A Comparison of Silver Carp (Hypopthalmichthys molitrix) Hatchery Strains and the role of Silver Carp on Livelihoods in Northwest Bangladesh Anthony O Shafer Institute of Aquaculture University of Stirling Stirling FK9 4LD Stirling, Scotland Dissertation submitted in partial fulfillment of the degree of Master of Science in Aquaculture at the University of Stirling June, 2000 ACKNOWLEDGEMENTS I would like to thank every one who assisted me with the thesis. Firstly to my supervisor Dr. David Little for his valuable suggestion during the experiment and revision of the writing. I would also like to thank Paul Bulcock with his help and suggestion and Malcolm Beveridge for his help while I was in Bangladesh. To all the NFEP-2 staff in Bangladesh. Thank you for all your help, assistance and hospitality during my stay. To the project director, for allowing me to carry out my research at the hatchery. Special thanks to Topan Dar and Mezzain Rahman for their friendship and assistance in the hatchery and hapa construction. To A.Z.M Chowdhury and Gazze for your assistance in the questionnaires. To Nagas and Purvis for their friendship, hospitality, and assistance. To Nick Mardall and Don Griffiths for their hospitality and help. Other members of the project staff that I would like to thank are: Mehrul, Faruque, Bilash, Mahmudu, Habiba, Sultana, Bepul, and special thanks to Babul, Mokbul, Ojean. I like to thank the drivers Idris and Alom. Thank you very much for your hospitality and friendship during the stay. I would like to thank my parents for the financial help and the support they have given me. CONTENTS PAGE Acknowledgements i Table of Contents ii List of figures vi List of Tables viii Abstract x 1. Introduction 1 1.1 General Background on Bangladesh 1 1.2 Resources and the Economy 2 1.3 Agriculture in Bangladesh 2 1.4 Aquaculture in Bangladesh 4 1.4.1 Role of Fisheries in the National Economy 4 1.5 Northwest Fisheries Extension Project, Second Phase (NFEP -2) 6 1.6 Constraints to Aquaculture 8 1.7 Aquaculture Production Potential 8 1.7.1 Fish Species in Bangladesh 9 1.7.2 Nutrition Related to Fish in Bangladesh 9 1.7.3 The Fish – Seed Trading Network in NW Bangladesh 10 1.8 The Perceived Quality of Seed 12 1.9 Objectives 15 2 Material and Methods 18 2.1 Experimental fish 18 2.1.2 Stocking of Broodstock 18 2.2 Experimental procedure 22 2.2.1 Feeding 24 2.2.2 Hapa construction 24 2.2.3 Pre-stocking of hapas and simulated transfer bag 25 2.2.4 Commercial hatcheries and stocking of simulated transfer bag 25 2.2.5 Treatments 26 2.2.5 Stocking 26 2.2.6 Data Analysis 27 2.3 Questionnaires 27 3 Results 28 3.1 Reproductive performance in incubators 28 3.2 Analysis of Hapa trial 31 3.3 Results of social survey questionnaires 32 3.3.1 The level of experience of nursers interviewed 32 3.3.2 Species of fish cultured by nursers in relation to the species produced 32 3.3.3 Species of fish considered the least problem to nurse 34 3.3.4 The reason why nursers perceive some species as being least problem 34 to nurse and also comprise a large proportion of their production 3.3.5 Assessment of hatchling quality by nursers 34 3.3.6 Species that are considered by nursers to have the best survival rate 35 3.4 Results Traders Questionnaire 36 3.4.1 The level of experience of traders interviewed 36 3.4.2 The fish species considered by traders to make up a large proportion 36 of fish trade and why 3.4.3 Fish species considered by trader the best for transferring and why 37 3.4.4 Indicators used to assess hatchling quality by traders 38 3.4.6 Assessment of silver carp quality in the past five years by traders 39 with over five years experience 3.5 Questionnaire Results for Male Aggregate Workers 39 3.5.2 Male aggregate worker fish purchasing results 42 3.5.3 The percent of species most frequently bought buy 42 male aggregate workers 3.5.4 Prices of the most frequently purchased table fish by male 42 aggregate workers 3.5.5 The species that are consider by male aggregate workers to be 43 the most available in markets 3.5.6 The percent of species most frequently bought in the 44 household of female road builders 3.5.7 Prices of the most frequently purchased table fish by female 44 road builders 3.5.8 The species that are consider by female road builders to 45 be the most available in markets 3.6 Market Survey Results 46 3.6.1 The proportion of species observed at Dinajpur market in 46 relation to the most quantities observed 3.6.2 The species retailers sell the most quantities at Dinajpur market 47 3.6.3 The average price in taka/kg in Dinajpur market of species 47 by weight per kilogram 4 DISCUSSION 49 4.1 Hatchery trials 49 4.1.2 Commercial Hatcheries 51 4.2.1 Hapa trial 52 4.3 The assessment of the importance of silver carp to the 55 livelihoods to the people of Northwest Bangladesh 4.3.1 Nursers 55 4.4.1 Traders 58 4.5.1 Consumers 60 4.6.1 Markets 63 5.1 CONCLUSION 63 5.2.1 Strain comparison trial 64 6.1 Bibliography 66 7.1 Appendix 69 7.1.1 Test Questionnaire for Nursers 69 7.1.2 Filled Questionnaire for Traders 70 7.1.3 Blank Form for Market Sampling 71 FIGURES PAGE Figure 1.5.1 Map of NFEP Command Area 7 Figure 2.2.1 Flow Chart of Protocol of Experiment One 19 Figure 2.2.2 Flow Chart of Protocol of Experiment Two 20 Figure 2.2.3 Flow Chart of Protocol is the same for Experiment One 21 and Experiment Two Figure 3.3.1 The level of experience of nursers interviewed 32 Figure 3.3.2 Species of fish cultured by nursers in relation to the 33 species produced Figure 3.3.3 Species of fish considered the least problem to nurse 33 Figure 3.3.5 Indicators used to assess hatchling quality by nursers 35 Figure 3.3.7 Fish species that are considered to have the best survival 35 rate during transfer from hatchery to nursery pond Figure 3.4.1 The level of experience of traders interviewed 36 Figure 3.4.2 The fish species considered by traders to make up a large 37 proportion of fish trade and why Figure 3.4.5 Indicators used to assess hatchling quality by nursers 38 Figure 3.4.7 Assessment of silver carp quality in the past five years by 39 traders with over five years experience Figure 3.5.3 The percent of species most frequently bought buy male 42 aggregate workers Figure 3.5.5 The species that are consider by male aggregate 43 workers to be the most available in markets Figure 3.5.6 The percent of species most frequently bought in the 44 household of female road builders Figure 3.5.8 The species that are consider by female road builders to 45 be the most available in markets Figure 3.6.1 The proportion of species observed at Dinajpur market 47 in relation to the most quantities observed TABLES PAGES Table 1 A simplified outline of the main actors, produce and 11 locations in the fish production and trading network Table 2.1 The mean weight and size of silver carp stocked in 18 broodstock pond Table 2.2.1 The seven treatments used for the hapa experiment 26 Table 3.1 Reproductive traits of two silver carp strains 29 Table 3.1.1 Size comparison of three day old post hatch NFEP 30 New and Old hatchery strain and commercial hatcheries Table 3.1.2 Silver Carp, survival, weigh, length and production 30 from 20 day post hatching Table 3.1.3 Gain in fork length in Silver Carp from 20 days post 30 Stocking Table 3.3.4 The reason why nursers perceive the following species 34 as being le ast problem to nurse and also comprise a large proportion of their production Table 3.4.3 Fish species considered by trader the best for transferring 38 and why Table 3.5.1 Background information on Male aggregate 41 workers/ Female road builders Table 3.5.4a The average price in taka per kilogram of the most 43 frequently purchased table fish by male aggregate workers Table 3.5.4b Current currency in Taka 43 Table 3.5.7 The average price in taka per kilogram of the most 45 frequently purchased fish in female road builder households. Table 3.6.3 The average price in taka/kg in Dinajpur market of species 48 By weight per kilogram Table 4.1.2 Background information of commercial hatcheries used 51 for the hapa trial Table 4.2.1 A summary of the amount of fingerlings produced for 57 1k of hatchling used by nursers, and the survival rates for 1994 and 1995 Table 4.3.1 The number and percent composition of fry and 59 fingerlings of silver carp and rui/mrigel/catla entering Parbatipur Railway Station for 1993 to 1996 Table 4.4.1 Sums of quantity in kg of table fish at 62 Rangpur market survey done in 1997 ABSTRACT Questionnaires were used to determine the impact that silver carp (Hypopthalmichthys molitrix) has on the livelihoods of people living in Northwest Bangladesh. Strain and hatchery comparison trials were conducted to assess the current performance of silver carp being used in aquaculture in the northwest. The comparison trials were conducted at the Northwest Fisheries Extension Project (NFEP), Parbatipur, Dinajpur Hatchery, with two strains of silver carp being investigated. The old strain (OS) was acquired from the existing stocks at the NFEP hatchery. The new strain (NS) was obtained from imported pure wild riverine fingerlings from the Yangzte River China in June 1994. Broodstock from each strain were stocked together in earthen ponds for six months. They were then spawned and seed production, fertilization rates, hatchability rates, and larvae survival were determined. Results indicate that the only significant difference between the strains was hatchability rates and fork length of larvae, which NS had the highest values. Larvae from NS and OS were then compared to three commercial hatcheries (CM-1,CM-2,CM-3) in a hapa- base trial. A further trial was created to mimic the effects of the transfer conditions the three commercial hatcheries larvae would experience before stocking into the hapas refer to as simulated transfer of new strain, (STN) and simulated transfer of old strain, (STO). Twenty day after stocking in hapas, survival, weight, length and total weight were determined for the 7 treatments. A significant difference in survival rate was observed, for STN, CM-2, CM-3 having greater survival rates than CM-1, STO, OS, and NS. The results did not indicate a clear difference between the seven treatments. The questionnaire survey ascertained that silver carp is a relatively important species to the people of Northwest Bangladesh, because it is one of the least expensive and most available cultured fish in the markets. 1. INTRODUCTION 1.1 General Background on Bangladesh The world community has identified Bangladesh either by its rampant poverty, over population, and tropical fertility. Among nations of significant size, Bangladesh probably has the highest number of people in absolute poverty per square kilometer. It is also one of the most densely populated areas in the world, with a population of 122.2 million people in a land area of 55,598 square miles. About a third of its population suffers from moderate to severe starvation during much of the year, and nearly close to another third of the population have an inadequate intake of calories during significant parts of the year. This persistent poverty within Bangladesh is often attributed to its limited resources for its growing population, but, it has also been suggested that it is not the lack of resources but the lack of governance of resources that is responsible for the persistence of poverty. Banglade sh gained independence in 1971, but despite the long struggle for democracy, the people of the country was politically oppressed at least until 1991, until the Bangladesh Nationalist Party (BNP) won its elections. Despite the fact that long absence of political stability tends to threaten governance and in turn impair development and poverty alleviation efforts, there are many non-government organizations (NGO’s) at work in Bangladesh. Their aim is to help develop the country and alleviate poverty. The Bangladesh Rural Advancement Committee (BRAC) is the largest of these NGO’s and has targeted the landless, marginal farmers, and women, and has tried to improve the livelihoods of these groups through aquaculture. 1.2 Resources and the Economy Bangladesh is still heavily aid dependent and is currently receiving approximately $1.5 billion per annum ($12 per head). Bangladesh does not have any productive natural mineral resource besides natural gas. Its industrial base is extremely narrow accounting for only 13% of the Gross Domestic Product GDP and 6% of employment, with the readymade garments industry and semi- intensive shrimp farming being the two major exported items (BBS 1998). Shrimp farming earns an estimated US$225 million each year, but is unsustainable due to environmental pollution and fishing methods, destroying the natural spawning grounds of the shrimp. However both these exports have recently taken off in the last decade due to the opening of the economy and easy access to the global market. Conversely, jute, the main export crop, is diminishing in its importance, due to the replacement of jute fiber with synthetic fiber on the global market and the replacement of jute cultivation with the growing of rice. Tea and leather are traditional but minor export crops that have been steadily contributing to the foreign exchange, but only make up a small portion. 1.3 Agriculture in Bangladesh Agriculture still remains the main component of the economy. However, its contribution to the GDP has fallen over the years from 46% in 1987 to 37% in 1998, but it still provides the most income and employment to currently 80% of the population (BBS, 1998). The main crops in Bangladesh are rice and jute, with other important crops being tea, sugarcane, oilseeds, fruits, vegetables, spices, wheat, potatoes, tobacco and cotton (NFEP-2, 2000). Rice makes up a large component of the Bengali diet with an average per capita consumption of 153.4 kg per year (BBS, 1998), supplying an average per capita consumption of 1,450 Kcal per day (NSP, 1998). In 1991 and 1992 Bangladesh had produce a near self sufficient rice crop, but it will continue remaining an importer of rice up to the year 2010. High Yield Variety (HYV) rice cultivation has been established as a mono-crop regime, which has replaced other food crops with rice, which has lead to a serious implication for food security and nutritional imbalance, especially for the poor households. The irrigation practices and chemical input used for HYV cultivation also has had an effect on depleting the capture fisheries. Additionally the change in cropping pattern has had an impact on the employment and food intake of rural Bangladeshi’s (Nabi, Datta, Chakrabarty, Begum, and Chaudhury, 1999). Land can be considered as an important income-generating asset especially for agriculture and aquaculture; therefore land ownership is closely linked to poverty. Landless households are not only income poor, but they also lack education and access to credit and other services, which prevent them from switching to non-farming occupations (NSP, 1998). The average per capita of land is very low (0.25 acres) and the amount of land is decreasing every decade with no scope for expansion. The disparity in landholding is also alarming, with only 10% of the rural households owning about 50 percent of the cultivable land and half of the rural population being completely landless. The landless depend exclusively for their sustenance on wages they usually earn from agriculture day labour and unskilled labour (NSP, 1998). A recent report from NSP Nutritional Surveillance Project, Annual Report 1998 showed that the average male landless agriculture earns 50Tk per day, which equates to just less than one US dollar. 1.4 Aquaculture in Banglades h Bangladesh has vast resources of water. It has many ponds, haors (natural depressions), baors (oxbow lakes), beels (swamps) and rivers. Approximately 34% of the country’s surface area is flooded with nutrient-rich water for nearly six months of the year, supplied by the seasonal flooding from the three major rivers Ganges, Brahmaputra and Meghna. Inland water bodies can be highly productive and contribute 73% of the country’s total fish production (14,920,000mt in 1997-98). The total inland water area of the country is 4.308m ha. Out of this 4.047 million ha are floodplains, 87,300 ha are brackish water and 0.114 million ha are natural depressions including beels and haors. Beels and haors produced 58,298 metric tons of fish during 1994-95. This represented 4.9% and 9.8% of the total country and capture fishery production. About 1.36 million ponds in the country cover a total area of 1.47 million ha, which makes up 5% of total surface are covered by l perennial water bodies, but it is believed that ess than half of these perennial water bodies are presently under any form of fish culture (World Bank, 1991). 1.4.1 Role of Fisheries in the National Economy The contribution of fisheries to the economy of Bangladesh is substantial, particularly with reference to food consumption, nutrition, employment and export. Fisheries contribute 4.7% to the national gross domestic product (GDP), 10 per cent of agricultural GDP and 10% of the export earnings (BBS, 1998). The average annual rate of growth of the fisheries sector over the years is 4.6%. Fisheries sectors provide full-time employment for 1.2 million professional fishermen, most of whom are landless, and 11 million part-time fisher folk, which is about 10% of the total population. Seasonal fishing in flooded areas is a vital component to the marginal and landless household’s subsistence activities. It is estimated that 8 per cent of the population depend upon fishing as their principal livelihoods with 75 per cent of the population engaging in subsistence fishing (New, Tacon, and Csavas, 1993). Fish is a cheap source of animal protein and constitutes 70% of animal intake for the population. However there have been considerable disparities in fish consumption between rich and poor households and this is believed to be widening (Gupta and Shah, 1992). A recent Northwest Fisheries Extension Project Baseline Survey in the Northwest Bangladesh demonstrated this disparity when they found that the average total intake of fish purchased from markets by relatively wealthily families (18 kg of market fish per year) was more than double the intake of poor families (8 kg of market fish per year) (Ali, 1998). The availability of wild fish in Bangladesh is declining. Inland capture fisheries declined from 63%(472,000Mt) of the total catch in 1983-84 to 50%(425,000Mt) in 1988-89. The per capita consumption of fish fell from 11.7 kg in 1972 to 7.5 kg in 1988, although recently increased slightly to 8.8kg per annum in 1995 (World Bank, 1991). There previously existed greater opportunities for capture fisheries for the poor in rivers, lakes and seasonally – flooded waterbodies. It has been reported that the average yield of major carps from rivers has declined from 13.3 kg/ha/year to 4.6/kg/ha/year during the period of 1970-80 (Nuruzzaman, 1990). Several factors have since reduced the availability of fish from inland capture fisheries. These include the steady decline in aquatic resources from increasing population, flood control, drainage and irrigation interventions, and a decline in stocks from over fishing and environmental degradation. The decline of capture fisheries has attracted considerable attention from government, international donors and NGO’s in Bangladesh to promote aquaculture. Some of these NGO’s have focused on aquaculture as a source of income generation for the poor or marginal households, while others have looked at local waterbody access rights for landless and poor women for both generating income and for household protein consumption. 1.5 Northwest Fisheries Extension Project, Second Phase (NFEP -2) The Northwest Fisheries Extension Project, Second Phase (NFEP-2) is jointly funded by the Department of Fisheries (DoF) of the Government of Bangladesh and the Department for Internationa l Development (DFID) of the United Kingdom. Since 1995, the main goal of NFEP-2 has been to increase aquaculture production and incomes of poor farmers in northwest Bangladesh. The project area covers eight divisions in the northwest of Bangladesh, which are Dinajpur, Gaibandha, Kurigram, Lalmonirhat, Nilphamari, Panchagar, Rangpur and Thakurgaon (Figure 1.5.1). These eight divisions cover an area MAP OF NFEP-2 PROJECT AREA of 16,058 square kilometers with a population of 9.8 million people and 88,941 ponds. The northwest region is considered one of the poorest areas in Bangladesh because of its low agricultural and fish production as a result of low fertility soils, extremes of climate, a six month dry period and sandy soils ( NFEP-2, 2000). 1.6 Constraints to Aquaculture Despite all the attention towards fish culture in Bangladesh, commonly existing aquaculture practices are not yet geared to meet the perceived shortfall in fish production. One of the major constraints to the improvement of the level of fish culture is lack of finance. As the farmers are the rural poor, they cannot afford to buy inputs that are not currently part of their on-farm resource base. Such items include lime, organic and inorganic fertilizers. Another major constraint to the development of fish culture is the low level of technical knowledge concerning optimum stocking levels, species mix, pond condition, disease prevention and fish feeding remain relatively low among many rural people (Lewis, Wood, and Gregory, 1996). Some of these constraints are being addressed by the NFEP-2 fry trader program, in which fry traders are trained to disseminate pond culture knowledge to the farmers. 1.7 Aquaculture Production Potential Currently the cultured ponds which make up 52% of the ponds in Bangladesh are producing 1.16 t/ha/year and culturable ponds 31% of the ponds in Bangladesh are producing .34 t/ha/year and derelict ponds comprising 17% of the ponds are producing 0.15 t/ha/year (New et al, 1992; Hoque 1995). The present mean production is .74t/ha/year in Bangladesh, which is low compared to China and India. Production has been achieved as high as 3.5 t/ha/year with an improved system of multi-species carp culture and low-quality supplementary feed. It has been suggested that if semi- intensive aquaculture was applied the potential yield could be as high as (0 .7 –1.4mt) per year in the existing water bodies alone. 1.7.1 Fish Species in Bangladesh The major species under culture in Bangladesh are the Indian major carp, which is C comprised of Rui (Labeo rohita), Catla ( atla catla ) and Mrigel (Cirrhinus mrigala). Indian major carps together with Chinese carps contribute about 47 percent of the total aquaculture production (New et al, 1992). The other main species being cultured are Silver carp (Hypophthalmichthys molitrix), Silver barb (Puntius gonionotus), Common carp (Cyprinus carpio), Mirror carp (Cyprinus carpio var. Specularis), Grass carp (Ctenopharyngodon idella), Bighead carp (Aristichthys nobilis), African catfish (Clarias gariepinus), and Tilapia (Oreochromis niloticus) and indigenous fish. 1.7.2 Nutrition Related to Fish in Bangladesh Fish play an integral part of the culture and tradition in the life of the people of Bangladesh. It is considered to be the cheapest source of animal protein, and provides 70 to 80% of the total animal protein intake for the resource-poor rural farmer. They are also an important source of calcium, fats, and vitamins and make up 8% of the total protein intake of the average Bengali diet. The average per capita per day caloric intake is 2251 Kcal in rural areas and 2209 Kcal in the urban areas (BBS, 1998). This is slightly under the FAO recommended intake of 2310 calories per day. The average per capita daily protein intake is 64 gm in rural areas and 67 gm in urban areas (BBS, 1998). The average fat content of the Bangladesh diet is 7 percent compared to the FAO recommended intake of 20-30 percent (Ali, 1998). Therefore, the promotion of aquaculture has the potential to increase the availability of fish, which can maintain or improve its affordability to the landless and the marginal poor. 1.7.3 The Fish – Seed Trading Network in NW Bangladesh The typical production cycle in Bangladeshi aquaculture follows a simple pattern. Fish seed, also known as hatchlings, are collected either from the wild in rivers or obtained from government or private hatcheries, where spawning is induced under artificial conditions. These hatchlings are then placed in nursery ponds, where they grow within a few days into fry and within a few weeks into fingerlings, which are several inches in length. The availability of fingerlings at the farmer’s pond side is dependent on some kind of transportation. There is a well-established fish seed transportation network in the Northwest part of Bangladesh (Lewis, Gregory, and Wood, 1993). The main fingerling market in the Northwest Bangladesh is located at night during the summer on the platforms of Parbatipur railway station, which forms the main rail junction for the Northwest and has been playing a vital role as a seed-trading center since 1930. Fish seed traders in the Northwest often procure fingerling from sources as far away a Jessore, Bogra and Rajshahi. Between 200 to 100 million fingerlings are usually traded there every year (Chowdhury and Mardall, 1997). There are two main types of fish-seed traders khuchra beparis, who transport fingerlings from nursery ponds distances of up to 100 miles by train for sale directly to pond-owners, or pikars, who transport large amount of seed from the nursery ponds to the commission agents (Lewis et al. 1993). Commission agents also known as arotdars who buy seed from traders in specialized markets such as the Parbatipur railway station. Fry traders still prefer the traditional methods of transportation of seed, which are namely patils (aluminum or earthen containers) and oil drums. Seed can also be transported in polyethylene bags with oxygen but the most preferred way is the patils, because it is relatively cheap, simple and materials are locally available. Once pond-owners purchase fingerlings they are stocked into larger ponds where they mature into food fish in between eight months and two years, depending on pond management and species. Some of the fish may be consumed at home or distributed through urban and rural markets for consumption. They can also be kept on as brood stock and sold to hatcheries, where they are spawned to produce more hatchlings to start the cycle again. Through out this process from hatchling source to food fish in ponds there is an elaborate network of relationships between different actors linked by a series of transactions (Table 1) (Lewis et al. 1993). Table 1. A simplified outline of the main actors, produce and lo cations in the fish production and trading network (After Lewis et al. 1993) Actors Produce Location Collectors Hatchlings River/hatcheries Fishermen Middlemen N u r s e r y p o n d-o w n e r s Fingerlings Nursery ponds Fingerling traders Wholesalers Local fingerling traders V i l l a g e p o n d -o w n e r s Food fish Food fish ponds C o m m e r cial pond operators Fishermen Food fish traders Rural and urban markets Currently the falling per capita consumption of fish and the declining availability of wild fish seed which the aquaculture sector has traditionally depended upon to stock their ponds, has put a greater reliance on cultured fish to meet this demand by both rural and urban people. For aquaculture to meet this demand fish seed has to be readily available. Currently there are over 700 hatcheries established in the private and public sectors in Bangladesh (Ali, 1998). These hatcheries have been breeding 13 endemic and 13 exotic fish species and contributing more than 98% (about 117 000kg) of the total hatchling production, the rest coming from natural river sources (Hussain and Mazid, 1999). 1.8 The Perceived Quality of Seed At present fish seed is currently abundant and cheap, but there has been an emerging concern that it is deteriorating in quality. Fish seed that has a poor growth performance and suffers high mortalities undermines the attempt to promote fish farming to new adopters and raises the cost of production, which in turns raises the price to the consumer making fish less available to the poor. The poor performance observed in pond stock fish can be due to sub-optimal seed quality or inadequate management by the farmer after stocking. If there is a decrease in performance in seed it could be genetic or management related. There are several stages and actors that are involved in the seed production and trading system, so this reduction in quality could occur at any stage. It also could be genetic, due to inbreeding and negative selection. This too can lead to a reduction in growth and reproductive performance (Kincaid, 1983). The rapid expansion of carp culture in Bangladesh has taken place as the controlled breeding and spawning of fish using the hypophysation technique has spread to the private sector. Hypophysation has allowed the life cycle of silver and other riverine carps being culture under static pond water conditions to be controlled. The closing of the life cycle requires little to no input of outside founder stock. Most hatcheries in Bangladesh rear their own broodstock and usually do not recruit broodstock from other farms or obtain them from natural sources. A study done by NFEP-2 showed that several districts in the Northwest region Thakurgaon, Panchagar, and Noagaon reared their own broodstock with no input of outside founder stock. Only one district Gaibandha collected some of its broodstock form the wild (Gaghat and Brahmaputra river). It can therefore can be considered that each hatchery is an isolated, self- sustaining genetically closed unit (Eknath and Doyle, 1990). Doyle (1983) has suggested that in genetically closed systems, the potential selective pressures exerted on small cultured population by various farm management practices such as choice of founder stock, the number of breeders maintained, the method of replenishing broodstock, stocking density and feeding regime, can result in indirect or negative selection, inbreeding and genetic drift. It has been well established in the scientific literature that inbreeding affects traits associated with reproductive and physiological performance (Falconer, 1981; Kincaid 1983). It has been suggested that in India fish hatcheries there have been a substantial variation in growth and reproductive performance of aquaculture stocks. Typically, what has been observed is a continue decline in the rate of fertilization, hatching, and fry survival. There is also has been a decrease in milt volume in silver carp and grass carp. Eknath and Doyle (1985, 1990) have related this decrease in performance from the inadvertent breeding of slow-growing and late maturing individuals. Hussain and Mazid (1997) showed that majority of hatcheries operators located in Jessore, Comilla and Mymensingh lacked knowledge of simple broodstock management practices and did not follow any principles or guidelines in selecting adequate sized breeders, injecting hypophysation dosage and mating unrelated male and female spawners. Not having a good understanding of appropriate hatchery procedures increases the inadvertent chance of negative selection due to the use of fish of undesirable size and leftover broodstock, and the mating of breeders’ generation after generation from closely related individuals. Inadvertent inbreeding in Bangladesh has in recent years led to a reduction in growth and reproductive performance, an increased incidence of diseases and morphological deformities and mortality of hatchery produced seeds of carps and silver barb (Hussain and Mazid, 1999). There has been a widespread concern that silver carp stocks have been deteriorating in Bangladesh through inbreeding and hybridization with other fish species. This has led to a reduction in both growth and reproductive performance. Silver carp is an economically important species in Bangladesh, especially in the Northwest. A hatchling production survey conducted by NFEP –2 in the Northwest region of Bangladesh found that silver carp was the species most produced by hatcheries during 1995 (Morrice, 1997). Silver carp is also the second most abundant seed being traded at Parbatipur railway station after the Indian major carps (Chowdhury and Mardall, 1998). Silver carp fingerling produced from nurseries in the NFEP –2 command area comprise 42 percent of the total production (Morrice, 1997). A market survey showed that silver carp was the most represented species in three regional markets in the Northwest (Thankurgaon, Dinajpur and Rangpur) during 1997 to 1998 (NFEP-2, market data). Silver carp generally has the lowest market price of the major cultured species (Morrice 1997). 1.9 Objectives The aim of the study is to not only assess the importance of silver carp to people of NW Bangladesh, but also to determine the quality of silver carp currently being used in aquaculture in NW Bangladesh. The combination of social surveys and technical on- farm trials will reveal a more complete “holistic” picture of the potential role silver carps might have in poverty alleviation in NW Bangladesh. Questionnaires were used to determine the importance of silver carp to the livelihoods of people in northwest Bangladesh and also the perceived quality of silver carp. This focused on interviewing nursers, traders, male and female unskilled labourers and market sampling. Technical Trials Trial 1 Compare the performance in hatchery incubators of hatchlings from existing silver carp stocks at the NFEP-2 farm at Parbatipur (old strain, OS) with the hatchling from imported pure Yangtze River silver carp (new strain, NS), by observing fertilization rates, hatching rates, and larvae survival and growth. The objective of the trial is to determine if there is any difference in performance between the new and old strain during early rearing in the hatchery. Trial 2 Compare the performance in hapas of larvae from the old strain with the larvae from the new strain, by determining percent survival, length, weight, and production at the end of 20 days. The objective of the trial is to determine if there is any difference in performance between the new and old strain during the nursing period in hapas. Trial 3 Compare the performance in hapas of larvae from the old and new strain with an induced period of stress (simulated transfer) before stocking in hapas with larvae transported from three commercial hatcheries, by determining the percent survival, length, weight, and production at the end of 20 days. The objective of the trial was to determine if government hatchery larvae perform better than larvae from commercial hatcheries, which might be less knowledgeable of broodstock management practices. Additionally, the performance of the new strain under conditions of simulated stress compared to the old strain was evaluated. Chapter 2 Material and Methods 2.1 Origin and maintenance of broodstock For the growth comparison trial, conducted at the Northwest Fisheries Extension Project (NFEP), Parbatipur, Dinajpur Hatchery, two strains of silver carp (Hypopthalmichthys molitrix) were investigated. The old strain (OS) was acquired from the existing stocks at the hatchery. The new strain (NS) was obtained from imported pure wild riverine fingerlings from a sanctuary found in the northern reaches of the Yangzte River China in June 1994. The new strain was tagged, isolated from the old strain and grown-on as broodstock. On October 15 th 1999, one 1600m2 decimal earthen pond was stocked with a total of 80 fish (twenty pairs of new stra in and twenty pairs of old strain silver carp) at a density of 2 fish 40m-2 . Table 2.1 summarizes the mean weight and size of the old and new silver carp. Table 2.1 The mean weight and size of silver carp stocked in a broodstock pond at NFEP Strain Weight (kg) Size (cm) New Strain 1.60 48.20 (0.30) (2.66) Old Strain 1.40 47.10 (0.28) (2.83) Standard errors of the means are in parentheses. The earthen pond was fertilized monthly using 100gm urea 40m-2 and 50gm TSP 40m-2 . and in addition was occasionally fertilized with cow manure to promote detritus. The broodstock feed was comprised of a dry ration of 40% mustard oil cake and 60% rice bran and fed at 1% of the stocked total biomass six times per week. 10/15/1999 2.2.1 Flow Chart Experiment one 20 Pairs of NS + 20 Pairs of OS Protocol Stocked in 40 decimal pond Experiment 1 4 Commercial Hatchery in Adamdigi is contacted to coordinate hypothesiation and pick up of larvae on 7/4 2/4/1999 Hormonal induction 5 Females NS + 5 Males NS 5 Females OS + 5 Males OS 3 Females NS + 3 Females OS Used for the comparison trial 1 Female NS 3/4/2000 3 Males NS Fertilization Eggs striped Pooled semen 5mls of pooled semen used to fertilize each female within a strain Fertilized eggs are divide into three equal parts than stocked in three separate incubators 3 Females NS use 9 incubators 3 Femal es OS use 9 incubators Total 18 incubators 2:30 PM Percent Fertilization Determined 9:30 AM 4/4/2000 Hatching Rate I Determined 4:00 PM 6/4/2000 First Feed 11:00 AM 7/4/2000 Total wet weight of larvae determined for each incubator 10/15/1999 Flow Chart 2.2.2 20 Pairs of NS + 20 Pairs of OS Protocol for Stocked in 40 decimal pond Experiment Two Experiment 2 20/4/2000 4 Commercial Hatchery in Adamdigi is contacted to coordinate hypothesiation and pick up of larvae on 22/4 22/4/1999 Hormonal induction 5 Females NS + 5 Males NS 5 Females OS + 5 Males OS 3 Females NS + 3 Females OS Used for the comparison trial 1 Female NS/OS 23/4/2000 3 Males NS/OS Fertilization Eggs striped Pooled semen 5mls of pooled semen used to fertilize each female within a strain Total fertilized eggs from one Female is stocked in one separate incubators 3 Females NS use 3 incubators 3 Females OS use 3 incubators Total 6 incubators 2:30 PM Percent Fertilization Determined 9:30 AM 24/4/2000 Hatching Rate I Determined 2:30 PM Hatching Rate II Determined 4:00 PM 26/4/2000 First Feed 11:00 AM 27/4/2000 Total wet weight of larvae determined for each incubator Flow Chart 2.2.3 The Protocol is the same for Experiment 1 and Experiment 2 11:00 AM 1 incubator out of each strain is made to contain equal amounts of larvae from the three females which will be used to stock the hapas 12:00 PM Simulated transfer bags are stocked with 150 gm of each strain, bags are filled with 1/4 water and 3/4 O2 bags are agitated for the next 4:30 hrs 7/4/2000 Stocking of Hapas Experiment 1 27/4/2000 Stocking of Hapas Experiment 2 Hapas stock in random block design Old strain (OS) and New Strain (NS) are stocked in the hapas at 1000 fish 2:30 PM per hapa which is determined volumemetrically 4 replicas of each strain Simulated transfer of new strain (STN) Simulated transfer of old strain (STO) are stocked in hapas at 1000 fish per hapa which is determined 4:30 PM volumemetrically 4 replicas of each strain Three commercial hatchery larvae arrive and are stocked at 1200 fish per hapa which is determined volumemetrically 5:00 PM 4 replicas of each strain 13/4/2000 Storm comes through and overturns all hapas Experiment 1 ends 17/5/2000 10:00 AM Harvest of hapas Experiment 2 ends 2.2 Experimental procedure Background The initial trial was conducted on April 2, 2000 and ended on April 13, 2000 due to a storm that came through and overturned all the hapas, releasing the experimental fish. The initial trial is referred to as experiment one (refer figure 2.2.1). a The tri l was reinitiated on April 22, 2000 and ended on May 17, 2000 and is referred to as experiment two (refer figure 2.2.2). The hapa procedure is described in figure 2.2.3. Selection of broodfish and induced breeding Broodstock were netted in the morning and examined immediately at the pond side to reduce the amount of stress. Ripe female broodstock were then determined by the presence of a swollen abdomen, soft post abdominal region and swollen genital papilla. Ripe male broodstock were identified by the presence of free running milt when pressure was applied to the abdomen. Following this the broodstock were scanned with a Trovan transducer to determine the new strain from the old, because only the new strain were PIT taged. Strains were placed in separate cotton sacks and transported to the hatchery and placed in one concrete holding tank, consisting of six NS females, 5 NS males and 6 OS females, 5 OS males. Fish were first measured for fork length, scanned and weighed then separated by sexes and placed into separated antechambers. The total weight of the broodstock was needed to determine the amount of hormone required for hypophysation later that evening. The primary injection of female broodstock occurred at 5:20 P.M. in the evening. Final gamete maturation was induced intraperitoneally through injecting at the base of the pectoral fin with human chorionic gonadotrophin hormone (HCG), obtained from Funda Hormone Factory in Xiamen, China, administering 200IU Kg-1 using 0.4mls of injected solution per Kg of fish. The resolving dose was administered at 12:30 A.M. The hormone was whole acetate dried pituitary gland (PG), obtained from Bangladesh. The homogenized PG resolving dose was also administered intraperitoneally through the base of the pectoral fin with females received 5mg in 0.4ml of water kg-1 . In contrast to the females, males received only a single dose of PG (2mg PG in 0.4ml of water kg-1 of fish) at the time of the primary female dose. Eight hours after the resolving dose was given, new strain and old strain were differentiated by scanning with the Trovan transponder, and stocked in different antechambers. Three males from each strain were then checked for the presence of free running milt. The milt was then hand stripped into a dried plastic weighing boat and the volume was determined volumetrically by surgical syringe and recorded for each male. Pooled semen from three males was used in doses of 5mL to fertilize the eggs from each female within a strain. Females were gently dried and eggs stripped into a dry bowl to prevent any water contamination. Eggs were then weighed on an analytical balance and a small sample was removed for a subsequent egg quality determination. Milt was mixed into each batch of eggs using a feather. A batch of fertilized eggs was divided into three equal weights and placed into three separate 800L concrete incubators during the first experiment. In the second experiment incubators were limited so all the eggs from one female were placed in one incubator s. Well water 26.5 o C flowed through the incubators. For the first experiment (refer to table 2.2.1) each female occupied three incubators, 9 incubators per strain, and 18 in total; for the second experiment one female per incubator, 3 incubators per strain, and 6 in total. The number of incubators and amount of eggs of each strain was recorded. Three hours after fertilization fecundity and egg weight was determined by counting five replicates of weighed batches of eggs. Fertilization rate was determined six and half hours after fertilization volumetrically, based on counting the number of eggs fertilized and the total number of eggs in three to five random 400mL samples. Egg hatchability was determined 24 hours after fertilization volumetrically, based on counting the number of larvae in three to five random 400mL samples over the total number of fertilized eggs. Hatching survival was determined by measuring the total weight of wet fish five days after fertilization over the weight of eggs initially stocked in each incubator. 2.2.1 Feeding First feeding began when hatched fish absorbed their egg yolk sac. This occurred approximately 52 hours after hatching at an incubator water temperature of 26.5 o C. They were then fed an egg solution of a half boiled egg blended and strained and brought to a final volume of 1 liter in water. Fish were fed at a rate of 1ml of egg solution g-1 of fish hatched every three hours. The initial feeding regime in hapas consisted of egg -1 solution twice a day once at 9:30 and again at 16:30 at 1ml g stocked for the first two days then switched to rice bran husk at 5% body weight twice a day at 9:30 and 16:30. Water temperature in hapas ranged from 28 – 33 o C. 2.2.2 Hapa construction Hapas were placed in a 5600m2 earthen pond (borrow pit). Twenty-eight hapas were constructed from fine mesh monofilament nylon cloth; each measuring 2m2 (2m x 1m) sewn together with nylon thread and lined with nylon tape. Hapas were fixed to a bamboo poles, by the corners at the top and bottom of hapa in the second experiment. Cones were then placed on bamboo in order to deter crabs from climbing up the poles and damaging the hapas. The base of the hapa was tied down above the cone. Extra bamboo was tied down to the hapa bamboo frame and extended to the bank and was fixed to a post for extra frame support in case of high winds and waves during the second trial. During the first experiment the bottom of the hapa was not fixed to the bamboo post (it was weighed down with bricks) and there was no extra support for the frame. 2.2.3 Pre-stocking of hapas and simulated transfer bag Before hapas and simulated transfer bags were stocked, one incubator from each strain was made to contain an equal number of larvae by weight from each female, so each contributing female will be equally represented in the treatments. Each incubator was let to settle for one hour before stocking of treatments. 2.2.4 Commercial hatcheries and stocking of simulated transfer bag Four commercial hatcheries were first contacted in Adamdigi, one week prior of stocking of hapas to coordinate hypophysation and stripping of females. Six hours before hapas were stocked commercial fish were purchased and had a four to five hour deliver time in a Land Rover to NFEP hatchery and were stocked at 150gm per bag. Five hours before commercial hatchery larvae arrived the simulated transfer bags were assembled. Fish were simulated transferred in a plastic bag filled with 1/4 water and containing ¾ pure oxygen and agitated regularly (every 15 – 20 minutes). Simulated transferred bags were stocked with 150 gm of fish and were placed inside the hatchery. There was one simulated transfer bag per strain. 2.2.5 Treatments The trial composed of seven treatments (New strain, Old strain, New strain with 5 hr simulated transfer time in bag, Old strain with 5 hr simulated transfer time in bag, Commercial hatchery 1,2,3) with four replicates of each treatment (refer to table 2.2.1). Fish were assigned to hapas laid out in a random blo ck design. Table 2.2.1 The seven treatments used for the hapa experiment Treatment 1 2 3 4 5 6 7 Description New strain Old strain Simulated transfer Simulated transfer Commercial Commercial Commercial of new strain of old strain hatchery 1 hatchery 2 hatchery 3 Replicates 4 4 4 4 4 4 4 2.2.5 Stocking New strain simulated transferred, old strain simulated transferred, and new strain and old -2 strain silver carp were stocked at a density of 500 fish m , while commercial hatchery silver carp were stocked at a density of 600 fish m-2 . The densities for stocking the hapas were initially determined volumetrically using the NFEP hatchery raised silver carp, which were larger than the commercial hatchery silver carp, thus giving a higher stocking density for the commercial hatchery silver carp. After the hapas were initially stocked the density actually stocked for the commercial hatchery silver carp was later determined volumetrically. Silver carp were stocked in hapas for 20 days, at which point the trial was concluded and weight, fork length, survival rate and production was determined. 2.2.6 Data Analysis Strain and female differences were compared using a one-way analysis of variance (ANOVA) followed by Tukey’s multiple comparison tests to determine significant P differences among the means ( <0.05). Percentage survival data was normalized if required by transformation to angular degrees. 2.3 Questionnaires Several structured questionnaires were used to ascertain the importance of silver carp to the livelihoods of the people in northwest Bangladesh. The questionnaires targeted several different ‘actors’ in the fish production cycle. The ‘actors’ that were interviewed were nursers n=15, fry traders n=15, male aggregate workers n=12, female aggregate workers n=12, and retail market stalls n=15. The interview followed a checklist of open- ended questions designed to gather qualitative data. All questionnaires were pretested in the field and necessary changes made before the full-scale interviews were undertaken. After several interviews with ‘actors’ a different location was used so respondents would not standardize their answers to previous given answers. Females were interviewed separately away from the public with a female interpreter present, to prevent intimidation. Trial 4 Compare the performance in hapas of larvae from the old and new strain with an induced period of stress (simulated transfer) before stocking in hapas with larvae from the old and new strain without a period of induced stress (no-simulated transfer). The objective of the trial was to determine if there was a decline in performance from larvae with simulated transfer than without simulated transfer. As there are many ‘actors’ in the seed trade the decline in seed quality might not only be associated with inbreeding depression but could also be due to the many handlers in the seed trade trading networks. CHAPTER 3 RESULTS 3.1 Reproductive performance in incubators Within the hatchery, significant strain differences were seen in the following reproductive trait1) female broodstock length, 2) hatching percentage 1 and 2 as described in Table 3.1. There was a significant difference between the strains in the female length used, with the new strain being significantly larger than the old strain. There were no differences in average relative fecundity (number of eggs produced per kilogram of fish) between the strain in both experiments this proved to be insignificant. Hatching percentage 1in the new strain (80.58, 51.44%) in both experiments was significantly higher than the old strain (42.46, 33.08%) respectively (ANOVA, F=11.140, P=0.001, df=3). The same was noted for hatching percentage 2, the new strain (70.20%) was significantly higher than the old strain (41.16%) (ANOVA, F=33.315, P=0.001, df=1). In hatching percentages one and two in the second experiment, the percent hatching for the new strain was almost double that of the old strain. The fertilization percentage in the old strain in the second experiment ranked higher than the new strain, which was in contrast to the first experiment where the new strain was higher than the old strain. However, the differences among the four values are not significant (ANOVA, F=1.058, P=0.371, df=3). There was a significant difference between strains for wet weight of three-day post hatch fish/ total amount of egg stocked in incubator when analysed by simple ANOVA but post-hoc analysis with Tukey’s test proved to be insignificant (ANOVA, F=3.22, P=0.047, df=3). Table 3.1 Reproductive traits of two silver carp strains New Strain 2 Old Strain 2 New Strain1 Old Strain1 a a a a Female weight (kg) 2.00 1.57 1.92 1.57 (0.12) (0.03) (0.30) (0.23) a a a a Male weight (kg) 1.88 2.03 1.98 1.83 (0.07) (0.12) (0.13) (0.08) a b a c Female length (cm) 52.67 44.67 53.83 46.4 (1.33) (0.33) (1.30) (1.22) a a a a Male Length (cm) 48.83 49.33 48.50 48.17 (0.83) (1.76) (1.26) (0.77) a a a a Weight of eggs (g)/g of body weight 0.10 0.10 0.14 0.10 (0.01) (0.01) (0.02) (0.03) a a a a Milt Volume (ml) 5.67 7.33 4.17 4.93 (0.17) (2.33) (1.23) (0.64) a a a a Percent Fertilization 78.43 86.97 80.51 77.55 (6.5hr post fertilization) (0.04) (0.02) (0.04) (0.03) a b bc bd Hatching percentage 1 80.58 42.46 51.44 33.08 (24hr post fertilization) (0.06) (0.04) (0.05) (0.04) a b Hatching percentage 2 70.20 41.16 (29hr post fertilization) (0.25) (0.29) a a a a Wet weight of 3 day post hatch fish (gm)/ 1.15 1.10 0.95 0.79 total amount of eggs stocked (gm) (0.10) (0.11) (0.07) (0.08) Values with different alphabetical superscript were significantly different from each other (P<0.05). Standard errors of the means are in parentheses. 1 Experiment 1 2 Experiment 2 Table 3.1.1 Mean fork length of three-day-old larvae for NFEP new and old hatchery strain and commercial hatcheries Description CM-1 CM-2 CM-3 NS OS a a a b c Fork length (mm) 6.46 6.37 6.32 7.21 6.95 (0.03) (0.04) (0.03) (0.06) (0.03) -3 CM-1(Bai Bai Hatchery), CM-2 (Sagar Hatchery), CM (Star Matshya Khamar hatchery) NS (New strain), OS (Old strain) Values with different alphabetical superscript were significantly different from each other (P<0.05). Standard errors of the means are in parentheses. Table 3.1.2 Mean survival, weight, length and production for silver carp from 20 days Description CM-1 post CM-2 stocking in CM-3 hapas STN STO NS OS a b b b a a a Survival % 55.62 87.71 76.31 91.36 62.63 68.62 65.6 (0.05) (0.05) (0.05) (0.03) (0.02) (0.05) (0.04) a b a a a a a Weight (gm) 0.0651 0.0346 0.0472 0.0499 0.0648 0.0605 0.0672 (0.01) (0.00) (0.00) (0.01) (0.00) (0.00) (0.01) a b a a a a a Fork length (cm) 1.52 1.21 1.33 1.40 1.51 1.50 1.51 (0.06) (0.03) (0.05) (0.04) (0.04) (0.03) (0.06) -2 a a a a a a a Production (g m ) 42.25 41.50 43.00 46.50 40.50 45.33 41.00 (1.32) (0.96) (2.88) (4.35) (2.06) (3.53) (1.29) -3 CM-1(Bai Bai Hatchery), CM-2 (Sagar Hatchery), CM (Star Matshya Khamar hatchery) STN (Simulated transfer new strain), STO (Simulated transfer old strain), NS (New strain), OS (Old strain) Values with different alphabetical superscript were significantly different from each other (P<0.05). Standard errors of the means are in parentheses. Table 3.1.3 Mean fork length gain for silver carp from 20 days post stocking in hapas Description CM-1 CM-2 CM-3 STN STO NS OS a b a a a a a Fork length gain (cm) 1.45 1.15 1.27 1.32 1.44 1.43 1.44 (0.06) (0.04) (0.05) (0.04) (0.04) (0.03) (0.06) -3 CM-1(Bai Bai Hatchery), CM-2 (Sagar Hatchery), CM (Star Matshya Khamar hatchery) STN (Simulated transfer new strain), STO (Simulated transfer old strain), NS (New strain), OS (Old strain) Values with different alphabetical superscript were significantly different from each other (P<0.05). Standard errors of the means are in parentheses. The fork length of larvae before stocking in hapas is summarized in Table 3.1.1 The new strain (7.21mm) was significantly greater in size than the old strain (6.95) followed by CM-1 (6.46), CM-2 (6.37) and CM-3 (6.32mm). Both new and old strains were significantly different from the commercial hatchery fish (ANOVA, F=72.22, P=0.001, df=4). 3.2 Analysis of Hapa trial Table 3.1.2 summarizes the data obtained for the 20-day hapa trail, it can be seen that there were significant differences in survival rate with STN (91.36%), CM-2 (87.71%), and CM-3 (76.31%), having a significant higher survival rate compared to NS (68.62%), OS (65.6%), and CM-1 (55.62%) (ANOVA, F=10.72, P=0.001, df=3). Statistical analyses showed that CM-1 (.065g), STO (.065), NS (.061), OS (.067) had a significantly higher mean weight than CM-2 (.0346g) (ANOVA, F=4.49, P=0.005, df=6). This same ranking was also noted with respect to fork length where CM-2 was smaller in size (1.21cm) compared to CM-1 (1.52), STO (1.51), NS (1.50), OS (1.51cm) (ANOVA, F=6.235, P=0.001, df=6). ). For production STN (46.5) and NS (45.33gm-2 ) had the highest mean value although there was no significant differences between the seven treatments (ANOVA, F=.768, P=0.604, df=6). Table 3.1.3 shows the mean fork length gain after 20 days of the trail period. Statistical analyses revealed that CM-1 (1.45cm), STO (1.44cm), NS (1.43cm), OS(1.44cm) had a significantly greater mean fork length than CM-2 (1.15cm) (ANOVA, F=6.03, P=0.001,df=6) 3.3 Results of social survey questionnaires 3.3.1 The level of experience of nursers interviewed Of the thirteen respondents surveyed the nursery questionnaire at NFEP hatchery, 4 (31 percent) had 1 to 5 years experience, 4 (31 percent) had 6 to 10 years experience, and 5 (38 percent) had over 10 years experience (refer to figure 3.3.1). 40 % of respondents 30 20 10 0 1-5 6 - 10 > 10 Years Figure 3.3.1 The level of experience of nursers interviewed 3.3.2 Species of fish cultured by nursers in relation to the species produced Responding to the question “what species did you nurse in your last production cycle and what species makes up a large proportion of what is produced”, 14 (100 percent) nurse silver carp and 12 (85.7 percent) replied it is the greatest proportion of fish cultured, 12 ( 85.7 percent) nurse rui/mrigel and 10 (71 percent) replied it is the greatest proportion of fish cultured, 8 (57 percent) nurse catla, 6 (43 percent) nurse grass carp, 11 (78 percent) nurse shorputi and 2 (14 percent) replied it is the greatest proportion of fish cultured, 3 (21 percent) nurse mirror carp, 6 (43 percent) nurse bighead carp, and 4 ( 28 percent) nurse common carp and 2 (14 percent) replied it is the greatest proportion of fish cultured (refer to figure 3.3.2). 3.3.3 Species of fish considered the least problem to nurse Respondents reported that silver carp was the least problem to nurse11 (79 percent) followed by rui/mrigel, 2 (14 percent) shorputi, 2 (14 percent) common carp, and 1 (7 percent) respondent catla was easy to nurse (refer to figure 3.3.3). 3.3.4 The reason why nursers perceive some species as being least problem to nurse and also comprise a large proportion of their production Of the fourteen respondents to the question why they considered some species the least problem to nurse and which also ha ve the highest production, silver carp had the highest number of responses 11 (79 percent) with demand receiving the most responses. Rui/mrigel had the second highest response 9 (64 percent) as being easy to produce and or has a high demand. The lowest was f shorputi and common carp with 2 (14 percent) each (refer to table 3.3.4). Table 3.3.4 The reason why nursers perceive the following species as being least problem to nurse and also comprise a large proportion of their production High Species survival Demand Fast growing Easy to produce Profit Total Row % Silver Carp 2 5 2 2 11 79 Rui/Mrigel 1 3 2 3 9 64 Shorputi 1 1 2 14 Common Carp 1 1 2 14 3.3.5 Assessment of hatchling quality by nursers Of the fourteen respondents asked how they assess hatchling quality, 9 (65 percent) reported that they assessed quality by how strong they swim against the current. 3 (21 percent) relies on the judgment of the hatchery manager. 1 (7 percent) answered quality by colour and another 1 (7 percent) by tapping on the container and seeing the hatchlings response (refer to figure 3.3.5). Strong swimmer/movement against 7% the current 7% Relies on hatchery manager colour 21% reacts to stimulus 65% Figure 3.3.5 Indicators used to assess hatchling quality by nursers 3.3.6 Species that are considered by nursers to have the best survival rate The eleven respondents to the question what species has the best survival rate during transfer from hatchery to nursery pond, 8 (73 percent) rui/mrigel, 2 (18 percent) catla, 2 (18 percent) shorputi, 3 (27 percent) silver carp, and 1 (9 percent) common carp (refer to figure 6). But overall it was suggested that silver carp had the lowest survival rate compared to other major cultured species except for shorputi. 80 % of respondents 70 60 50 40 30 20 10 0 Siver Carp Rui/mrigal Shorputi Common Carp Catla Figure 3.3.7 Fish species that are considered to have the best survival rate during transfer from hatchery to nursery pond 3.4 Results Traders Questionnaire Trader Background Traders were interviewed at the Parbatipur railway station in the evening. Two different locations on the railway platform were used to interview the traders. All traders interviewed sell directly to the pond farmer and the average travel time to the farmer was six hours traveling by using all forms of transportation (train, rickshaw, van, bike,) and all transferred fish by patil. 3.4.1 The level of experience of traders interviewed Of the fifteen respondents interviewed in the trader questionnaire at the Parbatipur railway station, 9 (60 percent) had less than five years experience, 4 (27 percent) had six to ten years experience and, 2 (13 percent) had over ten years experiences (refer figure 3.4.1). 80 % traders 60 40 20 0 1-5 6 - 10 >10 years Figure 3.4.1 The level of experience of traders interviewed 3.4.2 The fish species considered by traders to make up a large proportion of fish trade and why Of the fourteen respondents asked what is the most traded species and why are they traded the most 11 (73 percent) responded shorputi because they are fast growing (3 ) and demand is high (1). 14 (100 percent) for common carp because they are fast growing (2). For rui/mrigel 11 (73 percent) because the demand is high (3) and are fast growing (1). Bighead carp 4 (27 percent) was considered fast growing (1) by one respondent. Silver carp on the other hand 14(100 percent) was traded the most because of it fast growth (6) and high demand (5). Grass carp 6 (40 percent), catla 9 (60 percent) and African Catfish 1 (7 percent) had no response for why they were considered being traded the most. 250 Demand is high 200 % of respondents Fast Growing 150 Percent response 100 50 0 tla uti al p rp arp rp h Ca rig ar orp tfis Ca Ca rC sC i/m Ca Sh on ad lve Ru as m an he Si Gr m Big ric Co Af Figure 3.4.2 The fish species considered by traders to make up a large proportion of the fish trade and why 3.4.3 Fish species considered by trader the best for transferring and why The fifteen traders asked what fingerling is the best to transfer and why the most response was for common carp 14 (93 percent) because it is considered to be strong and hardy 10 (100 percent). Rui/mirgel and shorputi had the lowest response receiving 1 (7 percent) for transferring and 1 (10 percent) because it is strong hardy (refer to table 3.4.3). Table 3.4.3 Fish species considered by trader the best for transferring and why Species Number of responses Strong/Hardy Total n = 15 n =10 t Silver Carp 2 1 3 Rui/Mrigel 1 1 2 Shorputi 2 1 3 Common Carp 14 10 24 Grass Carp 4 3 7 Bighead Carp 2 2 4 3.4.4 Indicators used to assess hatchling quality by traders Traders were asked how they judge seed quality, the majority replied by movement 11 (59 percent). 4 (26 percent) assessed quality by strength. They judge strength by putting the fingerling in their hand and making a fist and see how hard it squirms. A small percent judge quality by eye colour and size 1 (5 percent), 2 (10 percent) respectively (refer to figure 3.4.4). 10% Movement 26% eye color strength 59% size 5% Figure 3.4.5 Indicators used to assess hatchling quality by nursers 3.4.6 Assessment of silver carp quality in the past five years by traders with over five years experience Traders with over five years experience were ask the quality of silver carp over the last past five years. Most traders felt the quality of silver carp declined 8 (62 percent), while some felt it increased 2 (25 percent) or remained the same 1 (13 percent)(refer to figure 3.4.7). 13% 25% Increase decrease same 62% Figure 3.4.7 Assessment of silver carp quality in the past five years by traders with over five years experience 3.5 Questionnaire Results for Male Aggregate Workers Male Aggregate Workers Background Male aggregate workers consisted of 12 males breaking bricks in various locations. Brick breaking involved breaking bricks throughout most of the day, by squatting or sitting on a brick and using a small hammer (head of the hammer the size of a 50 pence coin) and breaking large pieces of brick into smaller pieces close to pebble size. Brick breakers get paid by the amount of bricks they break or the amount of broken bricks they produce. Breaking brick is considered an unskilled labour and pay could range from 63 – 50 taka per day for males, while females might receive 40 to 46.6 taka per day (NSP 1998). Male aggregate workers were interviewed in three different locations, Dinajpur, Biral, and Rangpur. The youngest age was 12 and the oldest 70 (the average age was 41.5). The average days worked was 4.5 days a week and 80 percent of the respondents were landless (refer to table 3.5.1). Female Road Builder Background Female road builders consisted of 12 females doing different types of road construction. They were interviewed in two different locations but on the same road in Biral. The main labour the women were performing was carrying crushed rocks in baskets (weight 7- 10kg) on top of their heads and walking 10 meters and dumping it on the road. Other tasks involved filling the baskets or sweeping the road. The women were paid a set rate by the day, approximately 40 taka (personal communication, NFEP-2). The youngest age was 22 and the oldest was 50 and the average age was 32 (refer to table 3.5.1). Table 3.5.1 Background information on Male aggregate workers/ Female road builders Number of Number of Characteristics Male Female Respondents Male Respondents Female Average age 12 41.50 12 32.00 Average family size 12 4.90 12 5.00 The percent that own land 3/12 25 9/12 75 The average amount of land owned (decimals) 3 19 9 9.70 The percent landless (lives on government land) 9/12 75 3/12 25 Average worked days/week 12 4.5 12 4.40 The amount of times fish is consumed per year 9 63.6 12 26.88 The amount of times chicken is consumed per year 9 5.16 12 3.90 The amount of times other meat is consumed per year 9 37.92 12 3.80 Who buys the fish in the family Respondent 6/12 50 Husband 8/12 67 The percent of who consumes the most fish in the family Respondent 1/12 8 Husband 5/12 42 Mother 1/12 8 Daughter 2/12 17 Son 2/12 17 Equal 1/12 8 3.5.2 Male aggregate worker fish purchasing results 3.5.3 The percent of species most frequently bought buy male aggregate workers Of the twelve respondents the majority went to bazaars 10 (83 percent) and haats 5 (42 percent) to purchase fish. Of the twelve respondents asked what fish species is most frequently bought, 2 (17 percent) mrigel, 2 (17 percent) rui, 7 (58 percent) silver carp, 1 (8 percent) catla, 1 (8 percent) tengra, 1 (8 percent) mola, 1 (8percent) darika, 5 (42 percent) shorputi, and 2 (17 percent) sharti (refer to figure 3.5.3). 70 60 % of respondents 50 40 30 20 10 0 i ola ti Ru ra tla el ti a Pu ar p rik rig ng Ca ar M Sh Da M rC Te lve Si Figure 3.5.3 The percent of species most frequently bought buy male aggregate workers 3.5.4 Prices of the most frequently purchased table fish by male aggregate workers Male aggregate worker were asked what are the prices per kilogram of the most frequently bought fish. The results are summarized in table 3.5.4. The most expensive fish was rui 53 taka kg-1 the least expensive was darika 23 taka kg-1. Silver carp was the least expensive of the cultured fish 30 taka kg-1 compared to rui. Price convergence is listed in table 3.5.4a. Currency Taka 1 US dollar 51 1 Pound 72 Table 3.5.4a Taka/currency Table 3.5.4b The average price in taka per kilogram of table fish purchased most frequently by male aggregate workers -1 Species Taka kg Silver Carp 30 Rui 53 Mrigel 47 Puti 24 Sharti 30 Tengra 40 Mola 36 Chanda 30 Darika 23 3.5.5 The species that are consider by male aggregate workers to be the most available in markets Of the twelve respondent that were asked which fish species is the most available in the markets, 2 (17 percent) rui, 2 (17 percent) catla, 2 (17 percent ) ilish, 1 (8 percent), and 9 (75 percent) for silver carp (refer to figure 3.5.5). 80 % of respondents 70 60 50 40 30 20 10 0 Silver Boal Rui Catla Ilish Carp Figure 3.5.5 The species that are consider by male aggregate workers to be the most available in markets Female road builder fish purchasing results 3.5.6 The percent of species most frequently bought in the household of female road builders Of the twelve respondent 10 (83 percent) purchase fish from bazaars, 3 (25 percent) purchase from hats, and 3 (25 percent) from roadside market. When asked what fish is bought most frequently 4 (33 percent) replied sharti, 9 (75 percent) puti, 1 (8 percent) ilish, 3 (25 percent) silver carp, 1 (8 percent) rui, and 1 (8 percent) bata (refer to figure 3.5.6). 80 70 60 percent 50 40 30 20 10 0 Rui Bata Sharti Puti Ilish Silver carp Figure 3.5.6 The percent of species most frequently bought in the household of female road builders 3.5.7 Prices of the most frequently purchased table fish by female road builders Female road builders were asked what are the prices per kilogram of the most frequently bought fish. The results are summarized in table 3.5.7. The most expensive fish was chingri and ilish 80 taka kg-1 the least expensive was puti 19 taka kg-1 . Silver carp was the least expensive of the cultured fish 38 taka kg-1 compared to rui 40 taka kg-1 . Table 3.5.7 The average price in taka per kilogram of the most frequently purchased fish in female road builder households. -1 Species Taka kg Silver Carp 38 Rui 40 Ilish 80 Chingri 80 Puti 19 Sharti 26 Darika 12 3.5.8 The species that are considered by female road builders to be the most available in markets Of the twelve female respondent that were asked which fish species is the most available in the markets, 2 (17 percent) rui, 1 (8 percent) catla, 2 (17 percent) puti, 1 (8 percent) mola, 8 (67 percent) silver carp, and 1 (8 percent) sharti (refer to figure 3.5.8). 80 70 60 percent 50 40 30 20 10 0 Shol Puti Silver Mola Rui Sharti Catla carp Figure 3.5.8 The species that are consider by female road builders to be the most available in markets 3.6 Market Survey Results Background Dinajpur Dinajpur has three fish markets, with Bahadur Bazaar being the largest. The two other are Gudara Bazaar (Chalk ma rket) and Rail Bazaar that only opens on Thursday and Sunday. Bahadur Bazaar was sampled close to the end of the peak hours of fish buying which is early in the morning and evening. Rangpur Pour (Municipal) Market Is one of the biggest fish markets (in terms of supply and demand) in the northwest area and was established in 1962. On average throughout the year it is estimated that 5,000 consumers visit the fish market daily. There are 10 smaller markets within Rangpur, which are supplied from the Pour Market (NFEP, market data). Only a few retailers were interviewed at Pour market. Table Fish Market Questionnaires Dinajpur 3.6.1 The proportion of species observed at Dinajpur market in relation to the most quantities observed Of the 10 retailers stalls examined there were ten different species being sold. Rui and mrigel were the most prevalent species observed at the retailer stalls, 8 (21 percent) and 9 (24 percent) respectively. Silver carp was the third most prominent species observed at the retailer stalls 7 (18 percent). Silver carp might have been the third most prevalent species observed, but it had the highest quantities sold per retailer stall 6(50 percent) followed by mrigel 4 (33 percent) and rui 2 (17 percent). The lowest frequency of species observed was grass carp, tilapia, african magur, and common carp 1 (3 percent) (refer to figure 3.6.1). 60 Percent observed 50 Most quantities 40 30 20 10 0 i Ru tla el ia uti arp rp arp Ca rig ap orp ur Ca M Til ag rC sC Sh nM on ve as m Sil Gr ica m Afr Co Figure 3.6.1 The proportion of species observed at Dinajpur market in relation to the most quantities observed 3.6.2 The species retailers sell the most quantities at Dinajpur market Of the ten retailers asked which fish species do you sell the most of, 8 (80 percent) replied silver carp while 2 (20 percent) replied rui. 3.6.3 The average price in taka/kg in Dinajp ur market of species by weight per kilogram Average prices of table fish were assessed from the ten different retailer stalls; the results are summarized in Table 3.6.3. Silver carp had the lowest cost per kilogram at all weights, compared to rui, which was the most expensive table fish sampled. 3.6.3 The average price in taka kg- in Dinajpur market of species by weight per kilogram Species < 250g 251 – 500g 501 – 1000g Silver Carp 38 40 54 Rui 65 100 Mrigal 56 60 Catla 52 70 Grass Carp 70 Tilapia 60 African Magur 50 Shorputi 50 4 DISCUSSION 4.1 Hatchery trials Reproductive performance of strains The new strain showed no difference in fertilization rate when compared to the old strain, being the case for both experiment one and two. There was however, a significant difference in hatching rate with the new strain having a higher rate of hatching than the old strain at 24 hr and 29 hr post fertilization for the second experiment. The same was seen in the first experiment at 24 hr post fertilization. Overall the old strain showed a significantly lower hatching rate than the new strain, which could be an effect of inbreeding. Several investigators have suggested that inbreeding can cause a decline in fertilization rates and other reproductive performances (Kincaid, 1983; Falconer, 1981). Despite the low hatching rate found in the old strain, the amount of grams of wet weight of larvae produced at the end of the larval rearing period divided by the amount of grams of eggs stocks there found to be no significant differences between the old strain and new strain. This could be attributed to the difficulties in obtaining an accurate estimation in the incubators. Five to six replicates were carried out for hatching percentage one and ten replicates was conducted for hatching percentage two. Ten replicates were studied for hatching percentage two to determine if there were any differences. There was no significant difference between hatching percentage 1 and 2 for the new strain or for hatching percentage 1 and 2 for the old strain in the second experiment. The ineffective estimation is a result of the large volume of the incubator and the low stocking density. The incubators were 800l concrete funnels with the water coming up from a screenless bottom and leaving from the top. The flow rate was slightly different for each incubator and changed daily depending on the volume of the header tank. It was observed that a higher flow rate would have a denser pop ulation of eggs/larvae in the center of the incubator while a slightly lower flow rate will have a low density in the center of the incubator. The flow effect in the incubators could of influence the sampling results, by not giving a true estimation of what was being measured. At the completion of the hatchery trial there was found to be a significant difference in fork length between three day old larvae of the new and old strain, with the new strain being significantly larger than the old strain. This difference in size could be due to strain or parental differences. The old strain might have shown a reduction in size compared to the new strain due to inbreeding depression. It has been suggested by Hussain and Mazid (1999) that this depression due to inbreeding has led to a decline in the growth performance in carps in Bangladesh. In fact an earlier study by Moav and Wohlfarth (1976) showed that a single full sib mating of broodstock might result in 10 – 20% depression growth in progenies, exhibiting the potential “strength” of this factor. Alternatively the size and weight of the female could have an influence on the difference in size between the new and old strain larvae. The average weight of the new female (2.00kg) was larger than the old strain (1.57kg) but not significant. Larger fish tend to have a lower relative fecundity and bigger eggs than smaller fish. Thus the size difference could also be correlated to the fact that larger females tend to produce larger larvae compared to smaller females. 4.1.2 Commercial Hatcheries The general background of the three commercial hatcheries is summarized in Table 4.1.2 Table 4.1.2 Background information of commercial hatcheries used for the hapa trial Characteristics CM -1 CM -2 CM-3 Hatchery established 1990 1988 1989 No. of brood fish ponds 4 5 1 No. of circular tanks 4 2 2 No. of hatching jars 13 20 0 Source of broodfish Outside source Rear own Rear own Production last year 500 kg 1000 kg 250 kg Source NFEP-2 hatchery survey 1999 CM-1 (Bai Bai Hatchery), CM-2 (Sagar Hatchery), CM-3 (Star Matshya Khamar hatchery) The size of the broodstock used is unknown from the three commercial hatcheries, but looking at the background data it can be suggested that CM-2, CM-3 have more of a chance of inbreeding depression in broodstocks used than CM-1, who brings in new founder stock for breeding purposes. When interviewing CM-1, it was found that the owner brought in new broodstock to prevent inbreeding problems. The owner would then spawn three females from an outside source with two males on farm or vice/versa. By this recruitment of new breeders into the stock at regular intervals the chances of inbreeding depression, are reduced as is the risk of a reduction in growth and reproductive performance, increased incidence of diseases and morphological deformities (Eknath and Doyle 1990). With the growing concern about genetic deterioration in hatchery stocks, several scientists from the Bangladesh Fisheries Research Institute conducted a survey of hatcheries in Jessore and Commilla in 1994 to assess hatchery practices and found that the average weight of broodstock used for silver carp was 0.800kg – 1.00kg. Hussain (1998) suggesting that using undesirable sized breeders can results in early maturation and ultimately artificial seed production effort leads to stock deterioration problems because of poor growth and survival in the grow-out systems. The decrease in size in broodstock could be due to the effects of inbreeding or the unconscious ‘negative’ selection in hatcheries. Negative selection occurs when hatcheries collect fish from grow-out ponds after larger individuals (faster growing) have been sold and the smaller (hence slower growing) fish are used for broodstock replacement. Negative selection has been suggested as one of the major reasons of reducing the growth and other performance traits of farmed fish in Bangladesh (Hussain, 1998). There is evidence for this observation in India, where in small hatcheries younger individuals are exhibiting relatively slower growth rates, late maturity and also smaller size at first spawning, which has been related to the inadvertently breeding of slow- growing and late-maturing individuals (Eknath and Doyle, 1990). 4.2.1 Hapa trial There were significant differences in survival rates after 20 days of nursing. STN, CM-2, CM-3 had significantly higher survival rates when compared to NS, OS, and CM-1. STN had the highest value for survival rate (91.36) but the difference was not significant compared to CM-2, and CM-3. It would be expected that the non transferred larvae would have a higher survival rate than the simulated transferred fish and commercial hatchery fish, as the non-transferred fish were subjected to less stress. Handling, netting, and transportation are the most common causes of acute stress in fish, which can have detrimental effects on both survival and health (Barton and Iwama 1991) A possible explanation for higher survival observed in STN, CM-2, and CM-3 could be due to the samples taken from the transferred polyethylene bag. Simulated transfer fish and commercial hatchery fish were stocked at 150 gm of larvae per bag. The bag was then emptied over a fine mesh net to co llect the larvae. Stressed, dead or dying fish tend to inhabit the surface of the water, by emptying the bag it is possible that the overly stressed fish will tend to be located at the bottom of the fine mesh net while the stronger and healthy fish will tend to be at the surface of the strained fish. A beaker was used to collect a proportion of the fish. If the majority of the sample was taken near the surface of the pile it might contain a higher percentage of healthy and stronger larvae than if the sample was a cross section of the pile of fish containing a true representative of the distribution of fish in the bag. This might explain the observed differences in survival rate. But it was noted in experiment one there was a noticeably higher percentage of dead, dying and severely stressed fish at the water surface in STO then STN. This might suggest that the old strain is more sensitive to stress than the new strain, but further investigation is required to determine this effect. There was no difference in survival rate between NS and OS, which might suggest that the perceived quality difference might occur later on in rearing. The significant differences in weights and fork lengths after 20 days can be accounted for by the hapa survival/weight relationships, where low survival rates in hapas tend to produce larger size fish than hapas with high survival rates. This was observed with CM- 1 and CM-2. CM-2 (87.71%) had a significantly higher survival rate than CM-1 (55.62%), but CM-2 (1.21) had a significantly lower fork length than CM-1 (1.52). For production STN (46.5) and NS (45.33gm-2 ) had the highest mean values although there were no significant differences between the seven treatments. It might be that the new strain is more productive than the old strain but in the first twenty days it is too early to show a significant difference. A longer trial period might determine these effects. A comparison trail was done previously at NFEP-2 Parbatipur hatchery, where silver carp fingerlings produced from the existing old strain were compared to the new strain and a cross between the new and old strain. At the end of the trial period, lasting 271 days, the results showed that there was 21.6 percent increase in mean net production from the new strain of silver carp compared to the old strain. This suggest that the new strain does indeed perform better than the old strain in the NFEP-2 strain comparison, but unfortunately the strains were raised in different ponds and stocked at different initial weights which could have had an influence on the results. There are many factors which could have an impact on the overall performance of silver carp seed, such as maintenance of proper stocking density of broodfish and their balanced feeding, selecting broodfish of a desirable size and quality, injecting adequate hypophysation dosage, mating unrelated female and male breeder, basic disease control, and transport/handling stress. Some of these factors could of influence the performance of silver carp during the hatchery and hapa trial but there was no significant evidence except for the fact that the new strain produced larger size larvae than the old strain and commercial hatcheries, which could be caused by inbreeding depression and negative selection. The hapa trial period was to short too ascertain these effects on silver carp seed quality. Significant differences may be observed later in the rearing period, or the variety of actors (nursers, traders, farmers) potentially could have an impact on the quality of silver carp seed as it passes through their hands. 4.3 The assessment of the importance of silver carp to the livelihoods to the people of Northwest Bangladesh 4.3.1 Nursers The importance of silver carp and the assessment of seed quality can be observed through nursers. The nursers in northwest Bangladesh generally acquire hatchlings from private, government or NGO’s within the region. Questionnaire results found that 30 percent go to private hatcheries, 79 percent government, and 42 percent NGO’s. It has been suggested that experience nursers go to government hatcheries to buy hatchlings because the hatchery managers are more knowledgeable about the quality of seed (SOS, Bangladesh, 1999). Questionnaire results determined that majority of nursers judge hatchling quality by movement against the current or being a strong swimmer (65 percent). These findings are similar to the results from the SOS for Bangladesh. This might suggest that the majority of nursers are aware of the importance of initially having good quality seed before stocking in their nursery ponds. This finding is consistent with the data from the SOS report for Bangladesh, which determined that nursers are concerned with seed quality, and are willing to pay more to obtain good quality hatchlings to nurse. Nursers tend to nurse silver carp for 15 to 30 days before selling with the average stocking density being 40g 40m-2 . Some nursers kept these high stocking densities throughout the nursing period while some lower the densities after 15 days. By keeping the silver carp at high stocking density throughout the nursing period nurseries can produce fry that will be smaller than average size that will be sold to the traders. The use of high stocking densities to maximize water usage has been shown to exert severe adverse effects on growth (Kebus, Collins, Brownfield, Amundson, Kayes, and Malison 1992). Farmers than would be stocking ponds initially with smaller size seed that might be of lesser quality. In the SOS for Bangladesh it was determined that the Northwest produces a greater range of seed size through different stocking densities to meet the requirement of local traders as well as direct sales to the farmers. Through the questionnaire it was determined that nursers considered rui/mrigel to have a better survival rate than silver carp. The majority of the nursers interviewed felt that silver carp had one of the lowest survival rates other than shorputi compared to all the species they commonly nurse. This finding is consistent with the findings of the NFEP survey on fingerling production in NFEP command area (1997). Table 4.2.1 summarizes these findings. The NFEP results demonstrate that rui/mrigel have a higher survival rate than silver carp followed by shorputi, which supports the perceived survival rate by nursers. Table 4.2.1 A summary of the amount of fingerlings produced from 1k of hatchling used by nursers, and the survival rates for 1994 and 1995 Number of fingerlings Mean number of Percent Survival produced/kg of hatchling hatchlings in 1 kg Species 1994 1995 1994 1995 Silver Carp 63,299 61,171 200,000 31 31 Rui/mrigel 80,719 75,927 175,000 46 43 Shorputi 33,133 54,967 550,000 6 10 Source NFEP fingerling production 1997 Even though silver carp has a low survival rate it is still one of the predominant species being produced. Results show that 100 percent of the respondents interviewed nursed silver carp in their production cycle and 86 percent replied it was the species that was produced the most of by weight. It was also assessed in the questionnaire that over 50 percent of production was silver carp. A survey done by NFEP from 1994 to 1995 in the NFEP command area showed that 62,000 silver carp fingerlings were sold, which was second in amount to rui/mrigel where 78,000 fingerlings were sold, but one has to consider that two different species are being compared to one species. This would suggest that silver carp is probably the highest traded species. Results indicate that silver carp is considered the least problem to nurse by 70 percent, which was the highest species considered by the respondents. There were several reasons why nursers perceived silver carp the least problem to nurse and makes up a large proportion of their production. One of the main reasons was demand; some of the other reasons were high survival, fast growing and easy to produce. Mid -season and early season silver carp hatchling are perceived by nursers to be of good quality. These results are similar to the finding in the SOS for Bangladesh. However, late season seed is considered poor quality. This might be due to the effect of hatcheries selling off the large broodstock early in the season as table fish, which would require using undersized (possibly slow growing) individuals as breeders late in the season. Another possibility for poor quality seed late in the season is the repeated use of hormonal induction on the same broodstock through out the spawning season, which might be detrimental to ooycte development, thus effecting larval quality. 4.4.1 Traders The major aim of the trader questionnaire was to determine the quality of silver carp and its importance to the people in the northwest. Traders are the last link in the network linking hatcheries to farmers; they bear some of the greatest risk in the network and yet receive marginal benefits, that is why it is important to them to assess the quality of the seed before making the journey to the farmers pond side. The results determined that 59 percent of traders assess fingerling quality by movement and another 26 percent by strength. Strength is assessed by making a fist around the fingerling and feeling how strong it wiggles. It is important to the trader to start off with good quality seed, because low quality seed not only affects the farmer but also affects the trader’s livelihood. Traders transport many different species. The species they consider that makes up a large proportion of the trade are silver carp and common carp, which received a 100 percent response. There were several reasons why the trader considered silver carp as a high volume species, because of the demand by the farmer and it is fast growing. A study done by NFEP -2 showed a steady increase of silver carp entering Parbatipur railway station from 1993 to 1996, however they are the second most abundant species with rui/mrigel/catla being the most abundant in numbers and percent composition (refer to table 4.3.1) Silver Carp Rui/Mirgel/Catla Year Numbers Percent Numbers Percent 1993 12,895 20.50 27,395 43.50 1994 21,622 15.70 69,015 50.20 1995 37,878 14.00 143,982 53.00 1996 49,438 22.80 80,387 38.20 Source NFEP-2 Paper 20 Table 4.3.1 The number and percent composition of fry and fingerlings of silver carp and rui/mrigel/catla entering Parbatipur Railway Station for 1993 to 1996 The table illustrates the steady increase of fry and fingerlings of silver carp into Parbatipur Railway Station. Overall rui/mirgel/catla have higher numbers (in 1996 it they were close to double that of silver carp), but one has to consider that three different species are being compared to one species. This reasoning would suggest that silver carp might have close to the highest values in numbers and percentages during 1996 which would reflect the results that were seen in the questionnaire where silver carp made up one of the highest proportion of species being traded. Although the fry traders perceive silver carp as a dominant species being traded, it is considered one of the least favorites to transfer due to its low survival. Common carp was the species of preference due to its hardiness during transfer. These results agree with the results from the nursers that silver carp has a low survival rate compared to the majority of other species. It also demonstrates the importance of silver carp as a species, because in spite of its low survival rate it makes up a large proportion of the species being nursed and traded. Results for the questionnaire showed that 62 percent of the respondents perceived the quality of silver carp has declined in the past five years, while 25 percent thought it had increased due to better management, and the remainder 13 percent felt it has not change. One trader with ten years experience suggested that the decline in silver carp was due to hatcheries using undersized broodstock. This suggestion is consistent with the findings of Hussain and Mazid (1997) where the average size broodstock being used for production for silver carp was 0.8 – 1.0 kg, and their suggested minimum size is 1.5kg. The use of undesirable sized breeders can result in seed with poor growth and survival in the grow-out system. 4.5.1 Consumers The results indicate that the average days worked per week were approximately the same 4.4 for females road builders and 4.5 for males aggregate workers. There was a large difference between the amount of fish consumed per year with males eating fish 63.6 times per year (5.3/month) and females eating 26.88 times per year (2.24/month). Females responded 42 percent in the questionnaire that their husbands eat the most fish in the house. It is obvious that females within the household get far less fish than males. This is consistent with the finding of the IFPRI study, which found that adult women received a lower proportion of preferred foods (IFPRI, 1999). The women road builders also responded that after the husband the children eat the most fish in the house. Chicken was eaten 5 times per year by males aggregate workers and 4 times by females road builders. Males also ate a disproportionate amount of meat compared to females, consuming meat 38 times per year and females 4 times per year. The low numbers for meat consumption of the women might be influenced by their religious background. The questionnaire demonstrates that fish is the main source of protein for unskilled laborers who tend to be landless especially in the case for the men interviewed, and depend exclusively for their sustenance on their wages earned. These people walk on a thin line, ready to plunge down to severe malnutrition when any imbalance between their daily wages and the price and availability of fish occurs. The results indicate that silver carp was the most frequently bought table fish by male aggregate workers followed by puti then sharti. This was contrary to the female road builder response. They considered puti being the most frequently purchased table fish followed by sharti and silver carp. Although male aggregate workers purchased silver carp most frequently it is not dependent on the price. On the other hand female response to most frequently purchase table fish did follow a trend with price, where the least expensive table fish is the most purchased. The least expensive table fish quoted by male aggregate workers was puti and darika costing 24 taka kg-1 and 23 taka kg-1 respectively. Silver carp, shati, and chandra all were given the same price of 30 taka kg-1 . Silver carp was the least expensive of the culture species, with rui and mrigel being the most expensive. It was also determined that 50 percent of the female workers respondents purchase the fish while 67 percent were their husbands. This might reflect the discrepancy in table fish prices observed between the male and female workers. Availability might have more of an influence on purchasing factors for silver carp. 75 percent of the male worker respondent that silver carp was the most available table fish in the markets, while female road builders had a 67 percent response. This is consistent with NFEP sampling marketing data, of which a brief account is summarized in Table 4.4.1 It can be seen in Table 4.4.1 that silver carp has proportionally higher quantities than both tangra and mola, which confirm the perception that silver carp is the most available table fish in the market. Male and female workers responded that the reason why they purchase silver carp was because of the price, which received a 100 percent response. Table 4.4.1 Sums of quantity in kg of table fish at Rangpur market survey (1997) Species Rangpur Silver Carp 8,350 Tangra 797 Mola 80 Shorputi 3,930 Rui 5,830 Source NFEP-2 market data 4.6.1 Markets The results of the market questionnaire demonstrated that silver carp is the most frequent species occurring in the retailer stalls. It is the least expensive of the cultured fish, and the most purchased. The questionnaire also supported the dichotomy in size of silver carp observed in the market. Large and small silver carp appear to be for the rich and poor consumers respectively. 100 percent of the retailers confirmed this observation by responding that small size silver carp is purchased by the labours and the marginal poor, while large size silver carp is purchased by professionals, government workers, and the wealthy. 5.1 CONCLUSION Questionnaires were used to determine the impact that silver carp has on the livelihoods to the people of northwest Bangladesh. Strain comparison trials were used to assess the current quality of silver carp being used in aquaculture. The combination of social surveys and technical farm trials is a powerful tool to determine the importance and quality of silver carp in northwest Bangladesh. The questionnaires determined several main points: Nursers and Traders assess the quality of seed before purchase to ensure a better return on final sell One of the greatest proportions of species traded and nursed is silver carp Silver carp is considered by traders and nursers to have a low survival rate compared to other species Traders perceived that the quality of silver carp has deteriorated in the last five years Silver carp and small self recruiting species are purchased most frequently by unskilled workers with silver carp being the species frequently bought by male aggregate workers Silver carp is the most available in the markets sampled and is the least expensive of the cultured species The size of silver carp in the markets is purchased by different people: landless and marginal poor purchase the small sizes, while the large sizes are bought by the wealthy Silver carp was the most sold species in the markets sampled Through the survey it can be determine that silver carp is a relative important species to the people of northwest Bangladesh especially to the rural resource poor. It is one of the least expensive of the cultured species and is readily available. It is considerably one of the highest traded and nursed species next to rui/mrigel. 5.2.1 Strain comparison trial Hatchery trial There were no significant difference in fertilization rate between old strain and new strain A significant difference were seen however, in the hatching rates with the new strain having higher values then the old There were no significant differences observed in total wet harvest weight/amount of eggs stocked between the strains but there was significant difference in larvae fork length, with new strain being greater in size than old strain and commercial hatchery strain. In summary there was no clear difference between the new and old strain during the hatchery performance comparison. The new strain was significantly larger in size then the old strain and this could be related to inbreeding depression in the old strain or maternal differences in sizes between the strains The old and new strain from the NFEP hatchery was significantly larger in size than the commercial hatchery larvae and this could be the result of a lack of knowledge in commercial hatcheries on the subject of broodstock/hatchery management practices. Such fundamental knowledge includes the need for recruitment of new breeders into the broodstock at regular intervals, proper stocking densities of broodfish/hatchings and their balanced feeding. Further principles include the selection of broodstock of a desirable size, injecting adequate hypophysation dosage, and mating unrelated female and male breeders. These can all have an effect on seed quality. Hapa trial A significant difference in survival rate was observed, with two of the commercial hatcheries and the simulated transfer of new strain having greater survival rates than the commercial hatcheries one, simulated transfer of old strain, new strain, and old strain. Differences in survival rates might be due to stocking procedure. Commercial hatchery two had the significant lowest fork length and weight out of the seven treatments, but there was no significant difference in final harvest production In summary there were no clear difference between the seven treatments, which could be due to inadvertently stocking the hapas with ‘stronger’ larvae from the commercial and simulated transfer bags (CM-1, CM-2, CM-3, STN, STO). There were no differences between the old and new strain, which suggest that differences in performance could occur later on in the rearing cycle. 6.1 BILIOGRAPHY Ahmed, M., Rab, M. A. and Bimbao, M. A. P. (1995) Aquaculture technology adoption in Kapasia Thana, Bangladesh: some preliminary results from farm record-keeping data. ICLARM Technical Report TR44, Manila. Ali, M.L. (1998). Fishery resources development and management techniques. Fish Week ’98 compendium. Department of Fisheries and Ministry of Fisheries and Livestock, Dhaka, Bangladesh. Ali, Md. M. (28 Aug, 1998) Desirable Dietary Patterns for Bangladesh People, The Bangladesh Observer Bangladesh Bureau of Statistics (BBS) (1998) Statistical Pocket Book of Bangladesh, 1998. Statistics Division, Ministry of Planning, Government of the People’s Republic of Bangladesh. Department of Fisheries, Bangladesh. Barton B.A., and Iwama G.K. (1991) Physiological changes in fish from stress in aquaculture with emphasis on the response and effects of corticosteroids. Annual Reviews of Fish Diseases 10, 3-26. Chowdhury, A.Z.M., and Mardall, N. (1998) The fry and fingerling trade at Parbatipur railway station 1989, and 1992 to 1996. Northwest Fisheries Extension Project (Second Phase) Paper Number 20, Department For International Development DFID (Department for International Development) (1998) Fisheries and Aquatic Resources Sector Strategy and Programme 1995-2000. Doyle, R.W. (1983) An approach to the quantitative analysis of domestication selection in aquaculture. Aquaculture, 33: 167-185. Eknath, A.E. and Doyle, R.W. (1985) Maximum likelihood estimation of “unobseravable” growth and development rates using scale data: application to carp aquaculture in India. Aquaculture, 49: 55-71 Eknath, A.E. and Doyle, R.W. (1990) Effective population size and rate of inbreeding in aquaculture of Indian major carps. Aquaculture, 85: 293-305. Falconer, D.S. (1981) Introduction to Quantitative Genetics. Longman, New York, NY, 340pp. Gupta, M.V. and Shah, M.S. (1992) NGO linkages in development aquaculture as a sustainable farming activity-a case study from Bangladesh. Paper presented at Asian Farming Systems Symposium, Colombo. Griffiths, D., and Das, K.T. (1999) A trial to compare the grow-out of old, hybrid and imported silver carp fingerlings. Northwest Fisheries Extension Project (Second Phase) Paper Number 25, Department For International Development Hoque, M. T. (1995) Sustainable Agriculture: A perspective on Fish Culture for the Small-Scale Resource-Poor Farmers of Bangladesh. Journal of Sustainable Agriculture, Vol. 5(3) 97-112 Hussain, M.G. and Mazid, M.A. (1997) Carp genetic resources in Bangladesh. In: M.V. Gupta, M.M. Dey, R. Dunham and G. Bimbao (eds.), Proceedings of the Collaborative Research and Training on Genetic Improvement of Carp Species in Asia, 26-29 July 1997, Central Institute of Freshwater Aquaculture, Bhubaneswar, India. ICLARM Work. Doc. 1 (Unpublished). Hussain, M.G. and Mazid, M.A. (1999) Broodstock Management Status and Some suggestions to control negative selection and inbreeding in hatchery stocks in Bangladesh. Naga, 22 4: 24-27 International Food Policy Research Institute (1999) Bangladesh Adoption of Commercial Vegetables and Polyculture Fish Production: Impacts on Income, Household Allocation and Nutrition http://www.cgiar.ofg/ifpri/THEMES/MP17/bangl.htm Kebus M.J., Collins M.T., Brownfield M.S., Amundson C.H., Kayes T.B. and Malison J.A. (1992) Effects of rearing density on stress response and growth of rainbow trout. Journal of Aquatic Animal Health 4, 1-6. Kincaid, H.L. (1983) Inbreeding in fish populations used for aquaculture. Aquaculture, 33: 215-227. Lewis, D. J., (1997), Rethinking aquaculture for resource-poor farmers: perspectives from Bangladesh. Food Policy, Vol. 22 No 6, pp. 533-546 Lewis, D. J., (1997) NGO’s Donors, and the State in Bangladesh. ANNALS, AAPSS, 554, 33-45 Lewis, D. J., Gregory, R., and Wood, G.D. (1993) Indigenising Extension: Farmers, Fish-Seed Traders and Poverty-Focused Aquaculture in Bangladesh. Development Policy Review Vol. 11, 185-194 Lewis, D. J., Wood, G.D. and Gregory, R. (1996) Trading the Silver Seed: Local Knowledge and Market Moralities in Aquacultural Development. Intermediate Technology Publications of Dhaka: University Press Limited, London Li, S., Lu, W., Peng, C., and Zhao, P. (1987) A genetic study of the growth performance of silver carp from the Changjiang and Zhujiang Rivers, Aquaculture, 65:93-104. Morris, C.P. (1997) Annual Report on the 1995/96 Season of the Extension and Training Section of the Northwest Fisheries Extension Project. ODA Nabi, R., Datta, D., Chakrabarty, S., Begum, M., and Chaudhury, N.J. (1999) Consultation with the Poor- Participatory Poverty Assessment in Bangladesh. NGO Working Group on the World Bank Bangladesh 12-14 New M. B., Tacon A. G. J. and Csavas, I. (1993 ) Farm-Made Aquafeeds Proceedings of the FAO/AADCP Regional Expert Consultation on Farm- Made Aquafeeds 14-18 December 1992 Bangkok, Thailand The Regional Office for Asia and the Pacific, Food and Agriculture Organization of the United Nations and the ASEAN – EEC Aquaculture Development and Coordination Programme (AADCP) Northwest Fisheries Extension Project (second Phase) (2000) NFEP-2 Diary 2000 Department for International Development Nutrition Surveillance Project (NSP) Annual Report (1998) National and Divisional Trends Among Children and Households in Rural Bangladesh. Helen Keller International and Institute of Public Health Nutrition (August 1999) Nuruzzaman, A.K.M. (1990) Problems and potentials for development of inland capture fisheries of Bangladesh. World Aquaculture-1990 Conference, Halifax, Canada State of the System (SOS) Report Bangladesh (2000) Department For International Development (DFID) Aquaculture Research Programme World Bank (1991) Fisheries Sector Review, report Number 8830-BD. Washington DC. 7.1.1 Revised Questionnaire for Nursers Nursery Name Owner Location Length of time in business? What types of fish do you nurse? What fish do you produce more than the others? Why? What fish do you considered the least problem to nurse? Why? What do you look for in performance? What is the size of your ponds? How long do you nurse SC before selling? How much produce a yr? What % is SC. Is SC polycultured? What is the stocking density for SC? How much of the fish do you sell compare to keeping? Who do you usually sell too? Farmers Traders Market Others How much do you sell for? Where do you buy your hatchlings? Private Government Other Source When the best time to buy SC? How do you transport the fish? Walk Rickshaw Motor Other Do you add any chemicals? How do they compare to other fish in survival and quality during transportation? What is SC survival rate? 7.1.2 Trader Questionnaire Traders Name Location How many years experience do you have? What months do you trade fingerlings from? What type of seed do you trade the most? Why? What do you think is the best fry to transfer? Why? What is the size of fingerling of silver carp you generally trade? What size of silver carp gives the best survival? What prices do you sell silver carp fingerlings? What fish are more expensive? What fish is cheaper? Where do you buy your seed? Do you usually buy from the same place? What do you look for when buying fingerling? What form of transportation do you use to transport your fish? Who do you usually sell to? How long do you usually travel? What container do you use to transfer fish? Do you use chemicals? Quality of seed? 7.1.3 Market Sampling Name of Market Union Thana District Species observed Which one has the most quantities? Amount of SC Size and price What dose he sells the most of Who buys SC?
Pages to are hidden for
"A Comparison of Silver Carp ( Hypopthalmichthys molitrix )"Please download to view full document