62 



TEMPERATURE. 



Table 26. 



This table has been computed by taking the mean daily temperature amplitude on days 

 with cloud and 10 cloud (in June and July there were no clear days, so days less than 

 half overcast were used), and on days with mean vnnd velocitj'' greater and less than 4'5 

 miles per second. Tha values entered in the columns headed Cloud and Wind are the mean 

 values of these factors during the days consideied. 



Now it will be noticed that with one exception in spite of the increase of the amplitude 

 of the solar radiation from April to July the daily amplitude of temperature decreases in the 

 same period in every type of weather. 



This gives good reason to believe that if there had been no change in the type of 

 weather the temperature amplitude would still have decreased. Therefore we cannot ascribe the 

 decrease in temperature amplitude to changes in wind and cloud. 



Further tbe fact that the temperature amplitude was greater during April and May with 

 completely overcast skies than it was during June and July with sides less than half overcast, 

 gives further support to the conclusion that cloud was not the chief cause of the decrease 

 in the amplitude. On the other hand as the wind decreased every month from May to 

 August (see table 25), wind cannot have been the cause. 



Professor Meinardus's Explanation of the ' Fram' Type of Daily Temperal.iire Variation. — In 

 his discussion of the meteorological results of the Gauss Expedition, Meinardus says : — 



' The amplitude of the daily period of temperature is greatest during the spring months 



and in March ;* in the summer it is considerably less, probably because the midday 



temperature during this time is approximately that of the freezing point which 



cannot be passed on account of the heat during the process of melting. Also during 



the summer the effect of radiation on the air temperature is reduced owing to the 



shortening of the night and lengthening of the period of sunshine. Similar changes 



of the temperatm-e amplitude were found during the drift of the Belgica and the 



drift of the Fram. Also there the regular variations were greatest in the spring. 



On the contrary the greatest values occurred in January at the Discovery station 



where the freezing point was seldom reached.' 



Two probable explanations are given in this paragraph, neither of which was considered 



by Mohn : [a) a curtailing of the amplitude in the warmer months due to the temperature of 



the surface being unable to rise above the freezing point, and (6) a decrease in the amplitude 



due to the shortening of the night and the lengthening of the period of sunshine. 



The latter of these two is obviously based on a misconception ; the daily amplitude of 

 temperature does not depend on the amount of energy received from the sun, but on the 

 difierence between the amount of energy received at midday and at midnight. In any one 

 latitude the difference of energy received on a square centimetre of ground at midday and 

 midnight increases right up to the summer solstice. There is therefore no decrease in the 



* See (ootuote to page !iu. 



