4 io SCIENCE PROGRESS 



shown that rusting can take place in absence of carbonic acid." 

 Previous to this it had been considered that the action which 

 certain materials, such as lime and the alkalis, are known to 

 exercise in preventing the formation of rust, was due to their 

 power, as alkalis, of absorbing carbon dioxide ; but a number 

 of other substances were cited by Prof. Dunstan as sharing with 

 alkalis the power of inhibiting rusting. These were sodium 

 nitrite, potassium ferrocyanide, chromic acid, potassium chromate 

 and potassium dichromate. A new theory of rusting was 

 suggested which was based on the fact that hydrogen peroxide 

 was known to be formed in certain previously studied cases of 

 oxidation ; this had been established for certain physiological 

 processes by Hoppe Seyler and for the oxidation of certain 

 metals, such as zinc, by Schonbein and Moritz Traube. Prof. 

 Dunstan considered that the rusting of iron took place as 

 represented by the equations — ■ 



Fe + 2 + H 2 = FeO + H 2 O a ... (3) 

 2 FeO + H 2 2 = Fe 2 2 (OH) 2 . ... (4) 



According to this view the action of pure oxygen and water 

 on iron is to produce ferrous oxide and hydrogen peroxide, 

 the latter then interacting with the ferrous oxide, forming 

 hydrated ferric oxide or rust. This view, originally put forward 

 in 1898, was also advocated two years later by Prof. Dunstan 

 in a Report on the Atmospheric Corrosion of Steel Rails 

 presented to the Steel Rails Committee of the Board of Trade 

 (1900). It was considered to give a simple explanation of the 

 action of such substances as alkalis and oxidising agents in 

 preventing the conversion of iron into rust, this property being 

 attributed to the power possessed by these substances of decom- 

 posing hydrogen peroxide at the moment of its formation. It 

 was adduced as evidence in favour of this view that, if a plate of 

 highly polished steel is immersed in a strongly alkaline solu- 

 tion of the peroxide, decomposition of the latter occurs very 

 rapidly, bubbles of oxygen being liberated on the surface of 

 the steel ; yet no rusting is seen to occur. The composition 

 of several samples of ordinary iron rust was found to be 

 approximately that corresponding with the formula Fe 2 2 (OH) 2 ; 

 it was pointed out as "a curious coincidence, if nothing more, 

 that this is equivalent to — 



2 FeO + H 2 2 ." 



