Abstract
It is shown that the experimental results of Buijs and Choppin are not consistent with any thermodynamic model of liquid water based on three species of molecules, each with a sharp level of positive energy.