the expansive force of heat. An iron platform was built under the socket, and a brisk fire made upon it. The socket soon expanded, and the pin was now readily knocked out by the battering-ram, just as the stopper of the bottle is easily removed when the neck is heated. The walls of a very large building in Paris, which had bulged out and were in danger of falling, were restored to their upright position by the expansion of heat. [Illustration: Fig. 193.] It was done in this way: Long rods of iron were run through the walls after the plan represented in Fig. 193 (p. 213), their ends being made with a screw-thread, with nuts fitted to them. The rods marked _a_ were first heated, and as they lengthened the nuts were screwed up tight to the walls. On cooling, their contraction would of course draw the walls together. The other bars, _b_, were now heated and managed in the same way. The one set, you see, were made to hold on by their nuts to what had already been gained, while the other were expanding. By many repetitions of this process the walls were righted and the building saved. The same mode has been adopted successfully in other cases of a similar character. 275. =Expansion in Liquids.=--Liquids are expanded by heat more than solids are. But they are very unequally expanded by it. Thus water is expanded more than twice as much as mercury, and alcohol six times as much. We have a frequent example of the expansion of water by heat in our kitchens. If the tea-kettle be put over the fire filled to the brim, it will run over long before the water begins to boil. All liquids occupy more space in summer than in winter, and in the former case weigh less--that is, have less of real substance in them than in the latter. If, therefore, alcohol, or oil, or molasses be bought by the gallon in winter and sold in summer, there will be a profit afforded by the expansion. Twenty gallons of alcohol in winter becomes twenty-one in mid-summer. [Illustration: Fig. 194.] The influence of the expansion of heat upon the specific gravity of liquids may be very prettily shown by the following experiment: Let some little bits of amber--a substance which is nearly of the same specific gravity with water--be thrown into water in a glass vessel, and let the water be heated, as represented in Fig. 194, by a spirit-lamp. That portion of the water which is heated passes upward because it is made specifically lighter, and colder water continually comes down to take its place. The upward and downward currents are as indicated by the arrows, the upward passing up in the middle, the downward coming down at the sides. This will be made manifest by the little bits of amber.