What Makes a Thermometer Possible
The zeroth law looks trivial and is not: if A is in thermal equilibrium with C, and B is in thermal equilibrium with C, then A and B are in thermal equilibrium with each other. Thermal equilibrium is transitive.
That transitivity is exactly what licenses temperature. It means all mutually equilibrated systems share one common property, so a single number can label the equivalence class, and it means a thermometer works, since C can be a small instrument brought to each body in turn.
Empirical scales rely on some property that varies with heat: the length of a mercury column, a resistance, a gas pressure. These agree at their calibration points and drift between them, because different substances vary differently. The ideal gas scale is better behaved: at low density all gases converge on the same reading, giving from . The thermodynamic scale, defined later from Carnot efficiencies, turns out to coincide with it while depending on no substance whatever.
A substance's response to changes is captured by three coefficients:
the thermal expansivity, the isothermal compressibility, and the pressure coefficient. The minus sign in makes it positive, since raising the pressure always reduces the volume, a substance with negative would be mechanically unstable and would collapse.
The three are not independent. Because any two of , , fix the third, the chain rule ties them:
Common pitfall: treating the zeroth law as an empty statement. Transitivity is a genuine empirical fact about nature, not a logical necessity, and without it "temperature" would not be a well-defined property at all, since two bodies matching a third could still disagree with each other.