This change was made after a series of experiments had been conducted with targets representing the sides of the Simoom and other vessels. These vessels had strong transverse frames spaced only 1 foot apart; and it was found that a very serious amount of splintering took place from the side of the ship first struck, while the opposite side was considerably damaged. On the whole, it was considered that the damage done by solid and hollow spherical shot to these iron ships was likely to prove of a more destructive character to the crews than the corresponding damage in a wood ship. But it was remarked that iron plating above inch in thickness sufficed to break up the shell and hollow shot from the heaviest guns then mounted in ships. This feature was undoubtedly a very great advantage of the iron sides, as compared with wood; and the destruction of the Turkish fleet at Sinope, as well as the experience with our own ships during the Crimean War, proved how great was the danger of wood hulls exposed to the fire of shell guns. On the whole, however, the decision arrived at from the trials in the Simoom target still holds good; and from that time to this no fighting ship of the Royal Navy has been built with uncovered iron sides, and closely spaced frames, in wake of the gun decks. Iron hulls were confined to armoured ships until the construction of the swift cruiser class, of which the Inconstant was the earliest example. In order to secure the requisite structural strength, an iron hull was then considered necessary; but the transverse frames were widely spaced, and the shattering effect of projectiles was still further reduced by covering the thin iron plating with wood planking. The Simoom target experiments had shown that wood so applied reduced splintering and damage: subsequent experiments at Shoeburyness, with targets representing respectively the sides of a wood frigate and those of a swift cruiser, have confirmed the soundness of this view, even when the vessels are exposed to the fire of heavier guns than those in use twenty-five years ago.

Before and abaft the central batteries or citadels of iron

clad ships, there are frequently considerable portions of the topsides formed by thin uncovered iron plating; but in action these unprotected spaces would not be occupied by men, and splintering would not be productive of serious consequences. The coast-defence gunboats of the Comet class also have their thin iron plating exposed, but in them the single heavy gun is carried on the upper deck "in the open," and there is little to be feared from splinters: the chances of these little vessels being hit are also small. In fact, the only cases where guns are fought under cover of a deck, and behind thin plating unprotected by wood planking, are in the recently designed belted ships of the Nelson class in the Royal Navy; but in them very special arrangements are made to prevent splintering. The plating is of steel, about twice the thickness of an ordinary iron side: there are no numerous vertical frames behind it to be shattered; and any damage that may be done is restricted to a limited space by means of "traverse bulkheads" which are splinter-proof. On the whole, therefore, it may be safely asserted that the unarmoured or partially protected iron fighting-ships of the Royal Navy are not open to the objections which were fairly urged against the Simoom and her consorts thirty years ago, and which apply with considerable force to iron-built merchant ships of the present day.

In concluding this chapter, brief reference must be made to the substitution of steel for iron in shipbuilding.

The chief reason for such a substitution is the greater strength of steel. The varieties which have been used in shipbuilding have had tensile strengths from 26 to 50 tons per square inch; the corresponding strength for good iron varying from 17 to 22 tons. Hence with steel it is possible to use thinner plates and bars than with iron, in order to secure a certain strength: such thinner plates, &c., being, of course, much lighter, since steel is only about

2 per cent. heavier than iron. Numerous steel ships have been built with reductions of one-fourth or one-third from the scantlings which would have been considered necessary had iron been used: in some cases the reductions in scantlings have even reached one-half. The savings in weights of hull effected by using steel are very considerable, ranging, it is said, from 30 to 50 per cent. of the weight which the hull would have if built of iron and made equally strong. What is thus saved on hull is available for carrying power.

Notwithstanding the greater lightness of steel ships, it is a fact that comparatively few have been built as yet, the majority of those constructed being designed for special services where extremely shallow draught or exceptional lightness of hull was requisite. Vessels for river service, like that illustrated by Figs. 105 and 106, pages 339, 340, are frequently built of steel; and steel was also largely employed in the blockade-runners built during the Civil War in America; but taking the twenty years from 1850 to 1870, it appears that over 3,600,000 tons of iron shipping were built, while only 27,000 tons of steel ships were constructed. And this is still nearly as true as it was in 1870; the steel ships are comparatively few. In the Royal Navy, although steel has been for many years past extensively used for internal frames, stringers, and strengthenings, no ship was constructed wholly of steel until 1875; and at the present time only two such ships, the Iris and Mercury, armed despatch vessels, are being built, and the steel used in their construction is of a special character, only recently produced in this country.

Hitherto two main objections have existed to the use of steel its greater cost, and the want of uniformity in its character, and in the qualities essential to successful resistance to the wear and tear of service. Its cost as compared with iron varies with the relative and absolute qualities of the materials used. Iron "ship-plates," for instance, of different qualities vary greatly in price; and so would steel plates.

But a good average is probably that which takes steel as about twice as costly as the iron in common use.

The more serious disadvantage of most of the varieties of steel is the want of uniformity in strength, ductility, and malleability. Steel plates made under similar conditions, and presumably of the same quality, have displayed, when tested, singular differences in their qualities; and the shipbuilder has consequently been without any assurance of safety, such as he may obtain with iron. In the Royal Navy the best quality of iron ship-plates used is required to have a tensile strength of from 18 to 22 tons per square inch; and this condition the manufacturer can fulfil almost certainly in association with defined conditions of malleability. But with steel, until recently, it was necessary to allow a wider margin; the tensile strength desired being about 33 tons per square inch, and an upper limit of 40 tons being fixed. Moreover, it was found that the manipulation of steel during the various processes of building-punching holes, bending, riveting, &c.-required much greater care than did the corresponding operations with iron. And although of greater tensile strength than iron, and probably also of greater compressive strength, steel appeared, on the whole, less suited to withstand shocks or rough usage; its greater hardness and less malleability making it liable to fracture under conditions when good iron would simply bulge. Hence it resulted that in the ships of the Royal Navy steel was used only for internal strengthening, the outer bottom plating being of iron.

For many years the subject remained in much the same condition, notwithstanding frequent discussions and many attempts to produce steel of the quality and price suitable for shipbuilding. But during the last two years great

Full details of the use of steel in the Royal Navy prior to 1875, and tests illustrating the want of

uniformity in quality, will be found in Mr. Reed's Shipbuilding in Iron and Steel.

66

progress has been made both in France and England, and a so called "mild steel" has been produced free from the objectionable features mentioned above. The manufacture of this material in England is mainly due to the efforts of Mr. Barnaby, Director of Naval Construction, who had previously conducted most of the experiments on steel made in the Royal Dockyards, and done much to develop the use of the material.* "Mild steel" or "homogeneous iron" has a tensile strength of from 26 to 30 tons per square inch, being about 25 per cent. stronger than the best iron shipplates it is very ductile and malleable, and will not take a 'temper" even when heated to a cherry-red and plunged into nearly cold water. It will withstand all the operations of the shipyard quite as well as, if not better than, wrought iron; and its cost is not great as compared with the highclass iron used in the ships of the Royal Navy. In no respect is it inferior to iron, and in very many it is much superior. For equal strengths, probably a reduction of onethird or one-fourth in scantlings and weight would be possible with this mild steel as compared with iron; but the Iris and Mercury have structural arrangements of so special a character that even, when completed, it will not be possible to compare fairly the ratio borne by their weight of hull to their displacement with that of iron ships of ordinary type. Speaking generally, it seems a fair assumption that, should this mild steel come into general use, it will be possible to build ships say 20 to 25 per cent. lighter than iron ships of equal strength; and although the outlay on the original construction may continue greater for some time, until the process of manufacture has become better known, this addition to the carrying-power will be a permanent advantage, and may more than recoup the additional outlay on construction during the period of service. Take, for example, a

* See Mr. Barnaby's paper on this subject in the Transactions of the Institution of Naval Architects

for 1875; and its sequel in Mr. Riley's paper of the following year.