A multivariate approach to optimization of macronutrient composition in weaning diets for cod (Gadus morhua)

Atlantic cod, initial weight 0.26 g, were fed diets varying in added protein from 530 to 830 g kg−1, lipid from 50 to 300 g kg−1 and carbohydrate from 0 to 150 g kg−1 of dry weight, according to a three-component mixture design. Analysed values of protein and lipid were 500–770 g kg−1 and 30–270 g kg−1, respectively. Analysed carbohydrate levels were as added. Increasing levels of both lipid and carbohydrate had a positive effect on fish growth (P < 10−3), whereas protein levels above 600 g kg−1 gave a reduction in growth (P < 10−4). The effects on growth were evident in fish less than 4 g, whereas fish growth between 4 and 6 g was unaffected by the dietary variation. It is hypothesized that the reduction in growth at high protein levels in fish of less than 4 g could be owing to incomplete utilization of protein, as the stomach of cod is not fully developed before the fish is approximately 1 g. Mortality and cannibalism were high in fish less than 4 g but low when the fish grew from 4 to 6 g. There was a significant decrease in cannibalism with increasing dietary lipid during the first half of the experiment (P < 0.05) and cannibalism was consistently high in fish fed less than 150 g kg−1 lipid. The lipid level in whole fish increased with increasing dietary levels of lipid (P < 10−6) and carbohydrate (P < 10−4), whereas the liver lipid level increased with increasing dietary lipid up to 200 g kg−1 (P < 10−6) and decreased thereafter (P < 10−4). Whole body glycogen increased slightly with increasing levels of dietary carbohydrate (P < 0.05) and was not affected by the other dietary variables. Liver glycogen increased in response to increasing dietary carbohydrate (P < 10−5) and decreasing levels of dietary lipid (P < 10−5). An abrupt increase in liver glycogen was seen with the reduction in dietary lipid from 100 to 50 g kg−1. The hepatosomatic index increased in response to both dietary lipid and carbohydrate (P < 10−6). It is concluded that the protein requirement of young cod is less than 500 g kg−1 of dry diet. Fish of less than 4 g should not be given more than 620 g kg−1 protein and should be supplemented with 150–200 g kg−1 lipid. Carbohydrate up to 150 g kg−1 of dry diet promoted growth and did not seem to affect the fish negatively. Fish above 4 g can be given diets varying in protein and carbohydrate over the wide range of concentrations used in the present study, but lipid supplementation should be restricted to between 100 and 200 g kg−1.