Page:Elementary Text-book of Physics (Anthony, 1897).djvu/245

§ 231] to explain most of the laws of gases, though to explain others a rotation or something equivalent to it must be assumed.

The characteristics of the molecular motion assumed in the kinetic theory may be best explained by considering the motion in a gas. Let us suppose that a very large number of material particles is distributed uniformly throughout the region contained within a closed vessel, and that velocities are given to these molecules at a certain instant in various directions. If we further suppose that these molecules act on each other only by collision or by forces which are effective only when two molecules are extremely near each other, it is plain that the paths of the molecules thus assumed will in general be short straight lines, changing in direction with every encounter between two molecules. It is also evident that, no matter what the initial velocities were, they will not be maintained for any length of time, but that the velocity of any one molecule will change at each encounter, and that the velocities of the molecules in the mass will speedily acquire values ranging from zero to a very great or practically infinite velocity. It is also plain that very few molecules will possess these extreme velocities at any one time, and that most of them will possess velocities which do not depart far from a certain mean. An obvious condition to which the velocities must conform is that the kinetic energy of all the molecules in the mass must remain the same at all times, it being assumed that no energy enters the mass from without and that the encounters do not involve the loss of kinetic energy. It was shown by Clausius, and afterwards more rigorously by Maxwell, that the distribution of velocity among the molecules may be deduced by the theory of probabilities. Some idea of it may be got from the distribution of shots iu a target; if a rifleman shoot at a target a great many times, and if the distance of the shots from the centre of the bull's eye be measured, these distances conform to the same law of distribution. It is clearly infinitely improbable that any one of the shots will strike the exact centre of the bull's eye, and also infinitely improbable that any one will be sent directly away from the target, and it is very highly