The goal of this project is to address the problem of accessibility and preservation of research data. Today, the responsibility for most scientific data rests on individual researchers who typically do not have the means for long-term preservation of or access to their data. As a consequence, scientific data all too frequently are not available for comparative analysis and may become lost altogether due to funding changes, personnel turnover, and physical deterioration. Research universities, however, already rely on the library to provide institutional support for archiving, access, metadata, and outreach services for print and digital collections. This project will examine the opportunities for partnership between the research community and the library to ensure that research data are collected, preserved, and accessible to research communities. The project will focus on language acquisition data that have been collected by the Cornell Language Acquisition Laboratory (CLAL) over twenty years and across more than twenty languages and countries. The project will develop and document the principles of access, dissemination, and data preservation; develop and document a prototype for testing and demonstration; and examine possible protocols for inter-institutional library cooperation. The project will also assess both start-up and ongoing costs required for this planned infrastructure to facilitate extending the model to other research institutions and libraries. The project will have broad impact by exploring a new approach to long-term storage, preservation, access, and dissemination of scientific data which may be applicable to many disciplines.