A sorting-recall procedure was used to investigate how long-term memory in elderly subjects is affected by categorical organization. Sixty-four young adults (average age 20 years) and retirees (average age 67) sorted 48 unrelated words into two, four, six, or eight categories prior to recall. High- and low-frequency lists were tested, a manipulation that only affected the young adults. Surprisingly, overall initial recall was equally high for both groups, but the effect of increasing numbers of categories on recall differed dramatically for younger and older adults. For young subjects the correlation between recall and number of categories was .82, meaning that 67 percent of the variance in recall was accounted for by the number of categories in the sorting scheme. The older subjects, in contrast, recalled the same overall amount as younger adults, seemingly without the benefit of the chunking process provided by increasing the number of sorting categories. This finding shows that older subjects can bring resources to the memory situation that are highly effective, even if mysterious, compared to what is known about young adults. The fact that long-term forgetting was much greater in the elderly, however, suggests that in spite of excellent initial memory, the effects of decay or interference may be more pronounced in the elderly. (KC)