DOVE ON THE LAW OF STORMS. PA | 
points in 6. The phenomena in the southern hemisphere may 
be derived in the same manner; the rotation is in the opposite 
sense, and the change of direction on passing the boundary of 
the tropics is analogous*. 
As the West India hurricanes originate at the inner boundary 
of the trade winds, where at the limit of the so-called region of 
calms the air ascends and flows over the trade in an opposite 
direction, it is probable that portions of this upper current pene- 
trating through the lower one, give the first occasion of those 
storms. The high mountains of several of the islands, by offer- 
ing a mechanical impediment, may be one cause of this effect, 
as the air flows with redoubled velocity between two mountains. 
The reason of the course of the storm being in the first instance 
from S.E. to N.W. may be explained by the circumstance, that 
according to the theoretical deduction which has been given, this 
direction is the one most favourable to the origination of a rota- 
tory movement. If, as may happen, the first impulse should be 
from S.W. to N.E., the north-east trade blowing in the opposite 
direction would equally check all the points of the advancing 
line, so that no tendency to rotate would be produced. 
It is evident that if the above deduction of these phanomena 
be the true one, a similar whirlwind must be produced when- 
ever, owing to any other mechanical cause, a current flowing to- 
wards higher northern latitudes is more southerly on its eastern 
side than on its western, where the direction is more towards 
the east. Observations collected by Piddington+ show this to 
have been the case in the storm in the bay of Bengal on the 3rd, 
4th and 5th of June, 1838. It was one of those storms which 
usually accompany the change of the north-east to the south-west 
monsoon, which change takes place in the bay of Bengal between 
* The above derivation of the rotatory movement only applies when great 
masses of air, of a considerable extension in breadth, are set in motion; lesser 
whirls of wind or water, as water-spouts, &c., are produced by other causes, 
and therefore probably do not show either regular direction or definite opposi- 
tion in reference to the two hemispheres. Mr. Redfield observed a small tor- 
nado rotating like the greater storms; but Colonel Reid saw from the Govern- 
ment-house, Bermuda, a water-spout rotating in the opposite sense. The ob- 
servations of Akin at Greenbush near Albany, of Dwight at Stockbridge in 
Massachusetts, and of Dr. Cowles at Amherst, of whirlwinds of great force taking 
| place in forest fires in calm weather, show that a strong ascending current may 
| also produce a rotatory motion. 
+ Researches on the Gale and Hurricane in the Bay of Bengal on the 3rd, 
4th and 5th of June, 1838, being a first Memoir with reference to the Theory 
of Storms in India. Journal of the Asiatic Society of Bengal, No. 91, p. 550; 
Second Part, No. 52, p. 631. 
