Both the existing theory as well as the measurements that have traditionally been used to describe the water relations of plants in thermodynamic terms are not consistent with standard thermodynamic theory. In this article, three main errors are identified: (1) the assumption that the internal energy of a system includes the gravitational potential energy, when it does not; (2) the assumption that water spontaneously moves along gradients of chemical potential, when the Gibbs theory predicts that it would spontaneously move across phase boundaries, so that the chemical potential would be equal in adjacent phases; and (3) the assumption that plants and soil are single-phase systems, when they are in fact multi-phase systems. Examples are used to demonstrate the above errors. Measurements which can be used to separate plant and soil systems into bulk phases (i.e. solid, liquid and gas), which have long been used by wood and soil scientists, are shown to be ideal for the purpose of describing plant and soil systems using thermodynamic methods.