A memory model is proposed in which information enters an initial structuralchemical format within fractions of a second. Experimental amnesia is viewed as a failure to retrieve information from long-term memory rather than a failure of the information to be consolidated in long-term memory. Consolidation and retrieval failure explanations of experimental amnesia are found to differ primarily on the issue of recovery of memory. The nature of experimental amnesia paradigms is such that the consolidation position can never be supported; it may only be refuted by an occurrence of recovery of memory or left as an unresolved issue by a lack of recovery. The use of noncontingent stimuli to induce recovery from amnesia has been successful in several instances, thereby supporting the retrieval failure hypothesis. Success in producing amnesia for reactivated "old" memories further corroborates the retrieval position. Arguments that effectively delivered, high-intensity or high-dosage amnestic agents will disrupt consolidation beg the question and are presently without any empirical support. Two-process theories proposing both retrieval and consolidation failures are presently neither necessary nor parsimonious.