APPENDIX A; PETTERSSON 



- 14 — 



Anomaly of compensation in shape of a surplus of temperature in tlie surface layer during this and 

 ' orthe''wrter the foUowlng months until April 1903, the flow of the Atlantic current thence subsiding, 

 of the surface This hydrographic situation is still more pronounced in the temperature curves of 



Thorshavn (Fjeroes), where it appears as a retardation of the phase of the period, the 



maximum occurring a whole month 

 later than usual. 



This fact is still more obvious if 

 for the temperature variations at the 

 surface, we substitute those at 200 m. 

 depth, where the periodicity and the 

 disturbances of the Atlantic current 

 are fully developed. The surfacelayer 

 is in direct contact with the atmo- 

 sphere, and its variations of temper- 

 ature become more or less the retarded 

 reflection of changes in the terrestric 

 seasons. 



The Scottish station Sc. 8 lat. 

 éi°32' N. long. 3°io' E. is situated west 

 of the entrance to the Norwegian 

 channel. Here the Atlantic current 

 bifurcates, sending one branch polar- 

 wards through the Norwegian Sea, 

 another across the North Sea plateau 

 into the Skager Rak. In summer the 

 surface is covered by a layer of coastal 

 water up to 100 metres in thickness. 

 At 200 m., however, pure Atlantic 

 water is always found. This site, 

 therefore, is very favourable for sound- 

 ing the core of the Atlantic current. 



The tables on p. 1 5 give the tem- 

 perature and salinity at 200 m., the 

 store of heat in kilogram-calories of 

 the 200 — 300 ra. layer (calculated from 

 the freezing-point by using the ther- 

 Variations of mal coustauts of pure water), and the diff"erence in that heat store found to have arisen 

 in dM^rMayers ^^ ^^^e interval of two successive seasonal cruises, -|- denoting gain, — loss of heat. The 

 [North Sea] quantity d denotes the daily mean of this loss or gain. 



We find, that the maximum temperature and saHnity of both years (8°é7 C. and 

 35-26%o in 1902 and 8°79 C. and 35-3o''/oo in 1903) was recorded in November or 

 December for the 200 m. depth, whereas the minimum sahnity and temperature was 

 observed in May or August. This is a verification of the statement previously made, 

 that the flooding of the Atlantic current takes place at the end of the year. The max- 

 imum increase in heat occurred between August and November. November to May, 



Fig. 9. 



Average 1902 and 1903 



