THA: 75:008/80/WP/16 AN ASSESSMENT OF THE PERFORMANCE OF MACROBRACHIUM ROSENBERGII FED SIMPLE NON-VITAMIN-SUPPLEMENTED PELLETED DIETS

by

Michael B. New, Mali Boonyaratpalin, Paiboon Vorasayan
and Sombat Intarapichate

Joint Department of Fisheries and UNDP/FAO Programme for
the Expansion of Freshwater Prawn Farming in Thailand

Bangpakong, Chacheongsao
Thailand
1980

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.

AN ASSESSMENT OF THE PERFORMANCE OF MACROBRACHIUM ROSENBERGII
FED SIMPLE NON-VITAMIN-SUPPLEMENTED PELLETED DIETS

Michael B. New^a, Mali Boonyaratpalin^b, Paiboon Voarasayan^c
and Sombat Intarapichate^c

PREFACE

This is one of a series of 20 Working Papers which have been published by the UNDP/FAO Programme for the Expansion of Freshwater Prawn Farming in Thailand, which report on some aspects of its work during 1978, 1979 and 1980. This programme concludes on December 31, 1980 but its work will be continued by the Department of Fisheries of the Royal Thai Government alone.

INTRODUCTION

Work within our project on developing practical and cheap pelleted diets for Macrobrachium rosenbergii commenced in 1979. Much of the information available on commercial feeds is proprietary and unpublished (New, 1976). Some of the large farms in Thailand make their own feed (Trachai, 1980). Our objective was to design diets which could be made on small farms using simple machinery, and readily available ingredients. Results equal to or better than existing feeds, such as chicken feed, were sought. The two initial experiments, which are dealt with here, involved pelleted sun dried rations bound with guar gum. Later experiments, which are in progress now and will in due course be reported on by the National Inland Fisheries Institute (Thailand), involve the testing of alternative (cheaper) binder systems and of moist rations prepared on a daily basis. Other nutritional work is being planned and assistance will be sought for this work from the Network of Aquaculture Centres in Asia.

^a Senior Fisheries Biologist (Aquaculture), UNDP/FAO Programme for the Expansion of Freshwater Prawn Farming, P.O. Box 618, Bangkok 2, Thailand.

^b Chief, Nutrition Unit, National Inland Fisheries Institute, Kasetsart University Campus, Bangkhen, Bangkok 9, Thailand.

^c Fisheries Biologist, Chacheongsao Fisheries Station, c/o Department of Fisheries, Ministry of Agriculture and Cooperatives, Bangkok 2, Thailand.

The two completed experiments compared a) diets of two proteins levels with chicken starter crumbles in earthern ponds and b) diets having three protein levels with chicken starter crumbles in concrete ponds. Unlike the chicken starter control, the experimental diets utilised no vitamin or mineral supplements as it was suspected that these might not be necessary in earthern ponds when a balance diet was employed. The diets used in concrete ponds also had no supplements and it was expected that this would cause problems. Stahl (1979) has reported that an earthern substrate appears to supply a major growth factor lacking in applied feeds.

It was intended that the final or interim results of each experiment would be reported at ‘Giant Prawn 1980’, a conference held in Bangkok in June 1980. In fact the experiment in earthern ponds was terminated before the meeting began, due to human predation, and its results were not presented. The final results of the concrete pond experiment were reported and the paper (Boonyaratpalin and New, 1980) is attched as Appendix I.

As will be explained later, it is not possible accurately assess the results of the earthern pond experiment, but some prelimanary information was gained from this work. The results from the concrete pond experiment, however, were interesting and rather surprising. This working paper reports the composition of the diets used and the results obtained and discusses their implications for future work being planned in Thailand.

MATERIALS AND METHODS

Materials and methods for the concrete pond experiment are fully described in Appendix I. 15%, 25% and 35% protein feeds were compared to broiler starter feed.

In the earthern pond experiment six ponds varying in size from 1,643 m² to 2,755 m² on a commercial farm were chosen because we wanted the test to be conducted under practical conditions and because of the unavailability of suitable government facilities at the time. The layout of the ponds is given in Figure 1. 2 week old juveniles were stocked at 5/m².

The formulae of the rations used in each experiment is given in Table 1. The quantity of feed presented in the earthern pond experiment was based not on biomass but on visual water quality. Daily feeding rates rose from an initial 1 kg/rai* to 4.5 kg/rai after 4 months (Table 2). Diet preparation was as described in Appendix I.

Some intermediary information on growth rates in the earthern pond experiment was collected but is not recorded here, since sampling techniques were inconsistent. Final results of this experiment are given in Table 3.

* 1 rai = 1,600 m²

RESULTS

Detailed results of the concrete pond experiment are given in Appendix I. Differences in mean body weight were not significant between the three protein levels of the experimental diet and the control chicken starter feed. Total production varied between 639–698 kg/ha/119 days. Results from the use of a 15% protein diet were therefore at least as good as those on a 35% protein diet and those fed the control diet. All results were better than expected for concrete ponds and the lack of vitamin or mineral supplement in the experimental rations did not seem to have a deleterious effect. Food conversion efficiencies (FCE) ranged between 2.69 and 3.00, with the highest protein and control diets appearing to perform best (but not statistically different). The best protein efficiency ratio (PER) was obtained with the 15% protein level diet and the cost of feeding per hectare and cost of feed per kg of prawn produced was least with this diet.

It is not possible to statistically analyse the results of the earthern pond experiment because of management problems during the life of the experiment. Day to day control of the farm was not possible and a number of major problems were experienced. These included a ten day period when the farmer added extra food of his own manufacture to all the ponds; the accidental addition of broodstock to the control ponds (which were removed by partial draining and seining eleven days later); and, finally, substantial and uneven human predation from the ponds. This was disclosed at what was intended to be an intermediate harvest/ sampling (171 days) and the trial was terminated at that point.

The unknown effects of human predation on this experiment and of earlier errors in pond management make it impossible to analyse the results statistically. However, taking the results presented in Table 3 as minima, it is possible to make the following comments on them and draw some possible conclusions. The 25% protein diet seemed superior to both the 15% protein and the control rations, in terms of total harvested weights and survival. The control diet, however, appeared to produce animals of greater average size; a fact which is important if there is a price differential for size. Feed conversion efficiency was generally very poor, reflecting the problems of feeding an arbitrary amount based on visual water quality, instead of the percentage of biomass in the pond obtained by regular quantitative samplings. All ponds were fed at the same rate, so comparisons are still possible, however. The feed conversion efficiency of the 15% protein ration was the worst, while the control diet had an efficiency which was intermediate between the 15 and 25% protein diets. Since there is currently not a price differential for size in freshwater prawn culture in Thailand, the income from the prawns fed the 25% protein ration was higher than that on any of the other treatments; the price of the 25% protein ration compared favourably with that of the control and it's apparent feed efficiency was better.

Taking the results of the whole earthern experiment together, it will be seen that a total of 550 kg per hectare in a period of 171 days was produced, a figure which compares well with the average production rate in Thailand of 1,250 kg per hectare per year (Singholka et al., 1980)

DISCUSSION

Results from the two experiments described in this paper conflict. In the earthern pond, as was expected, the higher protein ration performed better than the lower protein experimental ration. A similar result was not obvious in the concrete pond experiment, however, where the 15% protein experimental diet performed at least as well as the 25%, 35% and control diets. Despite this, both experiments have shown that it is possible to improve the ration used extensively in Thailand at present, namely in terms of increased production and reduced feed costs.

Further work is in progress to identify and test alternative binding mechanisms for shrimp diets. The guar gum used in these experiments was successful but very costly, representing 0.52/kg of diet. Experiments are also underway to reconfirm the results obtained in the concrete pond, to determine whether the addition of a vitamin mix improves the performance of the diet, and to examine the effect of feeding ingredients in the form of a wet diet versus a sun dried diet.

It is not expected that the differences in the formula of the diets used in concrete and earthern pond experiments resulted in the different performance between high and low protein diets in these two experiments. The difference in formulation was principally that corn meal was excluded from the ration used in the concrete pond experiment and was replaced by a larger quantity of broken rice and rice bran. However, this contention needs to be confirmed in the future.

It is interesting to note that our preliminary results indicate that protein levels of 25%, or possibly less, produce acceptable results with Macrobrachium culture in earthern pond and concrete pond conditions. This is in contrast to the work of Millikin et al., (1980), who found, in indoor experiments in fibre glass tanks, that during the first 10 weeks, juveniles fed a 40 or 49% protein ration grew better than those fed 23% or 32% protein ration. After 14 weeks they found that the 40% diet produced a better result than 49% diet. Diets with much high levels would be extremely expensive, from a commercial farmer's point of view. Our work in Thailand is preliminary, but it gives encouragement that an improved diet for Macrobrachium can be developed which costs no more, and possibly less, than the ration currently in use.

It is further interesting to note, when comparing the two experiments, that the production rates achieved in concrete ponds were greater than those achieved in earthern ponds at Suphanburi. It may be concluded from this that:-

1. the production of Macrobrachium from concrete ponds can be surprisingly good, and

2. that the realisable production rate from earthern ponds has by no means been reached yet. Actual production rates at Suphanburi were higher than those recorded by us after predation.

ACKNOWLEDGEMENTS

The authors wish to thank Mr. Pornchai Pukanaprach owner of the freshwater prawn farm where the earthern pond experiment was carried out. Without his cooperation and that of his staff we would not have been able to complete the experiment. Thanks are also due to extension staff of the Chacheongsao Fisheries Station, to scientific staff and labourers of the National Inland Fisheries Institute, to Mr. Theera Changploy, who drew the figure and to Miss Wantanee Kate-Pasook, who typed this manuscript.

The support of the Department of Fisheries, Ministry of Agriculture and Cooperatives, Bangkok, and the Food and Agriculture Organisation of the United Nations, Rome, is gratefully acknowledged, as is their permission to publish this paper.

LITERATURE CITED

Millikin, M.R., Foster, A.R., Fair, P.H. and Sick, L.V., 1980. Influence of several dietary protein concentrations on growth, feed conversion and general metabolism of juvenile prawn, (Macrobrachium rosenbergii). Proc. World Maricult. Soc., 11: in press.

New, M.B., 1976. A review of dietary studies with shrimp and prawns. Aquaculture, 9: 101–144.

New, M.B., 1980. A bibliography of shrimp and prawn nutrition. Aquaculture, 21: 101–128.

Stahl, M.S., 1979. The role of natural productivity and applied feeds in the growth of Macrobrachium rosenbergii. Proc. World Maricult. Soc., 10: 92–108.

Trachai, P.T., 1980. Personal communication from the Thai Prawn Farm and Hatchery Co., Ltd., Bangkok.

Table 1: Experimental feed composition and cost.

¹ As in Appendix I

² As in Appendix I

³ Average of prices at beginning and end of experiments.

Table 2: Weight of food presented in earthern pond experiment to each pond¹.

¹ All ponds were fed at the same rate per unit area.

² 1 rai = 1,600 m²

³ For eight days between 28/12/79 and 4/1/80, 2.7 kg of a wet diet, based on fish meal and sticky rice, was given to each pond in addition. Thus a total of 21.6 kg extra food was given to each pond.

Table 3: Final harvest results of earthern pond experiments¹.

¹ Harvested weights were unevenly affected by earlier stealing from the ponds and, possibly, in the case of ponds 3 and 6, by the remnants of adult animals added by mistake during the experiment.

² Total harvest value = ฿104,784 (US$5,239); Total feed cost = ฿22,275 (US$1,113). (฿34/kg).

Figure 1: Layout sketch of the farm at Suphanburi, Thailand used for commercial evaluation of simple diet for Macrobrachium rosenbergii.

Appendix I

EVALUATION OF DIETS FOR Macrobrachium rosenbergii
REARED IN CONCRETE PONDS^c

Mali Boonyaratpalin^a and Michael B. New^b

ABSTRACT

Freshwater prawns are usually grown in earthen ponds where a variety of natural food is stimulated by fertilization, either directly or by feedstuff supplementation. Enquiries are received however as to the possibility of rearing Macrobrachium rosenbergii in concrete ponds. Pelleted diets with three protein levels (15%, 25% and 35%), derived from those formulated for a parallel experiment in commercial earthen ponds (New, M.B., 1980) have been tested in concrete ponds at this Institute. No attempt to modify the diets to provide a complete ration was made, for comparative purposes, and no vitamin supplement was provided.

Growth, production, survival and feed conversion of prawns fed 15%, 25%, 35% protein diets and broiler starter feed were not significantly different (P>0.05). The protein efficiency ratio of the prawns fed 15% protein diet was also not significantly different but appeared better than the other diets. This suggests that the 15% protein diet may be a desirable formula from a economical standpoint, at least for the first four months of rearing. The absence of a vitamin mix did not appear to cause problems but it is not yet known whether its inclusion would improve the performance of any of the test diets.

^a Nutrition Unit, National Inland Fisheries Institute, Department of Fisheries, Ministry of Agriculture and Cooperatives, Bangkok, Thailand.

^b UNDP/FAO Programme for the Expansion of Freshwater Prawn Farming, Bangkok, Thailand.

^c Paper presented at Giant Prawn 1980, Bangkok, June 15–21, 1980

INTRODUCTION

The freshwater prawn, Macrobrachium rosenbergii, presently has become the favourite species of commercial aquaculture in Thailand. These are now over 300 prawn farms in Thailand (New et al., 1980). Broiler chicken starter feed has been used as the basic for prawn in the first month; afterwards the farmers often make their own feed; usually the feed composed of 60% cooked broken rice, 20% rice bran and 20% fish or shrimp meal. It is made on a daily basis. Feed cost accounts for about 13–27% of the total annual cost of production (Shang & Fujimura, 1977). Protein is of dominant nutritional importance and also is responssible for the greatest cost component in any diet; it has therefore received more attention than any other nutrient to date.

The two Ralston Purina Marine Rations for shrimp used in the USA have 20% and 25% protein levels. The Evian Kyowa formula feed for prawn (Penaeus japonicus) contains 60% crude protein. Balazs et al., (1974) reported that total protein requirements of freshwater prawns were greater during the first two periods (0–56 and 57–119 days) and may have been in excess of 35%. The protein requirement decreased to 15% in the third period (120–175 days). There is thus a vary wide apparent range of optinum protein levels.

It is believed that diets for shrimps and prawns require good stability to prevent disintergration through exposure to water and from the manipulation process of the animal during ingestion. Among several types of binders, dry pellets bound with agar yielded higher assimilation rates than food bound with either cellulose or manucol (Forster, 1972). Alginates have proven to be effective binders (Mayer and Zein-Eldin, 1972). The amount of alginate needed was related to composition of diet, particle size and the processing procedure. Vegetable gums such as Guar gum have also been used (New, 1976).

The present preliminary study was undertaken to study the effect of protein level on growth, survival and feed conversion of juveniles to adult prawns in cement ponds. These results hopefully will lead to the formulation of inexpensive and complete diets.

MATERIALS AND METHODS

On October 30, 1979; 2,000 one month old post larvae Macrobrachium rosenbergii, average weight 0.12 g (N=1,000) were distributed at random into 8,5 × 10 m² cement ponds at the National Inland Fisheries Institute. Stocking density was approximately 5/m². Prawns used in this study were supplied free from the Macrobrachium hatchery at Chacheongsao Fisheries Station.

Each diet was fed to prawns in two concrete ponds twice daily at 9 AM and 4 PM. All ponds were fed at the same rate. The amount fed was adjusted to a level based on the highest amount which appeared to have been consumed by any of the eight ponds. Therefore some ponds may have been over-fed. At day 31, 65, 93 and 119 the average weight was measured on 2 random samples of prawns from each pond. In addition, at day 119, all animals in each pond were measured and weighed. Mean prawn weight and percent survival was determined. Intermidiary results are not presented here.

The data were subjected to analysis of variance and Duncans multiple range test to determine if the differences among means were statistically significant (Steel and Torrine, 1960).

RESULTS AND DISCUSSION

Mean repliate body weights of the prawns after 119 days were 18.4 g, 12.5 g, 18.7 g, and 15.8 g for the 15%, 25%, 35% and control diets respectively (Table 2). The differences in mean body weight were not significant (P>0.05). Mean total production was 639, 643, 666, and 698 kg/ha/119 days. Similarly these differences are not significant (P>0.05). The results from the use of the 15% diet seem therefore to be at least as good as those from the broiler starter control diet. No significant effect on survival rate was experienced between the treatments. Actual figures reflect the inaccuracies of stocking procedures. Apparent survival rates varied from 61% in Pond 1 to 90% in Pond 8, with 117% being recorded in Pond 61

The total production per individual pond, expressed as kg/ha, varied between 465 kg/ha and 820 kg/ha in 119 days. This result from concrete ponds compares favourably with production rates acheived by stocking post larvae, juveniles, and mixtures of the two, stocked at similar stocking rates (6.46/m²), in 0.25 ha earthen ponds in South Caroline, which ranged from 832.1 kg/ha to 1,250.7 kg/ha in a 153 day rearing period (Smith and Sandifer, 1980). Results were surprisingly good for concrete ponds, for although plankton was present in the water, there was no earthen substrate. Stahl (1979), in short-term studies in 20,000 litre fiber glass pools, found that natural foods alone, consisting of organic detritus, soil and manure, as well as phytoplankton, produced 180% greater growth than pelleted feed alone. However the addition of pelleted feed increased production rates in ponds with natural foods present. A later experiment by Stahl (1979) confirmed that an earthen substrate supplies a major growth factor. A production rate of 963 kg/ha in 182 days in an earthen ponds in Supanburi Province, Thailand was reported by Singholka (1979) but higher production rates have been acheived in Hawaii through continuous stocking / culling procedures. Average production from Hawaiian farms seems to be declining however, from 3,195 kg/ha/year in 1972 to 1,154 kg/ha/year in 1978, as the area under Macrobrachium farming has increased from 0.6 ha to 43.3 ha (Lee, 1979).

As shown in Table 3, mean feed conversion efficiencies (g. feed ÷ g. weight gain) for prawn fed 15%, 25%, 35% and control diets (commercial broiler starter feed) were 3.00, 3.21, 2.83, and 2.69. The 35% protein and the control diet appear to give a better feed conversion efficiency though differences in FCE's are not statistically significant (p>0.05). Food conversion efficiencies ranging from 1.66 to 2.33 were obtained by Smith and Sandifer (1980) in experiments with Macrobrachium rosenbergii stocked in earthen ponds. Our data should be accepted with reservation since it is based on the amount of food offered, which was equal in each pond; some ponds may therefore have been over-fed.

Protein efficiency ratio (g. wet weight gain/g. protein fed) was inversely related to dietary protein level in the experimental diets. The best PER (2.43) was obtained with the 15% protein level diet. However the differences between the diets were not statistically significant (p>0.05).

Feed is an important cost item (Shang and Fujimura, 1977). The cost of feed per kg of prawns produced (cost of feed × feed conversion efficiency) was ฿ 13.44/kg, ฿ 17.08/kg, ฿18.23/kg and ฿ 15.60 for the 15%, 25%, 35% and control diets respectively. Based on the production levels, the cost of feed per hectare (cost of feed/kg of prawns × production in kg/ha/119 days) over a 119 day period was ฿ 8,584/ha, ฿ 10,979/ha, ฿ 12,145/ha and ฿ 10,895/ha per the 15%, 25%, 35% and control diets respectively.

CONCLUSIONS

Based on the results of the current study, it appears that a simple, non-vitamin or mineral supplemented, 15% protein diet will give favourably comparable results to the broiler starter feed commonly used in Thailand, at least when freshwater prawns are reared in concrete ponds. Prawns grown on a 15% protein grew at least as much in average weight as those on the other treatments although survival rate appeared lower. Differences in average weight and survival rate were not significant, Although the food conversion efficiency of the 15% protein diet (3.00) appeared to be not so good as the broiler starter feed (2.69) it was not statistically different. The protein efficiency ratio appeared to be better than the control although the difference was not statistically different and it was the most economical feed to use in terms of cost per unit production. Though the three experimental diets were similar in terms of ingredients used and lipid levels were kept constant at 7–8%, it is not possible to compare them with the control diet in terms of protein content alone. The latter contained vitamin and mineral mixes designed for poultry and had a much lower lipid level (<3%) than normally used in shrimp diets.

Similar results have been obtained with comparably formulated 15% and 25% protein diets additionally containing corn, when compared with broiler starter feed and fed to Macrobrachium rosenbergii in earthen ponds in Thailand over 5.5 month growing period (New, 1980). These results have not been presented, however, owing to difficulties in interpretation caused by human predation. while lack of vitamin and mineral supplementation in the experimental diets was thought less likely to cause problems in earthen ponds, our results in concrete ponds were surprising. Clearly, the natural food present in non-earthen ponds with infrequent water change and well fertilised with formulated feeds has a significant effect on growth rate at least over the first four months of rearing. Balazs et al., (1976), using outdoor fibre glass tanks, obtained surprisingly good growth with no food at all over a 110 day period, when algae were present.

Further work is required to answer the many questions being raised by dietary studies with Macrobrachium rosenbergii. Based on our work and on the results of a parallel experiment (New, 1980), the following questions come first to mind:-

1. Can the costs of production during the first four months be further reduced, beyond the saving made by substituting a 15% protein diet for broiler starter feed, by replacing feed with a fertilisation regime alone?

2. Is vitamin supplementation necessary in diets for use in earthen ponds?

3. Is the application of a binder to feeds, in pond as opposed to laboratory tank experiments really necessary? Removal of the binder in our experimental diets would have saved at least ฿0.49/kg of diet and, if similar results had been obtained, reduced the cost of producing 1 kg of prawns to ฿ 12.95/kg for the 15% protein diet, compared with ฿ 15.60/kg for the broiler starter feed.

4. Can the efficiency of feed utilization and total production per unit area be increased by including an intermediary fish into the food chain, as occurs in Hawaiian freshwater prawn farming?

5. Do freshwater prawns perform better when a diet with same formulation is presented as a dry or a semi-moist pellet?

ACKNOWLEDGEMENTS

The authors wish to thank Chanintorn Sritongsuk of the Fish Culture Technique Unit at the National Inland Fisheries Institute for the use of facilities. Appreciation is also extended to Somsuk Singholka, of the Chacheongsao Fisheries Station for supplying the post larvae. Thanks are also due to FAO for supporting this work through the purchase of equipment and feed ingredients.

We wish to thank the Department of Fisheries, Royal Thai Government and the Food and Agriculture Organisation of the United Nations for permission to publish this paper.

LITERATURE CITED

Balazs, G.H., Ross, E., Brooks, C.C. and Fujimura, T., 1974. Effect of protein source and level on growth of the captive freshwater prawn Macrobrachium rosenbergii. Proceedings of the World Mariculture Society, 5: 1–4.

Balazs, G.H. and Ross, E., 1976. Effect of protein source and level on growth and performance of the captive freshwater prawn, Macrobrachium rosenbergii. Aquaculture, 7: 299–313.

Forster, J.R.M., 1972. Some methods of binding prawn diets and their effects on growth rate and assimilation. J. Cons. Int. Explor. Mer, 34: 200–216.

Lee, S.R., 1979. The Hawaiian prawn industry: a profile. Honolulu: Aquaculture Developement Program, DPED, State of Hawaii, Dec. 1979.

Meyers, S.P. and Zein-Eldin, Z.P., 1972. Binders and pellet stability in development of crustacean diets. Proceedings of the World Mariculture Society, 3: 351–364.

New, M.B., 1976. A review of dietary studies with shrimp and prawns. Aquaculture, 9: 101–144.

New, M.B., 1980. Personal communication.

New, M.B., Singholka, S. and Vorasayan, P., 1980. Current status of freshwater prawn farming in Thailand. Proceedings of International Conference on Freshwater Prawn Farming, Bangkok, Thailand (Giant Prawn 1980): in press. Shang, Y.C. and Fujimura, T., 1977. The production economics of freshwater prawn (Macrobrachium rosenbergii) farming in Hawaii. Aquaculture, 11: 99–110.

Singholka, S., 1979. Giant freshwater prawn farming in Supanburi, Thailand. Internal working paper of the Programme for the Expansion of Freshwater Prawn Farming. Ref. THA: 75/008/79/WP/7.

Stahl, M.S., 1979. The role of natural productivity and applied feeds in the growth of Macrobrachium rosenbergii. Proc. World Maricult. Soc., 10: in press.

Steel, R.G.D. and Torric, J.H., 1960. Principles and procedures in statistics. McGraw-Hill, New York. 481pp.

Smith, T.I.J., and Sandifer, P.A., 1980. Influence of three stocking strategies on the production of prawns, Macrobrachium rosenbergii, from ponds in South Carolina, U.S.A. Paper presented at the Symposium on Coastal Aquaculture, Marine Biological Association of India, Cochin, India, 12–18 January 1980.

Table 1. Composition and cost of experimental diets

¹ No manufacturing cost has been added to the cost of raw materials; it has been assumed that this would be equivalent approximately to the savings which would be acheived by bulk purchase of ingredients.

² Purchased in quantities up to 200 kg.

Table 2. Average body weight, total production and apparent survival of prawn on three dietary protein levels and one control diet (broiler starter feed) after 119 days.

¹ The differences in average body weight, total production and survival are not statistically significant (P>0.05).

Table 3. Analysed dietary protein levels, feed conversion efficiency, and protein efficiency ratio of three test diets and a control.

¹ Single analysis.