XI. A, 1 Heise and Aguilar: Oxygen Consumption of Waters 47 



Obviously, then, the presence of hypochlorites in appreciable 

 quantities introduces errors into the determination of oxygen con- 

 sumption, although this determination is at best so notoriously 

 inaccurate that the error introduced by free chlorine is generally 

 of minor importance unless great accuracy is desired. The 

 effect of hypochlorites can readily be eliminated by the addition 

 of a few milligrams of sodium sulphite. After a few hours the 

 excess sodium sulphite is decomposed and it has no appreciable 

 effect on the determination. Before proceeding to the final di- 

 gestion, however, it is best to digest the sample, which has been 

 . standing in contact with sodium sulphite, for three minutes with 

 sulphuric acid, in the cold, and to add a drop of permanganate 

 to make sure that the sulphite has all been destroyed. 



SUMMARY AND CONCLUSIONS 



Some of the factors influencing the determination of the 

 oxygen-consuming capacity by means of potassium perman- 

 ganate in acid solution have been studied quantitatively. Time 

 and temperature have so great an effect that concordant results 

 cannot be obtained unless great care is exercised to have all 

 determinations made under identical conditions. 



The influence of dissolved substances, especially of chlorides, 

 is very large. The effect of the latter is reduced, although not 

 eliminated, if digestion is carried on at or below room temper- 

 ature. The final titration of excess potassium permanganate 

 should be made with sodium thiosulphate (in the presence of 

 potassium iodide and starch indicator). The error due to chlo- 

 rides is reduced by digesting the water samples with potas- 

 sium permanganate in alkaline (Schulze method), instead of 

 in acid solution, the final titration being made with sodium 

 thiosulphate. 



Hypochlorites oxidize organic matter in cold or hot solution 

 and should be eliminated before a determination of the oxygen- 

 consuming power is made. 



The determination of oxygen consumption at best is not an 

 accurate measure of the organic content of a water, and it gives 

 such uncertain results that isolated determinations are of very 

 little value. It is only when a water supply is to be examined 

 repeatedly that the determination becomes very useful. 



