ANTARCTIC ATMOSPHERIC CIRCULATION 
more than affect the pressure and modify the strength 
of the prevailing easterly winds. A much stronger anti- 
cyclone, or one passing closer, would give Little America 
the “deep southerly current aloft (which) attended or 
preceded settled clear conditions and good visibility,” 
as Harrison [55] found on the first Byrd expedition, and 
as successive meteorologists at Little America have 
verified. 
To attribute all weather phenomena of the Ross Sea 
to dying cyclones from the northwest, as Palmer and 
Ramage have done, is to consider only half the picture. 
Perhaps the antarctic anticyclone develops lobes which 
become separate centers, one or more of which flow out 
929 
antarctic meteorology and help settle the acrimonious 
dispute over antarctic pressure waves. 
TEMPERATURES 
Upper-air temperatures and pressures were first 
measured over Antarctica in 1911 (Table IV), but only 
the nine-month series at Little America III during 
1940-41 [47] and the 1947 summertime observations are 
sufficiently extensive, in number and height, to shed 
much light on the atmospheric circulation. Simpson’s 
12 balloon meteorograph flights at Cape Evans in 1911 
[112], Barkow’s 128 kite meteorograph flights from the 
Deutschland in 1912 [101], and the 57 kite and airplane 
Taste IV, Antarctic anp SuBANTARCTIC UprER-AIR TEMPERATURE SOUNDINGS 
(Information to 1 January 1950 for all places south of 50°S, except South America.) 
From To Location Method* No. ee level 
13 Aug. 1911 25 Dec. 1911 Cape Evans, McMurdo Sound, Ross Sea B 12 6743 
— Mar. 1912 — Jan. 1913 Deutschland, Weddell Sea K 128 2400 
22 Sept. 1929 20 Dee. 1929 Little America, Ross Sea K 24 2559 
13 Nov. 1929 7 Dee. 1929 Little America, Ross Sea A 5 3548 
3 Sept. 1934 9 Jan. 1935 Little America, Ross Sea A 28 4929 
— Oct. 1934 — Nov. 1934 Deception Island, Palmer Peninsula R 6 — 
20 Jan. 1939 6 Feb. 1939 Schwabenland, South Atlantic R 36 — 
26 Apr. 1940 15 Jan. 1941 Little America, Ross Sea R 189 33750 
15 Dec. 1946 15 Mar. 1947 “Highjump,”’ all coastal waters R 348 19900 
— Dee. 1946 — Mar. 1947 Willem Barendsz, Northeast Weddell Sea R 90 = 
15 Dec. 1947 15 Feb. 1948 Edisto and Burton Island, 90°E to 70°W R 104 19183 
— Dee. 1947 — Heard Island, 53°01’S, 73°08’ R cont. = 
— Jan. 1948 — Macquarie Island, 54°30’S, 158°57’E R cont. = 
— Feb. 1949 — Feb. 1949 Commandant Charcot, 136° to 164°H R 25 — 
* Balloon meteorograph (B), kite meteorograph (K), airplane meteorograph (A), radiosonde (R). 
into the westerly circulation. This process may be 
caused by the invasion of air from lower latitudes in 
another quadrant, as Lamb suggested, or by the im- 
pinging of cyclones on the anticyclone in such places 
as the Ross and Bellingshausen Seas; possibly such 
impinging occurs only because the anticyclone already 
has broken into separate cells. In the present state of 
meteorological ignorance, these pomts cannot be de- 
termined definitely. 
Regardless of the first cause, such interplay of sub- 
antarctic cyclones and polar anticyclones is a plausible 
explanation of observed conditions. The anticyclones, 
travelling north or northwest, will give pressure maxima 
as they progress, so that pressure waves may be found 
moving in the same direction. The cyclones, moving 
east or southeast, will give pressure minima moving in 
that general direction. 
Probably most of the minima of the controversial 
antarctic pressure waves are due to subantarctic cy- 
clones, and many of the maxima may represent only the 
wedges between these cyclones, but the more pro- 
nounced maxima are apparently caused by outbreaks 
of the polar anticyclone. Application of this hypothesis 
to the available data, and examination of the newer 
data, particularly those for the summer of 1946-47, 
from this viewpoint, should provide much information on 
meteorograph ascents at Little America in 1928 and 
1934 [50] all ended well below the tropopause, and have 
been used chiefly to study the height and variations of 
the surface inversion [63, 99]. Antarctica’s first radio- 
sonde ascents, made in 1934 by Dr. Jérgen Holmboe, 
meteorologist on Lincoln Ellsworth’s attempted trans- 
antarctic flight, were so erratic that they have never 
been worked up. Records of the first successful radio- 
sonde ascents close to Antarctica, the 36 made from the 
Schwabenland in 1939 [94], were destroyed during the 
war, and only partial extracts are available [97] for the 
analyses which have recently appeared [8, 91]. 
Meteorological observations in phenomenal quantity 
were made in antarctic and subantarctic waters during 
January and February of 1947, including 348 radiosonde 
ascents by the three groups of the U. 8. Navy’s Opera- 
tion Highjump and about 90 by the Dutch whaler 
Willem Barendsz in the north Weddell Sea. All the 
Highjump ascents have been published [8], albeit only 
by standard pressure levels and arranged chronologi- 
cally regardless of location; the Dutch soundings men- 
tioned by Lamb [83] have not yet appeared. 
The following summer, two U. 8. Navy icebreakers 
obtained 104 more radiosonde observations around the 
continental edge; these are available on microfilm but 
have not been published because of the lack of prompt 
