objectives to the Congress and the community at large in terms of “themes” as described by its Office of Space Science. The Origins theme has been particularly resonant with Congress. The agency is to be commended for its proactive approach in clarifying its goals and in changing its research and analysis policies to attempt to ensure optimum science returns from the investment in missions.

The committee is concerned, however, that NASA maintain the diversity in mission size, from small to medium to large, needed to meet scientific objectives in a cost-effective manner.

• The committee recommends that NASA maintain diversity in its flight programs by ensuring that a suite of opportunities, including small, moderate, and major missions, is available to accomplish scientific goals.

There are compelling scientific, programmatic, technical, and educational reasons for ensuring some balance between the major flagship missions and missions of moderate and small size (and cost). Flagship missions such as the Compton Gamma Ray Observatory, Hubble Space Telescope, and Chandra X-ray Observatory occur about twice per decade and produce outstanding science that defines substantive new areas of research in astronomy. But because of their high public visibility and great costs, such missions must be designed and built so as to maximize the likelihood of operational success in orbit. Ensuring a high success rate tends to drive up costs and lengthen schedules. The future of astronomy in space will be at substantial risk if it must depend on the successful deployment of only a few missions per decade.

Small and moderate missions add important dimensions to NASA’s space astronomy program: respectively, rapid response and targeted science. The Explorer program, an effective response to the need for frequent small-mission opportunities, should be continued at its current level. Because they can deploy new technology on relatively short time scales or move rapidly to follow up on recent discoveries, newly conceived missions of moderate cost can at times scientifically outperform the large missions on particular problems. Given the lower costs of small and moderate missions, an occasional failure can be accepted, although no failure in space occurs without some political cost to the program. Compared with one or two larger missions, several moderate and many small missions will more likely provide greater opportunities for developing a diverse set of new technologies and for training experimental space