluctuations of the barometer,
and the relations of these fluctuations to temperature and aqueous
vapor. But these incidental causes of variation, in the atmospheric
pressure, prove to be of minor influence, and we are left to the
sufficient and only satisfactory solution of this marked phenomenon
which is found in the centrifugal force of rotation."

The average pressure of the atmosphere, at the surface of the ocean, or in
the interior of the country, allowing for elevation, is about equal to the
weight of a column of quicksilver, thirty inches in height; hence the
barometer is said to stand at about thirty inches at the level of the sea.

This is sufficiently accurate for the northern hemisphere, north of the N.
E. trades; but the average is somewhat lower in the trades and in the
southern hemisphere. Thus, the average of sixteen months, during which the
Grinnell expedition was absent, was 30.08/100.

From a large number of logs examined by Lieutenant Maury, the mean
elevation in the N. E. trades of the Atlantic was 29.97/100; the S. E.
trades of the Atlantic, 29.93/100; off Cape Horn, 29.23/100; S. E. trades
of the Pacific, 30.05/100; N. E. trades of the Pacific, 29.96/100. The
height of the barometer off Cape Horn is not a fair index of the general
elevation of the southern hemisphere, inasmuch as it stands lower there
than at the coast of Patagonia and Chili, or at most, if not all, other
stations in that hemisphere.

As the barometer is constantly oscillating up and down (irrespective of
its diurn