MATERIALS. 139 



to this subject, a vast amount of useful information has been 

 compressed. 



In connection with the employment of these metals in sea- 

 works, a very important factor has to be considered. I refer to 

 their durability, or power to resist the corrosive influences of 

 the saline constituents of the water, aided, as they are, by 

 carbonic acid gas, oxygen, and, under certain circumstances, by 

 galvanic action arising from various causes. 



The records of observations made on the corrosion of metals 

 in sea- water, under varied conditions, differ so widely from each 

 other, "and indeed appear to be so contradictory the one to the 

 other, that with our present knowledge it seems to be impossible 

 to lay down any rule for the rate of oxidation of metals in sea- 

 water, any more than it can be done for similar metals on land. 

 Nevertheless, records of what has actually taken place especially 

 if all the surrounding conditions have been carefully noted 

 enable deductions to be drawn by analogy, and we may hope 

 that, as these records multiply, it may be possible to predict what 

 the life of an iron or steel structure in the sea will be with more 

 accuracy than is possible at present. 



In the years 1884 and 1885, Mr. T. Andrews communicated 

 to the Institution of Civil Engineers the results of his careful 

 and valuable investigations extending over a period of about 

 four and a half years on galvanic action between wrought 

 iron, cast metals, and various steels during long exposure in 

 sea- water ; and also on the simple corrosion of these metals. 



In the first series of experiments, the test-bars operated 

 upon, in the galvanic tests, were 13 inches long, accurately 

 turned to exactly 3 inches diameter, and polished. These bars 

 were placed in galvanic connection, and were immersed in jars 

 or baths containing filtered sea-water, which was periodically 

 changed, and compensation made for evaporation by the addition 

 of distilled water. 



After 300 days' immersion the bars were accurately gauged 

 by a very delicate gauge, measuring absolutely to T oo 'o i ncn 5 anc ^ 

 it was found that the several specimens of steel had lost in 

 diameter, on an average, 0*0022 inch, the cast-metal specimens 

 0'0077 inch, and the wrought-iron nil, excepting in two specimens, 

 in which the loss amounted to O'OOIS and O'OOIO respectively. 

 The galvanic action going on between these metals was clearly 

 indicated and measured by galvanometers. 



