The main problem to be faced when explaining the a.c. electroluminescence (EL) in ZnS is the mechanism which allows electrons to be excited from the valence to the conduction band. In fact, the spectral features of the emitted EL light are the same as in photoor cathodoluminescence (1). This circumstance proves that, when an electron-hole pair is obtained, the radiative recombination takes place in the same way for all the ZnS luminescence phenomena. Former EL theories invoked, to explain the electron excitation, the impact ionization process. This, in practice, means to consider the EL as a sort of internal cathodoluminescence. According to ZALM, the impinging electrons are tunnel-emitted through barriers lying at the contact between ZnS and a Cu2S conducting phase, present at the crystal surface (2). These theories, even if modified (3), ceased to hold value when the accurate observation of the light-emitting regions inside the ZnS crystals evidenced the existence of the so-called comets (4). A new model was then proposed by FlSCI~ER (5), starting again from the accepted presence of a Cu2S phase, but assuming the electron-hole pairs to be produced by thermal excitation through the CthS forbidden band. As we are dealing ~ i th an experiment which contradicts the Fischer model, let us summarize its essential features. The CIhS precipitates are actually believed to be shaped like long and uninterrupted wires. When the electrid field is applied, at the opposite extremities of these precipitates, electrons and holes, coming, respectively, from the conduction and the valence band of CthS, are injected, according to the field direction, in the bulk lattice. This process, referred to as double injection, takes place by tunneling through potential