The electrical double layer (EDL), which arises at a metal-electrolyte interface, does not-only determine nost of the equilibrium properties of this interface (surface tension, zero-charge point potentials, capacitance, etc.), but also controls, to a large extent, the kinetics of electrode reactions. That is why the concept of the equilibrium EDL underlies almost an the branches of electrochemistry: corrosion and passivation of metals, catalysis and adsorption, galvanoplasties. The double layer consists of two parts: the diffuse part and the dense part. The diffuse part is of minor importance in all electrochenical processes. Even in dilute solutions the potential drop across this part is about one tenth of that across the dense part; in concentrated solutions the diffuse part virtually does not exist. However, the diffuse part of EDL has been studied to a higher degree than the dense part. At present, not only has the mechanism of the diffuse part formation been understood, but a reliable theory has also been developed that describes this part (the latter mainly due to the work of Gouy and Chapman). The situation is, however, much worse for the dense part; there is not even a comolete understanding of the factors responsible for the formation of this part. At least, there has been no common opinion in the literature concerning this point. Most of the authors stick to the molecular capacitor (MC) model proposed by Helmholtz. According to this model, the dense part of EDL is formed by a monolayer of solvent molecules adsorbed at the metal surface, the electrolyte ions being partially or comoletely “prohibited” from penetrating into this layer. The properties of the monolayer should, apparently, be determined by three different interactions: 1) between the metal and electrolyte ions; 2) between the metal and solvent molecules; and 3) between solvent molecules themselves. In any consistent theory of this phenomenon, all three interactions should be taken into account equally. At present, however, such a theory cannot be developed because of technical difficulties. One is, therefore, forced to construct various graphic models of the MC: the reliability and “heuristic power” of these models are very poor. Thus, most of the properties of the dense part of EDL (the mechanism of the formation of the zero-charge point potentials, surface tension, etc.) have not been explained within the framework of the MC theory. As early as in the time of Frenkel and Rice, another model of the dense part of EDL was known in which electrons of the metal, rather than solvent molecules, were assumed to play the major role. For a long time, this model did not attract the attention of electrochemists and was, in fact, completely forgotten. Quite recently, however, in connection with Ref. 1–10, in which the model was actually rediscovered, there has been a surge of interest in this model. Following Ref. 7, we shall call it the electronic capacitor (EC) model. At present, the concept of EC has enafcled a wide range of phenomena to be described not only qualitatively, but quantitativeliy as well (ref. 7–10). However, the theory developed in these papers is based on an approximation, rather than on straight-forward solution of the Schrödinger equation. It is always possible, therefore, to assert that the agreement between theory and experiment is accidental, and that a more rigorous theory of the molecular capacitor would better expalin experimental results. A thorough analysis of experimental data is needed to decide which of the two models (MC or EC) is “closer to the truth”. On the other hand, such an analysis is very difficult to perform without preliminary systematization of experimental data. The present review precisely deals with systematization and analysis of experimental results. At first glance, it would hardly seem reasonable to conduct a study with the sole purpose of choosing one model or another. It should not, however, be forgotten that electrochemistry is based entirely on the concept of EDL, so that any fact and any theory in the field has to be verified especially carefully. Furthermore, a simple systematization of experimental data reveals new laws and facts, which in itself is also interesting.