926 
in both hemispheres, parallel variations in circulation 
would be expected. 
As yet, no series of Southern Hemisphere maps is 
adequate to give zonal index data for comparison with 
the Northern Hemisphere, but current research should 
provide preliminary answers. With any correlation thus 
established between indexes of the two hemispheres, the 
Southern Hemisphere index can be estimated, from 
Northern Hemisphere values available back to 1899, for 
those periods of detailed antarctic observations. Pend- 
ing such further study, the possibility remains that the 
two periods whose weather has been studied intensively, 
1912 and the 1946-47 summer, were both rather ab- 
normal, 1912 having a very low zonal index with more 
storms and more meridional air exchange than usual, 
and the 1946-47 summer having a very high zonal 
index with fewer storms and much less meridional ex- 
change than usual. Furthermore, variations in the circu- 
lation must be considered in evaluating any meteorologi- 
cal data from Antarctica. 
WAVES 
Characteristics. Nonperiodic waves of atmospheric 
pressure, found on the barograms (actual or recon- 
structed from barometer readings) of all antarctic sta- 
tions with an amplitude of about 0.6 in., are a very 
controversial phenomenon of antarctic meteorology. 
Only the nature of the polar anticyclone has aroused 
more discussion than these waves, which are defined by 
agreement as having a change from minimum to maxi- 
mum of at least 0.2 in. or 5 mm (6.7 mb). 
That such waves can be found is not denied; the 
argument is whether they are “true pressure waves 
traversing the upper atmosphere, . .. travelling outwards 
from the center of the antarctic continent,” as postu- 
lated by Simpson [112] and supported by Loewe [87] 
and Lamb [81, 82, 83], or merely reflect the passage of 
ordinary fronts and depressions, as vehemently main- 
tained by Meiardus [14], Reuter [98], Kidson [74, 75], 
Palmer [89], and Ramage [64]. 
Unfortunately, these waves, about whose cause and 
exact place of origin Simpson did not speculate, have 
been often confused with the long-period surges, for 
which he gave both cause and place of origin. These 
surges, found from 10-day running means of pressure, 
have a constant period of about one month, but decrease 
in amplitude about 0.162 in. per thousand miles as they 
travel outward in all directions from 80°S, 120°W, where 
the extrapolated amplitude is 0.677 in. They are “‘due 
to increases and decreases in the intensity” of the upper- 
level cyclone “with the centre of lowest pressure over 
the low-lying region which we have reason to believe 
exists in the Pacific quadrant of the antarctic”; the 
epicenter was determined so that the observed ampli- 
tudes would fit the assumed law of linear decrease with 
distance. The surges explain why monthly pressure de- 
partures throughout Antarctica tend to vary similarly, 
and inversely from pressure departures in South America 
and the southern portions of Africa and Australia; surges 
are not reflected in local weather phenomena. 
The only visit thus far to the presumed epicenter of 
POLAR METEOROLOGY 
these surges, by Wllsworth [120] in the spring of 1985, 
encountered twelve days of variable bad weather, ap- 
parently from occlusions penetrating a deep bay to the 
north (see p. 921). Since it is likely that low pressure 
prevailed during this period, the daily pressure readings 
from Rymill’s Argentine Island base (65°15'S, 64°16’W) 
and Laurie Island, the only antarctic and subantarctic 
stations operating during 1935, could be tabulated in 
10-day running means to determine whether Ellsworth’s 
visit corresponded with a surge minimum. 
Waves and surges were discussed by Simpson sepa- 
rately, the only connection between the two being that 
the apparent line of wave propagation, extended back- 
ward, “passes very near to the position 80°S, 120°W 
about which the pressure surges were found to have 
their maximum intensity.” Yet Kidson, Ramage, and 
Lamb have given the surge epicenter as that of the 
waves, whose origin Simpson did not particularize any 
more closely than “the center of the antarctic conti- 
nent”; even Meinardus confused waves and surges. 
Tasie III. ANrarctic PRESSURE WAVES 
._, | Ampli- 
2 No. of | Period 
Location (westward) wears (eas) Gane) 
Framheim 1 163 -624 
McMurdo Sound 48 152 572 
o ee 5 L 152 560 
Cape Adare 1 119 549 
Macquarie Island 2 98 66 
Cape Denison 2L 102 OT 
we a -M 108 61 
Queen Mary Land 1 131 59 
Gauss 1 122 .642 
Kerguelen Island 1 69 634 
Deutschland 1 124 -61 
Laurie Island, South Orkneys - 91 642 
Snow Hill Island 2 107 567 
Belgica 18 124 630 
er 1M 134 63 
* Values cited by Simpson (S), Meinardus (M), Loewe (L). 
Average amplitudes and periods of waves found for 
all the earlier antarctic stations, and the number of 
years on which they are based, are given in Table III. 
Data of expeditions of the last two decades have not 
been analyzed for such pressure waves, largely because 
their hourly values have not been published, or even 
computed in many cases. Although from 1935 to 1949 
there were nearly 40 station-years of observations in the 
Palmer Peninsula area by the British, Americans, Ar- 
gentines, and Chileans, the only data published so far 
are brief monthly summaries for Rymill’s 2-year expe- 
dition [40] and Ronne’s recent wintering [37], and a note 
on the British observations [38]. Three years of hourly 
pressures for Little America have been published [47, 
50], but they have not been analyzed for waves. 
Movement. The crux of the pressure-wave dispute is 
the asserted northwestward travel of the waves. This 
direction of propagation is taken to prove that the 
pressure changes which they cause (and from which they 
are found) cannot be due to the eastward motion of 
subantarctic pressure systems. Constancy of wind speed 
