492 



mean of the corresponding corrected readings of the barometer taken on board the ship. Tho differences 

 between these two sets of mean values being then plotted to the corresponding mean distances were found 

 to exhibit the fairly regular curve shown in fig. 1, the difference increasing from - 0'03 inch at 10 miles on 

 the outward journey to - 29 inch at the extreme point reached, and then decreasing again to - 09 inch 

 at 10 miles from the vessel on the return. The correction required to make the aneroid when read at the 

 vessel agree with the ship's barometer, corrected for temperature and sea level, was applied throughout. 



By meaning the observations in this way the casual fluctuations of pressure were to a great extent 

 smoothed out, and probably sufficiently so to warrant us in regarding the mean for each distance-interval 

 as the mean pressure for the time occupied by the party in traversing it. The differences, therefore, 

 between those means and the means at the ship are due (1) to the difference in level between the two 

 places, and (2) to the barometric gradient which may have existed at the time. 



As regards (1), the distant barometer being always lower than that at the ship indicates that in so far as 

 that fact represents change of level the travellers had been rising ; whilst as regards (2), except so far as it is 

 modified by (1), the differences point to the presence of an area of relatively lorn pressure in the region 

 traversed. That the differences were chiefly due to change of level is, however, made clear by the fact 

 that throughout the journey the winds were, with only a few brief exceptions, very light, with a large 

 percentage of calms, and this would certainly not have been the case had the difference been to any 

 marked extent due to barometric gradient. There was also evidence of very considerable ice pressure, 

 and of a northerly movement of the Barrier ice-sheet, as to which Captain SCOTT remarks* : " it is certain 

 that there must be a supply of ice from some region to the south," which must necessarily have caused 

 some gradient, if only a small one. 



Some rise in level may therefore be safely assumed, and the question then arises whether its full extent 

 is indicated by the differences shown in the diagram, or whether these should not be increased (or 

 decreased) by an amount representing the barometric gradient. 



As they stand, assuming there was no barometric gradient, the differences indicate a change of level in the 

 ice of little more than 1 foot per mile, a gradient which could hardly be reduced in view of the pressure 

 phenomena which were observed upon the ice-sheet and of the remark of Captain SCOTT just quoted ; and 

 therefore the existence of an area of low pressure to the southward of Ross Island which would result from 

 a reduction of the ice gradient appears to be very improbable. 



On the other hand, if we may assume the existence of a relatively high pressure in the region traversed 

 by the party, with a sea-level gradient of about O'l inch for the distance of 170 miles represented by the 

 diagram, the corrected (augmented) differences would then give us a rise in level of the ice of about 2 feet 

 per mile along the line traversed, and although this gradient is still very slight it accords better with the 

 ice phenomena observed than does the smaller one, whilst at the same time we get a barometric gradient 

 sufficient to account fully, both as regards direction and force, for the wind distribution observed at the 

 " Discovery's " Winter Quarters at Ross Island. 



RESULTS AT OTHER STATIONS IN THE ANTARCTIC. 



Before leaving this part of the subject it may be well to refer to the results obtained in 1899 by the 

 " Southern Cross" Expedition at their winter station at Cape Adare in Lat. 71 S., Long. 170 E., rather 

 more than 400 miles north of the " Discovery's " position. 



The exposure for wind observations was probably less satisfactory there than it was at Discovery Bay, 

 as a range of high mountains rose abruptly to the south and west of the station ; but to the eastward and 

 round through north the exposure was open to the sea. It was from the south-east, however, that most of the 

 winds blew, and although the distribution was not so exclusively confined to the easterly quarters as was the 

 case at Discovery Bay, yet, as will be seen from the following Table VI., it was such as to lend considerable 

 support to the suggestion that the easterly air current which the " Discovery " experienced at Ross Island 

 extends, as the predominating current near the sea level, over a large extent of the surrounding region. 

 Calms were of much more frequent occurrence at Cape Adare than at Discovery Bay. 



* ' Voyage of the " Discovery," ' vol. ii., p. 428. 



