DOVE ON THE LAW OF STORMS. 915 
ter of observations at St. Thomas,—that dreadful pause which fills 
the heart of the bravest sailor with awe and fearful expectation,— 
receives a simple explanation on the rotatory theory, which re- 
quires that at the centre of the whirlwind the air should be in 
repose ; but appears irreconcileable with the supposition of a cen- 
tripetal inflowing, because two winds blowing towards each other 
from opposite directions must gradually neutralize each other, 
and thus their intensity must diminish more and more in ap- 
proaching their place of meeting. This takes place on the great 
scale in the trade winds ; and if the centripetal view of hurricanes 
were the just one, the same effect would necessarily be seen as 
the centre of the storm passed over the station of observation. 
But the phznomena shown by observation are widely different. 
At St. Thomas the violence of the tempest was constantly in- 
creasing up to 7230™ a.m., when a dead calm succeeded, and at 
85 10™ a.m. the hurricane recommenced as suddenly as it had 
intermitted. How can this be reconciled with the meeting of 
two winds? Besides, the air at Porto Rico should have been 
flowing towards St. Thomas at that time, and therefore should 
have been west, whereas it was N.N.E., just as is required by a 
- whirlwind of which St. Thomas was then the centre. 
A remark of the St. Thomas Observer, Hoskiaer, to the effect, 
that at each gust the mercury in the barometer sunk two lines 
and then immediately rose again to the same height as before, 
shows the diminution of atmospheric pressure to be not the 
cause, but rather a consequence attendant on the violent move- 
ment of the air. 
In considering the progressive advance of the whirlwind, we 
have not hitherto taken into account the resistance opposed to 
the motion of the air by the surface of the earth. This resist- 
ance, as Redfield justly remarks, causes the rotating cylinder to 
incline forwards in the direction of its advance, so that at any 
station the whirlwind begins in the higher regions of the atmo- 
sphere before it is felt on the surface of the earth, where there- 
fore the sinking of the barometer indicates its near approach. 
The inclined position of the axis causes a continual intermixture 
of the lower and warmer strata of air with the upper and colder 
ones, thereby occasioning heavy falls of rain, and proportionably 
violent electric explosions. The cold air appears to precipitate 
itself from the cloud, and the storm to assume the form called 
by the Greeks éxvepias. This may also explain the phanome- 
