WEIBEL AND THUKAS: RECORDING SALINITY 151 



is required it is only necessar}- to increase the flow through the 

 open cell and to make a correction for the error caused by the 

 rapid change in temperature. Since another recorder will be 

 used to measure the temperature of the sea water this tempera- 

 ture correction for the salinitj^ can easily be obtained. 



MULTIPLE TUBE CELLS 



Each of the 2 cells that have been designed (fig. 2) contains 

 6 parallel glass tubes 14 cm. long and 1 cm. in diameter. These 

 tubes are joined at each end to bulbs containing annular-shaped 

 platinum electrodes. Each electrode has an area of 5.3 sq. cm. 

 and is held rigidly in place by 4 platinum pins which are welded 

 to the electrode and sealed into the glass wall of the cell. The 

 cells are designed so that there are no pockets in which air can 

 collect, and the sea water is admitted in such a manner as to 

 sweep off any bubbles that might collect on the electrodes. 

 The inlet and outlet tubes are sufficiently large to insure a 

 thorough washing out of the cell in sufficient time to respond 

 to the maximum changes in salinity that are liable to be met 

 with on a vessel running at moderate speed. 



RECORDER 



In order to obtain a continuous record of sea-water salinity 

 the "NATieatstone bridge and galvanometer must be embodied in 

 a recorder mechanism such as that developed by the Leeds & 

 Northrup Company. The most important changes in their 

 present recorder are due to the use of alternating current. The 

 electrical connections are as shown in fig. 1. The current may 

 be obtained from the usual 60-cycle supply, but if only direct 

 current is available then the small direct-current motor used for 

 driving the recorder mechanism can be equipped with slip rings 

 and be operated as a converter. The recorder paper should be 

 ruled so that salinities can be read directly. 



INSTALLATION AND OPERATION ON BOARD SHIP 



To obtain a continuous record of surface salinity, the appara- 

 tus will be set up similarly to the temperature-recording appara- 



