Efficiencies of two types of bypass, a surface and a bottom sluice, were tested for the natural downstream migration of silver eels Anguilla anguilla at a small hydroelectric power plant at Halsou, on the River Nive in France. Naturally migrating eels were caught after their passage through either bypass. A total of 637 eels were trapped during the three-year study. Total efficiency for both bypasses, evaluated on the basis of downstream movement of radiotagged eels, ranged from 56% to 64%. Given a bias due to hydrological conditions at the time of the runs, the precise efficiency of each separate bypass was not calculated. However, preferred passage through the bottom bypass for both tagged and untagged eels was confirmed by telemetry, as three to four times as many eels transited through the bottom bypass compared to the surface one. The behaviour of 74 individuals released in the forebay was observed by radiotelemetry. Close to half of the radiotagged eels returned up the headrace after their release, and most eventually migrated downstream over the dam with appropriate environmental conditions. Upon arrival at the power plant, eels displayed foraging behaviour in the forebay with frequent displacement interrupted by long resting periods in zones with low current. The repulsive effect of the trashrack located in front of the turbine intake increased with increasing turbined discharge. The study indicated that a trashrack with a smaller bar-spacing (around 20 mm), associated with an appropriate bypass, could deflect a large proportion of the female eels from the turbines. However, this solution needs to be tested on site to quantify the risk of mortality due to impingement on the trashrack