946 
of the Canadian Archipelago. It rarely spreads over 
both. 
January. Among the many sources of “normal” pres- 
sure patterns it is believed that none except the 40-yr 
series have a January chart showing higher pressure 
over the Arctic Ocean than in the Canadian Yukon 
Territory. An occasional January mean map will have 
an Arctic Ocean high, January 1948 for example, but 
such a January is known to be relatively rare. By com- 
parison with Fig. 3, the Weather Bureau January map 
135" ae 90° 45” 
Fig. 3.—Average sea-level pressure (mb) during January. 
fails to indicate the true intensity and relative fre- 
quency of cyclonic activity in the Kara Sea, nor does 
it place sufficient emphasis on Greenland’s effect as a 
topographic barrier. 
Fie. 4.—Average sea-level pressure (mb) during April. 
April. By this time the mid-latitude continental 
warming has become well established although the 
Aretic remains cold with predominantly anticyclonic 
conditions. Pressures over the Canadian Archipelago 
and northern Greenland reach their annual maximum 
in April (Fig. 4). Arctic air flow southward over eastern 
POLAR METEOROLOGY 
North America is favored by the reduced zonal circula- 
tion. 
The principal weakness in the Weather Bureau 
“normal” pressure map for April is found in its failure 
to indicate the southwestward displacement of the 
North Atlantic cyclonic activity to an annual minimum 
latitude. The arctic anticyclonic circulation rarely de- 
parts from the single-cell pattern indicated for April, 
but its center may be located over northeast Greenland 
or the Canadian Archipelago when large-scale circula- 
tion anomalies disrupt the average pressure distribu- 
tion. 
Average 700-mb Contour Patterns. A survey of pre- 
vious research on the upper-level circulation over the 
Arctic was presented earlier in this article. The use of 
mean circulation methods developed by Rossby, Willett, 
and Namias has emphasized the importance of major 
anomalies in flow patterns aloft. Extended forecasting 
for the Arctic and elsewhere requires almost continuous 
reference to the upper-level “normal” maps published 
by the U.S. Weather Bureau [12, 13]. As a result of this 
writer’s research, it appears that many of the errors 
found in the 40-yr sea-level normals have been extra- 
polated up to the 700-mb charts. Namias [13] used 
many short-term radiosonde records, especially for high- 
latitude stations, by applying a correction based on ~ 
the concurrent departure from normal of the local sea- 
level pressure and surface temperature. In preparing 
July and January average 700-mb contour charts for 
this article, the elimination of errors attributable to 
incorrect ‘‘normal” sea-level pressures has received pri- 
mary emphasis. 
July. The suggested revision of the Namias 700-mb 
contours for July is shown in Fig. 5. A comparison with 
Fig. 5—Average 700-mb contours (ft) and isotherms 
(°C) during July. 
the corresponding Namias [12] chart shows the reduced 
heights over northern Baffin Bay and the Lapteyv Sea 
that are associated with the average greater than 
“normal” cyclonic activity mentioned in the earlier 
criticism of the July “normal” map. Reference has 
been made to Petterssen’s summer charts of 2-km and 
