150 



WEIBEL AND THURAS : RECORDING SALINITY 



utes be so near that of the open cell that the error will be less 

 than 0.02 in salinity. 



Although there. are as yet no observations on the rate of change 

 of salinity, it is assumed that the maximum change in a short 

 interval of time is not greater than 2 in salinity. This value is 

 estimated from the relative changes of temperature which are 

 discussed above. A response to this change to within 0.02 in 

 salinity in less than two minutes can be obtained by washing 

 out the cell at the rate of 3 or 4 volumes a minute. However, 

 if the temperature and salinity change at the same time, which is 



Elect t-olytic Cell 



c 



End view 



Plan view 



(m^ 



Side view 



Fig. 



¥^ 



quite probable, the lag of the recorder will be the sum of the 

 two and if they are both a maximum at the same time, which 

 is quite improbable, then the total time lag will be 4 minutes. 

 This is probably the worst possible condition that can occur, 

 but it is well to point out that as the regain of the true reading 

 is exponential the recorder Avill at first quickly approach its 

 correct value and although it takes four minutes to reach within 

 0.02 in salinity it will take much less than two minutes to obtain 

 an accuracy of 0.04 in salinity. However, if a more careful 

 study of the change in salinity between different water masses 



