Three composite carboys of seawater were prepared by the 

 Woods Hole Oceanographic Institution and designated numbers 

 1, 5, and 7. Another carboy of actual seawater collected from the 

 surface at about 50° N., 49° W., about 71/2 months earlier (25 

 July, 1957) was made available by the Coast Guard Oceanographic 

 Unit. This carboy, designated C-l-58, was a polyethylene plastic 

 container. The other carboys were glass and had been filled with 

 water stored for varying periods in glass containers. None of 

 the carboys had special seals until the beginning of the measure- 

 ments, when C-l-58 was placed under an oil seal. 



The WHOI laboratory ran 20 titrations on carboy 1, 14 titra- 

 tions on carboy 5, 9 titrations on carboy 7 and 10 titrations on 

 carboy C-l-58. The carboys were also measured on the WHOI 

 bridge and on the CGOU bridge. Mean values determined were 

 as follows : 



Carboy 

 1 



5 



7 



C-l-58 



Titration 



35.04 9 

 35.50! 



34.18e 



34.45r 



WHOI bridge 

 35.06s 



35.53 2 

 34.21s 



34.47e 



Whence the following differences were obtained 



Carboy 

 1 



5 



7 

 C-l-58 



Titration- 

 WHOI bridge 



-O.Ole 

 -0.03! 

 -0.03 2 

 -O.Olo 



Titration- 

 CGOU bridge 



+ O.OI5 

 + O.OO5 

 -O.OO5 

 + 0.00! 



CGOU bridge 



35.034 

 35.49a 

 34.19i 

 34.44„ 



WHOI bridge 

 CGOU bridge 



aver. +0.00 fl 



+ 0.03! 

 +0.03c 

 + 0.02 7 

 + 0.03i 



+ 0.03! 



Since both the CGOU bridge (model 4) and the WHOI bridge 

 (Bradshaw and Schleicher) were Wenner bridges they were both 

 more precise than titration, which is reflected in the smaller 

 variation in the difference between the two bridges than between 

 either bridge and titration. However, neither makes direct meas- 

 urements of salinity and each determines the electrical resistivity 

 of a sample of unknown salinity in terms of that of a sample of 

 known salinity, and hence depends upon a standardization. In the 

 case of the CGOU bridge this is accomplished by the construction 

 of a calibration curve on the basis of a number of samples of 

 actual seawater having a range of salinity and measured on the 

 bridge and by titration against Copenhagen standard water. In 

 the case of the WHOI bridge the standardization is accomplished 

 by using the Copenhagen standard water as a conductivity stand- 

 ard, assuming its conductivity is the same as the conductivity of a 

 sample of actual seawater whose chlorine content is the same as 

 the stated chlorine of the Copenhagen standard water. 



The calibration curve of the CGOU bridge used in the above 



32 



