THE HEIGHT OF THE ANTARCTIC CONTINENT, 



295 



Meinai'dus's work * may be summarised as follows :— 



The barometric pressure at any place is a direct measurement of the mass of air lyin^ 

 over that place. If therefore we integrate the pressure over the whole surface of the ulobe 

 (actual pressure, not pressure reduced to sea-level) we obtain the total mass of the earth's 

 atmosphere. This mass must be constant, hence the integral of the pressure over the whole 

 world must be the same at all seasons of the year. The pressure conditions are sufficiently 

 well known over the globe outside the Antarctic Circle to make it possible to effect this 

 integration, and Meinardus has made the necessary calculations for the two months January 

 and July from charts of the normal pressure for these two months. The result of the 

 calculations shows that outside the Antarctic Circle the total mass of air is greater in July 

 than in January. Thus in order to have the total mass of air the same all over the world 

 in these two months there must be less air over the Antarctic in July than in January. 

 In other words the integral of the pressure over the area within the. Antarctic Circle must be 

 greater in January than in July and the mean difference in pressure January-July must be 

 U mm. This seems contrary to what one would expect, for in July the temperature of the 

 air is much lower than in January, hence we should expect the air to be denser and there- 

 fore the pressure higher in July than in January. 



Meinardus shows that this apparent paradox can be explained by assuming that the land 

 surface within the Antarctic Circle is at a relatively great height above sea-level. To do 

 this he first discusses the mean pressure and temperature over the Antarctic in the months 

 of January and July respectively, using all the data available in 1909. This leads him to 

 believe that the mean pressure over the Antarctic at sea-level is the same in these two 

 months and equal to 745 mm. The mean temperature which he deduces are January -S^C, 

 July -26°C. He assumes further that the temperature decreases by 5''C. for each 1,000 

 metres of ascent during both months. 



Meinardus's reasoning then proceeds as follows : — • 



Let A, figure 88, represent a column of air over the Antarctic Continent in January. 

 The pressure at the base (sea-level) is then 745 mm. and the temiJerature -3°C. The 



a. 



4 = /"-/s 



A 

 January 



B 

 July 



Fig. 88. Diagram of vertical pressure, aumiuer and winter. 



* Meinardus, ' Die mutmassliclie mittle Hohe des antarktiachen Kontinents, ' Petormanns Mitteil, 1909 Heft 

 11 and 12. 



