When a solid is dissolved in a liquid, as, for example, sugar in water, the particles of sugar — its molecules — must obviously be separated from each other to a greater or less extent, according as much or little water be added. And it has been noticed that if the sugar be placed in the water, and not stirred up, the sugar will dissolve at the bottom of the vessel, and the strong solution of sugar will slowly mix up with the upper layer of water, and in course of time be equally distributed through the water; just as a heavy gas, such as carbonic acid gas, if placed in an open jar, will gradually escape into the lighter air above it. This process of mixing of two liquids or gases is termed ^Mdiffusion." It is now generally held that the pressure of a gas on the walls of the vessel which contains it is produced by the impacts of its molecules against the walls; and as the molecules are extremely numerous, and in a state of very rapid motion, they escape from an open vessel; so that even a heavy gas, like carbon dioxide, will escape upwards into a lighter gas ; and similarly, a light gas, like hydrogen, will escape downwards into a heavy gas, owing to the unceasing motion of its molecules. The fact that the molecules of sugar, which, by the way, becomes itself a liquid when dissolved in water, travel upwards, and diffuse through the lighter water, shows that they too are in motion; but the slowness of the diffusion, compared with the rate of diffusion of a gas, indicates that their motion is much impeded by the molecules of water, with which they are constantly coming into collision. And just as the motion of the molecules of a gas produces pressure, and causes the gas to escape through an opening, so the motion of the molecules of sugar, which causes them to rise through water against the attraction of the earth, may be taken to imply that they also exert a kind of pressure. But the molecules of water, with which the molecules of sugar are mixing, must also be held to exert pressure of the same kind, since they disperse themselves through the molecules of sugar. How is it possible to distinguish the pressure due to the sugar from that due to the water ? A parallel case with gases will help us to reply to this question.