METEOROLOGICAL ANALYSIS IN MIDDLE LATITUDES 
the investigations of pressure-jump phenomena show 
such promise that it seems timely to consider the steps 
which would have to be taken to revise our present 
procedures of atmospheric analysis with the purpose of 
taking full advantage of what we already know about 
this phenomenon. The first revision of synoptic practice 
that is indicated by this work would be at the observing 
level, namely, the development of the necessary instru- 
mental and observational techniques for detecting the 
passage of a pressure jump at each meteorological 
station. Secondly, we would be obliged to codify and 
transmit the data pertinent to pressure-jump passages 
from the observation point to the forecast and analysis 
centers. In all probability it would be necessary to make 
use of the code for special observations to report the 
position of the pressure jump when it is initially noted. 
Subsequently its position could be included in our three- 
and six-hourly coded reports. 
Finally, it would be essential that we decide upon 
conventions for detecting pressure-jump lines on our 
synoptic charts. This would entail, first, a symbol for 
entry on our synoptic charts to represent the pressure 
changes associated with pressure jumps and, secondly, 
a set of rules for enabling analysts to locate, uniquely, 
pressure-jump lines on our synoptic charts. The method 
of graphical representation of pressure, temperature, 
wind, and moisture data, suggested by Bellamy [3], 
appears to be well suited for showing the data pertinent 
to the detection of pressure jumps, while at the same 
time it lends supplementary emphasis to the other 
portions of the data used in the type of air-mass analysis 
customarily performed. 
Bellamy’s method for the graphical representation of 
data warrants the attention of all meteorologists since 
it constitutes, in point of fact, a convenient and practi- 
cable means for applying to analytical procedure the 
disclosures of many divergent investigations in the 
field of modern meteorological research. We have 
already touched upon the usefulness of Bellamy’s 
method apropos of pressure jumps. Turning our at- 
tention now to an entirely different topic, namely, the 
meteorological significance of the results of recent radar 
studies, we find that again Bellamy’s graphical plotting 
method may profitably be applied. 
Use of Radar as an Analytical Tool 
The advent of radar had sudden dramatic repercus- 
sions throughout the field of physical science. In mete- 
orology A. Bent and R. Wexler [23] were among the first 
to apply this new scientific tool, utilizmg radar in the 
investigation of the structure of precipitation clouds. 
Subsequently others have conducted further research 
along these lines. 
Recent radarscope movies, produced by A. C. Bemis 
at the Massachusetts Institute of Technology, where 
this work has been carried on for several years, virtually 
compel us to adopt some new type of cloud analysis. 
These radarscope pictures refute the classical model of 
an active cold front attended by a rather solid line of 
cumulonimbus with occasional holes between the cells, 
frequently replacing this model with the picture of the 
717 
cumulonimbus centers themselves, as well as the breaks 
in the clouds, organized in space in a series of rows, 
roughly parallel to each other and nearly parallel to 
the atmospheric flow above the frontal surface. Simi- 
larly, the disclosure that there are, as a rule, parallel 
bands of precipitation associated with warm fronts 
constitutes a marked departure from the accepted 
model. Bellamy’s cloud graph, with height above the 
surface as the ordinate and time as the abscissa, permits 
inferences concerning the horizontal distribution of 
frontal clouds. 
The same sort of graph has been used to represent 
the data obtained by turning a radarscope upwards to 
investigate the distribution in the vertical of clouds 
associated with fronts. At the time the radar studies 
were made, pilots and most meteorologists connected 
with airplane operations had already come to disagree 
with the model of the clouds associated with an ap- 
proaching frontal system which is found in most text- 
books: the familiar picture of cirrus gradually thickening 
and lowering, merging into altostratus, and finally 
stratus, with a solid cloud deck extending from the 
cirrus level down to the lowest cloud base. Radar 
studies have confirmed the findings of those who have 
maintained that the cloud pattern over a warm front 
consists of several distinct layers of clouds, generally 
not merging except over limited areas. Radar photo- 
graphs of the vertical structure of clouds are now being 
made at the Signal Corps Laboratories in Belmar, New 
Jersey, under the able direction of Dr. Michael J. 
Ference. One of these pictures, illustrating the cloud 
deck in advance of a warm front from a coastal storm 
in the eastern United States, is reproduced as Fig. 9 
(p. 1220) in the article by Dr. Ference in this Com- 
pendium. 
In view of these findings, a systematic investigation 
of our classical model of cloud structure should be under- 
taken to determine details concerning the apparent 
wave pattern and layer structure. Certainly these 
changes should be brought to the attention of pilots 
and all those who are responsible for the meteorological 
training of pilots. We need, too, to perfect and put into 
operation the instruments needed to make observations 
of both horizontal and vertical cloud patterns so that 
such data can be transmitted in our regular synoptic 
weather codes. Once these data have been received in a 
weather station, Bellamy’s method or some other 
method of graphic representation can furnish a practi- 
cable basis for the analysis of clouds. 
Upper-Air Analysis 
Since the structure of clouds is intimately connected 
with the configuration of upper-flow patterns, the re- 
vision of our model of cloud structure implies consonant 
changes in the technique of evaluating the data on 
upper-level charts to include the analysis of vertical 
motions. Although research meteorologists have already 
tackled this problem, unreliable data have forestalled 
its solution. For this reason we shall restrict ourselves 
to the consideration of the methods of analysis of upper- 
level charts in general use today. 
