This paper examines the concepts of ecology, ecosystems, and ecosystem management and then further examines the role of fish, wildlife, and plant ecology research in ecosystem management, past, present, and future. It is often assumed that research in support of ecosystem management will entail comprehensive studies of entire ecosystems whereas research programs that focus on one species do not constitute ecosystem management level research. The supposed dichotomy between single species and ecosystem level approaches has been the focus of considerable debate. However, this is a false dichotomy and ecosystem studies and single-species studies simply represent two ends of a spectrum of approaches for understanding ecological processes. Given that the level of scientific investigation (e.g., individual species, community, or ecosystem) does not differentiate ecosystem management research from more traditional approaches, what are the distinguishing features? Ecosystem management research is broader in scope than more traditional ecological studies. A greater emphasis is also placed on integrating results of various studies and programs to understand larger scale interactions and the structure and function of ecosystems. Model building also plays a greater role in ecosystem management research efforts as a means of not only understanding ecosystem processes but also as a means of generating hypotheses. Although the primary responsibilities of research and management are different, there is much room for interaction and integration of functions. Consequently, adaptive management has become an important part of ecosystem management and will likely become a larger part of basic research programs. However, adaptive management experiments should not be the endpoint. Instead, results from adaptive management studies should be used to generate hypotheses that can be tested with more traditional and rigorous scientific methods. As managers begin to deal at larger spatial and longer temporal scales changes in the end-products of research will be necessary. The task of assessing present as well as future conditions will greatly increase the need for user-friendly analytical tools (e.g., simulation models) that allow managers to visualize conditions on a large scale. A balance of adaptive management and traditional experimental designs will ultimately lead to better models of management.