THERMODYNAMICS IS the most empirical of the physical sciences. We do not mean that it is a mere mass of coordinated facts held together by their observational validity. Indeed the pattern of systems, observables, and states is impressed upon thermodynamics with extraordinary firmness, and it is by most meticulous attention to these methodical elements that this science proceeds. Its empirical character comes from the fact that it deals with a large class of systems whose laws are not identical in detail or derivable from some common source, but need to be fashioned after observations. Or, to put it in another way: the laws of thermodynamics1 do not enjoy the logical generality of the laws of electrodynamics. Outwardly this is evident in their form; instead of being differential equations, they are primitive relations between the variables of state, a fact which becomes significant when we recall that differential equations are generators of classes of primitive equations. The laws of thermodynamics, being particular primitive equations, are therefore of a lower order of generality than the laws of motion. These preliminary remarks will now be made more explicit and, we hope, more meaningful.
The systems of thermodynamics are ordinary bodies with clearly defined spatial boundaries, and our interest centers in their reaction to the various modes of heating them. Typical systems are gases, liquids, mixtures of gases and liquids enclosed in containers with fixed or variable volumes, and solid bodies.
We shall confine our attention for the most part to the simple____________________