DOVE ON THE LAW OF STORMS. 207 
In my first researches on the subject of the winds, [ had re- 
ferred both the law of rotation and the rotatory movement of 
storms to the mutual action of two currents of air, each tending 
to press aside the other; but a more close investigation of the 
phenomena has taught me to regard the law of rotation as rest- 
ing on more general conditions, and as being a simple and neces- 
sary consequence of the rotation of the earth. The principle of 
Hadley’s theory of the trade winds thus generalized, explained 
fully all the rules which had been found for the non-periodic 
variations of the meteorological instruments in the northern 
hemisphere, and permitted the prediction of rules for the south- 
ern hemisphere ; but it did not explain the rotatory movement 
of storms, and consequently when I published my Meteorolo- 
gische Untersuchungen, Berlin, 1837, which were made to embrace 
all that I had previously written on the subject, I was obliged 
to retain the earlier theoretical representation, since that which 
had been thus empirically deduced had been fully confirmed, 
but without its connexion with the principle of the general 
theory being shown. The object of the present memoir is to 
supply this deficiency. From the researches of Redfield and 
Reid we have the following facts :— 
1. Storms which originate within the tropics preserve the 
first direction of their path almost unaltered, until they enter 
either of the temperate zones, when their course becomes de- 
flected into one almost at right angles to the former. Thus the 
storms of the northern hemisphere move from S.E. to N.W., 
until they have past to the north of the tropic of Cancer, when 
their course becomes from S.W. to N.E.; and, on the other hand, 
the storms of the southern hemisphere, whose progress within 
the tropics is from N.E. to S.W., take a new direction on enter- 
ing the southern temperate zone, and then move from N.W. to 
S.E. 
2. The breadth of the whirlwind, which increases very gra- 
dually within the tropics, becomes suddenly greatly augmented 
at the time when the path undergoes the above-described flexure 
on passing those limits. The chart of the West India hurricane 
_ of the middle of August 1837, in Colonel Reid’s work, and that 
of the Mauritius storm of March 1809, in Berghaus’s atlas, are 
examples of these phenomena in either hemisphere. The course 
of storms is further illustrated by a chart of Redfield’s, in which 
the tracts of ten are laid down. The paths of two of these 
