506 report— 1879. 



3. On the Changes of Volume in Iron when passing from the Liquid to the 

 Solid State, and an Instrument for observing the same. By T. Wrightson, 

 Memb. Inst. C.E., F.G.S. 



That there are some considerable changes in the density of iron in passing from 

 the solid to the liquid state, is best illustrated by observing the behaviour of a piece 

 of cold iron when thrown into a ladle of molten iron. After being thrown in, it rises 

 to the surface, and as it becomes heated continues rising out of the metal until a 

 considerable portion of its bulk is raised above the molten surface. It then 

 appears for a time to remain without further change of volume until it reaches the 

 melting point, when it rapidly subsides into the general mass. 



These phenomena appeared to the writer to be well worthy of examination ; the 

 more so because much has been written upon the apparent anomaly that, although 

 iron when cast in a mould contracts in all its lineal dimensions about one per cent., 

 and should therefore, when cold, have a specific gravity higher than that of liquid 

 metal, it nevertheless floats on the surface when thrown into the molten iron. 



All kinds of ingenious explanations have been hazarded to account for this, but 

 no one appears to have taken the precaution to ascertain whether the anomaly was 

 real or only apparent. 



In order to exhibit the changes of specific gravity during the passage of the iron 

 from the solid to the liquid state, it occurred to the writer to submerge a ball of 

 cold iron in a vessel of molten iron to a certain depth, and to connect this ball 

 by means of a rod of refractory material to a spring balance ; any expansion or con- 

 traction of the ball woidd cause a greater or lesser displacement of the liquid iron,, 

 and the variation of buoyancy produced by this could be read off in ounces on the 

 index of the spring balance. 



A spring balance, with a circular dial plate, was accordingly suspended on a 

 wooden framing immediately above a large vessel of molten iron. A rigid rod 

 weighing two or three pounds was fastened to the moving slide of the balance, and 

 to the end of this rod was attached the cast iron ball to be experimented upon. 

 Before fixing the ball, the position of the index of the balance was marked in pencil 

 on a sheet of paper, surrounding but not covering the circular face of the dial. The 

 ball then being attached, another pencil mark, farther round the dial, indicated the 

 whole weight of the ball and rod, and one indication subtracted from the other re- 

 presented the actual weight of the ball. 



Now if the specific gravity of the cold ball were exactly the same as that of the 

 hot metal, there would be no tendency when the ball was lowered into the metal 

 either to sink or swim, and in that case the pointer would travel back exactly to 

 the mark, showing the weight of the rod alone. If the ball were of higher spe- 

 cific gravity there would be a sinking effect which would prevent the pointer 

 arriving at this mark, the space it fell short representing this sinking effect. As the 

 ball expanded in volume it would displace more liquid metal, producing an upward 

 notation equal to the difference between the weight of the ball and the weight of 

 the fluid displaced, and which, so long as the ball continues of the same weight and 

 is not allowed to rise to the surface, can be read off in ounces on the dial plate 

 while the operation proceeds. 



By carrying out this plan the author obtained an exact register of the succes- 

 sive alterations in volume taking place in the ball, though hidden from sight below 

 the surface of the metal. 



Immediately the ball was immersed and held two inches below the surface, 

 an assistant called out intervals of seconds, while the writer marked on the paper 

 round the dial plate the corresponding positions as the movement of the pointer 

 took place. After obtaining indications of various sizes of ball, the results were 

 laid down graphically. 



In the diagrams thus obtained we have a complete record of the changes in 

 volume and specific gravity from the cold solid to the commencement of the molten 

 state. 



An examination of the results obtained shows that in all cases there is 

 a sinking effect when the ball is first submerged ; in a few seconds this disappears 



