PHYSICAL OCEANOGRAPHY 



303 



from the mouth of the Sognefjord westwards, in the middle of 

 May, every year from 1901 to 1905. One of these series is 

 figured on p. 240. Nansen and the writer have calculated the 

 mean temperatures in the Atlantic water of this section, both 

 for the surface and for the deeper water. The variations in the 

 surface-temperature are represented in curve L, Fig. 208, curve 

 H. showing the variations in the growth of the pine in eastern 

 Norway during the following year. The low surface-temperature 

 in May 1902 corresponded to the small growth of the pine in 

 the succeeding year, 1903, and the high temperatures in the 

 surface of the Gulf Stream in May 1905 corresponded to a 

 great addition to the height of the pine trees in the year 1906. 

 This is explicable by the fact that the annual growth of the pine 

 is not determined by the meteorological conditions of the same 

 year, but by those of the year 

 before, when the bud was 

 formed, the growth mainly 

 depending on the formation 

 of the bud. Continued inves- 

 tigations will prove whether 

 the agreement strongly sug- 

 gested by the figure is really 

 a general rule, in which case 

 it may be possible, on the 

 basis of investigations in the 

 Norwegian Sea, to predict 

 with a high degree of probability how much the Norwegian pine 

 will grow in the following year. 



By calculating the mean temperature of the Atlantic water- 

 masses below the surface in the Sognefjord section, and 

 multiplying the ascertained value by the area of the transverse 

 section of these water-masses, an expression is obtained for the 

 amount of heat in the northern branch of the " Gulf Stream." 

 This has been done from the observations made during the 

 May cruises, and the results are exhibited in curves I. and II. in 

 Fig. 209 ; the two curves are obtained by two different methods 

 of calculation which need not be discussed here. The lower 

 curve shows the variations in the mean temperature of the air 

 in Norway during the winter months from the ist November to 

 the 30th April. The coincidence is striking ; when, for instance, 

 the amount of heat in the Gulf Stream was great in the 

 month of May, the air-temperature in Norway was high in the 

 following winter. This holds good throughout six years, 



Fig. 208. 

 mean temperature of the surface of the ' ' Guh 

 Stream " in the Norwegian Sea (Sognefjord 

 section, May) ; II., mean growth of the pine 

 in eastern Norway. 



