MILLIKIN: NUTRIENT REQUIREMENTS OE FISHES 



enceof 4% 18:lcu9 promoted better growth of tur- 

 bot than 1% 18:2w6 plus 4% 18:lo>9. Additionally, 

 1% cod liver oil plus 4% 18:lco9 produced better 

 growth rates of turbot than all other dietary 

 treatments (Cowey et al. 1976). Simultaneous 

 supplementation of 0.45% 18:2o>6 and 0.45% 18: 

 3a>3 as the only dietary constituents of the o»6 and 

 o»3 series did not increase the level of 20:5w3 or 

 22:6tu3 liver triglycerides in underyearling plaice 

 (Owen et al. 1972). Growth studies of red sea 

 bream indicated that a 2% dietary polyunsatu- 

 rated fatty acid mix (38% 20:5^3, 1.4% 22:5^6, 

 and 33.4% 22:6o»3) promoted significantly better 

 weight gain than fish fed up to 4.2% dietary 

 methyl linolenate (Fujii and Yone 1976). It was 

 concluded that 18:3a>3 is not essential for red sea 

 bream, since this species has little, if any, capa- 

 bility to elongate and desaturate 18-carbon fatty 

 acids to 20- and 22-carbon fatty acids, based upon 

 body fatty acid composition. 



Specific qualitative and quantitative essential 

 fatty acid requirements have not been deter- 

 mined for channel catfish fry (Yingst and Stick- 

 ney 1979) and channel catfish fingerlings (Na- 

 tional Research Council 1977; Stickney 1977). 

 Nevertheless, growth rates of channel catfish in 

 several studies comparing various lipid sources 

 (e.g., beef tallow, menhaden oil, safflower oil, 

 and corn oil) have consistently been high in indi- 

 viduals fed menhaden oil as the chief dietary 

 lipid (Stickney and Andrews 1971, 1972; Murray 

 et al. 1977; Dupree et al. 1979; Yingst and Stick- 

 ney 1979, 1980). Typically, menhaden oil con- 

 tains a large amount of fatty acids of the tu3 

 family (Stickney 1977), especially the polyun- 

 saturated fatty acids (e.g., 3% 18:3w3, 17% 20:5a>3, 

 and 13% 22:6w3). Therefore, w3 fatty acids may 

 be essential for channel catfish, especially in the 

 fry and early fingerling stages. 



Unlike most fish species studied, Tilapia zilli, 

 fingerlings require 1% 18:2cw6 or 20:4a>6 for opti- 

 mal weight gain as opposed to fatty acids of the 

 w3 series (Kanazawa et al. 1980a). However, this 

 same species has a greater relative ability to elon- 

 gate and desaturate 18:3w3 to 20:5a>3 and 22:6a>3, 

 than to elongate and desaturate 18:2cu6 to 20:4co6 

 (Kanazawa, Teshima, and Imai 1980). 



In a review of lipid requirements of fishes, 

 Castell (1979) discussed differences in fatty acid 

 composition of fishes due to salinity, tempera- 

 ture, diet composition, depth, seasonal variation, 

 and reproductive stage; requirements, metabo- 

 lism, and functions of dietary fatty acids for 

 fishes are also discussed. 



CARBOHYDRATES 



Carbohydrates are included in formulated 

 feeds for fish primarily as a low cost source of 

 energy to spare dietary protein for growth rather 

 than energy. Protein sparing action of dietary 

 carbohydrate was demonstrated in brook trout 

 fed marginal concentrations of dietary protein 

 (28 or 32%) with optimal protein-to-calorie ratios 

 of 75 mg protein/kcal (Ringrose 1971). 



Maximal dietary carbohydrate concentrations 

 that can be fed to fish without reducing growth 

 rate depend upon whether the fish species is car- 

 nivorous, omnivorous, or herbivorous. For exam- 

 ple, maximal dietary dextrin concentrations 

 that did not reduce growth rate were 10% for yel- 

 lowtail, Seriola quinqueradiata, 20% for red sea 

 bream and 30% for common carp (Furuichi and 

 Yone 1980). Rainbow trout subadults can be fed 

 38% wheat meal or 41% cooked wheat (17 to 25% 

 of dietary metabolizable energy) without dele- 

 terious effects on growth (Edwards et al. 1977). 

 Similarly, rainbow trout fed 32% wheat meal or 

 21% wheat meal plus 13% glucose (15 and 26% 

 metabolizable energy of the diet) did not have 

 significant differences in growth rate (Refstie 

 and Austreng 1981). However, dietary Cerelose 6 

 concentrations as low as 14%, substantially in- 

 creased liver glycogen concentrations of rainbow 

 trout compared with fish fed or 7% dietary 

 Cerelose (Hilton 1982). Additionally, low rearing 

 temperatures (10° vs. 15°C) for rainbow trout 

 resulted in increased liver glycogen concentra- 

 tions. Therefore, Hilton (1982) suggested that 

 stocking rainbow trout with high liver glycogen 

 concentrations into natural waters could result 

 in impaired liver function. Incipient lethal levels 

 of waterborne copper were reported to be lower 

 for rainbow trout fed higher concentrations of 

 available carbohydrate, probably as the result of 

 impaired liver function from high liver glycogen 

 content (Dixon and Hilton 1981). 



Digestibility of carbohydrates is generally in- 

 versely related to molecular complexity. Thus, 

 monosaccharides are more available nutrition- 

 ally to fishes than are disaccharides, which in 

 turn are more available than are polysaccha- 

 rides. Relative growth rates of chinook salmon 

 fingerlings fed 20% carbohydrate were as fol- 

 lows: glucose > sucrose > fructose > maltose > 

 dextrin > potato starch > galactose (Buhler and 



6 Reference to trade names does not imply endorsement by 

 the National Marine Fisheries Service, NOAA. 



665 



