The culture of the Giant Freshwater Prawn
(Macrobrachium rosenbergii de Man) in Cuba.

Report of first technical assessment mission 30 – 7 th May 1990

(Project Number TCP/CUB/8956[A]; Index 0047670)


J.F. Wickins

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.


Overall national objectives and current situation with relevance to freshwater prawn culture in Cuba.

Information from: FAO, S. Lopez and Asunciòn Prats Alemàn: Emp. Nac. Acuicultura, Gonzalo Diaz Pérez, José Fernàndez Calzat, é Gomez Hernandez, José Vazquez, Sergio Toledo Pérez; maronicultura, Melanio Borrero).


The Cuban government wish to expand the tourist industry as a sorce of much needed foreign exchange. Opportunites to supply and support the industry are actively being sought and include the provision of fish and shellfish foods from fisheries and aquaculture to supply the hotel and restaurant enterprises throughout the designated tourist areas. The culture of the Giant Freshwater prawn (Macrobrachium rosenbergii) is one such

Trials are being made with pond reared stock using three methods to improve breeding; i) eyestalk ablation; ii) photoperiod manipulation; iii) artificial insemination. In the hatchery larvae are fed algae followed by minced clam, squid, marine fish and Artemia but Nippai prepared feeds are also used. A nursery phase is employed lasting 30–40 days which takes the shrimp from 5–10 day old post-larvae to 0.5 to 1.0 g juveniles. Stocking rate is 100/sq m but trials, have been made with 1000–2000 in 70 t tanks. In the on-growing phase shrimp are stocked at 5/m2, feed is given at from 15 to 2.5% per day and salinity is 25%. Feed costs around 200–250 pesos per tonne and contains squid, shrimp meal (from processed P. schmitti caught at sea) and zeolite. Problems with unstable artificial feed are common. Production is around 400–500 kg/ha/cycle and at present there are 1.6 to 1.7 cycles/yr. Newness of the ponds, inexperience and climatic changes are constraints on production.

Early trials with P. notialis were not encouraging as growth stopped at 6–8 g, however new trials may be undertaken.


Artemia cysts are being cultured near Guantànamo and at present all production goes to the shrimp enterprise (30–50 kg/yr). The original source or cysts were San Francisco and USSR. New ponds are now commencing operation (0.5–1.0 ha) and are fertilized with chicken manure to produce the phytoplankton needed for Artemia food. Production is not as good as expected and a small predatory fish Cyprinodon sp. which tolerates salinities from 0 to 170 %, is a serious pest.


Cuba has 14 provinces each administering it's own freshwater aquaculture. The Island has no large rivers and many rivers are temporary, existing in the wet season, May to October. Approximately 180,000 ha of reservoirs have been built for irrigation as well as a large number of smaller microdams of 1–100 ha in area.

There are no native fish species of value and a variety of food species have been imported. These include channel catfish which are cultured in raceways, trout, Tilapia, and various carps. Seed are imported to some 15 nurseries and stocked into the reservoirs and microdams. In 1989 38 million seed were imported. Production amounts to about 15,000t (90%) of Tilapia, 1,200t of silver carp, 200t of common carp and 110t of channel catfish in raceways and cages. The total is 18,035t and the target for 1990 is 21,000t.

Research and extension support for freshwater aquaculture are provided from facilities at El Dique, the hatchery at Manzanillo and a small research hatchery at Pavon near Sagua La Chita, north of Santa Clara where there are 11 ha of ponds. Altogether there are 60 researchers and 100 skilled workers at the freshwater research stations of El Dique, Pavon and Manzanillo conducting 10 programmes of work including fish, genetics, nutrition, disease, population dynamics and aquaculture.

The FAO are providing technical and financial support to the Macrobrachium culture project.

History of Macrobrachium culture in Cuba.

(Information from: Emp. Nac. Acuicultura. Gonzalo Diaz Perez).

In the 1970's a survey of freshwater prawn distribution was conducted by Koreans specialists but their attempts at culture were not successful. In 1986 ten female Macrobrachium rosenbergii (and later, 2000 post-larvae) were imported from Panama and a Chinese specialist reared sufficient post-larvae to stock a pond from which 2000 broodstock were eventually obtained.


Conventional (concrete tank) hatcheries were established at El Dique and Manzanillo and in the last quarter of 1987 about 1 million post-larvae were produced at the latter. In 1989 approximately 1 tonne of prawns were harvested from ponds at Paso Malo near the hatchery and were distributed for use as broodstock. However the hatcheries do not have equipment suitable for seawater use and application was made for the construction of new hatcheries based on the advanced recirculation systems designed by France Aquaculture. Two advanced hatcheries have been established at Cienfuegos and Esmeralda, equipment has been ordered for a third under construction in Santiago de Cuba and a fourth is proposed for Matanzas.


Pond experiments at Manzanillo were conducted in 1987/8 in four 2000 m2 ponds stocked with 5, 9, 10, 16 newly metamorphosed prawns/m2. Feed was a Tilapia pellet (34% protein), given at 3–5% body weight/d. Fertilization with chicken manure (1 t/ha) plus 50kg superphosphate and urea/ha was made. Water flow was 3–5 1/sec continuously and silver and grass carp were added to reduce growth of bottom filamentous algae. After 6 months, survival ranged from 36 to 63 %, yields from 128.9 to 282.3 kg/pond, and food conversion from 5.1 : 1 to 11.5 : 1. Size distributions were positivly skewed showing the heterogeneous growth typical of Macrobrachium pond populations.

Outside Paso Malo, pond construction for Macrobrachium appears to be at an early stage of development and I was informed that land available for freshwater ponds was limited in Cuba. Land designated for Macrobrachium culture so far includes 250 ha at Marilla, 150 ha near Camaguey and with a prospect of 125 ha near Quantanamara, a total of 525 ha. In addition there are 50 ha likely to be available at the Santiago de Cuba hatchery and 10 ha at each of the hatcheries at Cienfuegos and Esmeralda. Ten ponds covering about 5 ha are being constructed at Cienfuegos with proposals for the further 5 ha of ponds to be implemented.


(Translated from spanish)

  1. To provide technical advice about highly productive spawning systems, recommending the details of the most suitable technology for the country.

  2. To check the construction projects for spawning centers and on-growing ponds and to propose a suitable design.

  3. To provide advice about the nutrition of prawns during the different phases of growth.

  4. To give two lectures to twenty researchers into aquatic organisms about the future development of prawn culture worldwide.

  5. To prepare a technical report and a rough draft of the final results of the mission.

  6. To visit centers of reproduction and on-growing to get to know the operation of growth technology for the species Macrobrachium rosenbergii.


7/5/90 Departed UK. Delayed in Madrid for 24 h.

8/5/90 Arrived Havana, Cuba, taken to Vedado Hotel.

9/5/90 Briefing at FAO (Asuncion Prats Alemàn), national situation and project details.

10/5/90 At El Dique, interview with José Fernàndez Calzat; tour of hatchery and other facilities.

Interview with José Vazquez, explaination of National Aquaculture Plan.

11/5/90 At El Dique, interview with Gonzalo Diaz Pérez, history of Macrobrachium project in Cuba.

Visit to La Turbera site, assessment of prospects for culture operations.

12/5/90 At El Dique, gave illustrated lecture “A global view of Macrobrachium culture and prospects”.

Discussions with Gonzalo Diaz Pérez. Interview with Sergio Toledo Pérez nutritionist, tour of nutrition laboratory.

13/5/90 Short tour of Old Havana. Consideration of information gained so far.

14/5/90 Interview with FAO representative, Cuba, S. Lopez. Outline of project requirements.

At El Dique, discussion and development of broodstock programmes.

15/5/90 Visit to Cienfuegos, inspection of Tunas de Zaza disused hatchery.

Discussions with Melanio Borrero about marine shrimp enterprise, shrimp nutrition and Artemia production.

16/5/90 Tour of Cienfuegos hatchery, discussions and teaching session on nitrification and filter operation and conditioning.

17/5/90 Tour of water supply arrangements for hatchery, tour of ponds and microdam.

18/5/90 Return to Havana via crocodile farm.

19–20/5/90 Collation of information gained to date, consideration of recommendations and schemes for improvements to facilities and operations seen.

21/5/90 Discussion at FAO regarding training needs.

Discussions at El Dique on improvements to Cienfuegos hatchery water supply, further teaching session with hatchery staff.

22/5/90 Full day of instruction with hatchery staff.

23/5/90 Preparation for discussions with Ministry Officials and of draft recommendations for FAO.

Visit to sugar mill cooling pond.

24/5/90 Visit to seed rice enterprise. El Palicio.

25-6/5/90 At El Dique, collection and collation of additional information.

27/5/90 Visit to Marina Hemmingway, preparation of draft report for FAO.

28/5/90 Presented draft recommendations to Sr. Lopez, FAO rep. Cuba.

Interview with Director of Science and Technology, Dr. Julio Baisre at Ministry of Fisheries, Havana.

29–30/5/90 Return to UK via Madrid.


El Dique hatchery problems

In 1988 larval survivals at El Dique were reasonable, 42–52% at densities of 40–80/1, but in 1989, the culture record was much less impressive, only one cycle reached full term with a survival of 2.5%. The mass mortalities generally occured at about stages 5–6. Careful examination of the procedures and facilities revealed a number of areas for action. Each of these (outlined below) was discussed and solutions suggested to the hatchery staff. The ideas are summarised in the RECOMMENDATIONS section below.


A change in the type of Artemia fed occured in 1989. In 1988, Artemia was imported from Japan (Nippai's own brand) but the consignment received in 1989 was Utah Salt Lake stock relabled with Nippai lables. In the past this brand has given cause for concern particularly when used in conjunction with poor quality supplementary diets. About 150 kg were imported in 1989 at about US$40–50/kg by the shrimp enterprise which then supplied the freshwater prawn enterprise. No problems obviously attributable to Artemia quality were reported from the shrimp hatcheries but there, algae was also present upon which the Artemia would feed. An Artemia production programme has been in operation in Cuba for some 4–7 years based on original San Francisco and USSR stocks but output of cysts has been low and supplied only to the shrimp enterprise.

Feeding was done at 10, 14 and 17.00 h daily but no effort was made to count the Artemia, 60 g are hatched in 35 1 and about 2 1 are added to each tank. After day 10 feeds of 50–60g of minced, sieved Tilapia flesh were given. Only two sizes of sieve were available and one of the sieves was broken.


Low water temperatures during October to February - Oct 23–24, Nov. 22 and Dec 15–18 C necessitated the installation of 2 kw stainless steel heaters and thermostats. Temperature is now controlled at 29–31°C. Water hardness is high 300–500 mg/l but no problems have been noted as, of course, the freshwater is mixed with seawater prior to use. Seawater comes by road in fibreglass tanks from about 35 km away, is used once and is then discarded. Measurements of pH were made with a Gallenkamp “pH stick” which was reportedly calibrated every 2–3 days in the chemistry laboratory. Readings were few and ranged from pH 7.9 – 8.7 in the records I saw. Ammonia was determined only when the pH was above 9.0. Minimum detectable level was 0.1 mg total ammonia /1, (which is reasonable for a test kit).

No arrangements appeared to exist for regular backwashing of sand filters.


The biological filters installed at El Dique were not likly to function correctly because of several design faults. The faults included incorrect dimensioning which led to overflows, the use of ungraded, small zeolite chips in the filter matrix which impeded flow and the ability of water to by-pass some of the filter sections. No aeration was provided to the filter while the flow was off (18 h in every 24 h). The expectation was that zeolite would remove ammonia but while this may occur in freshwater, it would not work in the salt water used during Macrobrachium culture.


The stocking was aimed at 80–100 larvae/1 but records seen indicated 182,000 and 112,000 per 2.5 cubic meter tank (45 – 73 larvae/1). One problem was that the 2.5 cubic meter capacity rectangular tanks were being stocked with between 3,000 and 11,000 /day over a period of 8–9 days because not enough females were hatching eggs on the same day. This would lead to unnecessary predation of larvae by post-larvas towards the end of a rearing cycle.


The aeration tended to concentrate the larvae towards the center of the tank but this did not seem to be a bad thing as it kept the later stages away from the walls. Late larvae jump a lot and often get stranded on tank walls. two aeration systems were in use, one was a compressor which requires an oil/water filter placed in the outlet line, the other was a blower but was only barely adequate for the tanks available.


General hygiene in the hatchery was poor with pipes and other equipment left scattered, often uncleaned, around the floor.

Cienfuegos hatchery problems


Several items of equipment ordered for use with recirculated seawater (brackish water) were found to be constructed of metals which are likely to cause problems of heavy metal toxicity or corrosion. Specifically the filter pumps had bronze impellors and bodies, the heat transfer pump had plastic coated copper coils which in my experience seldom remain durable under hatchery conditions. Another type of pump was of aluminium and would corrode rapidly in seawater.


The main freshwater supply to this expensive hatchery was severely constrained by it's distance from the microdam. The single galvanised pipe supplying the hatchery reservoir was fed from an old, leaking pump for which no replacement seemed readily available. The settlement tank installed in the supply line was found to be by-passed for no apparent reason. Water quality was poor especially when turbidity was high in the microdam.


The filters at the Cienfuegos hatchery contained coral rocks that were too big and would not have provided the most effective surface area. No ammonium tri-citrate was available for conditioning the biological filter and there were no test kits available for water quality analysis. In addition there was no backup blower in the hatchery.


Access of birds to the hatchery building together with pipes and other equipment left around the floor were hazardous and prevented satisfactory hygiene being maintained.

Pond problems at Cienfuegos


One site was visited at Cienfuegos where 10 ponds were under construction. One pond had recently been filled and almost immediately the bank surrounding the drain box had collapsed.

Inspection revealed that the pond bund (wall) was neither built over a compacted solid core of clay nor properly compacted or protected with soil of adequate clay content. The ponds were too shallow by about 60–80 cm, as no allowance had been made for freeboard. The gully beyond the fishing box was shear sided mud and would soon be undercut by water with the danger of collapse. No consideration had been made concerning the passage of waste water away from the ponds.


The iron water feed pipes to each pond were not sufficiently well protected from traffic nor adequately supported. Inlet and outlet were directly rather than diagonally opposite each other, a design that would not encourage water circulation within the pond.


During various discussions I received the impression that the amount of water required for Macrobrachium culture even at low stocking densities may have been underestimated. For example the Cienfuegos ponds cover about 5 ha at present. If the average depth is 1 m and the daily water exchange is at an absolute minimum of 4 % of the pond volume per day the annual demand will be 735,840 cubic meters of water per year. The microdam supplying these ponds and the hatchery holds only 800,000 cubic meters when full.

The disused hatchery at Tunas de Zaza.

A disused shrimp hatchery at Tunas de Zaza was visited, and although it contained some good tank and plastic plumbing items it would not be suitable for reconditioning as a hatchery for Macrobrachium. Primarily there was no adequate supply of freshwater. The town supply entered the site through a 1" galvanised pipe and I was informed that the pressure dropped significantly during the morning and evening periods of peak public demand. The nearest source of well water was likely to be some 8 km away. The site was also prone to flooding during storms and the main seawater supply pipe was attached to a flimsy jetty which would be easily damaged by waves under storm conditions. The equipment in the hatchery however might usefully be deployed elsewhere as it was all suitable for use in marine recirculation systems.

Potential sites for the culture of prawns.


This 500 ha swamp area was produced by the removal of top soil for enriching poor agricultural land elsewhere. A series of shallow channels and ponds resulted but are not suitable for prawn pond construction as the excavations are subject to upwelling ground water. Any attempt to construct ponds would involve enormous expenditure on clay or plastic lining material. There would be a high expenditure in removing vegetation and detritus and in deepening the channels. Water analysis revealed high chloride and hydrogen sulphide levels and low oxygen concentrations. Hardness was unduly high at 600–800 mg/l. Drainage and water exchange would both incur considerable pumping costs and necessitate the provision of electricity to the site. There is also a risk of pesticide spray carry-over from aerial crop spraying especially in spring and autumn.


A sugar mill SW of Havana was visited in order to assess the suitability of it's cooling pond for Macrobrachium culture. All sugar mills have similar ponds which could be available for aquaculture from about May to November or December each year. The pond was about 2000 square meters in area and sloped from 70 to 100 cm across the width. The pond contained many obstructions (platforms for the spray bars) and was supplied from a microdam. No idea of the volume of water available each day could be obtained. The surrounding area was extremely dirty and during rain it seemed likely that contamination would be unavoidable. The prawns would probably migrate to the deeper water even if fertilization was used to encourage phytoplankton growth (for shade). This would result in increased heterogenous growth. Expensive feed would be required as less natural production would develop than in an earth pond. It is understood that trials are about to start at another sugar mill and the results obtained will be of great interest.


A rice seed unit near Los Palaces was also inspected. Previous trials at this site made both with and without rice present yielded only 48 kg from 5 ha probably because the prawns remained in a 1 meter deep channel along one side of the field. The remaining three sides were bounded by a 40 cm deep channel while water depth in the center would be below 50 cm. In these experiments 120,000 post-larvae were stocked and harvested after 5 months.

Other species captured included 100 kg of Proclaimers cubes, a small crayfish and 4000 kg of small fish. This type of site is also unsuitable for Macrobrachium culture because of the restricted water flow (limited to the deep channel) and restricted depth which would effectively crowd the prawns into deep water. Frog predation, persist ant pesticide residues and competition from fish were judged to be additional problems in this area.



The attempt to culture Macrobrachium rosenbergii in Cuba is judged worthwhile because of the potential for supporting the expanding tourist industry. The decision to delay any attempts to export the product is valid because this species generally has a lower value on world markets and may also be more difficult to market than marine shrimp.


Cuba is particularly well equipped with traditional and advanced hatchery facilities at El Pique, Canaille, Cienfuegos and Esmeralda but other than at Paso Malo, pond construction is only just begining. Two other advanced hatcheries are planned, one under construction at Santiago de Cuba and another to be built at Matanzas (but see below). Despite this there are not yet sufficient trained technicians to operate any of the hatcheries at their maximum capacity. In particular there is no experience available for the efficient operation of the advanced recirculation system hatcheries.

The hatchery personnel I met were new to prawn culture and did not seem to fully appreciate the reasons for, or the importance of a number of the operations they conducted. Two and one half full days were spent explaining the basic principles and practical points fundamental to successful larval culture and recirculation system functioning.

It is RECOMMENDED that each hatchery technician (at ALL hatcheries, traditional and advanced) be given both a personal copy of the spanish version of the FAO manual of Macrobrachium culture by New and Singholka as well as a copy of the manual for the operation of the France Aquaculture type of hatchery. It is RECOMMENDED that the latter is retyped as the only copies I saw were very faint.


The following points of detail concerning the hatcheries visited at El Pique and Cienfuegos were noted. It is expected that many will apply to some extent at least, to other hatcheries in the country.

i) Several items of equipment ordered for use with recirculated seawater (brackish water) are constructed of metals which are likely to cause problems of heavy metal toxicity or corrosion. Specifically the filter pumps have bronze impellors and bodies, the heat transfer pump has plastic coated copper coils which are seldom durable. Another type of pump was of aluminium and will corrode rapidly in seawater.

It is RECOMMENDED that all future items ordered for use in brackishwater recirculation systems are of inert, non-toxic plastic (eg. Teflon, uPVC or ABS) and best quality stainless steel (eg 316 grade) or the more expensive titanium steel in the case of heat exchangers. Enquiries regarding the prospects for changing the specification of equipment not yet received have been put in hand and seem favourable. If this proves prohibitivly expensive, a less efficient but cheap indirect transfer system may be built from plastic materials (Figure 1).

In view of the cost of replacing all existing bronze components, the possible extent of the metal accumulation in the recycled water should be examined at first hand:-

It is RECOMMENDED that all recirculation systems are filled and run as if larvae were present. Duplicate samples of the brackish water should be taken from the in-coming mixed brackish water, and from each of the recirculation systems at weekly intervals for 50 days and sent for analysis at a laboratory specifically equipped and experienced in the analysis of metals in seawater. The laboratory will advise on the method of sample collection and presevation. The metals to be analysed are Cu, Zn, Fe, Cd, Pb, Sn, and Mn. The exercise should be repeated after 6 months of continuous pump operation to see if the rate of accumulation changes as the-pump wears. Periodic analysis of post larvae should be made to ensure no unexpected accumulation occurs in the prawns. The interpretation of the results should be discussed with France Aquaculture specialists.

Figure 1

FIGURE 1. Simple heat exchange system

ii) The biological filters installed at El Pique were not likly to function correctly because of several design faults. These were discussed and solutions suggested to the hatchery staff, it is not necessary to describe them in detail here. In essence, the faults included incorrect dimensioning which led to overflows, the use of ungraded, small zeolite chips in the filter matrix which impeded flow and the ability of water to by-pass some of the filter sections. No aeration was provided to the filter while the flow was off. The filters at the Cienfuegos hatchery contained coral rocks that were too big.

It is RECOMMENDED that the El Pique filters should be rebuilt to ensure correct passage of water without flooding and that pieces of coral or zeolite of about 2.5 cm diameter would be more suitable. The aluminium covers over the Cienfuegos filters should be replaced with plastic material.

Supplies of ammonium tri-citrate will be required for conditioning the biological filters at all hatcheries using recirculation systems. It would also be an advantage to purchase test kits for the analysis of ammonia in seawater at the low concentrations likely to be found in hatchery tanks, since most kits only produce good results in freshwater.

iii) At El Pique, two aeration systems were in use, one was a compressor which requires an oil/water filter placed in the outlet line, the other was a blower but this was only barely adequate for the tanks available. It is understood that two larger blower units have been ordered.

It is also understood that no backup blowers are available at the Cienfuegos and Esmeralda hatcheries and it is RECOMMENDED that each hatchery should have two servicable blower units each capable of supplying the total hatchery requirement. An allowance should be made for the use of large volumes of air for a few minutes each week to scour mechanical sand filters. The process involves bubbling air through a filter while it is full of water in order to break up compacted or congealed sand. Regular scouring reduces the need to laboriously remove and wash the sand periodically. A portable blower or oil-free compressor would be useful if sand filters are used away from the hatchery such as at the Cienfuegos hatchery freshwater treatment plant (see below).

iv) Problems experienced during larval culture in 1989 but not in 1988 suggested that a change in the quality of imported Artemia could have been partly resonsible.

It is RECOMMENDED that attempts are made to obtain Artemia cysts cultured in Cuba for testing in the hatchery and that planned trials to rear Artemia nauplii in continuous culture at El Pique be conducted as soon as adequate supplies of algal food can be assured. The latter will require separate cultivation, most conveniently perhaps in outside, fertilised shallow ponds.

This project would greatly benefit from an increased production of Artemia in Cuba. If the Salinas enterprise has insufficient ponds available for production then it is RECOMMENDED that purpose-built ponds should be constructed to:

  1. increase self-sufficiency in Artemia cyst supplies for prawn, shrimp and fish hatcheries, and

  2. ensure a more consistent product quality than has hitherto been available from Japan.

It was emphasised to hatchery staff that not all Artemia are of equal nutritional value and that the algae they feed on are also known to vary in their food value to Artemia. The Artemia Reference Centre, State University of Ghent, Rozier 44, Ghent, B-9000, Belgium can analyse quality and provide dietary supplements if required.

During the training sessions particular emphasis was placed on the importance of counting of Artemia (a simple, easily constructed counting device was described) and the preparation of wet food (fish flesh, egg custards) of the appropriate particle sizes. With regard to the latter it is RECOMMENDED that a selection of stainless steel sieves be purchased for each hatchery. These should be 8" or 10" diameter and graded to produce particles of 300, 400, 500, 700, 900, 1000, 1200 and 1400 microns.

Additionally, each hatchery should have a supply of good quality plastic mesh material of 120, 200, 250, 1200 microns for outlet screens and for the collection and washing of eggs, larvae and Artemia nauplii.

v) Alternative foods and ingredients for prepared feeds for larvae may be available in Cuba in the form of dried Chlorella, nematodes, shrimp meal and lobster processing wastes. It is RECOMMENDED that every effort should be made to investigate these possibilities and to conduct small scale laboratory trials to determine the suitability of formulations that can be made in Cuba without recourse to expensive imports. Literature containing formulations used in other countries were copied to hatchery staff and it was confirmed that nutritional expertise was available to the project through the experience and expertise of Sr. Sergio Toledo Pérez at El Pique.

If suitable pelleted rations can be found (for example, the pelleted formulations fed to marine shrimp), and proved to be satisfactory for Macrobrachium, they will need to be ground finely enough for late larvae and post-larvae. This would require the purchase of a suitable grinding machine capable of producing particles down to approximately 300–400 microns in size for each hatchery site.

If, in the future, artificial feeds are to be made from locally available dry ingredients, a small ball mill or similar milling machine will also be required to ensure that all feed components are finely ground prior to mixing and that they will therefore be evenly distributed throughout the final pellet or crumb. This requirement will be examined in 1991.

vi) The cost of bringing seawater to El Pique hatchery by road was considered high by hatchery staff and it is RECOMMENDED that the cost and suitability of making artificial seawater from evaporated sea salts (from the salinas) be determined.

An alternative approach described to hatchery staff was to set aside two mechanical and biological filtration systems for the re-conditioning of used seawater between culture cycles. Experimentation would be necessary to determine if a worthwhile saving could be made.

vii) Hygiene at the Cienfuegos hatchery must be improved by eliminating the access of birds and at all hatcheries by greater attention to pipes and other equipment left scattered, often uncleaned, around the floor. Care should be taken to ensure that all maintainance of machines is done well away from hatchery tanks and their water supplies, preferably outside hatchery buildings.

If the antibiotics recommended in the advanced hatchery manual are to be used during the culture of larvae, all staff must be made aware of the hazards to themselves and to the risks associated with the development of any resistant strains of bacteria.

viii) The main freshwater supply to the expensive Cienfuegos hatchery was jeopardised by it's distance from the microdam. A versatile system comprising dual pumps and a pressure sand filter was designed and discussed with staff at El Pique (Figure 2). It is RECOMMENDED that this and a second feed pipe is purchased and installed from the 16" main pipe to the hatchery to improve water quality and guard against failure.

ix) Efficient use of larvae tanks and of Artemia can only be achieved if larvae are stocked at the correct density within a 1–2 day period. Difficulty had been experienced in getting enough females to hatch on the same day and a scheme was designed that would enable a large number of gravid females with eggs at a known stage of development to be available throughout the year. The scheme uses 4 ponds and 6 hatchery tanks (Figure 3).


The ponds seen at Cienfuegos need substantial modification if they are to be made suitable for prawn culture. I was informed that engineers had not been involved in the design and construction of the 10 existing ponds but had only been called in later. Their recommendations, which I was not able to see, had not yet been implemented. It is essential to be able to dry a pond between cycles so the bottom should have a minimum gradient of 0.1% preferably 0.2–0.5% to ensure efficient drainage. All holes or depressions must be filled because they retain animals during harvesting and can harbour pests between crops. Tree stumps, peat and other organic material should also be removed. Drainage can be improved with channels cut into the pond bed. These may be arranged around the circumference of a pond, diagonally across the middle or in the shape of a fishbone with secondary ditches branching from a central channel.

Figure 2

FIGURE 2. Scheme for hatchery water supply.

Figure 3

Figure 3. Sample scheme for the year round production of broodstock using 2 nursery, 2 on-growing and 6 hatchery incubation tanks. Time spent in ponds varies with temperature, on-growing ponds stocked with 4 g juveniles at 6/m2; nursery ponds stocked with PL4–5 at 4/m2. Time to hatching in relation to egg colour: O = Orange, 19–21 days; B = Brown, 6–12 days; G = Grey, 1–3 days. The number of females obtained will depend on pond size and survival. Good quality feed essential, containing squid, shrimp meals, marine fish flesh. Expected yields (based on Kwei Lin and Boonyaratpalin 1988 Aquaculture 74 205–215): 60% survival, 4:1 female to male ratio; 64% of females with eggs; or 1536 ovigerous females per 1000 m2 of on-growing ponds from the two harvests (harvest 1 is selective, 2 is a drain down). S = stocking; H = harvest.

The first step in construction is often the formation of a perimeter drainage dike to prevent waterlogging of the construction area. On the other hand if the earth is too dry it may require moistening befored dike construction commences. It is extremely important to construct dikes correctly at the outset because mistakes are very difficult to remedy and dike failure can jeopardize the stock of a whole farm. The first step is to excavate down to a water-tight foundation. If porous sandy soil is to be used for the dikes then it may be necessary to use sealants such as bentonite or incorporate a clay barrier. The latter may take the form of a central core or a layer on the dike's inside surface but it it needs to be around 0.5 m thick and in contact with the impervious layer of the substrate. Dikes should be constructed in a series of layers of about 20 cm. Each layer must be thoroughly compacted and for this purpose soil may require moistening. Spaces can be left for the installation of inlet and outlet gates or alternatively the dike constructed intact and later excavated in the relevant spots using a back hoe. Allowances in design should be made for dike sinking which can be 10% of its height and for soil shrinkage which can be between 10– 20% of its volume.

In general dikes should be overbuilt because they will always be subject to erosion. Sandy dikes in particular are at risk and may need to be twice as wide as clay ones. This is one reason why clay soils are better for building deep dikes. Deep ponds can protect a crop against extremes of temperature but take longer to warm up and are harder to seine than shallow ponds. The width of a dike at its top should at least equal the height and never be less then around 1 m. 2–3 m is more usual and 3.5 – 5.0 m will be required if the dike must carry vehicles. A slope on exterior dry faces of 1:1 -2 is suitable whereas inside faces in contact with water need to be shallower with a minimum slope of 1:2 and preferably 1:3 or 4. The shallow slope is especially impoprtant if dikes are made of light earth or will be subject to strong wave action. Vegetation is valuable for stabilization and it can be encouraged with a layer of topsoil. Excessive growth however is generally a hindrance to harvesting and should be controlled. Trees should not be planted because their roots will weaken the dikes. To reduce dike erosion porous plastic sheeting can be placed on the dikes and, to limit wave damage on downwind dikes, an array of vertical wooden stakes can be effective.

To prevent the ingress of undesirable species as water enters a pond a net may be staked around the inlet and to retain the crop another may be positioned around the outlet. New ponds should always be filled slowly over a period of several days this allows the dikes to become fully saturated before they are subjected to the full weight of water.

It is RECOMMENDED that the Cienfuegos ponds be deepened to allow water depth of 1.0 to 1.3 m depth from shallow to deep end and with a bank height above the water of 60 cm all round. Engineers with experience in pond construction (perhaps those employed during the construction of the marine shrimp ponds) should supervise the work, advise wherever the clay content is inadequate and perscribe appropriate methods of compaction. I am advised that the sheeps-foot roller often used for pond compaction is not available in Cuba.

Strengthening of the bank in one corner is desirable for cleaning vehicle access. In addition, the inlet should be diagonally opposite the outlet to create circulation particularly since water exchange is likely to be minimal. The inlet pipes should be better supported and also protected from heavy vehicles. The use of slab cladding around the drain box is unsatisfactory and already has broken away from position. The drain must be integral with a substantial apron or built inside the pond as described in the FAO manual by New and Singholka.


This project would benefit if the basic shrimp and squid meal ingredients used in the manufacture of diets for the marine shrimp culture enterprise could be obtained. The waste from the lobster processing industry might also provide a valuable source of nutrients for use in crustacean diets. It is RECOMMENDED that these possibilities be explored and dietary trials be undertaken under controlled laboratory conditions before pond trials are attempted. It was confirmed that nutritional expertise was available to the project through the experience and skills of Sr. Sergio Toledo Pérez at El Pique.


The total area so far designated for Macrobrachium culture amounts to about 595 ha. I estimate that this area will be sufficient for the likely output from the advanced hatcheries during the next 3–4 years at least, bearing in mind that the ponds have yet to be built, the hatcheries have yet to achieve anything like full scale operation and much experience has still to be gained.

For example, if the hatcheries each produce on average 15 million post-larvae per year and these are stocked at 5 per square meter in a batch culture system with two cycles per year, the maximum pond utilisation will only be 600 ha. In the longer term land, but more likely water, will constrain development.


During various discussions I recieved the impression that the amount of water required for Macrobrachium culture even at low stocking densities may have been underestimated. As a rough guide the amounts of water required per day should be at least 4 and preferably 10% of the pond volume (assume 1 m depth) per day for batch culture and 15–20% per day for continuous culture for 365 days of the year. Account should be taken of other likely demands on the water source such as irrigation, cattle watering, hydroponics.

It is RECOMMENDED that the development of a further 5 ha of ponds at this site should be postponed until another water supply can be found.

It is RECOMMENDED that all future sites proposed for pond construction are carefully surveyed by qualified engineers paying particular regard to soil composition and seasonal water availability. Construction should also be professionally supervised at all stages.


In the light of the unproven performance of advanced hatcheries under Cuban conditions, the amount of land already designated for Macrobrachium ponds and the unsuitability of the La Turbera site for pond construction (see below), it is RECOMMENDED that the decision to build a fourth advanced hatchery at Matanzas be postponed.


A disused shrimp hatchery at Tunas de Zaza was visited, and although it contained some good tank and plastic plumbing items it would not be suitable for reconditioning as a hatchery for Macrobrachium. Primarily there was no adequate supply of freshwater. The town supply entered the site through a 1" galvanised pipe and I was informed that the pressure dropped significantly during the morning and evening periods of peak public demand. The nearest source of well water was likely to be some 8 km away. The site was also prone to flooding during storms and the main seawater supply pipe was attached to a flimsy jetty which would be easily damaged by waves under storm conditions. The equipment in the hatchery however belongs to the marine shrimp enterprise, but if unwanted, could usefully be deployed elsewhere as it was all suitable for use in marine recirculation systems.


Other sites that might be suitable for Macrobrachium culture were inspected during the mission. These were:

  1. La Turbera soil extraction channels;

  2. concrete cooling ponds used during the production of sugar;

  3. 5 ha seed rice units which are used in alternate 6 month periods.

i) La Turbera.

This area is clearly not suitable for pond construction from a geological standpoint.

The prospects for creating an artificial fishery for native or imported species of Macrobrachium were considered but since the large species of Macrobrachium require seawater to complete their development access canals may have to be constructed if a self-sustaining population is to be created. Alternatively, annual or twice yearly stocking with hatchery reared juveniles might support a small artesanal fishery but this prospect would need seperate evaluation.

It is possible that the whole area could be stocked with an imported species of swamp crayfish to create a self-sustaining population capable of supporting a small trap fishery. This is not recommended however for ecological reasons and because a large burrowing crayfish could cause damage to rice farm installations. As an alternative to crustacea, a fish, turtle or crocodile stocking programme might be considered.

ii) Sugar mill cooling ponds.

On the assumption that all sugar mills have similar cooling ponds and operational practices it would seem likely that the cost of cleaning the ponds and the surrounding area to a suitable standard each year plus the additional expense of fertiliser and food make the prospects for prawn culture look uneconomic even though they might be technically feasible. Moreover, there are less than 200 sugar mills in Cuba and if each has a pond of about 0.2 ha and yields equavalent to 500 kg/ha can be obtained once each year the total production will be only 20 t.

iii) Seed rice units.

On the assumption that all Cuban seed rice units are similar in design, size and water depth, it is unlikely that significant yields of prawns could be obtained without increasing water depth over the whole area and inducing circulation, for example with paddle wheels. Even so the risks of accumulation of pesticide residues in the flesh and the development of cost effective methods of predator control would need meticulous evaluation. Further trials do not seem worthwhile.


There is experience and expertise within several of the Cuban Enterprises that is relevant to all stages of Macrobrachium aquaculture but which is not being fully exploited by staff in the prawn culture project. In other words “the wheel was being unnecessarily re-invented.” Examples highlighted by the above recommendations include the experience and expertise existing in the Enterprises of Construction, Marine Shrimp Culture, Artemia Culture, Salinas, Marine fish and shellfish Processing and Animal Feed Manufacture.

It is strongly RECOMMENDED that contacts and exchange between the prawn culture project and all these other organisations be strengthened and expanded.


Several items of equipment are needed to bring the installations up to standard. It is understood that a small sum of money remains in the FAO budget for this project which might allow purchase of some of the smaller items and consumables (paragraphs 4, 6, 7, 10, 11).

  1. Air blowers. Each hatchery should have two blowers, each capable of supplying the total need. Two have been ordered for El Pique, the Canaille hatchery has two and the two existing advanced hatcheries have one each. A further 6 are required (4 if the Matanzas hatchery is postponed). A portable blower or compressor fitted with an oil filter would be useful for sand filter maintainance.

  2. Filter/pump units. Each advanced hatchery requires 7 units. These are in place at two of the advanced hatcheries. Four more have been ordered for the Santiago de Cuba hatchery. A further 10 are required (3 if the Matanzas hatchery is postponed). This requirement does not take into account spare equipment as it is assumed that an appropriate quantity of spare parts will automatically be ordered with each unit. All units must be suitable for use in seawater and contain no metal parts in contact with the water other than best quality stainless steel.

  3. Two heat pumps are required for the remaining two advanced hatcheries (1 if the Matanzas hatchery is postponed) and should not have metal parts in contact with the seawater other than titanium steel or best quality stainless steel. If these prove prohibitivly expensive or are unobtainable, a less efficient but cheap indirect transfer system may be built from plastic materials (Figure 1).

  4. The hatcheries at El Pique and Canaille also require heaters but because of the hatchery design and the need for flexibility, individual immersion heaters are recommended. These may be 2 kw, best quality stainless steel or titanium steel and, for convenience, contain an integral thermostat. Fifty are required.

  5. Six small grinding machines capable of producing particles sized between 300 and 1400 microns are required, (5 if the Matanzas hatchery is postponed), one for each hatchery. Domestic liquidisers can be used to fragment dry pellets but much food is often wasted as it is too quickly converted to a fine flour.

  6. Six sets of 8" or 10" stainless steel sieves are required, (5 if the Matanzas hatchery is postponed), one for each hatchery. The sieves should be capable of producing particles sized at 300, 400, 500, 700, 900, 1000, 1200, and 1400 microns.

  7. A quantity (5–10 meter lengths × 1 m width) of each of the following graded plastic mesh should be purchased for each hatchery: 120, 200, 250, and 1200 microns.

  8. Two freshwater pumps and associated control gear are needed for the Cienfuegos hatchery supply. Since these pumps will be used to recycle freshwater through a sand filter it is recommended that they are of plastic or stainless steel construction. The latter would have greater resistance to erosion by suspended material from the microdam. The site engineer will specify the size of equipment required.

  9. Approximately 17 off 2" valves will be required for the pumping scheme proposed for the Cienfuegos freshwater supply (Figure 1). Eight of these will be used in the recirculation pipework and should be plastic. The remainder are for single pass conditions and may be brass. All plastic plumbing MUST be protected from direct sunlight.

  10. Approximately 24 kg per year of ammonium tri-citrate will be required for conditioning the advanced hatchery filters.

  11. Test kits designed for measuring ammonia in seawater are desirable at each hatchery site. The level of detection need only be 0.1 mg of total ammonia as nitrogen per liter. A suitable example is given in Appendix 1 and is available from B D H Ltd., Poole, Dorset, UK.

    Further information on commercial test kits may be obtained from Dryden Aquaculture, Ltd., Abbeymount Techbase, 2 Easter Road, Edinburgh EH7 5AN, Great Britain.

 Aquaquant’ water testing kitscontinued
‘Aquaquant’ nitrite (NO2-) test kit
Graduations: 0-0 0.05-0012-0.02-0.03-0.04-0.05-0.06-0.08-0.1 mg/l (ppm)
‘MERCK’ (14408)
For about 110 determinations
16532 1B1 pack£38.9012x36.9636x35.98 c
  RT*811760382200 00 0
‘Aquaquant’ silicon (Si) test kit
Graduations: 0-001-002-0.04-006-008 0 1-0 15-0.2-0.25 mg/l (ppm)
‘MERCK’ (14410)
For about 150 determinations
16538 1N1 pack£38.9012×36.9636×35.98 c
  RT*8/1760382200 00 0
‘Aquaquant’ hydrogen sulphide (H2S) test kit
Graduations 0-002 004 006 008 0 1 0 13 0 16 02-0 25 mg/l (ppm)
‘MERCK’ (14416)
For about 100 determinations
16533 1D1 pack£38.9012×36.9636×35.98 c
  RT*8/1760382200 00 0
‘Aquaquant’ zinc (Zn) test kit
Graduations 0-0.1-0.2-0.3-0.4-0.5-0.7-1.0-2.0-5.0 mg/l (ppm)
‘MERCK’ (14412)
For about 120 determinations
16534 1F1 pack£38.9012x36.9636x35.98 c
  RT*8/1760382200 00 0
ITCArmand's stain ‘Gurr’ for tubercle bacilli (not suitable for export to tropical countries)
35003 4H500 ml£14.00  P
  RT*8/2920320419 00 9
R & SR: 10 S: 24/25
1,8-dihydroxynaphthalene-3,6-disulphonic acid trisodium salt: neothorone]
C16H10N2O11S2AsNa3M.W 614.27
Recommended reagent for thorium (IUPAC. 5th Report. 1964)
13010 3E5g£ 8.0012x7.60 36x7.40 A
  RT*6.1/1557293100 00 9
R & SR: 23/25-33 S:1/2-20/21-28-44
Arsenazo III [2.7 brs (o-arsonophenylazo)
1.8-dihydroxynaphthalene 3.6-disulphonic acid sodium salt]
C22H16As2N4O14S2Na2MW820 34
Reagent for thorium uranium and zirconium
13011 2F1 g£ 5.3012x5.0436x4.90 A
  RT*6.1/1557293100 00 9
R & SR: 23/25-33 S: 1/2-20/21-28-44
Arsenic broken lump
AsAt.W 74.92
27266 3T250 g£ 6.6012x6.2736x6.11 A
 7440-38-2RT*6.1/1558280480 00 0
R & SR: 23/25 S: 1/2-20/21-28-44
Arsenic powder 99.999%
AsAtW 74.92
27267 2U5 g£16.5012x15.6836x15.26 A
 7440-38-2RT*6.1/1562280480 00 0
R & SR: 23/25 S: 1/2-20/21-28-44
  Reagents recommended for Arsenic
Ammonium molybdate
Diethylammonium diethyldithiocarbamate
Silver diethyldithiocarbamate
Toluene-3,4-dithiol zinc derivative
Arsenic acid svrupy
approximately 75% w/w H3AsO4
Wt per ml at 20°CAbout 2.0 g
27268 4B500 ml£10.6012x10.0736x9.81 A
 7778-39-4RT*6.1/1553281119 00 0
R & SR: 25–33 S: 1/2-20/21-28-44
 Arsenic determination
Cotton-wool impregnated with lead acetate AsT.see page 165
Mercury(II) bromide test papers. see page 338
Mercury(II) chloride test papers. see page 339
 Arsenic determination
Mercuric bromide test papers for use in the stain test for arsenic
Stannous chloride solution AsT
Arsenic test strips
see under ‘Merckoquant’
Arsenic standard solution (1000 ppm)
c(As) = 1000±2 mg l-1 (13.3 mmol l-1)
Solute:arsenic trioxide 
Matrix:hydrochloric acid c(HCI) 
 = 0.5 mol l-1 
14086 2S100 ml New£ 4.8012x4.5636x4.44 A
14086 4U500 ml£ 7.9012x7.5136x7.31 A
  RT*6.1/2810281210 90 0
R & SR: 22
 Arsenic (III) chloride
see Arsenic trichloride
di- Arsenic pentaoxide GPR
(arsenic pentoxide)
As2O5M W 229.84 
Minimum assay (iodometric)97% 
Maximum Limits of Impurities 
Arsenous oxide0.2% 
Iron (Fe)0.1% 
27269 3C250 g£11.7012x11.1236x10.82 A
27269 4D500 g£20.1012x19.1036x18.59 A
 1303-28-2RT*61/1559281129 90 0
R & SR: 23/25-33 S: 1/2-20/21-28-44
 Arsenic pentoxide
see di- Arsenic pentaoxide

Head Office:

BDH Limited
Broom Road, Poole, Dorset BH12 4NN.
Telephone: (0202) 745520
Telex: 41186 and 418123 TETRA G
Fax: (0202) 738299

BDH Limited
Reagents Division

Broom Road, Poole, Dorset, BH12 4NN.
Telephone (0202) 745520
Telex: 41186 and 418123 TETRA G
Fax: (0202) 738299

For chemical reagent technical queries contact the Technical Services Department at Poole. (0202) 745520 Ext. 2360

BDH Limited
Diagnostics Division

Broom Road, Poole, Dorset, BH12 4NN.
Telephone: (0202) 737737
Telex: 41600 EMDIAG G
Fax: (0202) 734242

BDH Limited
Industrial Chemical Group

P.O. Box 11, Freshwater Road, Dagenham, Essex, RMB IQF
Telephone: 01-599 5141
Telex: 897479 TETRA G
Fax: 01-599 0876

For industrial chemical technical quenes contact the technical Services Department at Dagenham. 01-599 5141 Ext. 3078

BDH Limited
Apparatus Division

P.O. Box 8. Dagenham, Essex, RM8 1RY.
Telephone: 01-590 7700

Top of Page