EETURN OF THE ANTARCTIC EXPEDITION. 119 



doubtless throw light on the history of this portion of our 

 planet ; indications will there be found either of the per- 

 manence, or the reverse, of its glaciated condition during 

 past geological ages. On the other hand, observers on the 

 sea will be able to gain information as to its depth and the 

 vertical distribution of temperature, nature of sedimentary 

 deposits, the character and inter-relation of floatinu- ice in all 

 its varieties, the seasonal changes and movements of the ice 

 and their bearings on the temperature of the water and 

 on oceanic circulation generally throughout the whole 

 Southern Hemisphere. Much of this work will involve the 

 undertaking of voyages of circumnavigation outside the 

 areas where ice obstruction is a serious impediment. If these 

 are prosecuted simultaneously with observations taken within 

 the icy regions, of course the results will be much more 

 capable of effective co-orduiation. Passing from the domain 

 of physical geography proper, the next branch of science to 

 be mentioned is Meteorology. And here, more than in any 

 other department, is the necessity apj)arent for observations 

 continued for as long periods of time, and extending 

 to as many stations as possible. At present any view 

 we please as to meteorological conditions on the Ant- 

 arctic land may be held, because we have no facts by 

 which to regulate our speculations. Observations within 

 the Antarctic region proper should be conducted in 

 conjunction with the establishment of stations at such 

 places a;^ Cape Horn, the Falkland Islands, and the 

 South Shetlands, the Crozets and Kerguelen, Macquarie 

 Island, and the Auckland Islands. It will be seen that 

 the places mentioned, together with a few others in their 

 neighbouihood, fall naturally into three geographical groups, 

 which we may designate as the South Atlantic, the South 

 Indian Ocean, and the South Pacific. If with each group is 

 associated a station to the south of it, within the Antarctic 

 circle — say at Graham's Land, Enderby Island, and the 

 Balleny Islands respectively — there will be three observa- 

 tional areas situated at about equal intervals round the globe, 

 each of which would sufiice to determine exactly the track of 

 every cyclonic disturbance crossing it, while the three in con- 

 junction would probably suffice to render account of every 

 meteorological event of Antarctic origin that toot place during 

 the time the stations were maintained. This idea was suggested 

 at the last meeting of the British Association by M. Arctowski, 

 the meteorologist of the recent " Belgica " Expedition, who 

 brought it forward as a scheme for international co-operation, 

 and we may note that a similar scheme, due to Lieut. 

 Weyprecht, of a set of international circumpolar stations was 

 carried out some years ago in the Arctic regions with con- 

 siderable results to science. But in our hemisphere such a 



