Theories of the metal-electrolyte interface are reviewed with particular attention to the prediction of thermodynamic quantities. A comparison of the different theories is made using a range of experimental results on Hg-aqueous electrolyte systems. Certain widespread failings of the models are revealed. These are identi- fied as being the results of a neglect of solvent structure and of chemical interactions. Next a solvent-free metal-molten salt system is investigated using a model based on the MSA. Thermodynamic properties, parti- cularly the capacitance and the potential at the point of zero charge and their variation with temperature, are calculated. The magnitudes for a range of ionic species are in agreement with experiments for Pb-alkali halide systems, but the (relatively weak) temperature dependence is not predicted correctly. The success of this application of the MSA compared to applications to aqueous systems shows the great importance of the solvent in determining the properties of these systems. To attempt to improve on the MSA model a new model based on the EXP approximation is developed. The predicted trend of the temperature dependence of capacitance is now given correctly but in other respects the predictions are worse than those of the MSA. In our first MSA model the metal is treated as a hard wall. A more realistic representation is now introduced which incorporates electron spill-out. By a phenomenological analysis the extent of spill-out into the electrolyte is estimated. No direct measurement of this quantity has been made. It is consistent with theoretical estimates made for aqueous systems when allowance ,is made for the effect of the solvent.