APPENDIX C: KNUDSEN 



December 



January 



February 



March 



April 



May 



June 



July ■ 



August 



September 



October 



November 



The whole year 



That the air-temperature at Thorshavn has one minimum in January and 

 another in March, is explained by the fact that the water of the Polar Stream (on 



account of the melting ice) is 

 more slowly warmed in early 

 spring than the water of the 

 Gulf Stream. We can scarcely 

 detect any double maximum in 

 thesumnier,because the maxima 

 of the two components lie so 

 close together, that no partial 

 maximum is formed between 

 them. The table shows that 

 the two maxima are not quite 

 together, in consequence of 

 which the mean-value of the 

 summer -temperature remains 

 almost unchanged for a month 

 between 107° and io-8°. 



From the figures we see 

 further, that the air-temperature 

 conditioned by the, wind coming from the Gulf Stream, has a secondary maximum 

 in February. This is scarcely characteristic for the climate, however, and is probably 

 referable to the unusual number of warm days in February 1891. If temperature- 

 observations over several years were available, this maximum would undoubtedly 

 disappear. 



ÜCC. Jan Feb. Morz 



JHai Juni Juli Jug. Jepl. 

 Fig. 21 



Mk Dec 



■ The upper dotted curve signifies: "Mean-temperature over 25 years" 

 curve: "Mean-temperature from two components'". 



the lower continuous 



