The RF database (http://www.reuniting families.org) leverages existing open-source informatics technologies to collect and collate physical, phenotypic, and genotypic information from geographically dispersed agencies. The overall goal of the system design is to allow clients and administrators to search controlled data repositories in a two-tier format: general matches to physical evidence and specific identification based on molecular criteria. This two-tier approach is necessary because of the prohibitive cost of performing molecular tests on the large volume of public queries handled by RF. In addition to Web-based interfaces for data upload, searchers, and automated reporting, the RF data base system contains several tools that enhance its usefulness. These tools include stored queries, which allows the user to modify a query for future user-generated searches; multilevel user authentication; data uploads and automated reporting; and automated match discovery. This paper also describes database specifications. 2 figures and 12 references