THEORETICAL DISCUSSION. 263 



Barrier (see page 136). Tlie starting point was therefore a series of parallel lines over the 

 Ross Sea running approximately parallel to the Barrier edge. As the lowest pressure is doubt- 

 less over the coldest region, it has been shown over the low-lying area on the Pacific side 

 of the Pole. Round this area the isobars already drawn over the Ross Sea must pass. Leav- 

 ing for the moment the ends of the isobars over the Ross Sea, the other ends were continued 

 round the low pressure area imtil they came to 40° W. longitude. Returning then to the ends 

 over the Ross Sea, near the Western Mountains, it was realised that the pressure near 

 the edge of the plateau must be in equilibrium with the pressure at the same height 

 over the sea. The lines were therefore continued on to the plateau. But according to 

 our assumption the pressure of the plateau increases inland, this necessitated that the lines 

 should bend away from the centre and pass around the edges of the plateau. 



A few trials showed that these conditions were best met by carrying isobars Nos. 2 and 3 

 to the south and isobar No. i to the north. 



Isobars Nos. 2 and 3 after turning to the south run nearly parallel with the edge of 

 the plateau and join up with their other ends, completing the closed cyclonic system over 

 the low-lying land. 



A high pressure system was then indicated over the plateau by means of the closed 

 isobars 4 and 5. Round these two high and low pressure .systems encircling isobars were 

 drawn indicatmg that the pressure increases northwards over the whole region outside the 

 Antarctic. 



If this pressure distribution is correct, it will be in accordance with the isobars shown 

 in the vertical section of figure 80. The line along which this section has been made is 

 indicated in figure 82 by the thick line AB. We will pass along this line from A to B 

 and compare the pressure changes shown on it with the corresponding changes shown at 

 3,000 metres on figure 80. Starting on the left of each diagram we see that from 50° S. 

 to about latitude 64 S. the pressure lines at 3,000 metres fall in figure 80, and we cross 

 isobars 4 and 3 to a shallow trough of low pressure in figure 82. From 64° S. to 80° S. 

 the pressure lines at 3,000 metres rise on figure 80 and on figm-e 82 we cross isobars 3, 4 

 and 5 to the centre of the high pressure area over the plateau. From 80° S. to 75° S. 

 on the other side of the Pole the isobars fall on figure 80 and we pass across isobars 

 5, 4, 3, 2 and 1 to an area of low pressure on figure 82. From 75° S. to the right hand 

 edge of figure 80 the pressure lines at 3,000 metres rise, indicating increasing pressure, while 

 in the same distance in figure 82 we cross isobars 1, 2, 3, 4 and 5 in the direction of 

 increasing pressure. Thus the horizontal pressure distribution shown in figure 82 agrees with 

 the vertical pressure distribution represented in figure 80. 



We have now, from considerations of pressure and temperature alone, drawn a system 

 of isobars which enclose a high pressiu-e area over the plateau and a low pressui-e area over 

 the part of the Antarctic which is supposed to be at or near sea-level. The cnicial test 

 of this pressure distribution is whether the observations of wind direction agree with the 

 general rmi of the isobars. 



The pressure distribution at 3,000 metres has been chosen because this is the height 

 in the atmo.sphere from which we have the most information about the air motion. 



(a) Erebus smoke gives the air motion over the Ross Sea area at a height somewhat 

 greater than 3,000 metres. 



(6) The height of the plateau varies between 2,000 and 3,000 meters, hence wind obser- 

 vations made on its sm-face give useful information. 



