Alternative feed ingredients for intensive aquaculture.

Abstract

98 Alternative feed ingredients for intensive aquaculture G.L. Allan NSW Fisheries, Port Stephens Research Centre, Taylors Beach, NSW 2316 Summary The rapid expansion in aquaculture has accelerated the search for alternatives to feed ingredients of marine origin. Some Australian agricultural ingredients are already widely used in aquafeeds and considerable potential exists to increase their use both domestically and overseas. Plant breeding programs to reduce concentrations of anti-nutritional factors and increase essential nutrients, and processing to remove less digestible components such as fibre and carbohydrates will improve the usefulness of vegetable ingredients. For animal by-products, reducing ash and saturated fat levels and improving and standardising processing conditions will increase the scope for use of these products. For all ingredients the use of attractants, synthetic amino acids and enzymes offers the potential for formulation of successful aquaculture diets with reduced reliance on marine ingredients. and utilised. However, production of fishmeal already uses approximately 33% of the total global fish catch and the proportion of fishmeal used for aquaculture is predicted to rise to between 25-30% within the next decade (Tacon, 1996). In the same period, the proportion of fish oil used for aquaculture is expected to reach 3050% of total production (Tacon, 1996). As about 4 kg wet fish is needed to produce 1 kg fishrneal, if diets contain more than 17% fishmeal and/or the food conversion ratio exceeds 1.5: 1 or both, the aquaculture operation entails a net loss of fish protein. In this paper, Australian ingredients with the potential to replace fishmeal and other marine ingredients will be reviewed and constraints to their use discussed. Methods for evaluating these ingredients and increasing their use will be examined. Introduction Demand for seafood is escalating as global population rises and the popularity of seafood increases (Anon., 1994, Liao, 1996). Production fi-om capture fisheries is declining and aquaculture offers the only chance to meet this demand. Aquaculture production has risen from 15% of total fisheries production in 1989 to 22% in 1993 (Liao, 1996). A shift to more intensive culture practices, made possible by the availability of better, formulated diets, has been partly responsible for the increase in aquaculture production. From 1986 to 1990, Akiyama (199 1) estimated that demand for aquafeeds in Asia increased more than four-fold and New and Csavas (1993) predicted the Asian aquafeed market would reach 2.6 million tonnes by the year 2000. Marine based ingredients, especially fishmeal and fish oil, are preferred protein and energy sources as they provide high levels of essential amino and fatty acids, are low in carbohydrates and are well digested Recent Advances in Animal Nutrition in Australia 7997 University of New England, Armidale NSW 2351 , Australia Identifying alternative feed ingredients to those of marine origin To evaluate feed ingredients, information is needed on availability and price, the biochemical composition (proximates, amino acids, fatty acids, carbohydrates and anti-nutritional factors) and digestibility and availability to the target species. Although very little fishmeal is produced in Australia, abundant supplies of terrestrial agricultural ingredients are available and large volumes are exported. Table 1 lists production and export volume and average export price of major commodities. A large number of grains and grain by-products are used in aquafeeds, ranging from high quality soybean meal to cereals like wheat and rice. The most commonly available oilseed meals in Australia include soybean meal, canola meal, peanut meal, cottonseed meal and sunflower meal. Globally, soybean meal is probably the most widely used plant protein source for Alternative feed ingredients for intensive aquaculture 99 aquaculture diets (New et al. 1993). Grain legumes, excluding soybeans and peanuts which are typically considered as oilseeds, usually have a lower crude protein content but are also used widely as animal feed ingredients. Lupins, mungbeans, chick peas, cow peas and field peas are examples. Cereals generally have the lowest protein content but can be important sources of energy and useful for binding diets. Cereal byproducts, from which most of the starch has been removed, can be much higher in crude protein (>60%). Corn gluten meal and wheat gluten meal are examples. In Australia wheat, barley, oats, sorghum and triticale are produced in large quantities (Table 1). Animal by-products can be very usetil ingredients with a relatively high crude protein content and a valuable source of essential amino acids. Bloodmeal, meat and bone meals, ungraded slaughterhouse wastes and other by-products of beef, sheep, pigs and poultry are widely available. Brewing residues, such as distillers grains and solubles and brewers draff have been successfully used as ingredients in aquafeeds for a number of species (Kohler and Pagan-Font, 1978; Table 1 Production and export volumes of Australian Agricultural Feed Ingredients 1992/93'. Hughes, 1987; Webster et al. 1992). General discussions of benefits and limitations of different types of ingredients can be found in Evans (1985), Hardy (1989), HardyandDong(1995)andSwick(1995). Price is a major regulator of which ingredients will be considered for inclusion in aquaculture diets, especially where there are many to choose from. Fishmeal is the protein source of choice for most formulated aquafeeds and ranged in price in 1992 from $650-$1,300/t depending upon source, protein content and quality and country of purchase (New et al. 1993, NSW Agriculture, Sydney Retail Feed Ingredient Prices, 1992) [NSW Agriculture, Kite Street, Orange NSW 28001. Fish oil is also traded on the international commodity market and prices range from approximately $1,000-$2,000/t. Alternative protein feed ingredients will usually need to compete economically on a price per unit protein (or limiting amino acids) with fishmeal. Consistency of composition and availability is critical; feed manufacturers must be able to access ingredients when they need them and have confidence that the nutrient composition will be similar for different batches of the same ingredient. Biochemical composition of ingredients will determine their consideration as replacements for marine ingredients. For vegetable ingredients, composition will vary depending upon which cultivars are grown, soil and weather conditions, and processing and storage methods. For animal by-products, composition will depend upon the species composition, rendering equipment and methods and storage conditions. Even so, sufficient information is available for an initial assessment (for example Evans, 1985; AEC, 1987; Novus, 1992; NRC, 1993; New, 1987; Petterson and Mackintosh, 1994; New et al. 1993). Analysed dry matter, energy and protein for a range of ingredients used, or considered for use, in aquafeeds in Australia is presented in Table 2, and essential amino acid composition of a subset of these ingredients is presented in Table 3. Table 2 Analysed composition for dry matter, energy and protein for various feeds available in Australia (Allan et a/. unpublished data, 1993-l 996). 102 Al/an, G . Evaluating feed ingredients Armed with information on availability, price and composition, the next step is to determine digestibility for the target species. Measuring digestibility of an ingredient does not take into account losses which can occur in the production of urine and heat but normally accounts for the majority of the differences between ingredients for fish (Lovell, 1989). (Cho et al. 1982 or Cho and Kaushik 1990.) In addition, digestibility coefficients are additive (Cho et. al. 1982; Allan et al. unpublished data), so digestibility of diets can be calculated based on digestibility values for individual ingredients. Digestibility coefficients for a number of ingredients available in Australia fed to silver perch (Bidyanus bidyanus) are presented in Table 4. Aquaculture species have varying capacities to digest different ingredients, depending largely upon their digestive system and the presence and activity of various endogenous enzymes (Wee, 1992). Digestibility coefficients for energy and protein for a number of aquaculture species for some common feed ingredient, are presented in Table 5. Although the various methodologies used to calculate these values will have influenced the results, the comparison is useful to indicate differences between species with different digestive systems. Digestibility information allows ingredients to be compared on the basis of cost of digestible protein or digestible limiting amino acids (e.g. lysine). After digestibility information is available, the maximum amount of an ingredient which can be used without suppressing growth or causing other adverse effects, should be determined. With this information, diets containing alternative ingredients to those of marine origin can be formulated on a least-cost basis. Ultimately diets containing new ingredients need to be validated under commercial farming conditions. Table 4 Digestibility coefficients for dry matter, energy and protein for various feeds (values are means, n=3) fed to silver perch (Bidyanus bidyaws). I Alternative feed ingredients for intensive aquaculture 703 104 Allan, G . Table 6 Anti-nutritional factors present in oilseeds and grain legumes currently used in aquaculture diets'. Alternative feed ingredients for intensive aquaculture 705 Major constraints to replacing marine ingredients Compared with fishmeal, grains contain large amounts of carbohydrates, including fibre, and some species contain anti-nutrients or may be contaminated by mycotoxins. Grains and ingredients of animal origin are often deficient in essential amino acids and essential fatty acids, compared with marine ingredients, and this can suppress palatability and attractiveness of diets. Carbohydrate includes starches, non-starch polysaccharides, oligosaccharides and some free sugars. No requirement for carbohydrates has been demonstrated for fish (NRC, 1993), although they can provide an energy source and reduce the need for using expensive protein for energy. Carbohydrates, especially starch, also have an important role in binding extruded and pelleted feeds. The ability of different species of fish to utilise carbohydrate will limit the inclusion level of many unprocessed grains. Although enzymes necessary for carbohydrate digestion have been detected in fish, some species are clearly better able to digest carbohydrates than others (NRC, 1993). The digestibility of carbohydrates is influenced by the digestive system of the fish, with carnivorous species least equipped to digest carbohydrates. Processing, e.g. cooking or steam treatment also influences digestibility, as does the structural complexity of the carbohydrate (NRC, 1993 ; Robinson, 1989). Grains can also contain a number of antinutritional factors, including trypsin inhibitors, gossypol, glucosinolates, erucic acid, haemagglutinating agents, cyclopropenoic fatty acids and alkaloids (Table 6). Mycotoxins produced by fungi can also contaminate feed ingredients and formulated feeds. Peanut meal is particularly susceptible to contamination, but other grains are also affected. Contamination can occur during growth of the crop or storage and distribution of grains or feeds (Williams and Blaney, 1992). Another limitation with the use of grains is the presence of phytates. Phytates are found in all plant materials and are the major storage form of phosphorus in seeds (Reddy et al. 1982). The amount of phosphorus present as phytate in grains ranges f?om about 40-90% of the total phosphorus (Ravindran and Blair, 1992) and is considered to be almost unavailable to fish (NRC, 1993). In addition, phytates may reduce the bioavailability of protein and several essential minerals (NRC, 1993). Compared with marine ingredients, other ingredients are usually deficient in essential amino acids, particularly lysine and methionine. The essential amino acid content (as a percentage of protein) of a number of feed ingredients, including grains, is listed in Table 3. Maximum inclusion levels of ingredients in formulated diets will depend not only upon composition and digestibility, but also on the presence of antinutritional factors. Although meatmeal has fewer antinutritional factors than plant protein sources, it can contain high contents of bone fragments which can be deleterious. Excessive heat during the rendering process can damage proteins, especially lysine, and may contribute to lower protein digestibility. Consistent temperature throughout rendering facilities is important (Carpenter and Booth, 1973). Excessive amounts of hair or wool also make processing difficult, as can high contents of fat. In general, provided essential fatty acid requirements are met, saturated animal fats have no adverse effects on fish (Rein&, 1980) and they are a good, cheap source of energy. However, fish fed diets with high concentrations of saturated fat tend to have a body composition lower in unsaturated fatty acids which may become a marketing disadvantage. Reduction of fat content, through mechanical or chemical extraction, will result in meals with a higher protein content which is an advantage for aquaculture diet formulation. Contamination of meatmeal products with pesticides or bacteria, particularly salmonella, is a genuine concern and industry specifications on these contaminants are needed (Australasian Agribusiness Services, 1993). Concern with exotic diseases like bovine spongiform encephalopathy (or Mad Cow Disease) has reduced use of meat products in animal feeds in some countries (Australasian Agribusiness Services, 1993). One of the major factors which has prevented the use of meatmeals in animal feeds has been inconsistent composition. This was recognised in Australia in the review commissioned by the Meat Research Corporation into the meatmeal and tallow industry and markets (Australasian Agribusiness Services, 1993). The inconsistency in the composition of meat meals is especially notable when compared with vegetable protein sources such as soybean meal. The variability is a result of a number of factors, including the differing nature of raw materials, especially where mixed species are rendered. The practice of rendering processors to `take what's left' contributes to this variability. Improving the nutritional value of alternative ingredients Plant breeding programs have been very successful in improving the nutritive value of some grains. Varieties of maize which are high in lysine and tryptophan (Opaque -2) or lysine and methionine (Floury -2) are examples (Farrell, 1992). Low glucosinolate, low erucic acid varieties of rapeseed (called canola), low alkaloid varieties of lupins (to improve palatability) and tanninf?ee cultivars of faba beans have all been produced for livestock feeding (Farrell, 1992). f 0 6 Alan, G. Changes in the nutritive value of feeding ingredients can also be achieved by processing, including grinding, classification, sieving, mixing, heating, drying, and extrusion. Grinding to reduce particle size improves digestibility and is especially important for crustacean diets. Many forms of processing involve heat treatment, including pelleting and extrusion. Heat is important to deactivate some of the anti-nutritional factors present in grains, such as trypsin inhibitors (NRC, 1993), and can also be used to gelatin&e starch compounds which usually improves digestibility (Hardy, 1989; Table 7). Heating can also have detrimental effects including reducing the digestibility of some essential amino acids (Hardy, 1989). Lysine and cystine are the most likely amino acids to be adversely affected, but digestibility of arginine, threonine, leucine and tryptophan may also be affected (Ravindran and Blair, 1992). Heat sensitive vitamins, e.g. ascorbic acid, may also be damaged by some processing treatments which involve heating (Halver, 1989). Increasing the protein content of grains by removing some of the carbohydrate material should enable use of higher contents of grains in aquafeeds. Dehulling and protein fractionation are examples. Effects of dehulling and removing some of the carbohydrate fraction of some Australian grain legumes on dry matter, energy and protein digestibility to silver perch are presented in Table 8. Wheat or corn gluten meals, produced by removing starch, are generally highly digestible to fish (Table 4; Allan, 1995; unpublished data) although they are often expensive, and maximum inclusion of wheat gluten meal is limited by the agglutinating properties of this product. The Academy of Grain Technology in Australia is currently investigating ways to produce much cheaper wheat gluten, with reduced agglutinating properties, for use in animal feeds. There has been increasing interest in using exogenous enzymes to improve utilisation of nutrients in animal feeds. These include proteases, cellulases, pectinases, P-glucanases, lipases and phytases (Batterham, 1992). Use of these products offers the potential to increase the use of non-marine ingredients, especially grains, although efficacy for some products with fish has not yet been clearly demonstrated. For rainbow trout, supplementation of plant protein sources with phytase significantly increased phosphorus availability (Riche and Brown, 1996). Meat and bone meals and poultry waste products can be improved through the reduction of bone and fat. In studies with silver perch, digestible dry matter, energy and protein all increased for meat meals with more protein (through reduction in ash-bone) (Allan, 1994). Supplements of synthetic amino acids such as Llysine, DL-methionine and DL-threonine are used extensively and successfully in pig and poultry diets (Batterham, 1992). Unfortunately, synthetic amino acids leach rapidly in water and they are absorbed much more rapidly than protein-bound amino acids. Fish can utilise free amino acids, although their efficiency in overcoming deficiencies is the subject of some debate (Lovell, 1989; Cowey, 1992; Murai, 1992). Ingredients of marine origin contain various attractants and are usually highly palatable. Replacement of these ingredients can lead to problems with reduced feed intake and deterioration in performance. Not only may feeding stimulants be diluted or removed, but some ingredients actually contain Table 7 Digestibility coefficients for cooked and uncooked, reference diet (Allan and Rowland 1992), starch products and pregelled corn starch fed to silver perch (Bidyanus Myanus) . Alterna five feed ingredients for intensive aquaculture 707 Table 8 Composition and digestibility of processed and unprocessed legumes fed to silver perch (Bidyanus bidyanus) diets. deterrents (Mackie and Mitchell, 1985). It may be possible to address problems with reduced diet attractiveness and palatability by the addition of stimulants and `palatability enhancers' (Mackie and Mitchell, 1985; Viana et al. 1994). Examples of substances which have been shown to improve feeding behaviour include amino acids (especially in dipeptide linkages), betaine, inosine and organic acids (NRC, 1993). Carnivores tend to show the most positive response to alkaline and neutral additives such as glycine, proline, taurine, valine and betaine, while herbivores respond more positively to more acidic additives like aspartic acid and glutamic acid (NRC, 1993). Development Corporation recognised the importance of fishmeal replacement research and created a separate Sub-Program to coordinate national research in 1993. Additional funding for this research has been provided by the Australian Centre for International Agricultural Research (for collaborative research with the Thailand Department of Fisheries), the Australian Grains Research and Development Corporation, the Australian Meat Research Corporation and the Australian Academy of Grain Technology. The overall aim of the Sub-Program is to produce cost+Bective aquafeeds, specifically by replacing imported fishrneal (trash fish ' in Thailand) with cost-effective locally produced alternative protein sources. In summary, the methods used to achieve this objective include: l Fishmeal replacement research in Australia In Australia, with little fishmeal production but abundant agricultural production, aquaculture will not develop unless aquafeeds based on agricultural proteins are developed. The Australian Fisheries Research and Identifying and then evaluating alternative protein sources (digestibility, net energy utilisation, tracking stable isotopes, growth responses) Developing and evaluating processing methods to improve utilisation of ingredients Evaluating methods of increasing inclusion by using supplements, e.g. synthetic amino acids and enzymes l l 108 Al/an, G. 0 Examining the role of attractants Defining requirements for nutrients,e.g. amino acids and energy, where previously reported requirements are limiting fishmeal replacement. Allan, G (1994). Preliminary evaluation of meatmeal in aquaculture diets for silver perch (Bidyanus Port Stephens Research Centre, Taylors Beach, NSW, 2316. pp. 49. Allan, GL. (1996). Potential for replacement of marine ingredients in Asian aquafeeds. In: Proceedings, Victam Asia Conference, 1996, November 14-l 5 1996, Bangkok, Thailand. Anon. (1994). Status of World Aquaculture 1995. Aquaculture Magazine Buyers Guide 1996, pp. 6-27. Australasian Agribusiness Services. (1993). The meatmeal and tallow industry and itsmarkets. Main report prepared for the Meat Research Corporation. bi&anus). Final report to Meat Research Corporation, Project Number M.561. NSW Fisheries, l The research was perceived to be of value to most fish and crustacean species, but for practical purposes it was decided to concentrate efforts on four `representative' species in Australia; sea bass (Lates calcarifer), shrimp (Penaeus monodon), silver perch (Bid-yams bit&anus) and salmon (Salmo salar), and on hybrid walking catfish (Clarias macrocephalus X C. gariepinus) in Thailand. Australasian Agribusiness Services Pty Ltd, Australia., pp. 91. Acknowledgments Much of this information was extracted from a paper (Allan, 1996) presented at Victam International, Feed and Food Industries Show, November 14-15, 1996, Bangkok, Thailand. Feed Production on the Threshold of the Next Age. Victam International, Leidein, The Netherlands, pp. 125-152). Technical assistance for research with silver perch was provided by Scott Parkinson, David Stone, Jane Frances, Rebecca WarnerSmith, Ken O'Brien and Wendy Peasley. Jo Pickles and Jane Frances assisted with the preparation of the manuscript. Support from the Australian Fisheries Research and Development Corporation (Fishmeal Replacement Sub-Program), Australian Centre for International Agricultural Research, Australian Meat Research Corporation and the Australian Academy of Grains Technology for fishmeal replacement research is gratefully acknowledged. Batterham, E. S. (1992). Development of cost-effective diets for the pig industry: how to utilise low quality ingredients to formulate cost-effective diets. In: Proceedings of Aquaculture Nutrition Worbhop, (Eds. G L. Allan and W. Dali,), Salamander Bay, 1517 April 1991. NSW Fisheries, Brackish Water Fish Culture Research Station, Salamander Bay, Australia. pp. 112-l 17. Carpenter, K. J. and Booth, V. H. (1973). Damage to lysine in food processing: itsmeasurement and its significance. Nutrition Abstracts and Reviews, 43, 424A5 1. Cho, C. Y. and Kaushik, S. J. (1990). Nutritional energetics in fish: energy and protein utilisation in rainbow trout (Salmo gairdneri). World Review of Nutrition and Dietetics, 61, 132-172. Cho, C. Y., Slinger, S. J. and Bayley, D. S. (1982). Bioenergetics of salmonid fishes: energy intake, expenditure and productivity. Comparative Biochemistry and Physiology, 73 (B)(l), 25-41. Cowey, C. B. (1992). Nutrition: estimating requirements of rainbow trout. AquacuZture 100,177-l 89. ABARE Australian Bureau of Agricultural Resource Economics. (1994). Grainstatistics 1994. ABARE, Canberra, ACT, p. 49. AEC. (1987). Tables AK. Rhone Poulenc, Commentry, France, p. 86. Akiyama, D. M. (1991). Future consideration for the aquaculture feed industry. In: Proceedings of Aquaculture Feed Processing and Nutrition Workshop, (D. M. Akiyama and R. K. H. Tan Evans, M. (1985). Nutrient composition of feedstiffs for pigs andpoultry. Queensland Department of Primary Industries, Brisbane. pp. 135. Farrell, D. J. (1992). The use of non-traditional feed ingredients in poultry production and the potential for improving their nutritive value. In: Proceedings of Aquaculture Nutrition Worhhop (G. L. Allan and W. Dall, Eds.), Salamander Bay, 15-l 7 April 199 1. NSW Fisheries, Brackish Water Fish Culture Research Station, Salamander Bay, Australia. pp. 105-111. Gomes, E. F., Rema, P. and Kaushik, S. J. (1995). Replacement of fish meal by plant proteins in the diet of rainbow trout (Oncorhynchus mykiss): digestibility and growth performance. Aquaculture 130, 177-186. Hajen, W. E., Higgs, D. A., Beames, R. M. and Dosanjh, B. S. (1993). Digestibility of various feedstuffs by post-juvenile chinook salmon (Oncorhynchus tshawytscha) in sea water. Measurement of digestibility. Aquaculture 112,333-348. Editors), Thailand and Indonesia, September 19-25 1991. American Soybean Assoc., Singapore, pp. 5-9. Allan, G and Rowland, S. (1992). Development of an experimental diet for silver-perch (Bidyanus bidyanus). Austasia Aquaculture 6(3), 39-40. Allan GL. (1995). Growth ofjuvenile silver perch (Bidyanus bidyanus) on diets based on modified wheat gluten. Final report to Academy of Grain Technology and Fisheries Research and Developmelnt corportation. NSW Fisheries, Taylors Beach, NSW p. 23. Alternative feed ingredients for intensive aquaculture 1 09 Halver, J. E. (1989). The Vitamins. In Fish Nutrition (Ed J.E. Halver.) pp. 3 l-109. Academic Press Inc.: San Diego, California Hanley, F. (1987). The digestibility of foodstuffs and the effects of feeding selectivityon digestibility determinations in tilapia, Oreochroms niloticus. Aquaculture 66,163-l 79. Hardy, R. W. (1989). Diet preparation. In: Fish Nutrition (Ed. J.E. Halver) pp. 475-548. Academic Press Inc.: San Diego, California. Hardy, R. W. and Dong, F. M. (1995). Aquatic feed ingredients: quality standardsand methods of analysis. In: Proceedings of Feed Ingredients Asia 1995. Singapore International Conference and Exhibition Centre, Singapore 19-2 1 September 1995. Turret Group PLC., pp. 71-92. Hughes, S. G. (1987). Distillers products in salmonid diet. Proceeding of Distillers Feed Conference 42,27-3 1. Kohler, C. C. and Pagan-Font, F. A. (1978). Evaluations of rum distillation wastes, pharmaceutical wastes and chicken feed for rearing Tilapia aurea in Puerto Rico. AquacuZture 14, 339-347. Ravindran, V and Blair, R. (1992). Feed resources for poultry production in Asia and the Pacific. II. Plant protein sources. World 5 Poultry Science Journal 48, 205-23 1. Reddy, N. R., Sathe, S. K. and Salunkhe, D. K. (1982). Phytates in legumes and cereals. Advances in Food Research, 28, l-92. Reinitz, G. (1980). Acceptability of animal fat in diets for rainbow trout at two environmental temperatures. Progressive Fish Culturalist 42(4), 2 18-222. Riche, M. and Brown, P. B. (1996). Availability of phosphorus from feedstuffs fed to rainbow trout, Oncorhynchus mykiss. Aquaculture 142,269-282. Robinson, E. H. (1989). Channel catfish nutrition. Reviews in Aquaculture Sciences 1,365-39 1. Swick, R. A. (1995). Considerations in using common Asian protein meals. In: Proceedings of Feed Ingredients Asia 1995. Singapore International Conference and Exhibition Centre, 19-2 1 September 1995, Turret Group PLC., pp. 121-138. Tacon, A. G. J. (1993). Feed ingredients for crustaceans, natural foods and processed feedstuffs. FAO, United Nations: Rome, pp. 67. Tacon, G J., (1996). Global trends in aquaculture and aquafeed production. FAO/GLOBEFISH Research Programme Report, FAO: Rome (in press). Viana, M. T., Cervantes-Trujano, M. and SolanaSansores, R. (1994). Attraction and palatability activities in juvenile abalone (Haliotis fulgens): nine ingredients used in artificial diets. AquacuZture 127, 12-28. Liao, I. C. (1996). How can aquaculture help sustain world fisheries? Paper presented at 2nd World Fisheries Congress. Brisbane, JuZy 1996. Lovell, R. T. (1989). Nutrition and feeding offish. Van Nostrant Reinhold: New York, p. 260. Mackie, A. M. and Mitchell, A. I. (1985). Identification of gustatory feeding stimulants for fish - applications in aquaculture. In Nutrition and Feeding in Fish. pp. 177-189 (Eds. C. B. Cowey, A. M. Mackie and J. G. Bell). Academic Press: London. Murai, T. (1992). Protein nutrition of rainbow trout. Aquaculture 100, 19 l-207. New, M. B. (1987). Feed and Feeding of Fish and Shrimp. Webster, C. D., Tidwell, J. H., Goodgame, L. S., Yancey, D. H. and Mackay, L (1992). Use of soybeans and distillers grains and solubles as partial or total replacement of fishmeal in diets for channel catfish, Ictaluruspunctatus. Aquaculture 106,301-309. FAO: Rome, pp. 275. New, M. B. and Csavas, I. (1993). Aquafeeds in Asia - a regional overview. In: Farm-made Aquafeeds. Proceedings of Regional Expert Consultation on Farm-made Aquafeeds. 14-l 8 December 1992. Bangkok. Thailand. (Eds. M. B. New, A. G. J. Tacon and I Csavas) pp. 123. Wee, K. L. (1992). An overview of fish digestive physiology and the relevance to the formulation of artificial fish feeds. In: Proceedings of Aquaculture Nutrition Workshop, G. L. Allan and W. Dal1 (Eds.), Salamander Bay, 15-l 7 April 199 1. NSW Fisheries, Brackish Water Fish Culture Research Station, Salamander Bay, Australia, pp. 17-24. Williams, K. C. and Blaney, B. J. (1992). The potential of moulds to reduce the value of feedstuffs. In: Proceedings Aquaculture Nutrition Workshop, G. L. Allan and W. Dal1 (Eds), Salamander Bay, 15-17 April 199 1. NSW Fisheries, Brackish Water Fish Culture Research Station, Salamander Bay, Australia, pp. 205-213. Wilson, R. P. (1991). Channel catfish, Ictalurus punctatus. Handbook of Nutrient Requirements of Finfish. (Ed. R. P. Wilson) pp. 35-53. CRC Press: Boca Raton. New, M. B., Tacon, A. G J. and Csavas, I. (1993). Farmmade aquafeeds. Proceedings of FAOLAADCP Regional @ert consultation on farm-made aquafeeds, 14-I 8 December 1992, Bangkok Thailand. p. 434. Novus, (1992). Raw Material Compendium, 1st Edition. Novus International Inc.: St Louis, USA, pp. 5 11. Academy Press: Washington DC. NRC (1993). Nutrient requirements offish. National Petterson, D. S. and Mackintosh, J. B. (1994). The chemical composition and nutritive value of Australian grain legumes. Grains Research and Development Corporation: Canberra, pp. 68.