550 Proceedings of the Royal Society 
floating on the molten iron as frequently witnessed in foundry 
ladles. 
Some years ago the author made a number of experiments with 
pieces of cast iron of different shapes and sizes, which were placed 
on the surface of a large ladleful of molten cast iron, the results 
obtained being as follows : — 
Several pieces of pig iron were first tried ; these at first sank and 
a rush of hot metal took place upwards, but after a few seconds they 
rose to the surface, and floated 'with very little of their hulk 
above the molten metal. 
A piece of flattish irregularly-shaped cast iron floated with a 
small portion only of one of its corners appearing above the surface. 
Pieces of flat cast iron test-bars were carefully placed on the surface 
(the latter being well skimmed), these floated without going below 
the surface. One of the test-bar pieces, which was put in end on, 
kept in this position for a few seconds with its end above the 
surface, the other end then came up and the piece floated on its 
flat side. 
In another experiment, a cast iron ball, about three inches in 
diameter, was lowered by means of a wire on the molten surface ; 
the ball at first disappeared, but in a few seconds rose and floated 
with a small portion of its surface exposed. On being raised out of 
the molten metal it showed a red glow on the lower part, and when 
again lowered it did not sink, but floated with a larger portion above 
the surface than before. 
Some pieces of steel rails were put into a Siemens steel-making 
furnace filled w r ith molten metal; these pieces at first disappeared, 
but afterwards rose to the surface. Some other pieces, heated 
red-hot , were now put in ; these floated without sinking, and showed 
a little of their bulk above the surface. 
2. Different views are held as to the behaviour of cast iron when 
passing from the molten to the hot solid state, and finally thereafter 
to the cold (or ordinary temperature) state. 
Some hold that the molten metal on solidifying expands, as water 
does when passing into ice, and that it retains this expansion to such 
an extent that the cold solid is specifically lighter than the liquid 
metal. Others hold that no such expansion takes place, but that 
through cooling the cooled solid becomes specifically heavier than 
