3. FEEDING METHODS - FERTILIZATION AND SUPPLEMENTARY DIET FEEDING (contd.)

3.2.4.3 Manure fertilization through livestock integration

The use of fresh animal manures through livestock integration necessitates that the livestock living quarters (ie. the duck house, hen house or pig stye) be located adjacent or above the fish/shrimp pond. The advantages of integrating livestock production directly with aquaculture production are many, and include:

• The nutritional value of the manure and feed remnants is preserved because losses of nitrogen and energy due to natural wastage, fermentation, evaporation and non-reversible coagulation are eliminated (Barash et. al., 1982; Wohlfarth and Schroeder, 1979; Plavnik, Barash and Schroeder, 1983).

• Wastes resulting from livestock feeding (ie. uneaten feed residues) can be directly consumed by the cultured fish or shrimp (Delmendo, 1980; Edwards, 1980; Sin, 1980; Schroeder, 1980; Nugent, 1978; Woynarovich, 1980; Plavnik, Barash and Schroeder, 1983).

• Costs of manure collection, storage and transportation are eliminated (Wohlfarth and Schroeder, 1979; Edwards, 1980; Barash et. al., 1982; Plavnik, Barash and Schroeder, 1983; FAO, 1983).

• Saving of land area (sometimes rather valuable or in short supply) otherwise needed for the manure producing livestock, and consequent improvement in land/water productivity (Edwards, 1980; Edwards et. al., 1983; Barash et. al., 1982; Plavnik, Barash and Schroeder, 1983; FAO, 1983; Shang and Costa-Pierce, 1983; Sin, 1980; Vincke, 1985).

• Providing an alternative solution to manure waste disposal on land or at sea, and thereby reducing environmental pollution (Sin, 1980; Plavnik, Barash and Schroeder, 1983).

• Improved environment for the manure producing livestock. For example, ducks benefit from life above and in the water; pond integrated animals generally displaying superior growth, feed efficiency, survival and feather cleanliness when compared with land-based ducks (Woynarovich, 1980; Plavnik, Barash and Schroeder, 1983).

• Saving of livestock feed costs due to the natural food (ie. phytoplankton and aquatic plants) developing in the pond. For example, ducks benefit directly from eating aquatic plants, and consequently help to keep the water surface clean of algal blooms and floating aquatic macrophytes (Wohlfarth and Schroeder, 1979; Plavnik, Barash and Schroeder, 1983).

• The operating costs of fish or shrimp production are reduced by the on farm supply of fertilizer (manure) and feed (uneaten feed residues) inputs (ADCP, 1979; Sin, 1980; FAO, 1983; Shang and Costa-Pierce, 1983; Rajbanshi and Shrestha, 1980; Vincke, 1985).

• The ponds provide a continuous water supply for the livestock, either as drinking water or for cleaning out their living quarters (Sin, 1980).

• Integrated aquaculture/livestock farming systems are more efficient than independent fish or livestock farms in terms of the utilization of primary resources such as fertilizers and feeds, water, labour, land and transportation facilities (Sin, 1980; Woynarovich, 1979; Shang and Costa-Pierce, 1983; FAO, 1983).

• Sources of possible income to the fish or shrimp farmer are diversified, through additional meat, hide or egg sales (ADCP, 1979; FAO, 1983; Shang and Costa-Pierce, 1983; Rajbanshi and Shrestha, 1980).

Animals which have been successfully employed in integrated livestock-aquaculture farming systems include:

• Pig - Buck, Baur and Rose (1976, 1978), Nugent (1978), Woynarovich and Kunhold (1979), Woynarovich (1979, 1980), Delmendo (1980), Jhingran and Sharma (1980), Cruz and Shehadeh (1980), Hopkins and Cruz (1980), Malecha et. al., (1981), Hopkins et. al., (1981), FAO (1983), Vincke (1985), and Edwards et. al., (1986).

• Duck - Buck (1977), Nugent (1978), Chen and Li (1980), Woynarovich (1979, 1980a), Woynarovich and Kunhold (1979), Wohlfarth and Schroeder (1979), Sin (1980), Jhingran and Sharma (1980), Cruz and Shehadeh (1980), Edwards (1980, 1982), Lawson (1981), Hepher and Pruginin (1981), Barash et. al., (1982), Plavnik, Barash and Schroeder (1983), FAO (1983), Edwards et. al., (1983, 1986), Edwards and Kaewpaitoon (1984) and Vincke (1985).

• Chicken - Burns and Stickney (1980), Djajadiredja, Jangkaru and Junus (1980), Hopkins and Cruz (1980), Wetcharagarun (1980), Edwards (1982) and Vincke (1985).

• Goose - Sin (1980), Edwards (1982); Sheep - Djajadiredja, Jangkaru and Junus (1980); Cattle - Hepher and Pruginin (1981), Edwards (1982) Water buffalo - suggestion for investigation by Edwards (1983) and Edwards et. al., (1983).

Figure 18 shows a diagrammatic representation of a typical livestock-aquaculture integrated farming system.

Figure 18. Diagrammatic representation of a pig-fish integrated farming system employed in Thailand (Edwards et. al., 1983)

The amount of manure normally produced by individual livestock classes per day and the number of animals usually kept per unit area of pond water is shown in Table 21. However, it must always be remembered that the composition of animal manure is variable (depending on total live weight, age, feed characteristics, climate, and management; Edwards, 1983) and that the density of livestock kept/unit area of pond will be different from farm to farm or area to area depending on the natural productivity of the water body, the aquaculture species employed (polyculture ratio and density) and the water quality tolerance of the species stocked. The latter point is of particular importance; air breathing catfish (Pangasius sp) and Tilapia species, generally being more tolerant to poor water quality conditions and low dissolved oxygen concentrations (such as exists in heavily manured ponds) than asiatic carps, and consequently requiring less water surface to recycle a given quantity of animal manure (Delmendo, 1980; Jhingran and Sharma, 1980, Schroeder, 1980; Hepher and Pruginin, 1981). For additional information on integrated livestock-aquaculture farming systems readers should refer to the excellent reviews of Pullin and Shehadeh (1980) and Little and Muir (1987).

Table 21. Livestock manure production rates and recommended animal stocking densities/unit area of pond water surface ¹

PIG

1) Manure production and characteristics

¹ Composition of total wastes from piglets: dry matter 2.98%, total carbon 2.72%, total nitrogen 0.4%,ammonia nitrogen 0.24%, p₂o₅ 0.10%
² Composition of total wastes from fattening pig: dry matter 6.62%, organic matter 3.34%, total carbon3.35%, total nitrogen 0.57%, ammonia nitrogen 0.27%, P₂O₅ 0.12%
³ Composition of total wastes from sow: dry matter 7.95%, organic matter 4.76%, total carbon 4.0%,total nitrogen 0.68%, ammonia nitrogen 0.24%. P₂O₅ 0.10%

2) Animal stocking densities/unit area of pond water surface

DUCKS

1) Manure production and characteristics

2) Animal stocking densities/unit area of pond water surface

CHICKEN

1) Manure production and characteristics

2) Animal stocking densities/unit area of pond water surface

CATTLE/SHEEP/GOAT/GOOSE

1) Manure production and characteristics

2) Animal stocking densities/unit area of pond water surface

¹ The animal stocking densities presented refer to fish ponds, as little or no data exists for livestock: shrimp integrated farming systems.

3.2.4.4 Manure fertilization through composting and fermentation

In many parts of the world organic manures and wastes are first biologically stabilized by aerobic composting or by anaerobic fermentation prior to application as pond fertilizers. Both these stabilization processes rely on the controlled microbial decomposition of an organic waste substrate; the former (composting) in the presence of atmospheric oxygen, and the latter (fermentation) in the absence of atmospheric oxygen (Figure 19). The rationale behind the use of these stabilization techniques is to speed up the natural decomposition process and so reduce the time lag between fertilizer application and increased natural productivity. Apart from yielding useable by-products such as heat energy (composting) and biogas (mixture of methane and carbon dixide; anaerobic fermentation), these stabilization techniques permit the use of agricultural wastes which in their natural or undecomposed state would have a low fertilizer value (ie. such as coffee pulp, sugar cane waste, rice straw, palm oil waste), facilitate the destruction of potentially hazardous pathogens and parasites which may be present in the raw waste material (ie. human faecal wastes), reduce the bulk weight of the original waste material, and also reduce the oxygen demand of the stabilized waste when applied to a water body. For general reviews on composting and anaerobic fermentation stabilization techniques readers should refer to the reviews of Gotaas (1956), McGarry and Stainforth (1978), Hauck (1978), Taiganides (1980), NRC (1981a), Biddlestone, Ball and Gray (1981), Anon (1981), Gaur (1980) and Gasser (1985).

Composting

Composting is an aerobic process in which organic manures and wastes are partially decomposed into ‘humus’ by a mixed population of micro-organisms and invertebrates in a controlled, warm, moist environment. The process flow sheet for composting is shown in Figure 20 and the overall reaction can be represented as follows:

Organic matter		CO2 + H2O + inorganic nutrients + humus	+ heat
	microbial activity	(mineralization)	(energy)

The organisms involved in the composting process are principally micro-organisms (bacteria, actinomycetes, fungi, algae and protozoa: numbers/g moist compost commonly being 10⁸–10⁹, 10⁵–10⁸, 10⁴–10⁶, 10⁴ and 10⁴–10⁵ respectively) and to a lesser extent invertebrate animals (roundworms-nematodes, potworms-enchytraeids, earthworms, millipedes, centipedes, mites, bettles and fly larvae-diptera; Biddlestone, Ball and Gray, 1981; Figure 21). During the corse of the composting cycle over 50% of the organic carbon contained within the original composted material is lost as carbon dioxide and water, with a consequent reduction in bulk weight and an equivalent increase in nutrient density (both in terms of nutrient and live animal biomass content). The advantage of composting from a nutritional viewpoint is that agricultural by-products which would otherwise have a low nutritional value to farmed fish or shrimp can be nutritionally transformed and upgraded into a potentially useable commodity for aquaculture; either as an efficient pond fertilizer or a supplementary feed source. The major microbial control parameters for optimum composting are summarised in Table 22.

Figure 19. End products of organic decay (Fry, 1976)

Figure 20. The composting process (Biddlestone, Ball and Gray, 1981)

Figure 21. Food web of the compost pile (NRC, 1981a)

Table 22. Major physico-chemical factors affecting composting

1. Nutrient content - carbon (C): nitrogen (N) ratio:

The speed of the composting process is dependent on the C:N ratio of the organic material to be composted. The desired C:N and C:P ratio is 25–35:1 and 75–100:1, respectively (NRC, 1981a; Edwards, 1982; Biddlestone, Ball and Gray, 1981; De Bertoldi et. al., 1985). If there is an excess of C compared to N in the material to be composted (ie. C:N ratio > 35) the biological breakdown process is slow as the micro-organisms must go through many life cycles, oxidizing off the excess carbon, until a more convenient C:N ratio for their metabolism is reached (the C:N and N:P ratio of microorganisms on a dry weight basis being 10:1 and 5–20:1, respectively; Alexander, 1961). By contrast, a low C:N ratio in the starting material (ie. C:N ratio > 15) would result in the loss of nitrogen from the system through ammonification (particularly at high pH and temperatures; NRC, 1981a, De Bertoldi et. al., 1985).

2. Moisture content, aeration and particle size:

The optimum moisture content for efficient composting is about 50–70%; a low moisture content causing early dehydration and arresting microbial growth, a high moisture content interfering with aeration (by clogging the pore spaces between the compost particles with water) with the consequent development of anaerobic conditions and death of the aerobic microflora and fauna present (NRC, 1981a; Edwards, 1982; Biddlestone, Ball and Gray, 1981; De Bertoldi et. al., 1985). To facilitate adequate aeration and maintain aerobic conditions the compost pile should be ‘turned over’ periodically or aeration holes inserted into the pile (ie. use of hollow plant stalks, such as Typha, Phragmites or bamboo; NRC, 1981a, Guar, 1980). According to De Bertoldi et. al., (1985) the air within the composting mass should contain 15–20% oxygen and 0.5–5% carbon dioxide to maintain adequate aerobic conditions; oxygen levels below this favouring the growth of anaerobic microorganisms. Biddlestone, Ball and Gray (1981) have estimated the optimum air flow to be 0.6–1.8m³ air/day/kg volatile solids during the rapid thermophilic stage of composting.

The particle size of the composting material will also have an important bearing on the speed of the composting process; small particle sizes favouring microbial attachment and decomposition, but also incresing the danger of over compaction and the clogging of the pore spaces with water. The most desirable particle size for efficient composting is 1.5–7.5cm, although larger particles can be composted adequately (Guar, 1980; Biddlestone, Ball and Gray, 1981).

3. Temperature

Heat generated through microbial activity ranges from 25–35°C during the early days of the decomposition process (termed the mesophilic stage) to 45–70°C during the rapid or maximum decomposition phase (termed the thermophilic stage), falling again to 25–35°C during the cooling down and maturing phase when the digestible nutrients present within the composted material have been fully exhausted by the decomposing microflora and fauna. The heat generated during the thermophilic phase of composting has the added advantage of eliminating the heat labile animal/human pathogens and parasites which may be present in the composting material; temperatures of 55–60°C for up to one day being sufficient to destroy enteric viruses, salmonellae, shigellae, Escherichia coli, Cholera vibrio, Leptospires, Hookworm ova, Ascaris ova, Schistosome ova, Entamoeba histolytica cysts, Taenia saginata, Brucella obortus and Streptococcus pyogenes (Gotaas, 1956; McGarry and Stainforth, 1978; Feachem et. al., 1980; Muller, 1980; Gaur, 1980). However, the maintenance of compost temperatures of 65°C or more for extended periods should be avoided when ever possible to avoid undue nitrogen losses and bacterial/fungal thermokill; these high temperatures may be avoided by regularly turning the compost pile or through increased ventilation. By contrast, it is essential that the heat generated through microbial activity is conserved within the compost pile to a level of about 50–60°C (during the thermophilic composting stage) to sustain rapid microbial growth and waste decomposition; within open (‘windrow’) compost piles this may be achieved by insulating the compost heap with a layer of soil, by covering the heap with a black plastic sheet, by sheltering the pile from prevailing winds, or by controlled ventilation (McGarry and Stainforth, 1978).

4. pH level:

Although organic materials with a pH ranging from 3 to 11 can be composted, the optimum range is between 5.5–8; fungi prefering acid conditions and bacteria prefering near neutral conditions for rapid growth (NRC, 1981a; De Bertoldi et. al., 1985). However, pH control through liming is not usually required during the composting process (Gotaas, 1956; Biddlestone, Ball and Gray, 1981).

5. Compost heap size:

For heaps composting under natural aeration conditions, the composting material should not be piled over 1.5m high or 2.5m wide, but can be elongated into a ‘windrow’ of any length; a width or height above these dimensions would deminish the circulation of air into the centre of the compost pile and consequently would arrest microbial activity (Biddlestone, Ball and Gray, 1981).

6. Blending or proportioning of materials for composting:

The C:N ratio and moisture content are two important factors to be considered when proportioning and blending different kinds of materials for composting. To obtain the desired initial C:N ratio and compost moisture content, materials with a low C:N ratio (animal manures, sewage sludge, slaughterhouse wastes, animal wastes, leguminous plants, green weeds, aquatic macrophytes - water hyacinth) should be mixed with materials with a high C:N ratio (sawdust, paper, straw, wood chips, sugar cane trash, coffee pulp), and wet materials mixed with dry materials (ie. mixing fresh animal manures or sewage sludge with straw, sugar cane bagass or wood chips). Dry soil may also be added to reduce the moisture content of wet materials, and applied to composts with a high organic content to buffer acid conditions and as a diluent for retarding fermentation (Gotaas, 1956). The importance of the C:N ratio of the final blended material can be seen by its direct effect on the time period required for composting; for example, under optimal conditions, blended or individual waste materials with a low C:N ratio (25–30:1) can be composted in about 4 weeks (ie. composting of water hyacinth) whereas materials with a high C:N ratio ( 100:1) require 4–6 months to be composted (ie. rice straw, sugar cane trash, coffee pulp (Gotaas, 1956; Edwards, 1982). For the C:N ratio of individual organic wastes and composts see Tacon (1987a) and Misra and Hesse (1982).

Two ‘open’ composting methods can be considered for aquaculture purposes, turned pile composting and static pile composting; the former requiring manual turning at regular 2–4 weekly intervals to facilitate compost aeration and mixing, and the latter requiring artificial aeration using mechanical air blowing or air suction techniques. Figure 22 and Table 23 gives examples of composting techniques and methods which have been tested or can be used to produce low cost organic fertilizers for pond aquaculture production systems. In addition to the above conventional composting methods, the composting process may be accelerated through the culture of specific invertebrate animals, and in particular terrestrial earthworms, within the waste substrate; a process termed ‘vermicomposting’ or ‘vermiculture’. Terrestrial earthworms have the ability to grow on a wide variety of waste streams (including food processing wastes, paper pulp, animal manures, sewage sludges, spent mushroom compost) and feed directly on the resident microbial community. By virtue of their ability to directly ingest waste particles and soil, earth worms have a stimulatory effect on the composting process, by 1) increasing the surface area of egested particles, 2) increasing oxygen penetration and circulation within the compost pile through their burrowing activities, 3) removing senescent bacterial colonies and stimulating new bacterial growth, and 4) by increasing the interaction between microflora, protozoa and nematodes, and consequently improving the flow and exchange of nutrients (Mitchell, 1979, Mitchell et. al., 1980; NRC, 1981a; Edwards et. al., 1985). For example, the decomposition rate of aerobic sewage sludge has been shown to be accelerated 2 – 5 times by the presence of terrestrial red worms Eisenia foetida (NRC, 1981a). In addition to the their stimulatory effect on the microbial composting process, the earthworm biomass represents a valuable high quality complete diet for farmed fish and shrimp; on a dry weight basis earthworms containing 55–65% crude protein and 7.5 – 12.5% crude lipid (Guerrero, 1983; Hilton, 1983; Stafford, 1984; Tacon, 1987a; Stafford and Tacon, 1984, 1985; Tacon, Stafford and Edwards, 1983). The potential of earthworm biomass production during the composting process is considerable bearing in mind that conversion efficiencies of waste to worm biomass of 5 – 10% can be achieved under optimal culture conditions (Stafford, 1984). According to Edwards et. al., (1985), the optimum culture conditions for E. foetida (a temperate earthworm species) include a moisture substrate content of 80–90%, a worm to animal waste ratio of 1: 50, a substrate pH of 5 (range 4–9), an aerobic substrate with low ammonia content, a well drained waste substrate depth of 30 –40 cm, and a temperature of 25°C. By careful management of earthworm activity, Edwards et. al, (1985) reports that separated animal manure solids and straw or wood-shavings based animal wastes can be broken down very rapidly, ranging from 2 – 4 weeks for separated solids to 2 – 3 months for straw or shavings-based animal wastes. Although the culture conditions favouring the growth of earthworms differ somewhat from the optimal conditions for conventional turned pile composting mentioned previously (Table 22), a compromise between the two processes would yield a compost with excellent pond fertilizing properties and a high quality live supplementary diet for the farmed fish or shrimp species. Other invertebrate detritivores which may also be considered during the composting process include house fly larvae and pupae (Calvert, Martin and Morgan, 1969) and Soldier fly larvae (Hale, 1973; Bondari and Sheppard, 1981; Müller (1980). It may be sufficient to apply fresh animal manure onto the outside of a maturing (ie. cooling) compost pile two weeks before the end of the composting cycle; the fresh manure would attract flies to lay their eggs in the manure and within two weeks the outer layer of the compost pile would be full of fly larvae and pupae. Clearly, considerable further research is required in this very important subject area.

Figure 22. Simple turned pile composting techniques
a) FAO (1979) b) NRC (1981a)

Compost composition: 1/3rd earth, crop stalks and horse dung/ night soil/20kg /ton superphosphate

Figure 22. Continued: Simple turned pile composting techniques

1. Elongated ‘windrow’ compost (Gotaas, 1956)
2. Indian Indore heap method (Guar, 1982)
3. Chinese high temperature compost heap (Hauck, 1978) - compost is composed chiefly of human and animal excrement and chopped plant stalks, at a ratio of 1:4. The materials are placed in a heap in alternate layers, and the moisture content maintained at an optimum level by adding water.

While making the compost heap a number of bamboo poles are inserted to serve as vents or chimneys. After the heap is formed, it is sealed with a plaster of mud 3cm thick. The bamboo poles remain in position for a day or so and then are withdrawn, leaving the holes. When the temperature rises to 60–70 °C (which develops in 4–5 days) the holes are sealed. The heap is usually turned after 2 weeks to ensure even decomposition. While turning, water, animal or human excreta are added to make up any moisture deficiency. The turned material is repiled and sealed; the compost is usually ready after 2 months (Hauck, 1978).

The pit is filled to ground level with composting materials and covered with 5cm layer of earth. Successful compost- ing mixture includes 25% human excreta and urine, 25% domestic animal and poultry faeces, earth sweepings, including brushwood and grass ashes, weeds and leaves, and 25% soil (w/w). Ventillation channel protected by millet stalks

Figure 22. Continued: Simple turned pile composting techniques.

1. Chinese ground-surface continuous composting pile (McGarry and Stainforth, 1978)
2. Chinese large pit, fully aerobic composting (McGarry and Stainforth, 1978)
3. Thailand trench composting method (FAO, 1980)

Table 23. Examples of composting methods

Anaerobic fermentation

Anaerobic fermentation is a naturally occuring biological process in which organic manures and wastes are partially decomposed by a mixed population of bacteria in the absence of oxygen. The process flow sheet for anaerobic fermentation is shown in Figure 23 and the overall reaction, representing organic matter by the glucose molecule, can be summerised as follows:

The controlled anaerobic fermentation process can be divided into two consecutive phases; the liquification phase and the gasification phase. During the liquification phase faculative bacteria break down a large proportion of the carbonaceous waste matter into organic acids, and in particular acetic acid (collectively these are termed the volatile fatty acids, VFA). The VFA are then subsequently converted by methanogenic bacteria during the gasification phase to a mixture of methane and carbon dioxide (termed ‘bio-gas’). The major physico-chemical factors affecting the fermentation process are summarised in Table 24. Compared with aerobic stabilization techniques, anaerobic fermentation is slower (at normal ambient temperatures), produces less free energy as heat (and therefore is less efficient in terms of pathogen thermokill), contains a lower standing bacterial biomass (converting only about 10–20% of the carbonaceous waste substrate into new bacterial biomass), and produces an end product (digested sludge or slurry and liquid supernatant) with a higher biological oxygen demand (Khandelwal, 1981; Baines, Svoboda and Evans, 1985; Gaur, 1980; Edwards, 1982). Despite this, under controlled conditions, the anaerobic digestion of piggery slurry can reduce the total solids content by 40%, the chemical oxygen demand by 53%, and the biological oxygen demand by 83% over a 10-day fermentation period at 35°C (Baines, Svoboda and Evans, 1985). Furthermore, apart from the obvious economic value of biogas as a combustible domestic or industrial fuel (the calorific value of biogas ranges between 20 and 26MJ/m³; Verougstraete, Nyns and Naveau, 1985), the fermentation process also produces two by-products with aquaculture fertilizer potential, namely - stabilized solids (digestedsludge or slurry) and a liquid supernatant or effluent. According to Hauck (1978) a 10m³ capacity biogas plant in China ¹ (standard household size) produces about 10m³ digested sludge/year, 14m³ digester effluent/year, and 5m³ of biogas/day (the biogas production being sufficient to supply the household with enough fuel for cooking and lighting (Figure 24). Although most digester effluents in China are used for fertilizing land crops, in some areas they are used as fish pond inputs (Edwards, 1982; FAO, 1983). Barash and Schroeder (1984) found fermented cow manure to be an effective fertilizer in Israeli fish ponds, and Khandelwal (1981) cites the use of biogas slurry as a dietary feed ingredient by ‘progressive’ fish farmers in West Bengal since 1976. It is interesting to note that in China the traditional method of direct manure application (ie. piggeries and latrines placed on or over the fish pond) is rapidly being replaced by controlled fermentation, based on prior fermentation; animal manures, mixed with plant matter and silt being anaerobically fermented for a 10-day period, and human wastes being fermented for a 4-week period within sealed digesters, before being applied to fish ponds as fertilizers (FAO, 1983). However, it is not known whether this change of methodology is directed toward improving the sanitary aspects of waste disposal (Edwards, 1984) or toward the fuel savings and social benefits which may be gained from the construction of a family biogas plant (Hauck, 1978).

¹ For information on biogas plant design and construction see Hauck (1978), McGarry (1977), Fry (1976), Taiganides (1980), McGarry and Stainforth (1978), Verougstraete, Nyns and Naveau (1985), NRC (1981a) and FAO (1984).

Figure 23. Main biochemical pathways during the anaerobic fermentation of organic wastes (Taiganides, 1980)

Table 24. Optimal physico-chemical parameters for anaerobic fermentation and biogas production

In addition to the controlled fermentation of dilute waste streams within a biogas digester, wastes can also be fermented within anaerobic composting pits on agricultural land (Hauck, 1978) or directly within the fish pond (Schmidt and Vincke, 1981; Edwards, 1982; Vincke, 1985; Viveen et. al., 1985). The latter can be achieved by building a compost crib in the corner of the pond (1m crib radius, initially 1 crib/100m² pond water surface) and stacking the organic waste materials in alternating layers under the water surface (Figure 25). Waste materials can include chopped crop waste, spoiled fruit, grass cuttings, kitchen waste and animal manure. According to Vincke (1985) for a 100m² pond about 50–60kg of organic matter is required to start with (1m³ compost pile), followed by weekly doses of 8–10kg organic matter. In rural fish ponds in the Central African Republic fish yields of 1500kg/ha/year (T. nilotica) have been obtained using the above anaerobic composting technique (Schmidt and Vincke, 1981). The same authors also state that to collect, transport and pile the organic matter for composting a 100m² pond, a farmer may spend an average of 28 hours a year. In China, Delmendo (1980) reports that compost (anaerobically produced) is applied to fish ponds at levels ranging from 5 to over 10 tons/ha/year.

The value of composting at the rural or subsistence farming level is that it is simple to operate (requiring little training), requires only part-time labour inputs, and utilises locally available waste products at little or no cost to the farmer.

Figure 24. The ‘Biogas’ cycle in China (Hauck, 1978)

	Pit covered with mud and a 3–4cm water column to create anaerobic conditions
	1. Green manure (legumes) or water plants
	2. Silt or straw mixture
	3. Stable manure (pig)

Figure 25. Compost crib and anaerobic composting techniques.

1. Bamboo compost crib for a 100m² fish pond (FAO, 1979)
2. Compost crib filled with layers of grass alternating with fresh wastes such as kitchen offal (Viveen et al., 1985)
3. Green manure compost frame filled with green plant cuttings (2–4 frames/200m² pond; Edwards and Kaewpaitoon, 1984)
4. Chinese anaerobic composting pit (Hauck, 1978)

3.3 Supplementary diet feeding

In addition to the use of fertilizers for the production of natural food organisms within the water body, an external diet can also be fed as a ‘supplementary’ source of dietary nutrients for the cultured fish or shrimp; the dietary nutrient requirements of the farmed species being supplied by a combination of natural live food organisms and supplementary diet feeding. The relative importance of natural food organisms and supplementary feeds in the nutrition of fish and shrimp within extensive, semi-intensive and intensive pond culture systems is shown in Figure 26. The advantage of combining supplementary diet feeding with pond fertilization is that it permits the use of higher fish and shrimp stocking densities, facilitates faster fish and shrimp growth, and consequently results in higher fish and shrimp yields and crops over the growing season. For example, Sinha (1979) reports fish production (Indian carp/ Chinese carp polyculture) in ponds in India to be 1053kg/ha/year with no fertilizer or feed inputs, 1398–2303kg/ha/year with inorganic and organic fertilizer inputs (cowdung and 18:8:4 NPK fertilizer), 3314–4005kg/ha/year with supplementary feed inputs (1:1 mixture of rice or wheat bran and groundnut or mustard oilcake), and 4244–5506kg/ha/ year with both fertilizer and supplementary feed inputs.

Figure 26. The role of natural food organisms and artificial feeds in the nutrition of fish and shrimp within extensive, semi-intensive and intensive pond culture systems (Tacon, 1987)

It must be emphasised at the outset that the benefits of supplemental feeding will depend upon the composition and physical form of the feed used, the stocking density of the fish or shrimp species cultured, and the natural productivity of the water body in question. Furthermore, each pond ecosystem must be considered as being unique (depending on climate, location, soil type, water quality and fertilizer input) and so itshould be remembered that the success of a particular supplementary feeding regime in one location need not necessarily apply to another. This is in sharp contrast to the intensive culture system where strict controls are normally placed on water quality and feed inputs.

3.3.1 Selection of supplementary feeds for use by rural or subsistence farmers

In view of the general shortage of conventional feed ingredients in rural communities for human and livestock consumption, and the low cash income and purchasing power of rural subsistence farmers, feed selection must be based on the following criteria (Tacon, 1986b): in order of importance,

• Cost - the feed material should be available at little or no cost to the farmer.
• Availability - the feed material, when ever possible, should be available throughout the year.
• Handling and processing - the handling and processing requirement prior to feeding, including transportation, should be minimal or negligible.
• Nutritional value - feed materials with a high protein and low fibre content having a higher nutritional value than feed materials with a low protein and high fibre content.

Furthermore, by utilizing low quality and value products, and in particular those agricultural and industrial by-products which are not currently used for human and livestock feeding, aquaculture could be seen to be an asset to the community by increasing land productivity rather than a competitor with the traditional agricultural and live-stock farming activities.

Feed materials which could be considered for use as supplementary feeds at the rural farming level include:

Kitchen/cooking waste - uneaten food and household scraps
Beer waste - spent draff (grains) and yeast
Spoiled/contaminated animal feeds
Pasture/arable crop waste - leaves, stems, roots, tubers, peelings and seeds
Aquatic macrophytes - floating and emergent plants¹
Mill sweepings
Abattoir waste - animal offal, blood and rumen contents
Fruit wastes - peelings, damaged fruit, and pulp
Oilseed hulls and residues - hulls, brans and low value expeller cakes
Cereal grain hulls and residues - hulls and brans
Sugar-cane wastes - bagass (ascompost), molasses and filter press cake
Terrestrial invertebrates animals - earthworms, snails, and insects
(including larvae and pupae)
Aquatic animals - chironomid worms, polychaetes, chopped frogs, toads and tadpoles, crustaceans, and unwanted fish

¹ The use of aquatic macrophytes as a supplementary feed source will be discussed in a separate manual with this series (Arrivillaga Cortez, 1988 - In prep).

The reported food quotients (food conversion ratio; amount of food required per unit increase body weight) of some commonly used supplementary fish feeds is shown in Table 25.

Table 25. Food quotients of some common supplementary fish feeds ¹

¹ From Ling (1967)
² Food quotients for herbivorous fish species in China (Yang et.al., 1985)

3.3.2 Feed formulation and natural productivity

At present almost 95% of the available information on the dietary nutrient requirements of the major cultivated aquaculture species is derived from ‘laboratory based’ nutritional feeding trials; the animals (generally fingerlings or post-larvae) usually being housed within indoor artificial tanks at high density and subjected to controlled water quality conditions with no access to natural food organisms. To date the majority of these studies have been conducted within nutrition laboratories from North America, Europe and the Far East. Although commercial aquaculture production within these temperate industrialised countries is usually realised within intensive ‘clear water’ aquaculture production system (ie. cement tanks or raceways, and cages suspended in open water bodies), over 90% of aquaculture production within developing and third world countries (including Latin America and the Caribbean) is conducted within tropical or subtropical semi-intensive and extensive pond production systems. Consequently, whilst published information on dietary nutrient requirements may suffice for the formulation of complete pelleted feeds for use within intensive ‘clear water’ aquaculture systems, this information cannot be directly applied to the formulation of rations for use within semi-intensive or extensive pond aquaculture production systems.

In contrast to ‘complete’ diet feeding, where rations are formulated to a pre-set nutrient level for each fish or shrimp age class, the formulation of a supplementary diet is dependent upon the standing crop (ie. total biomass) of the fish or shrimp species present and the consequent availability of natural food organisms within the water body per animal or species. For example, no difference was observed in the growth of common carp (C. carpio) in earthen ponds in Israel when fed a cereal grain (ie. sorghum) or a formulated artificial ration containing 22.5% protein up to a standing crop of 800kg carp/ hectare, or between a ration containing 22.5% protein and a ration containing 27.5% protein up to a standing crop of 1400kg carp/hectare (Hepher, 1979). Similarly, no difference was observed in the growth of prawns (M. rosenbergii) in outdoor concrete ponds in Thailand when fed a 35% protein diet, a 15% protein diet or a broiler starter feed (stocking density 5 animals/m², pond water changed every three weeks; Boonyaratpalin and New, 1982), or in the growth and survival of shrimp (P. vannamei) in earthen ponds in Hawaii receiving no fertilizer and feed input, fertilization input only, or fed a commercial pelleted shrimp ration (stocking density 7.1–9.4 animals/m²; Lee and Shleser, 1984). Furthermore, no beneficial effect of dietary vitamin or mineral supplementation was observed with tilapia or common carp either in earthen ponds or cages (within the pond) at stocking densities of 2/m² and 100/m³ respectively (c. 400g fish; S. Viola, personal communication, Ashrat, Israel, 1985). The apparent non-essentiallity of dietary vitamin fortification under practical semi-intensive farming conditions has also been observed with freshwater prawns (Boonyaratpalin and New, 1982) and for filter feeding tilapia species reared in concrete tanks (Wee and Ng, 1986; Dickson, 1987) or in floating cages (Campbell, 1985; Guerrero, 1980; Pantastico and Baldia, 1979; Wannigama, Weerakoon and Muthukumarana, 1985). It is important to re-emphasise here that many fish species have the ability to filter fine particulate matter (ie. phytoplankton and detritus) from the water column (Bowen, 1982). For example silver carp (H. molitrix) and tilapia (O. mossambicus) were reported to grow from 15g to 260g and from 10g to 130g, respectively, over a 190-day period without artificial feeding in floating cages within fertilized ponds (Gaigher and Krause, 1983). Similarly, Cremer and Smitherman (1980) reported a weight increase from 22g to 270g for silver carp (H. molitrix) and 13g to 133g for bighead carp (A. nobilis) over a 159-day growth period without artificial feeding in floating cages housed within fertilized ponds. According to these authors, silver carp and bighead carp can filter particles from the water column as small as 8μm and 17μm, respectively (Cremer and Smitherman, 1980). In view of this unique filtering ability, it is perhaps not surprising that Wannigama, Weerakoon and Muthukumarana (1985) found no significant difference in the growth rate and feed efficiency of cage reared fingerling tilapia (O. niloticus) when fed a 29% protein diet or a 19% protein diet (containing 92% chicken mash) at stocking densities of 400, 600 and 800 fish/m³.

Apart from the classic studies of Balfour Hepher and his research colleagues in Israel, and the recommended supplemental vitamin allowances of NRC (1977, Table 26) very little information is available concerning the quantitative dietary nutrient requirements of fish or shrimp under semi-intensive or extensive pond rearing conditions. The importance of natural food organisms in the overall nutritional budget of pond reared fish or shrimp cannot be under stressed, especially during the early growth cycle of the cultured species and at low fish/shrimp stocking densities when total fish/shrimp biomass/m² is low and natural food availability per animal is high (Figure 27). From Figure 27 and the above discussion it follows that dietary energy is generally the first limiting factor at low fish/shrimp stocking densities (when natural food availability per fish or shrimp is high), whereas at high stocking densities and standing crops dietary protein and other essential nutrients become limiting and therefore have to be supplemented. From a farming viewpoint it can be seen that the protein and nutrient content of an artificial diet (intended for use within a semi-intensive pond culture system) will have to be progressively increased with increasing fish/shrimp biomass and standing crop, and decreasing natural food availability (Hepher, 1975; ADCP, 1984; Tacon, 1985; Figure 28). This relationship is the complete reverse situation of an intensive complete diet feeding strategy, where dietary nutrient levels usually decrease with increasing fish/shrimp weight and age (Tacon, 1987). Sadly, in the absence of published information, the majority of researchers, feed manufacturers and farmers alike still employ a gradually decreasing dietary nutrient density for the semiintensive culture of pond fish and shrimp; the diet in turn usually being formulated as a nutritionally complete pelleted ration with no allowance provided for natural food availability. Clearly, this situation must be remedied if maximum economic benefit is to be gained from semi-intensive pond aquaculture feeding strategies.

Figure 27. Theoretical relationship between fish/shrimp growth and the availability of natural food organisms within a semi-intensive pond aquaculture culture system

Table 26. Recommended allowance for vitamins in supplemental and complete diets for warm water fishes ¹

¹ Source: National Research Council (1977)

Proportion of sorghum to protein-rich pellets fed to carp/tilapia with increasing standing crop

Feeding strategy for a well populated and fertilized carp pond. During the early part of the culture period, the protein requirement of the fish is met by natural food, thus only supplemental energy is provided as wheat. For production targets below 1 ton/ha/year artificial feeding is unnecessary (common carp 20–30% population), for a production of 2 ton/ha/year (as common carp) feeding is necessary with grains, and productions of 3 ton/ha/year and above require the use of pellets containing additional protein, vitamins and minerals (50–70% population as common carp).

Figure 28. Examples of practical semi-intensive pond feeding strategies

1. For carp/tilapia polyculture - Israel (Hepher and Pruginin, 1981)
2. For carp polyculture - Hungary (ADCP, 1984)

The contribution of natural pond food organisms and artificial feeds to the overall nutritional budget of pond reared fish and shrimp may be determined by measuring the relative abundance of the stable isotopes of carbon 13C/12C (reported as δ C) within the natural pond biota, the external diet fed, and the cultured fish or shrimp before and after a pre-determined growth period; the δ C ratio of the fish/shrimp body tissue being directly related to its food intake. For a detailled discussion and review of the principles and use of stable isotopes to assess the flow of carbon and nitrogen in the aquatic food chain see Schroeder (1983, 1983a, 1983b), Anderson, Parker and Lawrence (1987), Anderson et. al., (1987) and Shan et. al., (1985). For example, according to the studies of Lilyestrom and Romaire (1987) and Anderson, Parker and Lawrence (1987) with pond reared prawns (M. rosenbergii) and shrimp (P. vannamei), natural pond biota accounted for 18–75% and 53–77% of the growth of the cultured species, respectively; in both studies animals were fed formulated artificial pellets.

3.3.3 Feed preparation and presentation

The success of a particular supplementary diet feeding regime will depend to a large extent upon the physical form of the diet fed (dry/wet mash or pellet) and the cost of the finished feed. At its simplest level feed presentation merely involves administering the feed in its fresh or ground state to the pond. This feeding strategy is most suited to ponds having a low stocking density (or standing crop) and high natural productivity. However, at high stocking densities there is no doubt that pelleted feeds (dry or moist) are more beneficial and economic in terms of feed conversion efficiency and growth. At low stocking densities however the beneficial effect of pelleting may not be so great. For example, in the Central African Republic no difference was observed in the growth response of pond reared T. nilotica fed a 30% protein supplementary diet either in ground or pelleted form (Miller, 1979). During this 62-day feeding trial low stocking densities were employed (2/m²; initial and final fish standing crop of 480 and 1200kg/ha, respectively) and fish fed at 4% of their body weight once daily (0900h). In view of the exceedingly high cost of the pelleting process reported, feed input cost/kg fish gain using the pelleted feed was found to be almost double that of the ground meal. Under these circumstances the additional costs of pelleting were not compensated by an equivalent increase in fish production. Clearly, special attention must be paid to the physical form of the supplementary diet used and the economic cost of processing.

For information on conventional feed preparation techniques, including pellet manufacture, readers should refer to the excellent review of New (1987) and to section 2.3 of the present manual. Figure 29 shows examples of two simple solar driers which can be used for drying wet feed materials (including moist pellets) prior to feeding, and Figure 30 presents examples of feeding methods currently employed for administering supplementary feeds to pond reared fish.

3.3.4 Feeding level and frequency

In contrast to complete diet feeding no ‘universal’ feeding tables exist for the use of supplementary feeds; supplementary diet feeding tables varying with the composition of the diet used, natural food availability, water quality (dissolved oxygen concentration and water temperature), and fish/shrimp species, age, stocking density and standing crop. Since natural food organisms play a gradually diminishing role in the overall nutritional budget of cultured species as the standing crop increases within the pond with time, it follows that the proportion of supplementary feed fed/unit body weight should be gradually increased over the course of the farming cycle. Sadly, however, the majority of farmers and researchers still employ a decreasing dietary feeding rate with time; once again, in the absence of published information to the contrary, a legacy from complete diet feeding regimes. Clearly, this situation must be remedied. For example, at the species level it has been suggested that the supplementary feeding rate for tilapias in ponds should be lower than that for common carp (Hepher and Pruginin, 1982). Similarly, the poor fish performance observed by Miller (1979) when feeding a pelleted ration may have resulted from the feeding regime employed (fish fed only once daily) and the poor water stability of the pelleted ration used (disintegration after 5 minutes of water immersion). Clearly, optimal feeding rates and frequency of feed presentation must be determined for individual feeds and farms. Figure 31 and Table 27 give some examples of dietary feeding rates and regimes which have been employed for the semi-intensive culture of selected pond fish and shrimp.

The AIT solar drier prototype shown is capable of drying 80kg of rice paddy in 2–3 days. The unit is built on a mound of earth. The air heater consists of a layer of burnt rice husks, to absorb solar radiation, and a clear plastic cover on a simple wooden frame-work to form an air duct lm wide and 10cm deep. The duct faces the prevailing wind and the warmed air passes through a shallow bed of paddy resting on a wire-mesh floor made of mosquito netting. The chimney box above is enclosed by clear plastic walls to keep the air inside warm.

The advantage of drying materials within a solar drier instead of the open air is that the drying process is much more rapid, the material is protected from rainfall, and vitamin losses are greatly reduced (the vitamins riboflavin, pyridoxine, folic acid, and ascorbic acid being easily oxidised on direct exposure to light/UV irradiation). One of the practical advantages of a solar drier is that liquid waste materials (ie. blood, rumen contents, brewery yeast slurry) can be progressively adsorbed onto dry feed materials (ie. middlings, cereal brans, ground corn cobs) and the resultant semi-moist feed mashes rapidly dried without spoilage (see also Schmidt and Vincke, 1981).

Figure 29. Examples of solar driers suitable for small-scale drying of feeds

1. Asian Institute of Technology solar drier (Excell and Kornsakou, 1978)

2. From New (1987)

Figure 30. Examples of supplementary diet feeding techniques

1. Manual feeding by broadcasting over water surface (FAO, 1981)
2. Manual feeding by placing floating feed items (ie. grass, green fodder plants, chopped aquatic macrophytes, dry powders-rice bran, floating pellets) into a floating or fixed surface bamboo frame (FAO, 1981)
3. Manual feeding by placing sinking feed items on a submersible feeding tray (Bay of Bengal Programme, 1984)
4. Use of electric light bulbs above pond surface to attract flying night insects (Edwards and Kaewpaitoon, 1984)
5. Tying bundles of plant/crop stems/leaves and fixing to pond bottom to encourage growth of attached microflora and fauna (Woynarovich, 1985)

Figure 30. Examples of supplementary diet feeding techniques (Cont.)

f) Demand feeder - bait rod feeder (Pitt, 1986)
g) Demand feeder - plastic/fertilizer bag feeder (Pitt, 1986)
h) ‘Farm Pond Harvest’ feeder with wind baffle

3.3.5 Economics of supplementary diet feeding and pond fertilization

Finally, but not least, simple economic analyses must be undertaken to ascertain the profitability of a particular supplementary diet or fertilization feeding strategy. For example, Table 28 shows a simplified economic analysis of two identical fish farming units, one receiving manure as the sole feed input and the other receiving pelleted feed as the sole feed input. For additional information see section 2.5 of this manual.

Pond feeding strategy employed for the culture of Tambaqui (C. Macropomum) in Brazil (Woynarovich, 1985)

Practical slide rule for common carp feeding in Israel (Marek, 1975)

Figure 31. Examples of practical semi-intensive pond feeding practices used in Brazil, Israel and China

Typical Chinese integrated fish/agriculture/livestock farming system (FAO, 1983; for specific examples see Edwards, 1982).

Figure 31. Examples of practical semi-intensive pond feeding practices used in Brazil, Israel and China (Cont.)

Table 27. Examples of practical dietary feeding strategies employed for the semi-intensive culture of pond fish and shrimp

FISH

1. Supplementary diet feeding table for common carp within fertilized ponds in Israel (Marek, 1975): ¹

¹ First number g 25% protein pellet/fish/day, second number g sorghumgrains/fish/day. The above feeding table refers to water temperaturesabove 24°C; for water temperatures of 20–24°C and 18–20°C use 70%and 50% of the above quantities, respectively.

2. Feeding regime for carp fingerlings in China (Pagan-Font and Zimet, 1980):

3. Feeding regime for common carp broodstock in Asia (Jhingran and Pullin, 1985):

Bangladesh - broodstock kept in fertilized ponds. Supplementary feeding is usually at 3% body weight with mustard oilcake:wheat bran (1:1); the mustard oilcake is mixed with water (2:3), soaked for 24h, and then this mixture is made into food balls with the wheat bran; spent fish transferred to well manured ponds and fed at 10% body weight/day during recovery.

Burma - some pond fertilization, but heavy reliance on supplemental feeding when natural feeds are in short supply; usually various polyculture species combinations. Various supplemental feeds and feeding rates with rice bran, peanut oilcake and chopped vegetation; eg. peanut oilcake: rice bran (1:2) plus equal volume of chopped green fodder (grasses or water hyacinth) during maturation, most spp. receive 3–4% body weight/day reducing to 1–3% in the prespawning period.

India - kept in manured ponds with supplemental feeding; various polyculture combinations; total stocking density 1000–2000kg/ha. Supplemental feeding at about 1% body weight/day with rice bran:oilcake (1:1).

Indonesia - in manured ponds but with heavy reliance on supplemental feeding; stocking density, usually 2000 /ha (some farmers use up to 3300 /ha but this is too crowded); individual fish weight are 300g – 2kg, 1–4kg. Rural farmers feed rice bran mixed with fresh vegetation, waste palm oil and waste groundnut oil; government and private hatcheries feed pellets at 2–3% body weight/day; pellets contain 20–25% protein and maximum 8% fat; a typical pellet mix is rice bran, 50%; fish meal, 25%; leaf meal, 12%; vitamin, mineral and antibiotic premix, 1%.

Pakistan - all species kept in manured/fertilized ponds with supplemental feeding. Various supplemental feeds depending on local availability; typical feed contain 30% maize, 30% rice, 20% horse gram, 20% cotton oilcake; some hatcheries use 20% fish meal from trash marine fish or tilapia fingerlings grown on site.

Philippines - all species kept in ponds and concrete tanks. Are fed rice bran and molasses or rice bran plus copra meal (1:1) or rice bran alone, all at 5% body weight/day.

Taiwan - all species kept in manured ponds. Various supplemental feeds - soybean cake, rice bran and peanut cake.

Thailand - all kept in manured/fertilized ponds; stocking densities one fish (2–4kg)/20–30m². Fed 25% protein fishmeal-based pelleted feeds 30–40 days up to expected spawning at various rates.

Sri Lanka - kept in manured ponds but also heavy reliance on supplemental feeds; monoculture 4000 /ha stocking density, mixed sexes. 1–2% body weight/day of rice bran: coconut residue cake (1:1) plus sometimes earth worms or silkworm pupae; 1–2% body weight of a 60% rice bran, 35% coconut residue ke, 5% fish meal feed.

Vietnam - all species kept in manured ponds; sometimes alone with sexes segregated, stocking 1kg/5–8m² or 1kg/10–20m². Supplementary feeding at 5–7% body weight/day with various feeds depending on local availability of materials; usually balanced 10–30% protein: 70–90% carbohydrates; a good feed is rice bran, 70%; fish meal, 5%; soybean cake, 12%; wheat flour, 10%; fish sauce waste, 3% plus microingredients in mg/kg dry food, CuSO₄ 4, KI 1, MnSO₄ 2, CaCL₂ 1–5. During the last two months before spawning fish also receive 1–2% body weight/day of germinated rice (assumed beneficial because of high vitamin E).

4. Feeding table for the African catfish (C. gariepinus) in stagnant ponds fed a 30% digestible protein pellet (Viveen et. al., 1985):

Initial fish stocking density 10/m² (1–2g fish)

5. Feeding table for grey mullet (M. cephalus) in Hong Kong (Ling, 1967): ¹

¹ Average stocking rate of 10,000 mullets and 1500 Chinese carps/ha.Quote average net fish production of 3500kg/ha over a 300-day cultureperiod; this production requiring 2500kg rice bran, 3000kg peanut cake,and animal manure application at 3 to 5 day intervals.

6. Feeding strategy for the freshwater prawn (M. rosenbergii) in fertilized ponds in Panama (MIDA, 1984):

SHRIMP

7. Nursery feeding schedule for P. monodon (Bay of Bengal Programme, 1984):

In Malaysia PL_5–6 are stocked at an initial density of 4/m², and for the first 7 days fed on 100% minced trash fish at a feeding rate of 20–40% body weight/day. The minced trash fish was scattered over the shallow part of the pond in specific locations by placing on feeding trays (see Fig. 30c). From the 8th to 30th day shrimp fed a diet consisting of 40% rice bran and 60% trash fish; the diet being mixed together and formed into a wet pellet and fed once daily during the evening. After the 30th day of the nursery culture cycle the animals are fed the same ration as above plus whole trash fish; the daily feeding rate being 20% from day 8–30 and 10–15% from day 30.

8. Recommended feeding rates for commercial shrimp pellets in Peru (Sales literature of Nicolini Hermanos S.A., Lima, Peru):

40% protein diet used for shrimp of 1 – 2g, 38% protein diet from 3 to 12g, and a 35% protein finisher diet from 13 to 25g.

9. Recommended daily feeding rates for commercial shrimp pellets in Ecuador (Sales literature of Nutril S.A Balanceados, Guayaquil, Ecuador):

Feeding rate described is for a 30% protein diet; at stocking densities above 8/m² a 35% protein diet is recommended.

10. Recommended daily feeding rates for commercial shrimp pellets in Brazil (Sales literature of Purina, Brazil):

Feeding rate described is for a 25% protein diet, and feeding once or twice daily.

Note 8. - 10. Reference to a commercial diet or brand name does not express any recommendation of the author (citation used for example purposes only).

Table 28. Profitability of two feeding systems in a pond environment using identical fish stocking and management techniques ¹

¹ Example solution based after Wohlfarth and Schroeder (1979).