Selection for growth of brown trout (Salmo trutta) affects feed intake but not feed efficiency

Brown trout (Salmo trutta) were selected for growth for 4 generations. We tested the effects of selection on voluntary feed intake measured by self-feeders, feed efficiency and size variability. The specific effects of a slight feed restriction and of food deprivation were also investigated. Fish were issued from groups of eggs of selected females fertilised with sperm of selected (S group) or control males (S1/2 group). According to the growth rates expected for the selected and control Lines, the S1/2 group was fertilised 12 days before the S group, so that all the fish reached 8 g at the same time. At 8 g, they were allotted to 8 tanks (500 fish per group) and 3 experimental periods followed. Fish were accustomed to self-feeders during a 28 days pre-experimental period. Then, half of the groups were fed ad libitum, and half were restricted (80% of the expected ad libitum level) for 171 days; growth and feed intake were recorded regularly and any uneaten food was weighed. Then followed a 56 days starvation period. At the end of the pre-experimental, feeding and starvation periods, individual weights and lengths were measured on 50 trout per tank. The response to selection at the end of the feeding period varied with the feeding level. In ad libitum fed groups, the mean final body weight of S was +6.1% higher than that of S1/2 and feed efficiency was similar (1.10 for S and S1/2). The higher growth of S compared to S1/2 was related to a higher feed intake of the S groups (+5.3%). When fish were restricted, the final body weight was lower in S (117.1 +/- 2.1 g) than in S1/2 groups (123.8 +/- 1.7 g). This was mainly related to a slightly lower feed efficiency of S compared to S1/2 at the beginning of the feeding period. Neither the group nor the feeding level affected the size variability of the fish. At the end of the starvation period, the relative loss of weight was equivalent for all the groups, and the variability of the weight was higher for S than for S1/2. The results highlight the fact that genetic gain can only be expressed when brown trout are fed ad libitum. (C) 2001 Ifremer/CNRS/Inra/IRD/Cemagref/Editions scientifiques et medicales Elsevier SAS.