To the chemical student the book had, and still has, many features of interest. It contains a number of important facts, based on original and fairly accurate observation. In the arrangement of these facts “I have been guided as much as possible,” says their author, “by obvious and simple analogies only. Hence, I have seldom entered into theoretical discussions, particularly concerning light, heat, and other agents, which are known only by isolated effects. Early experience has taught me the folly of hasty generalisation.” The work is divided into four main sections. The first chiefly relates to the production of nitrous oxide, and the analysis of nitrous gas and nitrous acid. He minutely studies the mode of decomposition of ammonium nitrate (first observed by Berthollet), and shows that it is an endothermic phenomenon, varying in character with the temperature and manner of heating. He is thus led to offer the following _Speculations on the Decompositions of Nitrate of Ammonia_:-- “All the phenomena of chemistry concur in proving that the affinity of one body, A, for another, B, is not destroyed by its combination with a third, C, but only modified; either by condensation or expansion, or by the attraction of C for B. On this principle the attraction of compound bodies for each other must be resolved into the reciprocal attractions of their constituents, and consequently the changes produced in them by variations of temperature explained from the alterations produced in the attractions of those constituents.”