PROCEEDINGS. Ixvii 



As it is bub a few years, not more than from 30 to 50 years, or less, since some 

 of the more important of these inventions came into usefuhiess, it is difficult for 

 us to realize how much they have contributed to possibilities that now seem to be 

 so easy of accomplishment. 



For a number of years after the art of rolling became known, only two rolls 

 arranged one above the other were used, and in consequence the bar or sheet of 

 metal, after passing through the rolls in one direction, had to be transferred 

 again to the side from which it had passed through, in order to make another 

 pass. This objectionable loss of time was overcome by arranging three rolls 

 superposed, one above another, called three high rolls, and this permitted the 

 passage of the bar or sheet of metal from either side between one pair of the 

 three, no matter on which side the sheet or bar might happen to be. This 

 arrangement not only saved time but caused the metal acted upon to be elonga- 

 ted alternately from each end instead of being acted upon alw ays in one direc- 

 tion. 



Another improvement was also made upon t'he rolls by which while the metal 

 was receiving the effect of one pair of rolls in one direction another pair of rolls 

 was at the same time acting upon it in a direction perpendicular to the plane of 

 action of the initial pair. This principle was only applicable to the rolling of 

 quarilateral or rectangular forms of cross section. 



But there was one serious defect which existed in and was absolutely insepar- 

 able from all malleable iron which was fashioned into merchantable shapes by 

 the rolling process, and that was the effect produced by the process of rolling 

 upon iron which had been welded up in its mass in the process of production. 

 The effe'.!t was, that there was organized and maintained by the rolling process a 

 well defined fibrous structure of the metal, which, while it did not diminish its 

 resistance in a direction parallel to the fibre, j'et it very materially interfered 

 with its strength in a direction at right angles to the fibre, and although for 

 many purposes this did not so much matter, yet there were other uses in which 

 this feature was a serious difficulty. This led to investigations with a view to 

 obtain a homogenous metal, and it was easily comprehended, at this stage in the 

 metal's history, that this condition required that such a metal slioujd be produced 

 by a process of fusioir and pouring instead of by the method of M-eldiug thereto- 

 fore practised. Several processes were devised with a view to obtain this much 

 desired homogeneity : some proposed to decarbonize the molten pig iron by 

 mixing with it as it poured from the furnace, powdered and highly oxidized ore, 

 to operate in reducing the portion of carbon to such a small degree that its 

 presence would not be objectionable, a proportion of one quarter of one per cent, 

 having been found to be no serious objection unless the iron is required for 

 smithing or welding purposes. 



At this stage it was proposed by Mr. Bessemer to put the molten metal into a 

 suitable vessel and blow air under pressure through the mass in order to burn up 

 the carbon by admixture with oxygen ; other processes were made passible by 

 the discovery by Mr. afterwards Sir \Vm. Siemens, of a means of obtaining much 

 more intense heat in furnaces by using the waste heat of a present conduct of 

 operations to heat up a nest of refractory brickwork, through which in turn the 



