[ xii ] 



Antarctic Manual' bear out this statement, and the idea was also put forward by Sir J. MURRAY and 

 Dr. BITHAX, and was referred to by Mr. BERNACCUI in his account of the observations at Cape Adare. 

 I gave very precise expression to the idea in the 'Proceedings of the Royal Society,' vol. 74, p. 20, 1904, 

 and showed that an east to west circulation at the surface, with a superposed west to east circulation in 

 the upper air was to be expected. 



Such were the conditions actually found by the " Discovery " at Winter Quarters, as well as by the 

 " Gauss " in a far distant part of the Antarctic. Mr. CURTIS has set out very clearly the full data on the 

 question, and it seems difficult not to regard the observations as confirming the inference from general 

 conditions, about which there can be no doubt. But Captain SCOTT and other members of the Expedition 

 brought back the general impression that the winds at Winter Quarters were a local phenomenon due to 

 local configuration (see p. 422), and that as soon as an "open" situation was reached the prevailing wind 

 was really from south-west. South-westerly winds were certainly noted at Cape Crozier on several 

 occasions, though some of the Cape Crozier observations were at an altitude of nearly 1000 feet; and 

 southerly winds helped to bring back Captain SCOTT from farthest south, across the barrier ice. In spite 

 of these facts it is difficult to imagine any cause for the existence of an easterly wind as a " local wind " at 

 Winter Quarters or Cape Armitage and the practical absence of observations of westerly or south-westerly 

 winds at these points. In order to place those who were discussing these questions in as favourable a 

 position as possible for realising the local conditions, I had a model in clay of Ross Island and the region 

 round the Winter Quarters, prepared in the office by Miss HUMPHREYS, who used all the indications that 

 could be collected from maps and photographs to give a true representation of the locality. The model 

 was exhibited at the Royal Society in 1907, and was seen by various members of the Expedition, who 

 used it to recall their experience in considering points put to them. It was approved as an accurate 

 representation. Two photographs of it are reproduced here, to show the general orographical conditions 

 in regard to the point in question. 



It is curious that endeavours to reach, by two separate crucial tests, a definite conclusion upon this 

 interesting point, as to whether the easterly wind at Winter Quarters is a local wind or a true general 

 wind implying a high pressure to the south, fail through very slight omissions in the observations or the 

 records. 



The first test is the experience of Lieutenant ROYDS on his journey across the ice barrier, November 10 

 to December 10, 1903 (p. 352), which should have been very favourable for extending the reach of the 

 observers beyond the open position of Cape Armitage. Mr. ROYDS' observations are nearly all printed 

 S.\V., but it is curious that the deviation of that direction from the easterly direction observed at Winter 

 Quarters and Cape Armitage is actually the deviation of the compass in that locality. There is no entry 

 in the registers to show how the direction was obtained, nor whether the original entries were magnetic or 

 true. Cross-examination of the members of the Expedition who visited the Office led to varying opinions, 

 but finally Mr. ROYDS wrote (2nd May, 1907), " I believe on some other occasions wind has been entered 

 as magnetic, but I do not remember doing it on that journey ; in fact I am sure I did not." They are 

 therefore printed as they stand. 



It is clear that when the magnetic variation is so great as to make the difference between a south- 

 westerly and an easterly direction of the wind, special care should be taken to distinguish between the 

 two specifications of direction. Indeed, anyone who has to deal with the uncertainties arising in a 

 question of this kind will agree that our geographical knowledge has now become too extensive for the 

 continued use of the symbol N. to mean true or magnetic north according to circumstances, and that 

 some other name for magnetic north ought to be devised when the point intended to be defined may be 

 within twenty-five degrees of due south. 



The second crucial test consists in endeavouring to identify the increase of pressure polewards that must 

 correspond with a normal wind from the east in south-polar regions. This also must be considered to have 

 failed, for Captain HEPWORTH and Mr. CURTIS have both attempted to use it and have felt themselves 

 justified in adopting different conclusions. The failure arises from the fact that the explorers brought 

 back no certain information about the amount of slope of the barrier surface towards the sea. The fact 

 that there is a definite " ice-fall " between White Island and Black Island leads one to infer that, in that 



