4.0 FORMULATED FEEDS FOR AQUACULTURE IN THE TROPICS

4.1 Micro-encapsulated larval diet
4.2 Diets for warm water omnivores
4.3 Diets for warm-water carnivores
4.4 Rainbow trout

Eleven of the twelve countries covered in this report lie within the climatic belt 30 degrees latitude from the equator. Tunisia, although situated north of this tropical belt, lies in the southern Mediterranean and therefore has a climate not unlike that of the tropics throughout much of the year.

Although trout culture has been traditional to at least three of the countries (India, Mexico and Venezuela), recent interest in all twelve countries has been development of aquaculture based on more rapidly growing indigenous warm water species. Development of appropriate diets for these species is therefore essential.

The main objectives in diet formulation for aquaculture are to: (a) satisfy all known nutrient requirements for growth of the species; (b) minimize feed cost; and (c) select and use ingredients that will result in products that are readily utilized to minimize wastage.

The nutrient requirements of most of the local species of fish are not known with certainty. It has been demonstrated, however, that most warm water species that take to artificial feed grow well on diets formulated for common carp, a species on which much of its nutrient requirements are already known. It is a matter of time before the nutrient requirements of the other species are determined. In the meantime, however, the need is for low cost diets that can be efficiently utilized for profitable farming of these species.

With the above objectives in mind, a number of diets were formulated for testing in each of the twelve countries visited. Some of these tests were undertaken during periods of the consultant's visits. The diets were formulated for most of the finfish species identified for mass culture in these countries. Table 4 lists those species and their natural feeding habits.

Tables 5, 6 and 7 specify the essential nutrients and their recommended levels in practical aquaculture diets.

The levels for warm water species are based on present recommendations set for common carp and channel catfish, respectively. Many warm water omnivores, including the tilapia, for which precise requirements of some nutrients, notably the essential amino acids, have not been determined, have responded well to carp diets. Similarly, diets formulated for the basically carnivorous channel catfish have been very successfully applied in Thailand in commercial culture of Clarias batrachus, another carnivorous species whose dietary requirements are still not known with precision. Therefore, until nutrient requirements of the other warm water species of interest to aquaculture are established, the present recommendations permit the formulation of diets that will support growth sufficiently to make commercial production of those species feasible.

The recommended nutrient levels for trout diets take into account availability and cost of raw material in the countries concerned. Because optimum growth in trout requires dietary protein levels that too often are difficult to achieve with local ingredients, the suggested diets are those that can be prepared under constraints of local feed ingredient supply, and at the same time provide for adequate growth rates.

Table 4 - Major Aquacultural Species (Finfish) raised on artificial feeds

Species Name	Common Name	Habitat	Natural Feeding Habits	Countries
Cyprinus carpio	Common carp	Fresh, warm water	Omnivorous	All twelve countries
Labeo rohita	Rohu	"	"	India, Nepal, Sri Lanka, Thailand
Catla catla	Catla	"	Plankton feeders	India, Nepal, Sri Lanka
Cirrhinus mrigala	Mrigal	"	Omnivorous	India, Nepal, Sri Lanka
Sarotherodon niloticus	Tilapia	"	"	All twelve countries
Tilapia guineensis	Tilapia	"	Herbivorous	Nigeria
Sarotherodon mossambicus	Tilapia	"	Omnivorous	Nigeria, The Philippines
Ophiocephalus striatus	Snakehead	"	Carnivorous	Thailand
Pangasius sutchi	Large catfish	"	"	Thailand
Clarias batrachus	Catfish	"	"	India, Thailand
Clarias lazera	Catfish	"	"	Egypt, Nigeria
Ictalurus punctatus	Catfish	"	"	Mexico
Pimelodus blochii	Catfish (Mandi)	"	"	Brazil
Prochilodus spp.	Pacu, curimbata	"	Omnivorous	Brazil, Venezuela
Colossoma spp.	Tambaqui, cachama	"	"	Brazil, Venezuela
Mugil spp.	Mullets	Brackish, warm water	Omnivorous	All 12 countries except Nepal
Epinephalus tauvina	Grouper	"	Carnivorous	Malaysia, Thailand
Dicentrarchus labrax	Sea bass	"	"	Egypt, Tunisia
Sparus auratus	Sea bream	"	"	Egypt, Tunisia
Lates calcerifer	Sea bass	"	"	Malaysia, Thailand
Oxyeleotris marmoratus	Sand goby	"	"	Thailand
Siganus spp.	Rabbit fish	"	Herbivorous	Malaysia
Salmo giardneri	Rainbow trout	Fresh, cold water	Carnivorous	India, Mexico, Venezuela

Table 5 Nutrient Specifications for Commercial Aquaculture Feeds

(Warm-water omnivorous species)

	Fry and Fingerlings	Juveniles and Growers	Brood Fish
Protein, % min.	30	25	30
Lipids, % min.	8	5	5
Ca, %, min.	0.8	0.5	0.8
Ca, % max.	1.5	1.8	1.5
P, avail, % min.	0.6	0.5	0.6
P, avail, % max.	1.0	1.0	1.0
Met + Cys, % min.	1.2	0.9	1.0
Lysine, % min.	2.0	1.6	1.8
DE Kcal/100 g, min.	310	280	280
Vitamins (Supplement) (per 100 kg)
A, i.u.	600 000	500 000	600 000
D3 i.u.	100 000	100 000	100 000
E, i.u.	6 000	5 000	6 000
K, g	1.2	1.0	1.0
C, g	24.0	20.0	24.0
Thiamine, g	2.4	2.0	2.4
Riboflavin, g	2.4	2.0	2.4
Pantothenic acid, g	6.0	5.0	6.0
Niacin, g	12.0	10.0	12.0
Pyridoxine, g	2.4	2.0	2.4
Biotin, g	0.024	0.02	0.024
Folic acid, g	0.6	0.5	0.6
Choline Cl, g	54.0	50.0	54.0
B-12, mg	2.4	2.0	2.4
Minerals (Supplement) (per 100 kg)
Iron, g	5.0	5.0	5.0
Copper, g	0.3	0.3	0.3
Manganese, g	2.0	2.0	2.0
Zinc, g	3.0	3.0	3.0
Iodine, mg	10.0	10.0	10.0
Cobalt, mg	1.0	1.0	1.0
Selenium, mg	10.0	10.0	10.0

Table 6 - Nutrient Specifications for Commercial Aquaculture Feeds

(Warm-water carnivorous species)

	Fry and Fingerlings	Juveniles and Growers	Brood Fish
Protein, % min.	36	30	36
Lipids, % min.
Ca, % min.	1.0	1.0	1.0
Ca, % max.	1.5	1.5	1.5
P, avail., % min.	0.5	0.5	0.5
P, avail., % max.	0.8	0.8	0.8
Met. + Cys., % min.	1.2	0.9	1.0
Lysine, % min.	2.0	1.6	1.8
DE, Kcal/100 g, min.	360	300	330
Vitamin and mineral supplements (See Table 5)

Table 7 - Nutrient Specifications for Commercial Aquaculture Feeds

(Rainbow Trout)

	Fry and Fingerlings	Juveniles and Growers	Brood Fish
Protein, % min.	43.0	40	40
Lipids, % min.	8.0	6.0	5.0
Ca, % min.	0.8	0.5	0.8
Ca, % max.	1.5	1.8	1.5
P, avail., % min.	0.6	0.5	0.6
P, avail., % max.	1.0	1.0	1.0
Met. + Cys., % min.	1.7	1.6	1.4
Lysine, % min.	2.9	2.7	2.4
DE, Kcal/100 g, min.	330	300	280
Vitamin and mineral supplements (See Table 5)

4.1 Micro-encapsulated larval diet

4.1.1 Results of tests with the micro-encapsulated egg diet

A micro encapsulated larval diet based on whole chicken egg has been described in an earlier report (ADCP/REP/80/11, pages 355-361). The table on the composition of chicken egg is reproduced below:

Table 8

Composition of (Chicken) Egg

	Whole egg	Egg white	Egg yolk
Protein, %	48.8	76.9	32.8
Fat, %	43.2	-	62.2
Gross energy, kcal/kg	5 830	3 070	6 910
Metabolizable energy (ME), kcal/kg	4 810	2 533	5 700
ME: protein ratio	9.8	3.3	17.3
Calcium, %	0.2063	0.0427	0.2653
Phosphorus, %	0.873	0.282	1.020
Amino Acids, %
Arginine	2.968	4.179	3.369
Cystine	0.837	1.282	0.526
Isoleucine	2.734	4.307	1.896
Leucine	4.063	6.273	2.790
Lysine	3.047	4.427	2.369
Methionine	1.563	2.700	1.663
Phenylalanine	2.500	4.427	1.316
Threonine	2.500	3.692	1.843
Tryptophan	0.837	1.350	0.577
Tyrosine	1.952	3.076	1.316
Valine	3.674	6.025	2.263

The encapsulated diet itself consists of whole egg supplemented with vitamins and a mineral source for calcium ^1/. The diet was tested with good results on fry of rohu (Labeo rohita) and clarias (Clarias batrachus) at RLCT Bangkhen ^2/. Two methods were used in its preparation. The first was as previously described, i.e. by rapidly adding boiling water with constant stirring to the homogenate. The product was a very fine suspension of micro-capsules which was then separated by medium speed (5 000 rpm for 15 minutes) centrifugation. A modified heat processing method was also tried. In this method, the homogenate was steam cooked for about 10 minutes and the custard-like product passed through a fine mesh nylon cloth to produce the micro-capsules. Since water (cold) was also required in this operation, the capsules were finally separated by centrifuging (at 3 000 rpm for 10 minutes). Product yield by the modified method was considerably higher and less fuel was probably required for processing. The modified method was also used to prepare a freeze-dried version of the larval feed at RLCI, Dhauli ^3/. Details of the freeze dried product are provided in the next section of this report.

^1/ The rate of addition of various vitamins was according to the recommended levels in diets for fingerlings as described in Table. The amount of calcium carbonate or limestone added to one egg weighing 50 g (10 g dry matter) was 0.2 g.

^2/ RLCT, FAO Regional Lead Centre for Aquaculture in Thailand

^3/ RLCI, FAO Regional Lead Centre for Aquaculture in India

4.1.1 Results of tests with the micro-encapsulated egg diet

The tests conducted at RLCT involved four species: the rohu, Clarias, sand goby and Pangasius.

Rohu (Labeo rohita)

Two preparations of encapsulated egg diet were tested against boiled egg yolk. The experiment was carried in replicate in glass aquaria each containing 250 fry that were less than 48 hours old. Diet rations were measured out daily and resuspended in water before feeding. The rest of the diet preparations were kept under refrigeration. Daily feed on a dry basis amounted to 20 percent estimated body weight of the fish and were given four times daily. The aquaria were cleaned daily with about 75 percent water replacement.

The larvae were kept on the diets for ten days, during which growth and survival were closely monitored. Fish on both encapsulated egg preparations grew normally and remained active during the period. There was less than 5 percent mortality among larvae fed the encapsulated diets. Fish that were fed the diet prepared by steaming appeared whitish in the abdominal region. However, there was no difference in survival rate between the two encapsulated diets.

Larvae fed boiled egg yolk suffered mortality from the fourth day of the test, and by the tenth day there were few survivors.

Clarias (Clarias batrachus)

The encapsulated egg diet was tested against live Moina and two artificial feeds: dried earthworms from Japan and a larval diet of New Zealand origin. Fish on Moina performed best in terms of growth and survival, followed by the encapsulated diet. Performance of fish on the two other diets was very poor.

Sand goby (Oxyelotris marmoratus)

Larvae of sand goby were fed the encapsulated diet for two weeks. Although there was no observed mortality up to the seventh day, large losses occurred beyond that time period. It was not known whether the losses were due to mortality or because of the increasing presence of water mites from the fourth day of the test.

Pangasius (Pangasius sutchi)

The test of the encapsulated diet on Pangasius larvae was unsuccessful. The aquaria were too densely stocked at the beginning, resulting in uncontrolled cannibalism among the young fish.

These preliminary studies indicate that encapsulated whole egg is a good substitute for live natural food for rohu and clarias larvae.

Earlier attempts were also made to test the encapsulated diet on the cachama (Colossoma spp.) at the Guanapito warm water fish station in Venezuela. However, due to improper procedures carried out in preparing the diet, the results were not conclusive. Attempts will be made at the Inter-regional Aquaculture Centre (IRAC) in Szarvas, Hungary to carry out properly controlled experiments with this new form of larval diet to establish its efficacy and commercial potential.

4.2 Diets for warm water omnivores

4.2.1 Carps and Tilapia
4.2.2 Pacu (Prochilodus spp.) and Tambaqui (Colossoma spp.)

4.2.1 Carps and Tilapia

The practice of carp polyculture in earthern ponds has gained prominence in India in recent years. In this culture system, Indian major carps, the catla (Catla catla), rohu (Labeo rohita) and mrigal (Cirrhinus mrigala) are stocked with the common carp (Cyprinus carpio), the grass carp (Ctenopharyngodon idella) and silver carp (Hypothalmichthys molitrix) in varying ratios, with a predominance of Indian major carps. The main food supply is the ponds' natural production of algae, plankton, detritus and benthic organisms. However, in order to support higher stocking densities, supplementary feeds are usually given. These feeds consist of a mixture of rice bran and oilcakes (usually groundnut oilcake or mustard oilcake). Sometimes aquatic as well as terrestrial weeds are also provided to augment the food supply. The rice bran/oilcake mixture is in a rough ratio of 1:1 by weight and is given daily at a rate of 1 to 4 percent body weight of fish. Although increased fish production has been attributed directly to supplementary feeding, an evaluation of its efficacy has not been possible given the many variations of stocking ratios and generally arbitrary way in which supplementary feeding rates are determined. Moreover the rice bran/oilcake mixture lacks minerals and vitamins and rapidly separates into its component ingredients during the feeding process. Considerable wastage is, therefore, expected and the effectiveness of such an artificial diet would also depend greatly on the natural productivity of the pond, especially for the supply of animal protein in the form of zooplankton and benthic organisms.

To determine if an improvement could be obtained by supplementing the diet mixture with minerals and vitamins, a test was conducted at RLCI, Dhauli ^1/. Two diets based on rice bran and groundnut oilcake to contain 28 percent crude protein were formulated. One of the diets also contained the vitamin and mineral supplement ^2/. Those diets were fed to rohu fingerlings averaging 3 g at the start of the experiment. Growth of fish fed mineral and vitamin fortified diet was found to be superior to that of control.

^1/ For full details of this test see "Carp nutrition research at the Freshwater Aqua-culture Research and Training Centre, Dhauli" FI:DP/IND/75/031, Field Document 4, May 1982

^2/ Control diet: rice bran, 40%; groundnut oilcake, 60%; test diet: rice bran, 32.5%; groundnut oilcake: 65,5%; dicalcium phosphate: 2%, and one normal strength multi-vitamin tablet (for human use) per 1 kg diet mixture

Other diets that were formulated for testing with carps at RLCI appear in Tables 9, 10 and 11. The test diets for each age class differ with respect to protein and/or energy sources. Vegetable protein supplements such as groundnut oilcake, sesame oilcake and soyabean meal are, for practical purposes, interchangeable where a choice does not exist. The same applies for energy feeds such as rice bran and wheat bran, or among cereal grains. The test diets were designed to determine which would be the preferred formulations, should choice of ingredients exist.

The experimental diets described were prepared in the following manner; the rice bran or wheat bran was cooked in water constituting 120 percent ^1/ of total weight of the diet mixture. The remaining components, after first being properly mixed, were then blended into the hot pasty slurry to form a dough. This was then extruded into pellets by means of a meat mincer fitted with a 3 mm die. The extruded pellets contained about 55 percent moisture and, unless dried in the sun, was used as moist type feed. For storage of the prepared feed, the pellets were spread out over a fine screen and dried in the sun. The dried pellets, which contained about 11 percent moisture, were found to have good water stability and shelf life of more than two months when kept in closed containers.

^1/ The amount of water used was determined by the absorptive characteristic of the bran. Higher quality rice bran, which contains more starch and less fibre, requires greater addition of water, as in this case

Following the installation of the laboratory model pellet mill at RLCI the dry diet mixtures were pressed out using a 2 mm die. The pelleting operation was carried out first with steam conditioning and then with 6 to 8 percent water blended into the dry mash before pelleting. Better results were obtained using the latter method ^2/. For dry type sinking pellets, water stability of the products was good (about 90 minutes). Also the percentage of "fines" was very small (less than 1 percent).

^2/ See section on Manufactured Foods for Aquaculture

In Egypt the most readily available protein feed is cottonseed meal. All others have to be imported, including fish meal and soyabean meal. Two diets were prepared for testing on common carp and tilapia in cages set in a 4 ha pond at El Zaweya. The two formulations tested are shown in Table 12 and details of the experiment are described in the relevant consultancy report prepared for the Project EGY/80/002 (See reference list). Growth measurements made at the end of the first two weeks indicated marginal weight gains in both dietary treatments. The slow growth was most probably due to low water temperatures during the test period (14°C in the early morning and a maximum of 24°C shortly after noon). Although meant to continue for three months, the experiment had to be terminated because of fish escape from all the cages during adjustment of water level in the pond, ten days after the experiment began. Fish that were recovered appeared well fed, and gain in average weight was registered for both carp and tilapia.

Formulation of diets for tilapia in Port Harcourt, Nigeria, was hampered by acute shortages of conventional feed ingredients, although abattoir waste (mainly fresh cattle blood) was available free in considerable quantities, and fish meal in limited quantities from the NIOMR fish meal pilot plant in Lagos. Table 13 shows the three diets that were formulated for testing at the African Regional Aquaculture Centre (ARAC) at Aluu. A problem associated with the use of fresh animal blood in diet making in the tropics is the rapid proliferation of mold in the product, especially if drying of the pellets is not completed rapidly. On one occasion it was observed that fungal spores (detected with a magnifying glass) had already contaminated the feed within a week after its preparation.

Table 9

Test Diets for Rohu Fingerlings at RLCI, Dhauli, India

Ingredients	Diet No. 1/
(%)	1	2	3	4	5
Groundnut oilcake	60	-	58	33	-
Sesame oilcake	-	78	-	33	77
Wheat bran	38	20	-	-	-
Rice bran	-	-	40	32	21
Dicalcium phosphate	1.5	1.5	1.5	1.5	1.5
Table salt	0.3	0.3	0.3	0.3	0.3
Trace minerals 2/	0.1	0.1	0.1	0.1	0.1
Vitamin mixture 3/	0.1	0.1	0.1	0.1	0.1
Calculated chemical composition:
Crude protein, %	29.3	27.9	29.2	28.6	27.9
Digestible energy, kcal/g	2.93	3.0	2.72	2.80	2.90
Lysine, %	0.88	0.61	1.05	0.91	0.71
Methionine +cystine, %	0.77	1.18	0.77	0.95	1.18
Cost/kg (excluding cost of mineral and vitamin supplements) 4/
I. Rs.	1.61	1.68	1.28	1.38	1.36

^1/ Diets were made into dry pellets as described in the text

^2/ Trace mineral mix made from CuSO₄ . 5H₂O, FeSO₄ . 7H₂0, MnSO₄ . H₂O, ZnO, CoCl₂ . 6H₂0, Kl and CaHPO (filler) to provide the following (ppm total diet): Cu, 10; Fe, 100; Mn, 50; Zn, 50; Co, 0.05; and I, 0.1.

^3/ Vitamin mixture to provide the following (per kg total diet): Vitamin A, 5 000 IU; vitamin D, 600 IU; thiamin, 10 mg; riboflavin, 20 mg; pantothenic acid, 30 mg; niacin, 50 mg; and ascorbic acid, 200 mg. Due to the unavailability of individual vitamins at the time of diet preparation. Miles Brand multivitamin tablets were used at the level of 1 tablet/kg diet.

^4/ For costs of individual ingredients, see Table 3.

Table 10

Recommended Test Diets for Carp Fry (India)

Ingredients	Diet No. 1/
%	1	2	3	4	5
Groundnut oilcake	61	45	28	20	61
Sesame oilcake	-	-	-	35	-
Fish meal	-	10	20	10	-
Rice bran	37	44	51.8	34	36.7
Dicalcium phosphate	1.5	0.5	-	0.5	1.5
Table salt	0.3	0.3	-	0.3	0.3
Trace minerals 2/	0.1	0.1	0.1	0.1	0.1
Vitamin mixture 3/	0.1	0.1	0.1	0.1	0.1
1-Lysine 4/	-	-	-	-	0.3
Calculated chemical composition
Crude protein, %	30.0	30.2	30.1	30.0	30.3
Digestible energy, kcal/g	2.77	2.74	2.71	2.82	2.77
Lysine, %	1.07	1.37	1.67	1.23	1.37
Methionine +cystine, %	0.79	0.81	0.96	1.07	0.79
Cost/kg diet (excluding cost of mineral and vitamin supplements) 5/
I. Rs.	1.32	1.57	1.76	1.58	1.47

^1/ Test diets to be made into dry pellets as described in text. Pellets should be ground to particle size not larger than 0.5 mm

^2/ As in Table 9

^3/ As in Table 9

^4/ Estimated cost: I. Rs. 50/kg

^5/ For cost of individual ingredients, see Table 3

Table 11

Recommended Test Diets for Carp Brood Stock (India)

Ingredients	Diet No. 1/
%	1	2	3	4
Groundnut oilcake	61	40	51.8	-
Sesame oilcake	-	-	-	60
Prawn head meal 2/	-	30	15	30
Rice bran	37	29.8	33	11.8
Dicalcium phosphate	1.5	-	-	-
Table salt	0.3	-	-	-
Trace minerals 3/	0.1	0.1	0.1	0.1
Vitamin mixture 4/	0.1	0.1	0.1	0.1
Calculated chemical composition
Crude protein, %	30.0	30.0	30.2	30.4
Digestible energy, kcal/g	2.77	2.63	2.73	2.87
Cost/kg diet (excluding cost of mineral and vitamin supplements) 5/
I. Rs.	1.32	1.50	1.43	1.75

^1/ Test diets to be made into dry pellets as described in text
^2/ No data available on lysine and sulphur amino acid content
^3/ As in Table 9
^4/ As in Table 9
^5/ For cost of individual ingredients, see Table 3

It is suggested that such feeds be prepared fresh for feeding, to avoid any health problem to fish, unless preservatives such as propionate or benzoate are available for addition to the diet mixture to prevent mold contamination.

Table 12

Composition of Test Diets for Tilapia and Carp Fingerlings for the El Zaweya Government Fish Farm (Egypt)

Ingredients, %	Diet 1/
	A	B
Cottonseed meal	53	-
Soyabean meal	-	50
Rice bran	43	46.2
Bone meal	3	3.5
Table salt	0.2	0.2
d1-Methionine	0.3	-
1-Lysine	0.4	-
Vitamin mixture 2/	0.1	0.1
Total	100.0	100.0
Calculated chemical composition
Protein, %	30	30
Lysine	1.6	1.6
Methionine + cystine	0.8	0.8
Digestible energy, kcal/g	2.25	2.25
Cost, piastres/kg 3/	7.7	15.9

^1/ Preparation of diets: rice bran and bone meal were mixed and boiled with water constituting 100 percent total weight of the diet. The purpose of boiling was to bring out the natural binding properties of starch to produce water-stable pellets. The slurry was then allowed to cool slightly before the remaining ingredients, previously mixed, were blended in. The resulting dough was then passed through a kitchen meat grinder fitted with a 2 mm die. The moist pellets produced were dried in the sun over a fine mesh screen. Pellets were crumbled to smaller particle size before feeding.

^2/ The vitamin mixture used was a commercial product for poultry and, as applied, provided the following (per kg diet): vitamin A, 5 000 IU; vitamin D₂, 600 IU; thiamine, 10 mg; riboflavin, 20 mg; pantothenic acid, 30 mg; niacin, 50 mg; pyridoxine, 2 mg; and ascorbic acid, 200 mg.

^3/ For cost of individual ingredients, see Table 3

Table 13

Test Diets for Tilapia Fingerlings at the African Regional Aquaculture Centre, Aluu (Nigeria)

Ingredients		Diet 1/
Fresh bovine blood		57	28	-
Wheat bran		43	44	66
Groundnut oil cake		-	28	-
Fish meal		-	-	34
Vitamin mix 2/		30	70	90
Vitamin C supplement 3/		10	20	20
Total		100.06	100.09	100.11
Approximate analysis (%)
	Dry matter	50	70	90
	Protein (dry basis)	30	30	30

^1/ Amounts in kg except for the vitamin supplements, which are in g. Preparation of diets: Diet 1, Fresh blood or thawed out frozen blood was weighed out in a large container; the pulverized vitamin mixtures were stirred into the blood, followed by addition of the wheat bran; after thorough mixing, the dough was extruded with a meat grinder; the pellets were then spread out over a wire mesh tray for drying in the sun. Diet 2. The wheat bran was cooked in 40 litres of water and blended with the fresh blood into which vitamins had previously been added; the groundnut oilcake was then blended into this mixture followed by extrusion into pellets. Diet 3. The rice bran was cooked in 100 litres of water and then blended with the mixture containing the fish meal and vitamin mixtures; the dough obtained was then extruded as described earlier.

^2/ The vitamin mixture (Coryl SP) was a commercial preparation meant for supplementing poultry diets at the rate of 1 g per 1 kg feed. It does not contain vitamin C.

^3/ "Redoxin" lozenges that contain 50 percent ascorbic acid.

Table 14

Composition of Mexican Feedstuffs ^1/

		% Dry Matter
Feedstuff		% Dry matter	Crude protein	Crude fat	Crude fibre	Ash	NFE 2/	Max. $/kg 3/
Maize, ground		88.8	10.0	4.5	2.8	2.4	80.3	6.5
Maize gluten		87.6	46.6	6.7	6.1	7.4	33.2	6.5
Maize germ meal		92.6	21.6	5.7	8.8	5.9	57.9	5.0
Sorghum		85.4	11.1	3.7	6.1	4.1	74.8	4.0
Wheat		91.5	16.2	3.4	3.1	2.8	74.6	5.5
Wheat bran		88.3	17.6	4.1	10.4	6.4	61.5	4.5
Wheat middlings		91.6	15.1	5.6	10.7	5.0	63.5	5.0
Rice bran, solvent extracted		94.3	12.9	1.3	21.8	22.1	41.9	4.5
Soyabean, whole		87.9	40.5	25.3	6.2	5.8	22.2	9.0 4/
Soyabean meal		88.5	54.2	1.2	5.3	7.9	31.4	8.5
Cottonseed meal		91.6	45.4	1.3	14.5	11.4	27.4	7.2
Safflowerseed meal, 22%		92.2	24.0	0.7	3.9	5.1	66.4	4.0
Safflowerseed meal, 36%		91.5	35.1	2.6	19.2	9.5	33.6	5.6
Sunflowerseed meal		91.8	29.8	4.4	21.5	7.9	36.4	4.5
Sesame oil meal		90.1	51.0	1.3	6.8	12.9	28.0	8.0
Copra meal		94.0	24.6	7.2	28.4	7.4	32.5	6.0
Alfalfa meal		90.1	18.9	2.8	24.6	11.4	42.4	4.5
Fish meal		93.1	70.4	10.7	0.9	14.9	3.8	13.0
Whey powder		90.1	15.3	2.1	0.2	9.9	72.5	20.0
Yeast powder		92.6	48.0	1.4	2.9	7.2	40.5	10.0
Brewers' grains		93.1	29.8	9.5	12.1	2.6	46.0	5.2
Vegetable oil		99.0	-	100	-	-	-	25.0
Molasses		85.0	5.4	0.3	9.9	10.3	74.0	1.5
Rock phosphate		Containing: phosphorus, 7%; calcium, 18%						3.0
Dicalcium phosphate		Containing: phosphorus, 18%; calcium, 24%						15.0
Limestone powder		Containing mainly CaCO3						2.0 4/
Salt								2.0
d1-Methionine		Containing: 98% d1-methionine						100.0
1-Lysine		Containing: 98% 1-lysine						84.0
Vitamin premix 5/								50.0
Mineral premix 6/								10.0
Mesquite:
	leaves with pods and seeds	-	9.9	1.5	23.3	3.1	62.2	NP 7/
	pods with seeds	-	13.8	3.2	27.8	4.9	50.4	NP 7/
	seeds	5.6	10.7	1.4	14.7	4.1	64.5	NP 7/

^1/ Source: Latin American Tables of Feed Composition, 1974, Florida and ALBAMEX. Prices provided by ALBAMEX

^2/ Nitrogen free extract

^3/ Mex.$ 25 = US$ 1 (approximately)

^4/ No prices provided by ALBAMEX. Prices indicated based on relative fair value of ingredients.

^5/ Each gramme of premix to contain the following: Vitamin A, 1 000 IU; Vitamin D₃, 200 IU; Vitamin E, 10 IU; (in mg) Vitamin K, 2, Thiamine, 4; Riboflavin, 4; Pantothenic acid, 10; Niacin, 20; Pyridoxine, 4; Biotin, 0.02; Folic acid, 1; Ascorbic acid, 40; Choline chloride, 90; Vitamin B₁₂, 0.004; Ethoxyquin (antioxidant), 16.

^6/ Each gramme of premix to contain the following (mg mineral element): Iron, 50; Copper, 3; Cobalt, 0.01; Manganese, 20; Zinc, 30; Iodine, 0.1; and Selenium, 0.1. The inclusion of selenium is necessary in view of its established role as an essential trace element in animal nutrition and its low levels in plant protein sources which are expected to replace selenium-rich fish meal in the least-cost diets.

Table 15

Least Cost Formulations for Warm Water Species ^1/

(Mexico)

		Fry and Fingerlings (I)	Grower (II)	Brood Fish (III)
A. CONSTRAINTS 2/
	RHS
	Protein, %	30.0	25.0	30.0
	Lysine, %	2.0	1.6	1.8
	Methionine + Cystine, %	1.2	0.9	1.0
	Lipid, %	8.0	-	-
	ME, Kcal/g*	3.1	2.8	2.8
	Calcium, 7.	0.8	0.5	0.8
	Phosphorus, total, %	0.8	0.8	0.8
	Ranges
	Calcium, %	0.7	1.0	0.7
	Phosphorus, %	0.5	0.5	0.5
	Bounds (%)
	Alfalfa meal (UP)	0.0	10.0	10.0
	Fish meal (LO)	5.0	-	-
	Molasses (UP)	0.0	5.0	5.0
	Sorghum (UP)	5.0	15.0	15.0
	Cottonseed meal (UP)	5.0	15.0	0.0
	Maize gluten (FX)	10.0	10.0	10.0
	Vitamin premix (FX)	0.6	0.5	0.6
	Mineral premix (FX)	0.1	0.1	0.1
B. SOLUTIONS
		Diet
		I	II	III
	Ingredients, kg/tonne diet 3/
	Extruded whole soyabeans	530.0	540.0	438.0
	Soyabean meal	-	-	31.5
	Sunflower seed meal	-	110.7	98.0
	Fish meal	140.0	-	-
	Maize gluten	100.0	- 4/	100.0
	Maize germ meal	-	73.4	-
	Sorghum	50.0	150.0	150.0
	Wheat	150.0	-	-
	Wheat bran	-	30.0	-
	Safflowerseed meal	-	-	100.0
	Molasses	-	50.0	50.0
	Calcium orthophosphate	1.5	13.5	16.0
	Limestone powder	17.0	25.4	6.6
	1-Lysine	-	-	2.9
	Vitamin premix	6.0	6.0 5/	6.0
	Mineral premix	1.0	1.0	1.0
	Analysis
	Metabolizable energy, Kcal/g	3.1	2.8	2.8
	Crude protein, %	35.5	26.9	30.0
	Lysine, %	2.22	1.6	1.8
	Methionine + Cystine, %	1.20	0.9	1.05
	Total phosphorus, %	0.8	0.8	0.8
	Calcium, %	1.5	1.5	0.8
	Fibre, %	5.0	8.0	8.5
	Lipid, %	12.0	10.0	8.5
	Raw Material Cost, Mex.$ per kg diet 6/	8.63	7.11	7.28

* All metabolizable energy values are for poultry since a list of ME values of individual ingredients for fish is not complete. In setting-up energy requirements for the species, this has been taken into consideration.

^1/ Formulations carried out with IBM 370 computer at ALBAMEX, using MPSX software package.

^2/ Constraints provided by FAO Fish Feed Technologist.

^3/ See Table 14, Table of Composition of Mexican Feedstuffs.

^4/ Error in set up of constraint for this ingredient. A fix bound at the 10 percent level was omitted in the data card deck.

^5/ Error in card punch for fix bound at the 0.5 percent level

^6/ Computed from raw material prices shown in Table 3.

Carp and tilapia farming under intensive culture systems is also being encouraged in Mexico and Venezuela. In Mexico, where the livestock feed industry is large and well-developed, many of the common conventional feedstuffs are available (Table 14). Since the Mexican Government also owns the country's largest feed milling enterprise, aquaculture will have the same access to these raw materials as the already well-established livestock industry. To benefit fully from this resource, however, formulation of aquaculture diets are being carried out using linear programming techniques on an electronic computer. Details of these methods are contained in an earlier report (ADCP/REP/80/11).

Least cost diets were formulated at one of the government mills for common carp and hybrid tilapia ^1/. The constraints for the computerized formulations are shown in Table 15, and the results appear at the bottom of the table. The diets were to be processed into dry type sinking pellets.

4.2.2 Pacu (Prochilodus spp.) and Tambaqui (Colossoma spp.)

The pacu and tambaqui have been observed to feed on fruits. Because these fish grow rapidly, their culture has been of great interest, especially in Brazil and Venezuela. In both countries there is a potentially large feed resource based on the fruit and vegetable processing industry. Tables 16 and 17 describe diets formulated for testing with these species of fish in Brazil and Venezuela respectively, incorporating some of the non-conventional feed ingredients indigenous to the two countries.

Table 16

Test Diets for Pacu and Tambaqui at the Latin American Regional Aquaculture Centre (CERLA), Pirassununga, Brazil

Ingredients	Diets for fingerlings 1/					Diets for young adults 2/
	1	2	3	4	5	1	2	3	4	5
Soyabean meal	37	35	30	30	35	36.5	36.5	29.5	41.5	45
Fish meal (anchovy)	15	15	15	20	20	-	-	-	-	-
Wheat bran	47	29	24	39	24	59	24	15	40	26
Orange pulp meal 3/	-	10	10	10	20	-	20	20	15	25
Orange seed meal 4/	-	10	10	-	-	-	15	30	-	-
Bone meal	-	-	-	-	-	3.5	3.5	3.5	3.5	3
Vitamin/mineral mix	1	1	1	1	1	1	1	1	1	1
Total	100	100	100	100	100	100	100	100	100	100

^1/ Approximate protein content of fingerling diets: 33 percent

^2/ Approximate protein content of adult fish diets: 25 percent

^3/ Proximate analysis (%): dry matter, 87.6; crude protein, 6.9; crude fat, 4.6; crude fibre, 10.3; NFE, 59.8; and ash 6.0

^4/ Proximate analysis (%): dry matter, 90.0; crude protein, 30.0; crude fat, 0.9; crude fibre, 19.8; NFE, 33.6; and ash 5.7

Table 17

Test Diets for Curimbata (Prochilodus spp.) and Cachama (Colossoma spp.) at the Guanapito Freshwater Fish Culture Station (Venezuela) ^1/

Ingredients	Diets for fingerlings 2/				Diets for young adults 3/
	1	2	3	4	5	6	7	8
Fish meal	20	20	-	-	-	-	-	-
Meat meal	15	14	26	25	17	16	16	15
Blood meal	-	-	10	10	5	5	5	5
Wheat bran	59	55	58	54	52	53	53	54
Tomato seed meal 4/	-	-	-	-	5	10	15	20
Soman seed meal 5/	5	10	5	10	20	15	10	5
Vitamin/mineral mix	1	1	1	1	1	1	1	1
Total	100	100	100	100	100	100	100	100

^1/ These diets are the author's improved versions of formulations carried out by national counterpart staff contained in a consultancy report prepared for the Project VEN/79/002

^2/ Approximate protein content of fingerling diets: 30 percent

^3/ Approximate protein content of diets for adult fish: 25 percent

^4/ Proximate analysis (%): dry matter, 91.1; crude protein, 25.0; crude fat, 19.6; crude fibre, 17.0; NFE, 22.9; ash, 6.6

^5/ Proximate analysis (%): dry matter, 85.0; crude protein, 18.0; crude fat, 1.4; crude fibre, 10.9; NFE, 65.1; ash, 4.6

4.3 Diets for warm-water carnivores

4.3.1 Catfishes (Clarias spp., Ictalurus punctatus)
4.3.2 Grouper (Epinephelus tauvina), sea bass (Lates calcarifer, Dicentrarchus labrax), sea bream (Sparus auratus)
4.3.3 Other diet formulations for warm water species

4.3.1 Catfishes (Clarias spp., Ictalurus punctatus)

Because of their voracious feeding habits and capacity for rapid growth, these fishes have been popular among aquaculturists in the tropics. Although most species are considered to be carnivorous, those that presently come under artificial culture readily accept feed in which ingredients of animal origin constitute only a small percentage of the total constituents.

The nutrient requirements for fresh water carnivores are not believed to differ much from those of omnivores. In the United States, recommended dietary protein levels for channel catfish range from 28 to 35 percent. In Thailand, Clarias are successfully grown on traditional diets consisting of trash fish, rice bran and broken rice mixed in ratios varying from 3:1:1 to 10:2:1 (S. Wattanutchariya and T. Panayotan, 1982). This comes to protein levels of 25 to 38 percent on dry diet basis. Most commercially produced dry feeds for Clarias in Thailand contain 30 percent protein. Table 18 shows diets for Clarias and channel catfish formul