894 
temperature gradients in the upper troposphere re- 
quired to reverse the circulation. Of the two soundings 
available from centers of tropical cyclones, one reached 
1714 km and the other 17 km; neither showed much 
evidence of having reached the tropopause, although 
the sounding in the October 1944 storm showed extraor- 
80 
100 
w 
fig <. 
a OCTOBER 8 
> 200 0100 EST 
I 
= 
300 
400 ae 
=90° -80° -70° -60° -50° -40° -30° -20° -10° 0° 
TEMPERATURE (°G) 
400 za 
500 \ 
oe OCTOBER 8 OIO0E 
o IGOOEST U% \ (IN THE EYE) 
& 600 % 
= OCTOBER 7 *, \ 
5 0400 EST :,\. 
2 700 , 
= 
‘800 
900 
1000 o 
-40° -30° -20° -10° 0° 10° 20° 30° 40° 50° 
TEMPERATURE (°C) 
Fre. 4—Soundings at Tampa, Florida, during the approach 
and arrival of a hurricane center, October 7-8, 1946. 
dinary stability im the middle troposphere. Figure 5 
shows probable tropopause variation in three tropical 
cyclones, and in Fig. 6 Simpson indicates the probable 
relation between temperature in the outer and inner 
vortex of a mature hurricane. 
ORIGIN OF TROPICAL CYCLONES 
Theories of Tropical Cyclone Development. The con- 
vective theory of the formation of tropical cyclones 
was developed many years ago and was generally ac- 
cepted until the early 1930’s. It was believed that the 
following series of events occurred in the equatorial 
trough: The warm moist surface air rose because of 
both insolation and widespread convergence, resulting 
in numerous cumulonimbus clouds and widespread 
heavy showers. Then the pressure began falling slowly 
(for reasons never very well explained), the cumulo- 
nimbus clouds gradually coalesced, and if the equa- 
torial trough was sufficiently far from the equator for 
the Coriolis force to be effective, a cyclonic circulation 
was initiated. The development of the circulation con- 
tinued until full intensity was obtained by release of 
latent heat through condensation. In some areas, no- 
TROPICAL METEOROLOGY 
tably the North Atlantic, it is now apparent that 
many, if not the majority, of the tropical cyclones do 
not develop in the equatorial trough. 
LAKE CHARLES 
SEPTEMBER 
1941 
19 CHARLESTON 
OCTOBER \ 
3 HOURS AFTER 
PASSAGE 
KILOMETERS 
Fie. 5.—Soundings at selected stations during the near- 
approach or arrival of a hurricane center. (®—height of 
tropopause; +—top of sounding, tropopause not reached.) 
(Hurricane Notes—Weather Bureau Training Paper No. 1, 
Washington, D. C., July 1948.) 
Some attempt was made in the middle 1980’s to 
apply Norwegian methods to tropical analysis. Hur- 
ricanes were supposed to develop on the equatorial 
front which at that time had several different defini- 
OUTER VORTEX 
20.000] 22] 
Fre. 6.—Probable relation between temperature in the 
outer and inner vortex of a mature hurricane. (After R. H. 
Simpson.) 
tions but which now is considered as a narrow zone, 
usually associated with the equatorial trough, in which 
air is undergoing horizontal convergence. However, 
there are no significant temperature differences along 
the equatorial front, there is no density discontinuity, 
and the air is essentially homogeneous. 
Tropical cyclones originate in easterly waves, in the 
intertropical convergence zone, and occasionally in the 
trailing southerly portions of old polar troughs. Cer- 
