DOVE ON THE LAW OF STORMS. 913 
mediate consequence of the bordering of the monsoon and the 
trade. Horsburgh says expressly*, that on the south coast of 
China the typhoons from July to September make the wind vanes 
when near the coast point successively N.W., N., N.E., E., S.K. 
and S., and that further off the coast they point instead to N., 
N.W., W., S.W., S. In other words, the typhoons are storms 
rotating S. E. N. W., passing along the coast from east to west, 
so that the northern half of the whirlwind impinges on the coast, 
and the southern half covers remoter portions of the sea. The 
Raleigh typhoon of the 5th of August, 1835, which passed from 
Bashee Island, between Luconia and Formosa towards Macao, 
in the direction from E.S.E. to W.N.W., is a recent example of 
these storms, and corresponds perfectly to the description which 
has just been given. 
But if these rotatory storms arise from the south-west mon- 
‘soon being more southerly on its eastern than on its western 
side, and if for this very reason they move from east to west, 
they will prevail by preference in the eastern part of the Indian 
ocean; and in fact it had been remarked by Dampier that on 
the coast of Coromandel storms are looked for in April and 
September, which are the months of the change of the monsoons, 
whilst on the Malabar coast they are frequent during the whole 
westerly monsoon. 
Having thus found in the typhoons a confirmation of the prin- 
ciples which were applied to the West India hurricanes, we may 
proceed to consider in greater detail the phenomena which ac- 
company these great disturbances of the atmosphere. 
_ When in the storms of the regions of the trades the rotating 
eylinder from the lower trade impinges on the upper current, it 
is evident that inasmuch as a south-westerly direction of the 
wind prevails above, the reasoning which has been made use of 
for the lower part of the same cylinder when it passed beyond 
the external limit of the trades, becomes applicable to its upper 
part, which will immediately spread, and will advance in a dif- 
ferent direction from that of the lower part of the whirlwind. 
Thus will arise the secondary phenomenon of suction in the 
middle of the whirlwind, producing a diminution of pressure on 
the surface of the earth, and this for two reasons, viz. inasmuch 
as the rotation causes the air to fly from the centre, and as more- 
over the whirlwind widens conically in ascending, and conse- 
* India Divectory, vol. ii. p. 288. 
