The influence of heat on reversible reactions is particularly instructive. If, for instance, a compound which is capable of being reproduced from its products of decomposition be heated up to the temperature at which decomposition begins, the decomposition of a mass of the substance contained in a definite volume is not immediately completed. Only a certain fraction of the substance is decomposed, the other portion remaining unchanged, and if the temperature be raised, the quantity of the substance decomposed increases; furthermore, for a given volume, the ratio between the part decomposed and the part unaltered corresponds with each definite rise in temperature until it reaches that at which the compound is entirely decomposed. This partial decomposition under the influence of heat is called _dissociation_. It is possible to distinguish between the temperatures at which dissociation begins and ends. Should dissociation proceed at a certain temperature, yet should the product or products of decomposition not remain in contact with the still undecomposed portion of the compound, then decomposition will go on to the end. Thus limestone is decomposed in a limekiln into lime and carbonic anhydride, because the latter is carried off by the draught of the furnace. But if a certain mass of limestone be enclosed in a definite volume--for instance, in a gun barrel--which is then sealed up, and heated to redness, then, as the carbonic anhydride cannot escape, a certain proportion only of the limestone will be decomposed for every increment of heat (rise in temperature) higher than that at which dissociation begins. Decomposition will cease when the carbonic anhydride evolved presents a maximum _dissociation pressure_ corresponding with each rise in temperature. If the pressure be increased by increasing the quantity of gas, then combination begins afresh; if the pressure be diminished decomposition will recommence. Decomposition in this case is exactly similar to evaporation; if the steam given off by evaporation cannot escape, its pressure will reach a maximum corresponding with the given temperature, and then evaporation will cease. Should steam be added it will be condensed in the liquid; if its quantity be diminished--_i.e._ if the pressure be lessened, the temperature being constant--then evaporation will go on. We shall afterwards discuss more fully these phenomena of dissociation, which were first discovered by Henri St. Claire Deville. We will only remark that the products of decomposition re-combine with greater facility the nearer their temperature is to that at which dissociation begins, or, in other words, that the initial temperature of dissociation is near to the initial temperature of combination.