SUMMARY The ability of ducks to derive magnetic direction information was tested in a conditioned procedure and the functional properties of the mechanism of magnetoreception investigated using common manipulations. Twelve ducks were trained to find a hidden imprinting stimulus behind one of three screens in a round arena. Once a criterion was reached, the directional choices of ducks were recorded in four treatments presented in a random order, separated with rewarded training trials to avoid extinction. In tests in the geomagnetic field, ducks preferred the screen in the training direction ( P =0.005). In the crucial tests of magnetic orientation with the magnetic field experimentally shifted by 120 deg, ducks showed a significant difference in the choice for the correct magnetic direction between treatments ( P =0.002). More specifically, they chose the correct magnetic direction more often than expected by chance ( P =0.03), indicating that they were deriving directional information from the magnetic field. Ducks also chose the correct magnetic direction more often than expected by chance in tests with the shifted field after the upper bill was anaesthetised with lignocaine ( P =0.05) or when the right eye was covered ( P =0.005), indicating that these manipulations did not impair the ability to choose the correct magnetic direction. Thus, Pekin ducks can be conditioned to magnetic directions, and the results are consistent with the hypothesis that magnetic orientation is based on a chemical magnetoreception mechanism that is not restricted to the right eye.