Existing theories of ‐powder‐EL are shown to be unable to explain the new findings of electroluminescent striations or lines described in Part I of this work. Several new hypotheses are discussed. Among these, a model seems to be likely which is based on simultaneous bipolar field‐emission from opposite ends of conducting imperfection lines into the insulating crystallite, with trapping of injected holes in activator centers, and recombination with mobile electrons at field‐reversal. This model, which is shown to be consistent with facts, is discussed in detail. A new impact ionization model is also described which is very similar to that of the field‐mission model. At present, it is experimentally indistinguishable from the field‐emission model.