The 'nootropics' are a new class of psychoactive substances that improve learning and memory. Their almost exclusive effect on memory may indicate that they act on processes specifically involved in information storage. When administered after the learning trial, these substances improve subsequent retention performance in mice, even if an interval of 8 h has elapsed between learning and treatment. CGS 5649B, a highly active new substance, is effective even after an interval of 24 h. Although consonant with the 'consolidation' hypothesis, the results may challenge prevailing notions about the formation of memory traces.