The foregoing remarks relate especially to the central and western
portions of the North Atlantic; they will however equally apply to the
remaining localities of storms. Under any circumstances, and in any
locality, a _high_ barometer not less than a low one should demand
particular attention, and if possible, _hourly_ readings taken some time
before and after the passage of the maximum: this will be referred to
more particularly under the next head.

_Preceding and Succeeding Accumulations of Pressure._--Mr. Redfield has
shown in his Memoir of the Cuba Hurricane of October, 1844, that two
associated storms were immediately preceded by a barometric wave, or
accumulation of pressure, the barometer rising above the usual or annual
mean. We have just referred to the importance of _hourly_ observations
on occasions of the readings being _high_ as capable of illustrating the
marginal phænomena of storms, and in connexion with these accumulations
of pressure in advance of storms we would reiterate the suggestion.
These strips of accumulated pressure are doubtless crests of atmospheric
waves rolling forwards. In some cases a ship in its progress may cut
them transversely in a direction at right angles to their _length_, in
others very obliquely; but in all cases, whatever section may be given
by the curve representing the observations, too much attention cannot be
bestowed on the barometer, the wet and dry bulb thermometer, the
direction and force of the wind, the state of the sky, and the
appearance of the ocean during the ship's passage _through_ such an
accumulation of pressure. When the barometer attains its mean altitude,
and is rapidly rising above it in any locality, then _hourly_
observations of the instruments and phænomena above noticed should be
commenced and continued until after the mercury had attained its highest
point and had sunk again to its mean state. In such observations