The aim of our study was to survey the geosmin (GSM) and 2-methylisoborneol (MIB) concentrations of the sediment and that of common carp (Cyprinus carpio) flesh in three different Hungarian fish farms (clayey, marshy, and sodic ponds).Results showed that the concentrations of off-flavour compounds of the sediment and fish fillets were related to the natural environment. The GSM concentration of bottom soil samples was higher, than MIB in each fish farm. Both off-flavour compounds were the highest in the marshy fish pond and the lowest in the sodic pond. In case of fish flesh, significant differences were found between the farms in GSM level and fat content. In the fish flesh the same tendency was found as in the sediments, but surprisingly, MIB concentration was higher in the fillets, referring to accumulation.

Fatty acid (FA) composition of the fillet and the intestinal content of dwarf common carp (Cyprinus car-pio carpio) living in Lake Hévíz was determined in wintertime collected samples and results were compared to widespread literature data on carp. Fillet FA profile of the thermally adapted (28 °C) Hévíz dwarf carps differed from profiles originated from divergent culture and feeding conditions in the overall level of saturation. Fillet myristic acid proportions largely exceeded all literature data in spite of poor dietary supply. Fillet fatty acid results indicate the effects of thermal adaptation (high saturation level) and the correlative effects of feed components rich in omega-3 fatty acids, with special respect to docosahexae-noic acid. With the application of discriminant factor analysis the Hévíz sample was accurately differentiated from the literature data on carp fillet fatty acid profile, mostly based on C14:0, C18:1 n9, C18:2 n6, C20:1 n9 and C20:4 n6 FAs. In summary, fillet FA profile suggested thermal adaptation, location specificity and the ingestion of algal and bacterial material.

Four important Hungarian common carp (Cyprinus carpio L.) breeds (Attala mirror, Attala scaled, Hortobagy scaled and Szeged mirror) harvested from different fish farms were analysed in this study. Main body indices, slaughter characteristics, fillet fat content, and conventional fish flesh quality characteristics were measured and compared. The slaughter value of the mirror strains tended to exceed that of the scaled type carps. For the calculated body indices (profile, cross-sectional, head and tail index) the influence of strain was statistically proven. Fillet fat content was significantly (P<0.01) affected by strains/ponds. The pH value of the fillet was significantly influenced by the strain as a fixed factor. Large, strain dependent variability was proven in fat content besides identical fillet dry matter contents. It was assumed that culture conditions and strain largely influence and lead to a marked variability of the body composition and flesh quality of the most important Hungarian fish species.

A feeding experiment was conducted on northern pike, Esox lucius L. (123.6 +/- 33.3 g initial body weight) applying graded dietary fish oil supplementation resulting three dietary fat levels (without supplementation: 6.2% fat and 11.7, 17.4% fat levels with supplementations) in a recirculation system. Feed consumption, feed efficiency and protein utilization of pike was not affected by the treatment. Whole body lipid content analysis showed that the composition of pike was significantly affected by the increasing level of fish oil supplementation, although no relationship was detected between the dietary and the fillet lipid content, as well as the protein content of fish bodies. High docosahexaenoic acid (DHA, C22:6n-3) proportions were found in the muscle lipids (groups fed fish oil supplementation), as compared to the dietary fatty acid compositions suggesting that with dietary fish oil supplementation the dietary precursors (mainly EPA) enable pike to convert long chain highly unsaturated fatty acids, especially DHA; resulting high DHA: EPA ratios in the fillet.

Pond-nursed pike-perch (Sander lucioperca) fry were trained to artificial diet and reared under controlled conditions maintained in aquaria using three stocking densities (1.25, 1.66 and 2.08 g/l). Two replicates per treatment were applied. The survival during the 4-week period of the experiment ranged from 44.2 to 49.6%. The majority of the losses were caused by cannibalism; only 8–14% could be attributed to natural mortality. Both cannibalism and natural mortality occurred in the first 2–3 weeks of rearing. The differences in the rate of cannibalism were independent of the stocking density. Natural mortality decreased with increasing density, consequently the highest survival was observed under the highest stocking density. Stocking density did not significantly influence growth, feed consumption and feed conversion ratio of the pike-perch fingerlings (p < 0.05).

Thecrude fat and crude protein content of fillets was predicted by X-ray computertomography (CT) and determined by chemical analyses. Between 20 to 40 serialscans were taken of 48 freshwater fish and samples from the ground fillet withskin were prepared for chemical analyses. Fish species investigated were asfollows: common carp (n = 18), grass carp (n = 10), silver carp (n = 10) andpike-perch (n = 10), all collected directly from fishponds and processedimmediately. The variables of the prediction equations were taken from thedensity values of the Hounsfield scale, on a range between −90 and+150, by summing the frequencies within each interval of 10 values. Withprincipal component analysis fillet fat and crude protein content could beestimated with R2 = 0.88 and R2 = 0.87 accuracy,respectively. Three dimensional"volumetric" estimation of fat tissue was also carried out on the basis of fatindices developed at our lab. The fillet composition of the Cyprinid speciescould be evaluated well with the applied methods, but not that of pike-perchwhich had an extremely low fillet and body fat content.