Field Document 3
June 1988


prepared for the project
Development of marine culture of fish


Pinij Kungvankij
Consultant (Hatchery Management)

This report was prepared during the course of the project identified on the title page. The conclusions and recommendations given in the report are those considered appropriate at the time of its preparation. They may be modified in the light of further knowledge gained at subsequent stages of the project.

The designations employed and the presentation of the material in this document do not imply the expression of any opinion whatsoever on the part of the United Nations or the Food and Agriculture Organization of the United Nations concerning the legal or constitutional status of any country, territory or sea area, or concerning the delimitation of frontiers.

Rome, 1988

Hyperlinks to non-FAO Internet sites do not imply any official endorsement of or responsibility for the opinions, ideas, data or products presented at these locations, or guarantee the validity of the information provided. The sole purpose of links to non-FAO sites is to indicate further information available on related topics.

This electronic document has been scanned using optical character recognition (OCR) software. FAO declines all responsibility for any discrepancies that may exist between the present document and its original printed version.



1.1 Terms of Reference
1.2 Background Information


2.1 Holding Tanks
2.2 Seawater Intake System
2.3 Water Supply System
2.4 Aeration System


3.1 Broodstock Development
3.2 Collection of Wild Broodstock

3.2.1 Capture of Spawners
3.2.2 Transporting Spawners
3.2.3 Determination of Sex and Maturity of Spawners
3.2.4 Selection of Suitable Broodstock

3.3 Broodstock

3.3.1 Broodstock Management
3.3.2 Factors Affecting Gonad Development

3.4 Spawning and Fertilization

3.4.1 Selection of Spawners
3.4.2 Care of Spawners in Pre-spawning Tanks

3.5 Spawning

3.5.1 Artificial Fertilization
3.5.2 Induced Spawning

3.6 Fertilization and Incubation
3.7 Larval Rearing
3.8 Factors Affecting Mass-Rearing of Marine Finfish Larvae
3.9 Rearing Environment
3.10 Feed and Feeding


1. Seawater intake through sump pit

2. Water system

3. Feed and feeding scheme for grey mullet



The Government of the People's Republic of China, assisted by the United Nations Development Programme and the Food and Agriculture Organization of the United Nations, is engaged in the project Development of marine culture of fish (CPR/81/014), whose main objectives are to develop the culture techniques including seed production of marine finfish in the coastal areas of China.

As part of the project operations, FAO assigned Mr Pinij Kungvankij as hatchery management consultant from 10 to 23 January 1988 and 29 January to 12 February 1988.

During the course of his assignment the consultant prepared this Guide.


Mugil cephalus Linneaus, commonly called the grey mullet, has been traditionally cultivated in the Mediterranean, Japan, Hawaii and China for more than a century. This extensive culture of mullet has made heavy demands on the supply of fry. At present, mullet fry come from the wild, and there are already signs of a shortage, particularly in China. It is thus becoming necessary to supplement the supply through artificial propagation.

Research into the artificial propagation of mullet has been carried out since the 1960s. In 1966 a comprehensive research programme was established at the Tungkang Fisheries Research Institute (Taiwan) for intensive culture and mass production of fry through artificial propagation. In the subsequent years the techniques of induced spawning of mature fish caught from the wild during their seasonal migration were refined and improved, and the survival of the larvae was extended to 30 days in 1973, a survival rate increase of 19.35%. While research on mullet was concentrated in Taiwan, work on the same species was also carried out in the USA mainly centred in Hawaii (Oceanic Institute), Texas and Louisiana.

China has a well-developed freshwater aquaculture, while its coastal aquaculture is still in the developing stage. However, the research findings and technology established in these places can be transferred and modified into a technology appropriate for the local conditions, the farmers being trained accordingly.

This manual draws on information provided in the report “Hatchery management techniques in marine fish culture development”, CPR/81/014, Field Document 1.


To ensure the smooth and effective implementation of operations at the marine finfish hatchery, the following facilities were required:


Holding tanks in the hatchery are used for various purposes such as for broodstock-conditioning, and subsequent spawning, incubation, larvae-rearing and production of natural food.


Seawater can be drawn directly from the sea or from the sump pit. If the water is relatively clear, it can be pumped directly into the overhead filter tank and stored in the reservoir or storage tank. Water is then gravity-fed to various culture tanks through delivery pipes. However, if the water is turbid and contains a high concentration of suspended solids, it must be pumped first into a sedimentation tank where the suspended solids are allowed to settle; only the clear upper stratum of water is pumped into the filter tank. In some areas where the source of water is far from the shoreline and during low tide when large quantities of water are needed continuously, the sump pit or tube well can be constructed inshore near the hatchery. The sump pit is connected to an underground pipe which is situated near the water source. The water continuously enters the sump pit through the underground pipe even during low tide. Water is then pumped directly from the sump pit or tube well (Fig. 1); this water is usually clear because it is filtered naturally through a layer of sand before entering the pipe so that it can be used directly. However, if very clear and clean water is required, it should be pumped through the filter tank before use.


The marine finfish hatchery requires different qualities of seawater for its specific purpose. It is essential to use the combined filtration system in the hatchery. Direct seawater can be used for the broodstock tank; while filtered seawater is necessary for larvae-rearing and culture of live food organisms (Fig. 2).


Aeration is essential during the entire larva-rearing process in order to maintain a sufficiently high dissolved oxygen concentration in the water, ensuring an even water temperature throughout the water column through turbulence and also helping to reduce the ammonia content in the water.

In culture tnaks of less than 2 m depth, air pressure of about 0.2–0.3 kg/cm and a volume of 4–5 1/m/min is sufficient to oxidize the dissolved organic matter in the tanks.

Since continuous aeration is essential for the survival of larvae stocked at high densities, any prolonged power interruption would seriously affect the culture organisms in the tank. Thus, it is essential to install an automtic switch which starts a standby generator whenever there is a power failure. A battery-operated awrning device to signal the crisis and the required operation of the standby generator should also be installed.



There are two sources of finfish broodstock: wild-caught adults and those reared in ponds or cages. It is advantageous to use pond or cage-reared broodstock as they are already used to culture conditions and are thus easier to develop into broodfish.


3.2.1 Capture of Spawners

Broodfish migrating towards the coastline or brackishwater areas, lakes and estuaries during their spawning season, are caught either by permanent fish traps, such as the otoshi-ami net or fish corrals, or by gillnetting or purse-seining. They are then transported to breeding stations where they are induced to spawn by hypophysation or by environmental manipulation.

3.2.2 Transporting Spawners

There are several ways of transporting spawners: from the most simple receptacle, such as plastic bags, to the most sophisticated, such as vats or special transport vehicles. Containers vary with the size, species and number of fish to be transported and the location of the collecting grounds. Short-distance transport

Large fish, e.g., seabass, grouper, caught in permanent traps near the breeding station, are transferred to a floating cage by means of a fine meshed scoop-net. The cage is then slowly pulled towards the shoreline with a pumpboat. When only 1 or 2 fish are caught, or in the case of small fish like mullet, these could be transferred directly to double plastic bags provided with water and aeration. The plastic bags containing the fish are then transported to the shoreline aboard a pumpboat. Upon reaching the shoreline, the fish (still in plastic bags) are brought on a stretcher to the experimental tanks. One can stock 5–6 large fish at a time in a 1.5 × 1.5 × 3 m floating cage but only one large fish could be transported in a double plastic bag of length 1.5 m and diameter 0.4 m. In the case of smaller fish like the grey mullet, 4–6 fish can be transported in a double plastic bag at a time. In other places, instead of a plastic bag, rubberized canvas is used to transport small fish. Long-distance transport

Fish caught in traps far from the breeding station are transferred to a 2-m diameter canvas tank one-third full of water and installed aboard a pick-up truck or in a fibreglass tank lined with rubberized canvas and one-third full of water. In both cases, the upper end of the canvas is tied with a rope such that only a very small opening is left to allow for the plastic tubing from the oxygen or compressed air tanks. This prevents the fish from being directly exposed to intense sunlight, minimizes evaporation, and prevents accumulation of dust or dirt during transport. The water is continuously aerated during transport either with compressed air or oxygen.

Long-distance transport of spawners by sea could be achieved by using a pumpboat. A canvas tank could be installed on a pumpboat or the fish could be placed directly into the hull of the pumpboat. Water is continuously aerated during transport. In the absence of aeration a third to a half of the water is changed every 30–60 min depending upon the number of fish being transported.

In highly developed countries, vats and special vehicles are used for long-distance transport of fish.

Vats. These are made either of metal or plastic, and are shaped like truncated cones or cylinders. They rest on their flat end which is either round or oval in circumference. They are generally used for transporting fish with oxygen.

Special vehicles. Large-scale fish-farmers usually have trucks mounted with tanks of different shapes. These tanks are generally made of metal; and sometimes they are covered with insulating materials. During transport, they are provided with aeration. There are several ways in which oxygen or compressed air is introduced into the tanks during transport. The more sophisticated the system is, the more expensive it becomes. So, money sets the limit for sophistication.

3.2.3 Determination of Sex and Maturity of Spawners

Two common aspects in the artificial propagation of finfish are the determination of sex and the maturity of spawners. Often, it is difficult to determine the sex of spawners through examining the external morphology of the fish. In some species, howwever, e.g., mullet, a gravid female exhibits a fuller profile than the ripe male; its abdomen is distended. Ripe males are easy to distinguish during the spawning season since milt oozes out from the urogenital pore as its abdomen is pressed. If the degree of maturity is right, the milt will be white and creamy; poor milt is watery and curdled. Milt which is not ripe will demand strong pressure and will be mixed with blood.

Assessment of gonadal maturation of broodstock is still a major difficulty in the artificial propagation of finfish. To date, the commonly-used method involves in vivo sampling of gametes. Gametes are removed from either an anaesthetized or unanaesthetized fish by using a polyethylene cannula. The inner diameter of the cannula to be used varies with the size of eggs to be sampled. The cannula is inserted 4–15 cm into the ovary or testis and gametes are drawn into the cannula by aspiration as the cannula is slowly withdrawn. The distance to which the cannula is inserted varies with the length of the ovary or testis. Samples from the middle portion, especially of the ovary, are generally considered to be the most representative.

The eggs collected are removed from the cannula by blowing them into a Petri dish. They are preserved in 1% formalin in 0.9% NaCl. The average egg diameter is determined from a batch of 50–10 by using a micrometer and their developmental stage is assessed under the microscope. Gonadal maturation is then expressed in terms of average egg diameter and the developmental stage of the eggs.

The milt collected is removed from the cannula by blowing it onto a clean dry Petri dish. A small portion of this is mixed with a drop of seawater or brackishwater, depending upon the species, and examined immediately under the microscope. Sperm motility and vitality are then assessed after Mounib (1978).

3.2.4 Selection of Suitable Broodstock

Spawners may come from either a captive or wild stock. There are, however, no established criteria regarding selection of broodfish. Nevertheless, in general, fish selected for broodstock should be fast-growing, lively fish, among the largest and strongest members of their age group and free from parasites and diseases.


An adequate supply of broodfish is essential for successful induced-breeding operations or artificial propagation, especially of the most important cultured species. Eggs for the mass production of economically important fishfry may be secured from either wild or captive stocks. The disadvantages of using wild stock is uncertainty of capturing them, the relatively large expenditure needed for their capture and transport, and the limited opportunities of obtaining good quality eggs.

3.3.1 Broodstock Management

Very little information exists regarding standard practice for broodstock management which is at present largely empirical or follows the traditional method.

3.3.2 Factors Affecting Gonad Development Nutrition

Most of the studies conducted on the nutrition (food) of cultured finfish deal with growth and body composition. There is paucity of information on the nutritional requirements of broodstock, and on suitable practical diets for broodstock. The standard practice for feeding broodstock is not well documented. At present, broodstock is fed following traditional or empirical lines. The formulated feed used are generally those commercially available as feed for rearing fish to marketable size.

Data accumulated to date indicate that poor nutrition can result in poor or no reproductive performance andf that lack of vitamin supplement could affect sperm quality. Further, mere reliance on natural food may lead to poor or variable reproductive performance. It has also been shown that the fatty acids and the fish (especially in the case of ovarian lipids) tend to utilize the w 3 polyunsaturated fatty acid. Results of other studies indicate that reduced feed levels may adversely affect fecundity and composition of ova. Environment

- Photoperiod

One of the factors considered to be of great importance to the inducement of sexual maturation and spawning is photoperiod. Photoperiod manipulation is now being employed to alter the normal reproduction of a few cultured species, for example, mullet, rabbitfish, rainbow trout, tilapia, carp and catfish. The greatest advantage of altering the spawning time of the cultured species is the availability of fry for stocking in ponds, pens and cages throughout the year.

- Temperature

Water temperature is another important factor which influences the maturation and spawning of fish. In some species of fish functional maturity is directly controlled by temperature; in others, the time of spawning is regulated by the day-length cycle such that it occurs when the temperature is optimum for survival and the food supply is adequate.

- Salinity

Some species of fish, e.g., salmon, migrate from the marine to the freshwater environment in order to spawn, while other species, such as eels, migrate from freshwater to the marine environment to complete their reproductive cycle. This confirms that salinity is somehow related to maturation and spawning.

It has been shown that the mullets, Mugil cephalus and Mugil capito do not reproduce spontaneously in freshwater; eggs reach the tertiary yolk globule stage but they are not ovulated (Abraham et al., 1960).

- Other environmental factors

In addition to photoperiod, temperature and salinity, there are other less obvious factors which may affect the maturation and spawning of broodstock. There is, however, paucity of information regarding the effects of these less obvious factors, which include rainfall, stress, sex ratios, stocking density, isolation from human disturbance, dissolved oxygen, social behaviour of fish, heavy metals, pesticides, and irradiation. Furthermore, the design of holding systems for broodstock such as ponds, tanks and cages is largely unknown.

Tanks on land are considered to be the most suitable for holding broodstock from which eggs could be obtained through environmental manipulation. The advantages of using tanks rather than ponds and cages include: (1) ease in manipulating environmental factors such as photoperiod, temperature, salinity, dissolved oxygen, etc.; (2) ease in the collection of naturally spawned eggs in nets attached to or stretched across the water outlets, and (3) the possibility of avoiding environmental contamination or pollutants. However, if tanks are used, the following points have to be considered:

  1. Photoperiod or temperature manipulation consumes a large amount of energy and is very expensive.

  2. Free-flowing seawater requires power and is relatively expensive. This is particularly true for large tanks and holding systems in elevated places.

  3. The size and shape of tanks must be suitable for the species. In particular, tanks must be deep and wide when larges fishes, such as milkfish, are to be reared. This implies, of course, high construction costs.

  4. Fish diseases can easily develop in landbased tanks. Epidemics can cause great damage and often mass mortality.

  5. In species such as milkfish, stress may impede maturation. Keeping the tanks clean would, therefore, pose another problem.


3.4.1 Selection of Spawners

The selection of spawners from the broodstock should be done months before the beginning of natural spawning to allow ample time for the fish to be conditioned to environmental and diet controls. Spawners are normally selected based on the following criteria:

Selected spawners are then transferred to the pre-spawning tank. The ratio of male and female stocked in the pre-spawning tanks is 1:1.

3.4.2 Care of Spawners in Pre-spawning Tanks

Immediately after stocking in the pre-spawning tank, the feeding is reduced from 5% to 1% of the total body weight and the fish are fed only once a day. This is to prevent them from becoming fat which could result in poor gonadal development.

Water in the spawning tank should be maintained in good condition. This can be achieved by changing the water about 50–60% daily.


Presently, two major techniques are employed in the massproduction of seabass fry in Southeast Asian countries: artificial fertilization and induced spawning.

3.5.1 Artificial Fertilization

Spawners are caught in natural spawning grounds near the mouth of the river or in saltwater lakes, where the water depth is about 10–20 m. Gillnets and seine-nets are commonly used. Normally, the fishermen will net the fish during spring tide 2–3 days before the new moon or full moon, up to 5–6 days after the new moon or full moon between 13.00 and 22.00 t at the rising tide.

The degree of maturity of the collected spawners should be immediately checked. If the female has ripe eggs and the milt of the male is at the running stage, stripping is done in the boat. The fertilized eggs can then be transported to the hatchery for subsequent hatching. In cases where only the male is caught, the milt is collected by stripping into a dry glass container and is then stored in an ice box or refrigerator. The milt can maintain its viability after a week in cold storage (5°–15°C). The preserved milt should be made available for immediate use when a ripe female is caught.

The dry method of fertilization is normally used in this case. The eggs are stripped directly from the female into a dry and clean container where the milt is added. A feather is used in mixing the milt and eggs for about 5 min. Filtered seawater is added to the mixture while stirring and then allowed to stand undisturbed for 5 min.

3.5.2 Induced Spawning

Two methods are normally used for inducing finfish to spawn in captivity, e.g., hormonal injection and environmental manipulation. Both methods would induce the fish to spawn naturally in the tank. This results in a monthly-spawning cycle until the gonads are spent.

A. Induced Spawning by Hormone Injection

To induce spawning, the hormones used include the following:

SPH - acetone-dried pituitary gland homogenate of coho salmon prepared by the British Columbia Research Council, Vancouver, Canada; 1 g of powder contains 17.6 mg gonadotropin.

HCG - human chorionic gonadotropin, manufactured by Ayerst Laboratories, New York.

Before injection, HCG is dissolved in 3 ml of its accompanying diluent. The solution is then used to homogenize the acetone-dried pituitary gland of salmon to be used for induced spawning.

Fishes having eggs with an average diameter equal to or greater than 0.65 mm are induced to spawn by injecting hormones intramuscularly a few centimetres below the dorsal fin. In the first injection the fish is given a combination of 10 mg SPH/kg body weight + 1 000–10 000 IU HCG/kg body weight. In the second injection, the fish is given a combination of 10 mg SPH/kg body weight + 2 000–20 000 IU HCG/kg body weight. Injections are administered intramuscularly a few centimetres below the dorsal fin after which the fish is completely anaesthetized by immersing it in seawater containing 100 ppm 2-phenoxyethanol. The timeinterval between injections is 24 hours for most marine fish. This interval was selected to ensure that final maturation of eggs is completed before the fish dies or before the eyes of the breeders are completely covered with a white opaque substance.

Usually, only two injections are needed to induce both captive and wild adult fish to spawn as long as the dosage and time-interval mentioned above are followed; however, badly injured fish may need a third injection. In such cases, the dosage of the third injection is that of the second injection. When a third injection is necessary, usually the fertilization and hatching rates are very low.

B. Induced Spawning by Environmental Manipulation

The method involves the simulation of the natural spawning environment in which temperature, artificial rainfall and tidal fluctuation are manipulated.

At the beginning of the new moon or full moon, the water temperature in the spawning tank is manipulated by reducing the water level in the tank to 30 cm deep at noon and exposing to the sun for 2–3 hours. This procedure increases water temperature in the spawning tank to 31°–32°C. Filtered seawater is then rapidly added to the tank to simulate the rising tide. In effect, the water temperature is drastically decreased to 27°–28°C.

The fish spawn immediately the night after manipulation (18.00–20.00 h) or, if no spawning occurs, manipulation is repeated for 2–3 more days until spawning is achieved.

Whether the fish are induced to spawn by hormone treatment or environmental manipulation, they would continue to spawn for 3–5 days after the first spawning provided the environmental factors that stimulate spawning are present, e.g., new or full moon, changes in salinity and temperature, etc.


The fish that are induced to spawn by hormone injection will be ready to spawn within 9–12 hours after the final injection. The schedule of injections for subsequent spawning must be synchronized with the natural spawning time of the fish which occurs in late evening between 18.00 and 24.00 h. On the other hand, in the stripping method, it is still necessary to sample the eggs from gonads by cannulation and examine them under the microscope. The fish has spawned only if at least 40% of the eggs are transparent.

Stripping is always done by gently pressing out the eggs with the thumb and forefingers, applying pressure just forward of the genital pore. As soon as the eggs are collected, fertilization is by the dry method, extracting the milt by hand stripping from the hormone-treated male.

The eggs and milt are mixed gently but thoroughly with turkey feathers. After at least 3 min, seawater (34 ppt) is added to the mixture while stirring. After 3 min, the fertilized eggs are transferred in a scoop net (mesh size 500 micron) and washed thoroughly with seawater isohaline in the water of the incubation tanks. The incubators are strongly aerated to prevent the eggs from clumping. The eggs are incubated at ambient temperature ranging from 25° to 30°C and at a salinity of 34 ppt. Six hours after the start of incubation, dead eggs are removed from time to time by stopping the aeration for about 5 min. Fertilized eggs float in seawater with a salinity of at least 34 ppt while unfertilized eggs sink.


The rearing tanks are usually made of plastic, fibreglass or concrete. The shape of the tanks can be retangular or circular. Volume ranges from 1 to 10m3. The tanks are usually protected from sunshine and heavy rain.

Five hours before hatching, the developing eggs are transferred to larvae-rearing tanks. The tanks are provided with mild aeration. The larvae start to hatch 16–25 h after fertilization depending on temperature and species. The usual stocking density of developing eggs is 100–200 eggs/l.


The most important environmental factors affecting larval growth and survival are: (1) light, (2) temperature, and (3) salinity.

(1) Light. The effect of light intensity and photoperiod on the growth and survival of larvae has received little attention in the past. Generally, fish larvae are reared either under continuous light or under day and night conditions. The few data accumulated to date show contradictory results. Rearing larvae under continuous light conditions has a positive effect on growth and survival for some species, but not for others.

It has been shown that high light intensities are detrimental particularly to the newly-hatched larvae. Light intensity must thus be controlled to optimize growth and survival. Young first feeding marine fish larvae have a feeding threshold of only 10 to 10 metre candle. Light intensities of 10 to 10 metre candle provide optimum feeding conditions.

(2) Temperature. Temperature can be either beneficial or detrimental to fish larvae. Temperature regimes outside the tolerance limits of a particular species will cause mortality of larvae while temperature regimes within the range that give good survival may be used to accelerate or even maximize growth of the larvae. High temperatures will shorten the time from hatching to metamorphosis, and consequently, mortality may be reduced.

The effects of temperature on the growth and survival of fish larvae must be determined for each species. Apparently, the eggs and larvae of tropical and subtropical species are generally stenothermal.

(3) Salinity. The effect of salinity on the growth and survival of fish larvae is primarily on larval osmoregulation. Survival of larvae of many species may be better at low salinities than higher salinities since low salinities are isosmotic to body fluids. Better survival for mullet was reported by Liao (1974) when the salinity of culture was gradually reduced to 25 ppt during the rearing period.


Good quality seawater at 30–31 ppt is required for larvaerearing. Water temperature is also important and should range from 26° to 28°C to promote fast growth of larvae.

Larval tanks are prepared one to two days prior to the transfer of newly-hatched larvae. Filtered seawater is added to the tanks and very mild aeration is provided. After stocking, unicellular algae (Tetraselmis sp. or Chlorella spp.) are added to the tank and maintained at a density of 8–10 × 10 or 3–4 × 10 per ml for Tetraselmis sp. and Chlorella spp., respectively. These algae serve a dual purpose: as a direct food to the larvae and rotifer and as a water conditioner in the rearing tank.

The day following stocking, the bottom of the larvae-rearing tank should be cleaned and every day thereafter. This is done by siphoning off unfertilized eggs, faeces, dead larvae and uneaten food accumulating on the bottom of the tank. About 20% of the tank water is changed daily for the first 25 days of the rearing period, then increased to 40–60% per day for the remaining culture period. Since seabass can also be cultured in freshwater, it is recommended to reduce the salinity of rearing water when the larvae are still in the hatchery, before transfer to a freshwater environment. Beginning from the twentieth day, salinity can be gradually lowered until freshwater condition is reached on the twenty-fifth day.


Feed and feeding schemes used most successfully are illustrated in Figure 3.

Fig. 1

Fig. 1 Seawater intake through sump pit

Fig. 2

Fig. 2 Water system

  1. Pump
  2. Sand filter or pressure filter
  3. Reservoir
  4. Larvae-rearing and algal culture tanks
  5. Broodstock development tanks
Fig. 3

Fig. 3 Feed and feeding scheme for grey mullet

Top of Page