..............................................................................................

Aquaculture Nutrition 2002 8;121^137

Eect of culture system on the nutrition and growth performance of Pacic white shrimp Litopenaeus vannamei (Boone) fed dierent diets
A.G.J. TACON1, J.J. CODY2, L.D. CONQUEST2, S. DIVAKARAN2, I.P. FORSTER2 & O.E. DECAMP2
1

Halliday Place, Kaneohe, HI, USA; 2The Oceanic Institute, Waimanalo, HI, USA

Abstract
Two 8-week feeding trials were conducted with juvenile Pacic white shrimp, Litopenaeus vannamei (Boone) to compare the growth and performance of animals fed a series of experimental and commercial pelleted shrimp and sh feeds and dietary feeding regimes within an indoor running-water culture system and an outdoor zero-water-exchange culture system. The best overall shrimp growth performance was observed for animals fed the experimental shrimp diet and all-day feeding regime under outdoor zero-water-exchange culture conditions. Final body weight and average weekly growth rate under these conditions were 2.8 and 3.4 times greater, respectively, than animals of similar size fed with the same diet under indoor running-water culture conditions. Although direct comparison between indoor and outdoor culture systems is dicult because of the lower indoor water temperatures, and consequently lower mean daily feed intake of animals, it is believed that the higher growth and feed performance of animals reared under outdoor green-water culture conditions was primarily due to their ability to obtain additional nutrients from food organisms endogenously produced within the zero-waterexchange culture system. The most promising features of zerowater-exchange culture systems are that they oer increased biosecurity, reduced feed costs and water use for the farmer, and by doing so provide a potential avenue of moving the shrimp culture industry along a path of greater sustainability and environmental compatibility.
KEY WORDS:

Introduction
Of the estimated 375 913 shrimp farms reportedly in existence in the world in 1999, 54% used extensive pondbased growout culture systems (stocking density below 2.5 m2, shrimp production 50500 kg ha1 year1, production costs US$13 kg1 live shrimp), 28% used semi-intensive pond-based growout culture systems (stocking density below 30 m2, shrimp production 5005000 kg ha1 year1, production costs US$26 kg1 live shrimp), and 18% used intensive pond-based growout culture systems (stocking density above 30 m2, shrimp production 500020 000 kg ha1 year1, production costs US$48 kg1 live shrimp; Rosenberry 1999). Moreover, although over 1.1 million tonnes of marine shrimp (valued at US$6.8 billion) were produced in 1998 (FAO 2000), little or no information exists concerning the optimum dietary nutrient levels for rearing these species under practical pond-based culture systems (Lawrence 1996; Tacon 1996). As a result of the pressure faced by the shrimp farming community for increased biosecurity, and disease and euent control (Bullis & Pruder 1999), there has been a trend within many countries towards the development of biosecure closed shrimp production systems, including zero-water-exchange or recirculating culture systems employing in situ (McIntosh 1999; Avnimelech 2000; McNeil 2000) or external bioltration techniques (Reid & Arnold 1992; Moss et al. 1998; Ogle & Lotz 2000; Van Wyk 2000). Trials of zero water exchange systems were conducted in Tahiti during the 1980s with Litopenaeus vannamei and Penaeus monodon and with yields of approximately 20 000 kg ha1 year1 (AQUACOP, personal communication). This paper describes two feeding trials conducted from July to September 1999 at the Oceanic Institute (OI), Hawaii, USA. The objective was to compare the growth and performance of juvenile Pacic white shrimp

diet, feeding regime, Litopenaeus vannamei, methodology, Pacic white shrimp, zero water exchange

Received 22 February 2001, accepted 15 August 2001 Correspondence: I.P. Forster, The Oceanic Institute, 41-202 Kalanianaole Highway, Waimanalo, HI 96795, USA. E-mail: iforster@oceanicinsitute. org

..............................................................................................

2002 Blackwell Science Ltd

121

122

A.G.J. Tacon et al.

L. vannamei (Boone) fed a series of dierent practical shrimp feeds and dietary feeding regimes within an indoor running-water culture system and an outdoor zero-waterexchange culture system. stocking density of 100 shrimp tank1 (equivalent to a shrimp density of 51 m2 cone surface area, 55 m2 at bottom surface area or 71 m3 water volume), with three tanks allotted per dietary treatment. Water within the microcosms was continuously mixed and aerated using six air lift tubes (to keep all particulate matter in suspension) and a zerowater-exchange green water management system operated within the tanks for the duration of the 56-day culture trial (for tank conguration and operation see Freeman & Duerr 1991). Air was continuously supplied to all experimental tanks with an EG&G Rotron 5 HP regenerative blower (Saugerties, NY, USA). Freshwater was used as required to replace evaporative losses. Diurnal water temperature, dissolved oxygen, pH and salinity measurements throughout the study were recorded (Table 1).

Materials and methods

Shrimp and experimental culture conditions
Pacic white shrimp L. vannamei (Boone) were obtained from the Oceanic Institute shrimp hatchery (industry production run, N-993 strain, Moss et al. 2001a) and fed initially with a 520-g kg1 protein commercial larval shrimp diet (Higashimaru Co. Ltd, Kagoshima, Japan), and later a 350400 g kg1 protein commercial nursery shrimp diet (Rangen, Inc., Buhl, ID, USA) prior to the start of the two 56-day feeding trials. In the indoor feeding trial, juvenile shrimp of mean initial weight 1.58 (0.05 standard deviation) g were stocked within indoor rectangular glass aquaria (0.76 0.31 0.31 m; 52-L water volume) at an initial stocking density of 24 shrimp aquaria1 (equivalent to a shrimp density of 100 m2 bottom surface area or 461 m3 water volume), with three aquaria allotted per dietary treatment [laboratory studies conducted at OI using these culture systems showed no dierence in the growth or survival of shrimp reared at densities of 50 m2 or 100 m2 (unpublished data)]. A seawater ow-through system with a water exchange rate of 100% hour1 was employed for the duration of the experiment (water temperature ranged from 26 to 27 C). The aquaria were cleaned every morning before rst feeding by siphoning out any uneaten feed, faeces, moults, or dead shrimp that were present. A 12-h photoperiod was maintained within the indoor laboratory using uorescent lighting (daylight hours from 06.00 to 18.00 hours). In the outdoor feeding trial, juvenile shrimp (of the same strain and size as above) were stocked within outdoor freestanding 1500 L cylindrical black-coated breglass microcosm tanks (1.52 m dia with a conical bottom) at an initial

Diets and feeding protocols

Tanks were randomly assigned one of four diets in both feeding trials: a sinking pelleted shrimp diet (OI shrimp diet) formulated to contain 350 g kg1 protein and 25 g kg1 squid meal (Tables 2 and 3); a commercially available sinking pelleted shrimp diet formulated to contain 350 g kg1 protein and 25 g kg1 squid meal; and a commercially available pelleted catsh diet in two forms (pelleted-oating, and crumbled-sinking) formulated to contain 370 g kg1 protein. The OI shrimp diet was prepared by rst mixing all the major dry feed ingredients (previously ground in a hammer mill to pass through a 60-mesh or 0.25 mm screen) for 15 min in a Hobart food mixer (Model D-300, Hobart Manufacturing Corporation, Troy, OH, USA). A warm (approximately 60 C) aqueous solution of sodium phosphate, potassium phosphate, choline chloride, and trace element premix, was then added to the dry ingredient mix, to bring the moisture content of the resulting mash to approximately 3435%. The mash was then blended for a further 15 min. Half the supplemental oil and lecithin and all the cholesterol were blended in a KitchenAid mixer (Model K5SS, KitchenAid, St Joseph, MI, USA), added to the mash

Table 1 Environmental conditions of culture containers in 8-week outdoor trial

Open cover Parameter Temperature (AM) Temperature (PM) Dissolved oxygen (mg L)1) pH Salinity (g L)1) Mean SD 28.2 31.3 6.0 7.8 33.9 1.0 1.3 0.5 0.4 1.1 Minimum^Maximum 25.3)30.4 27.1)34.0 4.7)8.2 6.7)9.2 31)38 Clear plastic cover Mean SD 29.6 32.6 5.8 7.7 34.0 1.0 1.2 0.5 0.4 1.1 Maximum^Minimum 25.6)31.5 29.7)35.1 4.6)7.1 7.2)8.8 31)36

..............................................................................................

2002 Blackwell Science Ltd Aquaculture Nutrition 8;121^137

Diet, culture and shrimp growth

Table 2 Formulation of the Oceanic Institute (OI) experimental sinking pelleted shrimp diet used in the 8-week feeding trial
Ingredient [crude protein (%)/crude lipid (%); ingredient cost US$ kg^1] Fishmeal ^ LT 94 (71.83/11.14; 1.28)1 Squid meal (58.94/4.19; 2.85)2 Soya bean meal, dehulled, solvent extracted (43.84/1.69; 0.16)3 Wheat, whole hard red winter (13.88/1.76; 0.12)4 Wheat gluten meal (72.97/1.06; 1.00)4 Brewers yeast (40.30/0.29; 0.62)5 Krill hydrolysate (59.38/10.45; 8.30)6 Soya lecithin, liquid (0.88)7 Marine fish oil, Menhaden (0.96)8 Cholesterol-FG (60%; potency 64%) (22.00)9 OI mineral premix LV99.1 (64.75)10 Potassium phosphate, dibasic (17.78% P, 44.9% K; 2.40)11 Calcium phosphate, monobasic (26.46% P, 17.12% Ca; 2.40)11 Sodium phosphate, dibasic (21.82% P, 32.39% Na; 2.40)11 OI vitamin premix ^ LV99.1 (47.74)12 Choline chloride (60%; 52% potency) (1.21)13 Vitamin C (35% ascorbic acid potency) (15.00)14
1 2

123

OI shrimp diet (g kg ^1 dry weight) 245.0 25.0 95.0 469.4 40.0 30.0 20.0 20.0 30.0 2.34 0.6 5.6 5.6

cabinet using an air blower at 38 C until the moisture level was below 10%. The vitamin premix and vitamin C source (Table 2) were then emulsied with the remaining oil and lecithin in a KitchenAid mixer and this mixture was added to the dry cooled pellets by top coating using a Hobart D300 food mixer with a whisk beater. The nished pellets were then stored in plastic bins at 1920 C until used.

Indoor protocols
Four diets were tested with four dierent feeding regimes and two pellet forms as follows: DFF NFF ADF OI shrimp diet; sinking pellet; xed ration; day feeding with feeders. OI shrimp diet; sinking pellet; xed ration; night feeding with feeders. OI shrimp diet; sinking pellet; xed ration; day and night feeding with feeders. OI shrimp diet; sinking pellet; fed to satiation; day feeding by hand. Commercial shrimp diet; sinking pellet; fed to satiation; day feeding by hand. Commercial catsh diet; oating pellet; fed to satiation; day feeding by hand. Commercial catsh diet; sinking crumble; fed to satiation; day feeding by hand.

5.6 4.0 1.154 0.714

DFS CSS CCFS CCCS

SSF Sildolje-og Sildemelindustriens Forskningsinstitut, Norway. Agribrands Purina Mexico, S.A. de C.V., Mexico (by courtesy of). 3 Land-o-Lakes, Seattle, WA, USA. 4 Hawaii Flour Mills, Honolulu, HI, USA. 5 Williams Bio-Products, Decatur, IL, USA (by courtesy of). 6 Specialty Marine Products,West Vancouver, BC, Canada (by courtesy of). 7 Central Soya Company Inc, Fort Wayne, IN, USA (by courtesy of). 8 Omega Protein Inc., Reedville,VA, USA. 9 Solvay Pharmaceuticals B.V.,Veenendaal,The Netherlands. 10 OI mineral premix LV99.1 ^ to supply the following elements (mg kg)1 diet): zinc (Zn, as sulphate) 72 mg, iron (Fe, as sulphate) 36 mg, manganese (Mn, as sulphate) 12 mg, copper (Cu, as sulphate) 24 mg, cobalt (Co, as chloride) 0.6 mg, iodine (I, as iodate) 1.2 mg, chromium (Cr, trivalent, as chloride) 0.8 mg, selenium (Se, as selenate) 0.2 mg, and molybdenum (Mo, as molybdate) 0.2 mg. 11 ICN Biomedicals, Inc., Aurora, OH, USA. 12 OI vitamin premix LV99.1 ^ to supply the following vitamins (mg or IU kg ^1 diet): thiamine 40 mg, riboavin 60 mg, pyridoxine 60 mg, pantothenic acid 180 mg, niacin 80 mg, biotin 0.6 mg, inositol 400 mg, folic acid 6 mg, cyanocobalamine 0.10 mg, vitamin A 6000 IU, vitamin D3 2000 IU, vitamin E 250 mg, vitamin K 40 mg, and astaxanthin 60 mg (premix prepared for OI by Roche V|tamins Inc, Parsippany, NJ, USA, by courtesy of). 13 Choline 60% ^ to supply 600 mg of active choline kg ^1 diet (Roche V|tamins Inc., Parsippany NJ; by courtesy of). 14 Stay-C 35% ^ to supply 250 mg of active vitamin C kg ^1 diet (Roche V|tamins Inc, Parsippany NJ; by courtesy of).

Day feedings. four times daily (08.00, 11.00, 14.00, 17.00 hours), Night feedings: four times nightly (20.00, 23.00, 02.00, 05.00 hours), All-day feedings: eight times during the day and night (at 08.00, 11.00, 14.00, 17.00, 20.00, 23.00, 02.00, and 05.00 hours) using battery operated Aquarium Fish Feeders (Fish Mate F14, Pet Mate Ltd, Hersham, Surrey, UK). The xed feeding ration employed was based on a shrimp daily feeding guide developed at the Oceanic Institute (Table 4). In the case of satiation feeding animals were fed to satiation four times daily (08.00, 11.00, 14.00, 17.00 hours); latex gloves were used for all feedings and handling of feed. All experimental animals were weighed individually at bi-weekly intervals for the duration of the experiment, and feeding rates adjusted weekly; animals blotted with an absorbent towel and weighed on a Mettler Toledo PB 3002 (Mettler-Toledo Inc., Hightstown, NJ, USA) electronic balance.

and mixed for a further 15 min. The resulting mash was then passed through a Hobart grinder tted with a 3-mm diameter die. The pellet temperature at the die was below 70 C. The resulting moist pellets were then dried overnight in a drying

Outdoor protocols
Four diets were tested with four dierent feeding regimes and two pellet forms as follows:

..............................................................................................

2002 Blackwell Science Ltd Aquaculture Nutrition 8;121^137

124

A.G.J. Tacon et al.

Table 3 Chemical composition (g kg^1 dry weight) of the test diets
Composition (g kg ^1 as feed basis except as noted) Proximate composition Moisture Crude protein (N 6.25) Crude lipid Cholesterol Ash Gross energy (MJ kg ^1) Amino acid composition Aspartic acid Serine Glutamic acid Glycine Alanine Taurine Cystine Tyrosine Isoleucine Leucine Methionine Phenylalanine Histidine Threonine Lysine Valine Arginine Tryptophan Mineral composition Phosphorus (P, g kg ^1) Potassium (K, g kg ^1) Calcium (Ca, g kg ^1) Magnesium (Mg, g kg ^1) Sodium (Na, g kg ^1) Manganese (Mn, mg kg ^1) Iron (Fe, mg kg ^1) Copper (Cu, mg kg ^1) Zinc (Zn, mg kg ^1) Boron (B, mg kg ^1) Feed cost5, US$ kg^1 Fatty acids (% total fatty acids) C12:0 C14:0 C14:1 C15:0 C16:0 C16:1n-7 C16:2n-4 C16:3n-4 C16:4n-1 C17:0 C18:0 C18:1n-9 C18:1n-7 C18:1n-5 C18:2n-6 C18:2n-4 C18:3c C18:3n-4 Diets1 DFF^DFS CSS CCFS CCCS

61.1 351.7 82.4 2.7 62.8 19.00 29.74 15.44 69.94 17.40 18.53 1.96 4.73 12.15 (A /E)2 14.70 (86) 31.03 (183) 8.64 (79)3 14.82 (159)4 7.82 (46) 13.41 (79) 20.51 (121) 17.25 (101) 21.17 (125) 3.70 (22) 4.912 8.09 5.733 1.635 4.103 32.88 135.9 27.35 104 1.164 1.04 0.1 5.1 0.2 0.4 19.2 5.3 1 0.9 0.1 0.3 2.4 15.0 nd 0.2 21.3 0.1 0.1 0.3

64.6 349.5 91.1 1.46 102.8 18.28 34.96 17.62 52.49 20.32 20.83 1.12 5.79 11.20 (A /E) 14.63 (85) 31.81 (186) 8.28 (82) 14.98 (153) 7.33 (43) 14.29 (83) 19.95 (116) 18.22 (106) 21.39 (125) 3.34 (19) 4.071 4.986 19.84 2.385 2.292 117.1 397.3 38.65 138.5 9.252 0.61 0.2 8.0 0.1 0.5 20.5 9.3 1.4 1.4 nd6 0.4 3 16.4 0.1 nd 16.1 0.2 0.2 0.4

47.2 374.7 54.3 0.97 103.4 18.45 37.07 17.65 59.97 29.36 24.01 0.98 4.73 11.51 (A /E) 12.72 (69) 34.18 (185) 7.83 (68) 17.07 (155) 10.45 (57) 14.72 (80) 23.78 (129) 20.45 (111) 23.39 (127) 3.78 (20) 6.121 7.186 25.23 2.182 2.544 116.8 430.7 8.789 122.8 9.409 0.43 0.1 4.6 0.2 0.4 20.3 6.1 1 0.9 nd 0.4 4.3 24 nd nd 14.2 nd 0.2 0.2

46.8 376.0 49.6 0.97 104.9 18.29 37.07 17.65 59.97 29.36 24.01 0.98 4.73 11.51 (A /E) 12.72 (69) 34.18 (185) 7.83 (68) 17.07 (155) 10.45 (57) 14.72 (80) 23.78 (129) 20.45 (111) 23.39 (127) 3.78 (20) 6.121 7.186 25.23 2.182 2.544 116.8 430.7 8.789 122.8 9.409 0.43 0.1 4.6 0.2 0.4 20.3 6.1 1 0.9 nd 0.4 4.3 24 nd nd 14.2 nd 0.2 0.2

..............................................................................................

2002 Blackwell Science Ltd Aquaculture Nutrition 8;121^137

Diet, culture and shrimp growth

Table 3 (continued)
Composition (g kg ^1 as feed basis except as noted) C18:3 n-3 C18:4 n-3 C18:4 n-1 C20:0 C20:1n-9 C20:1n-7 C20:2 n-6 C20:3 n-6 C20:3 n-3 C20:4 n-6 C20:4 n-3 C20:5 n-3 C21:5 n-3 C22:1n-11 C22:4 n-6 C22:5 n-3 C22:6 n-3 C23:0 C24:1 n-9 Total n-6 Total n-3 Unknown peaks
1

125

Diets1 DFF^DFS 2.5 1.6 0.1 0.1 2.7 0.1 0.1 0.2 0.1 0.5 0.6 6.5 0.1 2.2 0.2 1.1 7.6 0.3 0.2 22.3 20.1 1.3 CSS 1.7 1.1 0.2 0.1 0.7 0.1 0.1 0.2 0.1 0.9 0.6 8.0 0.1 0.1 0.2 1.4 3.8 0.5 nd 17.5 16.8 2.1 CCFS 1.6 1.1 nd nd 1.6 nd 0.2 0.3 nd 0.6 0.9 6.8 0.5 1.1 0.6 1.3 5.3 0.1 nd 15.9 17.5 1.0 CCCS 1.6 1.1 nd nd 1.6 nd 0.2 0.3 nd 0.6 0.9 6.8 0.5 1.1 0.6 1.3 5.3 0.1 nd 15.9 17.5 1.0

DFF^DFS = OI shrimp feed; CSS = commercial shrimp feed; CCFS = commercial catsh feed (pelleted-oating form); CCCS = commercial catsh feed (crumbled-sinking form). 2 A /E ratio is dened as [(each essential amino acid content/total essential amino acid content including cystine and tyrosine) 1000]. 3 Methionine + cystine. 4 Phenylalanine + tyrosine. 5 Feed costs: OI feed ^ US$ 1.04 kg ^1 diet ingredient costs only, excludes manufacturing costs; all other costs are from Rangen price list, F.O.B. Buhl, Idaho; FTL: full truck load quantities (5 July 1999). 6 nd = Not detected or value lower than 0.05%.

Table 4 Feeding rates for shrimp in 8-week feeding trials

Percentage of estimated shrimp biomass Shrimp body weight (g) 1^3 3^5 5^7 7^9 9^11 11^13 13^15 15^17 17^30

DFFP

21^24 C 8 7 6.5 6 5.5 5 4.5 4 3

24^28 C 6 5 4.5 4 3.5 3 2.5 2.5 2

28^32 C 7 6 5.5 5 4.5 4 3.5 3 2.5

OI shrimp diet; sinking pellet; xed ration; day feeding by hand; plastic tank cover. CCFF Commercial catsh diet; oating pellet; xed ration; day feeding by hand. CCCF Commercial catsh diet; sinking crumble; xed ration; day feeding by hand. CSF Commercial shrimp diet; sinking pellet; xed ration; day feeding by hand. Day feedings. four times daily (08.00, 11.00, 14.00, 17.00 hours); Night feedings: four times nightly (20.00, 23.00, 02.00, 05.00 hours); All-day feedings: eight times during the day and night (at 08.00, 11.00, 14.00, 17.00, 20.00, 23.00, 02.00, and 05.00 hours) fed manually by hand application. The xed feeding ration employed was based on the shrimp daily feeding guide described above (Table 4); allotted daily feed allocations were equally divided into four or eight portions per day as required. Sinking feed was applied using a feeding tube directly onto a 0.12-m2 submersible feeding tray placed 0.6 m below the water surface on one side of the tank. During the last 2 weeks of

DFF NFF ADF

OI shrimp diet; sinking pellet; xed ration; day feeding by hand. OI shrimp diet; sinking pellet; xed ration; night feeding by hand. OI shrimp diet; sinking pellet; xed ration; day and night feeding by hand.

..............................................................................................

2002 Blackwell Science Ltd Aquaculture Nutrition 8;121^137

126

A.G.J. Tacon et al.

the trial the feeding rates were reduced from the highest to the lowest temperature range feeding regime (Table 4) to avert the possible crash of the microbial community within the experimental tanks as a result of the high biomass loading brought on by the exceptional growth rates observed in some of the treatments. All experimental animals were weighed individually at the start and end of the 56-day feeding trial. At least 10% of the estimated remaining population of each tank was sampled bi-weekly using a net or minnow trap and this data was used to adjust feeding rates weekly. present within the water column of the outdoor microcosm tanks were collected at the end of the 8-week feeding trial. A stirred water sample (8 L) was collected from each tank and vacuum ltered through Whatman No. 1 hardened lter paper using a 20-cm Buchner funnel, and the ltrate then freeze-dried to constant weight using a Freezemobile 12 freeze-drier (Virtis Inc., Gardiner, NY, USA). Chemical analyses, including moisture, total nitrogen, crude lipid, and ash, were conducted in duplicate as described previously (Divakaran 1999). The gross caloric content of experimental diets and SPM were determined using a Parr 1261 Isoperibol Bomb Calorimeter (Parr Instrument Co, St Moline, IL, USA) using benzoic acid as the standard. Mineral analysis of diets, shrimp tissue, and SPM was undertaken by Inductively Coupled Plasma Atomic Emission Spectroscopy using a Model Atomscan 16 radial conguration instrument (Thermo Jarrel Ash, TJA Solutions, Franklin, MA, USA), after rst ashing the samples at 600 C for 6 h and then dissolving the ash in 3 N HCl prior to analysis (AOAC 1990a). Amino acids in freeze-dried SPM and shrimp tissue were analysed using a Beckman System 6300 Amino Acid Analyzer following hydrolysis in 6 N HCl for 20 h at 115 C (using norleucine as an internal standard) following the method of Hamilton (1963). For cystine/2 analysis, samples were oxidized at 50 C for 15 min with performic acid prior to hydrolysis, following the method of Hirs (1967). For tryptophan analysis, samples were hydrolysed in 4.2 N NaOH at 135 C for 48 h prior to neutralization and analysis (Hugli & Moore 1972). The crude lipid content of freeze-dried SPM was determined using the method of Hara & Radin (1978) with the following modications: samples were homogenized with a solution of 0.01 M MgCl2 and extracted with a chloroform:isopropanol 2:1 (v/v) mixture and 1 M HCl. The homogenate was then rinsed with the solvent mixture and centrifuged to recover the organic layer. The organic layer was then washed with 0.3 M HCl, and lipid was recovered by drying over a stream of nitrogen. Fatty acid analysis of experimental diets and freeze-dried SPM samples was undertaken by a modied direct methylation method (AOAC 1990b) using a Hewlett Packard 5890 Gas Chromatograph (Hewlett-Packard Co., Palo Alto, CA, USA) with a Flame Ionization Detector.

Chemical analyses
Water quality. Routine water quality testing was performed during each feeding trial. In the indoor trial, water temperature was measured daily (at about 08.00 hours) within all experimental tanks using a handheld mercury thermometer. All other water quality parameters were measured on a weekly basis, and included pH (using a Model 1001 Sentron pH meter, Sentron Inc., Gig Harbor, WA, USA), dissolved oxygen (using a Model 55 Yellow Springs Instrument oxygen meter), salinity (using a temperature compensated refractometer, Aquatic Eco-Systems Inc., Apopka, FL, USA), and total ammonia nitrogen (TAN) determined by the automated analysis method of Solorzano (1969) using a Technicon Auto-Analyzer II. In the outdoor feeding trial, water temperature and dissolved oxygen were measured twice daily (at about 08.00 and 16.00 hours), and pH, salinity and TAN twice weekly (Monday and Thursday at 13.00 hours), as described above. In addition, nutrient analyses were performed twice weekly, including chlorophyll a [following the method of Strickland & Parsons (1972) using a Turner Designs Fluorometer], total phosphorus and orthophosphate (total phosphorus Grassho et al. 1983; orthophosphate Murphy & Riley 1962), total nitrogen (DElia et al. 1977), and nitrate and nitrite using a Technicon Auto-Analyzer II (NitrateNitrite in water and seawater; Industrial method no. 158-71 W, December 1972; Technicon Industry Systems, Tarrytown, NY, USA). Diets, shrimp tissue and suspended particulate matter. Shrimp were collected at the start of the experiments (from a representative population sample) and from each tank at the end of the feeding trials (10 shrimp per indoor and outdoor tank) and frozen for subsequent analysis on an individual tank basis. In the case of the large animals harvested at the end of the feeding trials, shrimp were macerated, freeze-dried to a constant weight, and then ground prior to chemical analysis. Samples of the suspended particulate matter (SPM)

Calculations and statistics

Shrimp growth was measured by mean weight gain, weekly weight gain, and specic growth rate. Feed eciency was calculated as the mean weight gain divided by the amount of diet fed. Nitrogen and phosphorus eciency were calculated

..............................................................................................

2002 Blackwell Science Ltd Aquaculture Nutrition 8;121^137

Diet, culture and shrimp growth

in the outdoor trial as the accumulation of these elements in the shrimp whole body as a proportion of the total amount presented in the diets over the course of the trial. In the outdoor system, the eciency parameters (feed, nitrogen and phosphorus) are referred to as apparent eciency and are of more practical than biological signicance, because actual consumption of the diets could not be monitored, nor could the impact of cannibalism and consumption of natural food production be directly assessed. Data obtained from the experiments, which had a completely randomized design with three replicates per treatment, were analysed by one-way analysis of variance to determine if signicant dierences existed among treatment means (Snedecor & Cochran 1967). Arcsin transformation [sin1(x0.5)] was applied to the data prior to analysis. Tukeys test for mean separation was used to evaluate dierences among treatment means. All statistical analyses were performed in SigmaStat version 2.03 (SPSS Inc., Chicago, IL, USA 1997). Dierences were considered signicant at the 5% level of probability. animals exhibited a nal body weight 13.3 and 12.2% higher than those animals fed during the day-light and night-hours, respectively, and 24-h fed animals displayed a nal body weight 16.1% higher than animals fed with the same diet fed to satiation four times daily (Table 5). Moreover, there was no signicant dierence in the growth of shrimp fed during daylight hours or during the nighthours. However, survival and feed eciency were lowest among shrimp in the treatments fed during night-hours. Although the overall growth response and nal body weight of shrimp fed with the control OI shrimp feed (DFS) were higher than those of animals fed with the commercial shrimp diet (CSS), these dierences were not signicant. The poorer growth response and performance of the commercial shrimp diet (CSS) corresponded to the lower mean voluntary feed intake of animals fed with this ration compared with the OI shrimp diet (DFS). Animals rapidly (within a few minutes) learned to swim to the water surface to consume the oating pelleted catsh feed (CCFS), and grew as well as animals that were with fed the same diet in crumbled-sinking form.

127

Results
Indoor feeding trial
Shrimp growth (expressed in terms of nal body weight and weekly growth) was highest in those treatments which fed eight times daily over a 24-h feeding period; 24-h fed

Outdoor feeding trial

As with the indoor feeding trial, shrimp growth (expressed in terms of nal body weight and weekly growth) was highest in those treatments fed eight times daily over a 24-h feeding

Table 5 Shrimp growth and feed performance in an indoor ow-through culture system over the 8-week experimental period. Values within a row sharing a common superscript are not signicantly dierent (Tukeys test; P < 0.05; n = 3)
Dietary treatment1 Shrimp weight Mean initial body weight (g) Mean final body weight (g) Shrimp feed intake Mean feed intake (g shrimp ^1 day ^1) Shrimp growth response Total weight gain (%) Mean weekly weight gain (g week ^1) Specific growth rate (% day ^1)3 Shrimp feed utilization Feed efficiency (%)4 Total shrimp production Shrimp survival (%)
1

DFF 1.61a 5.92ab 0.21 268.4ab 0.54ab 2.33ab 35.8ab 93.1a

NFF 1.61a 5.98ab 0.24 272.6ab 0.55ab 2.34ab 28.9bc 77.8a

ADF 1.59a 6.71a 0.22 323.8a 0.64a 2.57a 40.4a 93.1a

DFS 1.59a 5.78b 0.17 264.5ab 0.52ab 2.31ab 42.4a 86.1a

CSS 1.60a 5.17b 0.14 223.7b 0.45b 2.10b 45.6a 92.7a

CCFS 1.57a 3.66c 0.12 133.4c 0.26c 1.51c 26.1bc 79.2a

CCCS 1.60a 3.67c 0.12 130.2c 0.26c 1.48c 21.8c 72.2a

SEM2

0.02 0.18

13.0 0.02 0.07 0.13 4.5

DFF = OI shrimp diet; sinking pellet; xed ration; day feeding with feeders; NFF = OI shrimp diet; sinking pellet; xed ration; night feeding with feeders; ADF = OI shrimp diet; sinking pellet; xed ration; day and night feeding with feeders; DFS = OI shrimp diet; sinking pellet; fed to satiation; day feeding by hand; CSS = commercial shrimp diet; sinking pellet; fed to satiation; day feeding by hand; CCFS = commercial catsh diet; oating pellet; fed to satiation; day feeding by hand; CCCS = commercial catsh diet; sinking crumble; fed to satiation; day feeding by hand. 2 Standard error of means. 3 Specic growth rate = [logenal body weight (g) ) logeinitial body weight (g)]/time (days) 100. 4 Feed eciency = [nal shrimp biomass (g) ) initial shrimp biomass (g)] 100/total feed oered (g, as fed basis).

..............................................................................................

2002 Blackwell Science Ltd Aquaculture Nutrition 8;121^137

128

A.G.J. Tacon et al.

Table 6 Shrimp growth and feed performance in an outdoor zero-water-exchange culture system over the 8-week experimental period. Values in a row sharing a common superscript are not signicantly dierent (Tukeys test; P < 0.05; n = 3)
Dietary treatment1 DFF NFF 1.56a 17.34a 0.44 1015.0a 1.97a 4.30a 63.1a 35.05a 46.56a 0.10a 0.92a 0.82a 1.94 79.3a 1.67 ADF 1.58a 18.89a 0.53 1098.2a 2.16a 4.43a 57.7a 32.01a 39.42ab 0.11a 0.89ab 0.78ab 2.03 69.7a 1.93 DFFP 1.59a 16.78a 0.54 956.6ab 1.90a 4.20ab 47.1ab 26.61ab 31.19abc 0.11a 0.68abc 0.57abc 1.82 61.0a 2.20 CCFF 1.60a 11.35b 1.76 609.5c 1.22b 3.50c 3.5c 1.79c 2.06d 0.11a 0.14 0.04 1.58 19.0b 46.74 CCCF 1.60a 10.46b 0.54 554.7c 1.11b 3.35c 25.6bc 13.01bc 12.05cd 0.11a 0.39c 0.29c 1.67 56.7a 1.71 CSF 1.57a 12.94b 0.55 728.0bc 1.42b 3.76bc 34.1ab 18.99ab 23.93bc 0.10a 0.50bc 0.40bc 1.76 58.3ab 1.87 52.1 0.09 0.10 6.17 3.46 4.14 0.01 0.08 0.08 6.2 SEM2

Shrimp weight Mean initial body weight (g) 1.60a Mean final body weight (g) 17.20a Shrimp feed intake Mean feed offered (g shrimp)1 day)1) 0.56 Shrimp growth response Total weight gain (%) 977.4ab Mean weekly weight gain (g week)1) 1.95a Specific growth rate (% day)1) 4.24a Shrimp feed utilization Apparent feed efficiency (%)3 49.7ab Apparent feed nitrogen efficiency (%)4 28.45ab Apparent feed phosphorus efficiency (%)5 33.44ab Total shrimp production Total initial shrimp biomass (kg m)3) 0.11a Total final shrimp biomass (kg m)3) 0.75abc Total biomass increase (kg m)3) 0.64abc Total feed offered (kg) (as fed basis) 1.93 Shrimp survival (%) 64.7a Feed cost kg shrimp production ($)
1 )1

0.03 0.68

2.29

DFF = OI shrimp diet; sinking pellet; xed ration; day feeding by hand; NFF = OI shrimp diet; sinking pellet; xed ration; night feeding by hand; ADF = OI shrimp diet; sinking pellet; xed ration; day and night feeding by hand; DFFP = OI shrimp diet; sinking pellet; xed ration; day hand feeding; plastic tank cover; CCFF = commercial catsh diet; oating pellet; xed ration; day feeding by hand; CCCF = commercial catsh diet; sinking crumble; xed ration; day feeding by hand; CSF = commercial shrimp diet; sinking pellet; xed ration; day feeding by hand. 2 Standard error of means. 3 Apparent feed eciency = [nal shrimp biomass (g) ^ initial shrimp biomass (g)] 100/total feed oered (g, as feed basis); value excludes the consumption of natural food organisms present within the culture system. 4 Apparent feed nitrogen eciency = whole body nitrogen gain 100/shrimp feed nitrogen oered. 5 Apparent feed phosphorus eciency = whole body phosphorus gain 100/shrimp feed phosphorus oered.

period (Table 6; Figs 1 & 2). Animals fed over a 24-h period exhibited a nal body weight 9.8 and 8.9% higher than animals fed during the daylight and night-hours, respectively, but these dierences were not statistically signicant. Moreover, there were no signicant dierences in the growth, feed eciency or survival of shrimp fed during daylight hours or during the night-hours. Although shrimp growth and body weight were satisfactory in those tanks with plastic covers and signicantly higher than the commercial control, water temperatures were high, ranging from a low of 25.6 C (AM) to a high of 35.1 C (PM) during the experiment mean 29.632.6 C, and were clearly in excess of or close to the reported lethal temperatures for shrimp. In contrast, tanks without covers displayed a range of 25.3 C (AM) to a high of 34.0 C (PM) mean 28.231.3 C. Despite this, dissolved oxygen concentrations in the covered tanks were generally satisfactory, ranging from 4.5 to 7.1 mg L1. However, the use of plastic covered tanks during the colder winter months may be more benecial (the current trial was conducted during the hotter summer months).

Shrimp fed with the OI shrimp diet had signicantly higher nal body weight and growth rate than shrimp fed with the commercial control diet (3046% higher), although both diet series had similar proximate composition (35% crude protein, 9% lipid) and both diets contained 2.5% squid meal. Calculated feed costs per kg of shrimp production ranged from $1.87 for the commercial shrimp feed, to $1.67 for the night-time OI feed, to $1.93 for the all-day OI feed (shrimp within this treatment were 46% larger by weight than the commercial shrimp feed; Table 6). The growth, feed eciency and survival of shrimp fed with the oating catsh feed were very low, primarily because the animals were not aware that feed was being administered (the feed eventually sank), and growth was equivalent to the crumbled (sinking) version of the same diet. Interestingly, there were no signicant dierences between the growth of the shrimp in the three commercial feed treatments (Table 6). Dietary mineral concentration had little eect on the nal tissue mineral concentrations of the experimental shrimp (Table 8), with tissue iron and copper concentrations actually decreasing over the 8-week period. This was particularly

..............................................................................................

2002 Blackwell Science Ltd Aquaculture Nutrition 8;121^137

Diet, culture and shrimp growth

7.4 mg L1 (range 69 mg L1). Nitrite values in the system were below 0.1 mg L1 until day 43. From day 43 onwards, nitrite increased concurrently with the decrease in TAN, reaching a level of 20 mg L1 by the end of the experiment (Fig. 5). Total nitrogen and total phosphorus accumulated steadily within the experimental tanks over the course of the study, from 2 to 30 mg L1 and 0.2 to 16 mg L1, respectively. By contrast, the pH of the system decreased progressively, from a high of 8.4 near the beginning of the experiment to a low of 7.1 by the end of the study (Fig. 5). Within all the microcosm tanks there was a rapid development of a microbial food chain, initially in the visible form of a green algal-based autotrophic microbial food web, with a bacterial-based heterotrophic microbial food web developing later; the latter was visible in the water column as suspended particulate matter or microbial oc (oc). Phytoplankton biomass, as indicated by chlorophyll a, peaked between day 25 and 32 at an average of 350 lg L1 (range 200580 lg L1). After the peak, chlorophyll a values declined slightly, averaging between 250 and 300 lg L1 until the end of the experiment (Fig. 5). Chemical analysis of the oc taken at the end of the trial revealed a valuable potential food source for the resident shrimp. As shown in Tables 9 and 10, oc was a valuable dietary source of amino acids, fatty acids, and minerals. Each tank was calculated to contain 87200 mg L1 of oc by the end of the experiment, or the equivalent of 122280 g of dry oc per 1.40 m3 of water (mean 201 g; Table 9).

129

24 22 20 18
DFF NFF ADF DFFP CCFF CCCF CSF

Shrimp weight (g)

16 14 12 10 8 6 4 2 0 0 2 4 Week 6 8

Figure 1 Growth response of outdoor shrimp fed with the experimental diets for 8 weeks (mean weight SD; n 3). DFF OI shrimp diet; sinking pellet; xed ration; day feeding by hand; NFF OI shrimp diet; sinking pellet; xed ration; night feeding by hand; ADF OI shrimp diet; sinking pellet; xed ration; day and night feeding by hand; DFFP OI shrimp diet; sinking pellet; xed ration; day hand feeding; plastic tank cover; CCFF Commercial catsh diet; oating pellet; xed ration; day feeding by hand; CCCF Commercial catsh diet; sinking crumble; xed ration; day feeding by hand; CSF Commercial shrimp diet; sinking pellet; xed ration; day feeding by hand.

unexpected, considering the initial dierences between dietary treatments in terms of the mineral composition of the diets fed (the OI diet had reduced levels of calcium, magnesium, iron and manganese compared with the commercial shrimp and sh rations tested; Table 3). However, shrimp fed exclusively during night-hours had signicantly lower carcass moisture content and elevated carcass zinc and phosphorus content than those fed with the commercially prepared feeds (Tables 7 and 8). Despite the high feed intake (Fig. 3) and growth (Fig. 4) of shrimp during the rst 6 weeks of the experiment, there was a progressive reduction in the growth response of shrimp during the nal 2 weeks of the experiment. This correlated with the progressively deteriorating water quality (elevated ammonia and nitrite and reduced pH) within all experimental tanks (Fig. 5) and the consequent need to reduce the daily feeding rates of all treatments from the higher 2832 C feeding rate range to the lower 2124 C feeding rate range (Table 4) to avoid tank or system crashes. Mean levels of TAN were initially low, peaking on day 42 at an average of

Discussion
The observation that in the indoor trial, the shrimp fed during the day grew as well as, and had better feed eciency and survival than, those fed at night, is in agreement with those of Robertson et al. (1993), who found that day feeding was as good as, or slightly better than, night feeding in terms of shrimp (L. vannamei) growth. In the outdoor trial, the higher feed eciency and survival among the shrimp fed during the night compared to day time feeding is in agreement with the ndings of Nunes et al. (1996), who found no signicant dierence between diurnal and nocturnal food consumption patterns of Farfantepenaeus subtilis, with animals displaying continuous feeding activity during day and night. However, Velasco et al. (1999) reported no benecial eect of increasing feeding frequency or ration size on the growth or survival of shrimp (L. vannamei) post larvae (185 mg body weight) fed with a diet containing 19.5% crude protein within an experimental zero-water-exchange culture system.

..............................................................................................

2002 Blackwell Science Ltd Aquaculture Nutrition 8;121^137

130

A.G.J. Tacon et al.

Figure 2 Histogram of mean shrimp body weight (g/shrimp) at the end of the outdoor feeding trial. DFF OI shrimp diet; sinking pellet; xed ration; day feeding by hand; NFF OI shrimp diet; sinking pellet; xed ration; night feeding by hand; ADF OI shrimp diet; sinking pellet; xed ration; day and night feeding by hand; DFFP OI shrimp diet; sinking pellet; xed ration; day hand feeding; plastic tank cover; CCFF Commercial catsh diet; oating pellet; xed ration; day feeding by hand; CCCF Commercial catsh diet; sinking crumble; xed ration; day feeding by hand; CSF Commercial shrimp diet; sinking pellet; xed ration; day feeding by hand.

Shrimp fed with the oating pellets indoors had growth similar to that of shrimp fed with the sinking crumbled feed. In this trial, a feed designed for catsh was used. If a oating feed specically formulated for shrimp was prepared for use in clear water conditions (where the shrimp can readily sense the presence of feed particles), perhaps growth rates would have markedly improved. Apart from the obvious nutritional benets of improved carbohydrate digestibility and water stability, the use of a oating shrimp feed within these

intensive clear water systems would allow the culturist to more accurately judge the correct amount of feed by observing the animals feeding at rst hand, instead of relying on feeding tables. The best overall shrimp growth performance was observed in animals fed with the OI shrimp diet and all-day feeding regime under outdoor zero-water exchange culture conditions; nal body weight and average weekly growth rate were 2.8 and 3.4 times greater, respectively, than animals of similar

..............................................................................................

2002 Blackwell Science Ltd Aquaculture Nutrition 8;121^137

Diet, culture and shrimp growth

Table 7 Proximate composition of shrimp carcass (whole body) at the beginning and end of the outdoor, zero-water-exchange 8-week experiment. Crude protein, crude lipid, ash, and NFE are reported on a shrimp live weight basis. Values within a column that share a common superscript are not signicantly dierent (Tukeys test; P < 0.05; n = 3 except initial n = 1)
Treatment1 Initial DFF NFF ADF DFFP CCFF CCCF CSF SEM3 Moisture (%) 75.88 74.54ab 73.89b 74.73ab 74.58ab 76.06a 75.29ab 76.65a 0.447 Crude protein (%) 17.41 19.82a 19.19a 19.41a 19.48a 18.49a 19.00a 17.86a 0.406 Crude lipid (%) 1.86 1.63a 1.49a 1.41a 1.67a 1.51a 1.53a 1.47a 0.116 Ash (%) 2.64 2.62a 2.88a 2.69a 2.61a 2.50a 2.64a 2.51a 0.112 NFE2 (%) 2.22 1.39 2.61 1.76 1.66 1.44 1.54 1.51

131

1 DFF = OI shrimp diet; sinking pellet; xed ration; day feeding by hand; NFF = OI shrimp diet; sinking pellet; xed ration; night feeding by hand; ADF = OI shrimp diet; sinking pellet; xed ration; day and night feeding by hand; DFFP = OI shrimp diet; sinking pellet; xed ration; day hand feeding; plastic tank cover; CCFF = commercial catsh diet; oating pellet; xed ration; day feeding by hand; CCCF = commercial catsh diet; sinking crumble; xed ration; day feeding by hand; CSF = commercial shrimp diet; sinking pellet; xed ration; day feeding by hand. 2 Nitrogen free extract calculated by dierence (100% ^ all other components). Includes carbohydrate and chitin. 3 Standard error of the means.

Table 8 Mineral composition of shrimp carcass (whole body) at the beginning and end of the 8-week experiment. Values within a column that share a common superscript are not signicantly dierent (Tukeys test; P < 0.05; n = 3 except initial n = 1)
Treatment1 Initial DFF NFF ADF DFFP CCFF CCCF CSF SEM2 P (g kg ^1) 2.50 3.20ab 3.50a 3.24ab 3.13bc 2.79bc 2.78bc 2.78bc 0.0787 K (g kg ^1) 2.16 2.39a 2.43a 2.31a 2.49a 2.43a 2.40a 2.58a 0.141 Ca (g kg ^1) 6.40 6.11a 7.05a 6.66a 6.04a 5.92a 6.55a 5.81a 0.60 Mg (g kg ^1) 0.73 0.75ab 0.89a 0.76ab 0.80ab 0.72ab 0.73ab 0.70b 0.040 Na (g kg ^1) 1.72 1.65a 1.94a 1.74a 1.72a 1.67a 1.59a 1.76a 0.13 Mn (mg kg ^1) 1.031 0.797a 0.906a 0.916a 0.815a 0.979a 1.259a 1.160a 0.0148 Fe (mg kg ^1) 107.302 15.687a 16.247a 22.445a 15.916a 15.004a 23.176a 21.652a 6.138 Cu (mg kg ^1) 28.485 15.091b 18.300ab 19.888ab 17.263ab 16.426ab 24.776a 11.641b 1.892 Zn (mg kg ^1) 16.715 18.358bc 21.522a 19.093ab 18.161bc 17.724bc 17.776bc 15.837c 0.6162 B (mg kg ^1) 2.116 1.933a 2.218a 1.870a 1.649a 2.087a 1.787a 1.819a 0.1433

1 DFF = OI shrimp diet; sinking pellet; xed ration; day feeding by hand; NFF = OI shrimp diet; sinking pellet; xed ration; night feeding by hand; ADF = OI shrimp diet; sinking pellet; xed ration; day and night feeding by hand; DFFP = OI shrimp diet; sinking pellet; xed ration; day hand feeding; plastic tank cover; CCFF = commercial catsh diet; oating pellet; xed ration; day feeding by hand; CCCF = commercial catsh diet; sinking crumble; xed ration; day feeding by hand; CSF = commercial shrimp diet; sinking pellet; xed ration; day feeding by hand. 2 Standard error of the means.

initial size fed with the same diet under indoor running-water culture conditions. Although direct comparison between experiments is not possible because of the lower indoor water temperatures (2627 C compared with 2831 C) and lower mean daily feed intake of animals (0.22 g shrimp1 compared with 0.53 g shrimp1), it is believed that the higher growth (as much as three times higher) and performance of animals reared under outdoor green-water culture conditions is due to their culture environment and ability to obtain additional nutrients from natural food organisms present within the water column and/or pond ecosystem (Leber & Pruder 1988; Moss 1995; Tacon 1996; Moriarty 1997). It is important to highlight here that outdoor zero-water exchange culture systems are completely closed farming

systems, with no water exchange for the duration of the culture cycle other than that added to the system to make up for evaporative losses. Moreover, shrimp growth is achieved through the simultaneous consumption of both exogenously supplied compound aquafeeds (the OI shrimp diet in this case), and endogenously produced living microbial feeds or microbial oc (oc), which is a complex mixture of microorganisms and invertebrates. For example, the biological diversity of the microbial food web within the microcosm tanks is evidenced by the presence of not only bacteria and algae (including diatoms), but also agellates, ciliates, amoebae, rotifers, nematodes, and gastrotrichs. It is interesting to note the similarity between the organisms associated with the macro-aggregates or ocs from the microcosm

..............................................................................................

2002 Blackwell Science Ltd Aquaculture Nutrition 8;121^137

132

A.G.J. Tacon et al.

Figure 3 Mean daily feed application in the outdoor trial. DFF OI shrimp diet; sinking pellet; xed ration; day feeding by hand; NFF OI shrimp diet; sinking pellet; xed ration; night feeding by hand; ADF OI shrimp diet; sinking pellet; xed ration; day and night feeding by hand; DFFP OI shrimp diet; sinking pellet; xed ration; day hand feeding; plastic tank cover; CCFF Commercial catsh diet; oating pellet; xed ration; day feeding by hand; CCCF Commercial catsh diet; sinking crumble; xed ration; day feeding by hand; CSF Commercial shrimp diet; sinking pellet; xed ration; day feeding by hand.

3.5
Week 0 2 Week 2 4 Week 6 8

3.0 2.5 Weekly growth (g) 2.0 1.5 1.0 0.5 0.0 DFF NFF ADF DFFP Treatment

Week 4 6

CCFF

CCCF

CSF

Figure 4 Mean weekly shrimp growth (g week1) in the outdoor trial. DFF OI shrimp diet; sinking pellet; xed ration; day feeding by hand; NFF OI shrimp diet; sinking pellet; xed ration; night feeding by hand; ADF OI shrimp diet; sinking pellet; xed ration; day and night feeding by hand; DFFP OI shrimp diet; sinking pellet; xed ration; day hand feeding; plastic tank cover; CCFF Commercial catsh diet; oating pellet; xed ration; day feeding by hand; CCCF ComCommercial catsh diet; sinking crumble; xed ration; day feeding by hand; CSF Commercial shrimp diet; sinking pellet; xed ration; day feeding by hand.

tanks and those normally encountered on marine and lake snow, and on oc from activated sludge (Curds 1992). The fundamental dierence between this culture system and the traditional open or running-water pond-based shrimp culture system is that the culture target is changed from a single-stomached animal (the shrimp), where micro-organisms generally play a limited (although important) role in digestion and nutrient supply, to the equivalent of a multistomached animal through the provision of an in situ microbial aerobic digester or bioreactor (the microcosm), where micro-organisms play a major role in digestion and nutrient supply, as they do in ruminants (Tacon et al. 1999).

Indeed, recent studies with shrimp (L. vannamei) within zerowater exchange culture systems have shown the nonessentiality of dietary vitamin and trace mineral supplementation within exogenously supplied compound aquafeeds (A.G.J. Tacon, unpublished data; Velasco & Lawrence 2000) and the ability of totally replacing shmeal in prepared feeds with rendered terrestrial animal by-product meals with little or no loss in growth and feed eciency (Tacon 2000). Of course, this oc also has important functions in removing and harnessing potentially toxic faecal wastes and metabolites (e.g. by nitrication) from the shrimp within the culture system.

..............................................................................................

2002 Blackwell Science Ltd Aquaculture Nutrition 8;121^137

Diet, culture and shrimp growth

133

Figure 5 Changes in water quality within the outdoor microcosm tanks over the course of the 56-day experimental test period. TN total nitrogen; Chl a Chlorophyll a; NO2 nitrite; TP total phosphorus; TAN total ammonia nitrogen.

It is also important to mention here the rapid growth rates observed for shrimp reared within the outdoor zero-water exchange culture systems, and in particular for those animals fed with the OI shrimp diet and all-day feeding regime; shrimp displayed an average weekly growth rate of 2.16 g week1, increasing from 1.58 to 18.89 g (market size) in only 8 weeks (shrimp stocking density 71 m3 water volume, Table 6). This growth rate was more than twice that reported for shrimp (L. vannamei) reared under commercial conditions within intensive zero-water exchange pond-based culture systems [0.81.0 g week1 at 112128 animals m2 or equivalent to 6069 animals m3 water volume (McIntosh & Carpenter 1999)]. Growth rates as high as 2.7 g week1 have been reported for juvenile shrimp (L. vannamei) reared within outdoor microcosm tanks and fed with a high-quality shrimp diet (containing 52% crude protein) over an 18-day experimental test period (Freeman & Duerr 1991). It is interesting to note that during the present outdoor feeding trial, average weekly growth rates peaked at 3.2 g week1 in one treatment (Fig. 4). As stated above, an important factor contributing to the very high growth rates of shrimp within these zero-exchange culture systems was likely the endogenous production and availability of microbial food organisms (oc) for the resident shrimp. Not surprisingly, nutritional analysis of the oc collected from the experimental tanks at the end of the feeding trial revealed a composition and nutrient prole comparable with that of similar ocs harvested from domestic waste water activated sludge treatment facilities (Tacon & Ferns 1978/1979; Tacon 1978/1979). In Tahiti, experiments were reportedly run utilizing domestic activated sludge as an inoculum for experimental tanks (AQUACOP, personal communication). Of particular note was the fact

that amino acids constituted over 25% of the oc by weight; compared with the estimated dietary amino acid requirement prole of shrimp (L. vannamei), the oc provided a rich source of threonine, valine, isoleucine and phenylalanine (plus tyrosine), although it was decient in lysine, histidine, and to a lesser extent, arginine and tryptophan (Table 9). Lipids constituted only 2.6% of the oc by weight, and fatty acid analyses revealed modest quantities [albeit rather low relative to levels found in the diets (Table 3)] of n-6 and n-3 polyunsaturated fatty acids, and in particular the highly unsaturated fatty acids arachidonic acid (1.65%), eicosapentaenoic acid (3.0%), and docosahexaenoic acid (1.35%; Table 10). The high proportion of unknown peaks (16.3% of total fatty acids) was probably related to the high number of branched or odd carbon number fatty acids commonly present in bacteria (Kharlamenko et al. 1999); the richness of the oc in 16:0, 16:1n-7 and 18:1n-7 fatty acids was similar to that reported for bacterial-based microbial communities from biological phosphate removal systems (Liu et al. 2000). Interestingly, 18:1n-7 (usually present at high levels in bacteria) was found to be present in the oc, but was not present in the experimental test diets (Table 2). The high ash content of the oc was similar to that reported for activated sludge and probably related to the presence of considerable amounts of acid-insoluble oxides and mixed silicates (Tacon & Ferns 1978/1979). Despite having relatively high sodium content (because of the seawater environment), the oc is a good source of essential minerals and trace elements (Table 9). Moreover, apart from serving as a direct source of nutrients to the shrimp, there is evidence that these organisms also exert a positive eect on the shrimp digestive enzyme activity and gut microora (Moss et al. 2001b).

..............................................................................................

2002 Blackwell Science Ltd Aquaculture Nutrition 8;121^137

134

A.G.J. Tacon et al.

Table 9 Composition of suspended particulate matter or microbial oc (oc) collected from the outdoor shrimp rearing tanks. Values are ranges and means (freeze-dried basis)
Floc AA /RAA (%) Diet1 AA /RAA (%) Shrimp2 AA /RAA (%)

Range Nutrient Suspended microbial floc (mg L ^1) Moisture (g kg ^1) Crude protein (N 6.25) (g kg ^1) Crude lipid (g kg ^1) Cholesterol (g kg ^1) Ash (g kg ^1) Gross energy (MJ kg ^1) Phosphorus (P) (g kg ^1) Potassium (K) (g kg ^1) Calcium (Ca) (g kg ^1) Magnesium (Mg), (g kg ^1) Sodium (Na) (g kg ^1) Manganese (Mn) (mg kg ^1) Iron (Fe) (mg kg ^1) Copper (Cu) (mg kg ^1) Zinc (Zn) (mg kg ^1) Boron (B) (mg kg ^1) Amino acid (g kg^1) Aspartic acid Serine Glutamic acid Proline Glycine Alanine Taurine Cystine Tyrosine Isoleucine Leucine Methionine Phenylalanine Histidine Threonine Lysine Valine Arginine Tryptophan Total amino acids (RAA) E/NE ratio6
1 2

Mean

87.3^200.8 58.6^73.1 292.0^343.3 25.7^26.3 0.47^0.49 255.5^318.1 10.3^12.8 3.6^21.2 1.3^8.9 5.6^28.6 1.2^4.5 4.1^43.1 8.9^46.8 170.8^521.0 3.8^88.6 78.3^577.9 8.8^45.7 30.3^31.1 12.7^13.8 31.7^34.3 12.1^12.8 16.8^17.6 17.6^19.4 0.34^0.36 3.9^4.1 9.9^10.1 12.1^12.6 17.8^19.7 4.7^5.2 14.2^15.3 4.3^4.5 14.4^15.0 9.0^9.6 16.6^18.0 14.6^16.3 1.8^2.2 245^263

157 66 312 26.0 0.48 282 12 13.5 6.4 17 2.6 27.5 28.5 320 22.8 338 27.3 31.1 13.2 33.0 12.5 17.2 18.5 0.35 4.0 10.0 12.4 18.7 4.9 14.8 4.4 14.7 9.3 17.3 15.4 2.0 254 12.25 5.21 13.01 4.91 6.78 7.31 0.14 1.57 3.93 (A /E)3 4.88 (97) 7.39 (146) 1.94 (70)4 5.83 (194)5 1.73 (34) 5.79 (115) 3.66 (73) 6.82 (135) 6.08 (120) 0.77 (16) 100 50.4:49.6 9.20 4.78 21.66 Not analysed 5.39 5.74 0.61 1.46 3.76 (A /E) 4.55 (86) 9.61 (183) 2.67 (79) 4.59 (159) 2.42 (46) 4.15 (79) 6.35 (121) 5.34 (101) 6.55 (125) 1.15 (22) 100 52.6:47.47 9.85 4.13 14.67 6.76 8.04 5.60 0.75 1.03 4.13 (A /E) 4.13 (82) 7.13 (142) 2.13 (63) 4.97(181) 2.16 (43) 4.00 (80) 5.35 (107) 4.57 (91) 9.70 (193) 0.91 (18) 100 50.2:49.8

OI control shrimp diet (Table 2). Estimated dietary amino acid requirement prole of shrimp (Litopenaeus vannamei) calculated according to the method of Ogino (1980) andTacon & Cowey (1985) upon the daily deposition of amino acids in whole body protein of rapidly growing shrimp (data calculated for outdoor shrimp fed the OI shrimp diet and all day feeding regime, with animals growing from an initial body weight of1.58 g (whole body tissue containing alanine 0.97%, arginine 1.42%, asparagine 1.70%, cystine 0.16%, glutamic acid 2.58%, glycine 1.20%, histidine 0.39%, isoleucine 0.70%, leucine 1.24%, lysine 0.90%, methionine 0.36%, phenylalanine1.02%, proline1.25%, serine 0.73%, taurine 0.15%, threonine 0.73%, tryptophan 0.16%, tyrosine 0.73% and valine 0.78% by weight) to a nal body weight of 18.36 g after a 8-week period (whole body tissue containing alanine 1.03%, arginine 1.76%, asparagine 1.81%, cystine 0.19%, glutamic acid 2.70%, glycine 1.46%, histidine 0.40%, isoleucine 0.76%, leucine 1.31%, lysine 0.98%, methionine 0.39%, phenylalanine 0.93%, proline 1.25%, serine 0.76%, taurine 0.14%, threonine 0.74%, tryptophan 0.17%, tyrosine 0.76% and valine 0.84% by weight). 3 A /E ratio ^ dened by Arai (1981) as [(essential amino acid/total essential amino acids plus cystine and tyrosine) 1000]. 4 Methionine + cystine. 5 Phenylalanine + tyrosine. 6 E/NE ratio ^ total essential amino acids, including cystine and tyrosine/nonessential amino acid ratio. 7 NE value is low compared with others because of the absence of a value for proline (not analysed).

..............................................................................................

2002 Blackwell Science Ltd Aquaculture Nutrition 8;121^137

Diet, culture and shrimp growth

Table 10 Fatty acid composition of lipids within microbial oc (oc) collected from the outdoor shrimp rearing tanks. Values are expressed as percentage total recovered fatty acids
Fatty acid C6:0 C8:0 C10:0 C12:0 C13:0 C14:0 C14:1 C15:0 C15:1 C16:0 C16:1n-7 C16:2n-4 C16:3n-4 C16:4n-1 C17:0 C17:1 C18:0 C18:1n-9 C18:1n-7 C18:1n-5 C18:2n-6 C18:2n-4 C18:3c C18:3n-4 C18:3n-3 C18:4n-3 C18:4n-1 C20:0 C20:1n-9 C20:1n-7 C20:2n-6 C20:3n-6 C20:3n-3 C20:4n-6 C20:4n-3 C20:5n-3 C21:5n-3 C22:0 C22:1n-11 C22:4n-6 C22:5n-3 C22:6n-3 C23:0 C24:0 C24:1n-9 Total saturated Total monounsaturated Totaln-6 Totaln-3 Unknown peaks Range 0.0^0.2 0.1^0.2 0.2 0.6 nd 5.2^7.2 4.0^4.7 1.1^1.2 nd 22.7^23.3 11.3^13.7 0.5^1.0 1.6^2.9 0.5 0.5^0.7 nd 2.1^2.5 5.5^6.2 4.2^5.5 0.2^0.3 3.9^4.6 nd 0.2^0.3 nd 3.4^4.5 0.2 nd 0.3 0.6^1.3 0.0^0.3 0.1^0.2 0.1^0.2 nd 1.2^2.1 0.0^0.1 2.1^3.9 0.1^0.2 nd 0.1^0.7 0.1^0.2 0.6^0.8 1.0^1.7 0.1^0.2 nd nd 34.5^35.0 28.4^30.2 6.3^6.4 9.3^9.4 15.9^16.7 Mean 0.1 0.15 0.2 0.6 nd 6.2 4.35 1.15 nd 23 12.5 0.75 2.25 0.5 0.6 nd 2.3 5.85 4.85 0.25 4.25 nd 0.25 nd 3.95 0.2 nd 0.3 0.95 0.15 0.15 0.15 nd 1.65 0.05 3 0.15 nd 0.4 0.15 0.7 1.35 0.15 nd nd 34.75 29.3 6.35 9.25 16.3

135

nd = Not detected (value lower than 0.05%).

As a result of the work carried out in this and other trials, further studies examining the relationship of dietary nutrient levels, development of oc throughout the growth cycle and

growth of shrimp in high-density culture conditions are being undertaken. In view of the fact that zero-water exchange culture systems are usually operated as closed farming systems (with no solids removal or water exchange in the present case), it is not surprising that many essential nutrients will be progressively depleted from the water column (including those additional mineral elements required by the oc and resident phytoplankton) and that other digestive/excretory metabolites or feed contaminants (including possible heavy metal contaminants) could progressively accumulate to toxic levels within the culture system with time (McNeil 2000). In the present instance, the last 2 weeks of the outdoor feeding trial saw a progressive deterioration in water quality, as evidenced by a decrease in pH, a marked increase in nitrite (following a peak in ammonia at week 6), and consequent reduced shrimp growth (Figs 4 & 5). Clearly, closed zero-water exchange culture systems can only biologically support a certain level of nutrient input and shrimp biomass without the system crashing and compromising shrimp growth and survival. For example, McIntosh (2000) reports that organic loadings could reach as high as 500 kg ha 1 day1 within zero-waterexchange shrimp ponds operated in Belize. Interestingly, this is equivalent to a daily loading rate of 50 g feed tank1 day1, which is similar to that reached during the nal weeks of the present outdoor feeding trial (Fig. 3). However, considerably higher loading rates and shrimp yields (as high as 8 kg m2) have been reported within experimental indoor zero-waterexchange culture systems in Montana (USA) operated with continuous illumination, buer input, cation addition, and solids management (R. McNeil, personal communication, August 2000; McNeil 2000). Despite the encouraging results obtained with zero-waterexchange culture systems, it is clear from the two feeding trials that the nutrition and feeding of shrimp reared under closed culture conditions will be dierent from that of animals reared under open running water culture conditions. Apart from the obvious dierences in terms of natural food availability, it is almost impossible to view shrimp feed consumption or feeding behaviour in zero-water-exchange culture systems because of oc production in the culture tanks. Clearly, the nutrition and feeding of the target species must be studied under conditions which mimic as closely as possible those of the intended farm production unit and environment (Tacon 1996). The most promising features of zero-water-exchange culture systems are that they oer both increased biosecurity (Bullis & Pruder 1999) and reduced feed costs and water use (Chamberlain & Hopkins 1994; Boyd 2000), and by doing so increase the possibility of moving the

..............................................................................................

2002 Blackwell Science Ltd Aquaculture Nutrition 8;121^137

136

A.G.J. Tacon et al.

shrimp culture industry along a path of greater sustainability and environmental compatibility.
Freeman, D.W. & Duerr, E.O. (1991) Design and use of outdoor microcosm laboratory tanks for the evaluation of shrimp diets. Aquacult. Engineer., 10, 8997. Grassho, K., Ehrhardt, M. & Kremling, K. eds. (1983) Methods of Seawater Analysis. Verlag Chemie, Weinheim, 419 pp. Hamilton, P.B. (1963) Ion exchange chromatography of amino acids. Anal. Chem., 35, 20552064. Hara, A. & Radin, N.S. (1978) Lipid extraction of tissues with a lowtoxicity solvent. Anal. Biochem., 90, 420426. Hirs, C.H.W. (1967) Determination of cystine as cysteic acid. In: Methods in Enzymology, Vol. XI, Enzyme Structure (Hirs, C.H.W. ed.), pp. 5962. Academic Press, New York. Hugli, T.E. & Moore, S. (1972) Determination of the tryptophan content of proteins by ion exchange chromatography of alkaline hydrolysates. J. Biol. Chem., 247, 28282834. Kharlamenko, V.I., Zhukova, N.V., Khotimchenko, S.V., Svetashev, V.I. & Kamenev, G.M. (1999) Fatty acids as markers of food sources in shallow-water hydrothermal ecosystem (Kraternaya Bight, Yankich Island, Kurile Islands). Mar. Ecol. Prog. Ser., 120, 231241. Lawrence, A. (1996) Shrimp feeds: match diet to production systems. Feed Int., 17, 1822. Leber, K.M. & Pruder, G.D. (1988) Using experimental microcosms in shrimp research: the growth-enhancing eect of shrimp pond water. J. World Aquacult. Soc., 19, 197203. Liu, W.-T., Linning, K.D., Nakamura, K., Mino, T., Matsuo, T. & Forney, L.J. (2000) Microbial community changes in biological phosphate-removal systems on altering sludge phosphorus content. Environ. Microb., 146, 10991107. McIntosh, R.P. (1999) Changing paradigms in shrimp farming. I. general description. Global Aquaculture Advocate, 2, 4247. McIntosh, R.P. (2000) Changing paradigms in shrimp farming. III. Pond design and operation considerations. Global Aquaculture Advocate, 3, 4245. McIntosh, R.P. & Carpenter, N. (1999) Changing paradigms in shrimp farming. II. Breeding for performance. Global Aquaculture Advocate, 2, 3639. McNeil, R. (2000) Zero exchange, aerobic, heterotrophic systems: key considerations. Global Aquaculture Advocate, 3, 7276. Moriarty, D.J.W. (1997) The role of microorganisms in aquaculture ponds. Aquaculture, 151, 333349. Moss, S.M. (1995) Production of growth enhancing particles in a plastic-lined shrimp pond. Aquaculture, 132, 253260. Moss, S.M., Reynolds, W.J. & Mahler, L.E. (1998) Design and economic analysis of a prototype biosecure shrimp growout facility. In: Proceedings of the US Marine Shrimp Farming Program Workshop (Moss, S.M. ed.), pp. 318. The Oceanic Institute, Waimanalo, HI. Moss, S.M., Arce, S.M., Argue, B.J., Otoshi, C.A., Calderon, F.R.O. & Tacon, A.G.J. (2001a) Greening of the blue revolution: eorts toward environmentally responsible shrimp culture. In: The New Wave, Proceedings of the Special Session on Sustainable Shrimp Culture, Aquaculture 2001 (Browdy, C.L. & Jory, D.E. eds). pp. 119. The World Aquaculture Society, Baton Rouge, LA, USA. Moss. S.M., Divakaran, S. & Kim, B.G. (2001b) Stimulating eects of pond water on digestive enzyme activity in the Pacic white shrimp, Litopenaeus vannamei (Boone). Aquaculture Res., 32, 125 132. Murphy, J. & Riley, J.P. (1962) A modied simple solution method for the determination of phosphate in natural waters. Anal. Chim. Acta, 27, 3136. Nunes, A.J.P., Goddard, S. & Gesteira, T.C.V. (1996) Feeding activity patterns of the Southern brown shrimp Penaeus subtilis

Acknowledgements
The authors appreciate the technical assistance of the following sta members of the Aquatic Feeds and Nutrition program at the Oceanic Institute: Michael C. Haring, Eric H. Beyer, Jesse H. Terpstra, Brent E. Larsen, Kekoa K. Nakachi, Gary A. Delanoy and William L. Mulherin. This paper was prepared as part of the activities of the Tropical Aquaculture Feeds and Culture Technology Development Project II: Development of Shrimp Feeds awarded to the Oceanic Institute by the US Department of Agriculture, Agricultural Research Service, under Agreement No. 59-5320-7-989. Mention of trade names or commercial products in this article is solely for the purpose of providing specic information and does not imply recommendation or endorsement by the authors.

References
AOAC (Association of Ocial Analytical Chemists) (1990a) AOAC method no. 968.08. In: Ocial Methods of Analysis of the Association of Ocial Analytical Chemists (Heilrich, K. ed.) 15th edn, p. 84. AOAC, Arlington, VA. AOAC (Association of Ocial Analytical Chemists) (1990b) Methyl esters of fatty acids in oils and fats: gas chromatographic method 963.22. In: Ocial Methods of Analysis of the Association of Ocial Analytical Chemists (Heilrich, K. ed.) 15th edn, p. 964. AOAC, Arlington, VA. Arai, S. (1981) A puried test diet for coho salmon, Oncorhynchus kisutch, fry. Nippon Suisan Gakk., 47, 547550. Avnimelech, Y. (2000) Nitrogen control and protein recycling: activated suspension ponds. Global Aquaculture Advocate, 3, 2324. Boyd, C.E. (2000) Water use in aquaculture. Global Aquaculture Advocate, 3, 1213. Bullis, R.A. & Pruder, G.D. eds. (1999) Controlled and Biosecure Production Systems: Evolution and Integration of Shrimp and Chicken Models. Proceedings of a Special Session of the World Aquaculture Society, Sydney, Australia, 2730 April 1999. 106 pp. The Oceanic Institute, Waimanalo, HI. Chamberlain, G.W. & Hopkins, J.S. (1994) Reducing water use and feed costs in intensive ponds. World Aquaculture, 25, 2932. Curds, C.R. (1992) Protozoa and the Water Industry. Cambridge University Press, Cambridge, UK, 122 pp. DElia, C.F., Steudler, P.A. & Corwin, N. (1977) Determination of total nitrogen in aqueous samples using persulfate digestion. Limnol. Oceanog., 22, 760764. Divakaran, S.D. (1999) AFIA Laboratory Methods Compendium II, Vol. 4: Aquaculture. American Feed Industry Association, Arlington, VA, 109 pp. FAO (Food and Agriculture Organization of the United Nations) (2000) Yearbook of Fishery Statistics 1998, Vol. 86/2. Aquaculture production. FAO Statistics Series No. 154 and Fisheries Series No. 56, FAO, Rome, 182 pp.

..............................................................................................

2002 Blackwell Science Ltd Aquaculture Nutrition 8;121^137

Diet, culture and shrimp growth

under semi-intensive culture in North East Brazil. Aquaculture, 144, 371386. Ogino, C. (1980) Requirements of carp and rainbow trout for essential amino acids. Nippon Suisan Gakk., 46, 171174. Ogle, J. & Lotz, J. (2000) Closed systems for maturation and growout of shrimp. Global Aquaculture Advocate, 3, 8990. Reid, B. & Arnold, C.R. (1992) The intensive culture of the Penaeid shrimp Penaeus vannamei Boone in a recirculating raceway system. J. World Aquacult. Soc., 23, 146153. Robertson, L., Lawrence, A.L. & Castille, F.L. (1993) Eect of feeding frequency and feeding time on growth of Penaeus vannamei (Boone). Aquacult. Fish. Manag., 24, 16. Rosenberry, R. (1999) World Shrimp Farming 1999. Shrimp News International, San Diego, CA, USA, 320 pp. Snedecor, G.W. & Cochran, W.G. (1967) Statistical Methods. The Iowa State University Press, Ames, IA, 593 pp. Solorzano, L. (1969) Determination of ammonia in natural waters by the phenolhypochlorite method. Limnol. Oceanogr., 14, 799801. Strickland, D.H. & Parsons, T.R. (1972) A practical handbook of seawater analysis. Bull. Fish. Res. Bd. Canada 167, 311 pp. Tacon, A.G.J. (1978/1979) Activated sewage sludge, a potential animal foodstu II. Nutritional characteristics. Agric. Envir., 4, 271279. Tacon, A.G.J. (1996) Nutritional studies in crustaceans and the problems of applying research ndings to practical farming systems. Aquacult. Nutr., 1, 165174. Tacon, A.G.J. (2000) Rendered animal by-products: a necessity in aquafeeds for the new millennium. Global Aquaculture Advocate, 3, 1819. Tacon, A.G.J., Conklin, D.E. & Pruder, G.D. (1999) Shrimp feeds and feeding: at the crossroads of a cultural revolution. In: Controlled and Biosecure Production Systems: Evolution and Integration of Shrimp and Chicken Models. Proceedings of a Special Session of the World Aquaculture Society, Sydney, Australia, 2730 April 1999 (Bullis, R.A. & Pruder, G.D. eds), pp. 5566. The Oceanic Institute, Waimanalo, HI. Tacon, A.G.J. & Cowey, C.B. (1985) Protein and amino acid requirements. In: Fish Energetics: New Perspectives (Tytler, P. & Calow, P. eds), pp. 155183. Croom-Helm, London, UK. Tacon, A.G.J. & Ferns, P.N. (1978/1979) Activated sewage sludge, a potential animal foodstu I. Proximate and mineral content: seasonal variation. Agric. Envir., 4, 257269. Van Wyk, P.M. (2000) Culture of Penaeus vannamei in single-phase and three-phase recirculating aquacultrue systems. Global Aquaculture Advocate, 3, 4143. Velasco, M. & Lawrence, A. (2000) Initial evaluation of shrimp vitamin requirements in laboratory tanks without water exchange. Global Aquaculture Advocate, 3, 23. Velasco, M., Lawrence, A.L. & Castille, F.L. (1999) Eect of variations in daily feeding frequency and ration size on growth of shrimp, Litopenaeus vannamei (Boone), in zero-water exchange culture tanks. Aquaculture, 179, 141148.

137

..............................................................................................

2002 Blackwell Science Ltd Aquaculture Nutrition 8;121^137