In 3 sections of the same interview, children (N = 107, ages 3–8 years) were asked about body touches during previous medical examinations that included genital and anal touches for some children. First, in a free recall phase all children were asked to describe what had happened during the medical procedures. In the second and third sections they answered questions about body touches in 2 conditions, with body diagrams (BDs) and without body diagrams (no-BDs), with the order of conditions counterbalanced. Within each interview condition, the children answered cued-recall questions about touching and a set of recognition (yes-no) questions about touches to individual body parts. Cued recall with BDs elicited a greater number of correct sexual touch reports, but also more forensically relevant errors from the younger group. Cued-recall performance with BDs was largely identical to recognition performance without BDs. Taken together, the paucity of research on BDs and the current findings suggest 2 interim conclusions: (a) the use of BDs to elicit touch disclosures is not yet an evidence-based practice, and (b) there is a pressing need for research that examines promising approaches for encouraging accurate disclosures of abuse.