APPENDIX IV. 



487 



small amount of precipitation in comparison with the number of days. 

 Often the amount of snow or water in the gauge was too small to be 

 measured, and not seldom it must have evaporated before the measure- 

 ment could take place. In the months where the figures for precipitation 

 are omitted, no gaugings were taken during the month. Even had a 

 complete figure for the whole year been obtained, it would be very low. 

 Of the other series of observations I have not worked out 

 summaries ; but if done they would hardly be in their place here. 



For this year also figures are wanting for the month of August 

 at winter quarters. The first week of the month, which was spent 

 there, was very warm, but after we left we soon had very low tempera- 

 tures. Still, the mean temperature might well be about that of June or 

 a little higher. The year's average would thus be 15'25 Cent., an 

 abnormally high figure, which was in great measure occasioned by the 

 unusually mild weather of December and February. During both these 

 months (and to a less degree also at other times) we had periods 

 of strong southerly winds causing the temperature to rise unusually 

 high. On February 9 and 12 positive temperature was reached several 

 times, and the mean for the first half of the month was 13'34 Cent. 

 This can only be described as abnormal ; but a rise of temperature in 

 February with a subsequent fall in the first half of March was, as 

 will be seen by the appended tables, a yearly recurring phenomenon, 

 which I cannot pretend to explain. A chance occurrence it cannot have 

 been, as it happened four years in succession. 



The absolute maximum for the year was 12'2 Cent., occurring on 

 June 27 ; the absolute minimum, 48'7 Cent , in January (17th), as 

 usual. 



