CONCLUSIONS. 245 



The weather characteristics which Meinardus found to accompany high easterly winds 

 at the Gauss Station are as easily explained by the travelling wave theory as they are on 

 the cyclonie theory. All these characteristics were found at McMurdo Sound to accompany 

 blizzards and therefore may be expected to occur during the high winds at the Gauss Station. 

 In fact they should be more pronounced at the latter station for the high winds there were 

 from slightly north of east and so indicate an indraft of air from the open ocean which 

 would on this account be warm and humid. 



It is true that there are considerable difficulties in explaining all the relationships between 

 wind and pressure at the Gauss Station on the wave theory, but there are none so great 

 as the cyclone theory encounters in the large pressure changes with a nearly constant wind 

 direction. Before the problem can be finally solved we must hava simultaneous observations 

 from neighbouring places as was the case in the Ross Sea area. It is possible that the 

 results of the Australian Expedition, which recently wintered near to the Gauss Station, will 

 help in this desired direction. 



The data for Kerguelen and Snow Hill have also been examined to see if the conditions 

 there are governed by pressure waves or cyclones. 



The observations at Snow Hill show large pressure changes with little or no change in 

 wind direction, but there are many cases in which the wind changes as the barometer rises 

 and falls. There can be little doubt that cyclones pass over this station, but it is equally 

 true that all the pressure changes cannot be ascribed to them. 



At Kerguelen there is a much more intimate relationship between the changes of wind 

 direction and changes of pressure than at any of the stations on or near the Antarctic 

 Continent. But even at this station, not infrequently, there are large barometer changes 

 without appreciable changes of wind direction, which may or may not be due to similar 

 pressure waves to those recognised over the Antarctic Continent. 



Turning now to the two top curves on the plates in Volume II giving the pressure and 

 winds at Melbourne in Australia and The Bluff in New Zealand we see a most close relation- 

 ship between the winds and the pressure changes. This is most clearly seen in the case of 

 Melbourne because on that curve the wind observations are more complete. The relationship 

 is extremely close : the falling barometer is accompanied by winds from some northerly direc- 

 tion, at the instant the barometer ceases to fall and commences to rise the wnd changes 

 and then blows from somj southerly direction while the barometer is rising. We know from 

 the daily weather charts of Australia that the weather at Melbourne and The Bluff is entirely 

 dominated by travelling cyclones and anticyclones. 



Conclusions. 



Cape Evans and Framheim. — At these stations the evidence is almost conclusive that the 

 barometer changes are not due to travelling cyclones and anticyclones but to real waves of 

 pressure which travel outwards from the Antarctic Continent. 



Gauss Station. — According to Meinardus the barometer changes at this station are due 

 entirely to travelling cyclones. Evidence has been brought forward to show that this is 

 impossible and reasons have been given for believing that at this station also pressure waves 

 are mainly responsible for the barometer changes. 



Cape Adare. — The evidence indicates very clearly that at Cape Adare the main barometer 

 changes are due to pressure waves while secondary changes are occasionally due to travelliug 

 cyclones and anticyclones. 



