studies on energy flow in the Silver Springs and the Silver River (Odum, 1956, 1957). Numerous studies have measured primary production, respiration, and flow of energy and materials through stream food webs, and this work has led to more recent paradigms such as nutrient spiraling (see next section).
Just as lake scientists shifted from a descriptive or observational approach to a more experimental approach 30 to 40 years ago, stream limnologists also have adopted experimental approaches in recent decades; the experimental manipulation of a small stream with sucrose to enhance trout production (Warren et al., 1964) was the precursor of many later manipulations (e.g., see Lamberti and Steinman, 1993). Stream experiments have involved both artificial channels and real streams. Artificial channels have the advantage of allowing replication. They have been used to study the effects of chemical contaminants on stream ecosystems and to develop a mechanistic understanding of lotic processes at several locations in both this country and Europe (Lamberti and Steinman, 1993). Real streams have been used to conduct acidification and remediation manipulations; for example, lime has been added continuously to several small streams in New York and Sweden to study rates and mechanisms of biological recovery from acidic conditions. In addition, a few watershed-stream experimental manipulations have provided valuable information about linkages between terrestrial and aquatic components of watersheds (Bormann and Likens, 1967). Perhaps the best known involved a small forested watershed in the Hubbard Brook Experimental Forest (Likens et al., 1970; Vitousek et al., 1979). A dramatic increase in nitrate export from the watershed was observed when the forested catchment was clear-cut and treated with a herbicide to prevent revegetation.
Organizing paradigms that treat streams as systems and that integrate across the physical, chemical, and biological aspects of stream science are few in number and recent in origin. The River Continuum Concept (RCC) (Vannote et al., 1980) is the first and perhaps most important of these integrating paradigms. The RCC views entire fluvial systems as a continuously integrated series of physical gradients and adjustments in the associated biota (Cummins et al., 1995). Geomorphological and hydrological characteristics provide the fundamental physical template, which changes longitudinally within a drainage basin from the headwaters to the river mouth in a predictable fashion. Biological communities and associated attributes, such as the nature of the functional community groups, develop in adaptation to the fundamental physical template. The model thus has a definite watershed orientation and focuses on terrestrial and aquatic interactions. It is useful at the basin and stream scale in predicting the