:\IOTION OF THE UPPER ATMOSPHERE. 135 



within 100 metres of its starting point. The instnmient record showed that it had risen 

 4,000 metres, hence up to this height the atmosphere was absolutely calm (see page 278). 



The directions of motion in the different layeis over Cape Evans are best discussed with 

 the aid of figure 47. In this figure the diagrams down the left liand side, Nos. i. 4, 7 

 and 10, represent the percentage freriuence with vvhich air motion from each direction was 

 recorded from the high clouds down to the ground, these diagi-ams will first be considered. 



Air motion at the grotnid.— -The lowest diagram. No. 10, shows the winds. It has already 

 been explained that the wind motion at Cape Evans is limited by the Western Mountains 

 on the west and Monnt Erebus on the east, so that there is practically no motion except 

 through McMurdo Sound : from the Barrier to the Ro.ss Sea, represented bv the winds between 

 south and east, and motion from the Sea to the Barrier, represented by winds between north- 

 west and north. 



Air motion, at the height of the medium clouds, 10,000 to 1.3,000 feet. The diagram for 



the medium clouds 7, shows that the motion at this height is also mainly limited to two 

 directions more or less parallel to tho.se of the wind at the surface, which might be expected 

 from the statement made above that the medium clouds were mainly at a lower level than 

 the top of Alonnt Erebus. A comparison of the diagram for the surface winds with the one 

 for the medium clouds shows an important difference, namely that the motion from northerly 

 direction is much more frequent at the level of the medium clouds than at the snrface 

 Even when periods only are considered during which the snrface winds are over 20 miles 

 an hour from the south the medium clouds move nearly a.« frequently from northerly as from 

 southerly direction {see figure 47(8)). 



Air motion at the height of Erebus smoke, 13,000 to 16,000 feel.—The sunnnit of Erebus 

 is above all the surrounding land masses and the direction of its snK)ke gives the first clue 

 to the unimpeded air motion over the Ross Sea area. It will be seen that the diagram for 

 the Erebus smoke, No. 4, differs from that for the medium clouds only in one particular^ namely 

 in the addition of considerable motion from the west and south-west; there is still little or 

 no motion from the east and north-east, while the lelative amounts fiom the Jiorth-west 

 and south-east are practically the same. 



The mean direction of the motion of the atmosphere at the height of Erebus is from 

 S. 82° W. which is slightly to the south of we.st. It will be shown later that the frequence 

 of south-west motion of Erebus smoke has probably been somewhat under-estimated at Cape 

 Evans, if this possible error were corrected it would bring the mean direction slightly further 

 south, at the same time it will be shown that the mean direction of Erebus smoke observed 

 by the Discovery Expedition, S. 56° W., was probably too nnich to the south. The true 

 direction is therefore probably between S. 82° W. and S. 56° W. and so mav be taken as 

 S. 70° W. or W.S.W. 



The direction of the motion of the air shown by Erebus smoke is in marked contrast 

 to the direction of the wind at ground level. The surface winds are almost entirely from an 

 easterly direction while those at the height of Erebus smoke are nearly as exclusively from 

 a westerly direction. This means that between the ground and about 15,000 feet the pressure 

 gradient is reversed, the low jiressure being over the Barrier and the high over the Ross Sea. 

 There can be little doubt that this is a direct consequence of the large difference of temijera- 

 ture between the Barrier and the Ross Sea. Owing to the higher temperature over the Ross 

 Sea than the Barrier the decrease in pressure with height is less over the former area than 

 the latter, thus at a certain height the low pressure over the Ross Sea is compensated 

 for by the less dense air colunm and the pressure difference between the Barrier and the Sea 



