Shaw, N.: Centres of Action in the Atmosphere. 
87 
minent examples are the Atlantic antieyclone south of the Azores and tho Ice- 
jandie area of low pressure as diselosed on monthly charts of normal pressure. 
We find it difficult to regard a region of high pressure as a “centre of action”, 
it is a region where pressure is maintained at a high level in spite of the general 
tendenecy of air to flow from high pressure to low. pressure; and, as higher 
pressure means increased mass per unit area, the access of mass which a per- 
manent anticyclone is always receiving suggests that it is a dumping ground for 
the air which has been removed from some real centre of action and not in it- 
self a centre to which such words can be applied, 
We would rather seek as the locality of centres of action those places where 
vigorous conveetion is apparently inevitable. One such area suggests itself in 
the North Atlantic to the east of the Grand Banks about lat, 45° N and long. 45° W, 
where the Gulf Stream loses itself in combination with the Labrador Current and 
the icebergs which are brought southward by it. 
In that region the sea is normally several degrees warmer than the air during 
{he winter months, though there is no such difference during the summer months, 
The difference is indieated in the maps of Fig. 1. The region indicated is notable 
as the scene of many wrecks through the ageney of the storms of winter or the 
ice of spring!). This is much more real as a centre of action than the centre 
of the Icelandie low pressure, which may be regarded as expressing a centre 
associated with the travel of dynamical systems already formed rather than with 
the influence of local conditions upon the formation or modification of a dyna- 
mical system, 
Some months ago when I was examining the question I was much interested 
to find, in a paper by V.IL Pettersson, data to show that the changes of tem- 
perature in the Atlantic to the East of that locality are not in correlation with 
‘he changes of temperature in the Gulf Stream, The conelusion at which I had 
arrived about the region near 45° N lat. and 45° W long. is expressed by Petters- 
son in words that I have used myself: “A real centre of action of the Ocean.” 
A study of the maps of the normal difference of temperature of sea and air 
over the Atlantic shows marked excess of sea temperature over air temperature 
during the winter months, especially in February; but little or no difference in 
the summer months, 
The consideration of F, M, Exner’s work on correlation coefficients between 
the pressure in Iceland and in other regions of the globe leads me to suggest 
that, from the point of view which I am now expressing, the western area of 
the Mediterranean Sea should also be regarded as a “centre of action”. In winter 
there is a high coefficient of negative correlation (5) between pressure in Iceland 
and that over the Western Mediterranean which on the principle foreshadowed 
above can only mean that the removal of air from the Mediterranean is to be 
associated with an increase of air over the Icelandie region and vice vers8a. 
Winter is the season of Mediterranean depressions, which are formed, or perhaps 
intensified, in the western part of that Sea. The normal pressure over that region 
as indicated in the map for January is a belt of pressure less than 1020 mb. 
between a region of high pressure 1020 mb. over Europe and Asia and another 
öf the same intensity over Northern Africa and the tropical Atlantic, Air cer- 
tainly does disappear from time to time from the Western Mediterranean and 
the apparent cause of its disappearance is the warmth of the water compared 
with that of the air. The “action” in this case is similar to that of the “real 
gentre of action” of the North West Atlantic, 
A reason for thinking that the removal of air from the Mediterranean and 
its dumping in the Iceland region is a more acceptable expression of the relation 
than the alternative of removing air from the Iceland region and dumping it in 
the Mediterranean is to be found in the fact that the high correlation does not 
1.As examples we may refer to the wreck of s, =. Pavonia on 3rd February 1899 in 421° N, 
49/,° W, and to ıhe wrecks of s, s. Antinoe in 47° N, 36° W. and s. s, Laristan in 459 N, 43° W. 
on, 27 January of the eunrrent year, (See The Times 6 February 1926 and The Meteorological 
Magazine, April 1926, p. 56.) The wreck of the 8, s, Titanic on 35 April 1912 occured in 41° N, 06 W.