MEAN TEMPERATURE OF THE ANTARCTIC. 



93 



being so much under the influence of the sea. There can be little doubt that the 

 temperature of all places on latitude 78" S. at any considerable distance from open water 

 have temperatures nearer that of Framheim than that of McMurdo Sound, in fact the tem- 

 perature of Framheim is probably higher than that of places on the same latitude but further 

 from the sea. So far as we know, the only other part of the Antarctic at which the sea 

 is ever open so far south a.s 78" S. is in the Weddel Sea, where Filchner penetrated in 1912 

 to 77° 4-)' S. The rest of the 78" S. parallel, in all probability, passes over ice as thick 

 as that of the Barrier. It is thus reasonable to conclude that the average temperature of 

 all places near sea level along latitude 78° S. is near to that found for Framheim and the 

 Barrier, and therefore must be below the value given by Meinardus. .4s all the temperature 

 data used by Meinardus have been obtained at stations either on the coast or over the 

 sea itself, they are open to the same objection when used to give the average temperature 

 over the Antarctic Continent. Cape Adare is surrounded on three sides by open water during 

 seven or eight months of the year, and during the remainder open water is never far away. 

 It is, therefore, almost certain that Cape Adare has a higher temperature than large tracts at 

 the same latitude which are far away from the sea, and have a snow surface similar to that 

 of the Barrier. But the temperature at Cape Adare has been given by Meinardus as the 

 mean of the latitude on which Cape Adare is situated. Thus we see that at latitudes 71° S. 

 and 78° S. Meinardus's mean temperatures for the whole latitude are too high, and at the 

 latter, the evidence points to an error of about 11°F. It is, therefore, safe to conclude that 

 all Meinardus's values south of the Antarctic Continent are too high, because he has based 

 them on coast temperatures, and not on inland temperatures. Considering how little we know 

 of the inland temperatures of the .Antarctic, and of the height of the continent, it seems 

 useless to carry the discussion further. We can, however, say definitely that the temperatures 

 H the various latitudes south of the Antarctic Circle are certainly lower than those given 

 by Meinardus. 



This conclusion is of importance when we compare the mean temperatures of the Arctic 

 and Antarctic for Meinardus's values are already considerably lower than those determined by 

 Mohn for corresponding latitudes in the north. 



Table 54. 



According to the above table the temperatures in north polar regions are between 4°F. 

 and 5''F. higher than at corresponding positions in the south and as Meinardus's values are 

 too high the real difference is considerably greater. 



