SUMMARY. 1, Microhabitats chosen by young grayling (Thymallus thymallus L.) were investigated by electrofishing. The first feeding habitat was marginal areas where water velocity did not exceed 15 cms−1. Fry fed on drift in the first 10 cm of the water column below the surface. This location was very stable during the first weeks of growth (mid‐April to mid‐May). When they were 3–3.5 cm long, fry began using benthic areas and gradually moved away from the banks towards the channel. 2. A downstream migration was noted from the beginning of emergence (mid‐April) to the end of May. This mainly nocturnal movement involved all fish sizes and caused a complete removal of young‐of‐the‐year by June from the spawning tributary.

No abstract

Trophic interactions, including “top‐down” predator‐prey interactions, are particularly important in influencing the structure of fish communities. While the varied impacts of piscivorous fish have been well investigated, the effects of fish‐eating birds on riverine fish behaviour and population dynamics still remain controversial, mainly because they are undervalued. Summer experiments were conducted in an experimental outdoor stream to evaluate the effects of avian predation threat, stream flow, and overhead cover on growth and behavioural tactics of wild juvenile chub (Leuciscus cephalus). Groups of fifteen chub maintained in riffle‐pool sequences were submitted to combinations of different conditions, namely absence or presence of a simulated fish‐eating bird, low or high flow, and absence or presence of medium or high cover. In the absence of predation threat, chub foraged in the riffles and maximized feeding opportunities. Under predation threat, they sheltered, foraged less and grew slowly and as expected, they increased their use of the riffles at high flow as water turbulence is an efficient shelter from birds but only in the absence of cover. In the presence of cover, fish sheltered exclusively under pool covers and were more prone to take risks at low flow because of higher costs in terms of lost feeding opportunities associated with these conditions. This result indicates that flow velocity altered cover use tactics through its impact on food supply, suggesting that it may affect the outcome of predator‐prey relationships. So, chub use cover in a flexible manner according to both the benefits in terms of predator avoidance and the costs in terms of lost feeding opportunities. A striking finding of the experiments is the drastic reduction in the range of growth variances amongst fish when they are maintained under predation threat, suggesting a homogenization of fitness between individuals. From all our results, we argue that in lowland streams, under summer field situations, fish‐eating birds may affect local prey population dynamics more through sub‐lethal effects on growth rates than directly through death rates.

An approach based on direct observation from the riverbank allowed a precise description of the size-dependent distribution patterns of young grayling. The smallest larvae (15–20 mm) were mainly observed in the upper layers of the water column and very close to the riverbank. With increasing age and size (20–35 mm), old larvae and young parr began to hold positions closer to the bottom and to the edge of the main channel. From a size of 35–40 mm. an increasing number of juveniles were observed in the river channel, holding a benthic position. A diel habitat shift was also shown for individuals of all stages, between their respective daytime habitats and dead zones, where they were observed resting on the bottom in very shallow water at night. The benefits of this shift might be expressed in terms of energy conservation and (or) decreased predation risk. This study points out the importance of marginal habitats in the first weeks of grayling life as (i) exclusive nursery areas for larvae that use them both by day and by night and (ii) resting habitats for juveniles at night.