804 REPORT—1899 
Shetland Islands, on South Orkney and on one of the Sandwich Islands, together 
with stations at Cape Pillar, Cape Virgins, Cape Horn, Staten Island and the 
Falklands. With such a system of observations it would be possible tu determine 
exactly the track of every cyclone crossing the polygon of stations. This is a 
matter of very great practical importance. These cyclones seem to travel in the 
general direction of the upper winds from west to east, and to follow the outline of 
Alexander, Graham, and Palmer Lands, but how and why this is so we cannot tell 
as yet. Between South America and the Antarctic land there isa belt of low 
pressure which seems to encircle the Antarctic region where there is apparently a 
permanent anticyclone; but observations are wanting to determine the associated 
conditions of atmospheric circulation. 
It seems scarcely necessary to insist on the advantages which two other 
polygons of stations would offer, one to the south of the Indian Ocean, the other 
between New Zealand and Victoria Land, The second polygon would be formed 
by the islands of Prince Edward, Crozet, Kerguelen, and a station on Enderby 
Land. The third polygon would include the Balleny, Macquarie, and Auckland 
Islands. This would be a particulsrly interesting polygon on account of its 
comparative proximity to the magnetic pole. 
The two vessels designed to winter in the pack should approach along the 
meridians of 145° W. and 35° E. Imprisoned in the pack, as the Belgica was, 
these vessels would be able to carry on oceanographical and zoological work, and 
also to collect magnetic and meteorological observations, thus adding two stations 
near the pole to the various polygons. From the meteorological point of view it 
would be extremely interesting for these vessels to reach high latitudes, for the 
region near the pole will probably differ greatly from the northern edge of the 
Antarctic lands in everything regarding atmospheric pressure, wind, and storms, 
As to the circumpolar expeditions I think that the vessel intended for this 
purpose should be quite independent of those which penetrate the pack. The 
region is too great to admit of the whole voyage being completed in one season— 
three would probably be necessary. It is not easy to indicate the route which 
should be followed, for everything depends on circumstances. Still it may be 
observed that—in summer at least—easterly winds predominate near the edge of 
the south polar pack, and therefore it would be advantageous to proceed from east 
to west. 
Leaving the River Plate in September the vessel might commence work at the 
South Shetlands at the beginning of October. The months from November to 
March might be occupied in the voyage from 60° W. to 150° W. along the pack, 
and thence a passage might be made to Melbourne. The following summer the 
extreme south of the Indian Ocean might occupy the vessel, and a third season 
might be devoted to the Antarctic Atlantic. 
This programme is doubtless but a dream. I often wished, on board the 
Belgica, that:I dared to propose it as a programme, because it seemed to me 
perfectly realisable. One may perhaps speak of it at the dawn of a new century, 
3. Notes on the Physical and Chemical Work of an Antarctic Expedition. 
By J. Y. Bucuanayn, F.R.S. 
In an Antarctic Expedition, the physical and chemical work to be done falls 
into two classes, according as it has to be done at sea or on land. 
The principal object at the outset of the expedition should be to push energeti- 
cally southwards, and effect a landing in the most suitable place in the highest 
possible southern latitude, and there establish the principal station. The locality 
should be chosen where the ship, or one of the ships, would find safe winter 
quarters. 
As the principal object is to establish the expedition as advantageously as 
possible on land, no time should be‘spent unnecessarily at sea, For this reason 
magnetic observations at sea should not be contemplated. They take up an 
