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



270 



200 - 



40C- 



800 



iOOO 



600 - 



Fig. 18. The vertical distribution of density at 3 pairs of stations in tlie eastern part of tlie Nortli Atlantic. 



summer. In May and June the temperature increases 

 fairly rapidly at this deptli but afterwards the rise of tem- 

 perature seems to be comparatively slow during some 

 weeks, and then it becomes quicker again. This course 

 of the variations may be explained by the variations in 

 the virtual conductivity of temperature. Fig. 18 illustrates 

 the variations of density from the end of May or begin- 

 ning of June to the end of August or beginning of Sep- 

 tember at three pairs of stations, one pair from each of 

 the areas in question. Both stations from area B are taken 

 at the same place; they are the saiue stations as repre- 

 sented in Fig. 10. The stations belonging to each of the 

 other pairs have not exactly the same position but are 

 situated near each other. The curves demonstrate that 

 the discontinuity layer develops very much from the end 

 of May to the end of August in intensity and thickness. 

 To begin with the discontinuity layer chiefly appears at 

 about 25 metres or a little deeper, in July it has evidently 

 developed so much that the conductivity of heat further 

 downwards is very much reduced with the result that the 

 rise of temperature at 50 metres is effectively retarded. 



All the time an absorption of heat radiation takes 

 place at 50 metres. As shown in section 30 this absorp- 

 tion gives a heating corresponding to an average rise of 

 temperature of about 0-004" C per day or 0-12"' C per 

 month in the summer half. The "temperature anomalies" 

 show in area B an increase of teiriperature in July and 

 August of 0-82° C or nearly 0-6° C more than accounted 

 for by the absorption of heat. According to the curve 

 in Fig. 16 the increase of temperature in July and August 

 is 0-25° C in the southernmo.st area (C) which corresponds 

 exactly to the heating calculated for tiie absorption. In 

 this case, no variations at 50 metres should take place 

 on account of conduction of heat during this time of the 

 year. It is quite obvious that the conditions so far are 

 different in different areas, as the currents and the degree 

 of stability are subjected to regional variations. 



At 100 metres a deformation of the curves similar to 

 that at 50 metres appears in suitimor. It is still more 

 luaiked in relation to the total seasonal variations than 

 at 50 metres. If the curves are rightly drawn they demon- 

 strate even a secondary miniiuum of temperature in August. 



