tive character than that which teaches us the composition of gases.
The movements of the gaseous molecules are, however, so wonderful that it is hardly surprising that those who are invited to believe these things should demand satisfactory evidence as to the existence of phenomena, which from the nature of the case seem to lie out of the reach of direct inspection by the senses. The methods by which our knowledge of the constitution of matter has been obtained, is by reasoning from the phenomena which our senses reveal to the more refined and supersensible conceptions of molecular physics. I could not undertake in a work of this description to give any complete account of the evidence by which these remarkable doctrines are sustained. I will therefore only indicate one of the main lines of argument, by which the necessity for the belief of an important part of the molecular doctrine of gases has been satisfactorily demonstrated.
There are no more fundamental properties possessed by a gas than those which are connected with the pressure which it exerts. If included within a closed chamber, a gas presses against the walls of that chamber. In this case there is a notable contrast between the behaviour of solids, of liquids, and of gases. Suppose, for instance, that a cubical box be fitted with a block of iron which fills it exactly. The metal presses, of course, on the bottom of the box, but not on all its sides. If the iron be removed and the cubical chamber be now filled with water, the liquid presses as before on the bottom with its entire weight, but in addition there will be a lateral pressure exerted by the liquid against the sides of the vessel. Even in this case, however, there is in general
