DAVIS STRAIT AND LABRADOR SEA 113 



slope band which from its derivation points to a swelling in the 

 west Greenland portion of the Labrador Current rather than an 

 increase in the Baffin Bay discharge. This is substantiated by the 

 excess in volumes recorded in 1933 (p. 54) for the West Greenland 

 Current off Cape Farewell and Ivigtut. 



The deficit of Labrador Current in 1931 along the American slope 

 is corroborated also (p. 50) by the fact that much of the West 

 Greenland Current off Cape Farewell branched out into the Labra- 

 dor Sea and was thus lost that summer to the Labrador Current. 

 Attention has already been called to the great scarcity of icebergs 

 south of Newfoundland in 1931, a total of 13 compared with the 

 average number of 420. 



As a further analysis of the components of the Labrador Current, 

 temperature-salinity correlation curves have been drawn based upon 

 all 1931 observations below a depth of 50 meters in the Labrador 

 Current of the American sector. The two solid lines embrace the 

 pattern of the temperature-salinity plots, the greater distance be- 

 tween the two curves near the bottom of the graph signifies a con- 

 sistent scattering of the correlation points in the cold low-salinity 

 water and a concentration in the higher temperature and salinity 

 brackets. The broken line is illustrative of the mean temperature- 

 salinity correlation for all the sections in the American sector the 

 summer of 1931. The lower left-hand portion of this curve repre- 

 sents the Baffin Land Current component and the upper right-hand 

 the West Greenland Current component. A point about halfway 

 along the broken line may be taken as representative of the division 

 between the mixture. A computation of the average volumes of the 

 current for the American sector in 1931 with reference to this 

 boundary results in 2.1 million cubic meters per second for the West 

 Greenland Current water and 1.4 million cubic meters per second for 

 the Baffin Land Current water. These proportions of 3 to 2 cor- 

 respond to previous estimates of the composition of the Labrador 

 Current based on the observations of 1928. (See p. 85.) 



The horizontal distribution of temperature and salinity, surface 

 to 600 meters, in the American sector during the summers of 1931, 

 1933, and 1934 is portrayed on figures 77 to 86. When these are 

 compared with figures 52 to 61, the temperature and salinity maps 

 for 1928, good agreement in the general form and position of the 

 isotherms and isohalines fs observed. The resemblance between 

 the 1928 and 1931 surface isotherms and isohalines is especially 

 striking, the 1928 temperatures in the American sector being gen- 

 erally about 1° C. lower than those in 1931. A minimum tempera- 

 ture about —1.5° C. was found on the American shelf in all of the 

 summers. A feature common to all of the 400- and 600-meter 

 temperature maps is a band of water warmer than its surroundings 

 which extended southward along the American slope to the latitude 

 of Belle Isle. This is undoubtedly the thermal influence of the west 

 Greenland water in the slope band of the Labrador Current which, 

 as remarked on page 100, was cooled as it progressed southward. 



Attention is especially called to the distribution of temperature 

 and salinity at a depth of 100 meters in the latitude of Belle Isle 

 just offshore of the continental edge (fig. 84). Temperatures of 



