AQUAFEEDS IN ASIA - A REGIONAL OVERVIEW (Continued)

Few have hitherto attempted to predict the growth of non-carnivorous finfish culture, which constitutes the bulk of aquaculture production in Asia, as well as globally. Similarly, the demand for supplementary feed ingredients which are more commonly used in semi-intensive aquaculture has not been addressed. A case in point is evident from Andhra Pradesh (Table 5) where nearly 18,340 kg/ha/yr of de-oiled rice bran is utilized for supplemental feed. Swaminathan (1992) drew attention to the global problem of environmental and food security. He focused on the urgent need to increase production of certain produce such as, for example, to double the production of rice by year 2000. For the aquaculture industry in rural Asia to conform to the envisaged growth the supply of feed ingredients which can be incorporated into supplemental feeds need to be readily available.

The availability of some of the common ingredients used in supplemental feeds in Asia over the last 6 years has been relatively static (Table 12). If pulses (rarely used by farmers) are not taken into account the total availability (for all uses) of the major ingredients in 1990 would amount to about 80 million t. Assuming that 10% of this is available to the aquaculture industry, the total amount of ingredients available will be 8 million t, an increase of only 12% over the six-year period 1985-1990, or averaging 2% per year. This obviously is much less than the rate at which the aquaculture industry has grown over the last decade, as well as its envisaged growth in the present decade. In such a scenario it is conceivable that, in the ensuing years, the cost of commonly used supplemental feeds and/or ingredients will increase, and accordingly the cost of supplemental feeding will also increase. Apart from direct cost increases, these ingredients are beginning to have competing uses and thereby also influence their ready availability to the farmer. In other words, a more serious “trap” in respect of supplemental feeds is in the offing. Perhaps it is opportune for us to pay more attention to availability and efficient usage of supplementary feed ingredients and not to literally put all the eggs in the “fishmeal basket”, before it becomes too late.

* rice-bran
** wheat bran
*** based on extraction rate for cotton seed

Sources: FAO (1991); Devendra (1977)

New (1990) advocated simple farm-made moist feeds, and/or sun-dried feeds derived from these, for farms capable of producing up to 300 t of shrimp per year. Lippert (1990), on the other hand, pointed out that increasing dependence on farm-made feeds will decrease the incentive for major feed companies to develop better feeds. In shrimp farming, most of which borders on the upper scale of semi-intensive culture, utilization of practical, compounded feed is an accepted practice. Moreover, the profit margins in shrimp culture permit dependence on commercial feeds, or investment in small-scale machinery to make on-farm feeds a practical proposition. This scenario might not necessarily hold true for semi-intensive culture of non-carnivorous finfish, the great bulk of aquaculture produce in Asia. Therefore, the future strategies for this sector may have to be different.

The growth of the shrimp industry has relatively higher physical constraints and therefore higher physical limitations (based on the availability of suitable sites) in comparison to semi-intensive culture of finfish. Whilst both industries face the common problem of availability of fishmeal, admittedly to varying degrees, the sheer volume of production and the higher degree of dependence of the semi-intensive finfish culture industry on agricultural by-products tends to make this industry in some ways more vulnerable in the future.

It is essential that future research efforts on supplemental feed development take a different direction. It is the author's belief that there is an urgent need to move away from the traditional dose-effect studies on nutritional requirements of cultured species in Asia and on the emphasis of development of fishmeal substitutions by agricultural by-products for compounded diets, as has been pointed out earlier by De Silva and Davy (1992). There is an urgent need to work in conjunction with the practioners and for research on supplemental feeds to be based on improvements in ongoing farm practices. The “top-down” approach does not appear to have delivered the goods. Perhaps a more pragmatic “bottom-up” approach is urgently needed.

LIST OF REFERENCES

Albrecht, M.L. and B. Breitsprecher. 1969. Untersuchungen ueber die chemische Zusammensetzung von Fischenahrtieren und Fischfutter-mitteln. Z-fish N.F. 17: 143-163.

Bailey, C. and M.Skladany. 1991. Aquaculture development in tropical Asia. A re-evaluation. p. 66-73. Natural Resources Forum, February 1991.

Boonyaratapalin. M. and D.M. Akiyama. 1989. The aquaculture industry in Southeast Asia. p. 41-56. In Proceedings of the Third International Symposium on Feeding and Nutrition in Fish, Toba, Japan.

Bowen. S.H. 1987. Dietary protein requirements in fishes - a reassessment. Canadian J. Fish Aquat. Sci. 44: 1995-2000.

Castell, J.D., D.E. Conklin, J.S. Carigie, S.P. Lall and K. Norman-Boudreau. 1988. Aquaculture nutrition, p. 291-308. In M. Bilio, H. Rosenthal and C.J. Sindermann (eds.) Realisms in Aquaculture: achievements constraints, perspectives. European Aquaculture Society, Belgium.

Castille, F.L. and A.L. Lawrence. 1989. The effects of deleting dietary constituents from pelleted feed on the growth of shrimps in the presence of natural food in ponds. J. World Aquaculture Society 20(1):22A (Abstract).

Chiu, Y.N. 1989. Consideration for feeding experiments to quantify dietary requirements of essential nutrients in fish, p. 46-57. In S.S. De Silva (ed.) Fish Nutrition Research in Asia. Asian Fisheries Society, Special Publication 4, Manila, Philippines.

Csavas, I. 1989. Development of semi-intensive and intensive aquaculture in Asia and the Pacific. Paper presented at 29th Annual Conference of the International Association of Fish Meal Manufacturers, November 1989, Hong Kong.

Csavas, I. 1990. Aquaculture development and environmental issues in the developing countries of Asia. Paper presented at the Conference on Environmental Issues in Third World Aquaculture Development, September 1990, Bellagio, Italy.

De Silva, S.S. 1985. Performance of Oreochromis niloticus (L.) fry maintained on mixed feeding schedules of different protein content. Aquaculture and Fisheries Management 16:621-633.

De Silva, S.S. and F.B. Davy. 1992. Strategies for finfish nutrition research for semi-intensive aquaculture in Asia. Asian Fisheries Science 5:129-144.

De Silva, S.S. and R.M.G. Gunasekera. 1991. An evaluation of the growth of Indian and Chinese major carps in relation to the dietary protein content. Aquaculture 92:237-241.

De Silva, S.S. and D.E.M. Weerakoon. 1981. Growth food intake and evacuation rates of grass carp, Ctenopharyngodon idella fry. Aquaculture 25:67-76.

De Silva, S.S., M.K. Perera and P. Maitipe. 1984. The composition, nutritional status and digestibility of the diets of Sarotherodon mossambicus from nine man-made lakes in Sri Lanka. Environmental Biology of Fishes 11:205-19.

De Silva, S.S., R.M.G. Gunasekera and D. Atapattu. 1989. The dietary protein requirements of young tilapia and an evaluation of the least cost dietary protein levels. Aquaculture 80:271-284.

Devendra, C. 1977. The utilization of agro-industrial by-products in Asia and the Far East, p. 111-124. In New Feed Resources. FAO Animal Production and Health Paper No.4. FAO, Rome, Italy.

Devendra, C. 1985. Non-conventional feed resouces in Asia and the Pacific. 2nd ed. FAO/ APHCA Publication No. 6. FAO, Bangkok. 140 p.

Dickson, M.W. 1987. The supply of vitamins in feeds for intensive tilapia farming in Zambia. Aquaculture and Fisheries Management 18:221-230.

FAO. 1991. FAO Yearbook - Production. Vol.44, (1990). FAO Statistics Service No. 99. FAO, Rome, Italy. 283 p.

FAO. 1992. Aquaculture production (1984-1990). FAO Fisheries Circular No. 815, Revision 4. FAO, Rome, Italy. 206 p.

Gaillard, J. 1991. Scientists in the Third World. The University Press of Kentucky, USA. 191 p.

Green, B.W. 1992. Substitution of organic manure for pelleted feed in tilapia production. Aquaculture 101:213-222.

Hanley, F. 1991. Effects of feeding supplementary diets containing varying levels of lipid on growth, food conversion, and body composition of Nile tilapia, Oreochromis niloticus (L). Aquaculture 93:323-334.

Hepher, B. 1975. Supplementary feeding in fish culture. Proceedings International Congress on Nutrition, 9(3):183-198.

Hepher, B. 1978. Ecological aspects of warm water fish pond management, p.447-465. In S.D. Gerking (ed.) Ecology of Freshwater Fish Production. Black well Scientific Publisher, Oxford.

Hepher, B. 1988. Principles of fish nutrition, p.121-142. In M. Shilo and S. Sarig (eds.) Fish Culture in Warm Water Systems: Problems & Trends. CRC Press, NY, USA.

Hepher, B., J. Chervinski and H. Tugari. 1971. Studies on carp nutrition III. Experiments on the effect on fish yields of dietary protein source and concentration. Bamidgeh 23:11-37.

Jana, B.B. and R. Chakraborty. 1990. Exogenous introduction of live plankton, a better approach to carp growth than direct-manure systems. Prog. Fish-Cult. 52:252-260.

Kiaerskou, J. 1991. Production and economics of “low pollution diets” for the aquaculture industry, p. 65-76. In C.B. Cowey and C.Y Cho (eds.) Nutritional Strategies and Aquaculture Wastes. University of Guelph, Canada.

Ling, S.W. 1967. Feeds and feeding of warm-water fishes in ponds, p.291-309. In T.V.R. Pillay (ed.) Proceedings of the World Symposium on Warm Water Pond Fish Culture. FAO Fisheries Report No. 44(3).

Lippert, I.E. 1990. Shrimp farming and its markets. Fish Farming International 17(2):26-29. Lovell, T. 1989. Energy requirements for growth : fish versus farm animals. Aquaculture Magazine 15(4):65-66.

Murai, T. 1992. Protein nutrition of rainbow trout. Aquaculture 100:191-207.

Nandeesha, M.C., K.D. Murthy, S.S. De Silva and K. Dathatri. (in preparation). Evaluation of mixed feeding schedules in carp culture - field trials on catla, rohu and common carp.

New, M.B. 1987. Feed and feeding of fish and shrimp. FAO Field Document ADCP/REP/87/ 26. FAO, Rome, Italy. 275p.

New, M.B. 1990. Compound feedstuffs for shrimp culture, p.79-123. In M.B. New, H. de Saram and T. Singh (eds.) Technical and Economic Aspects of Shrimp Farming. INFOFISH, Kuala Lumpur, Malaysia.

New, M.B. 1991. Compound feeds: a world review. Fish Farmer 14(2):39-42.

Odum, W. 1970. Utilization of the direct grazing and plant detritus food chain by Mugil cephalus, p. 222-240. In H. Steele (ed.) Marine Food Chains. Oliver and Boyd, London.

Pantastico, J.B. 1988. Non-conventional feed resources in aquaculture: An overview of work done in the Philippines, p.71-87. In S.S. De Silva (ed.) Finfish Nutrition Research in Asia. Heinemann Publishers Asia Pte. Ltd, Singapore.

Prus, T. 1970. Calorific value of animals as an element of bioenergetical investigations. Polish Archives Hydrobiologie 17:183-199.

Santiago. C.B. and O.S. Reyes. 1991. Optimum dietary protein level for growth of bighead carp (Aristichthys nobilis) fry in a static water system. Aquaculture 93: 155-165.

Schroeder, G.L. 1978. Autotrophic and heterotrophic production of micro - organisms in intensely-manured fish ponds, and related fish yields. Aquaculture 14: 303-325.

Sinha. V.R.P. 1979. Contribution of supplementary feed in increasing fish production through composite fish culture in India, p. 565-574. In J.E. Halver and K. Tiews (eds.) Finfish Nutrition and Fish Feed Technology. Vol. 1, Heenemann, Berlin.

Spataru, P., B. Hepher and A. Halevy. 1980. The effect of the method of supplementary feed application on the feeding habits of carp (Cyprinus carpio L.) with regard to natural food. Hydrobiologia 72: 171-178.

Sumagaysay, N.S., F.E. Marquez and Y.N. Chiu-Chern. 1991. Evaluation of different supplemental feeds for milkfish (Chanos chanos) reared in brackish water ponds. Aquaculture 93:177-189.

Swaminathan, M.S. 1992. Environment and global food security. Food Tech. 46(7): 89-95.

Swingle, H.S. 1958. Experiments on growing fingerling channel catfish to marketable size in ponds. Proc. Southeast Association of Game Fish Commissioners 12:63-72.

Tacon, A.G.J. 1988. The nutrition and feeding of farmed fish and shrimp - a training manual. 3. Feeding methods. FAO Field Document, Project GCP/RLA/075/ITA. Field Document 7/E, FAO, Brasilia, Brazil. 208 p.

Tacon, A.G.J. and A.J. Jackson. 1985. Utilization of conventional and non-conventional protein sources in practical feeds, p. 119-145. In C.B. Cowey, A.M. Mackie and J.G. Bell (eds.) Nutrition and Feeding in Fish. Academic Press, London.

Trino, A.T., V.D. Penaflorida and E. Bolivar. 1992. Growth and survival of Penaeus monodon juveniles fed a diet lacking vitamin supplements in a modified extensive culture system. Aquaculture 101:25-32.

Tuang, P-H. and S.-Y. Shiau. 1991. Effect of meal frequency on growth performance of hybrid tilapia, Oreochromis niloticus x O. aureus, fed different carbohydrate diets. Aquaculture 92: 343-350.

Veerina, S.S., M.C. Nandeesha and S.S. De Silva. (in preparation). Analysis of input - output interrelationship in the Indian major carp farming system in Andhra Pradesh, India.

Veerina, S.S., M.C. Nandeesha, K. Rao Gopal and S.S. De Silva. (in preparation). Status and technology of Indian major carp farming in Andhra Pradesh, India. Asian Fisheries Society, Indian Branch, Special Publication 9, Mangalore.

Weatherly, A.H. 1963. Notions of niche and competition among animals with special reference to freshwater fish. Nature (Lond.) 197:14-17.

Wee, K.L. 1991. Use of non-conventional feedstuffs of plant origin as fish feeds: is it practical and economically feasible? p.13-22. In S.S. De Silva (ed.) Fish Nutrition in Asia. Asian Fisheries Society Special Publication, 5. Asian Fisheries Society, Manila, Philippines.

Werner, E.E. and D.J. Hall. 1976. Niche shifts in sunfishes, experimental evidence and significance. Science 191:404-406.

Wijkstrom, U.N. and M.B. New. 1989. Fish for feed: a help or a hindrance to aquaculture in 2000? Infofish International 6/89: 48-52.

Wilson, R.P. (ed.) 1991. Handbook of nutrient requirements of finfish. CRC Press, Boston, USA. 94 p.

Wohlfarth, G.W. and G. Hulata. 1987. Use of manures in aquaculture, p. 353-367. In D.J.W. Moriarty and R.S.V. Pullin (eds.) Detritus and Microbial Ecology in Aquaculture. ICLARM Conference Proc. 14, Manila, Philippines.

Wohlfarth, G.W. and G.L. Schroeder. 1979. Use of manure in fish farming - a review. Agricultural Wastes 1:279-299.

Yakupitiyage, A., P. Edwards and K.L. Wee. 1991. Supplementary feeding of fish in duck-fish integrated system I. The effect of rice bran, p. 143-157. In S.S. De Silva (ed.) Fish Nutrition Research in Asia. Asian Fisheries Society Special Publication, Manila, Philippines.

Yang, H.Z. et al. 1985. Chinese integrated fish farming system and their effects. Freshwater Fisheries Research Center, Wuxi, China. [cited in Tacon (1988)]

Zhu, Y, Y. Yang, J. Wan, D. Hua and J.A. Mathias. 1990. The effect of manure application rate and frequency upon fish yield in integrated fish farm ponds. Aquaculture 91:233-251.

FEED FORMULATION AND ON-FARM FEED MANAGEMENT
Albert G.J. Tacon

Inland Water Resources and Aquaculture Service FAO, Via delle Terme di Caracalla, 00100 Rome, Italy

TACON, A.G.J. 1993. Feed formulation and on-farm feed management, p. 61-74. In M.B. New, A.G.J. Tacon and I. Csavas (eds.) Farm-made aquafeeds. Proceedings of the FAO/ AADCP Regional Expert Consultation on Farm-Made Aquafeeds, 14-18 December 1992, Bangkok, Thailand. FAO-RAPA/AADCP, Bangkok, Thailand, 434 p.

INTRODUCTION

The nutritional performance and success of an artificial diet for farmed fish or shrimp is dependent upon a number of important factors which can be broadly considered under two major headings: feed formulation and manufacture; and on-farm feed management and aquatic environment (Figure 1). Sadly, all too often it is believed that the role of the nutritionist and aquafeed manufacturer ends once the diet has been formulated and physically produced. It does not; the ultimate success of an artificial diet usually being dependent upon the management of the finished feed on the farm. This paper attempts to provide some general guidelines for the formulation and on-farm management of artificial diets for fish and shrimp, with particular reference to home-made aquafeeds.

FEED FORMULATION

The formulation of a practical diet for fish or shrimp is based upon a series of technical and economical considerations. These have been discussed previously (New 1987; Tacon 1988; Lall 1991) and can be summarized as follows:

• the market value of the species to be fed - the farm gate value of the cultured species will set the upper limit which can be spent on the formulation and finished diet (i.e. as a general rule total dry diet cost should not exceed 25% of the farm gate value of the cultured species);

• the financial resources of the farmer (if home-made aquafeeds are to be produced) - capital available for the purchase of feed ingredients and feed manufacturing equipment (i.e. the formulation will depend upon the money in the farmer's pocket and the availability of services such as electricity or fuel);

  

  Figure 1. Major factors determining the nutritional performance and success of an artificial diet feeding regime

• dietary nutrient requirements of the species to be fed (or closest relative) - including dietary protein, amino acid, fatty acid, mineral, vitamin and energy requirements for each phase of the culture cycle (i.e. for larvae/fry, juveniles, production and broodstock, if known);

• natural feeding habits of the species in question - in the absence of detailed information on the dietary nutrient requirements of the species an analysis of its natural feeding habits in the wild will indicate its position in the aquatic food chain (i.e. herbivore, detritivore, omnivore or carnivore), preferred food items and feed size (i.e. plant and animal feedstuffs), feeding station (i.e. surface, mid-water or benthic feeder), and feeding behaviour (i.e. daylight or nocturnal feeder, visual or olfactory feeder, rapid or slow feeder, continuous or intermittent feeder). Apart from determining dietary food preferences, an analysis of the natural feeding behaviour of the target species will also determine the physical characteristics of the artificial diet to be produced (i.e. feed size and shape, texture, palatability, buoyancy and water stability);

• available feed ingredient sources, composition and cost - including seasonal availability, current usage, proximate composition, quality control, digestibility and nutrient availability, and additional ingredient handling and processing cost prior to mixing or pelleting;

• intended feed manufacturing process to be used - grinding, mixing, cold pelleting, steam pelleting, expansion pelleting, flaking, and/or microencapsulation (i.e. the formulation will depend upon the manufacturing process employed and whether the diet is to be produced as a mash, crumble, paste, ball, moist pellet or dry pellet); and

• intended stocking density and farm production unit - extensive, semi-intensive or intensive fish/shrimp production systems within tanks, raceways, concrete lined ponds, earthen ponds, pens or floating cages (i.e. the intended stocking density and production unit will dictate the relative availability of natural food organisms per unit biomass of cultured fish/shrimp over the production cycle and the consequent necessity for a complete or supplementary diet formulation).

Despite the obvious importance of all the above mentioned factors, sadly little or no consideration is usually given by nutritionists and feed manufacturers concerning the intended fish/shrimp stocking density and consequent availability of natural food organisms within the proposed production unit. For example, at present almost all commercial feeds produced for semi-intensive pond farming systems are formulated as nutritionally complete diets irrespective of the intended stocking density employed and natural food availability (Akiyama 1991; Tacon 1991a; Chamberlain 1992; De Silva and Davy 1992). To a large extent this has been due to the almost total absence of published information concerning the dietary nutrient requirements of fish and shrimp within semi-intensive pond farming systems and the apparent difficulty of predicting the nutritional contribution of natural food organisms within the overall nutritional budget of the pond raised species (for review see Tacon 1987, 1988; Hepher 1988).

It follows from the above that the formulation of a supplementary diet is dependent upon the density and total biomass or standing crop of the cultured fish or shrimp species present, and the fertility and consequent availability of natural food organisms within the water body. As a general rule the contribution of natural pond biota toward the overall nutritional budget of the cultured fish or shrimp will be highest at low stocking densities or at the start of the culture cycle when the total biomass or standing crop is lowest. The consequent availability and relative contribution of natural pond biota thereafter decreases over the course of the production cycle with increasing fish or shrimp size and standing crop. For example, Figure 2 shows the influence of shrimp (Penaeus monodon) market size and pond stocking density on the reported performance (as indicated by the recorded apparent food conversion ratio - AFCR) of a pelleted shrimp feed within commercial shrimp farms in Thailand (ponds stocked with ca. 30-50 postlarvae/m²; Charoen Pokphand Aquaculture Co., Ltd., personal communication). From the data it is clear that the lower AFCRs observed at the lowest shrimp market sizes and stocking densities were almost certainly due to the higher contribution of natural pond biota toward the overall nutritional budget of the farmed shrimp at these lower densities and market sizes.

Figure 2. Influence of market size and stocking density on food conversion ratio of pond cultured black tiger shrimp fed CP feed in Thailand

The nutritional and economic importance of natural food organisms within semi-intensive pond farming systems cannot be over emphasized. For example, the tracer studies of Lilyestrom et al. (1987) and Anderson et al. (1987) with pond reared prawns (Macrobrachium rosenbergii; 0.5-1.5 animals/m²) and shrimp (Penaeus vannamei; 20 animals/m²) showed that natural pond organisms accounted for 18-75% and 53-77% of the growth of the cultured species, respectively; in both studies animals were fed complete pelleted feeds. In view of these results it is perhaps not surprising therefore that no difference was observed in the growth of prawns (Macrobrachium rosenbergii) in outdoor concrete ponds in Thailand when fed a 35% protein diet, a 15% protein diet or a broiler starter diet (stocking density 5 animals/m², pond water exchanged every three weeks; Boonyaratpalin and New 1982), or in the growth and survival of shrimp (Penaeus vannamei) in earthen ponds in Hawaii receiving no fertilizer and feed input, fertilization only, or fed a commercial pelleted shrimp diet (stocking density 7.1-9.4 animals/m²; Lee and Shleser 1984). Similarly, Bostock (1991) reported that there was no difference in the growth of pond reared shrimp (Penaeus monodon; 10/m²) in India fed a high-nutrient pelleted diet or a locally produced dough-ball costing one third of the price. In addition, numerous studies have been conducted concerning the necessity for dietary vitamin supplements within artificial feeds. In general the results have shown dietary vitamins to be dispensable under practical farming conditions for tilapias and shrimp/prawns when reared under semi-intensive pond farming conditions (Boonyaratpalin and New 1982; Castille and Lawrence 1989; Viola 1989; Tacon 1991b; Trino et al. 1992).

On the basis of the studies of Hepher and co-workers in Israel it has been found that dietary energy is generally the first limiting nutrient within earthen ponds at low carp (Cyprinus carpio) stocking densities and that, as the density and standing crop increases, so the requirement for supplemental dietary nutrients such as proteins, minerals and vitamins becomes greater (Figure 3). It follows therefore that the dietary protein and essential micronutrient content of supplementary feeds intended for use within semi-intensive pond farming systems should be low at the start of the culture operation and then progressively increased with increasing fish size and standing crop (Tacon 1988). This is the complete reverse of an intensive clear-water farming system where dietary nutrient levels usually decrease with increasing fish or shrimp size and age. Sadly, the majority of researchers, feed manufacturers and farmers alike still employ gradually decreasing dietary nutrient levels with increasing fish/shrimp age groups for semi-intensive fish and shrimp production systems. Clearly, this situation will have to be rectified if feed and production costs are to be reduced and maximum benefit is to be gained from semi-intensive farming systems. However, it is encouraging to note that the tide may soon be turning. For example, a commercial shrimp feed manufacturer in the Philippines has recently introduced two new feed lines on to the market; a conventional shrimp grower/ finisher diet containing 37% crude protein and a new diet designed for extensive to semi-intensive farming systems (for ponds larger than one hectare with a shrimp stocking density of not more than 10/m²) containing 30% crude protein.

Figure 3. Schematic model of the contribution of nutrients from natural and supplemental food for increasingly intensive carp culture in Israel (Viola 1989)

ON-FARM FEED MANAGEMENT

As mentioned before, the success of an artificial diet is dependent not only upon the formulation and manufacturing process used to produce the diet but also upon the management of the diet on the farm.

Feed handling and storage

Since aquafeeds are composed of highly perishable nutrients it is essential that adequate handling and storage procedures are employed to protect the finished feed from the natural elements (i.e. light, heat, humidity, air and water) from the time the feed emerges from the die plate to the time the feed is consumed by the fish or shrimp. The economic and nutritional consequences of poor feed handling and storage can be both profound and immediate, with a reduction in the nutritional performance of the feed (as indicated by increased FCR and reduced fish or shrimp growth), either through mechanical damage (i.e. increased “fines” content), oxidative damage (i.e. lipid peroxidation and vitamin destruction), or microbial/pest infestation (i.e. nutrient destruction and mycotoxin production). For a review of feed handling and storage techniques and information concerning the nutritional consequences of poor feed storage, including the pathological effects of oxidized lipids and mycotoxin contamination, see Chow (1980), New (1987), Akiyama (1989), De la Cruz et al. (1989), Tsai (1989a), Coelho (1991), Lovell (1992) and Tacon (1988, 1992).

Although the negative effects of feed handling and storage may be minimised by the use of dietary anti-fungal compounds and higher dietary antioxidant/vitamin fortification levels it is important to note that many dietary vitamins can be lost from finished feeds in a matter of minutes, just by exposing the surface of feed bags to the warming and destructive ultra violet rays of the sun. The latter is a common phenomenon during feed transportation and when bags of feed are left by the side of a tank or pond for feeding later the same day. Clearly, farmers must be made aware of the serious consequences of poor feed handling/storage and provided with adequate guidelines concerning proper on-farm feed management procedures. For example, Annex 1 shows the recommended feed handling and storage guidelines given to farmers by a commercial shrimp feed manufacturer in the Philippines. In addition to the guidelines listed in Annex 1, it should also be stated that feeds should not be stored in direct sunlight.

Feed application method and feeding regime

Maximum benefit from an artificial diet will only be achieved if the diet is ingested in its entirety and supplied to the fish or shrimp at a sufficient rate to satisfy its requirements for optimal growth and feed efficiency. Sadly, dietary feeding regimes are more often than not related to the convenience of the feeding technician or farm worker during his or her working day rather than to the behavioral feeding requirements of the cultured fish or shrimp species. For example, despite the nocturnal feeding habit of most pond reared shrimp and the rapid loss of water soluble nutrients from rations on prolonged immersion in water (Cuzon et al. 1982; Brown et al. 1989; Lim and Cuzon 1992; Table 1), the majority of commercial shrimp farmers in Latin and North America still only feed their animals once or twice daily during the morning or late afternoon (Hirono 1989; Jaenike 1989; Villalon 1991). Clearly, farmers must be made aware of the nutritional and economic losses they will be incurring through dietary leaching for every additional moment their diet remains uneaten in water; a feed pellet behaving in much the same way as a dissolving sugar cube in a hot cup of tea! It follows from the above that pond reared shrimp should be fed on a “little and often” basis, with the bulk of the feed fed during the night (i.e. from sun-set to sun-rise) when their feeding activity is highest (Tacon 1988; Akiyama 1989; Villalon 1991). The benefits of increasing the frequency of feed application are immediate and include reduced leaching and feed loss, and improved growth and feed efficiency (Villalon 1991; Sumudra 1992).

Table 1. Reported nutrient loss from a pelleted shrimp diet after a one hour seawater immersion period* (Cuzon et al. 1982)

Nutrient	Level at start	Level after 1 hour	Percent loss (%)
Dry matter (%)	100	81	19
Crude protein (%)	52	41	21
Carbohydrate (%)	16	8	50
Vitamin C (mg/kg)	3,089	332	89
Thiamine (mg/kg)	29.5	0.7	98
Riboflavin (mg/kg)	55	7.5	86
Pyridoxine (mg/kg)	14	1	93
Pantothenate (mg/kg)	100	5.9	94
Niacin (mg/kg)	120	17	86
Inositol (mg/kg)	4,000	1,928	52
Choline (mg/kg)	3,368	1,835	45

* diet contained 15% wheat gluten as a binder to maintain pellet shape in waterfor at least 48 hours

It is also important to mention here that each diet has an ideal dietary feeding rate range where both growth and feed efficiency are optimal for each fish/shrimp age class; the optimal feeding level in turn varying with body weight, dietary nutrient level, water quality, stocking density and natural food availability (for review see Hepher 1988, and Tacon 1988). Sadly, the majority of feeding tables recommended by feed manufacturers for use within semi-intensive farming systems are generally hypothetical and arbitrarily set, irrespective of dietary composition, pond fertilization rate, natural food availability, stocking density and standing crop. Furthermore, since natural food organisms play a gradually diminishing role in the overall nutritional budget of pond reared shrimp and fish as the body size and standing crop increases with time, it follows that the feeding rate of the artificial diet should be gradually increased over the course of the culture cycle. However, at present almost all feed manufacturers are recommending semi-intensive farmers to use a decreasing dietary feeding rate with increasing fish/shrimp size and body weight (Akiyama 1989; Villalon 1991; Sumudra 1992). Clearly, optimal feeding rates and frequency of feed presentation must be determined for individual feeds and farms by carefully monitoring feed consumption, growth and feed efficiency over several growing seasons (for good pond feed monitoring methods see Akiyama 1989). For example, Figure 4 shows the observed variation in reported FCR of a commercial pelleted shrimp diet within 174 shrimp farms in Thailand (Charoen Pokphand Aquaculture Co., Ltd., personal communication). The fact that the reported FCR ranged from 1 to 2.6 clearly shows the wide variation in farm management practices employed by the shrimp farmers using the same diet (see also Figure 2). From an economic viewpoint, feed costs alone ranged from as low as US$ 1.20/kg to as high as US$ 2.60/kg of shrimp harvested, depending upon the management procedure employed. The results clearly show that there is considerable room for improvement concerning on-farm feed management programmes; the difference between good and poor feed management being the difference between economic success and failure!

Figure 4. Variation in food conversion ratio in 174 black tiger shrimp farms using CP shrimp feed in Thailand

Aquatic environment and natural food availability

Finally, but not least, the performance of a diet is dependent upon the aquatic environment in which the fish or shrimp is cultured, including:

intended fish or shrimp stocking density and natural food availability (see previous sections and Figures 2-3);

intended farm production unit - tank, raceway, concrete lined pond, earthen pond, pen or floating cage. For example, aquafeed distribution and management within large 5-10 ha ponds is much more difficult (from a practical point of view) and consequently far less effective (from a nutritional point of view in terms of fish/shrimp growth and feed performance or FCR) than within smaller 0.5-1 ha ponds; and

water quality and water/pond management - water temperature, dissolved oxygen concentration, salinity, turbidity, mineral concentration, pH, water depth, water exchange rate, pond preparation and fertilization, water circulation pattern and artificial aeration. For example, Featherstone (1981) found that the performance of salmon starter diets was related to their physical characteristics (i.e. feed size and density) and consequent spatial behaviour with respect to the water circulation pattern within the culture tanks used. Similarly, high pond water exchange rates have a negative effect on the action of fertilizers and on natural pond productivity. Furthermore, under certain circumstances the action of fertilizers and performance of aquafeeds can be improved by using artificial pond aeration devices. For a review of the importance of water quality, pond preparation and pond aeration to the success of a semi-intensive pond feeding strategy see Tacon (1988), Boyd (1989), Chiang et al. (1989) and Tsai (1989b).

CONCLUSIONS

On the basis of the above discussion it is clear that the nutritional and economic success of an aquafeed is based upon a variety of inter-linked factors and that no one factor can be considered in isolation (Figure 1). It is also apparent that the development of dietary feeding strategies for semi-intensive pond farming systems has proceeded without due regard to fish or shrimp stocking density and natural food availability. This is particularly surprising since the majority of world fish and shrimp aquaculture production is realised within semi-intensive pond farming systems. In fact, one of the major stumbling blocks to aquaculture development within many developing countries has been due to the direct transfer and application of intensive complete diet feeding strategies to semi-intensive pond farming systems. Furthermore, increasing raw ingredient and feed manufacturing costs, and a generally static and/or decreasing market value for the majority of cultured aquaculture species, necessitates that farmers reduce production costs to maintain profitability. Clearly, the road is open for improvements to be made in the area of feed formulation and feed management practices. These should be designed to reduce production costs and maximize the role played by natural food organisms in the overall nutritional budget of the cultured species. The special and unique needs of the semi-intensive fish or shrimp farmer should also be addressed. It is hoped that this paper will help to sow a seed along this road.

LIST OF REFERENCES

Akiyama, D.M. 1989. Shrimp feed requirements and feed management, p. 75-82. In D.M. Akiyama (ed.) Proceedings of the Southeast Asian Shrimp Farm Management Workshop, Philippines, Indonesia, Thailand, 26 July-11 August 1989. American Soybean Association, Singapore.

Akiyama, D.M. 1991. Future considerations for the aquaculture feed industry, p. 5-9. In D.M. Akiyama and R.K.H. Tan (eds.) Proceedings of the Aquaculture Processing and Nutrition Workshop, Thailand and Indonesia, September 19-25, 1991. American Soybean Association, Singapore.

Anderson, R.K., P.L. Parker and A.L. Lawrence. 1987. A 13C/12C tracer study of the utilization of presented feed by a commercially important shrimp Penaeus vannamei in a pond growout system. J.World Aquaculture Soc. 18(3): 148-155.

Boonyaratpalin, M. and M.B. New. 1982. Evaluation of diets for Macrobrachium rosenbergii reared in concrete ponds, p. 249-256. In M.B. New (ed.) Giant Prawn Farming. Elsevier Science Publ. Corp., Amsterdam, Holland.

Bostock, T. 1991. Better feeds for small-scale shrimp farmers. Bay of Bengal News 42:22-26. Boyd, C.E. 1989. Aeration of shrimp ponds, p. 134-140. In D.M. Akiyama (ed.) Proceedings of the Southeast Asian Shrimp Farm Management Workshop, Philippines, Indonesia, Thailand, 26 July - 11 August 1989. American Soybean Association, Singapore.

Brown, P.B., E.H. Robinson and G. Finne. 1989. Nutrient concentrations of catfish feces and practical diets after immersion in water. J. World Mariculture Soc. 20(4):245-249.

Castille, F.L. and A.L. Lawrence. 1989. The effects of deleting dietary constituents from pelleted feeds on the growth of shrimp in the presence of natural foods in ponds. J.World Aquaculture Soc. 20(1): 22A (Abstract).

Chamberlain, G. 1992. Shrimp culture in Indonesia - 4. Feeds for the shrimp industry. World Aquaculture 23(2): 38-40.

Chiang, P.D-R., C-M. Kuo and C-F. Liu. 1989. Pond preparation for shrimp growout, p. 48-55. In D.M. Akiyama (ed.) Proceedings of the Southeast Asian Shrimp Farm Management Workshop, Philippines, Indonesia, Thailand, 26 July - 11 August 1989. American Soybean Association, Singapore.

Chow, K.W. 1980. Storage problems of feedstuffs, p. 215-224. In Fish Feed Technology. FAO Field Document, FAO/UNDP Report ADCP/REP/80/11, Rome, Italy.

Coelho, M.B. 1991. Effects of processing and storage on vitamin stability. Feed International 12(12):39-45.

Cuzon, G.,M. Hew, D. Cognie and P. Soletchnik. 1982. Time lag effect of feeding on growth of juvenile shrimp Penaeus japonicus Bate. Aquaculture 29:33-44.

De la Cruz, M.C., G. Erazo and M.N. Bautista. 1989. Effect of storage temperature on the quality of diets for the prawn, Penaeus monodon Fabricus. Aquaculture 80:87-95.

De Silva, S.S. and F.B. Davy. 1992. Fish nutrition research for semi-intensive culture systems in Asia. Asian Fisheries Science 5:129-144.

Featherstone, P.B. 1981. A comparison of various fish feeding diets for the Atlantic salmon (Salmo salar L.). M.Sc thesis. University of Stirling, Stirling, Scotland, 75 p.

Hepher, B. 1988. Nutrition of pond fishes. Cambridge University Press, England. 388 p.

Hirono, Y. 1989. Shrimp farm management in Ecuador, p. 2-10. In D.M. Akiyama (ed.) Proceedings of the Southeast Asian Shrimp Farm Management Workshop, Philippines, Indonesia, Thailand, 26 July - 11 August 1989. American Soybean Association, Singapore.

Jaenike, F. 1989. Management of a shrimp farm in Texas, p. 11-21. In D.M. Akiyama (ed.) Proceedings of the Southeast Asian Shrimp Farm Management Workshop, Philippines, Indonesia, Thailand, 26 July - 11 August 1989. American Soybean Association, Singapore.

Lall, S.P. 1991. Concepts in the formulation and preparation of a complete fish diet, p. 1-12. In S.S. De Silva (ed.) Fish Nutrition Research in Asia. Proceedings of the Fourth Asian Fish Nutrition Workshop. Asian Fisheries Society Special Publication No.5, Asian Fisheries Society, Manila, Philippines.

Lee, C-S. and R.A. Shleser. 1984. Production of Penaeus vannamei in cattle manure-enriched ecosystems in Hawaii. J. World Maricul. Soc. 15:52-60.

Lilyestrom, C.G., R.P. Romaire and P. Aharon. 1987. Diet and food assimilation by channel catfish and Malaysian prawns in polyculture as determined by stomach content analysis and stable carbon isotope ratios. J. World Aquaculture Soc. 18(4): 278-288.

Lim, C. and G. Cuzon. 1992. Water stability of shrimp pellets. Paper presented at the Third Asian Fisheries Forum: Fisheries Towards 2000, 26-30 October 1992. Asian Fisheries Society, Singapore.

Lovell, R.T. 1992. Mycotoxins: hazardous to farmed fish. Feed International 13(3):24-28.

New, M.B. 1987. Feed and feeding of fish and shrimp. FAO Field Document, FAO/UNDP Report ADCP/REP/87/26:275 p., Rome, Italy.

Samudra, H.D. 1992. Management of a shrimp farm in Indonesia. ASA Technical Bulletin, AQ35 1992/8: 18 p., American Soybean Association, Singapore.

Tacon, A.G.J. 1987. The nutrition and feeding of farmed fish and shrimp- A training manual. 1. The essential nutrients. FAO Field Document, Project GCP/RLA/075/ITA, Field Document No.2, 117 p. Brasilia, Brazil.

Tacon, A.G.J. 1988. The nutrition and feeding of farmed fish and shrimp- A training manual. 3. Feeding methods. FAO Field Document, Project GCP/RLA/075/ITA, Field Document No. 7,208 p. Brasilia, Brazil.

Tacon, A.G.J. 1991a. Aquaculture nutrition and feeding in developing countries: A practical approach to research and development. Paper presented at the IV International Symposium on Fish Nutrition and Feeding, 24-27 June 1991. Biarritz, France.

Tacon, A.G.J. 1991b. Vitamin nutrition in shrimp and fish, p. 10-41. In D.M. Akiyama and R.K.H. Tan (eds.) Proceedings of the Aquaculture Processing and Nutrition Workshop, Thailand and Indonesia, September 19-25, 1991. American Soybean Association, Singapore.

Tacon, A.G.J. 1992. Nutritional fish pathology. Morphological signs of nutrient deficiency and toxicity in farmed fish. FAO Fish. Tech. Paper No. 330: 75 p. Rome, Italy.

Trino, A.T., V.D. Penaflorida and E.C. Bolivar. 1992. Growth and survival of Penaeus monodon juveniles fed a diet lacking vitamin supplements in a modified extensive culture system. Aquaculture 100: 25-32.

Tsai, C.K. 1989a. Quality control in formulated feeds, p. 270-287. In D.M. Akiyama (ed.) Proceedings of the People's Republic of China Aquaculture and Feed Workshop, September 17-30, 1989. American Soybean Association, Singapore.

Tsai, C.K. 1989b. Water quality management, p. 56-63. In D.M. Akiyama (ed.) Proceedings of the People's Republic of China Aquaculture and Feed Workshop, September 17-30, 1989. American Soybean Association, Singapore.

Villalon, J.R. 1991. Practical manual for semi-intensive commercial production of marine shrimp. Texas A & M Sea Grant College Program, TAMU-SG-91-501, Galveston, Texas. 103 p.

Viola, S. 1989. Production of commercial feeds for warm water fish, p. 143-162. In M. Shilo and S. Sarig (editors) Fish Culture in Warm Water Systems: Problems and Trends. CRC Press Inc., Boca Raton, Florida.

Annex 1. Example of feed handling and storage guidelines for a commercial shrimp feed in the Philippines

When loading and unloading feeds, always see to it that the bags are carried and put down gently

During transit, make sure that the floor of the vehicle is clean and dry

When transporting feeds, cover the bags on all sides with plastic or any waterproof blanket

Store the feeds in a cool dry place

Keep the feeds away from rats and other animals

Do not store the feeds in the same room with pesticides and other poisonous substances

Keep the feeds at least one foot away from the wall of the storeroom

Do not stack the bags more than 10 bags high. In a large stack of several piles, keep each pile one foot away from the next one. Raise the bottommost layer of each pile four inches above the ground with a grilled and hollow pallet. This will ensure that every stack is ventilated on all sides

When a new batch of feeds arrives in the storeroom, separate it from the older batch and make sure it is used only after the older feeds are consumed (first in, first out)

To ensure that your shrimps receive the feeds in the best possible condition, do not stock the feeds for more than one month. To avoid having excess feeds, purchase smaller quantities of feeds at a time, but more frequently during the grow-out cycle.

Source: San Miguel Foods, Inc, Manila, Philippines

FEED INGREDIENTS AND QUALITY CONTROL
Prasert Sitasit

National Inland Fisheries Institute Kasetsart University Campus, Bangkok 10900, Thailand

SITASIT, P. 1993. Feed ingredients and quality control, p.75-86. In M.B. New, A.G.J. Tacon and I. Csavas (eds.) Farm-made aquafeeds. Proceedings of the FAO/AADCP Regional Expert Consultation on Farm-Made Aquafeeds, 14-18 December 1992, Bangkok, Thailand. FAO-RAPA/AADCP, Bangkok, Thailand, 434p.

INTRODUCTION

Feed ingredients and finished feeds have played an important role in poultry and livestock production in Thailand for many decades. Many raw materials and finished feeds have been imported and many feed manufacturers were established to produce feeds to satisfy local demand. To protect farmers from the use of poor quality raw materials and finished feeds the Thai government promulgated the first law on animal feed quality control in 1963. Due to advances in animal nutrition and rapid progress in feed manufacturing, this law was revised in 1982.

CURRENT THAI FEEDSTUFF LEGISLATION

According to the 1982 law (Annex 1), animal feeds produced or imported for commercial use have to be licensed by the Department of Livestock Development. Species covered by the law were chicken, ducks, pigs, cattle and buffalo; no aquafeeds were included. This was because, when the law was promulgated, aquaculture was a small enterprise, and most aquafeeds were farm-made or consisted of kitchen wastes. Since 1986, aquaculture production in Thailand has been booming and there has been an increased demand for aquafeeds. Domestic aquafeed production in 1991 was approximately 300,000 t and was expected to reach 410,000 t in 1992. The rapid increase in aquafeed production has caused shortages of feed ingredients and resulted in inconsistent feed quality. Poorly formulated feeds and/or improper manufacturing processes could yield feeds which are poorly digested by aquatic species and have poor water stability. Such low-quality feed will pollute the aquatic environment. Furthermore, feed additives such as growth promoters, hormones, probiotics, and antibiotics, if improperly used in feed formulae, could be a hazard to aquatic species and to public health.

In 1991, to prevent such problems, the Department of Fisheries, together with the Department of Livestock through the support of the Minister of Agriculture and Cooperatives, issued a law controlling the quality of aquafeeds. According to this law aquafeed producers and distributors have to follow the rules, procedures and conditions prescribed (Agriculture and Cooperatives Ministerial Regulation 1991).

According to the regulations, many kinds of feed ingredients, premixes, feed additives and commercial aquatic feeds have to be licensed. At present, commercial feeds and premixes for marine shrimp, freshwater prawns, Clarias spp. and herbivorous fish feeds have to be registered with the Department of Fisheries. Some details are given below.

Raw materials

Feed ingredients sold in the market must have guaranteed proximate analyses as shown in Table 1.

Aquatic commercial feeds

Finished feeds sold in the market are not required to have defined nutritional characteristics but the nutritional value and the quality of the feed stated on the container labels have to comply with those registered with the Department of Fisheries.

Probiotics

Many kinds of probiotics such as Lactobacillus and Pediococcus spp. are approved for use in aquafeeds.

Hormones

Iodinated casein or thyroprotein, which are used to increase growth rate and feed efficiency in pigs, can be used in fish feeds upon approval of the Department of Fisheries.

Antibiotics

Many antibiotics are currently registered in Thailand for subtherapeutic use as growth promoters in poultry and livestock. However, for aquafeeds, all antibiotics are prohibited in finished feeds.

Containers

Aquafeed containers for feed distribution have to have the following characteristics:

• all containers have to be new, dry and have the capability to protect feeds from moisture;

• metal surfaces inside containers have to be coated with antirust materials which are not dangerous to aquatic animal health;

• vehicles transporting aquafeeds have to be dry and uncontaminated.

Feed preservatives and feed additives

Feed preservatives and additives that are permitted in finished feeds are listed in Table 2.

* this list applies equally to aquafeeds and other animal feeds

Non nutrients

The non-nutrient compounds which can be added in finished feed at the appropriate level are shown in Annex 2.

Feed form

No restrictions are placed on the forms of aquafeed, as for poultry and livestock feeds. Aquafeeds can be produced in pellet or extruded form, or as meals, crumbles and flakes. The one exception is that shrimp feeds are specially required to be of the sinking type.

Quality control procedures

The Department of Fisheries is responsible for the quality control of aquafeeds. Routine inspections are carried out as follows:

• aquafeeds distributed in the market or on-farm are sampled and analyzed for protein, fat, fibre and moisture and compared to the levels shown on the label and registration documents held by the Department of Fisheries;

• routine inspections of the cleanliness of feed manufacturing premises and equipment and of the raw materials used in feed manufacturing are carried out;

• the examination of finished products emphasizes a check on the presence of antibiotics.

GENERAL NOTES ON QUALITY CONTROL

In feed manufacturing, the purpose of the quality control of raw materials is to ensure that minimum contract specifications are met. These specifications are usually determined by a committee consisting of the nutritionist, management personnel, and the quality control manager. To meet the specifications, sampling, physical inspection and laboratory analysis have to be conducted before the ingredients are accepted and placed in the store. To minimize losses in nutritive value and deteriorative changes in feed ingredients during storage, they must be kept cool in as dry a place as possible. Physical examination and laboratory analysis is also essential for finished aquafeeds, to ensure that specifications are met.

To assure good feed quality, feed ingredients and finished feeds need to be closely monitored: this is known as quality control (QC). QC involves the verification of the quality standards established for each feed ingredient through the process of feed manufacturing and continues as they finally go into storage as finished feed. At first glance, feed ingredients and finished feed may appear to be nutritious and of high quality; however, unless they are systematically analyzed - through physical, chemical, and/or biological means - there is no way that one can be sure of their true value to aquatic animals.

Using good quality food ingredients in feed manufacturing partially guarantees good quality in final products. The purpose of the QC of feed ingredients is to verify that they meet the specifications of both the feedmill and the supplier. Moreover, QC provides knowledge concerning the exact composition of raw materials and the levels of toxic substances present. Feed ingredients arriving in the mill need visual inspection for evidence of wetting or mould growth, presence of scrap metals, stones, or other non-biological contaminants, and presence of insects. Feed ingredients which pass the visual examination will be sampled for further laboratory analysis. Good sampling is very important in QC. An analysis of feed ingredients is only as good as the sample obtained therefrom. If the sample is not representative of the entire batch, the evaluation is useless, no matter how extensively the sample is analyzed. Sampling requires certain techniques.

Feed ingredients received in bulk are sampled by using a scoop. Two samples per ton are taken for bulk consignments smaller than 10 t. Larger consignments, up to 100 t, require one sample per ton or one sample for every two tons, depending on the size of the consignment. Sampling of bagged ingredients is done with a spear probe. Samples from at least 5% of the bags are required. Samples taken should be pooled, mixed, and ground to the size suitable for chemical analysis. Samples submitted to the quality control laboratory should be stored in tightly sealed containers.

Chemical analysis of feed ingredients is the next step in QC. Proximate analyses are usually conducted to reveal the major nutritional characteristics of the samples. These analyses include moisture, crude protein, crude lipid, crude fibre, and ash. Nitrogen-free extract (NFE) is determined by difference. Calcium and phosphorus determinations are necessary for feed ingredients with high ash content and for mineral supplements, such as fish meal, meat and bone meal, calcium phosphate and calcium carbonate sources.

Feed ingredients liable to have high levels of certain anti-nutrients need to be tested prior to use. Trypsin inhibitor and urease content in soybean products, gossypol in cottonseed, aflatoxin in corn and peanut meal, mimosine in Leuceana leaf meal, and rancidity in fats, are natural toxins that, at high enough levels, are growth inhibitory and sometimes fatal to the animal consuming them. Therefore, it is advisable to analyze these feed ingredients for their toxic content.

The quality of finished feeds reflects the quality of the ingredients used. However, feed manufacturing and storage can alter the finished feed quality to a certain extent. Therefore, the purpose of QC of the final products is to ensure whether the feed ingredients were added in the proportions required by the formulation and to examine the homogeneity of the final product.

Finished feeds can be sampled during bagging-off by taking a handful from every fifth bag of 40-50 bags and pooling the individual samplings. The products are first visually inspected for homogeneity of size and colour and the amount of fines. The samples are then placed in tightly sealed containers and submitted for physical and chemical evaluation.

Physical evaluation of finished feeds includes the floating ability test for expanded feeds. The feed will be rejected if less than 85% floats after 15 minutes in water. For shrimp feed, water stability is of utmost importance in addition to nutritional quality. In general, water stability up to 3 hours is considered satisfactory.

Chemical analysis of the finished feeds is important to check the specifications, or claims, established for the finished products. Proximate analysis for moisture, crude protein, crude lipid, crude fibre, ash, and nitrogen-free extract (by difference) of the finished feeds is necessary to confirm nutrient content. This analysis also checks the manufacturing process. However, if raw material quality control is properly conducted and if process control is adequate, then the only chemical tests necessary for finished products on a regular basis are for moisture and crude protein. Periodic analysis of other components should also be scheduled, however.

Totally effective quality control for farm-made aquafeeds is impossible. However, physical inspections are appropriate and feasible and will ensure that the correct feeds are used. Inspection of the external features of the ingredients before use, which requires skill and experience, should include:

• assessing whether the ingredient is the normal colour;

• evaluating the odour of each ingredient, which will reflect its freshness and the effectiveness of its method of preservation. Odours such as “stuffy musk”, “rancid”, or “burning” indicate potential problems;

• feeling if the ingredient is moist (dry ingredients only!). If so, this may mean that the ingredient is mouldy or otherwise in poor condition;

• looking for the presence of insect and mould growth, indicating deteriorative changes in the feed and poor hygiene. Presence of scrap metals, stones, dirt or other non-biological contaminants indicate its impurity.

LIST OF REFERENCES

Agriculture and Cooperatives Ministerial Regulation. 1991. Name, category, type or description of animal feed. Government Gazette, Vol. 108, Part 218, 12 December 1991, Bangkok, Thailand. 103 p.

Chow, K.W. 1980. Quality control in fish feed manufacturing, p. 369-385. In Fish feed technology. ADCP/REP/80/11. FAO, Rome, Italy.

Department of Livestock Development. 1990. Animal Feed Quality Act, B.E. 2525. Department of Livestock Development, Ministry of Agriculture and Cooperatives, Bangkok, Thailand. 177 p.

Somsueb, P. 1993. Aquafeeds and feeding strategies in Thailand, p. 365-385. In M.B. New, A.G.J. Tacon and I. Csavas (eds.) Farm-made aquafeeds. Proceedings of the FAO/ AADCP Regional Expert Consultation on Farm-Made Aquafeeds, 14-18 December 1992, Bangkok, Thailand. FAO-RAPA/AADCP, Bangkok, Thailand.

Annex 1. Thai feedstuff regulation definitions

Current rules and regulations include the following definitions:

• materials intended for animal feed used as published in the Government Gazette by the Minister upon the advice of the animal feed quality control committee, are termed to be “animal feeds”;

• “produce” means manufacture, mix, transform, prepare, change the features of, or repack;

• “sell” means distribute, dispose of, give or exchange for commercial purposes and also includes possession for sale;

• “import” covers ordering and entry of materials into Thailand;

• “container” means any material specifically used for packing or wrapping animal feed;

• “label” includes any picture, imprint or statement shown on an animal feed container.

Further details of the regulations are provided by Somsueb (1993).

Annex 2. Non-nutrient compounds permitted in aquafeeds in Thailand

Anise oil
Vanillin
Monosodium glutamate
Saccharin sodium
All natural products and corresponding synthetic products
Pulverized bentonite
Hemicellulose plus lignin
Polymethylolcarbamide
Silicon dioxide or colloidal silica
Citric acid
Sodium, potassium and calcium stearates
Dilicic acid
Diatomaceous earth
Calcium silicate
Sodium aluminosilicate
Kaolin and kaolinitic clays
Natural mixtures of steatite and chlorite
Vermiculite
Lignosulphonates
Bentonite and other montmorillonite clays
Ball clays
Aluminium calcium silicate
Magnesium silicate
Hydrated sodium calcium aluminosilicate
Tricalcium silicate
Lecithins
Alginic acid
Alginate salts
Propylene glycol alginate
Propane-1, 2-diol alginate
Agar
Carrageenan
Furcellaran
Locust bean gum
Tamarind seed flour
Guar gum or guar flour
Tragacanth
Annex 2. (cont.)
Acacia or gum arabic
Xanthan gum
D-glucitol or sorbitol
Mannitol
Glycerol
Pectins
Microcrystalline cellulose
Methylcellulose
Ethylcellulose
Hydroxypropylcellulose
Hydroxypropylmethylcellulose
Ethylmethylcellulose
Chrondrus extract
Gum ghatti
Agar-agar
Carboxymethylcellulose
Monoacyl and diacylglycerols
Monoacyl and diacylglycerols esterified with the following acids: acetic, lactic, citric, tartaric, monoacetyltartaric and diacetyl-tartaric
Sucrose esters of fatty acids
Sucroglycerides
Polyglycerol esters of non-polymerized edible fatty acids
Propylene glycol esters of fatty acids, propane-1, 2-diol esters of fatty acids
Stearoyl 2-lactylic acid, sodium stearoyl-2-lactylate, calcium stearoyl-2-lactylate
Stearoyl tartrate
Glycerol poly (ethylene glycol) ricinoleate
Dextrans
Caramel
Sorbitan monostearate
Sorbitan tristearate
Sorbitan monolaurate
Sorbitan monopalmitate
Sorbitan mono-oleate
Polyoxyethylene (20) sorbitan monostearate
Polyoxyethylene (20) sorbitan tristearate
Polyoxyethylene (20) sorbitan monolaurate
Polyoxyethylene (20) sorbitan mono-oleate
Polyoxyethylene (20) sorbitan monopalmitate
Polyoxyethylene (20) sorbitan trioleate
Polyoxyethylene stearate
Penta sodium triphosphate
Polyethyleneglycol esters of fatty acids from soya oil
Polyoxyethylated glycerides of tallow fatty acids
Esters of polyglycerol and of alcohols obtained by the reduction of oleic and palmitic acids
Polypoly ethyelene glycol 6000
Polyoxypropylene-polyoxythylene polymers (M.W. 6,800-9,000)
Propane-1, 2-diol
Sodium, potassium and calcium salts of edible fatty acids of castor oil (polyglycerol polyricinoleate)
Patente blue V
Acid brilliant green BS or lissamine green
Carotenoids
Capsanthin
Beta-apo-8-carotenol
Ethylester of beta-apo-8 carotenoic acid
Lutein
Xanthophylls
Cryptoxanthin
Violaxanthin
Canthaxanthin
Zeaxanthin
Citranaxanthin

ON-FARM FEED PREPARATION AND FEEDING STRATEGIES FOR CARPS AND TILAPIAS
Amararatne Yakupitiyage

Division of Agricultural and Food Engineering Asian Institute of Technology GPO Box 2754, Bangkok, Thailand

YAKUPITIYAGE, A. 1993. On-farm feed preparation and feeding strategies for carps and tilapias, p.87-100. In M.B. New, A.G.J. Tacon and I. Csavas (eds.) Farm-made aquafeeds. Proceedings of the FAO/AADCP Regional Expert Consultation on Farm-Made Aquafeeds, 14-18 December 1992, Bangkok, Thailand. FAO-RAPA/AADCP, Bangkok, Thailand, 434 p.

INTRODUCTION

Finfish production dominates the inland aquaculture sector in the Asia and Pacific region. Out of nearly 7 million t of finfish produced by inland fish culture in 1990, fresh water fish feeding low in the food chain contributed 93% - 6.5 million t (Csavas 1992). This included 86% carps (5.6 million t) and 5% tilapias (0.33 million t), showing the importance of these species in human nutrition in the Asian region.

A notable feature of carp and tilapia culture in most developing countries is their relatively low farm gate value compared to marine or freshwater carnivorous fish. They are usually produced for local consumption rather than export. These two factors influence the type of feed used in the culture system, the feed preparation method and the feeding practice. For the same reasons, the use of commercial pelleted feed is prohibitively expensive in many countries. For example, in Thailand the farm gate value of carps and tilapias is about US $1.00/kg but it may require 2 kg of commercial feed costing US $0.50/kg to produce them!

The majority of carp and tilapia culture systems, therefore, are at least partially dependent upon natural food. This paper attempts to delineate possible feed preparation methods and feeding strategies for these fish by reviewing the existing literature.

OVERVIEW OF ON-FARM FEED PREPARATION

Farms are delineated by a boundary across which there is inflow and outflow of materials (Figure 1). Farmers also require knowledge of the nutrient requirements of fish, the physical and chemical properties of ingredients, and feed formulation and processing methods, which depend on past experience or have been contributed by research and development (R&D) institutes. Equipment and raw materials need to be purchased outside the farm boundary. Generally available multipurpose equipment may also be used depending upon the scale of operation. Capital and manpower requirements may also be directly related to the size of production.

Figure 1. Components of on-farm feed production

The chain of events comprises the acquisition of raw materials, their preprocessing, final processing, storage and/or feeding. Each step adds biological and economic value to the product. Added biological value represents an improvement of palatability (i.e. suitable particle size, taste), enhancement of nutritional quality (i.e. removing anti-nutritional factors) and an increment of nutrient digestibility. However, the value of improved product biological quality must exceed the accompanying cost of the improvement. Since carps and tilapias fetch a relatively low market price, a cost effective feeding regime is dependent upon the careful selection of feedstuffs and feed ingredients, utilization of an optimum feed formulation, selection of multi-purpose equipment for feed preparation, optimization of feed dispersion and minimization of feed waste. The remaining sections of this paper, therefore, are aimed at elucidating the above factors and at suggesting methods for optimal use of available feed resources for carp and tilapia culture.

SELECTION OF FEEDSTUFFS AND FEED INGREDIENTS

Schaeperclaus (1933) identified two types of farms on the basis of their feed/feeding regime in carp culture in Europe, namely:

• through food or natural food supplied by the pond, equivalent to pasture feeding of terrestrial animals;

• through feeding or feed supplied by the farmer, analogous to stall feeding of terrestrial animals.

  Using the same classification, on-farm feeds for carps and tilapias may be divided into two major groups:

• natural food produced by pond fertilization within the culture system;

• addition of pre-prepared feed to the culture system (Figure 2).

Figure 2. Classification of on-farm feeds

Use of natural feeds through pond fertilization

The main objective of fertilization is to enhance autotrophic and heterotrophic food production. The method of fertilization will be mainly dependent upon the feeding behaviour of fish. For example, enhanced autotrophic production may be sought for species such as tilapias, Indian major carps (rohu, catla) and Chinese carps (silver carp and bighead carp). However, if the target species feed upon benthic organisms or detritus, autotrophic production may indirectly contribute to the fish yield. Hence, a common practice is to grow the bottom feeding carp with filter feeding tilapias in polyculture in order to utilize the different niches of the fish pond ecosystem.

The in situ production of natural feed is mainly fulfilled by fertilizing fish ponds with organic manure, inorganic fertilizers or a combination of both. Fertilizing fish ponds is still an art rather than a science. Coleman and Edwards (1987) extensively reviewed different fertilization methods suitable for carp and tilapia culture. Further details on practical fertilization methods used by different authors can be found in Tacon (1988). Edwards (1990) showed that even though the nitrogen (N) and phosphorus (P) conversion efficiency of organic and/or inorganic fertilization of carp/tilapia culture systems are marginally less than feeding fish with complete feed, fertilized systems pollute the environment less than intensive complete feed-based culture systems with water exchange to the surrounding environment. Edwards (1990) also emphasized that a reduction in feed cost can be mainly achieved through fertilizing fish ponds with N and P, rather than attempting to increase yields through the use of complete diets. Potential yields of tilapias and carps under different fertilization regimes are shown in Table 1.

Fertilization rates are dependent upon the micro-ecological conditions including soil type and pond history. It has been demonstrated that the maximum N and P fertilization rates of Asian Institute of Technology (AIT) ponds in acid sulphate soils with considerable P binding capacity are 4 kg N and 2 kg P/ha/day (AIT/CRSP unpublished data). Beyond the above N loading rate, ammonia toxicity develops and the fish standing stock may collapse. However, it should be emphasized that these rates may not be appropriate for other sites where the pond ecology differs from AIT conditions. A number of site specific fertilization rates for carp and tilapia pond culture are summarized by Tacon (1988).

Apart from conventional fertilization techniques, a number of novel fertilization methods have been attempted to produce on-farm natural feeds. For example, the use of waste arising from luxury fish species raised intensively on complete feeds as food for species that feed lower in the aquatic food chain has been attempted, including:

• stock manipulation (i.e. the polyculture of catfish and tilapia at a ratio of 10:1 is an existing practice in Thailand);

• luxury species such as walking catfish (Clarias spp.) are cultured in a cage utilizing formulated feed with faecal waste passing to the pond functioning as fertilizer for tilapia stocked outside the cage (Lin, unpublished data); and

• feeding large fish with formulated feed within a cage and producing fingerlings through the fertilizing effect of fish excreta in the pond in which the cage is suspended (Lin, unpublished data).

* polyculture = tilapia + silver barb + common carpSource: AIT (unpublished data)

Artificial feeds

The phrase “on-farm feeds” creates an instant picture of artificial feed in the mind of the majority of fish nutritionists. Feeding fish either a single feed (e.g. rice bran or leaves) or a simple mixture is usually undertaken. These mixtures may be further processed into a wet dough or pellet to minimize dispersion losses. Pellet production provides additional advantages such as an ability to prepare feed in advance, and ease of storage.

Terrestrial based live feed, such as earthworms and maggots, can also be produced on-farm using various organic wastes. Naturally occurring terrestrial live feeds such as termites and snails can also be used depending upon their availability.

Energy supplements

The addition of energy supplements has been mainly used in carp culture. A closer look at the species used in these systems indicates that benthic feeders mainly benefit from energy supplements rather than phytoplankton feeders (Table 2). A plausible reason for this phenomenon is that the relative scarcity of benthic organisms makes a pronounced protein sparing effect of the energy supplement. As phytoplankton is abundant in well fertilized culture systems, the protein sparing action of the energy supplement may not be important to plankton feeding fish.

Source:
(1) Yakupitiyage et al. (1989); 
(2) Xin (1989);
(3) Spataru et al. (1980)

Szumeic (1969) and Gurzeda (1969) recommended that for common carp every unit of ingested natural feed should be supplemented with three units of an energy supplement (i.e. barley) by weight. Viola et al. (1988) reported that at least 2% lipid supplement was necessary within animal protein-free formulated feeds for tilapia. However, if the ponds were well fertilized, added oil had no significant positive effect on tilapia growth (Hung 1989).

Commonly used energy supplements in Asia include agricultural byproducts such as rice bran, broken rice and maize. If the particle size is small, special preparation of these ingredients is not necessary. However, cooked carbohydrate (i.e. cooked cassava tubers) has certain advantages over non-cooked feed due to its higher palatability and digestibility, and the ability to use it as a wet dough. Feeding rates for rice bran, as reported by Ling (1967), are presented in Table 3.

Source: Ling (1967)

Protein supplements

Empirical data show that approximately 10 g of digestible protein/m² may be available to plankton feeding fish at maximum fertilization rates, leading to maximum autotrophic production (Yakupitiyage, unpublished data). Fish dependent upon secondary production certainly receive less protein than those dependent upon autotrophic production. A protein deficit may develop if the absolute protein requirement of the fish population per unit area is higher than the available protein from natural food.

There are two important factors relating to fish which should be consid- ered: stocking density and body weight. For example, it has been estimated that a protein deficit usually develops at an individual body weight of fish higher than 200 g, at a stocking density of 2 fish/m². However, if the stocking density is 0.25 fish/m², a protein supplement is required only after fish attain 1 kg body weight (Yakupitiyage, unpublished data). The latter stocking density is widely used in Indian carp polyculture systems.

Several examples of protein supplements used in different countries are summarized in Table 4. It has been demonstrated by a number of authors that plant protein can be used as the main protein source for carps and tilapias without need for the addition of an animal protein component (Viola et al. 1988; Nandeesha et al. 1989; Hung 1989). Other possible ingredients are listed by Tacon (1987). However, the presence of anti-nutritional factors should be taken into consideration during the selection of ingredients for feed preparation.

Source:
(1) Ling (1967);
(2) Nandeesha et al. (1988);
(3) Hung (1989)

Mineral and vitamin supplementation

Viola et al. (1988) demonstrated that the addition of 2% di-calcium phosphate was necessary for both carps and tilapia to use soybean-based diets efficiently. Hung (1989) also demonstrated the positive growth effect of the addition of di-calcium phosphate in a well fertilized culture system for tilapia.

However, it is questionable whether calcium is required in these culture systems. As calcium and phosphorus were supplied together, it is reasonable to conclude that either, or both, should be added to supplementary feeds. No data is available concerning the requirement for other minerals. However, it is a common practice in China and India to use common salt as an ingredient in formulated feeds.

The addition of vitamins to supplementary feeds is controversial. Hepher (1972) showed that there was an interactive effect between dietary vitamin supplementation and carp stocking density (Table 5). For example, although added vitamins had no growth stimulating effect in carp stocked at 0.2 fish/m², when stocking density was raised to 0.6 fish/m², there was a marginal improvement in food conversion efficiency (FCR) and a 15-20% increase in fish yield. Clearly, further research in this field is warranted.

Source: Hepher(1972)

FEED PREPARATION

Existing feed preparation techniques for carps and tilapias are simple. Terrestrial based live feeds such as termites and silkworm pupae are added directly to fish ponds. Snails are crushed before being offered to fish. Leaves are usually chopped when the fish are small but larger fish (>50g) are fed directly in Chinese carp culture systems. If dry ingredients are used, a well ground mixture can be offered in feeding bags or the mixture simply dispersed throughout the pond. A reduction in feed waste can be achieved by cooking the carbohydrate source and mixing other ingredients with the gelatinized starch. Such mixtures can be either directly fed as a wet dough or can be pelleted by a simple extrusion process. Prepared pellets may be sun dried and stored for future use. Examples of several protein supplements used as feed for carp and tilapia, including reported FCR, are given in Table 4.

FEEDING

There are three important factors which should be considered when feeding carps and tilapias, namely:

• fluctuation of feed intake and digestibility at ad libitum feeding (De Silva and Perera 1983, 1984; De Silva and Gunasekera 1989; Yakup-itiyage 1989);

• total dry matter loading rate into the system (AIT, unpublished data); and

• total nitrogen (N) loading rate into the system (AIT, unpublished data).

Feed intake fluctuation

It has been observed under laboratory conditions that there is a fluctuation of feed intake and protein digestibility in fish. De Silva and Perera (1983) suggested that this problem could be overcome by feeding fish with two different diets: a high protein diet followed by a low protein diet. Since it is cumbersome to prepare two diets of different composition and a low protein diet increases the dry matter loading rate to the system, Yakupitiyage (1989) suggested that feeding fish at different rates (i.e. a high feeding rate followed by low feeding rate) may be more suitable than using two different feeds. However, as supplementary feeds are usually not used for ad libitum feeding, more on-farm feeding trials should be conducted before recommending any of these feeding practices to farmers.

Total dry matter loading rate

The dry matter loading rate into a static culture system will directly influence the early morning dissolved oxygen concentration. The possible upper dry matter loading rate for a short duration (3-6 months) for tilapia/carp polyculture systems was approximately 100 kg feed DM/ha/day to maintain dissolved oxygen at approximately 1 mg/litre at dawn (AIT unpublished data). However, the exact loading rate may depend upon the pond history, as previous organic matter accumulation in the pond bottom may affect the community respiration rate. Ling (1967) reported that 13 t of green fodder, 7.2 t of agricultural by-products and 2 t of animal products were used in Chinese carp polyculture systems at a stocking density of 250 grass carp, 2,000 silver carp and bighead carp, 3,000 mullet, 2,000 milkfish, 1,000 mud carp and 500 common carp for a 10 month period. Assuming green fodder contains 75% moisture and other feed materials contain 10% moisture, the loading rate was approximately 38 kg DM/ha/day.

Total nitrogen loading rate

Table 6 shows the potential N loading rate with pelleted diets of different dietary protein levels. It was assumed that the maximum possible N loading rate was 4 kg N/ha/day, as above these N loading rates ammonia toxicity affects fish growth (AIT, unpublished data). Maximum feeding rate was assumed to be 100 kg DM/ha/day with 80% protein digestibility and a 50% retention of digested protein. The table also demonstrates that reducing dietary protein level may result in reducing the fish population density which can be supported by the supplementary feed, as the absolute protein supplement to the system decreases with decreasing dietary protein content.

* calculations are based upon AIT conditions (unpublished data)

CONCLUSIONS

There is great scope for producing fish economically by utilizing natural food through proper pond fertilization rather than using more expensive formulated feeds. Unfortunately, many fish nutritionists do not consider pond fertilization as a method of producing fish feed and advocate the use of formulated feeds in carp and tilapia culture. Before attempting to draw up recommendations on possible farm-made formulated feeds, investigations on the potential nutritional limiting factors of these culture systems are required.

Farming systems may be divided into industrial agriculture systems, resource-rich farming systems and resource-poor farming systems. Industrial large scale enterprises have apparently had little interest in carp and tilapia culture and the majority of culture systems in Asia belong to the resource-rich or resource-poor farming systems. Formulated feed may have a role in these farming systems provided that the cost of feed is relatively low and the stocking density or fish biomass has been deliberately manipulated to maximise the utilization of the natural food present within the system (i.e. that fish biomass is higher than the critical standing crop). People who are most in need of on-farm made aquafeeds for intensifying their culture operation are generally resource-poor farmers and their use of formulated feed will be mainly dependent upon the availability of feed ingredients. A question that requires immediate attention is how to make best use of the low quality feed ingredients available to resource-poor farmers. There may be several answers to this question but current strategy is to use them as supplementary feed in fertilized systems rather than as an ingredients within complete feeds.

LIST OF REFERENCES

Coleman, J.A. and P. Edwards, 1987. Feeding pathways and environmental constraints in waste-fed aquaculture: balance and optimization, p. 240-279. In D.W.J. Moriarity and R.S.V. Pullin (eds.) Detritus and microbial ecology in aquaculture. Proceedings of the Conference on Detrital Systems for Aquaculture, 26-31 August 1985, Bellagio, Como, Italy. ICLARM Conference Proceedings 14. ICLARM, Manila, Philippines.

Csavas, I. 1992. Recent developments and issues in aquaculture in Asia and Pacific. Paper presented at the APO Seminar on Aquaculture, 25 August-4 September, Tokyo, Japan.

De Silva, S.S. and R.M. Gunasekera. 1989. Effect of dietary protein level and amount of plant ingredient (Phaseolus aureus) incorporated into the diets on consumption, growth performance and carcass composition in Oreochromis niloticus fry. Aquaculture 80:121-133.

De Silva. S.S. and M.K. Perera. 1983. Digestibility of an aquatic macrophyte by the cichlid Etroplus suratensis with observations on the relative merits of three indigenous components as markers and daily changes in protein digestibility. J. Fish Biol. 23:675-684.

De Silva. S.S. and M.K. Perera. 1984. Digestibility in Sarotherodon niloticus fry: effect of dietary protein level and salinity with further observations on variability in digestibility. Aquaculture 38:293-306.

De Silva. S.S., R.M. Gunasekera and C. Keembiyahetty. 1986. Optimum ration and feeding frequency in Oreochromis niloticus young, p. 559-564. In J.L. Maclean, L.B. Dizon and L.V. Hosillos (eds.) The First Asian Fisheries Forum. Asian Fisheries Society, Manila, Philippines.

Edwards, P. 1990. Environmental issues in integrated agriculture-aquaculture and waste water-fed fish culture systems. Paper presented at Bellagio conference on Environmental and Third World Aquaculture Development, September 1990, Bellagio, Italy.

Gurzeda, A. 1969. Relationship between natural and artificial food in carp nutrition, p. 118-130. In New developments in carp and trout nutrition. Papers submitted to a symposium held in conjunction with the 5th session of EIFAC, Rome, May 1968. EIFAC Tech. Paper 9. FAO, Rome, Italy.

Hepher, B. 1972. Supplementary feeding in fish culture, p. 183-198. In Proceedings of the 9th International Congress of Nutrition, Mexico 1972, Vol. 3.

Hung, L.T. 1989. Evaluation of soybean meal as a supplementary feed for red tilapia. M.Sc. Thesis, 79 p. Asian Institute of Technology, Bangkok, Thailand.

Ling, S.W. 1967. Feeds and feeding of warm-water fishes in ponds in Asia and the Far East. FAO Fish. Rep. No. 44 (3):291-309.

Nandeesha, M.C., G.K. Sriknath, T.J. Varghese, P.Keshwanath and H.P.C. Shetty. 1989. Growth performance of Indian major carp, Catla catla on fish meal free diet, p. 137-142. In Fish Nutrition Research in Asia. Proceedings of the Third Asian Fish Nutrition Network Meeting. Asian Fish. Soc. Spec. Publ. 4. Asian Fisheries Society, Manila, Philippines.

Schaeperclaus, W. 1933. Textbook of pond culture. Paul Parely Book Publishing House, Berlin. 261 p.

Spataru, P., B. Hepher and A. Halevy. 1980. The effect of the method of supplementary feed application on the feeding habits of carp (Cyprinus carpio. L) with regard to natural food in ponds. Hydrobiologia 72:171-178.

Szumeic, J. 1969. Relationship between natural and artificial food in carp feeding, p. 140-160. In New developments in carp and trout nutrition. Papers submitted to a symposium held in conjunction with the 5th session of EIFAC, Rome, May 1968. EIFAC Tech. Paper 9. FAO, Rome, Italy.

Tacon, A.G.J. 1987. The nutrition and feeding of farmed fish and shrimp - A training manual. 2. Nutrient sources and composition. FAO Field Document, Project GCP/RLA/075/ ITA. Field Document 5/E, Brasilia, Brazil. 129 p.

Tacon, A.G.J. 1988. The nutrition and feeding of farmed fish and shrimp - A training manual. 3. Feeding methods. FAO Field Document. Project GCP/RLA/075/ITA. Field Document 7/E, Brasilia, Brazil. 208 p.

Viola, S., O. Rappaport and G. Zohar, 1988. Animal protein free feeds for hybrid tilapia (O.niloticus x O. aureus) in intensive culture. Aquaculture 75: 115-125.

Xin, Zhang. 1989. Rice bran as an energy-rich supplementary feed in a fertilized system for nile tilapia. M.Sc. Thesis, 87 p. Asian Institute of Technology, Bangkok, Thailand.

Yakupitiyage, A. 1989. Quantitative and qualitative aspects of energy acquisition of the cichlid fish Oreochromis niloticus. Ph.D. Thesis, 237 p. Institute of Aquaculture, University of Stirling.

Yakupitiyage, A.,P. Edwards and K.L. Wee. 1989. Supplementary feeding of fish in a duck-fish integrated system. I. The effect of rice-bran, p. 143-157. In S.S. De Silva (ed.) Fish Nutrition Research in Asia. Proceedings of the Fourth Asian Fish Nutrition Workshop. Asian Soc. Spec. Publ. 5. 205 p. Asian Fisheries Society, Manila, Philippines.

ON-FARM FEED PREPARATION AND FEEDING STRATEGIES FOR CATFISH AND SNAKEHEAD
Wimol Jantrarotai and Patanai Jantrarotai

National Inland Fisheries Institute Kasetsart University Campus, Bangkok 10900, Thailand

JANTRAROTAI, W. and P. JANTRAROTAI. 1993. On-farm feed preparation and feeding strategies for catfish and snakehead, p.101-119. In M.B. New, A.G.J. Tacon and I. Csavas (eds.) Farm-made aquafeeds. Proceedings of the FAO/AADCP Regional Expert Consultation on Farm-Made Aquafeeds, 14-18 December 1992, Bangkok, Thailand. FAO-RAPA/AADCP, Bangkok, Thailand, 434 p.

INTRODUCTION

Inland aquaculture has been practised in Thailand for over five decades with various degrees of diversity. There are about 27 species of fish being cultivated with numerous types of systems ranging from super-intensive farming for commercial production to extensive, mainly for home consumption. More than 60,000 inland farms with a total cultivated area of 49,752 ha exist. Over 95% of the total area consists of pond and paddy-field type culture systems. The remainder are dammed-up ditches, swampy areas and cage culture systems. Total Thai freshwater aquaculture production for the year 1989 was 92,000 t with the top seven species being Tilapia nilotica, Pangasius sutchi, Puntius gonionotus, Trichogaster pectoralis, Clarias spp, Macrobrachium rosenbergii and Channa striatus.

Out of the 27 species being reared, Clarias spp. (catfish) and Channa striatus (snakehead) are the most intensively cultured. The popularity of catfish farming has been due to its short culture cycle, rapid growth rate and extreme tolerance to inferior water quality and limited water supply. Snakehead, on the other hand, has drawn farmers' attention due to its high farm-gate value. According to official statistics, the production of catfish and snakehead together represents 18% of the total freshwater aquaculture production, with an annual value of US$21 million. Catfish production has merely doubled since 10 years ago and that of snakehead is little changed. The reason for the slow increase in total production is that catfish farmers have been facing difficulties such as disease and limited availability of, and high cost, fish feeds. Over-production also resulted in depressed prices, rendering catfish culture less profitable. For snakehead, a decline in wild seed due to overfishing, destruction of spawning grounds and the unreliable availability and rising price of trash fish (the main feed for snakehead) has hindered expansion.

Despite the problems facing further expansion of catfish and snakehead farming, production is quite considerable; according to official statistics an average of about 12,000 t of catfish and 4,000 t of snakehead have been produced annually over the past five years. 75-95% of the production was from the 25 provinces in the central part of Thailand. Culture techniques for these two species have not changed much in the last 10 years. From the nutritional point of view, most farmers still rely on fresh moist feeds prepared on-farm, the so called “farm-made aquafeeds”. However, in the future, some modification in feed type and composition may be possible to suit the economic situation of these farms and to increase the farm-gate value of the fish. The purpose of the present paper is to review the current status of farm-made feed preparation and feeding strategies for catfish and snakehead, including an economic comparison between the use of farm-made feeds and commercial feeds.

CULTURE PRACTICES

Catfish

Clarias catfish belong to the family Clariidae and are found throughout Southeast Asia, the Indian subcontinent, and Africa. They are distinguished by the possession of an accessory air breathing organ, which enables them to exist for a long time out of water, or indefinitely in oxygen-poor waters (Bardach et. al. 1975). Culture of catfish is established in Thailand where Clarias batrachus and C. macrocephalus support a thriving domestic demand. Recently, hybrid catfish (C. macrocephalus x C. gariepinus) were introduced because of their faster growth rate than either of the local species.

Grow-out catfish ponds are 300-2,000 m² in area, and 1-1.5 m deep. Little preparation is required for firming the banks or pond bottom since available soils usually retain water adequately. Water supply is introduced and discharged by a simple canal and weir, or is pumped in and out by long-tail pumps. 2-3 cm fry are stocked in production ponds at rates ranging from 60-300/m², depending upon their availability and price as well as the intended production. Most operations achieve two crops per year. Each crop requires 3-5 months to produce marketable sized fish of 120-200 g. The incidence of mortality and the quantity and quality of feed used determines the fish yields obtained, which range from 3-12 kg/m² per crop.

Snakehead

Voracious carnivores, snakeheads are intensively cultured by farmers with access to capital. The operations of small farms are restricted by the very high capital requirement, largely due to high feed costs. Snakehead grow-out ponds, which are more or less similar to those used for growing catfish, range from 400-3,200 m², with a depth of 1.5-2 m. The rearing system depends on adequate, good quality water which is supplied by gravity or by daily water exchange through pumping. Wild fry, with an average length of 1.5-2.0 cm, are stocked by weight at the rate of 0.1-0.4 kg/m². Fry numbers are extremely variable ranging from 80-800 fry/m². The reason farmers stock at such a relatively high density is to compensate for high mortality rate caused by diseases and cannibalism. Thus, they expect that enough fry will survive to produce a profitable crop. The survival rate of snakehead in grow-out ponds is fairly low, from 7-24%. Culture time to a marketable size of 500-1,000 g ranges from 7-10 months, depending on stocking density and the feed used. Production from successful operations can be as high as 8-16 kg/m²/crop.

Many snakehead farmers take less risk by avoiding the high mortality experienced during the first two months by stocking ponds with 10-15 g fingerlings or 50-70 g sub-adults at rates of 30-50 fish/m² and 20 fish/m², respectively. By stocking larger fish, the survival rate until harvesting may be as high as 85-90%, and the time to reach marketable size is reduced from 9-10 to 6-8 months.

NUTRIENT REQUIREMENTS

Clarias catfish

The success in rearing catfish to date, using a simple 8:2 trash fish: rice bran diet has created little interest in exploring the nutrient requirements of the fish per se. Rather, most studies have emphasized comparisons of trash fish based diets with mixed feeds, by assessing their quality and cost effectiveness. These studies have been based on the comparative growth performance of fish fed various ingredient combinations. Meeting specific nutrient requirements has not been an important criteria for feed formulation, since little was known about the nutrient requirements of catfish.

Thongutai (1969) reported the superior growth of catfish fed 9:1 trash fish:rice bran diet to those receiving Auburn No. 2 pelleted feed, in which fish meal and soybean meal were included as protein sources. Sitasit (1970) found that pelleted feed with 28% protein was not as good for catfish as trash fish/rice bran in a 6 month trial. Nutritional imbalance in the pelleted feed was suspected. Srisuwantach et al. (1981) reported similar findings when they compared the effects of trash fish and pelleted feed in catfish grow-out operations. However, a huge outbreak of disease in pond-reared clarias catfish in late 1982, which was traced to the use of poor quality trash fish, shifted attention to mixed feeds. Research to determine the basic nutrient requirements of catfish began in early 1980. Using a diet with a 2:1 fish meal:soybean meal ratio as the protein source, fed at graded levels to catfish for 45 days, Chotiyarnwong and Chuapoehuk (1981) found that the optimum protein requirement for grow-out Clarias batrachus was 30%. Boonyaratpalin (1988) estimated the protein requirement for catfish to be 35-40%, 25-35% and 28-32% for fry, grow-out and broodstock, respectively. Clarias catfish appear to require the same ten essential amino acids as other fish. However, quantitative amino acid requirements have not yet been determined. Requirements for total lipids and essential fatty acids are also not yet known. However, it has been suggested that the dietary inclusion of 6-8% fish oil or soybean oil would provide sufficient essential fatty acids for catfish (Boonyaratpalin 1988). Tacon (1981) implied that clarias catfish are probably able to equally utilize n-3 or n-6 fatty acids, as are channel catfish.

Little is known about the ability of catfish to utilize carbohydrates for energy. Broken rice, a high carbohydrate ingredient, has long been used at high levels in combination with trash fish and rice bran for maturing catfish (Sitasit 1968), however. Based on the analysis of the partition between protein and non-protein energy retention, Luquet and Moreau (1990) showed that clarias catfish efficiently utilize non-protein energy from carbohydrates and hence can improve protein retention. Jantrarotai et al. (1992) reported that catfish can well utilize up to 49% carbohydrate in diets without any detrimental effects. However, reduction in growth was observed when dietary carbohydrate was increased to 54%.

The optimum energy to protein ratio for best growth of Clarias batrachus was reported by Tanomkiate (1984) to be 5.8 kcal/g of protein. This value was much lower than those reported for channel catfish, which ranged from 7.5-11 kcal/g of protein (Lovell and Prather 1973; Garling and Wilson 1976).

Vitamins and minerals play an important role in clarias catfish nutrition in the sense of increasing fry survivval. Taechajanta and Sitasit (1981) found that fry survival was 79% for a group fed mixed feed with vitamin and mineral premixes, in comparison to 35% survival rate in the group fed a diet with no premix. The individual vitamins required by catfish were later determined by Butthep et al. (1983) and Sitasit et al. (1984). The requirements and deficiency signs of some vitamins are summarized in Table 1. No study has been done to determine the mineral requirements of clarias catfish but the skull fractures often observed in catfish are believed to result from calcium deficiency (Sitasit 1970). Bone meal supplementation of the feed corrects the problem.

* N = not required in the diet
** RD = required but dose was not determined

Snakehead

The nutrient requirements for snakehead are much less well known than those for clarias catfish. Snakehead is a strict carnivore which is believed to have a high protein requirement. Boonyaratpalin (1980) reported that the protein requirement for snakehead fed on isocaloric diets (3.1 kcal/g) was 43% and 36% for snakehead fry and one month old fish, respectively. Effects of lipid levels on growth, survival and feed conversion of young snakehead were also evaluated. The minimum level of lipid, which could be n-3 or n-6, required for maximum growth, survival and feed conversion of the fish, appeared to be 6% (Boonyaratpalin 1981a). In another study, Boonyaratpalin (1981b) reported that pantothenic acid is a particularly important vitamin for snakehead growth and survival.

FEEDS OF CHOICE

Clarias catfish

Feeding of catfish in the past was heavily based on trash fish, rice bran and broken rice. The normal on-growing diet was 8:2 trash fish:rice bran. As a fattening diet, 8:1:1 trash fish:rice bran:broken rice was used. The mixtures were hand mixed and minced in a meat mincer to form a slurry. The slurry was hand distributed into ponds until it was all used up or until the fish no longer responded to the food.

Problems with trash fish based feeds

Although trash fish based diets were regarded as one of the best feeds for catfish, many limitations have forced farmers to modify the diets to some extent. As mentioned earlier, catfish farms in Thailand are mainly situated in the central plain region. Trash fish, therefore, had to be transported from port areas. Without proper storage, trash fish spoils easily and its quality can deteriorate during transportation. This can result in risks of transmitting disease. The national disease outbreak in pond-reared freshwater fish in late 1982 was partially blamed on unsanitary trash fish used for feed.

The rapid expansion of shrimp farming in Thailand in the past seven years has led to a high demand for shrimp feeds. Approximately 160,000 t of shrimp feeds are produced annually. This, in turn, results in high competition for raw trash fish for making fish meal, a major ingredient of shrimp feeds. Raw trash fish has, therefore, become more scarce for use in catfish grow-out. Limited availability of trash fish will further dictate an increase in price.

Demand for catfish is mainly for domestic consumption, which cannot be expanded very rapidly. However, when fast growing Clarias hybrids were recently introduced, catfish production increased beyond demand causing farmgate value to fall. The use of trash fish diets will no longer be feasible as their cost may exceed the value of the fish produced.

Since 1980, many commercial catfish feeds have been available on the market. Although catfish feeds have many advantages over trash fish-rice bran diets they were rarely accepted by fish farmers at first. The high unit cost and low bulk volume of commercial feeds in comparison to moist trash fish-rice bran diets were the main concerns of the fish farmers. However, the situation has now completely changed. More and more catfish farmers, particularly those who have no access to trash fish or other cheap ingredients, rely on commercial feeds.

According to a survey in 1992, feeds of choice for intensive clarias catfish farming can be classified into 3 categories: commercial, fresh feeds and farm-made feeds.

Commercial feeds

Commercial feeds are widely used in catfish operations. According to industry sources, approximately 60,000 t of commercial catfish feeds are produced annually, which would place catfish production, allowing for that also produced through the use of farm-made feeds, at about 45,000 t, much higher than official statistics indicate. Catfish feeds are manufactured in several types according to fish size and their nutrient requirements (Table 2). The declared ingredients in commercial catfish feeds are fish meal, soybean meal, full fat soybean meal, corn, broken rice, fresh rice bran, coconut oil meal, vitamins and minerals. Commercial feeds possess good water stability and floating ability and allow easy management of feeding rate. The average price for commercial grower feeds is US$ 500/t. The feed conversion ratio (FCR) for commercial catfish feeds ranges from 1.5-2.0:1, depending upon stocking density and pond management. Taking an average FCR of 1.7:1, the feed cost is US$ 0.85/kg of catfish produced. Most catfish farmers admit that if the farm-gate value of catfish drops below US$ 1.00, the use of commercial feeds is not viable.