MECHANISM OF PRESSURE CHANGE 
The vertical accelerations are the most questionable 
of the observational data. However, it is believed that 
the comparatively slow displacement of the high-level 
field of vertical velocities as compared with the large 
horizontal velocities makes it possible to estimate the 
magnitude of the vertical accelerations in the upper 
atmosphere. The vertical accelerations in Fig. 3 were 
estimated from such considerations. This idealized 
model serves to illustrate those features of pressure 
changes which have been discussed earlier. Two other 
features of this diagram deserve attention. 
1. The pressure trough and the zone of maximum 
temperature contrast are approximately coincident in 
the lower atmosphere, consistent with the general con- 
cept of a frontal surface. However, above the middle 
troposphere this coincidence is no longer present. The 
pressure troughs and ridges in the higher atmosphere 
are near regions of maxima and minima in the tempera- 
ture field. The pressure systems move along with the 
low-level temperature field at a speed which is not 
vastly different from the horizontal wind speeds. Above 
the lower troposphere, however, the prominent features 
of the temperature and pressure field move much slower 
than the horizontal wind speeds at the respective levels. 
This behavior of the higher atmosphere may suggest 
that the primary cause of pressure changes is to be 
found in the lower troposphere. However, before such 
a conclusion can be reached it is necessary to explain 
the origin of the vertical accelerations at high levels. 
2. A developing pressure system is usually associated 
with an intensification of the high-level pressure trough 
and the associated temperature maximum in the strato- 
sphere. Figure 3 shows that this maximum arises from 
descending motion upstream from the point of maxi- 
mum temperature. Consequently a developing pressure 
system must be accompanied by downward accelera- 
tions many hundreds of miles upstream from the strato- 
spheric temperature maximum. This evidence on the 
high-level accelerational field indicates that the devel- 
opment of a pressure system is accompanied by signifi- 
cant changes over a wide area. 
Hence it appears that the problem of pressure changes 
cannot be localized to processes within the immediate 
vicinity of an isallobaric system. 
Conclusions 
Tt has been shown that there have been various ap- 
proaches to the problem of the mechanism of pressure 
change. At present the physical picture is incomplete 
and, therefore, no definite conclusion can be reached as 
to the most desirable line of approach. It is evident that 
a theory must explain an increase or decrease of mass 
in an air column. In this review the emphasis has been 
placed upon the thermal explanation of pressure 
changes since it appears to satisfy many of the observed 
facts concerning the temperature field. Moreover, a 
thermal approach is in accord with the general concept 
that heating and cooling are the primary causes of at- 
mospheric motion. However, it is apparent that the 
thermal theory has not been fully developed. Many prob- 
lems remain: 
637 
1. Theoretical analyses of the development of diver- 
gence and vertical-motion fields with pressure changes 
appear to be inadequate. 
2. The empirical picture of divergence and vertical 
motion is well established for stationary heat sources; 
however, it is evident that there are gaps in our knowl- 
edge concerning the manner in which divergence and 
vertical-motion fields develop and move with migratory 
pressure systems. Much of the empirical information 
is based upon twelve-hour changes and it is possible 
that such data give an incorrect impression of the mo- 
tion im areas of pressure change. 
3. The direct influence of nonadiabatic temperature 
changes on pressure changes requires further considera- 
tion. 
4. The details of horizontal advection require more 
investigation in order to ascertain whether advection 
should be considered a cause or an effect of pressure 
change. 
5. The complexity of the temperature field in the 
vicinity of the tropopause warrants further considera- 
tion of the stability of this field. 
6. The influence of surface friction must be further 
investigated. 
7. It is necessary to develop a thermal model which 
gives a pressure change in a quiescent region, followed 
by a moving pressure-change field with its associated 
changes at various levels, and then the end stage of a 
cold cyclone or warm anticyclone. 
Finally, it is important to note that the empirical 
data indicate that the explanation of pressure changes 
cannot be localized to processes taking place over a 
small area. This concept of the extent of the field of 
influence applies to all analyses of the problem of pres- 
sure change. For this reason there is some advantage 
In specifying a certain unbalanced condition as an initial 
state in a region where the pressure is changing and 
later investigating how such a particular initial state 
might have originated. Such a procedure, which has 
been followed in some dynamic approaches, might well 
be adopted in a thermal or convectional approach. 
Even though the pressure-tendency equations do not 
appear to lead to fundamental explanations, they play 
an important role in research on pressure change. A 
preliminary step toward an explanation of pressure 
change is an accurate description of all features of the 
wind, pressure, and temperature fields in the vicinity 
of pressure changes. The pressure tendency equations 
make it possible to check the internal consistency of 
such pictures. In connection with this important aspect 
of pressure-change research the author would like to 
suggest the desirability of utilizing to a greater degree 
other relationships which may be derived from the 
equations of motion. For example, vorticity-divergence 
relationships should also be helpful in determining the 
reality of an empirical model of a stationary or moving 
pressure-change system. 
This breakdown of approaches to the pressure-change 
mechanism into the two categories of thermal and dy- 
namic is perhaps artificial. The classification is consistent 
with the conventional use of the adjectives thermal and 
