Apart from the strong correlation between length
and proportion of piscivorous trout found in our study,
this belief rests further on our own observations in the
autumn of 2001, where trout were preying on 0+
cyprinids in the littoral zone and on a study by NIVA
(1999), where trout performed a relatively fast switch
from invertebrate prey to fish prey, as soon as this became abundant. We finally conclude (i) that for an effective
food-web manipulation stocked brown trout must be
relatively large (> 300 ram), (ii) that a high proportion of
large trout (> 500 mm) must be present and (iii) that their
proportion of the total fish community should be close to
the upper limit of the range of 30-40% recommended by
BENNYDORF (1995) for successful biomanipulation.
Acknowledgements
We thank Falk Wieland for technical assistance in the field and
Hans-Peter Uhlig (Talsperrenmeisterei Freiberger Mulde!
Zschopau) for providing the temperature data. The cooperation
of the Anglerverband Stidsachsen Mulde/Elster e.V. is
greatly acknowledged. Two anonymous referees gave helpful
comments to improve the manuscript, we are grateful for
these. This study was supported by the Landestalsperrenverwaltung
Sachsen and the Arbeitsgemeinschaft Trinkwassertalsperren
e.V. (ATT).