Feeding experiments with the European Atlantic sturgeon, Acipenser sturio L., 1758 to accustom large juveniles to a new feed item and the influence of tank size and stocking density on growth

27 Acipenser sturio juveniles of a 1995 artificially reproduced F-1 population from the Gironde, Cemagref/Bordeaux were used in feeding experiments and in trials on the influence of tank size and stocking density on growth lasting for 26 months. In the beginning the 66 month old fish had a median total length of 98 cm (range 80110 cm), at the end the fish measured 106 cm (median, range 88123 cm). Fish were kept in two small (6.8 m(3)) and two large (11.6 m(3)) experimental tanks at 20 +/- 1.5 degrees C and under a natural photoperiod in freshwater in a semi-closed indoor system. Basic food were frozen large chironomids (FLC) offered after thawing ad libitum (feeding rate 2.5-4 % body weight day(-1), exceptionally 8 %). An experiment for six weeks to wean the fish to a pelleted trout feed was not successful. For 28 weeks the juveniles were fed FLC plus sprat pieces which resulted in an increase of specific growth rates (SGR) at the end of the weaning trials. Sprat pieces fed alone during the 4(th), the 10(th) and the 22(nd) week led to significantly lower SGR values. Fish kept in large tanks at stocking densities of 5.3 kg m(-3) for 8 weeks only after the fourth week showed higher SGR values than the fish kept a higher stocking density of 9.7 kg m(-3) in the smaller tanks. During two experimental periods of six weeks each time the fish were kept at stocking densities of 4.4-4.8 kg m(-3) and 2.5-2.9 kg m(-3), respectively. Only during the second experiment the fish kept at the lower stocking densities showed significantly higher growth rates. Fish kept at stocking densities of 3.5-3.8 kg m(-3) kept in large tanks (11.6 m (3)) did not grow better than those kept in small tanks (6.8 m(3)) kept at the same stocking density. Increased feeding levels (8 % body weight day(-1)) for 4 weeks (food distributed for 10 and 24 hours) only led to an increase of the SGR values after the second week of feeding, whereas after the fourth week the SGR values dropped to normal values. A. sturio showed a high variation in both the SGR and the individual SGR values thus limiting the interpretation of these quantitative results.