877 



January Febiiiaiy March April May June July August 

 3« 2 S«^ 2 3M 5^ 5 7^ 2 9'^ 7 10'^ 8 10« 8 



September October November December. 

 9*^ 4 6« 7 5° O 3*^ 5 



The mean annual temperature is 6*^ 5 C. 

 These show, that winter and early spring (December-March) 

 have nearly the same temperature, about 3*^3— 3° 2 C. Consequently 

 there is no winter to speak of with prolonged periods of tempera- 

 tures below zero. So far as vegetation is concerned, however, the 

 mean readings for any period are of less value than the real reading, 

 because it is more especially the extreme ranges of temperature, 

 which are of importance in this connection. The lowest tempera- 

 ture in Thorshavn during 30 years is — 11° 6 C. It is furthermore 

 of interest to note, that the annual number of »ice-days« (i. e. 24- 

 hour periods during which the temperature does not exceed zero) 

 is only 8, and there are 70 »frost-days« (i. e. 24 hours during which 

 the temperature falls for some time below zero). These are com- 

 paratively small numbers. Even on »frost-days« the temperature 

 is seldom much below zero, as is proved by the faet that in Thors- 

 havn during a period of 30 years only 9 days have yielded a tem- 

 perature below — 10*^ C. 



It must also be considered whether the changes of temperature 

 occur often and suddenly, or whether the periods are long, and the 

 transitions slow and uniform. It is not easy to obtain good data 

 to answer this question, but all things considered the changes seem 

 to be rapid. An abrupt change from frost to thaw, especially in 

 spring, will invariably have an injurious effect upon the vegetation, 

 and on the Færoes one frequently hears of damage from this cause. 

 These sudden changes are confirmed by the observations on the 

 snow-covering in the winters 1901 — 02 and 1902 — 03, made by my 

 friend Dr. Knud Poulsen. These I have given in full on a later 

 page, but one part of his notes is particularly applicable here: De- 

 cember 30th and 31st, 1902, heavy fall of snow, and a fairly deep 

 snow-covering; January Ist, 1903, rain; Jan. 2nd, snow disappeared 

 in the lowlands; Jan. 4th, black-frost; Jan. oth and 6th, continuous 

 fall of snow; Jan. 7th, snow with wind; Jan. 8th, rain, the snow 

 melting; Jan. 9th, black-frost: etc. Thus within ten days the di- 

 strict has been twice covered with snow, twice cleared of snow by 

 rain, and twice in the grip of black-frost. One can easily conceive 

 that changes of this kind must have some influence on the growth 



