TROPICAL METEOROLOGY 
for those storms that clearly originate in the doldrums 
but that he denies such an origin in cases where the 
storm can be traced to an easterly wave. Temperature 
differences have been reported at the equatorial front 
in other parts of the world, for example in India [60] 
and in equatorial Africa [67]; this probably predisposed 
Dunn to concede that a frontal origin of doldrum hur- 
ricanes was possible. : 
This concession, either in the form of an outright 
admission of a frontal origin of some tropical storms, 
or the lesser concession that storms that originate in 
the neighborhood of the equatorial low-pressure trough 
differ fundamentally from those that can be traced 
back to easterly waves, has been a characteristic of 
workers of the perturbation school. It has, in fact, done 
much to prevent their presenting a unified theory of 
the origin of tropical storms. Apart from this admission, 
however, the workers as a group have been rather ex- 
treme in denying the existence of fronts in the tropics. 
They have, rightly I think, insisted on the absence or 
irrelevance of temperature contrasts in the Caribbean 
and the western Pacific; but their zeal has also led them 
to minimize the well-authenticated accounts of cloud 
systems and wind-shift lines near the equator, lines 
that look like the cold fronts of high latitudes. As I have 
said, they may admit that such lines can exist in the 
equatorial low-pressure trough but deny their presence 
in the easterly wave in the Caribbean. Yet nothing is 
easier, with good aircraft reports, than to demonstrate 
the presence of such lines in certain Caribbean easterly 
waves. To admit this, perhaps, would seem to them to 
destroy the value of the contributions they have made; 
we would be back in the confusion that prevailed during 
the ascendancy of the frontal theory. The resolution 
of their difficulty must be left to the next section. 
The detailed structure of the easterly wave was 
worked out by Riehl [53]. He assumes that the basic 
undisturbed current in the Caribbean is a deep easterly. 
Tf this assumption is conceded, it is a simple matter to 
define and to detect the chief axes of the easterly wave, 
both at the surface and aloft, for the axes, horizontal 
and vertical, along which the wind is easterly will be 
the “trough” and ‘‘wedge” axes. In the Northern Hemi- 
sphere we shall find northeast winds ahead of the trough 
and southeast winds behind it. By the analysis of 
upper-level wind maps, supplemented by time cross- 
sections at individual stations, it is then possible to 
discuss the slope of the systems. Riehl showed that the 
perturbations of the wind field were of greater ampli- 
tude than those in the pressure field, and that the 
amplitude of the former also seemed greater in the 
middle troposphere than at low levels. Generally, the 
trough axis slopes toward the east with height. Further, 
the lower wind fields are positively divergent? ahead 
of the trough and negatively divergent behind it; with 
this distribution of the horizontal velocity divergence, 
4. Riehl is careful to show that the total horizontal velocity 
divergence, not merely the streamline divergence, varies in 
this manner. The distinction was never clear to workers of the 
air-mass school. 
869 
and (to a high degree of approximation) of the mass 
divergence, the westward passage of the wave is ex- 
plained, as is also the “rolling up”’ of the trade inversion 
behind the wave axis and the consequent distribution 
of cloud and precipitation in the perturbation. The 
virtue of Riehl’s work is that he establishes the correla- 
tions of the fields of motion, pressure, composition, and 
temperature to form a simple integrated picture of the 
perturbation; this has hardly been achieved for any 
perturbation in high latitudes except, perhaps, the 
frontal wave. 
Riehl also discusses another type of perturbation 
observed in the Caribbean area, one we have mentioned 
before. During the winter in that region, the east wind 
near and at the surface is frequently accompanied aloft 
by winds with a westerly component; in other words 
there may be a typical trade-antitrade regime. In the 
lower easterlies, perturbations that look superficially 
like easterly waves on the low-level synoptic map ap- 
pear; however, these disturbances move, not toward 
the west, but toward the east, against the lower easterly 
current. It is always possible to show that the trough 
in the easterlies is continuous with a surface trough in 
the westerlies of the higher latitudes and with an upper- 
level trough in the westerlies of the tropics. This trough 
Riehl called the polar trough; he accounted for the east- 
ward movement of the system in the lower tropical 
atmosphere as being a reflection of the movement of the 
trough aloft i the circumpolar westerlies. This explana- 
tion had been advanced before, of course, but in a 
different form. The older explanations were in terms of 
an upper front in the westerlies—Riehl simply omitted 
the front, thus effecting a great simplification in this 
part of the model. However, it is a fact that true surface 
fronts move southward over the United States in winter- 
time and that they may move off the continent into the 
Gulf of Mexico and the Caribbean. The surface front 
becomes weak, but can often be traced far to the south 
into Central America. It is then found that it is dis- 
sociated from the polar trough (especially in the 
pressure field) and that the chief cloudiness and pre- 
cipitation accompany the eastern part of that trough 
rather than the surface front. This picture has been 
elaborated by Cressman [16], Simpson, and others. As 
time goes on more and more complexities are introduced 
into it; one suspects that all is not well with the polar- 
trough model and that several different synoptic 
phenomena are being confused under one heading. At 
present, however, little research is being done on winter 
disturbances in the tropics and the time seems to be 
ripe for new investigations. In the meantime, if current 
synoptic maps from weather stations in the Pacific are 
a guide, there seem to be two rival explanations of 
winter disturbances in the higher tropical latitudes: 
the polar trough explanation and the older one which 
accounts for all winter precipitation in the trades in 
terms of classical fronts traveling from higher latitudes 
toward the equator. 
In 1948 Freeman [80], on the basis of wartime ex- 
periences in the New Guinea area, introduced a new 
