328 SCIENCE PROGRESS 



and a similar cast slab of silicate of iron (ball furnace cinder) 

 were put into a beaker containing the viscous indicator just 

 about to set ; they were connected externally by copper wire 

 through a galvanometer. A strong current flowed immediately 

 and the gelatin surrounding the iron began to turn blue. Next 

 day, the slag had a large pink halo around it and the iron one 

 of intense blue topped by a thin ring of pink on the surface of the 

 jelly. Mr. Cobb attributes the production of the pink ring at 

 the surface to a current flowing from one part to another of the 

 iron bar, in consequence of the difference in potential existing 

 between neighbouring portions owing to electrical inequalities 

 in the bar ; he does not fully realise apparently that the current 

 which gives rise to the pink is conditioned initially and deter- 

 mined in direction by the oxygen polarisation of the bar at the 

 surface of the liquid — though he is evidently alive to the fact 

 that such polarisation is operative. This and other illustrations 

 given by Mr. Cobb are, however, particularly interesting ; in 

 fact, his paper is one which better repays close study than any 

 other I have met with on the subject. 



After reading the various books and essays on the corrosion 

 of iron, the impression that remains is that a very large amount 

 of work has been done to little or no purpose, owing to the 

 lack of guiding principles and the absence of clear thinking. 

 On the practical side, little need be added beyond what was 

 said in the earlier section of this essay : as there pointed out, 

 it can scarcely be regarded as otherwise than established that 

 rusting takes place only in presence of acid, although, para- 

 doxical as it may seem, this condition is compatible with the 

 manifestation of a certain degree of alkalinity, as in the case of 

 weak solutions of alkaline carbonates exposed to the air. 



Apparently, neutral salts have no direct effect in promoting 

 rusting : certain salts, such as the chromates, have a distinct 

 inhibiting effect but the manner in which protection is secured 

 by their use is by no means clear. The ultimate conclusion is, 

 without question, that, if it be desired to prevent iron from 

 rusting, the access of liquid water together with that of acid 

 and oxygen must be guarded against in every possible way : 

 in other words, as acid is always present in our atmosphere, 

 the iron must always be protected against moisture. 



