DAVIS STRAIT AND LABRADOR SEA 



189 



corresponding to the minimum practically attainable temperature 

 of the bottom-water component shown on figure 146. Thus frorn 

 figure 146 it will be seen that if there is no horizontal motion, verti- 

 cal convection to bottom may be established at stations 22, 23, 24, 25, 

 and 26 when the water columns have been cooled by winter chilling 

 to temperatures of 2.65°, 1.35°, 1.75°, 1.60°, and 2.80° C, respectively. 

 However, the rapid horizontal circulation in the upper levels at 

 stations 22 and 26 eliminate the possibility of vertical convection 

 there, and, of the remaining three stations, 25 is close enough to 

 the West Greenland Current to make it uncertain Avhether or not 

 deep vertical convection is possible. Attention is called to the fact 



18 19 20 21 22 



23 24 25 26 27 28 



M 



500 . 



0.13 0.50 0.41 



5.83 4.00 4.47 



10 



20 



25 



30 



25 50 MILES 



Figure 149.- — Maximum temperature to establish vertical convection, surface to bottom, 

 Resolution Island to Fiskernaesett. (From the Oudthaah's observations taken .Tune 11- 

 16, 1928.) Inset shows area in which bottom water is formed in the wintertime according 

 to the authors' views. 



that the average salinity of the water columns at stations 23, 24, and 

 25 lie between 34.81 and 34.91%o, the ran^e within which the bottom- 

 water component must fall. This indicates that a small central 

 part of this section lies in the area where the bottom water of the 

 Labrador Sea is formed. That this section passes close to the north- 

 ern boundary of the area of bottom-water formation is evident when 

 one remembers the horizontal components of the westward branch- 

 ing of the West Greenland Current south of Davis Strait. From 

 similar computations of the average salinity and maximum tem- 

 perature for vertical convection to bottom in the region of Davis 

 Strait Ridge two conclusions were reached — {a) that even if there 

 were no horizontal currents vertical convection to bottom could be 

 produced only at temperatures very close to the freezing point and 



