216 Proceedings of the Koyal Society of Edinburgh. [Sess. 
last, which covers the period 1912-14, and that this has been associated 
with a progressive rise in the mean annual temperature of St Helena 
similarly bloxamed since the three-year group 1906-08.*' The values in 
metres per second of mean wind velocity at the South Orkneys, and the 
air temperatures at St Helena are as follows: — 
Period .... 
1904- 
1906. 
1905- 
1907. 
1906- 
1908. 
1907- 
1909. 
1908- 
1910. 
1909- 
1911. 
1910- 
1912. 
1911- 
1913. 
1912- 
1914. 
Wind velocity. South 
Orkneys 
6-9 
6-3 
5-5 
5-3 
5-2 
5*0 
4-7 
4-7 
4-5 
Temperature, St Helena 
0° F. 
61T 
61-2 
60-6 
60-7 
60-9 
61-3 
61*7 
62-4 
62-7 
The wind velocity thus shows a steady diminution throughout, and this 
is associated since 1906-08 with a rise in the St Helena mean annual 
temperature. The wind velocity at the South Orkneys was thus a third 
less in the last group of years than in the first, while the St Helena tem- 
perature increased 2°T during the period 1906-08 to 1912-14. There is 
no doubt that in the Antarctic circulation there will occur from one year 
to another, or even between groups of years, certain great changes which 
must profoundly modify conditions in lower latitudes. At M‘Murdo 
Sound the mean wind velocity during Scott’s first expedition (1902-04) 
was only half that recorded during the second expedition (1911-12). In 
the Weddell sea area, as shown by the data from the South Orkneys and 
Snow Hill, storms were frequent and violent from 1902-06, while from 
1907-14 few storms have been experienced. 
There is steadily growing evidence regarding the important relations 
that exist between Antarctic conditions and those in lower latitudes, but 
the discussion is hampered by the sporadic manner in which the circum- 
polar material accumulates. In general the data refer to relatively short 
periods covered by expeditions, working as a rule in widely separate 
reofions. The establishment of fixed observatories in high southern 
latitudes would certainly lead to many practical results of the first im- 
portance, especially in the direction of utilising the material as an aid to 
long-range weather forecasting in Australia, South America, and South 
Africa ; but for this we should require synchronous data extending over 
many years from several typically Antarctic stations. 
* Data to 1908 from The Trade Winds of the South Atlantic, M.O. 203, supplemented by 
later data, kindly forwarded by Sir Napier Shaw, F.K.S. 
{Issued separately September 27, 1915.) 
