450 SCIENCE PROGRESS 



passivity of the iron is destroyed and rusting proceeds (see also 

 Dunstan, Nature, 191 1, p. 381)." 



It would seem at first that a point has been scored by 

 Dunstan and Hill in showing that iron and other metals are 

 rendered "passive" even by treatment with alkali — but reflection 

 shows that in giving this proof they have cut the ground from 

 under their own feet and have disproved not only the correct- 

 ness of their assumption that metals undergo aerial oxidation 

 by direct action of oxygen and water but also of their assumption 

 that acid is not necessary for rusting and that it neither 

 conditions nor determines rusting. 



The passivity induced by alkali is destroyed, we are told, 

 by mere washing in four or five changes of distilled water — 

 such passivity is scarcely worth speaking of: presumably it 

 disappears because of the dissolution of the coating of oxide 

 by the carbonic acid in the water. In any case such an 

 observation is sufficient to dispose of the idea that the coating 

 has particular resistant properties. 



Moreover, assuming that in Friend's experiment the iron 

 was superficially oxidised, if rusting be a process in which 

 oxygen and water act directly, as Dunstan and Hill affirm, 

 why did not the iron rust and continue to rust when 

 kept in oxygen and water ? Being so easily removed it 

 cannot be supposed for one moment that the coating of oxide 

 which conditions passivity produced by subjecting iron to 

 the action of a weak alkali is of so special a character as to 

 be rust-proof. The same argument applies to Moody's experi- 

 ments. It must be held, in fact, that Moody and Friend have 

 proved definitely that rusting is not a process of direct oxidation 

 by oxygen dissolved in water. And as rusting takes place 

 the moment acid is free to act together with oxygen and 

 not before, it is difficult to see how it can be denied that acid 

 is the determining cause of rusting and to be as necessary 

 as oxygen. In point of fact, Dunstan and Hill seem to have 

 supplied whatever evidence was needed to place it beyond 

 doubt that rusting is an electrolytic process in which an 

 acid is involved together with oxygen and water. 



They seem to have difficulty in understanding the action 

 of alkalies on iron as in discussing their experiments on 

 passivity they remark : 



