400 REPORT — 1882. 



breaking point was 25'2, which was higher than in the original bar. The 

 elongation remained nearly the same, being 34 per cent., so that the 

 mere heating to a welding temperature without disturbing the particles 

 by hammering had no serious detrimental effect. 



I then took a piece of the steel in the restored condition, and after 

 heating it to the welding point, delivered upon it in that state a single 

 blow of a hammer sufficient to crush it into half its thickness. The 

 result was that the flattened piece divided into fissures all round the 

 edges. For the purpose of comparison, I took a piece of wrought iron 

 selected at random from a scrap heap, and treated it in exactly the same 

 manner. The result was that the iron bore the blow, flattening it to the 

 same extent as the steel, without showing the slightest fissure on the 

 edges. These two pieces are now on the table, and it is impossible to 

 examine them without perceiving that the steel, though difiering so little 

 from iron in the amount of its carbonisation, was yet, when heated to 

 the welding point, in a state of friability, while the iron remained per- 

 fectly plastic. The conclusion was thus confirmed that it is the dis- 

 turbance of the particles in this friable state, and not the mere heating, 

 which exercises the injurious effect in the welding process. 



I was not surprised to find that the coil itself had derived no benefit 

 from the tempering, because although steel, so low in carbon as this 

 sample, is considerably improved by tempering when the piece subjected 

 to the process is of small dimensions, yet when the bulk is considerable 

 the cooling in the oil is not sufiBciently rapid to produce any decided 

 effect. 



My next experiments were made upon a block of gun steel contain- 

 ing '^4 per cent, of carbon, which had been rejected on account of its 

 deficient tensile strength. A test piece cut from this block as received 

 from the maker began to stretch permanently at 11 tons per square inch, 

 breaking at 29*4 tons per square inch, with an elongation of 24'25 per 

 cent. ; but a piece of the same steel drawn out under the hammer at a red 

 heat from a thickness of 5 inches to a thickness of 1^ inches, resisted 

 19 tons instead of 11 without stretch, and a breaking strain of 33'2 tons 

 against 29"4, with an elongation of 27*5 against 24"25. A piece of the 

 same steel 5 inches long by 4 inches thick, having been tempered in oil, 

 gave a test piece which showed a further increase of strength, with little 

 diminution of ductility. It began to stretch at 23 tons, breaking at 

 36'5 tons, and elongating 21 per cent. Various attempts were made to 

 weld this steel in a pile of slabs, but it was found impossible to make 

 sound joints, and the steel was even more deteriorated than had been the 

 case with the previous sample ; but a piece of this material spoiled in the 

 attempt to weld it, having been drawn out into a bar of an inch square, 

 proved to be far stronger than in the original state. It stood 24 tons per 

 square inch before stretching against 11 tons in the previous untempered 

 state, and 33"6 tons before breaking against 29'4, but the elongation was 

 reduced from 2425 per cent, to 15 per cent. The fracture in this case 

 was of the same character as in the original test piece, and showed no 

 indication whatever of the previous injury it had sustained by the attempt 

 to weld it, A piece of the same block heated to a welding temperature, 

 and allowed to cool without hammering, gave a test piece which, so far 

 from showing any injury by the heating, resisted a considerably higher 

 strain than the sample taken from the block as it came from the maker. 

 Its stretching point was 16 tons per square inch, its breaking point 332 



