ME. EOBEET MALLET ON VOLCANIC ENEEGY. 
197 
Barrow for Bessemer steel ingot-moulds), were employed as receptacles and measures 
for the liquid slag. 
The form of these and their dimensions are shown in Plate IX. fig. 4. Elat cast-iron 
base-plates. were provided for these, so that when standing thereon with the wider end 
of the cone downwards it formed a close vessel, with its mouth or top free. 
In the same vertical line at one side of one of these cones were bored partly through 
its thickness three holes, at top, mid height, and bottom, sloping downwards, to contain 
mercury or water, in which to insert a thermometer to ascertain the temperature of the 
mass of iron in the cone itself, the thickness of which, as seen by Plate IX. fig. 4 (No. 2), 
is great. The content of anyone of these cones exceeds 8000 cubic inches, or above 4'6 
cubic feet. 
149. At successive periods each of these cones was fixed perfectly upright upon its base- 
plate, the latter being upon a secure and level foundation, and placed close to one of the 
blast-furnaces, and conveniently posited so that the slag could be run directly from the 
furnace into the cone, with arrangements whereby the 'run of slag could be instantly 
stopped upon the cone being filled. 
The cone being at the temperature of the atmosphere at the place, about 51° Fahr., 
was then run up full of liquid slag, the run being stopped without difficulty precisely as 
the slag reached the level of the top or brim of the cone. 
The slag in every case ran as liquid as very hot cast iron, and formed a pure 
unbroken and flat surface on the top of the cone, with a slight rounded edge due to 
capillarity all round the circle. 
Means were taken that the stream of slag should come pure and unmixed with 
scoria; Szc. from the centre of the volume at the time contained in the furnace, which 
was for a short time previously “ dammed up ” to admit of its accumulation. 
150. The slag so filled in continued for some time liquid ; its top surface rapidly com- 
menced to descend below the brim of the cone as the liquid slag began to lose heat by 
conduction to the mass of iron in the cone, and so to contract and the cone itself to 
expand. 
Very soon, however, a solidified crust began to form over the top, the thickness of 
which was judged of by its resistance to penetration with a light pointed steel rod, and 
very soon after the before flat and level crust commenced to get hollow or concave 
towards the centre. This announced that a self-supporting crust had then formed also 
all round the cone, whose interior, however, was still liquid. This therefore marked 
the moment at which the true dimensions of the cast-iron cone itself must be known, 
as giving the true volume of the slag when consolidating at its surfaces. The 
dimensions of the cast-iron cones having been taken with accuracy when cold, i. e. at 
49°, 51°, and 52° Fahr. respectively, we only need the temperature of the cone of cast 
iron to fix its capacity at the instant of incipient consolidation of the slag within it. 
151. The cones were run up full of slag as rapidly as possible, the time of complete 
filling being from 4 to 8 minutes from commencement. 
2 D 2 
