74 



BJ0RN HELLAND-MANSEN 



[REP. OF THE "MICHAEL SARS" NORTH 



been lower if tlie insulating capacity of the water-bottle 

 had been perfect. A temperature /// situ of 2-27° corre- 

 sponds very nearly to neutral equilibrium. Perhaps 

 the temperature should be put lower still. In any case 

 it seems certain that the temperature at this station de- 

 creased between 4000 and 4750 metres. 



At the "Michael Sars" stations 10 A, 49 C and 63 

 the temperature in situ increased downwards at great 

 depths. The determinations were made by means of 

 RiCHTER reversing thermometers, and the increase is greater 

 than the error of observation. 



From Stat. 10 A we have no determinations of sa- 

 linity, but we have rea.son to believe that the salinity was 

 uniform from 3000 metres downwards. At Stat. 49 C 

 the salinity was 34-90 "oo both at 3950 and at 4950 

 metres. At Stat. 63 the salinity at 4000 metres was a 

 little lower (34-88 %o) than at the two other stations. 

 This agrees with the supposition that the deep water has 

 a sligthly higher salinity in the eastern part of the North 

 Atlantic than in the western. The salinity at 4850 metres 

 at Stat. 63 was not determined, but probably it was the 

 same as at 4000 metres. 



dS 

 Provided that — m^ O in the three cases where 

 Sz 



Tm increases downwards, we may compare the potential 

 temperatures from the point of view of adiabatic equi- 

 librium. We then find that the temperatures in situ did 

 not increase as much downwards as would correspond to 

 adiabatic equilibrium. We had at all three stations a 

 state of positive stability, however slightly developed. 

 The vertical variation of potential temperature per 1000 

 metres was — 0-04° at Stat. 10 A, — 0-09= at Stat. 49 C 

 and — 0-11° at Stat. 63. It may be that the differences 

 between these values correspond to characteristic regional 

 variations in the North Atlantic. 



Temperatures near the bottom at great depths in the 

 North Atlantic have been taken by several expeditions. 

 Most of them are from the "Challenger" Expedition, the 

 observations being made in 1873 and 1876 by means of 

 the Miller-Casella maximum and minimum thermometers. 

 These observations are not so accurate as observations 

 with the modern reversing thermometers. The thermo- 

 meters used on board the "Challenger" were mostly read 

 in 0-1° F., but the instrumental error was probably greater 

 in many cases, and the corrections applied are rather 

 uncertain. 



The principle of the Miller-Casella thermometers does 

 not allow the registration of secondary maxima appearing 

 at depths passed by the instruments when they are hauled 

 up. A slight increase of temperature towards the bottom 

 cannot be detected by these thermometers. Observations 

 ostensibly taken near the bottom may, in reality, show 



a minimum of temperature occurring in the water at some 

 distance above the bottom. In such cases the tempera- 

 tures recorded as bottom temperatures are too low, if 

 the observations are otherwise correct. For instance in 

 the Philippine Deep where reversing thermometers — as 

 mentioned in the previous section — showed an increase 

 of temperature from 1-50" C. at 5000 metres to 2-60° at 

 9788, Miller-Casella thermometers would have uniformly 

 shown 1-50° C. all the the way from 5000 metres down 

 to the bottom. The bottom temperatures recorded by the 

 "Challenger" Expedition cannot therefore be used for a 

 discussion of the adiabatic variations in the deep parts of 

 the ocean in the same way as the more modern observ- 

 ations from the Pacific deeps. It may be added, how- 

 ever, that the determinations of temperature from the 

 "Challenger" Expedition seem otherwise, in the main, to 

 be remarkably good. 



The chart. Fig. 26, shows the position of the "Chal- 

 lenger" stations and some stations from other expeditions, 

 where observations have been taken near the bottom at 

 depths of 4000 metres or more. The four "Michael Sars" 

 stations, referred to above, are also included. In the chart, 

 isobaths are drawn for 4000, 5000 and 6000 metres. The 

 central longitudinal ridge stands out very clearly, dividing 

 the ocean at great depths into an eastern and a western 

 region. 



Fig. 27 exhibits the 'bottom temperatures' from depths 

 below 4000 metres at the "Challenger" stations. The 

 temperatures in situ (rm) are shown to the left in the 

 figure, and the corresponding potential temperatures (Q) 

 to the right. The black dots refer to stations on the 

 eastern side of the central ridge, and the crosses refer to 

 stations on the western side. Stations south of 10° N. 

 are omitted. 



The latter figure shows that the temperatures of the 

 bottom-water are generally higher in the eastern part of 

 the North Atlantic than in the western. When we group 

 the temperatures for some intervals of depth and calculate 

 the averages (in C°), we find: 



The difference between the mean temperature in situ 

 in the eastern part and that in the western part becomes 

 more marked the deeper one goes. It is relatively small, 

 about 0-1° C, at depths from 4000 to 4500 metres, in- 



