NO. 4 TEMPERATURE VARIATION'S IX THE NORTH ATLANTIC 



55 



good agreement between the single curves for the fields is therefore 

 not to be expected. These curves each by itself can make no pre- 

 tense that it represents with absolute correctness the conditions of 

 its region. 



However, it is clear that representative values may be obtained 

 by taking the mean for the 2° fields over a considerable region. We 

 have, as we have already said, therefore divided our whole northerly 

 region between 10° and 70° west longitude in six fields, each of 10° 

 longitude. Within each of these 10° longitude fields, we have taken 



1898 99 J900 1 



hO°N. 



6'f-6S' 



68-6^0 



Figure 19. Continuation of figure 18. Curves for fields between 60° and 

 70° west longitude, 40° and 41° north latitude. 



the mean of the anomalies of the mean temperatures of the chosen 

 2° fields. The temperature anomalies thus obtained for both decade 

 groups are given in table 2-W and are graphically expressed by 

 the curves in figure 20. These curves show a great correspondence, 

 and it is therefore doubtless to be assumed that they correspontl to 

 the actual temperature relations. 



This applies also for the field between 50° and 60° west longitude 

 where the curves for the single 2° fields are somewhat irregular. 

 Among other things the correspondence between the curves of the 

 first and the second decade groups for these 10° longitude fields 

 warrants the belief that the variations which they show depend in 



