The classical theory of simple electrode processes predicts that the rate of electron transfer depends on the electrochemical potentials of the reactant and product 1. Since these species are present in the vicinity of a charged interface, the electrostatic potential experienced by the reactant at the reaction site is different from that in the bulk of solution 2. The double layer terms normally included in the rate equation involve the macroscopic or average potential, and thus ignore variations in potential in planes parallel to the electrode due to the discrete nature of charge in the solution 3'4. Discreteness-of-charge (self-atmosphere) effects have been cited by several authors in cases where the usual double layer corrections failed to account quantitatively for the observed kinetic results, some examples being the reduction of hydrogen ion 5'6, of zinc and gallium cations 7'8, and of polyvalent anions 9. Hale 1ยฐ has recently derived corrections to the usual double layer terms in order to account for self-atmosphere effects for the case that the reacting particle is in the diffuse layer in the transition state. However, there is some evidence that the reaction site is closer to the electrode in the inner layer 1~-~4. Parsons 12 has presented a model in which discreteness effects due to specifically adsorbed nonreacting ions are expressed as a change in the activity coefficient of the activated complex in the inner layer, and specific adsorption of the reactant is ignored. Krishtalik 14 has criticised this model, pointing out that when specific adsorption of non-reacting ions accelerates the reaction, one would still expect to observe significant specific adsorption of the reactant if the Boltzmann factor governing its concentration is solely electrostatic and there is no potential independent work of adsorption. Furthermore, he pointed out that if the classical theory is valid, then a change in the activity coefficient of the activated complex would result in corresponding changes for the reactant and product. In order to overcome this contradiction, Krishtalik postulated that both reactant and product can be present in the inner layer without being significantly adsorbed. In the light of existing experimental data, discreteness-of-charge effects must be considered in the development of an improved model for double layer effects in electrode kinetics. In the present paper the concepts of Parsons and Krishtalik are extended in this direction.