WORLD WEATHER NETWORK 
dynamically and thermodynamically. The Southern Hem- 
isphere circumpolar vortex is evidently a more intense dy- 
namic phenomenon than that of the Northern Hemisphere 
and consequently might be expected to be the more im- 
portant of the two as the seat of disturbances or changes 
of the general circulation as a whole. 
He emphasizes also: 
The tropics must contain the principal heat source of the 
general circulation and also constitute the zone of inter- 
action of the two great hemispherical cyclonic vortices. 
Yet it is in the tropics and in the Southern Hemisphere 
as a whole that our principal lack of aerological data 
lies and, pending the establishment of a good over-all 
world network, the distribution of upper-air stations 
along two or three selected meridians would be most 
helpful. Two sections which have been suggested by the 
author and which are largely in effect are (1) a maritime 
one, from the North Pole through Alaska, the Pacific 
Islands and New Zealand, to Antarctica; and (2) a 
continental section from the North Pole through Can- 
ada, the United States, Central and South America, to 
Antarctica [12]. 
Another measure which would be of great help to 
the basic theoretical studies of the general circulation 
is the provision for the right kind of measurements 
from ships and island stations for treating the heat 
budget of the atmosphere-ocean complex as a whole. 
To put it very simply, this budget necessitates knowing 
the ingoing and outgoing water at the ocean surface 
(rainfall and evaporation) and the ingoing and out- 
going radiation at the same surface. While evaporation 
can be estimated from sea and air temperatures com- 
bined with wind speeds which are currently part of ship 
measurements, rainfall for the oceans is derived exclu- 
sively from the records of island stations. Consideration 
should be given to the measurement of rainfall on mov- 
ing ships at sea. Hven though the interpretation of the 
catch of rainfall from a moving vessel presents certain 
difficulties, it is probable that such measurements would 
prove far more satisfactory than the present practice of 
estimating ocean rainfall from island stations. Small is- 
lands in extensive oceanic regions usually exert a pro- 
found and a very local orographical influence and, there- 
fore, properly interpreted rainfall data from ships in 
transit may prove far more representative. 
The establishment of radiation-measuring stations 
on properly dispersed islands in the oceans and on 
a few selected ships seems also to be a desirable meas- 
ure [13]. 
Summary 
The principal lines along which further work should 
proceed in order to fill out the world network may be 
summarized as follows: 
1. The development of automatic stations for land 
and ocean, tropical and polar use. 
2. The filling out of the land station network in 
tropical areas. 
3. The establishment of manned or automatic surface 
stations on suitable islands and reefs. 
709 
4. The establishment of upper-air stations on selected 
islands and reefs. These would be manned, but, at the 
same time, it would be desirable to initiate research on 
indirect methods of atmospheric sounding without flight 
equipment, which would lend themselves ultimately 
to automatic operation. 
5. The development of buoys, which could be 
anchored in deep water, to carry manned stations 
or automatic stations for remote oceanic points. 
6. The development of methods of getting in and 
out of arctic and antarctic areas and of maintaining 
manned stations and automatic stations in those 
regions. 
These are tremendous engineering problems which 
the humility of the meteorologist may give him pause 
to urge or to undertake. It will be recalled, however, 
that the pioneering work on rockets in this country was 
undertaken with the meteorological sounding of the 
upper air as its primary objective. Furthermore, all 
such developments contribute directly to problems com- 
mon to other geophysical fields and the support of these 
fields should be enlisted to justify the effort. The other 
aspect of the world network problem (that of inter- 
national cooperation) is no less difficult of solution 
than the technical developments involved. It is, how- 
ever, merely an extension and growth of the coopera- 
tion which is going forward in the International Meteor- 
ological Organization, the International Civil Aviation 
Organization, and in the North Atlantic (and other) 
weather ship programs. Only by beginning an attack 
on this immense problem of the world network will 
suitable data ultimately become available both for 
synoptic studies of the general circulation and for 
the kind of information which is needed in the objective 
weather-forecasting approach promised by the use of 
electronic computers. 
REFERENCES 
1. Deacon, G.E. R., ‘‘Waves and Swell.” Quart. J. R. meteor. 
Soc., 75:227-288 (1949). 
2. pEWinp7, J., ‘“‘Sur les distances moyennes A la céte dans 
les océans,’’ Mémoires couronnés et mémoires des savants 
étrangers, Tome LVII. L’Académie Royale des Sciences, 
des Lettres, et des Beaux-Arts de Belgique, 1898. 
3. Fuercuer, J. O., Floating Islands in the Arctic. Paper 
presented at the Alaskan Science Conference of the 
National Academy of Sciences-National Research Coun- 
cil, Washington, D. C., November 9-11, 1950. 
4. Garctra, R., ‘Discussion on International Co-operation in 
Obtaining Radio Soundings in the Antarctic with a 
View to Representing the Desirability of Such Obser- 
vations to the I. M. O.”’ P. V. Météor. Un. géod. géophys. 
int., Oslo, 1948. Ucele (1948). (See p. 113) 
5. Haynes, B. C., The Weather Bureaw’s Arctic Observation 
Program outside of Alaska. Paper presented at the 
Alaskan Science Conference of the National Academy 
of Sciences-National Research Council, Washington, 
D. C., November 9-11, 1950. 
6. IntTeRNATIONAL Mernoronocican Oreanization, List of 
Resolutions. Conference of Directors, Washington, D. C., 
Sept. 22-Oct. 11, 1947. Secretariat, I. M. O., Lausanne, 
1948. 
