When a liquid is heated, the attraction of the molecules operates against the action of the heat, which tends to tear them asunder. At a certain point the force of heat triumphs, the cohesion is overcome, and the liquid boils. But supposing we assist the attraction of the molecules by applying an external pressure, the difficulty of tearing them asunder will be increased; more heat will be required for this purpose; and hence we say that the _boiling point_ of the liquid has been _elevated_ by the pressure. [Sidenote: INFLUENCE OF PRESSURE ON FUSING POINT.] If molten sulphur be poured into a bullet-mould, it will be found on cooling to contract, so as to leave a large hollow space in the middle of each sphere. Cast musket-bullets are thus always found to possess a small cavity within them produced by the contraction of the lead. Conceive the bullet placed within its mould and the latter heated; to produce fusion it is necessary that the sulphur or the lead should _swell_. Here, as in the case of the heated water, the tendency to expand is opposed by the attraction of the molecules; with a certain amount of heat however this attraction is overcome and the solid _melts_. But suppose we assist the molecular attraction by a suitable force applied externally, a greater amount of heat than before will be necessary to tear them asunder; and hence we say that the _fusing point_ has been _elevated_ by the pressure. This fact has been experimentally established by Messrs. Hopkins and Fairbairn, who applied to spermaceti and other substances pressures so great as to raise their points of fusion a considerable number of degrees. Let us now consider the case of the metal bismuth. If the molten metal be poured into a bullet-mould it will _expand_ on solidifying. I have myself filled a strong cast-iron bottle with the metal, and found its expansion on cooling sufficiently great to split the bottle from neck to bottom. Hence, in order to fuse the bismuth the substance must _contract_; and it is manifest that an external pressure which tends to squeeze the molecules more closely together here _assists_ the heat instead of opposing it. Hence, to fuse bismuth under great pressure, a less amount of heat will be required than when the pressure is removed; or, in other words, the fusing point of bismuth is _lowered_ by the pressure. Now, in passing from the solid to the liquid state, _ice_, like bismuth, contracts, and if the contraction be promoted by external pressure, as shown by the Messrs. Thomson, a less amount of heat suffices to liquefy it. [Sidenote: EXPERIMENTS.]