THE PHYSICAL BASIS FOR THE GENERAL CIRCULATION 
large fraction of the kinetic energy so transferred should 
remain without being destroyed by the mean conver- 
gence, while if these pressures are high practically all 
the kinetic energy fed into the cap should disappear 
because of the mean convergence. For this purpose it 
was decided to examine 7 with reference to the area- 
mean pressure deficit north of 45°N as compared to the 
pressure at 45°N. This pressure deficit was denoted by 
a and was computed from sea-level pressures. It be- 
came apparent that abnormally large values of 7 are, 
generally speaking, encountered only when the pressure 
deficit is small or negative, in other words when the 
capacity of the polar cap to destroy kinetic energy is 
abnormally great. 
From these general observations it would appear that 
large meridional transports of internal heat energy and 
of kinetic energy take place during those periods when 
the mean meridional pressure distribution at lower 
levels is relatively flat north of the subtropics with no 
marked polar low present. How can a situation such as 
this maintain itself over appreciable intervals of time, 
as is known to be the case? Is it a quasi-steady state or 
is it a dislocation which becomes progressively more 
difficult to maintain? We can only speculate about the 
answers, but it is rather interesting to do so. 
First let us consider the fact that it is during these 
low-index conditions that very cold air masses are in- 
jected into southerly latitudes. This in turn means a 
very strong heating of the atmosphere in middle lati- 
tudes and the subtropics, especially over ocean areas, 
so that a rapid replenishment of heat and also a rapid 
generation of kinetic energy would be a normal ac- 
companiment of this state of affairs. This consideration 
would argue for the possibility of quasi-steady state. 
On the other hand in order to maintain a quasi-steady 
condition, means would have to be present to dispose 
of large quantities of heat in the polar cap. It is unlikely 
that the net heat loss through radiation could be 
stepped up enough to meet this requirement. Progres- 
sive melting of ice in the polar regions could absorb 
large quantities of heat, but probably exactly the op- 
posite is actually the case, since very low temperatures 
are apt to prevail at low levels in the arctic and sub- 
arctic regions under these circumstances. 
Observationally there is some qualitative evidence 
that during prolonged low-index conditions there is apt 
to be present in polar regions an accumulation of ab- 
normally warm air a little distance above the surface 
in the troposphere and in the stratosphere. Also there 
is some evidence that there is actually a progressive 
depletion of heat energy from middle latitudes (in spite 
of the strong surface heating) with a consequent pro- 
gressive southward shift of the latitude of the maximum 
westerlies at higher levels. What ultimately puts a stop 
to the trend of developments is not obvious. 
Contrasted with the low-index regime, during periods 
of a strong polar vortex at low levels the heat and 
kinetic energy transport is weak. Relatively warm air 
streams across the continents and oceans in a more 
zonal manner so that no very vigorous heating takes 
place. So far no difficulties seem to be present as far 
547 
as the more southerly regions are concerned. It would 
appear, however, that progressive cooling should now 
take place in the arctic regions since the abnormally 
low heat input would appear to be insufficient to main- 
tain the status quo. 
On the whole there may be good and sufficient physi- 
cal reasons, such as those mentioned above, why the 
general circulation does not persist indefinitely in one 
or the other abnormal conditions but rather tends to 
shift about a general average state. There is neverthe- 
less the important question why the average state does 
not persist without such pronounced departures as 
those just described. The writer’s conjecture on the 
matter is that the average state can too easily be dis- 
rupted by the effects of the nonzonal continentality 
found especially in the Northern Hemisphere. Such 
effects may be partly due to the fact that the continents 
provide avenues for the southward penetration of in- 
tensely cold air masses which can thus arrive at rela- 
tively low latitudes without undue modification. The 
occasional outpourings of such cold masses over ad- 
jacent water areas at fairly low latitudes could easily 
result in abnormal heating of the atmosphere sufficient 
to disrupt the average regime. 
Also it should not be forgotten that large mountain 
ranges extending over an appreciable spread of latitude 
can exert powerful mechanical effects upon the atmos- 
phere, as pointed out by White [16]. By way of example 
large pressure differences across the Rockies in North 
America could perhaps disrupt the angular momentum 
balance in the belt of the prevailing westerlies. 
Finally the suggestion made by Willett [18] that the 
seat of the variation could perhaps be found in the 
variations in solar output, and hence the heating of 
the atmosphere, deserves study. For it may be that the 
final answer is bound up in a combination of effects, 
especially if long-term aberrations of the general circu- 
lation during historic and geological periods are also 
considered. 
From what has been said in the several preceding 
paragraphs it might appear that too little emphasis has 
been placed upon the influence of the conditions in the 
upper troposphere and in the stratosphere. In the final 
analysis it is obvious that the entire atmosphere must 
form an integrated system. It would seem, however, 
that the lower layers furnish the seat of thermodynamic 
activity and provide the motive force for the general 
circulation. The writer’s colleague, Dr. H. L. Kuo, is 
currently engaged in a theoretical study of the effects 
which might result at the level of the jet stream from 
forced disturbances originating in the lower levels. From 
his preliminary work it would appear that the genera- 
tion and maintenance of the jet stream itself are a 
necessary consequence of such impulses received from 
below. Likewise the characteristics of the angular mo- 
mentum balance as outlined previously, including such 
features as the tilted (and properly curved) trough and 
ridge lines, as well as the development of regions of 
sharp shear to the north and to the south of the jet 
follow as consequences of the analysis. 
Having made an excursion into some of the ramifi- 
