402 EEPOET— 1882. 



Frederick Bramwell, that we should eventually be able to produce in this 

 manner continuous umvelded cylinders for boiler- making purposes. Steel 

 cylinders thus made and tempered in oil will be in a highly favourable 

 condition for the construction of ordnance, but in order to make them 

 available for longitudinal as well as for lateral sti'engths it will be essential 

 to have them in much greater widths than existing machines are com- 

 petent to produce. 



All ductile metals derive additional strength bv beinsj stretched, but 

 steel does so in a pre-eminent degree. Roughly speaking its modulus of 

 elasticity may be taken as equal to xo'iTo'''Ii of an inch per foot for every 

 ton per square inch of tensional strain. This measure of elasticity applies 

 equally or nearly equally to all kinds of steel, but the range of elasticity 

 becomes greater as the strength is increased. Thus steel that will bear 

 20 tons without permanent stretch will retract y-|}yths of an inch per foot 

 of length on being released from its load, while steel that will bear 40 

 tons without permanent stretch will recover yj^^ths of an inch per foot on 

 the removal of the strain. So also if the weaker specimen which recovers 

 only yf7Qths of an inch be stretched to a point at which it will sustain 40 

 tons per square inch, it will be in exactly the same condition in regard to 

 elasticity as the stronger specimen which bore that load in the first 

 instance — that is to say, its range of elasticity will be doubled. This is a 

 very valuable quality, enabling the steel to gather strength as it yields to 

 an imprudent increase of load. As an illustration of the extraordinary 

 strengthening effect of stretch upon mild steel I may mention that a 

 sample of the steel taken from the welded coil to which I have adverted, 

 and which in its original state showed a tensile strength very slightly 

 exceeding that of wrought iron, sustained a load of nearly 85 tons per 

 square inch measured on the attenuated section of fracture. 



But much as steel gains in strength by the process of rolling it gains 

 still more by that of wire-drawing. No form of steel is comparable in 

 respect of strength and toughness to that which has been drawn into 

 the form of wire or riband, and in the case of its application in that form 

 to the strengthening of a cylinder it has the additional advantage of 

 admitting of being laid on with a more favourable adjustment of tension 

 than is practicable with a solid hoop of considerable thickness. Bat, 

 even with wire, the best tensional condition for giving strength to a 

 cylinder can only be approximately attained, owing to the fact, which is 

 commonly overlooked, that, in bending a wire over a cylinder, it is im- 

 possible to give the proper degree of stretch to both of its sides. The 

 outer side having a larger circle to describe must necessarily undergo 

 greater elongation than the inside, and in fact, unless the wire be laid on 

 at a far higher strain than would be necessary or beneficial in the case of 

 rings, the inside, acting as a fulcrum to stretch the outside, will assume a 

 state of compression which can only be taken ofi" by expanding the 

 cylinder after the wire has been laid on. The thinner the wire the less 

 will this disadvantage be felt, and for this reason a given area of section 

 is much better in the flat or riband form than either round or square. 

 Great additional strength is given to steel wire by tempering in man- 

 ganese, and the highest strength is attained by passing the wire repeatedly 

 through the die as a final operation after the tempering process. The effect 

 of this treatment is to put a very hard skin upon the wire, which, though 

 greatly adding to the strength, is very unfavourable for bending, and a 

 very slight injury to the surface greatly conduces to fracture. Ductility 



