1. The European eel population has decreased drastically during recent decades, and new EU-legislation calls for measures to change this negative trend. This decline has been attributed to a number of factors, including habitat fragmentation by structural barriers that prevent eels moving between freshwater and the sea. The success of downstream migrating adult silver eels migrating past a hydroelectric plant (HEP) in Sweden was examined by radio-telemetry, and the results were considered in a historical context by analysing catch data from the river for 1957-2006. 2. The choice of routes and passage success were quantified for three treatment groups and one control group of silver eels. The first treatment, the reservoir group (n = 50), was released into the reservoir upstream of the HEP, and these fish could proceed downstream by passing through the HEP (20 mm rack and turbines) or by entering the spill gates into the former channel, bypassing the HEP. The second treatment group (inside rack, n = 15) was released downstream of the 20-mm rack and had to pass through the turbines to continue migration to the sea. The third treatment group consisted of dead radio-tagged eels (n = 6) that were released into the turbines to study the extent of drifting by dead individuals. Finally, the control group (n = 50) was released downstream of the HEP to test for effects of confounding factors. 3. Most live individuals displayed migratory behaviour and continued to proceed downstream after release. Only 8% of the fish released in the reservoir or downstream of the HEP (control) did not migrate. The probability of reaching the next HEP, 24 km further downstream, was high for the control group (96%) and the reservoir-released individuals that passed the HEP via the spill gates and the former channel (83%). Survival was low and size-dependent for the individuals that passed the turbines (40%) and even lower for the individuals that had to pass through the rack and the turbines (26%). The overall passage success for eels released in the reservoir was 30%, including both routes. 4. Annual catch data from 1957 to 2006 showed that the number of eels in the River Ä tran has decreased. Despite this decrease, escapement biomass has remained unchanged, because of the fact that the mean size of eels has doubled. Passage data from 2007 show that changes in size and abundance have resulted in a reduction of relative escapement to the sea to values that are 21-24% of what they were in 1957-66. However, this low level of 2167 escapement could potentially be rectified if appropriate measures facilitating HEP passage are successfully implemented, since the potential escapement biomass in the river, owing to the large size of the eels, has changed little since the 1950s.

We evaluated the effects of a rehabilitation project, whose goal was to re-establish longitudinal connectivity for anadromous trout in the regulated river Emån. We used a holistic approach, by tagging and following both upstream-migrating spawners (N ¼ 348) and downstream-migrating smolts (N ¼ 80) and kelts as they passed two hydroelectric plants (HEP 2-3) with naturelike fishways.When migrating upstream, 84-88% of the spawners stopped, primarily at spawning grounds, before reaching HEP2. The proportion of stoppers was lower (56%) for fish that had been to the fishways in previous years, indicating that the recolonization rate is likely to increase over time. Of the spawners that approached the fishway at HEP2, 77% rapidly located the fishway situated next to the tail-race, resulting in an attraction efficiency of 81% and a passage efficiency of 95%. The time required to locate the fishway inside the former channel at HEP3 was substantial, but the attraction efficiency (89%) and passage efficiency (97%) were nevertheless high.The kelts swam downstream mainly in spring, using spill gates and the fishways, to swim past HEP2 and 3 and continue downstream to the Baltic Sea. Iteroparity was confirmed by the fact that 20% of the spawners were tagged in previous years. Smolt loss was about 30% for both HEPs, with a higher turbine-induced loss 30% for fish passing through Francis runners than a Kaplan runner. Fifteen per cent of the tagged smolt reached the sea and none of these fish had swum through the Francis runners.It will probably take many years before longitudinal connectivity is fully re-established in the river Emån, due to substantial losses of both upstream-migrating spawners (35% loss) and downstream-migrating smolts (50%) and kelts. In addition, smolt production in areas upstream of HEP3 is far below carrying capacity. Thus, additional measures that not only facilitate movement of upstream spawners, but also reduce mortality and injuries of downstream migrants are urgently needed to create a self-sustaining fish population.

We evaluated the function of two nature-like fishways for re-establishing connectivity for anadromous salmonids in the regulated River Emån. Between 90 and 100% of the salmonids that entered the fishways actually passed through them, with median speeds of 180-190 m h À1 . Only 50% of the anadromous brown trout that passed the first fishway also passed the second one, indicating that the fish might have had problems locating the upstream fishway. The fishways were also observed to function as a passage for downstream post-spawning migrants. The densities of brown trout yearlings upstream of the fishways were higher in 2002, after the fishways were built, than during pre-fishway years. In control sites in other parts of the river as well as in a nearby river, no changes in yearling densities were observed. Thus, the fishways are working for upstream spawners, albeit at a recolonization rate that is slower than expected.