TROPICAL METEOROLOGY 
trades. A large part of the trade air, both at the 
ground and aloft, passes westward as a solid east 
current, quite analogous to a jet of fluid being directed 
into another fluid at rest. Like that jet, the east stream 
spreads downstream into its surroundings. It is also un- 
stable, so that the slightest disturbance of the stream 
leads to waves which grow in amplitude and ultimately 
form a series of vortices which either move out of the 
stream into higher latitudes or end stagnating in the 
“monsoon low’ over the land in the extreme west. 
Downstream the air is thoroughly mixed, there are 
only small temperature gradients, and it is here, if any- 
where, that the dynamic mechanisms suggested by 
Rossby in his latest paper will be valid (see Fig. 4). 
We have not, it will be noticed, done more than 
suggest a working hypothesis to apply to the large- 
ISS 3 
mixing — }westward moving 
vortical EEirculations with a ™ 
wide! range /of lintensity i 
ey F- — I] 
Region of | ri 
] . 
unst 2 
nstable a 
Ny), “Sa |: 
877 
school. The elucidation of the dynamics of easterly 
waves in very low latitudes is badly needed—an elu- 
cidation that, by including vertical motions as well as 
horizontal, would cover the main features of both the 
equatorial and the Freeman wave, since the latter is 
clearly only a limiting case of the former. It is probably 
too early to ask for a dynamic explanation of the kine- 
matic features we now know attend the transformation 
of an equatorial wave into a vortex, since these will 
clearly involve perturbations of large and changing 
amplitude, with the consequent difficulty of solving 
nonlinear equations. But we may fairly ask the dy- 
namic meteorologists for a treatment of the relation 
between the energy of the equatorial perturbations and 
that of the east current, particularly as the synoptic 
evidence suggests that kinetic energy is transferred 
ai Region of northeast trade winds 
Region of 2 
stable_ equatorial~: 
(_X_,)—of 
. . = 
equatorial ---Dry :zone|of Central P 
east current 
Fie. 4.—Schematic explanation of the mean circulation of Fig.3 in terms of a “general circulation” (trades and equatorial 
easterlies) and its perturbations, stable and unstable. 
scale features of the motion in a single ocean. It is 
premature to advance beyond this point while we have 
so little data over the land. But a great step forward 
would be taken if we could explain the main features 
of the circulation over one ocean. The gap in the data 
is clearly in the eastern oceanic sectors. The only ex- 
tended series of observations in such a sector are those 
of the Meteor in the Atlantic [18]. We require similar 
data for the eastern sectors of the North and South 
Pacific and the South Indian Oceans, and clearly the 
first region to investigate should be the eastern sectors 
of the tropical Pacific, about which little is known. A 
full program of upper-air soundings in this region would 
enable us to decide whether the direct-cell model of 
the trades is to be taken seriously, for, with compara- 
tive isolation by the high mountain barriers to the 
east, it is here that we would expect the most favorable 
conditions for a direct thermal circulation to develop. 
There remain other projects which we have sug- 
gested in the previous discussion of the perturbation 
from the east current to the perturbations as the latter 
move downstream. The approach of Kuo [40] here 
seems the most promising. 
The problem of tropical-cyclone formation is at pres- 
ent so different in aspect from what it was ten years 
ago that one is encouraged to think that the rate of 
progress will be maintained and will result in a com- 
plete solution in the near future. It now appears that 
the problem is that of deciding the sufficient and neces- 
sary conditions for the rapid deepening of a vortex 
already formed from an easterly wave, whether this be 
in high tropical latitudes, as in the Caribbean, or in 
low latitudes, as in the eastern Atlantic and the Pacific. 
We know that this deepening never occurs over the 
land, in spite of a plentiful supply of vortices; clearly 
a very moist atmosphere is necessary for the process 
to occur. But, though we have cause for optimism con- 
cerning the future of hurricane research, the problems 
connected with forecasting the movement, especially 
the recurvature, of tropical storms, remain as they were 
