Prof. Schonbein on the Polarization of Oxygen. 511 



gradual combustion of phosphorus as the type of all the slow 

 oxidations which inorganic and organic bodies undergo in moist 

 atmospheric air or pure common oxygen^ I suspected that the 

 peroxide of hydrogen might be produced, if not in all (from 

 secondary reasons), at least in a great number of cases, and 

 directed therefore my attention first to the slow oxidation of the 

 more readily oxidable metallic bodies. My conjecture proves 

 correct, as I have already found that half a dozen of the metals, 

 during their slow oxidation, give rise to the formation of very 

 appreciable quantities of HO^, as you will perceive from the 

 following statement. To detect with full certainty small quan- 

 tities of that compound, I required in the first place proper {i. e. 

 most delicate) tests for HO^, and I fully succeeded in finding out 

 more than one of that description ; in corroboration of which I 

 may tell you that by means of them I am able to detect the 

 millionth part of the said peroxide contained in water, and even 

 less. These tests depend upon the oxidizing and reducing 

 efi'ects produced by HO^upon certain substances. Dilute paste 

 of starch containing some iodide of potassium, if it be mixed 

 with water containing but half a millionth of HO^ is within 

 a very short time coloured dark blue on adding some drops 

 of a weak solution of any proto-salt of iron to the mixture. 

 The dilute solution of HO^ slightly acidulated by SO^ dis- 

 charges the red colour of an acidulated solution of the per- 

 manganate of potash (by reducing the acid of that salt to the 

 protoxide of manganese). HO'^ even in a most dilute state, 

 throws down prussian blue from a mixture of most dilute 

 solutions of the red cyanide of potassium and any persalt of 

 iron (by reducing Fe^O^ to FeO). Most dilute H0-, coloured 

 blue by some indigo solution, is rapidly discoloured on adding 

 some drops of a dilute solution of iron vitriol to the mixture. A 

 dilute solution of chromic acid is certainly a less delicate test 

 for HO^ than those mentioned; but its property of being 

 coloured azure-blue by water containing but 2 o^iroth of HO^, 

 makes it in many cases a valuable and practical test, which I 

 always use when I have to deal with water somewhat rich in HO^. 

 Now by means of these tests I have of late ascertained that, 

 during the slow oxidation of zinc, cadmium, lead, tin, bismuth, 

 and copper (effected by moist common oxygen or atmospheric 

 air), perceptible quantities of HO^ are always formed conjointly 

 with the oxides of those metals. To produce HO^ some of the 

 metals, in a state of mechanical division, as zinc, cadmium, 

 and lead, have only to be put in contact with pure water and 

 atmospheric air for a very short time ; but I find it more conve- 

 nient to amalgamate first the metals with mercury. Take, for 

 instance, 100 grms. of zinc-filings and the same quantity of 



