284 



[chap. 4 



North Atlantic, If we make, as did Iselin (1940), the hypothesis that the pro- 

 cesses which produce the warm surface masses of the central Atlantic water 

 are more or less constant in time, then an increasing transport in the North 

 Atlantic gyre, accompanied by the deepening of the thermocline in the Sargasso 

 Seal (ag shown in the top right of Fig. 90), must be accompanied by a radial 

 shrinkage of the whole sea-current system. Thus the Gulf Stream should shift 

 to the south, leading to cooling at higher Atlantic latitudes! We are thus left 

 with three comparative deductions between the two test periods which, though 

 not provably incompatible, are not yet demonstrably otherwise, namely: 



Worm layer 

 Deep water 



c o 



o >. 



5 c 



< 19 



i 



Jig 



Warm layer 

 Deep woter 



INITIAL LATER 



STEADY STATE 



Fig. 90. Zonal, vertical profiles showing schematically the creation and maintenance of 

 maximum thickness of warm oceanic surface layer under the influence of anti- 

 cyclonic wind stress, and minimum thickness of same layer under cyclonic wind 

 stress. (After Bjerknes, 1959, Fig. 3. By courtesy of the author and the publishers.) 

 Full lines : sea surface and interior isobaric surfaces. Dashed line : density dis- 

 continuity surface. Leftward displacement of center of oceanic deformation relative 

 to air-circulation center due to variation of Coriolis parameter with latitude. (For 

 explanation see, for example, Stommel, 1958, chap. VII.) 



increased wind stress and inferred faster water circulation around the Atlantic 

 gyre, apparent marked warming in the meander region off Grand Banks, and 

 seemingly unaltered geostrophic mass transport and density field in the main 

 portion of the Gulf Stream. 



It is plain that such long-period changes, while they are valuable to identify 

 where possible, and their origins intriguing to debate, are at the borderline of 

 subject matter that our present tools and physical theories can relate to one 

 another meaningfully, beyond partial quantitative descriptions, statistics, and 

 courageous hypotheses, as in the works of Namias and Bjerknes. The latter 



1 Stommel produces some weak confirmatory evidence for this occurrence. 



