38 TEMPERATURE. 



Turning now to figure 11 we again see the slow fall of temperature over the open 

 sea and a rapid fall as the pack is reached. The rate of decrease of temperature falls 

 off over the sea ice and when the heavy permanent ice is reached in latitude 70° S. the 

 decrease of temperature as we proceed towards the south becomes slow and finally reaches 

 the same rate as over the Barrier. 



It must be borne in mind that a great deal of what has just been written is based 

 only on conjecture ; it would no doubt have been more accurate to have used the word 

 ' probably ' before many of the statements, but for clearness that has not been done. As 

 further actual observations are not likely to be obtained in the near future the above des- 

 cription of the probable temperature conditions in the Ross Sea area must stand as best 

 representing our present knowledge. 



Plateau Temperatures. 



Captain Amundsen reached the South Polar Plateau on 21st November, 1911, and left 

 it on January 5, 1912, while Captain Scott reached it on 22nd December, 1911, and left 

 it on February 6, 1912. Thus observations were made on the plateau continuously from 

 21st November, 1911 to February 6, 1912. As the plateau is not level the temperature 

 varies on account of changes in height, hence it is necessary to reduce all the temperatures 

 measured to a constant height before they can be compared. It has become an international 

 convention to reduce the temperature to sea-level by applying a constant temperature cor- 

 rection of +-5°C. per 100 metres of ascent (■274°F. per 100 feet).* In his discussion of 

 the observations made on the plateau by Amundsen. Professor Mohn did not follow this 

 convention, but from the observations concluded that the actual temperature gradient was 

 •53° C. per 100 metres. He therefore reduced the temperature to sea-level by applying 

 this correction. For the reasons given by Hann it is more satisfactory to retain in all parts 

 of the world the same value of the temperature gradient and this has been done in the 

 following discussion 



In reducing the plateau temperatures to sea-level it is necessary to know the height 

 at which each observation was made. Mohn has calculated the height of each place where 

 Amundsen's temperatures were measured and the same has been done for Scott's journey 

 (see Chapter IX). Unfortunately the two determinations of the height of the South Pole do 

 not agree, Mohn making it 2,454 metres while I make it 2,765 metres. As the observations 

 must be reduced to the same standard it is necessary to take one or other of the above 

 values. I have therefore increased all the plateau heights given by Mohn by 311 metres 

 in order to make the height of the only common station agree in the two sets of observa- 

 tions. Amundsen's temperature observations were taken at irregular times, on some days only 

 two observations were made and on others three, four or six. I have taken as the mean 

 temperature of the day the mean of two observations taken as far as possible at the same 

 hour in the morning and evening. Thus on December 1st observations were made at 8-30 a.m., 

 1 P.M. and 5-30 p.m., the mean of the observations at 8-30 a.m. and 5-30 p.m. after 

 reduction to sea-level has been taken as the mean temperature of the day. 



Scott's observations have been treated somewhat differently. All the temperature observa- 

 tions were plotted and a curve drawn through the points as already described for the 

 Barrier. This curve was measured at four-hourly intervals and the mean temperature for 

 the 24 hours calculated for each day. The height at the lunch camp was taken as the height 

 for the dav. 



* Sec Hanu Lehrbiich d. Mctcorologie, 3i(l edition, page 138. 



