HO HARBOUR CONSTRUCTION. 



For a distance of 3 inches below the water-line all the bars 

 were coated with a reddish-yellow peroxide of iron, from absorp- 

 tion of oxygen ; and below this level they were covered with a 

 fine black carbonaceous deposit, which it was considered would 

 develop local galvanic action and cause pitting. 1 



It might appear that small differences in the composition of 

 various irons and steels would not be capable of setting up any 

 appreciable galvanic action when these metals are in contact, or 

 of greatly affecting the rate of their corrosion. The experiments 

 conducted by Mr. Andrews, however, show that such is far from 

 being the case, and that what might be considered quite insig- 

 nificant chemical or structural differences constitute a very 

 important element of destruction. 



The rate of corrosion of various qualities of steel plates, in 

 galvanic connection with wrought iron, was increased, beyond 

 the rate of simple corrosion of similar plates, by about 27 per 

 cent, on the average of the tests made. 



In experiments on simple corrosion, bright plates being 

 exposed singly and separately to sea- water for a period of 110 

 weeks, Mr. Andrews found the loss by oxidation, per square foot 

 of surface exposed, amounted on an average to 



260*82 grains in the wrought-iron plates = ^Vy * ncn ^ n 

 thickness per annum. 



313*06 grains in the various kinds of steel plates = y-sW * ncn 

 in thickness per annum. 



38515 grains in the cast-metal plates = T5 ^ ff inch in thick- 

 ness per annum. 2 



These results do not agree well with measurements which 

 I shall presently give, indicative of the rate of corrosion of 

 wrought-iron plates and bars in the wreck of the s.s. Gambia. 



This wreck lay in the Indian Ocean, fully exposed, not only 

 to the corrosive, but also to the erosive power of the waves ; 

 the water in contact with the plates was therefore being 

 constantly renewed. All the plates and bars that were measured 

 lay between high and low water, and were thus exposed to the 

 combined action of air and water. 



These unfavourable conditions would doubtless account for a 

 good deal of the difference to which I have referred, but it is, 

 nevertheless, so large as to be rather startling. 



1 Min. Proc. Inst. C.R, vol. Ixxvii. 



2 Compiled from tables given by Mr. Andrews (Min. Proc. In*t. C.E., vol. 

 Ixxxii. p. 297). 



