ANTICYCLONES 
view of the behavior of the westerlies and their in- 
fluence on anticyclones both of the polar and warm 
types. 
Most of this paragraph is based on Rossby and 
Willett’s discussion of the circulation changes accom- 
panying a complete index cycle [29]. If the westerlies 
are zonal over the hemisphere, corresponding to “high 
index,” then the anticyclones are located to the south 
and are large in area, but not abnormally intense, with 
axes oriented west-east. As the mndex decreases, the 
westerlies move southward and become less zonal, per- 
mitting the formation of stronger polar anticyclones to 
the north. With further decrease of the zonal index to 
its minimum value, the westerlies develop strong ridges 
and troughs, with cutting-off of warm anticyclones in 
the higher latitudes and cold cyclones in the lower 
latitudes. This is the period of maximum dynamic 
anticyclogenesis of polar anticyclones to the north. 
When the index begins to increase again, the westerly 
waves decrease their amplitude and become longer, 
the higher-latitude anticyclones dissipate, and the 
“eycle” is ready to begin over again. 
This simplified account of the relationship of the 
westerlies and anticyclones is one of association only, 
and does not pretend to isolate causes and effects. 
What is quite clear and outstandimg, however, is that 
the phenomena must be considered on a hemisphere- 
wide scale and not as a local development closed off 
from the rest of the atmosphere. It is true that there 
are important local influences pertaining primarily to 
the character of the earth’s surface, some of which will 
be discussed later, but these must be considered as 
secondary effects superimposed on the primary pattern 
of the general circulation. 
There is definite evidence that the beginning of the 
mechanism that transforms a high-index, essentially 
zonal flow pattern into a low-index, essentially merid- 
ional flow pattern is the appearance of a “blocking 
action” in some portion of the westerlies which effec- 
tively obstructs the normally eastward motion of waves 
in the upper westerlies, or their sea-level counterpart 
in the form of anticyclones and cyclones. In the ex- 
cellent aerological study of a case of the breakdown of 
zonal westerlies, Berggren, Bolin, and Rossby [1] showed 
how a nearly zonal flow pattern in the 500-mb wester- 
lies, in which were imbedded wave cyclones, moving 
with uniform speed across the North Atlantic, encoun- 
tered “blocking” in the eastern Atlantic which caused 
the westerly waves to fold up like an accordion as their 
wave length decreased and amplitudes increased. The 
increase in amplitudes finally resulted in the appearance 
of “cut-off” cold lows in the south and warm highs in 
the north. The authors showed that the appearance of 
blocking is somewhat similar to the “hydraulic jump” 
which develops in open-channel flow when the depth 
of the water current drops below a critical depth, and 
where a certain amount of the initial energy of flow 
is transformed into turbulent or eddy energy. If the 
analogy with atmospheric ‘“‘blocking”’ is accepted, then 
perhaps a certain supercritical speed of the westerly 
jet results in the sudden breakdown of a zonal flow 
623 
into the large-scale anticyclonic and cyclonic eddies. 
The exact manner under which this remarkable change 
in flow pattern occurs is unknown, nor is the process 
known whereby the blocking, once formed, proceeds 
upstream for considerable distances—in many cases as 
a recognizable pressure rise at high latitudes capable 
of being followed westward for more than 180° of 
longitude. The importance of blocking in forecasting 
was first pointed out by Garriott [12] and has received 
further attention recently by long-range forecasters 
(9, 21]. 
The immediate origin of kinetic energy of atmospheric 
large-scale motions is a knotty problem that has not 
been fully solved. Rossby and Willett, m their deserip- 
tion of the index cycle cited above, explain that the 
return of the high index, or speed-up of the westerlies, 
is caused by the re-establishment of the normal, north- 
south thermal gradient caused by radiative cooling in 
higher latitudes and radiative heating in lower latitudes. 
But it is not clear how the new sources of atmospheric 
energy, in the form of internal heat energy or potential 
energy, are “tapped” to provide the mcreased kinetic 
energy. 
Let us turn now to the second type of dynamical 
explanation of warm anticyclones, based on horizontal 
mass convergence arising from cross-isobaric flow as- 
sociated with supergradient winds. Many ingenious 
solutions have been proposed to explain supergradient 
winds. One popular explanation is that based on the 
removal of an air parcel from its high momentum 
environment to an environment of lower momentum 
where the winds are gradient. The foreign parcel, main- 
taining its higher momentum at least momentarily, 
will, by virtue of the greater Coriolis force acting on it, 
be flung across isobars towards higher pressure. This 
explanation was first proposed by Helmholtz [16] in 
studying the stability of the circular vortex. 
Recently many investigators, while agreeing in their 
use of this basic reasoning in explaining pressure changes, 
have differed widely in their explanations as to how 
the air parcels are removed from an environment of 
high momentum to one of low momentum. Rossby 
[26] proposed that current systems possessing curved 
lateral velocity profiles would create frictionally driven 
eddies in isentropic surfaces, and that these eddies 
would transport momentum laterally thus producing 
unbalanced flow. In the free atmosphere the lateral 
frictional stresses within isentropic surfaces would be 
of greater importance than the vertical frictional stresses 
arising as a result of vertical wind shear since, in the 
latter case, the normal thermal stability of the atmos- 
phere would resist vertical motions. Rossby found from 
a simplified model that the percentual increase in total 
pressure drop across the current, resulting from the 
diffusion of momentum, would amount to 8 per cent. 
The significance of reasoning of this type is enhanced 
by the recent discovery of the narrow, fast, westerly 
““et-stream’”’ flanked by exceedingly strong shears and 
curvatures of the lateral velocity profile (e.g., see Pal- 
mén and Nagler [23)). 
Another explanation based on changes in environ- 
