THE point which Mr. Browne (p. 102) raises is covered by Voigt (“Thermodynamik,” vol. ii., § 69, pp. 209 et seq.). Ordinary diffusion of two gases at equal pressure and temperature is an irreversible process involving loss of available energy, but when the diffusion takes place through porous membranes this available energy can be utilised in a greater or less degree in the form of work, and this is the case in Mr. Browne's experiment. By introducing the conception of “semi-permeable partitions,” Voigt obtains a reversible method of mixing or separating gases. In this case the partial pressure of the mixture is equal to the sum of the partial pressures of the components. In ordinary diffusion the volume of the mixture is equal to the sum of the volumes of the components. The work of expansion from the former to the latter final state can be utilised if a reversible transformation is employed. It is lost in the case of ordinary diffusion. An equal amount of work must, however, be supplied from without to separate the gases. The results are fully in accordance with the second law.