There are three issues that directly relate juvenile sockeye production to the types of fish, zooplankton and algae that are available in 3 specific sockeye nursery lakes. When large bodied herbivores are abundant, we might predict that sockeye growth and production will increase. Conversely, when large invertebrates or competing fish species (such as stickleback) are abundant, we might predict that sockeye production will be lower. When these two factors are combined, we might predict that fertilizer induced stimulation of algae will cascade up to the invertebrate predator rather than sockeye. On the other hand, when invertebrate carnivores are rare, and large herbivores are abundant, fertilization is likely to cascade up to the sockeye, resulting in increased rates of growth and production. Finally, all of this is predicated on the assumption that fertilizer additions will stimulate algae that can be grazed by zooplankton. To determine which of this series of pelagic food web pathways were most important in Woss Lake, we undertook an analysis of food web structure in both Woss (fertilized) and Vernon (control) Lakes. Since its inception this has been a collaborative program. During the first three years (2000-02) the program was run by the Gwa'ni Hatchery, Nimpkish Resource Management Board, Fisheries and Oceans Canada and York University. More recently (2003-08) the work has been funded by the Pacific Salmon Endowment fund and run by Gwa'ni Hatchery, Nimpkish Resource Management Board, and Fisheries and Oceans Canada. The program has been structured to deliver the following analyses: (1) Beginning at the top of the food web, we measure sockeye growth rates, diets, diel migration patterns and water temperatures. These data are then used in a bioenergetic analysis to estimate food consumption by juvenile sockeye. (2) We then assess zooplankton population abundances, biomasses, length-weights, egg numbers. We know that all zooplankton were not going to be equally good food sources for juvenile sockeye, so it is not enough to calculate overall zooplankton density and production rates. What we need to calculate is species-specific rates of production by zooplankton and consumption by sockeye. This analysis is intended to allow us to determine whether or not the sockeye can graze down all the 'good food species', leaving other species that are much less suitable as food sources. (3) We then calculate the percentage of species-specific zooplankton production that is consumed by sockeye. When consumption approaches or exceeds production, we know that food limitation is a problem and in the long term we can react to this problem by adding more or less fertilizer or by increasing or decreasing rates of sockeye fry recruitment (by increasing spawner abundance or through fry outplanting). (4) Finally, because 'bottom-up' sockeye enhancement theory is based on the assumption that fertilizer additions will stimulate algae that can be grazed by zooplankton, we monitor rates of nutrient addition and we complete species-specific assessments of algal species composition. Our objective is to determine whether the algal genera that were stimulated by the fertilizer additions were also grazable by zooplankton.