pressure, and the proportions of such increase which result from heating a gas. Similarly, Charles's important law, that the volume of a given mass of gas under a constant pressure varies directly as its temperature, follows obviously from the hypothesis.
Priestley was the first to remark that gases diffuse through each other. This fact is familiarly illustrated by the passage of odorous gases through the atmosphere. If a bottle of ether is opened in a room, its vapor diffuses through the air, and its presence is soon recognized by the sense of smell. In this case, the ether-molecules may be figured as issuing from the bottle with great velocity; and, if their course were not interrupted by striking against the molecules of the air, the room would be instantaneously permeated by their odor. But the molecular particles of both air and ether are so inconceivably numerous, that they can not avoid striking one another frequently in their flight. Every time a collision occurs between two molecules, the paths of both are changed; and the course of each is so continually altered that it is a long time in making any great progress from the point at which it set out, notwithstanding its great velocity.
We must next inquire how these velocities are measured, and what is their amount. We have seen that the pressure exerted by a gas is due to what may be appropriately called the molecular bombardment of the walls of its containing vessel; and, knowing this pressure, we can calculate the velocity of the projectiles, if we can ascertain their weight, just as we can estimate the speed of a bullet when its weight and mechanical effect are known. Now, a cubic centimetre of hydrogen at a pressure of one atmosphere weighs about one thousandth part of a gramme; we have, therefore, to find at what rate this mass must move—whether altogether or in separate molecules makes no difference—to produce this pressure on the sides of a cubic centimetre. The result gives six thousand feet per second as the velocity of the molecule of hydrogen, while in other gases the speed is much less.
The question of molecular weights brings us face to face with the chemical aspect of the hypothesis; and we have now to examine the support which is given to it by chemical phenomena, and show how wonderfully these are correlated with the physical proofs. Bearing in mind the distinction between physical and chemical changes, we know that we can make a mixture of finely divided sulphur and iron, for example, in any proportion. But these bodies when heated combine chemically to form a new substance called sulphide of iron; and the two classes of products exhibit great differences, which are indicated by a most remarkable characteristic. Chemical combination, unlike mechanical mixture, always takes place in certain definite proportions. Thus, fifty-six grains of iron combine with exactly thirty-two grains of sulphur; and, if there is any excess of either substance, it remains uncombined. This principle is known as the law of definite combining proportions, and the atomic theory, which, in one shape or an-
