Routine salinity measurements, as in previous years, were made with 

 a Wenner salinity bridge. Prior to the beginning of the 1957 season the 

 bridge was cleaned and the calibration curve redetermined. For this 

 purpose several composite samples of actual sea water collected during 

 the 1956 field work and well distributed over the range of salinities 

 encountered in the Grand Banks region and the Labrador Sea were 

 measured on the Wenner bridge and by silver nitrate titration. Assuming 

 that, over the range of salinity involved, the relationship between 

 conductivity and salinity has the form 



C = C" (K + LS) 

 where C is the conductivity corresponding to any salinity S and C is 

 the conductivity corresponding to salinity S', and K and L are constants, 

 the calibration curve of the bridge has the form 



a 



S = c 



b + m 

 where S is the salinity, m is the reading of the X-dials of the bridge at 

 balance and a, b, and c are constants. 



The constant b was measured electrically and mean values of a and c 

 were determined from the measurements of the several samples by 

 bridge and titration. The resulting calibration curve was found to be 



9817.725 



S = 4.2534 



200.2 + m 

 This method permits the arbitrary selection of a single point on the 

 calibration curve. This point was so selected as to bring the salinity 

 of Copenhagen normal sea water and the salinity of the deep water near 

 the middle of the range of the bridge, a salinity of 35°/ 00 corresponding 

 to a dial reading of 49.911. 



As the calibration curve, determined in this manner, is defined by the 

 silver nitrate titration of the several samples used, the accuracy of the 

 measurement of salinities by the bridge is no better than the accuracy 

 of the silver nitrate titrations used in the calibration. The precision 

 with which the salinities may be measured with the bridge is better, 

 however, and is considered to be about 0.005 / 00 . 



Water from an oil-sealed carboy of sea water was used as a working- 

 standard for the routine measurements. At least twice during each 

 run samples of Copenhagen normal sea water were measured as unknowns. 

 These measurements indicated corrections of less than .005°/ oo except 

 for stations 6504 through 6519 for which a correction of +.01°/ oo was 

 indicated and applied. Copenhagen water of the batch P 22 was used 

 as the reference standard and a series of measurements were also made 

 to compare the conductivity-salinity relationship of the batch P 2 2 with 

 that of batch P 2 3 to permit a subsequent shift to the latter as a future 

 reference standard. 



83 



