This study shows that chronic, long-hospitalized schizophrenics perform (tested by the Famous Events Questionnaire) at a higher level on remote memory than they do on word list recall. A post hoc, matched-tasks check suggested that this finding was not due to differences in the discriminating power of the tasks. Relative to normals, patients showed no differential performance on the three time periods (1973/74, 1979/80, 1984) represented in the remote memory questionnaire, thus giving no evidence for an amnesic gradient. The findings are discussed in relation to amnesia and dementia hypotheses in chronic schizophrenia.