THE RUSTING OF IRON 411 



The hydrogen peroxide theory is, however, at the outset 

 open to objection. First, the formation of hydrogen peroxide 

 during the rusting of iron has never been directly proved, in 

 spite of the fact that hydrogen peroxide is a substance easily 

 detected when present in even the most minute quantity and 

 that, according to equations 3 and 4, it might be expected 

 to accumulate to some extent during the process of rusting. 

 Secondly, there is no reason why the destruction of hydrogen 

 peroxide, the product of the action represented by equation 3, 

 should prevent the action of oxygen and water on metallic iron 

 from continuing ; it might, indeed, be thought that it would 

 favour such an action. 



It must here be emphasised that Prof. Dunstan considered 

 that, in view of his results, the older theory which attributed 

 rusting to carbonic acid was practically to be abandoned 

 altogether : " So far as ordinary rusting is concerned un- 

 doubtedly it must be, because carbonic acid gas is not necessary 

 for rusting to occur." And again : " This hypothesis has been 

 proved by the results of the present investigation to be quite 

 untenable, since it has been shown that rusting can take place 

 in the absence of carbonic acid." An alternative method of 

 explaining rusting had, indeed, appeared necessary when all 

 attempts to arrest the change by excluding carbon dioxide had 

 failed. The validity of the hydrogen peroxide theory, therefore, 

 depends on the question whether pure water and pure oxygen 

 are together sufficient of themselves to cause iron to rust. It 

 therefore becomes of importance to consider the method by 

 which the air and water were purified in Prof. Dunstan's experi- 

 ments, so as to ensure the absence of traces of carbon dioxide. 



The apparatus used is illustrated in fig. i. 1 The iron, the 

 surface of which had been thoroughly cleaned and polished, 

 was contained in the tube A, which was attached at c by the 

 only rubber joint in the whole apparatus. Hydrogen, which 

 had been passed through several wash bottles and towers 

 containing caustic potash, was driven from B through the 

 distilling flask C, which contained water, lime and potassium 

 permanganate, so as to fill the apparatus completely, the clamp 

 c being closed. A quantity of water amounting to about 100 c.c. 

 was then distilled in a current of hydrogen from the flask C into 



1 I am indebted to the Chemical Society for permission to use the blocks 

 illustrating this article. 



