This can be proved. Iron expands 0·001235 of its length for every 180° Fahr. Divided by 180 it gives us the expansion for 1°, which is 0·00000686, taking the difference of winter and summer at 70° Fahr. Multiply these together, and the result (0·00048620 of its length) by the number of yards in 400 miles, and we find our answer 338 yards. Expansion acts in solids and most liquids by the destruction of cohesion between the particles. Gases, however, having much less cohesion amid the particles, will expand far more under a given heat than either solids or liquids, and liquids expand more than solids for the same reason, and more rapidly at a high temperature than at a low one. We have spoken of expansion. We may give an instance in which the subsequent contraction of heated metal is useful. Walls sometimes get out of the perpendicular, and require pulling together. No force which can be conveniently applied would accomplish this so well as the cooling force due to the potential energy of iron. Rods are passed through the walls and braced up by nuts. The rods are then heated, and as they cool they contract and pull the walls with them. When glass is suddenly cooled, the inner skin, as it were, presses with great force against the cooled surface, but as it is quite tight no explosion can follow. But break the tail, or scratch it with a diamond, and the strain is taken off. The glass drop crumbles with the effect of the explosion, as in the cases of Prince Rupert’s drops, and the Bologna flasks; the continuity is broken, and pulverization results. But a very curious exception to the general laws of expansion is noticed in the case of nearly freezing water. We know water expands by heat, at first gradually, and then to an enormous extent in steam. But when cooling water, instead of getting more and more contracted, only contracts down to 39·2° Fahr., it then begins to expand, and at the moment it freezes into ice it expands very much—about one-twelfth of its volume, but according to Professor Huxley it weighs exactly the same, and the steam produced from that given quantity of water will weigh just exactly what the water and the ice produced by it weigh individually. At 39·2° Fahr. water is at its maximum density, or in other words, a vessel of a certain size will hold more water when it is at 39° Fahr. than at any other time. Whether the water be heated or cooled at this temperature, it _expands_ to the boiling or freezing point when it becomes steam or ice, as the case may be. Water, when heated, is lighter than cold water. You can prove this in filling a bath from two taps of hot and cold water at the same time. The cold falls to the bottom, and if you do not stir up the water when mixed you will have a hot surface and a cold foundation. The heat increases the volume of water, it becomes lighter, and comes uppermost.