ACTION OF WATER ON IRON. 269 



structure, is less acted upon ; and the varieties of iron which 

 present scarcely any symptoms of a crystalline texture at all, 

 but still are grained or mottled, and can barely be touched by 

 the file, turned, or bored, are those which, while they are still 

 capable of being used for almost every purpose to which cast 

 ii'on is applicable, are the least susceptible of alteration or decay. 



40. The officers of the French artillery, amongst whom M. 

 Born has been most conspicuous, have made a number of expe- 

 riments on this branch of our subject. They have found that 

 the corrosion of iron by air and water is greater in proportion 

 to the purity or goodness of the coke with which the iron is 

 made, and that it is altered less when made with charcoal than 

 with coke. In the former case, it is probable this arises from 

 the iron containing the largest dose of uncombined carbon or 

 graphite ; and in the latter, namely, in that made with charcoal, 

 it seems to arise from the less quantity of silicium contained in 

 this cast iron. Various careful analyses made by Berzelius, 

 Karsten, Berthier, and others, show that, while coke-made iron 

 contains from 0*025 to 0*045 of silicium, that made with char- 

 coal only contains from 0*002 to 0*013 ; and it is certain that 

 the presence of silicium disposes iron to corrode, although in 

 dissolving in menstrua it may sometimes act as a mechanical 

 protector, covering it with a coat of silex. 



41. M. Born has also observed that iron cast in "dry sand," 

 or "in loam" moulds faced with charcoal, oxidates much less 

 speedily than when cast in green sand ; and that "chilled" cast 

 iron, or that cast in iron moulds, is the least of all susceptible 

 of this change. — {Comptes Rendus, 1837.) 



42. Becquerel, in remarking upon these statements, observes, 

 that cannon which are cast of close gray iron, and in "dry sand," 

 sustain little alteration further than a single coat of rust, or 

 browning like a gun-barrel, which seems to suspend further 

 action. This he attributes to the charcoal facing of the mouldy 

 and adds, that if it were possible to carburate the surfaces of ob- 

 jects cast in iron in the operation of moulding, this alone would 

 preserve them from further oxidation. There appears, how- 

 ever, here to be a serious mistake ; the presence of a carbona- 

 ceous coat on the surface of cast iron, unless impervious to air 

 and water, cannot preserve it from rust, however uniformly 

 spread. That it should do so, would be at variance not only 

 with the observed facts, and with the circumstance that the coat 

 of plumbago formed by the action of sea-water on iron does not 

 'preserve the remainder, but is at variance with an experiment 

 of Becquerel himself, in which he shows that the application of 

 a piece of common charcoal to the surface of iron in a solution 



