recorder (PESR) in an effort to place the low- 

 est Nansen bottle at approximately 25 meters 

 above the bottom. The PESR had been damaged 

 and was not operating properly; therefore, it 

 was not possible to use the pinger as effectively 

 as would have been possible otherwise. 



Two protected reversing thermometers were 

 used on each sampling bottle. At 200 meters and 

 deeper, unprotected reversing thermometers 

 were used on alternate bottles. Salinities were 

 determined aboard ship using inductive type 

 salinometers, and dissolved oxygen content was 

 determined using a modified Winkler oxygen 

 determination method ( Jacobsen, Robinson, and 

 Thompson, 1950). Thermometer corrections, 

 thermometric depths, sigma-t values and speci- 

 fic volume anomalies were determined using the 

 computer programs described by O'Hagan 

 (1964). These values were used in preparation 

 of the figures included in this report. Processed 

 temperature, salinity, and oxygen data were 

 submitted to the National Oceanographic Data 

 Center (NODC) for further processing and the 

 results are included in Appendix 1 (NODC 

 cruise numbers 31-938 and 31-963). 



DATA ANALYSIS PROCEDURE 



At the outset of the analysis of the data it 

 was planned to use the volume transports 

 through the sections as a basis of comparison. 

 It immediately became apparent that this would 

 not provide a legitimate quantitative means of 

 comparison, because in most cases the differ- 

 ence in dynamic height between stations used 

 in the computation of volume transports was 

 of the same order of magnitude as the possible 

 error in measurement. Kollmeyer (1964), in his 

 discussion of errors in dynamic height deter- 

 mination, points out that a realistic instrumen- 

 tal accuracy of ±0.02°C in the measurements of 

 temperature, and ±0.02%= in the measurement 

 of salinity, and a similar interpolation error, 

 would result in a possible error of ±0.0338 

 dynamic meters over 1000 meters. The effect of 

 this error on computed current speed, assum- 

 ing a latitude of 61 °N, station spacing of 35 

 miles and no navigational error, is 8.17 cm/sec. 

 This exceeds the computed current speed in al- 

 most all instances and thus makes the value of 

 the computed volume transports doubtful. As a 

 result, it was necessary to utilize a more quali- 

 tative method of comparison of the sections. 



T-S ANALYSIS 



Possibly the best means of comparing the 

 water mass characteristics of one section with 

 another is by analysis of T-S diagrams (figs. 

 2-14). The diagrams are arranged to allow com- 

 parisons of observations taken at approximate- 

 ly the same position at different times. Included 

 in the diagrams are dashed lines connecting the 

 point representing North Atlantic Water (T = 

 9°C, S = 35.31%„) and Irminger Sea Water (T 

 = 3°C, S = 34.89%o). A mixture of these two 

 types of water would fall along this line. A 

 second dashed line on each diagram connects 

 the point representing North Atlantic water 

 with that of Norwegian Sea Water (T = 0°C, S 

 = 34.95%o, off diagram) which overflows the 

 ridge. 



The most apparent features indicated by the 

 T-S diagrams were the similarity of the repeat- 

 ed stations at about the same position and the 

 deviations from the Atlantic-Irminger line on 

 part of the section (EVERGREEN stations 17- 

 21) between Iceland and Outer Bailey Bank 

 (figs. 2-6). The deviations from the Atlantic- 

 Irminger line toward the Norwegian Sea line 

 indicated the presence of varying quantities of 

 Norwegian Sea Water in the three-component 

 mixture. These observations were quite similar 

 to those described by Steele, et al. (1962). He 

 pointed out that at stations near Iceland, 

 Norwegian Sea Water was present as a 

 component of the water mixture from rela- 

 tively shallow depths to the bottom; where- 

 as at stations further away from Iceland, 

 it was limited to the bottom layer. These condi- 

 tions are shown in figures 17-21 ; the former by 

 the proportion of Norwegian Sea Water found 

 at CGC EVERGREEN stations 17A-D, and 

 the latter by the proportion of Norwegian Sea 

 Water near the bottom at CGC EVERGREEN 

 stations 18-21. In general the water between 

 Iceland and Outer Bailey Bank was a mixture 

 mostly of North Atlantic Water and Irminger 

 Sea Water. This was indicated by the T-S val- 

 ues falling near the Atlantic-Irminger line 

 (figs. 3-6). At stations between Outer Bailey 

 Bank and Scotland, the T-S curves tended 

 more closely to the line that represents a mix- 

 ture of North Atlantic Water and Norwegian 

 Sea Water (figs. 10-13). 



An exception to the temporal constancy of 

 T-S characteristics was observed by comparing 



