Hutchinson — On the Estimation of Oxygen in Water. 137 



renders the solution less absorbent of free oxygen, although it acts 

 very rapidly upon dissolved oxygen. Before use this was further 

 diluted with three times its bulk of distilled water, recently boiled. 



The liquid recommended by which the change of colour detects 

 the completion of the process is either carmine indigo (sulphindigo- 

 tate of soda, Cg II4 JSTa NOSO3), or Coupler's aniline blue. 10 grammes 

 of the carmine indigo are recommended to be dissolved in one litre 

 of water, the product being kept in well-stoppered bottles also in the 

 dark. 



An ammoniacal solution of pure copper sulphate is also recom- 

 mended to be made, containing 4*46 grammes (or, more correctly, 

 4*471 grammes) of the crystallized salt per litre. This is to be used 

 for the standardization of the above two solutions. 



Since the reducing agent is so sensitive to the presence of oxygen, 

 it is necessary to make the estimations in an atmosphere of pure hydro- 

 gen. To ensure the purity of the hydrogen, I passed it through a 

 solution of nitrate of silver, in addition to the sulphuric acid, and the 

 tube containing pieces of caustic potash. 



We begin by finding the volume relation between the indigo and 

 hyposulphite. The burettes of the apparatus are filled, one with 

 indigo carmine solution, the other with hyposulphite ; a rapid ciu-- 

 rent of hydrogen is passed through the apparatus, a small quantity of 

 warm distilled water added, and this coloured by the addition of a 

 small quantity of indigo. "We now add cautiously the hyposulphite ; 

 the blue solution turns first green, and finally to a clear yellow tint. 

 If the whole of the air has been expelled from the apparatus, the 

 yellow tint will remain unchanged ; the slightest trace of oxygen causes 

 the surface of the liquid to become blue. A known volume of indigo 

 (25 cub. cents.) is now added, and the hyposulphite solution again 

 run in until the yellow tint appears, indicative of the reduction of the 

 whole of the indigo. The colour change is exceedingly sharp, one 

 drop being sufficient to change the colour froni green to yellow. If 

 the solution be acid, the blue colour changes first to red, and finally 

 the yellow tint appears. 



We next require to find the reducing power of the hyposulphite 

 in terms of oxygen, finding from this the amount of oxygen any volume 

 of the indigo will yield. This being a stable solution, the hyposul- 

 phite (being liable to change) can be readily standardized at any future 

 time. 



Two methods can be used, by which this reducing power can be 

 found : 



Pirst. By finding the quantity necessary to reduce the ammonia 

 copper solution, i.e., the amount which brings the blue solution to a 

 colourless state, by the reduction of the cupric to cuprous oxide, 10 

 cub. cents, of this solution yields 1 cub. cent, of oxygen (0°C. 760 

 m.ra.s. pres.) to the redwcer. 25 cub. cents, are operated on in 

 a smaller apparatus, similar to the one used for the water estima- 

 tions. I find, however, that the C/Oilour change in this plaii is so ijide- 



L2 



