We estimated benthic organic matter standing stocks, macroinvertebrate production, food-web structure, and stream metabolism along a longitudinal continuum of Kings Creek at Konza Prairie Biological Station in Kansas. Benthic macroinvertebrates were sampled for 1 yr from three stream reaches (grass, shrub, gallery forest). Total habitat-weighted production ranged from 9.1 g to 31.9 g ash-free dry weight m 22 yr 21 with the grass headwaters accounting for the highest production. Functional group contributions to production were similar along the continuum, with collector-gatherers (40-56% of total) and predators (23-38%) dominating all reaches. Detrital pathways appeared to account for most energy flow in all reaches, and all reaches were net heterotrophic on all dates when metabolism was measured, except for the grass headwater reach in spring. Predators were wellrepresented and consumed 65-107% of total macroinvertebrate production, whereas scrapers (11-15%), shredders (2-4%), and filterers (, 1%) consumed relatively small percentages of available resources. Dominance of predators suggested top-down controls are important in this system. However, changes in energy flow through other groups reflected longitudinal variation in resource availability. This prairie stream continuum conforms to some predictions of conceptual stream continua models, but biotic interactions, which are often not considered, may override some predicted responses to changes in the physical template.

-Diet interactions between native and non-native fishes may influence the establishment of native species within their historical range (i.e., reintroduction). Therefore, we illustrated the food web structure of and followed the transition of the federally endangered humpback chub Gila cypha into a novel food web following translocation and determined the potential for a non-native species, rainbow trout Oncorhynchus mykiss, to influence translocation success. Humpback chub and rainbow trout used resources high in the food web and assimilated similar proportions of native fishes, suggesting non-native rainbow trout may occupy an ecological role similar to humpback chub. Subsequently, humpback chub may be well suited to colonise tributaries because of their ability to consume resources high in the food web. Additionally, diet partitioning may occur between all members of the fish community as indicated by separation in trophic niche space and little trophic overlap; although all species, particularly bluehead sucker Catostomus discobolus, used a broad range of food resources. Rainbow trout stomach content analysis corroborated stable isotope analysis and suggested rainbow trout diet consisted of aquatic and terrestrial macroinvertebrates, while larger rainbow trout (>120 mm total length) consumed a greater proportion of fish (incidence of piscivory = 5.3%). Trophic interactions may reveal an underutilized niche space or biotic resistance to the establishment of translocated native fishes. Continued translocation of humpback chub into tributaries appears to be one option for conservation. However, successful establishment of humpback chub may depend on continued removal of non-native trout, increasing availability of diet sources at higher trophic levels.

Abstract.-We examined the influence of water temperature, diet, and feeding sites on growth differences of brown trout Salmo trutta in southeast Minnesota streams. Streams were classified as having either fast or slow growth based on back-calculated lengths at age and incremental growth measurements from scales collected in 2002 and 2003. Mean back-calculated lengths at age in fast-growth streams exceeded 141 mm at age 1, 229 mm at age 2, and 289 mm at age 3. In slowgrowth streams, mean back-calculated lengths did not exceed 127 mm at age 1, 208 mm at age 2, and 268 mm at age 3. The frequency of feeding sites was low in both stream groups, although possibly higher in the fast-growth streams. Water temperatures were significantly warmer in the fast-growth streams in summer and somewhat cooler in winter. We used a Wisconsin-style bioenergetics model for brown trout to independently test the effects of water temperature and diet, holding other parameters constant. Differences in temperature regime alone were not sufficient to explain differences in growth, and, in fact, predicted slower growth in the fast-growth streams in summer. Different diets resulted in differing predicted weight gains when temperature was held constant; diets with the highest energy densities always resulted in the greatest predicted weight gains. For younger cohorts, diet items with high energy densities included amphipods and mayflies (Ephemeroptera). Estimated consumption rates increased with summer temperatures. Our data suggested that growth differences among streams were likely due to differences in diet and prey availability, with more energy-rich diets available in fast-growth streams. We hypothesize that differences in diet and prey availability are mediated by water temperatures.

The longfin sculpin, Jordania zonope Starks, 1895, and the scalyhead sculpin, Artedius harringtoni (Starks, 1896), arecommon subtidal cottids in the San Juan Archipelago, Washington, but little of their autecology has been investigated. Comparisons of distribution, microhabitat density, foraging behaviour and diet wereconducted for the two species. Even though the two species occurred close together, microhabitat use differed between them, as did foraging strategies. Jordania zonope had a broader foraging range and tended to take bites out of prey items, whereas A. harringtoni had a more limited foraging range and tended to swallow prey whole. Partitioning of spatial and nutritional resources as well as morphological differences supported the hypothesis that J. zonope and A. harringtoni minimize interspecific competition in spite of microhabitat overlap and often side-by-side feeding.