The greater amount of energy required to thread steel pipe with common dies was perhaps not the chief factor operating to prevent the more extensive use of steel pipe, so much as was the widespread belief that pipe steel possessed a great amount of carbon in its composition, and, therefore, was more easily corroded under ordinary conditions than was wrought-iron. As a matter of fact, however, there is but little difference between the composition of iron and of pipe steel. The steel used in the manufacture of pipe is really a refined iron made by the Bessemer process. It is similar to wrought-iron in that the carbon is reduced to a few hundredths of one per cent., but differs from wrought-iron in that it is free from cinder, scale, and other impurities. Pipe steel, practically speaking, is a superior grade of wrought-iron, the cinder of which has been removed from the molten metal in the process of refining the pig iron.

Numerous experiments have been made to determine the relative corrodibility of wrought-iron and steel. The result of various investigations up to 1892 has been collected and discussed in the "'Metallurgy of Steel," by H. M. Howe, Professor of Metallurgy at Columbia University, formerly of Harvard, from which book most of the following data were obtained. It might be remarked in passing, that great improvement has been made in the manufacture of pipe steel since 1892, and that tests made within recent years show far better results that those cited here.

The result of Professor Howe's study of the relative corrosion of iron and steel is summed up in his opinion that "there is probably no important difference in the rate at which these two classes of iron corrode under ordinary conditions." There are conditions, however, under which wrought-iron is more resistant than steel, others under which steel is more resistant than wrought-iron, while for most uses there is no appreciable difference in the corrodi-bility of the two metals. For instance, wrought-iron corrodes more rapidly than steel when immersed in foul river water, also exposed to acidulated waters. On the other hand, steel corrodes faster than wrought-iron when immersed in hot sea water or when exposed to the weather, and the corrosion of wrought-iron and steel is equal when immersed in cold water other than pure sea water; in pure river water, in sewage-bearing sea waters and in bilge water.

It might further be stated that while steel corrodes faster than wrought-iron when exposed to hot sea water, the loss is so infinitesimal that it in no way interferes with the use of steel for marine work, and that all plates and tubes used in marine boilers and in surface condensers are made of steel. Indeed, the use of steel for marine work is not only permitted but preferred by naval engineers, and is approved by the Lloyds Register, whose chief engineer-surveyor, in a private communication to Professor Howe, stated that "Experience has proved that steel does resist corrosion equally as well as iron, and is used almost exclusively in the manufacture of marine boilers. Ninety-nine out of every hundred boilers constructed under the inspection of this society's surveyors are made of steel. In fact, the use of iron for marine boiler-making is a thing of the past."

In Table LXVIII are shown the results of experiments of two different observers to determine the loss of wrought-iron and steel from corrosion, in pounds per square foot of surface per annum. From the results tabulated it can be seen that the difference in corrodibility between the two metals is so slight that it might be due to peculiarities of structure in the samples used, and would indicate that the difference in corrodibility between the two metals is so slight that for practical purposes it can be assumed that the corrosion is equal.

Table LXIX shows the relative corrosion of wrought-iron and steel in small scale tests under different conditions of exposure.

The results of experiments on the corrodibility of wrought-iron and steel all point to the fact that when exposed to the weather steel will corrode more rapidly than will wrought-iron. This fact would probably assign to steel pipe a lower value than wrought-iron pipe for durability, were it not that coating or covering pipe protects it from the weather, and most exposed pipe is either painted, covered or galvanized. In this connection Mallet's observations, in Table LXVIII, on the corrodibility of galvanized wrought-iron pipe are interesting as showing the protection offered by a coating of zinc. The fact that galvanizing protects steel pipe in the same manner as it does wrought-iron would seem to indicate that economy would be effected in the life of wrought pipe by having it galvanized before use. This would be particularly desirable or advantageous for steel pipe, which, owing to the less cost of its manufacture, could probably be put on the market galvanized at the cost of plain wrought-iron pipe.