1898-1902. No. 4.] THE TEMPERATURE OF THE AIR. 61 



screen was standing on the ice near the Fram; but afterwards it stood 

 on board on the bowsprit just ahead of the forestay, at a height of 

 5.5 metres above the sea or ice. 



The Tables give the corrected temperature for every alternate hour 

 during the stay of the Fram in her winter quarters. In the few cases 

 in which the observations have been omitted, the temperatures inserted 

 in the Tables have been taken from the thermograph curves, or, when 

 such are wanting, computed by interpolation. 



The Tables give, for each day, the mean of the 12 observed tem- 

 peratures, the minimum and the maximum, and their difference. The 

 spirit minimum thermometer was read every day at 8 a. m. and at 

 8 p. m., and the observer noted the reading of the index and of the 

 top of the spirit column. The comparison of the latter with the reading 

 of the dry thermometer gives the error of the minimum thermometer, 

 or the correction to be applied to the reading of the index. The cor- 

 rected index reading can in most cases be taken as the minimum tem- 

 perature of the day. But as the day is reckoned from midnight to 

 midnight and not from 8 a. m. or 8 p. m. it sometimes happens that 

 the minimum registered does not fall within the adopted limits for the 

 day. In some cases the minimum temperature of the day occurs when 

 the temperature is falling just at midnight, and the corresponding read- 

 ing of the dry thermometer has then been taken as the minimum. In 

 some cases the curves of the thermograph have been consulted, but it 

 has been impossible to take the extreme temperatures from them, as 

 they have no marks for the exact time every day; and there are more 

 or less lengthy periods in which the thermograph did not work. In 

 the few cases in which the extreme-thermometers were out of order, 

 the lowest and highest temperatures given in the tables have been taken 

 directly from the corresponding readings of the dry thermometer. 



The maximum thermometer was a mercury thermometer. Its error, 

 as far as I am avvare, was not determined by direct comparison, but 

 it seems to have been rather small, judging from its indications and 

 the daily march of the temperature. When the temperature was below 

 — 35° C, it could not be used. In such cases, and in the few other 

 cases in which it was not read, the highest reading of the dry thermo- 

 meter has been taken as the maximum temperature. 



The series of daily extreme temperatures given in the Tables is 

 thus not quite homogeneous, but they will give a fairly good represen- 

 tation of the aperiodic range of the temperature of the air in the space 

 of one day. 



