reading in microamperes is always exactly the 

 same as the first setting at 10° C. This pro- 

 cedure will give a series of counter -dial numbers 

 which will correspond to the temperature of the 

 water. In subsequent determinations, these 

 numbers are set on the counter dial to corre- 

 spond with the water temperature at the time of 

 oxygen determination. Throughout the calibra- 

 tion period it is important that the stirring be 

 sufficient to give the maximum read-out, and 

 that this stirring be uniform throughout the 

 duration of the calibration period. The magnetic 

 stirrer is ideally suited for this purpose. When 

 the temperature calibration curve is obtained, 

 remove the electrode and wash any oil from the 

 assembly with detergent. Remove the old mem- 

 brane, prepare the electrode according to in- 

 structions and reassemble . The instrument is 

 now ready for field or laboratory determinations 

 of oxygen . 



Table 1; — A comparison between Instrument readings and Winkler 

 determinations of oxygen content In thermo-chemlcally 

 equilibrated water obtained at various depths. The 

 differences between Winkler determinations are probably 

 within the limits of error of the method. 



Depth (feet) Temperature 



Instrument Reading 

 (microamperes) 



Winkler Oxygen 

 (Parts per million) 



OTHER VARIABLES 



In addition to being highly temperature 

 sensitive, the electrode is slightly light sensi- 

 tive, and the difference in light in the laboratory 

 and bright sunlight may make a diffe rence of 2 

 or 3 percent in oxygen determinations. This 

 variable can be minimized by placing a dark- 

 colored plastic baby bottle over the electrode 

 assembly. A large hole should be cut in the 

 bottom of the bottle and several small holes 

 drilled in the "shoulder" of the bottle to permit 

 circulation of water. This arrangement will 

 also protect the membrane when the instrument 

 is used in brush and rocks, or when it is lowered 

 to the bottom mud in lakes or ponds . 



The possibility that the electrode assem- 

 bly would be pressure sensitive was also in- 

 vestigated. The instrument was taken to a lock- 

 and-dam system and instrument readings were 

 compared with Winkler determinations at various 

 depths. The first series of determinations in- 

 dicated variations in oxygen content due to strat- 

 ification of water in the lock, but later the lock 

 was drained and rapidly refilled to permit the 

 passage of a barge. Temperature readings in- 

 dicated that thermal equilibrium was established, 

 and presumably chemical equilibrium as well. 

 Table 1 lists the results of the test to a depth 

 of 21 feet. Additional tests to a depth of 54 feet 



permitted the same conclusion -- that pressures 

 due to depth did not appreciably influence oxygen 

 determinations. This may only be true, how- 

 ever, if the electrode is properly constructed 

 and the membrane is placed tightly over the 

 platinum electrode. Then the pressure does not 

 significantly alter the thickness of the membrane 

 or the film of electrolyte behind it . 



RESPONSE TIME 



The response time of the electrode was 

 checked with a recorder. Two beakers of water 

 of the same temperature (24° G.) were placed 

 on a magnetic stirrer. Sodium sulfite was 

 added to one sample to remove the oxygen. The 

 other water was saturated with oxygen. The 

 electrode was placed in the zero -oxygen water 

 and the recorder output was adjusted to zero 

 The recorder was adjusted to read 100 percent 

 in the saturated water. The record is shown in 

 Figure 8 . The electrode output falls to zero 

 within about three minutes, but of course will 



