72 



OBSERVATIONS AND RESULTS IN PHYSICAL OCEANOGRAPHY 



either the initial or final standards. In the absence of 

 any evidence to the contrary, it was assumed that this 

 difference was the result of a gradual change and the 

 difference was therefore proportioned according to the 

 number of samples measured in the cell. After this 

 shearing correction was applied to the slide-wire read- 

 ings, they were converted to salinities by means of the 

 previously discussed equation and corrections. The 

 final standards were allowed to remain in the measur- 

 ing cells until the bridge was used again. A specimen 

 set of observations and computations is shown in table 3. 



The design feature of having similar electrolytic 

 cells form two adjacent arms of the bridge has, as one 

 of its objectives, lessening the importance of accurate 

 temperature control. In other words, it was hoped that 

 by this device the effective temperature coefficient of 

 the instrument would be much less than that of sea water. 

 The efficacy of this arrangement was tested on the Car - 

 negie as follows: When the regulating temperature of 



the water bath was changed from 30° to 40° C, Copen- 

 hagen standard water, which was used as the final stand- 

 ard at the end of the last 30° C routine salinity run, was 

 left in the cells and was remeasured on the following day 

 at 40° C, with the auxiliary resistance Q having the 

 same setting as was used at 30° C. The differences in 

 slide-wire readings were converted into differences in 

 salinity and considered to be the effect produced by a 

 10° C change in temperature of a sample. This was done 

 on a basis of 1.0 unit on the slide wire corresponding to 

 a change of 0.01 per mille in salinity. This procedure 

 further was based on the assumptions that during the 

 period of about 24 hours the salinity of the solution in 

 the auxiliary or Y cell did not change and that the cell 

 constants did not change. Such assumptions were justi- 

 fiable as only a rough determination was made. The 

 slide-wire reading at the balance of the initial standard 

 was 698.5 in each case, by definition. Either because of 

 changes in cell constants or changes in the auxiliary cell 



Table 3. Specimen set of observations and computations of salinity 



Cell A 



711.0 



hO.13/7 



Cell B 



-15.8 



933 



649 



3189 



23 



94 



377 



1888 



1412 







46 



187 



473 



2765 



2335 



5 



70 



281 



666 



aCopenhagen standard water. b 34.25 per mille by titration by J. H. P., May 28, 1929. c 34.99 

 per mille by titration by J. H. P., May 28, 1929. d 34.53 per mille by titration by J. H. P., May 28, 

 1929. 6 34.55 per mille by titration by J. H. P., May 28, 1929. f 34.64 per mille by titration by 

 J. H. P., May 28, 1929. g 34.79 per mille by titration by J. H. P., May 28, 1929. "Initial standard 



discarded as being probably in error. 



