By means of electrofishing, we examined seasonal and size-class variation in habitat preference by juvenile brown trout (Salmo trutta) in a third-order river in northern Finland. Larger trout preferred deeper stream areas than young-of-the-year fish. At the onset of winter, all trout size-classes moved into shallower water, but this mainly reflected seasonal variation in habitat availability. In winter, trout preferred slowly flowing stream areas, whereas in other seasons the mean water velocities used by trout parallelled habitat availability. In summer and autumn, age-0 fish favoured stream areas with large amounts of aquatic vegetation to provide cover. The largest trout (16-22 cm) occupied habitats with little cover throughout the year, and in winter, all trout avoided areas with high instream cover. In summer, all size-classes preferred small substrates, whereas in winter, areas with cobble-boulder substrates were preferred, especially by trout larger than 10 cm. Wintering trout often shelter among the interstitial spaces of coarse substrates, and to facilitate the survival of juvenile trout through winter, stream management programmes need to ensure that such particles are abundantly available in trout wintering areas.

1. The predominantly nocturnal constrained drift of stream invertebrates is commonly regarded as a behaviour that avoids encounters with visually foraging fish in the water column. The alternative explanation, that drift peaks are caused by bottom‐feeding, nocturnal predators, has rarely been tested. 
2. We examined these hypotheses by collecting invertebrate drift in five streams in northern Finland: one with brown trout (Salmotrutta, a drift‐feeding fish), one with alpine bullhead (Cottuspoecilopus, a benthic fish), one with both species, and two fishless streams. 
3. Drift by Baetis mayflies was aperiodic or slightly diurnal in both fishless streams on all sampling occasions. In contrast, drift was nocturnal in streams with trout and, to a lesser extent, in the stream with bullhead. Non‐dipteran prey drifted mainly nocturnally in all streams with fish, whereas Diptera larvae were less responsive to the presence of fish. 
4. In laboratory experiments, bullheads were night‐active, causing a much higher frequency of drift by touching Baetis at night than during the day. Thus, increased nocturnal drift may serve to avoid both visual predators (a pre‐contact response) and benthic fish (a post‐contact response). In streams with bottom‐feeding fish, nocturnal drift should be caused by increased drift by night rather than by reduced drift by day.

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.. British Ecological Society is collaborating with JSTOR to digitize, preserve and extend access to Journal of Animal Ecology. Summary 1. Prey preference is determined by active predator choice and by the relative vulnerability of prey taxa. In this study, we addressed the mechanisms of prey preference in the perlodid stonefly Diura bicaudata. 2. Components of the predator-prey interaction between the stonefly and its prey were quantified in laboratory observations. These data were compared to prey selection in preference trials and to gut contents of field-collected stoneflies. Experiments were conducted in spring (May) and in autumn (September), using prey taxa commonly available in each season. 3. In the September trials, Diura exhibited positive selection for black fly larvae, whereas Heptagenia, Ephemerella and large Baetis mayflies were avoided. Encounter rates did not affect preference: these were highest for heptageniids and lowest for black flies. Once contacted, black flies were practically always attacked with a high capture probability. Attack propensity and capture success were very low for all other prey types, including Baetis mayflies. 4. In May, female Diura avoided Ephemerella mayflies and Asellus isopods, but showed a positive, albeit non-significant, preference for Nemoura stoneflies. Males did not select any of the prey types. Again, encounter rate was the least important determinant of preference: nemourid stoneflies were encountered less frequently than other prey, especially by female Diura. Females attacked Asellus more frequently than other prey types. Baetis was not a preferred prey for either of the sexes. 5. Our results show that D. bicaudata prefers sedentary or slowly moving prey types. Preference was determined both by active predator choice and differential prey vulnerability. We suggest that although mobile prey such as Baetis are encountered frequently, they are difficult to capture, and are thus relatively safe from stonefly predation when sedentary prey are also available. 6. Microhabitat overlap between predator and prey may determine encounter rates in the field, but this may not translate into prey preferences. Prey with efficient antipredatory behaviours can risk predator encounters, whereas prey with less efficient escape mechanisms may have to select microhabitats avoided by the predator. It is thus essential that laboratory systems incorporate at least some of the structural complexity of natural streams. However, even relatively simple laboratory systems may provide the complexity needed, as long as they contain prey refuges.

No abstract

The recovery of benthic macroinvertebrates after disturbance from stream rehabilitation was studied in the River Livojoki, northern Finland. The stream that had been channelized for log transport was rehabilitated on 1 July 1992 by digging holes and inserting boulders. We measured habitat characteristics and sampled benthic animals before and after rehabilitation, including an unrehabilitated control site. The immediate effect of rehabilitation was a slight decrease in the abundances of benthic insects. Recolonization occurred rapidly, within 10 days. Disturbance of the rehabilitation did not have a detectable effect on the macroinvertebrate community. Most species-level changes and community patterns reflected seasonal life history events. Timing of such rehabilitation work can be critical for the recovery rate, which depends on the colonization abilities of the species present after disturbance. We suggest that many disturbances (including minor floods and moderate rehabilitation procedures) may have only small, short-term effects on benthic communities. We emphasize the importance of considering seasonality in studies of disturbance in streams.