The central problem of this dissertation was to determine the relative importance of the direct and indirect effects of temperature and ration on growth of trout. I approached this problem by combining field observations and a growth experiment to address the relative contribution of temperature, macroinvertebrates (i.e., potential ration), and the indirect effects of temperature through ration. Laboratory experiments are useful for eliciting direct causal effects by manipulating one factor while controlling others. Laboratory studies of temperature and ration on growth under controlled conditions have provided empirical data that fits a theoretical framework based on animal physiology. A great advantage of these studies is that direct causal linkages can be determined between parameters. However, experimentally determining the outcomes of interactions between many potentially interacting factors is a daunting task; and using experimentally derived direct effects to determine the consequences of complex indirect interactions is not feasible (e.g., due to synergisms). Furthermore, trout seldom experience these conditions in natural streams where daily and seasonal variability in ecological conditions is common. Observations from field studies have the advantages of complexity, realism, and connection to natural processes. However, direct causal linkages are difficult to ascertain due to inherent variability in natural systems. Experimentally derived growth models predict that temperature should be an important controlling factor for fish in coldwater streams. I found temperature summaries to be positively correlated with growth of juvenile trout (Chapter 2, Chapter 3) and macroinvertebrate standing crop (Chapter 3). Furthermore, temperature summaries explained a large fraction of the variation in growth of trout and suggest that temperature is the major factor affecting growth in small coldwater streams (Chapter 2, Chapter 3). Two causal path models were developed to examine the causal linkages between temperature and growth of juvenile brook trout. The first simple model contained direct causal paths from temperature and macroinvertebrate parameters to trout growth (Figure 3.3). Analysis found the total effect of temperature on variation in growth of juvenile trout to be 50% greater than the combined effect of the macroinvertebrate standing stocks. In addition, direct and indirect effects of temperature on trout growth were of the same magnitude. The model and analysis indicate that for streams with suboptimal temperatures increases in stream temperature can lead to growth increases approximately double those predicted from physiological considerations of temperature alone. Since causal path models are analyzed as linear models with direct effects complex interactions between parameters can only be modeled as direct effects. The second path model is similar to the first but contains an additional direct causal path to growth from the physiological growth potential (PGP) of trout. This parameter (PGP) represents the physiological effects of temperature on growth when fish are provided with excess ration and was developed from the growth experiment described in Chapter 4 (Figure 4.6). PGP had the strongest direct effect in the model and was more than 50% greater than the direct effect of invertebrate standing stock. The direct effect of temperature on growth was also significant and of the same magnitude as the indirect effect. However, since PGP estimates the effects of temperature in the absence of ration, I interpret this path as a temperature mediated effect on the availability of prey to trout. Results of the model parameterization are consistent with this view. Direct and indirect effects of temperature were of the same magnitude in each path model and the total temperature effect made up about two-thirds of the total effect of all parameters on growth of trout. Temperature thus influences trout growth rate via two causal pathways: (1) the direct physiological effect on metabolism, and (2) by influencing the produ tion of availability of macroinvertebrate prey. These data suggest that small increases in temperature in streams with sub-optimal temperatures will improve juvenile trout growth. Macroinvertebrate standing stock was also shown to be a significant predictor of growth of trout (Chapter 3). Decomposition of variance into year, site, and year*site interactions for growth rate of brook trout, temperature, and macroinvertebrate standing crop indicate that macroinvertebrate biomass is more variable over time within sites than temperature or growth rate (Figure 3.2). Thus, variability in prey abundance may be responsible for much of the observed year to year variation in brook trout growth, while temperature may be the primary factor influencing average growth differences that occur between sites. I used a natural experiment to explore the effect of an alteration of the benthic community on predator populations in lotic ecosystems (Chapter 5). Perlodid stonefiles and juvenile trout had significant increases in their densities (Isogenoides spp., Isoperla spp. juvenile brown trout) or decreased growth rates (juvenile brook trout) associated with pathogen-induced reductions in Glossosoma density. This study gives the first documented evidence of the impact of a pathogen-induced trophic cascade on lotic macroinvertebrate predator and trout populations and that benthic predator populations were affected by the interaction of the grazing caddisfly Glossosoma nigrior with its parasite Cougourdella. Futhermore, the causal path model developed in this study indicated a strong direct effect of Glossosoma and a strong negative indirect effect of the microsporidian pathogen Cougourdella sp. on the growth rate of juvenile brook trout. Since trout rarely consume Glossosoma I suggest that this direct path represents a trait mediated indirect effect that increases the vulnerability of other macroinvertebrates to trout. Major implications of the trophic cascade for brook trout populations include lower juvenile growth rates, lower overwinter survival, lower standing stocks of both juvenile and mature trout, and increases competition with brown trout.