28 



H. MOHN. 



M.-N. KI. 



Baric Wind-Roses. 



The numbers in this Table and the following Wind-Roses are all weigh- 

 ted and smoothed means, computed from the formula 



Pa+^P+Pc 



M being the weighted and smoothed value for the respective wind-direction, 

 b the mean value for that direction, a and c the values for the adjacent 

 directions, p^, p^ and p^ the respective weights or number of observations. 

 The last row contains only weighted means of the numbers in the respec- 

 tive column (wind direction), and not smoothed. 



The mean pressure found for a single wind is the mean of values 

 differing often considerably from that mean. Any wind may accompany 

 both high and low pressure. 



The baric wind-rose shows maxima with E, SW and NW, and mi- 

 nima with N, S and W. There is no regular march round the compass. 



The Temperature ofthe Air. 

 As the Thermograms were unsuitable for" the computation of the 

 daily and monthly mean temperatures, I have taken the means of the ob- 

 servations at 8 a. m and 8 p. m. as the mean temperature of the months. 



April May June July Aug. Sept. Oct. Nov. Dec. Jan. 



0000000000 

 Mean — 27.6 — 35.4 — 34.4 _ 36.5 - 44.8 - 37.5 — 24.2 — 15.5 — 6.7 — 9.7 



20.0 — 10.6 — 12.5 — 24.5 — 9.5 — 9.0 — 5.2 - 0.2 — 2.7 

 50.6 — 58.2 — 54.5 — 59.0 — 53.5 — 40.2 — 28.0 — 16.8 — 17.2 

 30.6 — 47.6 — 42.0 — 34.5 — 44.0 —31.2 - 22.8 — 16.6 — 14.5 



Prof. Julius v. Hann has reduced the Framheim temperatures to the 

 5-years' means of temperature from MacMurdo Sound ^ 



1 Met. Z. 1914, p. 553. 



