GEOCHEMISTRY AND PHYSICS OF OCEAN CIRCULATION 305 



results for the Atlantic for latitudes greater than 60° N have been 

 published by Fonselius and Ostlund of the Stockholm group and 

 results for the western South Pacific by Rafter and Fergusson. 

 For simplicity of presentation averages for various reservoirs are 

 given in Table III along with the number of samples analyzed 

 and the standard deviation of the results from the average for 

 each group. For each group the results agree nearly within the 

 error, and no significant variation with current or with geographic 

 location is apparent. 



Since many of the samples were collected subsequent to large- 

 scale production of C^^ during nuclear testing, a small correction 

 for contamination from this source is necessary. This correction 

 has been estimated both by considering only pre- 1956 samples 

 (column 2) and by applying a correction based on estimated 

 mixing rates within the system. Fortunately the corrections are 

 small, the largest being ten per mil for the South Pacific. No 

 significant change in the pattern of results is introduced. The 

 resulting corrected averages are summarized in Table IV in the 

 column for 1955. 



A further correction for the dilution of the surface ocean C^^ by 

 industrial CO2 is necessary. Fergusson has shown through C^^ 

 measurements on tree rings that the average dilution of the world's 

 atmospheric C^^ was close to 2% in 1955. Ocean mixing model 

 calculations suggest a dilution of about two-thirds the atmospheric 



Table II. Conversion of C'^ and C^^ Data to per Mil Scale 



A A* 



sfH _ - ^ sample ^ 1890 wood i nnn 



■^ 1890 wood 



-pi 3 - '^sample -'^atandard •( rvrvrv 



■'^atandard 



aC'^ = 8C'' - 25C1M 1 + ^j - 50.0 



measured isotope fractionation constant chosen 



difference correction to make AC^^sgo wood = 



(Analysis precision averages ±8 per mil) 



