rate in these tests can be detected in less than 90 min, showing that feeding depression is a rapid, as well as general indicator of toxic stress. Because feeding depression is a rapid, sensitive, and ecologically relevant indicator of toxic stress in Daphnia spp s it is an ideal candidate for use as a bioassay endpoint. In situ bioassays previously have been deployed that use cladocerans as test organisms and have successfully detected toxic impacts in rivers(15-19]. Cladocerans adapt well to deployment in the field and are suitable as test organisms because survival rates in uncontaminated conditions are fre- quently greater than 85% for C dubia[15] and greater than 90% for magna[16]. However, most in situ bioassays that cladocerans have used lethality as an endpoint[15,16,18], which may not be the most sensitive endpoint to use. Sublethal, physiological endpoints generally are recognized lo be more sensitive than lethal endpoints because physiological endpoints often the initial reaction of organisms to stress and can be detected before mortality occurs ll9]. Therefore, feeding de- pression could be a useful sublethal endpoint to measure in in bioassays with D. magna and already has been success- fully employed as an in situ endpoint with Gammans pulex[16h. To gain an accurate measure of the toxicity of a water body by a D. magna in situ bioassay, animals must be con- tinually exposed, in cages, to the surrounding environment. For a bioassay that uses feeding depression as an endpoint, this would introduce some difficulty into the experimental de- sign. Directly measuring feeding rates of daphnids in the field while simultaneously exposing them to the surrounding en vironment is extremely difficult because the microparticles such as algae, that constitute the primary food source of D magna would leak from exposure chambers and compromise measurements of feeding rate. Taylor et al. [20] found that D, magna previously exposed to sublethal concentrations of eadmium exhibited a persistent