AEROLOGY OF TROPICAL STORMS 
important frictional term in the equation of motion. 
As a prelude for further objective work, an extensive 
theoretical study should be executed. 
Upper Outflow. The laws governing the outflow aloft 
(Fig. 2) have been described more satisfactorily than 
those governing the lower inflow. The pressure-gradient 
force decreases upward on account of the temperature 
distribution and the gradient actually reverses above 
10-13 km [87]. Outflow will take place throughout the 
layer of decreasing pressure gradient if the air conserves 
its absolute angular momentum [13]. This conserva- 
tion principle is more likely to hold true for the upper 
outflow, since ground friction is absent. Above the 
friction layer, the principal problem is the almost com- 
plete absence of data. There are many reasons that 
suggest an upward decrease of the circulation. Such 
wind measurements as have been made have shown the 
decrease and eventual disappearance of the vortices 
between 300 mb and 100 mb. In some cases, the circula- 
tion has actually become anticyclonic at 200 mb. 
Z 
_ Fic. 3.—Same as Fig. 2, outside the eye. Solid lines are 
isobars and dashed lines are isotherms. 
Dynamical Aspects of the Eye. The only explanation 
given for the eye so far is that the law, vr = const, can- 
not hold near a vortex center, as wind speeds and 
kinetic energy there would have to become infinite. 
Friction alone would act to prevent this. This negative 
statement is reasonable, but it hardly explains the 
absolute calms frequently observed or the variable 
diameter of eyes. We also have no knowledge concern- 
ing the rate of vertical motion in the eye. The direction 
evidently is downward, at least below 300 mb. Con- 
vergence must exist in the upper troposphere and diverg- 
ence in the lower layers [28]. This is indicated by the 
stability distribution in the few available eye sound- 
ings (Fig. la). It follows from continuity reasons that 
air must constantly seep through the boundary of the 
eye in the low troposphere and enter the rain area. 
Beyond this statement, little progress is likely to take 
place in the future until actual observations of winds 
aloft in and near eyes become available. 
Maintenance of Tropical Storms. Qualitatively, it is 
907 
easier to understand the maintenance of tropical cy- 
clones than of extratropical storms. The pressure de- 
creases toward the center and the temperature rises 
(Fig. 3). Air moves inward near the ground and upward 
in the central region of convergence, while the outflow 
is aloft. Thus the circulation about the vertical solenoids. 
is in the kinetic-energy-producing sense (excepting the 
eye). This solenoid field will be maintained as long as 
colder air with lower moisture content does not enter 
the center; and as long as the air evicted aloft does not 
descend in the immediate outskirts but is removed 
considerable distances. A hurricane and its surround- 
ings cannot be a closed thermodynamic system. 
As the maintenance of tropical storms does not appear 
to offer very difficult problems in principle, future re- 
search can turn to quantitative considerations: the 
measurement of heat generation, conversion into kinetic 
energy, frictional dissipation, etc. Again the need for 
accurate data stands out as the foremost requirement 
to advance such research. 
FORMATION OF HURRICANES 
Seanty as is the literature on storm structure, the 
number of papers written on formation is very large. 
Up to very recent days, investigators generally have 
sought some grounds leading to extensive surface con- 
vergence and attendant development of cyclonic circu- 
lation. As Brunt [4] points out, the question of the 
upper outflow necessary to account for the observed 
surface pressure falls has not been answered. Conver- 
gence near the ground should lead to pressure rise and 
not fall at the center. This is a basic paradox of meteorol- 
ogy, an obstacle to the explanation of extratropical as 
well as tropical cyclones. It is true that upper warming 
caused by convection currents will produce radial out- 
ward acceleration aloft. Yet it is not clear—and has not 
been shown—why this outflow should become greater 
than the inflow. If the low levels lead in producing 
storms, the surface convergence must precede the high- 
level divergence. It is only in the last few years that 
attempts have been made to reverse the problem and 
consider the high-level events as leading [30, 35]. If 
these attempts should sustain the test of time, it would 
follow that the low-level convergence develops as a 
reaction to the pressure changes imposed from aloft. 
Convection Theory. The classical thermodynamic 
(convective) theory of hurricane formation has few 
defenders today. Many of the heaviest tropical rain- 
falls occur without the existence of a cyclonic circula- 
tion, and even closed depressions with torrential pre- 
cipitation often fail to deepen. Such centers have been 
observed to travel in relatively steady state over dis- 
tances in excess of 1000-1500 miles. It follows that 
convection is not a sufficient criterion for storm de- 
velopment. 
No one, however, denies that convection is necessary 
for generation of tropical storms. As stated before, the 
air that fills the body of a hurricane is all drawn from 
the mixed layer near the surface. Therefore the prop- 
erties of the surface air and their variation with time in 
part determine whether or not storms will form. Pal- 
