torms is about 13 miles an hour; in none of
those observed has it exceeded 14 miles nor fallen below 11 miles. The
diameter of the exterior revolving circle of the storm varies from 40
to 130 miles, and the diameter of the inner circle or calm region, may
be estimated at from 8 to 15 miles. The duration of the true typhoon
at any one place is never longer than ten hours and generally much
less. These storms are always accompanied by abundant rain, with low,
dense clouds, which at times limit the horizon to a few yards distance,
and are generally accompanied by electrical discharges. The barometer
falls slowly for some days before the typhoon, then falls rapidly
on its near approach, and reaches its lowest when the vortex is but
a little way off. It then rises rapidly as the vortex passes away,
and then slowly when it has gained some distance. Near the vortex
there are usually marked oscillations. The typhoon generally begins
with a northerly wind, light drizzling rain, weather squally and
threatening, a falling barometer and the wind veering to the eastward,
when the observer is to the northward of the path of the storm, and
backing to the westward when he is to the southward of it; the wind
and rain increase as the wind shifts, and the storm generally ends
with a southerly wind after abating gradually.

The following warnings of the approach of a typhoon, and directions
for avoiding the most dangerous part of it, are taken from the China
Sea Directory: The earlier signs of a typhoon are clouds of a cirrus
type, looking like fine hair, feathers or small white tufts of wool,
traveling from east or north, a slight rise in the barometer, clear
and dry weather, and light winds. These signs are followed by the
usual ugly and threatening appearance of the weather which forbodes
most storms, and the increasing number and severity of the gusts with
the rising of the wind. In some cases one of the earliest signs is
a long heavy swell and confused sea, which comes from the direction
in which the storm is approaching and travels more rapidly than the
storm's center. The best and surest of all warnings, however, will
be found in the barometer. In every case there is great barometric
disturbance. Accordingly, if the barometer falls rapidly, or even if
the regularity of its diurnal variation be interrupted, danger may
be apprehended. No positive rule can be given as to the amount of
depression to be expected, but at the center of som