A number of nondestructive techniques for analyzing the timing, frequency, and magnitude of natural disturbances in forest stands are discussed in this paper. Intensive age determination of trees is desirable for reconstructing forest disturbance history, but age distrubution alone is not always a sufficient basis for a disturbance chronology. Frequently all-aged forests have undergone severe past disturbance which cannot be readily identified from the distribution of trees among age-classes. Radial growth patterns provide more direct evidence of past canopy tree deaths. In cases where a large sample of tree ages and growth records is not feasible, structural attributes of forests can provide valuable supplementary evidence. Evidence of the effects of disturbance history and age structure on diameter distributions is examined, and principles for valid interpretation proposed. Crown area distributions and diameter distributions of nonsuppressed trees potentially appear to be more reliable structural indicators of past canopy disturbance than simple size distributions. For regional studies of disturbance frequency, random dispersal of medium to large plots across large landscape units is recommended as a sample design.© National Research Council of Canada, NRC Research Press. Abstract reproduced by permission.