Gooch and Austin — Oxidation of Manganese. 263 



by Deshayes's titration of permanganic acid in nitric acid by 

 means of standard arsenious acid.* Our experience in follow- 

 ing out this idea shows that the precipitated oxides of manga- 

 nese, as well as the soluble permanganate, may be easily 

 reduced with the aid of gentle heating by arsenious acid in 

 the presence of sulphuric acid, and that the determination of 

 the excess of arsenious acid by titration with iodine after neu- 

 tralization of the free sulphuric acid by an alkaline carbonate 

 gives exact data for estimating the oxidizing power of the 

 manganese compound. We found, however, that if the iodine 

 is allowed to come into contact with the manganous carbonate 

 thrown down by the alkaline carbonate, as is inevitable 

 for at least short intervals during the titration of the 

 arsenious acid in presence of the floating carbonate, the 

 danger arises of more or less reoxidation of the manganous 

 carbonate by the iodine and the consequent introduction of 

 error. Fortunately the difficulty may be obviated by adding 

 to the solution while still acid enough tartaric acid, or alkaline 

 tartrate, to prevent the precipitation of the manganese in the 

 subsequent neutralization by the bicarbonate. 



Confining our attention to the last two simpler iodometric 

 methods — the reduction of the higher oxide by an acidified 

 iodide on the one hand and by arsenious acid on the other — we 

 made, first, some experiments to determine the accuracy with 

 which manganese may be thus estimated. We used for the 

 manganese compound of known oxidizing power a solution of 

 potassium permanganate filtered carefully through asbestos and 

 standardized against ammonium oxalate which had been found 

 to be the exact equivalent of a specially prepared lead oxalate. 

 For each experiment a definite portion of this solution was 

 drawn from a burette and treated with a solution of pure man- 

 ganous sulphate until the color of the permanganate had van- 

 ished, thus precipitating a hydrous oxide approximating quite 

 closely probably to the condition of oxidation of the dioxide, 

 but containing at all events, whatever its actual composition 

 might be, the exact amount of available oxygen originally in 

 the permanganate- In" the experiments of the following table 

 this precipitate was treated with a solution of potassium iodide 

 (6 grm.) and tartaric acid (10 grm.), by which the freshly pre- 

 pared hydrate is dissolved quite as well as by the iodide and 

 hydrochloric acid of Pickering's original method and with less 

 risk of evolution of iodine outside the main reaction. From 

 the iodine found by titration with this sulphate we have cal- 

 culated the weight of manganese dioxide which would liberate 

 it; and a comparison of this value with the amount of the 

 dioxide theoretically precipitated by the interaction of the 



*Bull. Soc. Chim., xxix, 541. 



