It has been suggested that post-traumatic stress is related to the nature of an individual's trauma memories. While this hypothesis has received support in adults, few studies have examined this in children and adolescents. This article describes the development and validation of a measure of the nature of children's trauma memories, the Trauma Memory Quality Questionnaire (TMQQ), that might test this hypothesis and be of clinical use. The measure was standardised in two samples, a cross-sectional sample of non-clinic referred secondary school pupils (n=254), and a sample participating in a prospective study of children and adolescents who had attended a hospital Accident and Emergency department following an assault or a road traffic accident (n=106). The TMQQ was found to possess good internal consistency, criterion validity, and construct validity, but test-retest reliability has yet to be established.