Based on the results of this consultancy, the following recommendations are made:
The new “Improved Vitamin Concentrate” (Appendix 3) should be obtained from one of the suggested manufacturers as soon as possible. After the appropriate dilution, it should be used at Sg. Merbok in the seabass and shrimp feed formulations given in this report. Until this new vitamin concentrate becomes available, the Agrimate fish premix should be used at the rates given in section 4.1 of this report and noted in the various appendixes dealing with formulation.
When animals are available, the broodstock feed for seabass and the maturation diet for shrimp should be tried. The formulae and method of preparation are given in Appendix 15. Some broodstock should be kept on existing feeding regimes for comparative purposes.
The new post-larval feed for P. monodon, of which a trial quantity was made during this consultancy with project staff and whose effectiveness has been demonstrated, should be in future used in all post-larval production in the Sg. Merbok hatchery. The use of micro-encapsulated feed for this purpose should be abandoned as inadequate, cumbersome, and expensive. The formulation and method of manufacture of the new post-larval shrimp feed are given in Appendix 8. This diet should also be tested, with and without supplementary Artemia, on P5-P10 animals. If Artemia can be omitted as a post-larval feed, further savings will be effected in time, equipment and money.
A batch of one or both of the new grow-out feeds formulated during this consultancy should be made on site. An experiment, perhaps using several pens within one pond, should be set up to compare the results of feeding these diets with those achieved by feeding commercial brands of shrimp feed. Replication is necessary; growth rate, survival rate, and feed conversion ratio must be determined so that the cost-effectiveness of each dietary regime can be compared. The formulae of the new grow-out diets are given in Appendix 10. Appendixes 9, 11 and 13 should be referred to as examples of how experiments should be planned, executed, and interpreted. Cost-effectiveness needs to be determined for each feed tested.
It is recommended that further trials with young fingerling seabass be carried out to:
Test the new seabass fingerling diets, SB No. 2 and SB No. 3, (Appendix 14) in moist form using filleted trash fish as a control.
Test diets SB No. 2 and SB No. 3 should also be tried in dry form. The basic formulation is the same as in Appendix 14, but the feed would be sun-dried for easier storage. If sufficient facilities and animals are available this trial could be combined with (a) above. It should also test the feeding of the dry version of the diets in the dry form itself and in a form re-moistened just before feeding.
If seabass can be persuaded to accept a dry diet, and if they thrive on it, there would be an enormous impact on ease of diet preparation and storage both within the project and, potentially, commercially.
The project should try the use of the first seabass juvenile feed (SB No. 1) formulated during this consultancy (Appendix 13) for rearing seabass or grouper to market size, compared to the use of trash fish. This would determine its cost effectiveness and ease of production and use, and would be an insurance against future shortages and cost escalations of trash fish. As in order trials, replication is necessary and growth rate, survival, and feed conversion ratios should be measured. It is hoped that, based on the results of this trial and that outlined in section 5.5., a series of seabass diets with specifications suitable for each size category can be designed in future (see section 5.15). As mentioned in section 5.5 for young fingerling seabass, it would also be useful to try using diet SB No. 1 as a grow-out feed, presented in dry or re-moistened form.
It is suggested that further trials with grow-out seabass be carried out to determine:
The effectiveness of adding Payzone (nitrovin) to seabass diet SB No. 1. The Payzone should be added to the feed at the rate of 0.1% of the concentrate (containing not less than 2% nitrovin).
The effectiveness of adding a vitamin mix to trash fish. Either 2.5% of the diluted (1:6) improved vitmix (Appendix 3) or (less satisfactorily) 0.04% of the Agrimate fish premix should be thoroughly mixed with the trash fish prior to mincing.
The post-larval shrimp feeding experiment carried out during this consultancy has demonstrated not only the use of daily records in interpreting final results but also the losses of animals that occur during routine tank maintenance (removal of excess feed, water change, harvesting, etc.). Staff at all levels should be made aware of the financial effect of such losses. In a batch where 1 million post-larvae are expected to be reared, a loss of 1% (a mere 500 per day during a 20-day rearing period) would be a loss of about M$ 400. Such unexplained losses are often written off in hatcheries as “mortalities” whereas they are in fact “missing animals”, many of which may have been inadvertently thrown away.
Daily records (containing such information as weight and type of feed presented, time of feeding, observed mortalities, water exchange rate, notes on any unusual environmental conditions, equipment malfunction, etc.) must be kept for all production batches, not just for experimental work. This applies to broodstock, larvae, post-larvae and grow-out animals. These records help to determine the cause of problems or to explain the reasons why one particular batch was successful. They are also essential to evaluate feed conversion efficiency. Basic records of stocking rate and density, survival, growth rate, etc., are also essential. Length/weight relationships for each species should also be constructed, together with typical growth rate curves. The construction of a fish measuring board would help to facilitate (and make more accurate) such measurements. Without basic information mentioned above, there are no criteria of success or failure. Records of feed ingredient costs should also be maintained so that true feeding costs can be calculated. Examples of how comparative experiments should be planned, executed, and interpreted are given in Appendixes 9, 11, and 13. It is essential that reports of experiments contain not just the bare results but a discussion of them and the conclusions drawn from the trial.
During the author's briefing meeting, a request was made for a set of instructions on how to plan, execute, monitor and analyse experiments. Appendixes 9, 11 and 13 contain detailed examples of the way in which this should be done by project staff, and how the results should be recorded. In addition, Appendix 17 contains a simple list of general “DO's” and “DO NOT's” for experimental work.
It is essential that complete reports of each trial are circulated to all relevant on-site project staff, not just retained in LKIM Headquarters. It is also requested that this report, in its entirety, be circulated to project staff at Sg. Merbok. The importance of adequate experimental protocol has been discussed on site during the consultancy.
It is strongly recommended that LKIM obtain the documents listed in section 6 of this report. They should be made available to project as well as to LKIM Headquarters staff. LKIM should also try to obtain any locally published information on the analytical characteristics of Malaysian feed ingredients (such reviews are often available from ministries of agriculture 1 universities, poultry and veterinary institutes, etc.). The collection of such data, together with ingredient price trends, should be one of the duties of the “feeds specialist” (see section 5.14). Another duty should be to scan internationally-published aquaculture literature for new information on the feeding of the species of interest to LKIM. Two papers, reference in section 6 (New, 1976, 1984), review the literature on feeds for shrimp and the marine Percoidae, which include seabass and groupers. However, nutritional knowledge is not static; papers are continually being published.
Two of the officers at Sg. Merbok have been shown how to prepare and mix ingredients, to extrude moist diets, and to produce sun-dried pellets or crumbles. These officers (Rosli bin Ismail and Ahmed Fuad Ismail) should be referred to by other LKIM staff wishing to make feeds. Notes on feed preparation are contained in several appendixes to this report, especially Appendix 8. Making larger-scale batches of feed (see section 5.12), since the moist extrusion technique is suggested for this too, requires exactly the same techniques as used in small-scale manufacture. Until the equipment mentioned in Appendix 17 is available it is suggested that sun-drying be made more efficient by constructing trays from wire mesh or mosquito netting which could be raised from the ground on frames. It is recommended that, if it is decided to make shrimp feeds on a routine basis at LKIM project sites, high quality fish meals should be substituted for the Thai fishmeals in the formulations in this report. Sg. Merbok staff have been shown how to adjust ingredient inclusion rates according to the different analyses of alternative ingredients. An example of this is given in Appendix 19.
It is suggested that the Sg. Merbok project purchase one set of the equipment listed in Appendix 17 and discussed in section 4.9. Current equipment is inadequate if the project is to make its own feeds. Even the quantity of the new post-larval shrimp diet which will be required when the hatchery is in full production cannot be made properly with the existing machinery.
No tables of feeding rates have been provided in this report because of the danger inherent in all feeding tables, namely that they may be rigidly applied without reference to water quality, survival and daily consumption patterns. It is suggested that seabass and groupers (all sizes) be fed “to demand”. Post-larval shrimp should be fed “according to the volume of water” rather than the number of shrimp and the rate should also be matched to demand. For grow-out shrimp in ponds being fed dried feeds the rates given in Gold Coin literature may be applied. When moist feeds or trash fish are fed, greater quantities will be required. The equivalent quantity can be calculated from feeding rate table for dry feed as follows:
| where: | X | = feeding rate (% of biomass) of moist feed |
| Y | = feeding rate (% of biomass) of dry feed | |
| DM | = dry matter content (%) of feed (= 100-%H2O) |
Example:
Say trash fish is 75% moisture and the feeding rate for dry shrimp feed for the current animal weight is 5% (of biomass) per day, then:
feeding rate for trash fish = 5 × 100 ÷ 25 = 20%
Say a moist feed of 45% moisture is used in the above example, then:
feeding rate of moist feed = 5 × 100 ÷ 55 = 9.1%
LKIM should appoint a feeds specialist, preferably to work at the project site (s) rather than in Kuala Lumpur. Someone with a biochemical or nutritional background would be required. He/she could be recruited from the feedstuff industry, from local universities or from agricultural research institutions. Alternatively, an existing LKIM staff member with an appropriate background could be trained for this duty. It is emphasized that, whichever method of recruitment is used, the selected candidate and employers should be prepared to make at least a 5-year commitment, otherwise the training and experience acquired are wasted. Also, as now, there would be no continuity in feed development or ultimate responsibility for it, which would cause inefficiency in this key field.
If it is decided to send existing staff for (say 3–4 months) training in feed development abroad, one of the institutions listed in Appendix 1 (some of which run occasional structured short-term courses on this topic) would be suitable:
It is recommended that, until LKIM appoints its own feeds specialist (see section 5.14) further assistance should be sought, in say a year's time, from donor agencies to review progress in this field at Sg. Merbok and to assist with future planning.
A number of experimental trials have been suggested in the earlier parts of this section of the report. These are summarized as follows:
| Experimental Trials | Animals | Duration | |
| (a) | Test shrimp maturation feed | Broodstock shrimp | 2 months |
| (b) | Test new post-larval feed with and without Artemia | Post-larval shrimp P5-P10 | 6 days |
| (c) | Test grow-out diet against Gold Coin, Ang Hock, etc. | P25-market size shrimp | 3–4 months |
| (d) | Test fish broodstock feed | Broodstock seabass | 3 months |
| (e) | Test new seabass fingerling feeds (SB No. 2 and SB No.3) | Fingerling seabass | 3–4 weeks |
| (f) | Test grow-out diet (SB No. 1) | Fingerling to market size seabass | 6–8 months |
| (g) | Test use of Payzone | Grow-out seabass | 2 months |
| (h) | Test use of vitmix in trash fish | Grow-out sea-bass | 3–4 weeks |
| (i) | Test use of dried form of SB No. 1, both dry and remoistened, with seabass fingerlings | Grow-out finger-lings | 3–4 weeks |
| (j) | Test use of dried form of SB Nos. 2 and 3 with growout fish | Grow-out seabass | 6–8 months |
