An Austrian polar explorer, Karl Weyprecht, was the first to advance a definite scheme for investigating the polar regions on an international level. Weyprecht's idea was that each interested government should establish one or more stations in the polar regions, and that scientific work should be done simultaneously at all stations according to a previously co-ordinated plan. Weyprecht's plan was discussed by an international conference which met at Hamburg in 1879. The delegates at this conference formed themselves into a permanent International Polar Commission whose task was to make further and more detailed plans. In 1880 a Second International Polar Conference met at Berne, and a Third met at St Petersburg in 1881. As a result of the work done by these conferences the First International Polar Year was organised in 1882–83. Eleven countries—Austria, Denmark, Finland, France, Germany, Great Britain, Holland, Norway, Russia, Sweden and U.S.A.—set up and manned for a year twelve stations in the Arctic and two in the Antarctic. The field work completed, the Fourth* and Fifth5 International Polar Conferences met in Vienna in 1884 and Munich in 1891, and arranged publication of the scientific results, which filled 27 volumes. At the Fifth Conference the International Polar Commission was dissolved, its work being completed.

The conditions of surface and weather in which sledge journeys are carried out, together with distances travelled and altitudes attained, are of great interest to anyone attempting to analyse and compare performances by various travellers.

To obtain the necessary data the investigator must usually turn to the narrative of the journey concerned and extract the data himself, setting them out in some suitable form. Gordon Hayes adopts this procedure with many journeys in his Antarctica (London, 1928), and John Rymill gives an appendix, analysing the main journeys of the British Graham Land Expedition, 1934–37, in Southern Lights (London, 1938). Many other writers have made similar tables, but essential data always appear to be lacking, and there is no standard form to which such analyses can be reduced.