Search

Resume Brown trout and rainbow trout (average weight 100 g) were reared in fresh water at 12OC under the same conditions beforc transfer of brown trout to sea water, in order to compare nitrogen utilisation in the two species. Apparent protein digestibility (ADC), nitrogen (ammonia and urea) excretion, protein productive value (PPV) and actual observed nitrogen mass balance were determined. Rainbow trout raised in fresh water had a higher growth rate (1.1 vs 0.8Yb.d-'), better food conversion ratio (0.7 vs 1.0), better ADC (91 vs 85%) and PPV (45 vs 35%) and lower ammonia excretion rates than brown trout reared in fresh water. Transfemng brown trout to sea water induced lower PPV (30%) and ammonia and urea excretion. Salinity did not rnodify metabolic efficiency in brown trout. Fat content was higher in brown trout (7.7-8.9% ww) than in rainbow trout (5.7-7.6% ww). Nitrogen mass balance indicated that compounds other than arnmonia and urea were produced in higher quantities by seawater brown trout. Behaviour, less domestication and specific ability to utilise protein could explain the differences between the two species.

Protein digestibility of soyabean meal, casein-gelatin mixture and fish hydrolysate was tested in the European sea bass, using increasing substitution levels in a reference diet. The protein digestibility of the latter composed of Norwegian fish meal had already been established (91 %). An accurate estimation of the apparent digestibility coefficient was made by linear regression, for the casein-gelatin mixture and soyabean meal, but not for fish hydrolysate (non significant correlation, p > 0.05). Protein digestibility estimated by extrapolation varied with the substitution level. The estimations obtained with the highest level used for each ingredient were close to those obtained by linear regression. However they exhibited a high variability for soyabean meal and, to a lesser extent, for fish hydrolysate. The availability of individual amino acids was also investigated in each protein source. For some amino acids it was different from that of total nitrogen, suggesting that the availability of essential amino acids should be used for accurate feed formulation. Metabolic fecal nitrogen and amino acids were determined using a protein-free diet, in order to estimate true availabilities. These coefficients were not significantly different from apparent availability coefficients.