ATLANT. DEEP-SEA EXPED. 1910. VOL. i| PHYSICAL OCEANOGRAPHY AND METEOROLOGY 



25 



12° 13' 



400 



700 



Fig. 1. The vertical distribution of temperature at the Danish station (Da. May 23.-26.), the '-Michael Sars" station 

 (N, August 13.— 14.) and the Scottish station {Sc, August 13.— 14.) in the Faeroe-Shetland Cliannel. 



115 



Fig. 1 illustrates the vertical distribution of temperature 

 at our Stat. 115 (the curve marked N) and the Scottish 

 station (Sc.) in August, and at the Danish station (Da.) 

 in May. The average temperatures found by harmonic 

 analysis (r^ in the table, section 23) have been used for 

 the construction of the curves. 



The temperature at intermediate depths is much 

 higher at the Scottish station than at Stat. 115, so that 

 for instance the teinperature found at the latter station at 

 200 metres is found at about 300 metres at the former, 

 and 300 metres at Stat. 115 corresponds to 400 at the 

 other with regard to temperature. Similar conditions are 

 displayed by the salinities. The direct observations of 

 salinity give the following mean values: 



In the same way we find the following average va- 



lues of <;,: 



The difference in depth of the intermediate water-layers 

 at the two stations is not found to be exactly the same when 

 estimated from salinity or density as when estimated from 

 temperature. The discrepancy is but very small, however, 

 and we have evidently to deal with parts of the same 

 water-layers even if the salinity anomaly is a little different, 

 be it due to errors of observation or to different mi.xing 

 processes. We may at any rate conclude that the same 

 intermediate water- layers are situated about 100 metres 

 deeper at the Scottish station than at Stat. 115. 



The increase of surface temperature from May to 

 August is conspicuous (see Fig. 1), but even in May the 

 heating at the surface is so advanced that a considerable 

 temperature-gradient is found in the uppermost water-strata, 

 between the surface and 25 metres. This quasi discon- 

 tinuity is much more marked in August when the tem- 

 perature at the stations in question decreases very rapidly 



