The global production of freshwater fish, crustaceans, and molluscs increased two-fold within 10 years from 21.8 billion t (in 2001) to 41.7 billion t (in 2010)  . Among freshwater salmonids, farming of rainbow trout (Oncorhynchus mykiss) keeps also augmenting from 0.5 to 0.7 billion tons (2001-2010)  . Such rapidly increasing freshwater fish production requires equally rapid scientific development of sustainable and economical nutritious diets. It is crucial that such alternative diets contain all essential amino acids, fatty acids, vitamins, and minerals to satisfy the fish’s physiological requirements because dietary lack thereof may compromise somatic fish growth and health and consequently may also affect healthy food supply for humans.
A promising approach to reduce the increasingly more expensive fish meal (FM) in pellet feeds for fishes and also crustaceans are to fully or at least partially replace FM with plant sources    or other animals sources, such as poultry by-products   . Although recent research suggests that plant sources as full or partial FM replacement yield similar fish growth as conventional FM feeds   , there is experimental evidence that marine and freshwater salmonids fed diets with vegetable oils contain less omega-3 (n-3) long-chain polyunsaturated fatty acids (PUFA), in particular docosahexaenoic acids (DHA; 22:6n-3)   . Contrary to amino acids, fish can store and thus accumulate dietary fatty acids. Moreover, in addition to proteins, lipids and their fatty acids support various physiological processes, including somatic growth    , reproduction, behavior and vision, as well as immune response of fishes  . It is thus a scientific challenge to find dietary substitutions that result in similar growth and fatty acid retention in fishes and other consumers (e.g., crustaceans  ) than conventional FM feeds.
The use of regional animal by-products for freshwater aquaculture is of economic and ecological importance, especially for countries without access to the sea. Feeding studies using animal by-products as partial replacement of FM are thus far mostly conducted on marine fish (but see, e.g.,  ). For example, poultry by-products as partial replacement (30%) of FM in salmon diets resulted in reduced growth of Chinook salmon (Oncorhynchus tshawytscha), which may have been due to reduced palatability  . Using pork in fish diets as partial replacement of marine FM, no significant growth difference in Coho salmon (Oncorhynchus kisutch) was reported  , suggesting that such animal by-product meals may also be valuable for other salmonids. Thus far, it is not known how terrestrial animal by-products or vegetable feeds as partial replacement of FM affect somatic growth and, from a human nutrition perspective, highly precious n-3 PUFA in freshwater salmonids.
We conducted a feeding study to examine the effect of partial replacement of dietary FM with poultry or pork by-products, or partial replacement with vegetable feeds on the growth rate and tissue fatty acid profiles in the consumer-sized sparctic charr Salvelinus alpines X fontinalis, commonly known as Alsatian charr, used in freshwater aquaculture particularly in Europe  . We tested the null-hypothesis that there is no difference in fish growth rates or tissue fatty acid profiles in charr feeding on the different diets over their entire growth cycle. This test underlies the assumption that these dietary substitutions in fish feeds can fully replace commonly used FM resulting in equal fish growth rates. Alternatively, if fish growth is enhanced by certain fish feeds it is expected that PUFA contents in muscle tissues decrease relative to more slowly growing charr (effect of growth dilution).
2. Materials and Methods
2.1. Fish, Husbandry, and Experimental Diets
The Alsatian charr was descended from female brook trout (S. fontinalis) x male Arctic charr (S. alpinus) cross. Fish (15 - 20 g body weight at the beginning of the feeding experiments; n = 600) were held for 180 days at the aquarium facilities at the Wasser Cluster Research Centre Lunz, Austria, from June until November 2014. The experiment was conducted in a flow-through system containing twelve 1000-L rectangular tanks with a continuous supply of gravel filtered spring water (ca. 25 L min−1). Wastewater was drained using a sinkhole covered by a 5 mm mesh screen. Fish were subjected to natural photoperiod (latitude = 47.8604˚N), delivered by artificial fluorescent lighting and adjusted weekly. A total of 600 juvenile Alsatian charr were randomly distributed as 50 fish of mixed sexes per tank. Feeds were supplied to two replicate tanks per dietary treatment, thus it was not possible to statistically test for tank effects. However, we collected enough fish to conduct analysis of variance (see below) and test for differences among treatments.
Dissolved oxygen, pH and water temperature were recorded daily. Throughout this long-term feeding experiment, fish were exposed to natural variability of water temperature (9˚C to 12˚C; mean = 10.5˚C), dissolved oxygen (7 to 11 mg L−1) and slightly above neutral pH values (~8), typical for water in the limestone Alps.
Six isocaloric fish feeds were formulated (GarantTM, Austria) to provide sufficient lipid and protein to meet somatic requirements for salmonids  . Fish in duplicate tanks were fed 1 of the 6 different diets that contained 32.5% FM (F1,control) or only 17.5% FM (F2, poultry; F3, Pork SP; F4, Pork D; F5, vegetable feeds) or 25% FM (F6, Poultry Plus)as partial FM replacements (Table 1(a)). All feeds contained wheat, marine fish and rapeseed oil, wheat flower, blood meal, rapeseed cake, and wheat gluten (Table 1(a)). Diets were dispensed daily
Table 1. (a) Feed components (in %) with decreasing contents in fish meal and increase in animal by-product meals and vegetable feed, and relative mean contents (in %) of saturated, mono-unsaturated, and omega-3 and -6 polyunsaturated fatty acids; (b) proximate composition of experimental diets (g/100 g diet).
into the tank by a clockwork belt feeder (Dryden Aqua Ltd) over a 12 hrs feeding period. The daily feed ration exceeded the recommended feeding rate for salmonids for the prevailing water temperature.
2.2. Sampling Procedure
During the entire feeding experiment, every 2 wks one third of the fish in each tank was randomly selected, weighed (g) and measured (cm) for the assessment of specific growth rates and biomass. After the biomass assessments, all fish were returned to their respective tanks. The specific growth rate (SGR, % body weight day−1) was calculated as
where W0 and W1 are weights in grams per fish at the start and at the end of the feeding period, respectively, and t is the time of feeding in days. At the end of the feeding trial (after 180 days), twelve fish were selected at random from each treatment to determine lipid contents and fatty acid composition in the dorsal muscle tissue. Fish were rendered unconscious (blow on the head) and then killed by cardiac incision following the Federal Act on the Protection of Animals, Austria (http://www.ris.bka.gv.at), then a sub-sample of muscle was dissected and stored in plastic vials (8 mL). Muscle samples were obtained by cutting a fillet from above the lateral line as a border between the dorsal and ventral tissue. No skin or bone was included in the sample. All tissue samples were stored at −80˚C and subsequently freeze dried before analysis.
2.3. Proximate Analysis
The gross nutrient composition of the 6experimental diets was determined as below (Table 1(b)). Moisture was determined by drying to constant weight in an oven at 110˚C for 24 h  . Sample weight was recorded before drying and after removal from the oven. This process was repeated at 1 h intervals until the weight change was <5 mg. Total protein content in experimental diets was determined by modified Bradford assay  and total lipids by solvent extraction and gravimetric determination  . Ash content was determined by placing pre-weighed diets in a muffle furnace at 550˚C for 8 h or until white ash was obtained  that was subsequently weighed.
2.4. Lipid Extraction and Fatty acid Analysis
Total lipids from homogenized, freeze-dried dorsal muscle samples (25 - 35 mg) were analyzed as in Heissenberger et al.  . In brief, samples were sonicated and vortexed (4X) in a chloroform-methanol (2:1) mixture. Organic layers were removed and transferred into solvent-rinsed vials. For gravimetrical determination of total lipid contents (i.e., mg lipids g dry weight−1), subsamples (100 µL) of the extracts (duplicates) were evaporated and weighed. Fatty acids were derivatized to obtain fatty acid methyl esters (FAME) using toluene and sulphuric acid-methanol solution (1% v/v, 16 h at 50˚C). In contrast to Heissenberger et al.  , hexane without butylated hydroxytoluene (BHT) was used for each washing step after methylation to avoid BHT-related peak interference in chromatograms (data not shown). FAME were analyzed using a gas chromatograph (TRACE GC THERMO, Detector: FID 260˚C, Carrier gas: H2: 40 mL/min, N2: 45 mL/min, air: 450 mL/min, temperature ramp: 140˚C (5 min) − 4˚C/min − 240˚C (20 min) = 50 min) equipped with a temperature-programmable injector and an autosampler. A SupelcoTM SP-2560 column (100 m, 25 mm i.d., 0.2 µm film thickness) was used for FAME separation. Excalibur 1.4TM (Thermo Electron Corporation) was used for calculation and, if necessary, manual resetting of the chromatograms. Fatty acid concentrations were calculated using calibration curves based on known standard concentrations.
2.5. Data Analysis
Principle components analysis (PCA) was used to reduce the number of individual FA into a single FA composition score and to analyse the difference between dietary and tissue FA compositions. Significant differences in somatic growth, total lipids and fatty acid contents among dietary treatments were determined by one-way ANOVA with subsequent Tukey’s HSD tests to examine between treatment effects. Data identified as non-homogeneous, using variance test, were subjected to log transformation before applying the statistical tests. SPSS 15.0 was used for data analysis. Fatty acid retention ratios were determined as the quotient of FA in fish muscle tissues (mg FA per unit biomass) and FA in the respective diet. We define retention as the ability of fish to regulate and control ingested FA. To evaluate the effect of somatic growth differences on PUFA contents in the obtained biomassper fish after 180 d (end of this feeding experiment; PUFAfish), we calculated PUFAfish = (PUFAmt/5) × BM180, where PUFAmt was the PUFA content in muscle tissues, “5” was the measured conversion factor from freeze-dried (dry weight; dw) to wet tissue weight, and BM180 the measured biomass (wet weight; ww) of fish in all diet treatments after 180 d of feeding. Finally, we calculated the variance of PUFA (n-3 and n-6) retention as a measure of the dispersion of the PUFA around the mean.
3.1. Diet Composition
All feeds contained similar contents of total proteins (~43%), total lipids (~20%), total ash (~10%), crude fiber (1.8% - 2.7%), and moisture (~6% - 9%; Table 1(b)). As partial FM replacement in feeds F2-F6, F2 and F6 contained 15% poultry by-product meals, F3 and F417% pork by-product meals, and F5 13% soy grain GMO-free. All feeds contained wheat, marine fish and rapeseed oil, wheat flower, blood meal, rapeseed cake, and wheat gluten (Table 1(a)). The total lipid concentrations ranged from 125 - 189 mg g dry weight−1 (mean values) and were not significantly different among the 6 feeds (p = 0.31; Table 2). The fatty acid contents (mg per unit biomass) for total saturated (SAFA), mono-(MUFA), and PUFA were similar in all feeds (Table 2).
3.2. Fish Biomass and Specific Growth Rate
There was an overall statistically significant difference in fish biomass among these six feeding treatments in this experiment (F(5, 78) = 4.0, p = 0.003), with fish feeding on F6 being significantly larger than those feeding on F2 (p = 0.004), F3
Table 2. Contents (mg per g dry weight; n = 6 per treatment) of total lipids, fatty acids (>0.5 mg per g dry weight), and grouped in omega-3 (n-3) and omega-6 (n-6) polyunsaturated fatty acids (PUFA) in fish feeds and fish (Alsatian charr; Salvelinus alpinus X fontinalis). F1 = Control, F2 = Poultry, F3 = Pork SP, F4 = Pork D, F5 = Vegetable, F6 = Poultry plus.
OW-ANOVA: aF = 1.29. p = 0.31; bF = 1.91. p = 0.12; cF = 1.56. p = 0.21; dF = 1.94. p = 0.12; eF = 1.29. p = 0.31; fF = 1.79. p = 0.15; g mean n-3/n-6 values.
(p = 0.031), and F4 (p = 0.014). At the end of the feeding experiment, fish feeding on F6 enhanced their biomass by 15% relative to F1 and had the higher mean biomass (264 mg wet weight). However, fish feeding on feeds with pork by-product meals (F3 and F4) resulted on average in a 25% lower biomass gain (205 and 210 mg wet weight, respectively; Figure 1) compared to fish feeding on F6.
Somatic growth rates among the different feeding treatments were not affected by water temperatures that ranged from 9˚C - 12˚C during this experiment (p > 0.2 for all diet treatments). Fish feeding on F6 had the highest growth rate (0.6 ± 0.08), while F1, F2, and F5 fish had lower growth rates (0.54 ± 0.13, 0.54 ± 0.08, and 0.53 ± 0.16, respectively), and the lowest was detected in fish feeding on feeds containing poultry by-products (F3: 0.46 ± 0.14 and F4: 0.46 ± 0.08). The overall feed conversion ratios (FCR), determined as dry feed weight per total wet weight gain for the entire feeding period, were lower in the F1 (FCR = 1.08), F6 (FCR = 1.09) and F2 (FCR = 1.12) than in F3 (FCR = 1.28), F5 (FCR = 1.29), and F4 (FCR = 1.30).
3.3. Total Lipids and PUFA Composition in Fish
Total lipid concentrations in dorsal fish muscle tissues (n = 6 per treatment; mean values ranged from 66 - 109 mg g dry weight−1) did not significantly differ among these 6 feeding treatments (p = 0.12; Table 2). Similarly, there was no statistically significant difference observed in n-3 (p = 0.12) or n-6 (p = 0.15) PUFA in fish (Table 2). We tested also for differences in concentrations of the individual n-3 PUFA alpha-linolenic acid (ALA; 18:3n-3), eicosapentaenoic acid (EPA; 20:5n-3), and docosahexaenoic acid (DHA; 22:6n-3) because freshwater salmonids are an important dietary supply of long-chain n-3 PUFA to humans that are otherwise well provided by marine fish (   ); there was no significant
Figure 1. Average biomass (in g) of Alsatian charr (Salvelinus alpinus X fontinalis) fed diets containing 32.5% fish meal (F1; control diet), 17.5% fish meal and 15% poultry by-product meal (F2), 17.5% fish meal and 17% pork by-product meals (F3 and F4), 17.5% fish meal and 13% soy grain feeds (F5), and 25% fish meal and 15% poultry by-product meal (F6).
difference in dorsal muscle tissues of Alsatian charr in ALA (F = 1.39; p = 0.25), EPA (F = 1.06; p = 0.4), or DHA concentrations (F = 0.97; p = 0.45) among these 6 different feeding treatments. Fish in these treatments contained on average 29% - 34% PUFA per total lipid content and retained on average 1.2 - 1.3X more PUFA per total lipids than their diets; there was no significant difference in PUFA retention (normalized to total lipids; p = 0.43). However, relative to its dietary supply, DHA was less strongly retained in dorsal muscle tissues of F1 fish (only 1.6X more DHA) than in all other fish that were feeding on meals partially replaced by other animal or vegetable by-products (2 - 2.5X more DHA than their dietary DHA supply). Similarly, the retention of the omega-6 long-chain PUFA arachidonic acid (ARA; 20:4n-6) in dorsal muscle tissues was lower in F1 fish (1.3X) than in other fish (1.4 - 1.8X; Table 3).
The first and second principle components (PC1 and PC2) together of the PCA explained 66% of the variability in eigenvalues (PC1: 40.5%, PC2: 25.5%). The proportions of n-3 PUFA and SAFA were positively correlated, whereas the proportions of the long-chain SAFA 20:0, 22:0, and 24:0, used as fatty acid markers of terrestrial sources, were negatively correlated to the PC1 scores (Figure 2). Omega-6 PUFA were negatively correlated to the PC2. Thus, feeds were clearly separated by terrestrial fatty acids, whereas fish were strongly associated with n-3 PUFA.
Omega-3 PUFA contents per total individual fish weight at the end of the feeding experiment (after 180 days) showed lowest n-3 PUFA contents (ww) in F1-fish (640 ± 90 mg per fish; ww) and highest in F6-fish (866 ± 99 mg per fish), however, without any significant differences among these 6 treatments (OW-ANOVA; F = 1.7, p = 0.16; Figure 3(a)). Omega-6 PUFA contents in fish were also lowest in F1-fish (322 ± 85 mg per fish) and highest in F3-fish (845 ± 308 mg per fish), but not significantly different among these treatments
Table 3. Retention ratios of polyunsaturated fatty acids (PUFA) contents (mg g dry weight−1; calculated as ratios between PUFA contents in fish and feeds) in Alsatian charr (S. fontinalis X alpinus) fed on 6 different feeds: F1 = Control, F2 = Poultry, F3 = Pork SP, F4 = Pork D, F5 = Vegetable, F6 = Poultry plus.
Figure 2. Principle components analysis of fatty acids in experimental feeds and dorsal muscle tissues in Alsatian charr (Salvelinus alpinus X fontinalis) supplied with 6 different feeds: 32.5% fish meal (F1; control diet), 17.5% fish meal and 15% poultry by-product meal (F2), 17.5% fish meal and 17% pork by-product meals (F3 and F4), 17.5% fish meal and 13% soy grain feeds (F5), and 25% fish meal and 15% poultry by-product meal (F6).
(OW-ANOVA; F = 1.5, p = 0.22; Figure 3(b)). Finally, the variance of retained n-6 PUFA (55181) in all fish was almost twice as large as that of n-3 PUFA (30685).
This long-term feeding experiment shows that poultry by-products as partial FM replacement (~25% less FM) caused the highest weight gain in Alsatian charr (F6) during 180 days of feeding. The final weight gain in Alsatian charr was very
Figure 3. Contents (mg; mean ± standard deviation) of omega-3 (grey bars) and omega-6 (white bars) polyunsaturated fatty acids (PUFA) per total weight of individual fish at the end of the feeding experiment (after 180 days) in 6 feeding treatments: 32.5% fish meal (F1; control diet), 17.5% fish meal and 15% poultry by-product meal (F2), 17.5% fish meal and 17% pork by-product meals (F3 and F4), 17.5% fish meal and 13% soy grain feeds (F5), and 25% fish meal and 15% poultry by-product meal (F6).
similar when feeding on conventional FM feed (F1; containing 32.5% FM), feeds containing poultry by-products (F2) or vegetable diets (F5) as partial FM replacement (~50% less FM). However, pork by-products as partial FM replacement in feeds F3 and F4 (~50% less FM) resulted in the lowest fish weight gain, indicating that animal by-products in isocaloric fish feeds can still cause differences in somatic growth of this freshwater salmonid. At the same time, such diet composition induced differences in fish weight did not significantly affect the retention of fatty acid contents in muscle tissues.
The largest weight gain in F6 coincides with the highest dietary contribution of animal meals in this feed (40%), compared to 32.5% FM in F1, 34.5% animal meals in F2, F3, and F4, or only 17.5% FM in F5. All of these feeds were isocaloric and differed only in their ingredient composition. It is likely that F6 with the highest animal meal contribution and slightly higher amounts of wheat in the feed caused higher weight gain in F6-fish. Bureau et al.  reported similar digestibility of animal protein ingredients in feeds for rainbow trout and Dosanjh et al.  found no biomass differences in marine salmon that were fed on pork by-product meals. This is in contrast to our finding that charr grew less efficiently when feeding on feeds containing pork by-product meals (F3 and F4). It is likely that pork by-product meals may have been less digestible for charr than F1 and the poultry by-product feeds as indicated by their higher FCR. Lower digestibility of pork than poultry by-product meals was recently reported in a feeding study on Pacific white shrimp (Litopenaeus vannamei)  . It is also possible that other unidentified ingredients in pork by-product meals caused lower weight gain because charr feeding on F5 (vegetable feed by-products) had similar FCR as F3 and F4, yet higher weight gain. These results indicate, however, that the higher contribution of animal meals in F3 and F4 (34.5%) compared to F5 (with only 17.5% animal meal) did not account for the higher weight gain measured in F5 charr. Furthermore, the similar weight gain in fish feeding on F1, F2, or F5 demonstrates that this freshwater salmonid can be supplied with 50% lessmarine FM in their diet and that poultry by-products or vegetable feeds as partial FM replacement result in very similar final fish biomass. Thus, 50% reduction of FM and proportional replacement by poultry by-products or vegetable feeds can fully replace commonly used FM in feeds for Alsatian charr.
Contrary to our hypothesis that charr with highest somatic growth will result in lowest PUFA contents in muscle tissues (effect of growth dilution), there was no significant difference in charr PUFA contents among these treatments. This suggests that independent of weight gain the dietary PUFA supply was not limiting PUFA retention in Alsatian charr. It can be consequently speculated that, in addition to FM replacements, slight reductions of dietary PUFA supply may have little effects on PUFA retention in this sparctic charr. It is, however, clear that full replacement of fish oil with vegetable oil may also reduce fish-derived long-chain n-3 PUFA content in charr or other aquaculture-raised freshwater salmonids  , although fish weight gain may not be equally compromised  . Thus, further research should aim at identifying by how much dietary fish oil can be replaced with other oil sources in a combined effort to reduce both FM and highly valuable fish oil in aquaculture feed production for freshwater salmonids and other freshwater fish.
Principle components analysis revealed that charr of all feeding treatments were strongly associated with n-3 PUFA, independent of diet ingredients and somatic growth differences. The most highly retained n-3 PUFA was DHA in fish of all feeding treatments, whereas all other identified n-3 PUFA contents were lower than in their dietary supply, indicating that DHA is the most required n-3 PUFA in Alsatian charr, as is the case of freshwater fish in general  . However, F1-fish retain on average only 1.6 times more DHA relative to its dietary supply, whereas it was 2.4 to 2.5 times more in F4 and F3-fish, respectively. Similarly, the n-6 PUFA gamma-linolenic acid (18:3n-6) and ARA were effectively retained in fish of all treatment, but the lowest retention in F1-fish and highest in F3 and F4-fish suggests that PUFA retention efficiency in Alsatian charris affected by the feed composition and not directly regulated by dietary fatty acid supply. Finally, the higher variance in n-6 than n-3 PUFA retention in fish among these different feeding treatments suggests that, independent of these diet compositions, Alsatian charr retained n-3 PUFA more consistently than n-6 PUFA.
In conclusion, this long-term feeding trial of 180 days furthers our knowledge that reducing marine fish meal by 50% and using poultry by-product meal and vegetable feeds, both used as partial replacements of marine fish meal, results in similar weight gains of this sparctic charr. When marine fish meal was only reduced by 25% and substituted by poultry by-product meal, the fish biomass was clearly enhanced by 15% relative to conventional fish meal feeds. However, it became evident that pork by-product meal as partial replacement yielded lower biomass gain than the conventional marine fish meal. The use of poultry or pork by-product or vegetable meals as partial fish meal replacements did not significantly change the total lipids or fatty acid contents in these fish, suggesting that these alternative feeds, yet with similar lipid sources, have very little effect on the fatty acid composition in this freshwater salmonid. Further studies are warranted to assess the effect of increasing use of animal by-product and vegetable meals as well as lower fish oil contribution to fish feeds on weight gain and fatty acid composition in freshwater salmonids that are increasingly raised and eaten by humans worldwide  .
This work was financially supported by GARANT Tiernahrung Austria and the Province of Lower Austria. We are grateful for technical assistance and support from Katharina Winter, Katharina Hader and Stefanie Danner for their laboratory assistance.
 Hixson, S.M., Sharma, B., Kainz, M.J., Wacker, A. and Arts, M.T. (2015) Production, Distribution, and Abundance of Long-Chain Omega-3 Polyunsaturated Fatty Acids: A Fundamental Dichotomy between Freshwater and Terrestrial Ecosystems. Environmental Reviews, 23, 414-424.
 Murray, D.S., Hager, H.H., Tocher, D.R. and Kainz, M.J. (2014) Effect of Partial Replacement of Dietary Fish Meal and Oil by Pumpkin Kernel Cake and Rapeseed Oil on Fatty Acid Composition and Metabolism in Arctic Charr (Salvelinus alpinus). Aquaculture, 431, 85-91.
 Benedito-Palos, L., Navarro, J.C., Sitja-Bobadilla, A., Bell, J.G., Kaushik, S. and Perez-Sanchez, J. (2008) High Levels of Vegetable Oils in Plant Protein-Rich Diets Fed to Gilthead Sea Bream (Sparus aurata L.): Growth Performance, Muscle Fatty Acid Profiles and Histological Alterations of Target Tissues. British Journal of Nutrition, 100, 992-1003.
 Fuertes, J.B., Celada, J.D., Carral, J.M., Saez-Royuela, M. and Gonzalez-Rodriguez, A. (2013) Replacement of Fish Meal with Poultry By-Product Meal in Practical Diets for Juvenile Crayfish (Pacifastacus leniusculus Dana, Astacidae) from the Onset of Exogenous Feeding. Aquaculture, 404, 22-27.
 Subhadra, B., Lochmann, R., Rawles, S. and Chen, R. (2006) Effect of Fish-Meal Replacement with Poultry By-Product Meal on the Growth, Tissue Composition and Hematological Parameters of Largemouth Bass (Micropterus salmoides) Fed Diets Containing Different Lipids. Aquaculture, 260, 221-231.
 Stubhaug, I., Lie, O. and Torstensen, B.E. (2007) Fatty Acid Productive Value and Beta-Oxidation Capacity in Atlantic Salmon (Salmo salar L.) Fed on Different Lipid Sources along the Whole Growth Period. Aquaculture Nutrition, 13, 145-155.
 Tocher, D.R., Dick, J.R., MacGlaughlin, P. and Bell, J.G. (2006) Effect of Diets Enriched in Delta 6 Desaturated Fatty Acids (18 : 3n-6 and 18 : 4n-3), on Growth, Fatty Acid Composition and Highly Unsaturated Fatty Acid Synthesis in Two Populations of Arctic Charr (Salvelinus alpinus L.). Comparative Biochemistry and Physiology B-Biochemistry & Molecular Biology, 144, 245-253.
 Bell, J.G., Henderson, R.J., Tocher, D.R. and Sargent, J.R. (2004) Replacement of Dietary Fish Oil with Increasing Levels of Linseed Oil: Modification of Flesh Fatty Acid Compositions in Atlantic salmon (Salmo salar) Using a Fish Oil Finishing Diet. Lipids, 39, 223-232.
 Bendiksen, E.A., Arnesen, A.M. and Jobling, M. (2003) Effects of Dietary Fatty Acid Profile and Fat Content on Smolting and Seawater Performance in Atlantic Salmon (Salmo salar L.). Aquaculture, 225, 149-163.
 Caballero, M.J., Obach, A., Rosenlund, G., Montero, D., Gisvold, M. and Izquierdo, M.S. (2002) Impact of Different Dietary Lipid Sources on Growth, Lipid Digestibility, Tissue Fatty Acid Composition and Histology of Rainbow Trout, Oncorhynchus mykiss. Aquaculture, 214, 253-271.
 Izquierdo, M.S., Socorro, J., Arantzamendi, L. and Hernandez-Cruz, C.M. (2000) Recent Advances in Lipid Nutrition in Fish Larvae. Fish Physiology and Biochemistry, 22, 97-107.
 Arts, M.T. and Kohler, C.C. (2009) Health and Condition in Fish: The Influence of lipids on Membrane Competency and Immune Response. In: Arts, M.T., Brett, M.T. and Kainz, M.J., Eds., Lipids in Aquatic Ecosystems, Springer, New York, 237-256.
 Dosanjh, B.S., Higgs, D.A., Plotnikoff, M.D., McBride, J.R., Markert, J.R. and Buckley, J.T. (1984) Efficacy of Canola Oil, Pork Lard and Marine Oil Singly and in Combination as Supplemental Dietary-Lipid Sources for Juvenile Coho Salmon (Oncorhynchus kisutch). Aquaculture, 36, 333-345.
 Gross, R., Gum, B., Reiter, R. and Kuhn, R. (2004) Genetic Introgression between Arctic Charr (Salvelinus alpinus) and Brook Trout (Salvelinus fontinalis) in Bavarian Hatchery Stocks Inferred from Nuclear and Mitochondrial DNA Markers. Aquaculture International, 12, 19-32.
 Bell, J.G., Tocher, D.R., Henderson, R.J., Dick, J.R. and Crampton, V.O. (2003) Altered Fatty Acid Compositions in Atlantic Salmon (Salmo salar) Fed Diets Containing Linseed and Rapeseed Oils Can Be Partially Restored by a Subsequent Fish Oil Finishing Diet. Journal of Nutrition, 133, 2793-2801.
 Murray, D.S., Bain, M.M. and Adams, C.E. (2013) Adhesion Mechanisms in European Whitefish Coregonus lavaretus Eggs: Is This a Survival Mechanism for High-Energy Spawning Grounds? Journal of Fish Biology, 83, 1221-1233.
 Heissenberger, M., Watzke, J. and Kainz, M.J. (2010) Effect of Nutrition on Fatty Acid Profiles of Riverine, Lacustrine and Aquaculture-Raised Salmonids of Pre-Alpine Habitats. Hydrobiologia, 650, 243-254.
 Kainz, M., Arts, M.T. and Mazumder, A. (2004) Essential Fatty Acids in the Planktonic Food Web and Their Ecological Role for Higher Trophic Levels. Limnology and Oceanography, 49, 1784-1793.
 Sidhu, K.S. (2003) Health Benefits and Potential Risks Related to Consumption of Fish or Fish Oil. Regulatory Toxicology and Pharmacology, 38, 336-344.
 Bureau, D.P., Harris, A.M. and Cho, C.Y. (1999) Apparent Digestibility of Rendered Animal Protein Ingredients for Rainbow Trout (Oncorhynchus mykiss). Aquaculture, 180, 345-358.
 Villarreal-Cavazos, D.A., Ricque-Marie, D., Pena-Rodriguez, A., Nieto-Lopez, M., Tapia-Salazar, M., Lemme, A., Gamboa-Delgado, J. and Cruz-Suarez, L.E. (2014) Apparent Digestibility of Dry Matter, Crude Protein, and Amino Acids of Six Rendered By-Products in Juvenile Litopenaeus vannamei. Ciencias Marinas, 40, 163-172.
 Torstensen, B.E., Bell, J.G., Rosenlund, G., Henderson, R.J., Graff, I.E., Tocher, D.R., Lie, O. and Sargent, J.R. (2005) Tailoring of Atlantic Salmon (Salmo salar L.) Flesh Lipid Composition and Sensory Quality by Replacing Fish Oil with a Vegetable Oil Blend. Journal of Agricultural and Food Chemistry, 53, 10166-10178.