Another phenomenon connected with solutions, which has received much attention, is that of osmotic pressure. A salt in solution exercises a definite pressure, quantitatively measurable, which is curiously analogous to the pressure exerted by gases. In a gas the molecules are widely separated, and move about with much freedom. In a very dilute solution the molecules of a salt are similarly separated, and are also comparatively free to move. The kinetic theory of gases, therefore, is now paralleled by a kinetic theory of solutions, founded by Van t’Hoff in 1887, which is now generally accepted. All the well-established laws connecting pressure, temperature, and volume among gases find their equivalents in the phenomena exhibited by solutions. In Avogadro’s law we learn that equal volumes of gases, under like conditions of temperature and pressure, contain equal numbers of molecules. According to the new generalizations, equal volumes of different solutions, _if they exert the same osmotic pressure_, also contain equal numbers of molecules. The parallelism is perfect. With these relations the freezing- and boiling-point phenomena are directly connected. But, both for gases and for solutions some apparent anomalies existed. Certain compounds, when vaporized, seemed not to conform to Avogadro’s law, and called for explanation. This proved to be simple, and was supplied by the fact that the anomalous compound, as such, did not exist as vapor, but was split up, _dissociated_, into other things. For instance, ammonium chloride, above a certain temperature, is decomposed into a mixture of two gases--hydrochloric acid and ammonia--which, on cooling, reunite and reproduce the original compound. Twice as much vapor as is required by theory, and specifically half as heavy, is produced by this transformation, which is only one of a large class, all well understood. In the case of solutions it was found that certain compounds, notably the acids, alkalies, and metallic salts, caused a depression of freezing point which was twice as great as ought to be expected. This fact was illuminated by the phenomena observed in gases, and soon it was seen that here too a splitting up of molecules, a true dissociation, occurred. These anomalous solutions, moreover, were electrolytes--that is, they conducted electricity and underwent electrolytic decompositions--while normal substances, especially solutions of carbon compounds, such as sugar, were not.