42 
mospheric pressure ; and careful watching will 
pretty surely show a rising barometer, as the 
annulus of high pressure that surrounds the 
storm (see fig. 8) moves over the observer. 
The weather may still be clear, and the wind 
moderate and from its normal quarter ; but this 
change in the glass demands renewed watch- 
fulness. Let.us suppose that such an observa- 
tion be made on board a vessel lying east of 
the Lesser Antilles. The chart shows the cap- 
tain that he is in the stormy belt. He may be 
directly in the path of the advancing storm, 
where he will feel its full violence; and he 
must make the best of his way out of it. Fol- 
lowing the rising pressure, three other signs 
of increasing danger may be observed, — first, 
faint streamers of high cirrus-clouds may be 
seen, slowly advancing from the south-east to 
the north-west, or from the east to the west, 
in the high overflow from the storm’s centre ; 
this unpropitious change may accompany the 
rising of the barometer, or may be first seen 
when the barometer is highest : second, the ba- 
rometer begins to fall, slowly at first, but more 
and more quickly when it reaches and passes 
twenty-nine inches; the vessel is then within 
the limits of the storm: third, the wind has 
shifted so as to blow from a distinctly northern 
quarter, and its strength goes on increasing ; 
this is the indraught, blowing spirally toward 
the centre. There is then no longer any ques- 
tion that a storm is approaching; and as soon 
as a heavy bank of clouds makes itself seen, 
moving southward across the eastern horizon, 
then the central part of the storm is in sight. 
These clouds are the condensed vapor in the 
rising central spirals, and rain is falling from 
them. In deciding on a course to be pursued, 
the first point to be determined is, where is the 
storm’s centre? That being known, its prob- 
able path can be laid down with considerable 
certainty in this part of the ocean; and then, 
perhaps, the greatest danger may be avoided. 
But here a very practical difficulty arises. To 
find the direction of the storm-centre, we must 
know the incurving angle of the wind’s spiral, 
—the angle of inward inclination that it makes 
with a circle whose centre is at the storm’s cen- 
tre. ‘The earlier students of the question — 
Dove, Redfield, Reid, and Piddington— cone 
sidered the course of wind to be concentric cir- 
cles, or inward spirals of very gradual pitch; 
so that they said the inclination of the wind is 
practically zero, and a line at right angles to its 
course must be a radius leading to the centre. 
Later studies showed this to be incorrect. The 
inclination of the wind inward from the circle’s 
tangent was found to vary from 20° to 40° or 
SCIENCE. 
50°: but it was. thought that this inclination 
was symmetrical on all sides; so that, with an 
average inclination of 30°, the storm’s centre 
must always bear 60° to the left of the wind’s 
course. Finally, the most recent results seem 
to show that the wind’s course is neither circu- 
lar nor symmetrically spiral; that the wind’s 
inclination is very distinctly different in differ- 
ent latitudes, on different sides of the storm, 
in the different conditions found on sea and 
land, at different distances from the centre and 
at different altitudes. In so complicated a 
case, much judgment will be required to find 
where the storm-centre lies. 
First, in regard to the latitude of a storm. 
Without considering its progression, the nearer 
it is to the equator, the less its indraught winds 
will be deflected to the right by the earth’s ro- 
tation, —the more nearly radial they will be. 
But, as they move with much energy, they will 
gain in rotary motion rapidly as they ap- 
proach the centre, and there will whirl around 
in almost perfect circles. Storms in low lati- 
tudes will therefore tend to have a compara- 
tively small but violent central whirl, only one 
or two hundred miles in diameter, within which 
the winds may be almost circular; and the 
centre will there be nearly at right angles to 
the wind’s course. Farther from the centre, the 
winds would be nearly radial; and, if storms 
could arise on the equator, they would have 
simply radial indraughts with a very small cen- 
tral whirl. On the other hand, in the temperate 
zone the inflowing winds will be strongly de- 
flected to the right of their intended path; and 
they must depart widely from a direct line to 
the centre of low pressure, forming a whirl 
often one thousand miles in diameter: but, un- 
less they inclined inward at a distinct angle, it 
would take them too long to reach the centre, 
and their strength would be lost in overcoming 
friction on the way. ‘Their average inclination 
is therefore well marked. The steeper incli- 
nation of the winds close to the centre, ob- 
served in some northern storms (Toynbee), 
may be an effect of the tornado action in the 
cyclone, yet to be described. 
Second, in regard to the sides of the storm, 
as affected by its progression. The inclination — 
will generally be less than the average in front 
and on the right, and greater in the rear and on 
the left of the centre; for in whatever manner 
the storm advances, either by bodily transfer- 
rence or by successive transplanting, the motion 
of the wind must partake both of the direction 
of whirling and direction of progress, when 
seen by an observer not moving in either of 
these directions. In the case of bodily trans- 
