FACULTEIT LANDBOUWKUNDIGE WETENSCHAPPEN Academiejaar 2004- · PDF fileWETENSCHAPPEN . Academiejaar 2004-2005 . ... Prof. Koen Dewettinck and ... and the Belgian MSc student Stijn Vandendriessche

FACULTEIT LANDBOUWKUNDIGE EN TOEGEPASTE BIOLOGISCHE

WETENSCHAPPEN

Academiejaar 2004-2005

OPTIMIZATION OF MUD CRAB (SCYLLA PARAMAMOSAIN) LARVICULTURE IN VIETNAM

OPTIMALISATIE VAN DE LARVICULTUUR VAN DE TROPISCHE KRABSOORT SCYLLA PARAMAMOSAIN IN VIETNAM

by/door

Truong Trong Nghia

Thesis submitted in fulfillment of the requirements for the degree of Doctor (Ph. D.) in Applied Biological Sciences

Proefschrift voorgedragen tot het bekomen van de graad van Doctor in de Toegepaste Biologische Wetenschappen

On the authority of Op gezag van

Rector : Prof. dr. A. DE LEENHEER

Decaan : Promotor:

Prof. dr. ir. H. VAN LANGENHOVE Prof. dr. P. SORGELOOS

Auteur en promotor geven de toelating dit doctoraatswerk voor consultatie beschikbaar te srellen en delen ervan te copiëren voor persoonlijk gebruik. Elk ander gebruik valt onder de beperkingen van het auteursrecht, in het bijzonder met betrekking tot de verplichting uitdrukkelijk de bron te vermelden bij het aanhalen van de resultaten van dit werk. The author and the promoter give the authorization to consult and to copy parts of this work for personal use only. Every other use is subject to copyright laws. Permission to reproduce any material contained in this work should be obtained from the authour. Ghent, October 2004 De promotor/Promoter De auteur/Author Prof. dr. Patrick Sorgeloos Truong Trong Nghia

Acknowledgements My deepest sincere gratitude to my promoter and Professor dr. Patrick Sorgeloos for his scientific

orientation and assistantce, especially his patience in correcting the first and final thesis drafts during his already busy time.

I think highly of Dr. Patrick Lavens who came up with the initial plan for my research before he

left ARC for INVE Technologies.

I am very grateful to my “second promoter” Mathieu Wille for his proper suggestions on experiment design and scientific discussion, especially for his devoted and thoughtful revision and recommendations in the preparation and completion of all chapters of the thesis. I also appreciate

An Van Der Eecken for her contribution to my thesis correction.

I am especially thankful to the members of the reading committee, Prof. Peter Bossier (Ghent University, Belgium), Dr. Lewis Le Vay (University of Wales Bangor, the UK), Dr. Geoff Allan

(NSW Department of Primary Industries, Australia), Prof. Koen Dewettinck and Prof. Frans Ollevier (Leuven University, Belgium) for their critical reviews and extremely valuable

suggestions to improve this thesis.

I deeply thank the administrative support headed by Magda Vanhooren and her assistants Dorina Tack and Alex Pieters for arranging my accommodation and money; Marc Verstraeghen

and Tom Baelemans for purchasing computer wares and experimental facilities; Katerina Gamrotova for helping to access references and especially Els Vanden Bergh for her

kindness in guiding the paperwork for my PhD annual registration and defense.

I greatly appreciate the encouragement and support from my former superiors Prof. Nguyen Kim Quang (Ex-Vice Rector of Can Tho University), Prof. dr. Tran Phuoc Duong

(Ex-Rector of Can Tho Unversity), Prof. dr. Tran Thuong Tuan (Ex-Rector of Can Tho University) and Dr. Vu Do Quynh (Ex-Director of Shrimp Artemia R&D Institute, Can Tho University) and

the current Rectorate of Can Tho University for the fulfillment of the thesis.

I am greatly indebted to all teachers, farmers and friends who have transferred unconditionally their valuable knowledge and experience in aquaculture since I have involved in this field.

The experiments during seven years were possible thanks to the very valuable cooperation received

from the research assistants Tran Cong Binh, Nguyen Van Danh, Nguyen Minh Dat, Tran Thi Dep, Huynh Thi Ngoc Hien, Tran Tan Huy, Nguyen Hong Loc, Ngo Le Ngoc Luong, Pham Thi Tuyet Ngan, Tran Suong Ngoc, To Cong Tam, Hoang Phuoc Thanh, Ngo Thi Thu

Thao, Nguyen Thi Thanh Thao, Cao Phuoc Tho, Nguyen Thi Hong Van, Quach The Vinh and Vu Ngoc Ut; especially Geert Vandewiele for analizing the HUFA samples

and the Belgian MSc student Stijn Vandendriessche.

I would like to express my warmest feelings to all my friends and my colleagues in various institutions and universities, Can Tho University and its College of Aquaculture and Fisheries, who

always were concerned about my PhD completion.

My PhD program was financed through scholarship and research grants from the International Foundation for Science (IFS), Binh Chau Company (established by Mr. Trinh Vinh Binh), the Flemish Inter-University Council (Vl.I.R.-IUC) and the European Commission (INCO-DC).

My family would be very happy with my success and this is a gift for them.

I hope I did not forget anyone, but just in case… thank you.

List of abbreviations Σ Total ANOVA Analysis of variance ARA Arachidonic acid BIARC Bribie Island Aquaculture Research Centre in Australia C1 First crab stage C2 Second crab stage DAH Day(s) after hatch DHA Docosahexaenoic acid DIS Dry Immune Selco® (INVE Aquaculture, Belgium) EC5 Effective concentration at 5 % endpoint EFA Essential fatty acid EPA Eicosapentaenoic acid FAME Fatty acid methyl ester FAO Food and agriculture organization GRIM Gondol Research Institute for Mariculture in Indonesia HSD Honest significant difference HUFA Highly unsaturated fatty acid ICES International council for the exploration of the sea IFS International Foundation for Science INCO - DC International cooperation for developing countries LC50 Lethal concentration at 50 % endpoint LSI Larval stage index M Megalopa(e) MDS Moult death syndrome MR1 First metamorphosis rate MR2 Second metamorphosis rate PNR Point of no return PNR50 Point of no return at 50 % endpoint PUFA Poly unsaturated fatty acids PVC Polyvinyl chloride RIA3 Research Institute for Aquaculture No III in Vietnam TAN Total ammonia nitrogen VASEP Vietnam Association of Seafood Exporters and Producers Vl.I.R - IUC Vlaamse Interuniversitaire Raad (Flemish Interuniversity Council) - Institutional University Co-operation WSSV White spot syndrome virus Z1 - Z5 Zoeal stages 1 - 5

Table of contents CHAPTER 1............................................................................................................................1 Introduction ...........................................................................................................................1 1. World production and demand of high-value crustaceans ..................................................1 2. Production of high-value crustaceans in Vietnam...............................................................2 3. Mud crab culture..................................................................................................................3 4. Reclassification of mud crab species...................................................................................4 5. Importance of mud crab larviculture in Vietnam ................................................................6 6. Aims and outline of the thesis .............................................................................................7 CHAPTER 2..........................................................................................................................11 Current status of mud crab (Scylla spp.) hatchery technology .......................................11 Abstract..................................................................................................................................11 1. Introduction .......................................................................................................................12 2. Broodstock.........................................................................................................................13

Sourcing and maturation ...................................................................................................13 Spawning (egg extrusion) ..................................................................................................16 Incubation and hatching ....................................................................................................17

3. Larval rearing ....................................................................................................................20 Selection and stocking .......................................................................................................20 Water quality and parameters ...........................................................................................20 Culture systems..................................................................................................................22 Feeding and nutrition ........................................................................................................24

4. Nursery ..............................................................................................................................27 5. Bottlenecks to commercial production ..............................................................................29 6. Discussion..........................................................................................................................31 CHAPTER 3..........................................................................................................................33 Reproductive performance of captive mud crab (Scylla paramamosain) broodstock in

Vietnam.............................................................................................................................33Abstract..................................................................................................................................33 1. Introduction .......................................................................................................................34 2. Materials and methods.......................................................................................................35

2.1. Broodstock ..................................................................................................................35 Broodstock source .........................................................................................................35 Rearing systems and culture conditions ........................................................................35 Broodstock management................................................................................................36

2.2. Egg incubation............................................................................................................37 2.3. Reproductive performance .........................................................................................37 2.4. Statistical analysis ......................................................................................................39

3. Results ...............................................................................................................................40 3.1. Effect of selected management and environmental parameters on reproductive performance.......................................................................................................................40

Eyestalk ablation ...........................................................................................................40 Rearing system...............................................................................................................40 Broodstock source .........................................................................................................41 Month and seasonal cycle..............................................................................................41 Female weight................................................................................................................43

Contents

ii

Time to spawn ............................................................................................................... 44 3.2. Artificial incubation of eggs and egg diameter during incubation ............................ 44

4. Discussion ......................................................................................................................... 44 4.1. Effect of selected management and environmental parameters on reproductive performance ...................................................................................................................... 44

Eyestalk ablation ........................................................................................................... 44 Types of rearing system................................................................................................. 45 Broodstock source ......................................................................................................... 47 Month and seasonal cycle ............................................................................................. 47 Female weight ............................................................................................................... 50 Time to spawn ............................................................................................................... 51

4.2. Artificial incubation of eggs and egg diameter in function of incubation time ......... 52 5. Conclusions and suggestions............................................................................................. 52 CHAPTER 4 ......................................................................................................................... 61 Optimal feeding schedule for mud crab (Scylla paramamosain) larvae......................... 61Abstract ................................................................................................................................. 61 1. Introduction ....................................................................................................................... 62 2. Materials and methods ...................................................................................................... 63

2.1. Broodstock rearing..................................................................................................... 63 2.2. Live feed culture and enrichment ............................................................................... 64 2.3. Larval rearing: objectives, experimental design and techniques .............................. 65

Experiment 1 ................................................................................................................. 65 Experiment 2 ................................................................................................................. 66 Experiment 3 ................................................................................................................. 67 Experiment 4 ................................................................................................................. 67 Experiment 5 ................................................................................................................. 68 Experiment 6 ................................................................................................................. 68 Experiment 7 ................................................................................................................. 69 Experiment 8 ................................................................................................................. 69

2.4. Evaluation criteria ..................................................................................................... 69 2.5. Statistical analysis...................................................................................................... 70

3. Results ............................................................................................................................... 70 4. Discussion ......................................................................................................................... 73

4.1. Ability of S. paramamosain zoeae to catch instar-1 Artemia..................................... 74 4.2. Suitable first feed........................................................................................................ 75 4.3. Alternative Artemia forms as first feed ...................................................................... 78 4.4. Feeding schedule........................................................................................................ 80

5. Conclusions and suggestions............................................................................................. 81 CHAPTER 5 ......................................................................................................................... 89 Influence of the content of highly unsaturated fatty acids in the live feed on

larviculture success of mud crab (Scylla paramamosain) ............................................ 89Abstract ................................................................................................................................. 89 1. Introduction ....................................................................................................................... 90 2. Materials and methods ...................................................................................................... 91

2.1. Source of larvae ......................................................................................................... 91 2.2. Larval rearing ............................................................................................................ 92

Larval rearing systems and procedures ........................................................................ 92 Live feed culture and enrichment .................................................................................. 93

Contents iii

Feeding ..........................................................................................................................94 2.3. Experimental design ...................................................................................................94 2.4. Evaluation criteria......................................................................................................95

Fatty acid composition ..................................................................................................95 Larval performance .......................................................................................................96

2.5. Statistical analysis ......................................................................................................97 3. Results ...............................................................................................................................98

3.1. Fatty acid composition of live feed and crab larvae (experiment 3)..........................98 3.2. Zoeal survival .............................................................................................................99 3.3. Larval development rate during the zoeal stages .......................................................99

Experiment 1..................................................................................................................99 Experiment 2................................................................................................................100 Experiment 3................................................................................................................100 Experiment 4................................................................................................................100

3.4. Metamorphosis .........................................................................................................101 Experiment 3................................................................................................................101 Experiment 4................................................................................................................102

3.5. Correlation between the fatty acid composition of the live feed and the crab larvae, and larval development rate and metamorphosis success...............................................103

LSI in relation to the fatty acid composition of the live feed .......................................103 LSI in relation to the fatty acid composition of the crab larvae..................................103 Metamorphosis success in relation to the fatty acid composition of the live feed and the crab larvae...................................................................................................................103

4. Discussion........................................................................................................................104 4.1. Fatty acid composition of live feed and crab larvae ................................................104 4.2. Survival in the zoeal stages ......................................................................................104 4.3. Larval development rate during the zoeal stages .....................................................106 4.4. Metamorphosis .........................................................................................................109

Metamorphosis rate .....................................................................................................109 Timing and duration of metamorphosis.......................................................................110

4.5. Survival of Z1 to megalopa and the first crab (M/Z1 and C1/Z1 survival rates).....111 5. Conclusions and suggestions ...........................................................................................113 CHAPTER 6........................................................................................................................125 Improved larval rearing techniques for mud crab (Scylla paramamosain) .................125Abstract................................................................................................................................125 1. Introduction .....................................................................................................................126 2. Materials and methods.....................................................................................................127

2.1. Source of larvae........................................................................................................127 2.2. Food and feeding ......................................................................................................128

Micro-algae culture .....................................................................................................128 Rotifer culture and enrichment....................................................................................128 Artemia culture and enrichment..................................................................................128 Feeding ........................................................................................................................129

2.3. Larval rearing experiments: objectives and experimental design............................129 Experiment 1................................................................................................................130 Experiment 2................................................................................................................130 Experiment 3................................................................................................................131 Experiments 4, 5 and 6 ................................................................................................131 Experiments 7 and 8 ....................................................................................................131

Contents

iv

2.4. Evaluation criteria ................................................................................................... 132 2.5. Statistical analysis.................................................................................................... 132

3. Results ............................................................................................................................. 133 Experiment 1 ................................................................................................................... 133 Experiment 2 ................................................................................................................... 133 Experiment 3 ................................................................................................................... 133 Experiment 4 ................................................................................................................... 134 Experiment 5 ................................................................................................................... 134 Experiment 6 ................................................................................................................... 134 Experiment 7 ................................................................................................................... 135 Experiment 8 ................................................................................................................... 135

4. Discussion ....................................................................................................................... 135 4.1. Rearing system ......................................................................................................... 135

Recirculation ............................................................................................................... 135 Role of supplemented micro-algae.............................................................................. 136 Choice of system.......................................................................................................... 137

4.2. Other zootechnics..................................................................................................... 138 Z1 stocking density ...................................................................................................... 138 Rotifer density for feeding early larval stages (Z1-Z2 stages).................................... 138 Artemia for feeding later larval stages (from Z3 onwards) ........................................ 139 Prophylactic chemicals ............................................................................................... 141 Cannnibalism .............................................................................................................. 142

5. Conclusions and suggestions........................................................................................... 143 CHAPTER 7 ....................................................................................................................... 149 General discussion............................................................................................................. 149 1. Challenges and perspectives for mud crab larviculture .................................................. 149

Feasibility of mud crab larviculture compared to other crustaceans............................. 149 Challenges for mud crab larviculture ............................................................................. 150 Perspectives for larviculture of mud crabs ..................................................................... 151

2. Broodstock management (Chapter 3).............................................................................. 152 3. Optimal feeding for the larvae (Chapter 4) ..................................................................... 154

Rotifers and Artemia as suitable live feed ...................................................................... 154 Early zoeae of S. paramamosain might be more rotifer-dependent compared to other Scylla species .................................................................................................................. 154 Alternatives for rotifers ................................................................................................... 155

4. Live feed quality (Chapter 5) .......................................................................................... 156 Role of essential fatty acids in crustaceans..................................................................... 156 Effects of DHA, EPA and ARA on mud crab larvae ....................................................... 157

5. Rearing systems and other zootechnics (Chapter 6) ....................................................... 158 Selecting an efficient rearing system............................................................................... 158 Other zootechnics............................................................................................................ 161

6. Overall conclusions and suggestions for further research............................................... 162 References .......................................................................................................................... 165 Summary ............................................................................................................................ 179 Curriculum vitae ............................................................................................................... 189

CHAPTER 1

Introduction

1. World production and demand of high-value crustaceans

According to FAO (2002), reported global capture fisheries in 2000 returned to the

level of the early 1990s, reaching about 94.8 million tonnes. Unlike capture fisheries,

aquaculture production has continued to increase markedly (35.6 million tonnes) with an

annual growth rate ranging from 5.3 - 7.1 % over the two last decades. It is believed that

aquaculture potential still exists in many areas and for many species. More than half of

global aquaculture production originated from marine or brackish coastal waters. High-

value crustaceans and finfish predominate in brackish water, and molluscs and aquatic

plants in marine waters. Although brackish water production represented only 4.6 % of total

global aquaculture production by weight in 2000, it comprised 15.7 % of total production by

value.

Crustaceans are amongst the most highly valued foods, but of the 26,000 species of

crustaceans (Ruppert and Barnes, 1994) only penaeid shrimps, mitten crab (Eriocheir

sinensis), freshwater prawns (Macrobrachium spp.) and red swamp crawfish (Procambarus

clarkii) are being produced on an industrial scale (Wickins and Lee, 2002). The availability

of crustaceans per capita more than tripled from 0.4 to 1.4 kg (between 1961 and 1999),

largely because of the production of shrimps and prawns from aquaculture practices. Shrimp

is already the most traded seafood product internationally, and about 26 % of total

production now comes from aquaculture (1.1 million tonnes in 2000). Similarly, trade in

crab species has increased with growing aquaculture production (140,300 tonnes in 2000)

(FAO, 2002). Mud crabs (Scylla spp.) are considered luxury seafood items due to their large

size and delicate flavor, and therefore have high market value and are in great demand

(Dorairaj and Roy, 1996; Triño and Rodriguez, 2002). They are highly sought after in East

Asia (particularly in Hong Kong, Japan, Taiwan and Singapore), where live crabs

(especially gravid females) command premium prices (Agbayani, 2001; Keenan, 1999a).

There is also a growing market in the USA for frozen, soft-shelled mud crab and the

demand for mud crab meat for value added products is expanding internationally (Cholik,

1999; Keenan, 1999a; Tan, 1999; Wickins and Lee, 2002).

CHAPTER 1 – Introduction

2

Statistics on mud crab production are much more fragmentary and variable than for

shrimp because the species are often fished and cultured artisanally and therefore not

regularly reported. Depending on the source, reported mud crab production figures range

from a few hundred to several thousands tonnes, mainly from countries in the Indo-Pacific

region (Agbayani, 2001; Camacho and Aypa, 2001; Chaitanawisuti and Krittsanapuntu,

1998; Cholik, 1999; Fortes, 1999; Liong, 1992). The official global catch of mud crabs in

2001 was less than 17,000 tonnes (FAO, 2002). World capture and culture production of the

genus Scylla were 14,841 and 5,883 tonnes, respectively (FAO, 1998). Capture production

is predicted to decline because worldwide exploitation of mud crabs is aiming at all size-

classes and is increasing constantly (Le Vay, 2001; Le Vay et al., 2001). As a result,

declining crab landings and smaller maximum sizes have been reported over the last two

decades (Angell, 1992).

For the period up to 2030, gross trends project that, for developed countries,

consumption patterns will reflect demand for, and imports of, high-cost/high-value species;

and in developing countries, trade flows will reflect the exportation of high-cost/high-value

species and the importation of low-cost/low-value species (FAO, 2002).

As capture production of mud crabs has been declining and the demand for high-value

species in developed countries is predicted on the rise (FAO, 2002), culture of mud crabs is

expected to increase. Especially since (in contrast to other crustaceans and even other crab

species) aquaculture production to date only makes up a small proportion of the availability

of this high-cost/high-value species.

2. Production of high-value crustaceans in Vietnam

According to the most recent report of the Vietnamese Ministry of Fisheries (MOFI,

2003), between 1990 and 2002, capture fisheries and aquaculture doubled (709,000 -

1,434,800 tonnes) and respectively tripled (310,000 - 976,100 tonnes) their output, resulting

in a ten-fold increase in export value. Although world seafood consumption in 2002

remained high, the market is subjected to harsh competition that required Vietnamese

seafood exporters to struggle continuously to protect their tenth position on the world

market. In 2002, Vietnamese seafood export reached 459,000 tonnes, valued over 2 billion

US$. Export was was composed of four main groups, namely shrimp, fish, squid and others

CHAPTER 1 – Introduction 3

(bivalves, marine crabs, seaweed) valued proportionally 47, 22, 12 and 19 %. Proportions in

weight of these groups were 25, 16, 29 and 30 %, respectively.

During the period 1950 - 2000, marine crab production has increased sixteen-fold

(2,000 to 32,000 tonnes) in Vietnam (FAO, 2002). The highest production was recorded in

1998 (48,000 tonnes), which represents about 4 % of the highest reported global crab

production figure in 2000.

The Mekong Delta is the most important region in Vietnam for both capture fisheries

and aquaculture (including mud crabs), accounting for 43 and 67 % of the nation’s total

production, respectively and 57 % of the total export values in 2003 (Huy, 2004). Central

Vietnam is less suited for mud-crab farming because it lacks extensive pond areas, which

are plentiful in the two large delta regions of the Red River (North Vietnam) and Mekong

River (South Vietnam). Culture potential in North Vietnam however, is limited due to its

temperate climate.

Data on mud crab production in Vietnam are also variable and not trustworthily

reported for the same reasons (artisanal fishing and culture) as for the rest of the world. In

1993, reported culture production of Scylla was 3,800 tonnes (Dau, 1998). The total mud

crab production in 1995 was estimated at 4,500 to 5,500 tonnes (IFEP, 1996). A 1995

survey of the 6 eastern coastal provinces in the Mekong Delta estimated a total mud crab

production in that region of about 1,644 tonnes (Tuan et al., 1996). In 1999, the mud crab

production in Ca Mau province (the largest fisheries producer accounting for 7.8 % of the

country total) amounted to 5,000 tonnes, of which 20 % came from farming (Xuan, 2001). It

is therefore seems, mud crab production in Vietnam has followed the trend of increasing

global production and trade.

3. Mud crab culture

Mud crabs are attractive candidates for culture because in their post-larval stages they

are hardy to fluctuations in water quality and temperature, relatively resistant to disease and

grow quickly on a wide variety of diets (Williams and Primavera, 2001).

Scylla spp. feed on a variety of food items. Juveniles, which are more mobile, feed on

prawns, smaller crabs, fish and other small invertebrates (Joel and Sanjeevaraj, 1986).

Larger crabs are primarily carnivorous, eating benthic molluscs and crustaceans (Hill 1979;

Joel and Sanjeevaraj, 1986; Paterson and Whitfield, 1997), but are also opportunistic


4

omnivores (Warner, 1977) and will eat a wide variety of animal protein and even vegetable

matter such as submerged aquatic weeds (Hill, 1979), filamentous algae (Williams and

Primavera, 2001), detritus (Hill, 1979) and cooked maize (Rodríguez et al., 2003).

Cholik (1999) classified three types of mud crab culture in ponds based on the final

product, i.e. grow-out from juvenile to consumption size, fattening and gravid female

production. Triño and Rodríguez (2002) reported pen culture of mud crab in tidal flats with

mangrove trees is a newly profitable culture model. In its simplest form, mud crabs are

cultured by fishers in order to add value to poorer quality crabs. Newly-moulted mud crabs

have flaccid, watery flesh and thus a low market value and females with immature ovaries

are worth less than gravid females. Fishers stock “empty” or “thin” crabs and immature

females in bamboo pots, cages or penned enclosures where they are fed trash fish and other

“waste” material for 15 - 40 days until fattened (300 - 800 g) or ripe (in the case of females)

and thus more valuable in the marketplace (Cholik and Hanafi, 1992; Dat, 1999a; Tan

1999). This is highly profitable and has evolved into more sophisticated farming where wild

caught juveniles are bought from fishers and stocked into ponds or enclosures and cultured

until they reach market size (Chong, 1993; Tan, 1999).

In Vietnam, mud-crab farming occurs along the entire coast, especially in areas where

there is abundance of wild populations as a source of seed stock (i.e. in mangrove areas)

(Dat, 1999a). Felix et al. (1995) described the various crab culture types practiced in

Vietnam (“moulting”, “soft-shell” and “mature” female crab culture). Keenan (1999a)

indicated that in the Mekong Delta, one of the culture models which produces very high

numbers of crab is the extensive culture in mixed mangrove aquasilviculture systems. With

this model, no supplemental feed is added and the crabs forage across the forest floor for

natural food. Tung (1995) reported that mud crab culture can deliver high profits within a

couple of months, with limited risk of disease and low inputs compared to shrimp culture.

4. Reclassification of mud crab species

In the original descriptions, mud crabs were identified as belonging to the genus

Scylla, but made up of different species (Keenan, 1999b). By observing color and

morphological features (color of the carapace, polygonal pigmented patterns, the

anterolateral teeth of the carapace, the “H” mark on the carapace, the length of chelipeds,

and size attained), the mud crabs in the Philippines were classified into three species and


one variety, i.e. S. olivacea, S. tranquebarica, S. serrata and S. serrata var. paramamosain

(Estampador, 1949; cited by Keenan, 1999b). This classification was supported by Serene

(1952; cited by Keenan, 1999b) based on a study which examined spination and color of

Scylla populations in Nha Trang, Central Vietnam.

Recently, new samples (from near Hong Kong, the Mekong Delta, Vietnam and near

Semarang, Central Java, Indonesia) were found to be closely interrelated, but distinctly

different based on morphology, DNA sequencing and allozyme electrophoresis data from

the other three species, indicating they all belonged to a fourth species of Scylla, S.

paramamosain. Furthermore, the absence of heterozygotes (i.e. hybrids) of the different

species provides strong evidence that there is no genetic exchange between them (Keenan,

1999b).

In Vietnam two types of mud crabs have been distinguished: the large-sized or

greenish crab (S. paramamosain) and the small-sized or reddish crab (S. olivacea). The

former has the widest distribution and is preferred for culture and consumption. Logically,

the large-sized species (S. paramamosain) has been selected as a priority for research.

The reclassification into four species (Keenan et al., 1998; Keenan, 1999b) also

imposed the need to review all previous publication carefully and created the need to

differentiate research for all four species within the genus as there could be significant

species-dependant differences in requirements (e.g. environmental and nutritional

requirements).

Little research has been conducted on S. paramamosain (Dat, 1999b and Nghia et al.,

2001b in Vietnam; Djunaidah et al., 2001a in Indonesia; Zeng and Li, 1999 in China)

compared to S. serrata most probably because the latter species has a much wider

distribution (Davis, 2003; Keenan et al., 1998; Le Vay, 2001). Most research centers dealing

with mud crab are also located in the regions where S. serrata is the dominating species

(Baylon and Failaman, 1999 and Quinitio et al., 1999 in the Philippines; Brick, 1974 in

Hawaii, USA; Davis et al., 2001 and Davis, 2003 in South Africa; Mann et al., 1999b and

Williams et al., 1998 in Australia). In addition, the larger size and the higher market value

of S. serrata (Carpenter and Niem, 1998) compared to other Scylla species, have fascinated

the culturists and therefore was the preferred species (even in areas where several species

are co-existing like in the Philippines.


6

5. Importance of mud crab larviculture in Vietnam

During the past two decades, there has been a rapid expansion of traditional shrimp

culture in the Mekong Delta, Vietnam. Unfortunately, this expansion has been at the

expense of mangrove clearance at a rate of approximately 5,000 ha year-1 to less than half

their original 280,000 ha in 1990 (Hong and San, 1993).

Disease outbreaks in 1993 - 1994 led to a dramatic decline in shrimp yields, with farm

incomes falling to 10 % of the previous year (Johnston and Keenan, 1999). This was the

first time that white spot syndrome virus (WSSV) appeared in Vietnam. There was a need

for diversification of cultured species besides shrimp. Mud crab has been the first priority

because it is the only other species that has been cultured traditionally in the coastal area

besides shrimp. Expansion of mud crab culture requires a reliable supply of quality seed

stock. The availability of wild seed has however considerably declined due to the over-

fishing of natural resources at all size classes (Le Vay, 2001; Le Vay et al., 2001) and the

destruction of mangrove forests being the natural nursery for many marine species including

mud crabs. When the research described in this thesis started, artificial seed production in

hatcheries had not commenced.

In conclusion, the need for development of mud crab larviculture in Vietnam is

justified for the following reasons: (i) mud crab is a high-value species with an increasing

demand on global markets, (ii) mud crab farming is a readily applicable alternative for

shrimp culture which is suffering serious disease problems, (iii) hatchery technology is

necessary to overcome the shortage of wild seed due to overfishing and habitat destruction,

(iv) mudcrab is a hardy species that can be cultured using simple and traditional practices

requiring low initial investment but generating considerable profit and (v) Scylla

paramamosain can be considered a “new species”, especially after the reclassification of the

genus Scylla, for which only very limited information exists.

The first research on mud crab larviculture in Vietnam dates back to 1993 - 1995, but

it was only several years later that the results of this study was published (Dat, 1999b).

However, back then, Vietnamese researchers were too much isolated from their

international colleagues working on the same species, and could not benefit from the

knowledge exchanged at the first specialized workshop on larval rearing of Scylla in Broom,

Australia in 1995.


The first experiments for this study started in 1996 thanks to the financial support

from the International Foundation for Science (IFS). After some preliminary success under

the IFS grant, we had the opportunity to join in an EU sponsored INCO research project

with partners from Belgium, Indonesia and the UK aiming to optimise and standardise

larviculture of S. paramamosain for stock enhancement in the Mekong Delta. With the

support of the Laboratory of Aquaculture & Artemia Reference Center (Ghent University,

Belgium), through this project, our knowledge of crustacean larviculture has improved and

research efforts strengthened. Research findings of this programme were presented at the

international workshop on “Mud crab rearing, ecology and fisheries” held in Can Tho at the

beginning of 2001 (see www.dec.ctu.edu.vn/sardi/AacrabCWare/index.htm). This event

strengthened the network of mud crab researchers and even extended it with new partners

from Australia, China, Thailand and the Philippines.

6. Aims and outline of the thesis

The aim of this thesis is to develop technologies that could contribute to the

standardization and optimization of the larviculture of Scylla paramamosain in Vietnam.

Nutrition, zootechnics and disease control are the three main areas of research, which

have led to commercial hatchery production of marine fish and crustacean larvae (Sorgeloos

and Léger, 1992). The three aspects are interconnected to some extent and developing

hatchery technology for new species is not possible unless all three are addressed.

There has been a great deal of progress in marine larval rearing technology ever since

the pioneering years of the 1960’s (Howell et al., 1998; Shelbourne, 1964). Many of the

more technical aspects developed in the past can be directly applied to new species. Only

minor modifications are usually required to achieve acceptable survival rates, but if mass

mortality persists, more specific research becomes necessary.

Crab larvae need a proper environment to feed and develop. In culture conditions, that

environment is the rearing tank, filled with high quality water and conditioned by a number

of controlling factors (e.g. aeration, filtration, prophylactic chemicals …). All these together

make up the culture zootechnics. When the medium deteriorates by some causes (e.g.

insufficient water exchange, unsuitable feed, overfeeding, too high larval densities …) then

the risk of disease appears. Therefore, nutrition and zootechnics need to be considered at the

same time and are the research fields that have to be addressed first. Next but not less


8

important are disease studies, especially on the interference of bacteria on larval

performance. However this study requires more specialised equipment and staff and was

beyond the scope of this thesis. For these reasons, the content of this Ph.D. thesis is limited

to the main aspects of nutrition and zootechnics. The thesis outline can be summarized as

follows:

- CHAPTER 1 (Introduction, this chapter) outlines the need for and importance of mud crab

larviculture in Vietnam based on figures of global and local production of Scylla crabs in

relation to the total production of high-value seafood (shrimp and other crab groups). The

specific circumstances in Vietnam, particularly in the Mekong Delta are also considered to

verify the need for artificial reproduction of Scylla paramamosain.

- CHAPTER 2 (Current status of mud crab Scylla spp. hatchery technology) reviews the

status of mud crab larviculture. A review of the literature is normally presented as an

introduction to the subject of a thesis. But because mud crab aquaculture is still an emerging

industry, particularly the hatchery phase of production, very little peer reviewed literature

exists. Research projects have been conducted in Australia, China, India, Indonesia, Japan,

Malaysia, the Philippines and Vietnam over the past decade, but most of this information

has remained in-house, was not published in English or has appeared as papers or abstracts

in conference proceedings. Because a more empirical approach is commonly taken while

establishing new hatchery technology, the most pertinent information is only available as

personal observations and unpublished results collected by the various scientists active in

the field. This was also the case with our project and the complementary Vl.I.R. Own

Initiative project between Rhodes University (South Africa) and the Laboratory of

Aquaculture. At the initiative of Prof. Patrick Sorgeloos and after consultation of scientists

from Australia, Belgium, the Philippines and the UK, it was decided to prepare a

collaborative paper which would describe the “state of the art” of mud crab larviculture

technology.

- CHAPTER 3 (Reproductive performance of captive mud crab Scylla paramamosain

broodstock in Vietnam) covers the effect of various culture parameters on some

reproductive characteristics. Broodstock availability and management are the first concerns

for those who wish to develop a new species for aquaculture. Environmental parameters,


broodstock management conditions and reproductive characteristics were recorded as a

basis for improving broodstock culture with the ultimate purpose to fully domesticate and

control seed production for this species.

- CHAPTER 4 (Optimal feeding schedule for mud crab Scylla paramamosain larvae)

identifies the most suitable prey type and size (rotifers or Artemia) for the early stages of

mud crabs because initial feed for larvae is a critical research aspect that has to be solved

first. The earliest appropriate time to shift from rotifers to Artemia was proposed since

rotifer culture is very laborious and not well mastered by most hatchery managers in

Vietnam. The possibilities are reviewed to replace rotifers in early stages by various forms

of processed Artemia awaiting the availability of suitable micro-bound diets for mud crab.

- CHAPTER 5 (Influence of the content of highly unsaturated fatty acids in the live feed on

larviculture success of mud crab Scylla paramamosain) describes the effects of different

standard HUFA (highly unsaturated fatty acid) live feed enrichment emulsions on survival

and growth of crab larvae. In captivity, rotifers and Artemia nauplii support growth and

survival of mud crab larvae, but this simplified diet is not ideal. As for other species, the

phenomenon of moult death syndrome (MDS), i.e. high mortality during or after the moult

from Z5 to megalopa, has been observed. Poor nutrition, even if confined to the early larval

stages, has been suggested as a cause for MDS. Inferior nutrition may also be a factor

contributing to the highly variable survival and the high susceptibility to disease often

recorded in mud crab larviculture.

- CHAPTER 6 (Improved larval rearing techniques for mud crab Scylla paramamosain)

compares six different rearing systems varying in the level of micro-algae supplementation

and the form of water exchange. In areas where clean and full strength seawater is limited,

as in the Mekong Delta, the combination of “green-water” and recirculation is

recommended. Other important zootechnics (zoea 1 stocking density, live feed density,

prophylactic chemicals and cannibalism) were also discussed.

- CHAPTER 7 (General discussion) reviews the results of the experiments and observations

presented in the technical chapters (Chapters 3 - 6) in order to value achievements and

pinpoint unsolved issues of this study. Based on the general discussion, some suggestions

for further research are proposed.

CHAPTER 2 Current status of mud crab (Scylla spp.) hatchery

technology

Davis, J.A.*1,2, Nghia, T.T.2,3, Wille, M.2, Mann, D.4, Quinitio, E.T.5, Williams, G.6, Fushimi, H.7, Hecht, T.1, Shelley, C.6, Churchill, G.J.1 and Sorgeloos, P.2

1 Department of Ichthyology and Fisheries Science, Rhodes University, South Africa. 2 Laboratory of Aquaculture & Artemia Reference Center, Ghent University, Ghent, Belgium. 3 College of Aquaculture and Fisheries, Can Tho University, Can Tho City, Vietnam. 4 Bribie Island Aquaculture Research Center, Bribie Island, Queensland, Australia. 5 Aquaculture Department Southeast Asian Fisheries Development Center, Tigbauan, Iloilo, the Philippines. 6 Darwin Aquaculture Center, Department of Primary Industry and Fisheries, Channel Island, Darwin, Australia. 7 Laboratory of Aquaculture and Stock Enhancement, Department of Marine Biotechnology, Fukuyama University, Ohama, Innoshima, Hiroshima, Japan. Abstract

A major bottleneck to the expansion of mud crab (Scylla spp.) aquaculture is a lack of

hatchery produced seed. Although research on egg production and larval rearing techniques has been undertaken for the past 30 years (and intensively for the last decade), little information is available in the primary literature. Most of the technical information has been published in the “grey” literature and in research reports and is largely unavailable to the wider scientific community. This paper attempts to collate and summarize the published and unpublished data together with observations made, and techniques used by scientists currently involved in mud crab hatchery research. The paper describes techniques for broodstock sourcing, maturation, spawning, egg incubation and hatching. Larval rearing techniques are outlined with regard to stocking, water quality requirements, culture systems, feeding techniques and nutrition. Rearing of megalopae to the juvenile stage is also described. Bottlenecks to commercial production are discussed and possible solutions proposed.

Manuscript submitted to Aquaculture

CHAPTER 2 – Current status

12

1. Introduction

The four species of Scylla; S. serrata, S. tranquebarica, S. paramamosain and S.

olivacea are large, primarily carnivorous portunid crabs that are found in coastal and

estuarine waters throughout the tropical and subtropical Indo-Pacific region (Keenan et al.,

1998). Growth is rapid and the life cycle can be closed within a year. The post-larval stages

are highly resistant to disease and can tolerate a wide range of environmental conditions.

Mud crabs in all their marketed forms have a high value (Agbayani, 2001; Wickins and Lee,

2002) and are suitable for aquaculture (Williams and Primavera, 2001). Currently mud crab

aquaculture is reliant on juveniles or adults caught from the wild (Le Vay, 2001). Despite a

long history of culture, records of which date back to 1890 (Shen and Lai, 1994), world

production is low. In 2001 slightly more than 10,000 tonnes were produced (FAO, 2002).

The expansion of the industry is currently limited by the lack of hatchery-produced seed

(Williams and Primavera, 2001).

Research on the breeding of mud crabs started approximately 40 years ago and several

early publications describe captive spawning and larval rearing (Brick, 1974; Du Plessis,

1971; Hill, 1974; Ong, 1964; Ong, 1966). In Japan, research on hatchery production of seed

for restocking purposes has been ongoing since 1979 (Fukunaga and Uzumaki, 1982;

Hamasaki, 2002; Hasegawa, 1989; Horiuti and Yamamoto, 1987). In the last decade there

has been a substantial increase in research, fueled in part by the disease problems in the

shrimp industry and partly by declining mud crab fisheries in some parts of South East Asia

(Le Vay, 2001). The results of this research have been presented in workshops and forums

in Western Australia (1995), Northern Territories, Australia (1997), the Philippines (1998)

and in Vietnam (2001), amongst others. Although some of the results have been published

in the primary literature the majority of information on mud crab hatchery technology and

aquaculture appears in the “grey” literature and in unpublished research reports. This paper

reviews the available literature from all sources, summarizes the current hatchery and

rearing techniques, identifies research problems and provides comment on current research

trends. This will inform those currently involved, or those wishing to become involved in

mud crab hatchery production.


13

2. Broodstock

Sourcing and maturation

Broodstock are most commonly sourced from the wild. They are either caught

specifically for this purpose (Mann et al., 1999a), are bought from markets (Dat, 1999a;

Quinitio et al., 2001) or directly from fishers (Dat, 1999a; Marichamy and Rajapackiam,

2001). In areas where commercial culture is practiced, mature breeders are sourced from

production ponds (Millamena and Quinitio, 2000). The life cycle can be closed within 9

months (Quinitio et al., 2001).

Mud crabs readily mate in captivity. This occurs while the female is in a soft-shell

condition following molting. However, adult females caught from the wild have usually

mated (Robertson and Kruger, 1994) and carry spermatophores for extended periods,

producing viable eggs up to 6 months after capture without the need for mating (Nghia et

al., 2001a).

Before introducing wild broodstock into the hatchery, the crabs are commonly

scrubbed to remove mud, encrusting algae, infestations and detritus. Individual crabs are

usually marked for identification by engraving (Churchill, 2003; Djunaidah et al., 2003;

Millamena and Quinitio, 2000) or by gluing an identification tag to the shell. They are then

disinfected in a strong formalin bath (50 - 100 µl l-1) for periods ranging from 1 hour

(Millamena and Bangcaya, 2001) to overnight (Mann et al., 1999b). Shell disease is

commonly observed, especially amongst older crabs that have long intermoult periods

(Robertson, 1987). The disease seldom causes mortality and regular cleaning of the shell

can minimize its effects (Lavilla-Pitogo et al., 2001). Various parasites have also been

reported, none of which present serious threats to broodstock or egg production (Lavilla-

Pitogo et al., 2001).

Neither the size nor type of vessel seems to affect maturation or spawning (egg

extrusion). Brood crabs have been successfully maintained individually in small bins (60 -

300 l) or communally in large tanks ranging from 1 - 12 m3 (Baylon et al., 2001a; Dat,

1999b; Hamasaki, 2002; Mann et al., 1999a; Millamena and Quinitio, 2000; Williams et al.,

1998). Successful spawning has been achieved at water depths of 25 - 150 cm (Dat, 1999b;

Djunaidah et al., 2001b; Millamena and Bangcaya, 2001). The average stocking densities in

maturation systems is approximately 1.5 crabs m-2 (Mann et al., 1999a, 1999b) but


14

broodstock can be stocked up to 5 m-2 (Djunaidah et al., 2001b). Broodstock are often kept

in separate tanks or cages which eliminate the risk of cannibalism and allows for controlled

feeding.

Shelters are usually provided to reduce stress and prevent cannibalism (Djunaidah et

al., 2003; Hamasaki, 2002; Millamena and Quinitio, 2000; Millamena and Bangcaya, 2001).

Shelters should be easy to remove so that brood crabs can be observed and the tank cleaned.

Mud crabs require a substratum for spawning. The crab digs a depression in the

substratum over which the abdominal flap is extended and into which the eggs are extruded

(Rusdi et al., 1994). The pleopods then create a current that assists the attachment of eggs to

the tertiary setae. Absence of a substratum results in poor egg attachment and potential loss

of the batch. Spawning of S. paramamosain is improved by providing a mud substratum

rather than sand (Djunaidah et al., 2001b). It has been observed that the Zoothamnium spp.

load on the eggs is reduced when the crabs spawn on a mud substrate, which may be a

consequence of turbidity (Nguyen Co Thach, Research Institute of Aquaculture III, Nha

Trang, Vietnam, pers. comm.). However, sand substratum is more commonly used since it

does not affect water quality and allows for the use of under-gravel filters (Churchill, 2003).

The entire bottom of the tank can be covered with spawning substratum (Williams et al.,

1998; Millamena and Quinitio, 2000). However, it is more common to contain the

substratum in a tray (Mann et al., 1999a) allowing the rest of the tank to be cleaned more

easily. Substratum depths of 50, 80 and 200 mm have been used successfully (Baylon et al.,

2001a; Djunaidah et al., 2001b; Millamena and Bangcaya, 2001).

Females also mature their ovaries and extrude eggs in earthen ponds or netted pens.

There is some indication in Vietnam that a higher percentage of crabs spawn in ponds,

possibly due to reduced stress. The diet available to broodstock in ponds is more varied and

probably closer to the one they would encounter in the wild. However, managing

broodstock in ponds is more difficult than in tanks and the eggs from broodstock spawned in

ponds are occasionally heavily infested with parasites (Quinitio and Parado-Estepa, 2003).

In smaller broodstock maturation systems the water is recirculated, while some larger

systems operate on a flow-through basis (Hamasaki, 2002; Marichamy and Rajapackiam,

2001; Millamena and Bangcaya, 2001; Quinitio et al., 2001) or on partial re-circulation

(Baylon et al., 2001a; Mann et al., 1999a; 1999b; Millamena and Quinitio, 2000; Millamena

and Bangcaya, 2001, Williams et al., 1998).


15

Full strength seawater (30 - 35 g l-1) is normally used for captive maturation and

spawning of S. serrata in the hatchery (Mann et al., 1999a). Lower salinities (20 - 25 g l-1)

can be used for S. tranquebarica, S. olivacea and S. paramamosain (Ng, 1998). However,

full strength seawater is still commonly used for maturing and spawning all four species.

Broodstock are generally kept between 25 °C and 35 °C. A lower critical temperature

for egg development of 18 °C has been identified (Hamasaki, 2002; Heasman and Fielder,

1983) but successful ovulation and spawning have been observed at water temperatures

ranging from 20 - 30 °C (Davis et al., 2003; Mann et al., 1999a; Millamena and Bangcaya,

2001). S. serrata broodstock extrude significantly larger eggs with higher hatch rates during

cooler months (18 - 22 °C) Mann et al. (1999a) suggesting that lower temperatures than

normal should be applied.

Mud crabs are nocturnal and normally inhabit turbid estuaries (Dat, 1999a; Barnes et

al., 2002; Hill, 1978). For this reason broodstock crabs are maintained under low light

conditions (Mann et al., 1999a; Quinitio et al., 2001; Williams et al., 1998), although

darkness is not required for spawning (Nghia et al., 2001a). Spawning tanks are normally

securely covered in order to prevent the crabs from escaping.

Broodstock are normally fed fresh or frozen foods including penaeid shrimp, squid,

fish, bivalves such as clams and mussel, gastropod snails and annelid worms (Dat, 1999a;

Hamasaki, 2002; Mann et al., 1999a; Millamena and Bangcaya, 2001; Quinitio et al., 2001).

Enriched Artemia biomass has also been used as food, presented as agar bound pellets

(Djunaidah et al., 2003). The natural diet of mud crabs consists mostly of benthic molluscs

and crustaceans (Hill, 1979) and it is generally agreed that these organisms must be present

in the diet. The source and health status of feed organisms, particularly crustaceans, need to

be carefully assessed as they can act as vectors for disease as WSSV (Chen et al., 2000).

Formulated diets developed for penaeid broodstock have been fed to mud crab

broodstock to investigate the effects on fecundity and spawning. Best results have been

achieved when formulated feeds have been used as a supplement to fresh or frozen feeds

(Djunaidah et al., 2003; Millamena and Quinitio, 2000; Millamena and Bangcaya, 2001).

Formulated diets are currently not used extensively.

Broodstock is usually fed on an ad libitum basis (Mann et al., 1999a; Williams et al.,

1999b), but when rationed, crabs are fed at rates ranging from 3 - 5 % (Dat, 1999b), 6 - 10

% (Millamena and Bangcaya, 2001) to 20 % body weight day-1 (Baylon and Failaman,

2001). Owing to their nocturnal feeding habits (Barnes et al., 2002; Hill, 1978) food is


16

normally provided in the evening, but feeding twice (Dat, 1999a) or three times daily

(Millamena and Bangcaya, 2001) allows for more careful monitoring of food consumption

and better maintenance of water quality.

Li et al. (1999) highlighted the importance of n-6 and n-3 highly unsaturated fatty

acids (HUFA) in broodstock diets. Levels of HUFA and protein in newly-hatched larvae can

be manipulated by altering the levels in the diet fed to the broodstock (Djunaidah et al.,

2003). No studies have yet been undertaken to test the effect of HUFA levels and ratios in

broodstock diets on egg quality or larval performance.

Spawning (egg extrusion)

Mud crabs adapt well to artificial conditions and spawn readily in captivity without

intervention. Over 85 % of wild caught crabs brought into the laboratory or hatchery

normally spawn, mostly within 40 days of stocking (Davis et al., 2003; Mann et al., 1999a;

Williams et al., 1998). The status of the ovary can be assessed by biopsy (Mann et al.,

1999a; Millamena and Bangcaya, 2001) or by pushing down on the first abdominal segment

which exposes the ovary under the carapace (Djunaidah et al., 2003; Quinitio and Parado-

Estepa, 2003). Mature ovaries are bright yellow to deep orange in colour and fill the body

cavity (Quinitio and Parado-Estepa, 2003).

Natural spawning is the norm in most hatcheries. However, eyestalk ablation can be

used to shorten the pre-spawning period. It is only necessary to ablate one of the eyestalks

(Baylon and Failaman, 1997; Baylon and Failaman, 2001; Baylon et al., 2001a; Dat, 1999b;

Djunaidah et al., 1998; Mann et al., 1999a; Marichamy and Rajapackiam, 2001; Millamena

and Quinitio, 2000; Nghia et al., 2001a; Quinitio et al., 2001). Prior to eyestalk ablation,

brood crabs are anaesthetized in an aerated chloroform bath (1 - 3 µl l-1). Hot pincers are

used to crimp the eyestalk and cauterize the wound. Ablation does not seem to stress the

crabs as there is no reduction in survival, fecundity or fertilization rate (Millamena and

Bangcaya, 2001; Nghia et al., 2001a). However, there is little and conflicting information on

the effect of ablation on egg and larval quality (Millamena and Quinitio, 2000). Mann et al.

(1999a) reported that ablated crabs produced larger eggs with better hatch rates while

Millamena and Bangcaya (2001) recorded lower fertilization and hatch rates of eggs from

ablated females.


17

Captive mud crabs produce viable eggs year round. Some authors have reported

seasonal changes and peaks in spawning activity (Davis et al., 2003; Hamasaki 2002; Mann

et al., 1999a; Marichamy and Rajapackiam 2001; Nghia et al., 2001a) especially in areas

that have extremes in seasonal water temperatures or salinity (Li et al., 1999; Le Vay, 2001).

However, these are usually not distinct and spawning patterns seem to vary between

locations (Hai et al., 2001; Nghia et al., 2001a). Mud crabs are capable of spawning up to

three times from a single mating (Dat 1999a; Marichamy and Rajapackiam, 2001; Quinitio

et al., 2001). Rematuration of the ovary in captive crabs occurs about one month after

spawning (Hai et al., 2001; Djunaidah et al., 2003; Marichamy and Rajapackiam, 2001).

Although some authors report no reduction in fecundity or fertilization rate with repeated

spawning (Millamena and Quinitio, 2000), Dat (1999b) recorded a substantial decrease in

egg production. The effect of repeated spawning on egg and larval quality has not been

quantified. Female S. serrata are capable of more than one post pubertal moult (Robertson

and Kruger, 1994) and spermatophores are retained after molting.

Scylla are highly fecund and typically produce more than 1 million eggs batch-1. Egg

production ranges widely between species, locations and individual crabs. S. serrata seems

to be more fecund than other Scylla species (Dat, 1999a; Djunaidah et al., 2001b; Hai et al.,

2001; Jyamanna and Jinadasa, 1993; Mann et al., 1999a; Marichamy and Rajapackiam,

1992; Millamena and Bangcaya, 2001; Quinitio et al., 2001; Srinivasagam et al., 2000;

Zainodin, 1992). Batches of 50,000 are considered small and the largest batch recorded in

captivity is 10 million eggs (Davis et al., 2003). Egg number is not a limiting factor to

juvenile production. Broodstock matured in captivity are more fecund and produce eggs

with a higher rate of fertilization than wild caught crabs. This is apparently due to enhanced

environmental conditions and nutrition (Millamena and Bangcaya, 2001; Quinitio et al.,

2001). Broodstock domestication is a high priority area of research.

Incubation and hatching

Brood crabs which have spawned are conspicuous. The large, bright yellow to orange

egg mass (“sponge” or “berry”) is carried prominently under the abdominal flap. Berried

brood crabs are transferred to incubators as soon as possible after spawning. Flat bottomed

glass aquaria of 60 1, plastic containers of 100 l or larger fiberglass tanks of approximately


18

300 - 500 l (Hamasaki, 2002; Mann et al., 1999a; Millamena and Quinitio, 2000; Millamena

and Bangcaya, 2001) and 1000 l (Williams et al., 1998) have been used as incubators.

Water in the incubators is usually sterilized with UV light and exchanged either on a

flow-through basis or recirculated through biofilters (Hamasaki, 2002; Mann et al., 1999a;

Williams et al., 1998). Because of the need for a stable environment during incubation,

recirculation is preferred. To maintain water quality and to prevent contamination of the

eggs, berried crabs are not fed. Mild aeration is provided and incubators are siphoned clean

daily (Hamasaki, 2002; Mann et al., 1999b).

Poor attachment of the eggs can result in the gradual loss of a brood during

incubation. The causes of poorly attached eggs are not well understood, but may include

poor initial attachment of the eggs, poor egg quality, or parasitic and fungal infections (Hai

et al., 2001; Quinitio et al., 2001). When eggs are lost during incubation, the hatch rate of

the remaining eggs is generally low (Dat, 1999b). Where space is limiting, brood crabs with

poorly attached eggs are removed to make space for crabs carrying well attached eggs

(Churchill, 2003; Dat 1999b; Davis et al., 2003). Dropped eggs can be incubated. However

hatch rates are highly variable and fungal infections and eggs adhering to the sides of

containers has so far made this practice impractical (Hai et al., 2001; Churchill, 2003).

The eggs can be infected with a variety of parasitic worms, fungus (Haliphthoros,

Sirolpidium, Atkinsiella and Lagenidium spp.) and ciliates (Zoothamnium spp.) (Churchill,

2003; Hamasaki and Hatai, 1993a). Parasitic infections generally occur if water quality is

not adequately maintained and may result in poor embryogenesis or egg loss as a result of

brood crabs tearing at the egg mass. The vulnerability of eggs to fungal infection decreases

as their development progresses (Hamasaki and Hatai, 1993a). Berried brood crabs can be

bathed in formalin and/or malachite green as a prophylactic, or can be treated if infestations

occur. There is evidence that formalin (25 µl l-1) is toxic to eggs up to one day after

spawning, although older eggs can tolerate high doses of both formalin (50 - 150 µl l-1) and

malachite green (50 µl l-1) (Churchill, 2003; Davis et al., 2003; Hamasaki and Hatai, 1993b;

Kaji et al., 1991). Heavy aeration is applied during the bathing process. Antifungal agents

such as Treflan®1 (44 % trifuralin) (0.05 - 0.1 µl l-1) or formalin (25 µl l-1) can be added to

the incubation water resulting in a reduced fungal load of the eggs and preventing

transmission of the fungus from the egg surface to the larvae at hatch (Kaji et al., 1991;

Quinitio et al., 2001). Preventative measures, such as maintaining a hygienic incubation

1 Mention of a branded product does not mean endorsement by the authors


19

environment and good water flow in the tank are also effective and are preferable to anti-

microbial chemicals.

At 27 - 28 °C first cleavage occurs 5 - 8 hours after extrusion (Quinitio and Parado-

Estepa, 2003). The developing embryo is visible under a dissecting microscope from day 4

after spawning (Djunaidah et al., 2003). Embryonic development can occur successfully

between 20 and 30 °C but temperatures above 26 °C are preferred (Li et al., 1999; Mann et

al., 1999a; Quinitio et al., 2001). The incubation period is strongly temperature dependent.

For S. serrata, the relationship is best represented by the equation y = 6,030 - 407 x + 7.2 x2

where y = incubation period and x = temperature (Churchill, 2003) Embryonic development

is infinitely protracted below 17 °C (Heasman and Fielder, 1983). Depending on

temperature, mud crab eggs usually hatch within 9 - 12 days after extrusion (Dat, 1999b;

Baylon and Failaman, 2001; Quinitio et al., 2001). Time of hatch can be predicted by

monitoring development of the embryo. Vigorous limb movement, a beating heart and well

developed eye spots with a purplish patch are all indications of imminent hatching (Figure

1). Hatching can also be predicted by measuring egg diameter, which increases during

incubation (Mann et al., 1999b; Nghia et al., 2001a). For S. serrata the relationship is best

represented by the polynomial equation y = 299 + 2 x + x2 where y = egg diameter (µm) and

x = incubation period in days (Churchill, 2003).

Hatching usually occurs in the morning (Hai et al., 2001; Hamasaki, 2002). Brood

crabs are often transferred to a separate tank 1 - 2 days prior to hatching (Hamasaki, 2002;

Mann et al., 1999a, 1999b). Hatching tanks tend to be larger than 500 l, providing sufficient

space for the hatched larvae to disperse and to delay deterioration of water quality. The

water in the hatching tank is usually full strength seawater (30 - 35 g l-1) that has been

filtered and treated to reduce the microbial load. A small number of larvae (100s to a few

1000s) normally hatch one day before the main hatch. Hatching is normally a spontaneous

and rapid event with 90 % of the larvae hatching within 10 minutes. The larvae hatch as pre-

zoeae, a non-swimming phase that typically moults into the first zoeal stage within 10

minutes post-hatch (Figure 1). The persistence of pre-zoeae after the normal initial moult

time has elapsed is used as an indicator of poor batch quality (Dat, 1999b; Hai et al., 2001,

Mann et al., 1999a).


20

3. Larval rearing

Selection and stocking

Once the eggs have hatched, the brood crab is removed from the hatching tank. Hatch

success is generally high, ranging from 80 - 90 %. Larvae are negatively buoyant and swim

vigorously in order to maintain their position in the water column. In the Philippines, larvae

are left in the hatching tank for approximately one hour after hatching, providing sufficient

time for “poorer quality” larvae to sink to the bottom so that “better quality” larvae at the

surface can be used for rearing (Quinitio and Parado-Estepa, 2003). The value of this

practice needs to be weighed against the fact that Z1 larvae are highly susceptible to fungal

infections both from a variety of sources including the egg envelope (Hamasaki and Hatai,

1993a) and in a typical hatching tank of 1 m3, bacterial numbers rise rapidly within an hour

of hatching (Mann et al., 1999b). This suggests that larvae be removed soon after hatching

(10 - 15 minutes). Larvae are sometimes rinsed with clean seawater before transferring them

to rearing tanks to reduce the bacterial load (Mann et al., 1999b). To transfer the larvae to

rearing tanks they are scooped from the water surface (Baylon and Failaman, 1997; Baylon

et al., 2001a; Djunaidah et al., 2003, Mann et al., 1999a). Nets cannot be used because the

long spines of the larvae become entangled in the mesh and are damaged. Larvae have been

reared at densities ranging from 10 - 200 larvae l-1 (Baylon and Failaman, 1999; Dat, 1999b;

Djunaidah et al., 1998; Quinitio et al., 1999; Quinitio et al., 2001; Williams et al., 1998)

although densities of 30 - 60 l-1 are more commonly used. No correlation has been found

between stocking density and survival although in the Philippines vibriosis has been

associated with stocking densities higher than 100 larvae l-1.

Water quality and parameters

The larvae require high quality water, free of potential pathogens, predators or

parasites. Seawater is commonly filtered to 1 µm and then either chlorinated overnight and

dechlorinated with sodium thiosulphate (Mann et al., 1999b; Parado-Estepa and Quinitio,

1998; Quinitio et al., 2001; Williams et al., 1998; Williams et al., 1999b;), ozonated and

then recirculated through a biofilter, inoculated with nitrifying bacteria and settled for

several days before use (Baylon and Failaman, 1999; Williams et al., 2002) and/or re-


21

filtered through activated carbon and sterilized with UV light (Dat, 1999b). Water in the

rearing vessels is also allowed to stabilize for several days before introduction of the larvae

(Mann et al., 1999b; Parado-Estepa and Quinitio, 1998) and background algae are

sometimes added for apparent antimicrobial qualities. Water pretreatment has been found to

significantly improve survival through to megalopa (M) (Mann, 1999b; Williams et al.,

2002).

Early stage S. serrata have a lower temperature tolerance of 12 °C (Hill, 1974).

Larvae have been successfully reared at 25 - 30 °C, but temperatures in the upper range (29

- 30 °C) shorten development time (Dat, 1999b; Li et al., 1999; Mann et al., 2001; Quinitio

et al., 1999; Quinitio et al., 2001). Larvae are extremely sensitive to abrupt changes in

temperature. Temperatures are maintained within 1 °C from hatch to harvest of megalopae.

Larvae (particularly during the early stages) are vulnerable to temperature gradients

generated by immersion heaters of the type used in shrimp hatcheries (Mann et al., 1999b).

If temperatures must be increased, low capacity aquarium heaters should be used or heating

can be applied indirectly by incubating rearing vessels in heated baths or with remote

heaters in the sump of recirculating systems. Heating is unnecessary in large-scale vessels

under tropical conditions, as long as diurnal temperature variation is not extreme.

At temperatures above 25 °C S. serrata Z1 have lower and upper 24 hour salinity

LC50s of 17.5 g l-1 (Hill, 1974) and 40 g l-1 respectively (Churchill, 2003). Scylla Z1 - Z4 are

usually reared in seawater ranging from 30 - 35 g l-1, depending on locality (Djunaidah et

al., 1998; Mann et al., 2001; Quinitio et al., 2001) although in Japan, S. tranquebarica are

reared at 25 g l-1 in order to control fungal infections.

Reducing the salinity at Z5 to 20 - 24 g l-1 for Z5 S. tranquebarica and 20 g l-1 for S.

olivacea significantly improves metamorphosis to megalopa (Baylon and Failaman, 2001;

Quinitio et al., 2001) and reducing salinity from 30 to 25 g l-1 has been found to trigger

metamorphosis of Z5 S. paramamosain (Dat, 1999b). Seawater is therefore gradually

diluted (1 - 2 g l-1 daily) at the end of the Z5 stage or at the beginning of the megalopa stage

for these three species. S. serrata does not seem to require reduced salinity at

metamorphosis and the larvae are commonly reared from hatch through to the first crab

stage (C1) in full strength seawater (Baylon et al., 2001b; Cowan, 1984; Mann et al., 2001).

Early (Z1 and Z2) larvae are strongly photopositive and light is often used to keep

early larvae close to the water surface. The larvae are visual predators, particularly in the

latter stages of development (Z3 - M), and high light intensities (1800 - 4000 lux) are


22

typically applied (Mann et al., 2001). Survival and development after the Z3 stage are

significantly compromised in the absence of light or under low (50 lux) light intensities.

Where possible, larvae are reared under natural light (Takeuchi et al., 2000; Williams et al.,

1998).

Larvae seem to require a dark phase (Djunaidah, et al., 1998), but no significant effect

on survival was recorded between 12 and 18 hour photoperiods (Nghia et al., 2001b).

Photoperiods of at least 12 hours are generally provided (Mann et al., 2001; Quinitio et al.,

2001).

S. serrata larvae can tolerate relatively high levels of nitrogenous waste. A 24 hour

LC50 of 39.7 ± 2.0 mg l-1 total ammonia nitrogen (TAN) at pH 8.2 was determined for S.

serrata Z1 in South Africa (Churchill, 2003), while in Australia a level of 62 mg l-1 was

determined (Ravi Fotedar, Aquatic Science Research Unit, Muresk Institute, Curtin

University, Western Australia, pers. comm.). The Australian study also determined a 24

hour LC50 for later instars of approximately 50 mg l-1. Larval growth rate was reduced by 5

% after 96 hours in comparison to a control (96 hour EC5) at 5 - 7 mg l-1 TAN. TAN is

generally maintained below 1 mg l-1 in hatchery runs. S. serrata Z1 have a 96 hour LC50 of

3 mg l-1 ammonia (NH3-N) (Quinitio and Parado-Estepa, 2001). All zoeal stages have a 96

hour LC50 of 80 mg l-1 nitrite (NO2-N) (Mary Lyn Seneriches-Abiera, Mindanao State

University, General Santos City, the Philippines, pers. comm.). Little research on ammonia

tolerance has been conducted for the larvae of the other species although S. paramamosain

zoeae survive concentrations of 5 mg l-1 NH3-N in recirculating systems in Vietnam.

No research has been conducted on the tolerance of Scylla larvae to extremes in pH

and oxygen concentration.

Culture systems

For highly controlled experimental conditions, the vessels used for larval rearing are

small, ranging from 100 ml to 5 l in capacity (Baylon and Failaman, 1999; Baylon et al.,

2001a; Mann et al., 2001; Quinitio et al., 1999; Williams et al., 1999a; Zeng, 1998; Zeng

and Li 1999). Highly predictable survival (60 - 90 % up to megalopa) can be achieved in

these systems, particularly in the presence of antibiotics.

Larger cylindro-conical, fiberglass tanks are used for investigating zootechnical

aspects or for nutritional studies. Under pilot and commercial scale conditions larvae are


23

reared in plastic, fiberglass or reinforced concrete tanks ranging from 1 to 200 m3 capacity

(Dat, 1999b; Fukunaga and Uzumaki, 1982; Hamasaki et al., 2002b; Millamena and

Bangcaya, 2001; Quinitio et al., 2001; Williams et al., 1999b). Tank colour does not appear

to be an important factor in mud crab larval rearing. A range of tank colours has been used

successfully.

Water is exchanged either on a constant flow-through basis, or by draining or

siphoning 50 - 85 % of the tank volume daily and replacing it with clean seawater, or by

recirculation through a biofilter (100 % every 2 - 3 hours) (Nghia et al., 2001b). Under

green-water culture conditions water is not exchanged for the first three days. Thereafter,

water exchange is slowly increased from 10 - 20 % day-1 for Z2 - Z3 to between 40 and 50

% day-1 at the end of the rearing cycle (Z4 - M) (Mann et al., 1999b; Quinitio et al., 2001).

In Japan a mesocosm system is used for culturing larvae in larger tanks (> 10 m3). The tanks

are partially filled with green-water at Z1 (20 - 25 % volume). The tank is filled with clean

seawater during the course of the Z2 - Z3 stages and during the Z4 and M stages water is

exchanged on flow-through basis (Hamasaki et al., 2002b).

Dead larvae and uneaten food that accumulate on the tank bottom are generally

siphoned out of rearing vessels daily (Baylon and Failaman, 2001; Quinitio et al., 2001) and

care must be taken to avoid siphoning out larvae which have sunk to the bottom of the

container. A biofilm develops on the sides of the tank during culture. Williams et al. (1998)

achieved significantly higher survival in 5-l bowls when the biofilm was removed daily.

Cleaning the biofilm from the tank sides can however release large amounts of bacterial

flock into the water column and cleaning must be done by careful vacuuming or the tank

must first be drained down before cleaning.

Larvae ingest micro-algae by chance when swallowing water. Although the presence

of micro-algae in the water prolongs the survival of Z1, they cannot moult to Z2 unless the

diet is supplemented with zooplankton (Brick, 1974). Several genera of microalgae

including Tetraselmis, Skeletonema, Chlorella, Nannochloropsis, Chaetoceros and

Isochrysis have been added to the rearing water during larval rearing at densities ranging

from 5 104 to 5 105 cells ml-1 (Dat, 1999b; Djunaidah, et al., 1998; Mann et al., 1999b; Mann

et al., 2001; Quinitio et al., 2001; Williams et al., 1999b; Zeng and Li, 1999) in order to

“condition” the water and to serve as food for rotifers and Artemia. Species with high

HUFA levels such as N. oculata and I. galbana are added in order to continually enrich

rotifers and Artemia. The effect of background algae on larval survival and growth is


24

however not clear. The current trend in mud crab larval rearing is to use algae at least during

the rotifer feeding stages and for megalopae.

In conclusion, it appears that the current hatchery systems around the world are more

a reflection of available facilities (e.g. tank sizes and live food production capacity) than

best practice and are indicative that a range of approaches to mud crab larviculture can be

successful. An ideal system for mud crab larval rearing has not yet been perfected.

Feeding and nutrition

Larvae are fed as soon as possible after transfer into the rearing vessels. High protease

activity in newly-hatched zoeae indicates their ability to digest food immediately after hatch

(Li et al., 1999). Delaying feeding for up to 24 hours after hatching has no significant effect

on survival (Lumasag and Quinitio, 1998). However the effects of such starvation on the

later stages are not known. Starving larvae for longer than 48 hours induces high mortality

despite resumption of normal feeding (Djunaidah et al., 2003; Lumasag and Quinitio, 1998).

The point of no return (PNR) for newly-hatched larvae has been estimated at 30 hours and

96 hours for S. paramamosain and S. serrata, respectively (Li et al., 1998; 1999). In an

earlier study Mann and Parlato (1995) found that S. serrata Z1 could survive for up to 142

hours without feeding but were not able to moult to Z2. Temperature significantly

influences the PNR. Newly-hatched S. serrata have a PNR50 of 57.6 hours at 28 ºC and 91

hours at 24 ºC (Lumasag and Quinitio, 1998).

Z1 and Z2 are usually fed on rotifers (Brachionus spp.) (Li et al., 1999; Mann et al.,

1999b; Takeuchi et al., 2000, Zeng and Li, 1999). Larvae fed rotifers benefit from the

supplementation of Artemia nauplii as early as Z1 but Z2 utilize Artemia more efficiently

(Li et al., 1998). If Artemia are withheld beyond Z3, growth and survival are compromised

(Li et al., 1998; Suprayudi et al., 2002a; Takeuchi et al., 2000; Zeng and Li 1999).

Although larvae can be reared on Artemia nauplii from hatch, survival is usually

enhanced by the addition of rotifers to the diet (Baylon and Failaman, 1999; Ong, 1966).

Larvae are commonly reared on rotifers during Z1 and Z2 while Artemia are usually

introduced at Z3 (Li et al., 1999; Mann et al., 1999b; Nghia et al., 2001b; Takeuchi et al.,

2000; Quinitio et al., 2001). However, in large systems (10 - 100 m3 tanks) that cannot be

flushed regularly, feeding with Artemia is delayed to Z4 (Hamasaki et al., 2002b). Although

Artemia are typically provided as newly-hatched nauplii, Z3 are large enough to consume


25

metanauplii, allowing for the delivery of supplementary nutrients via bio-encapsulation (see

below). On-grown Artemia (5 days old to adult) provide a larger sized prey item and are fed

to Z5 and megalopae (Mann et al., 1999b; Quinitio and Parado-Estepa, 2003).

Z1 are functionally passive feeders relying on chance encounters with their food

(Heasman and Fielder, 1983). Laboratory based studies have indicated that high densities of

rotifers (30 - 80 ml-1) significantly enhance survival (Djunaidah, et al., 1998; Suprayudi et

al., 2002a; Zeng and Li, 1999). However, in mass production systems practical

considerations such as maintenance of water quality and rotifer production capacity

sometimes dictate that lower densities (10 - 20 ml-1) are used (Baylon et al., 2001b; Mann et

al., 1999b; Quinitio et al., 2001; Williams et al., 1998). Artemia are generally provided at

0.5 - 10 ml-1 (Baylon et al., 2001a; Mann et al., 1999b; Mann et al., 2001; Quinitio et al.,

2001; Williams et al., 1999b; Zeng and Li 1999;) although when larvae are densely stocked

(100 larvae l-1) Artemia densities of up to 20 ml-1 are used (Nghia et al., 2001b). The

optimum density of Artemia has not been investigated and it appears that a practical

approach is commonly taken with Artemia density based on the stocking density of larvae,

the culture system used and the available budget.

Mud crab larvae accept inert food. Formulated feeds adapted from shrimp larval diets

(Quinitio et al., 1999) and dried Artemia flakes (Nghia et al., 2001b) have been used with

some success. Neither is an effective replacement for live food, but they would appear to

have some potential as supplements. In the Philippines, it is common practice to supplement

live food with 2 mg l-1 day-1 formulated feeds in large tanks (> 10 m3) from Z1 to Z5

(Quinitio et al., 2001).

Inferior nutrition may be a factor contributing to the highly variable survival and the

high susceptibility to disease often recorded in mud crab larviculture. Poor nutrition, even if

confined to the early larval stages, has been suggested as a cause for the phenomenon of

moult death syndrome (MDS) - high mortality during or after the moult from Z5 to M

(Hamasaki et al., 2002a, b; Li et al., 1999; Mann et al., 2001; Marichamy and Rajapackiam,

2001; Quinitio et al., 2001; Suprayudi et al., 2002b; Zeng and Li, 1999).

It is generally accepted that larvae eat zooplankton in the wild which are nutritionally

superior to rotifers and Artemia. Copepods commonly caught from the wild such as Acartia

tsuensis and Pseudodiaptomus spp. have been used as live feeds (Toledo et al., 1998) but

ensuring a regular supply is difficult as they cannot be cultured consistently at high densities

(Delbare et al., 1996). The nutritional quality of rotifers and Artemia can be improved by


26

enriching them with nutrients in a process known as bioencapsulation (Coutteau et al., 1997;

Kanazawa and Koshio, 1994; Rees et al., 1994; Wouters et al., 1997). The effect of

enriching the live food with essential fatty acids (EFAs) contained in algae, yeasts and

formulated emulsions on survival and growth of mud crab larvae has been tested.

Suprayudi et al. (2002b) recorded significantly improved survival of S. serrata larvae

after boosting the total n-3 highly unsaturated fatty acid (Σn-3 HUFA) content of rotifers

from 3 - 5 mg g-1 to 7.6 - 8 mg g-1. However, Hamasaki et al. (2002a) reported abnormal

development of S. serrata Z5 leading to high mortality at metamorphosis to megalopa as a

result of feeding the larvae rotifers containing Σn-3 HUFA levels above 6 mg g-1. Whereas

Suprayudi (2002b) recorded high mortality through the moult to megalopa and first crab

after feeding rotifers boosted to 31 mg g-1 Σn-3 HUFA, in Vietnam enriching rotifers with

emulsions containing 30 % Σn-3 HUFA enhanced growth of S. paramamosain larvae. The

contradictory results of these studies indicate that the requirement for Σn-3 HUFA may

differ between batches or species. They also indicate that specific n-3 HUFAs may be more

important than the absolute levels of n-3 HUFA.

Suprayudi et al. (2002a) found that the low levels of eicosapentaenoic acid (20:5 n-3)

(EPA) (3 mg g-1) and docosahexaenoic acid (22:6 n-3) (DHA) (1 mg g-1) found in Artemia

are sufficient for good survival through the moults to M and C1 for S. serrata. These results

were supported by Mann et al. (2001) who found that boosting levels of EPA (39 mg g-1)

and DHA (15 mg g-1) in Artemia did not lead to significant improvements in survival.

Kobayashi et al. (2000) even suggested that levels of 16 - 35 mg g-1 EPA and 17 - 29 mg g-1

DHA in Artemia were excessive, compromising survival of S. tranquebarica. However,

Kobayashi et al., (2000) found that enriching Artemia with EPA at 13 mg g-1 while

maintaining DHA at trace levels enhanced survival indicating that the ratio of DHA/EPA in

Artemia may be more important than absolute levels. There are indications that the

DHA/EPA ratio in the rotifers is also important. Kobayashi et al. (2000) found that a

relatively low DHA/EPA ratio in rotifers (0.07) produced significantly better survival than

higher ratios (0.8 - 0.9) for S. tranquebarica larvae. This is supported by Suprayudi (2002b)

who recorded best survival through first metamorphosis for S. serrata when feeding rotifers

containing a relatively low DHA/EPA ratio of 0.3. However, in Vietnam, zoeal growth and

first metamorphosis of S. paramamosain larvae were significantly enhanced when the

DHA/EPA ratios in the enriching emulsions were high (0.6 - 4) (Vandendriessche, 2003).


27

This is another indication of possible differences in fatty acid requirements between the

different Scylla species.

High arachidonic acid ARA/EPA ratios (0.23) in unenriched as opposed to enriched

(0.02 - 0.05) Artemia significantly enhanced larval survival of S. tranquebarica (Kobayashi

et al., 2000), but no other studies on the ARA requirements of the larvae has been

conducted.

The results of the experiments conducted on the enrichment of rotifers and Artemia

with EFAs have generally been contradictory and our understanding of the requirements of

larvae for fatty acids is scant. The experiments described above represent work on four

different species at different laboratories using different larval rearing techniques. Most of

the studies have concentrated on the enrichment of either rotifers or Artemia, whereas work

in South Africa and Vietnam has revealed that both need to be enriched with EFAs to

significantly improve larval performance. It also appears that the larvae may require

different quantities and ratios of EFAs at different stages of development. Despite limited

information, several inferences concerning the fatty acid requirements of larvae can be

made. Firstly, although HUFA rich algae can maintain the nutritional value of previously

boosted live food and may supply other essential nutrients, the addition of algae to the

rearing tank does not adequately enrich live food with the necessary EFA levels. Live food

needs to be artificially boosted before being fed to the larvae. Secondly, Σn-3 HUFA levels

ranging from 8-10 mg g-1 in live food may be beneficial, but larvae seem unable to tolerate

excessive levels in the diet. Thirdly, EPA and DHA enrichment of the rotifers is required,

but the absolute amounts do not seem to be as important as the DHA/EPA ratio.

4. Nursery

Cannibalism is a common problem and can account for a large percentage of mortality

(30 - 50 %) at both first and second metamorphosis (Dat, 1999b; Quinitio et al., 2001;

Suprayudi et al., 2002a). Asynchronous moulting exacerbates the problem (Quinitio et al.,

2001). Moulting synchronicity can be improved by enriching the live food (Takeuchi et al.,

1999). Megalopae are commonly transferred to separate rearing systems which reduces the

rate of cannibalism. Megalopae change from planktonic to benthic orientation at 4 - 5 days

after first metamorphosis and are typically transferred prior to this change. Metamorphosis

to C1 occurs 7 - 10 days after metamorphosis to M (Baylon and Failaman, 1999; Dat,


28

1999b). S. serrata megalopae tolerate handling (Quinitio and Parado-Estepa, 2000),

however megalopae of S. paramamosain appear to be more delicate and it is common

practice in Vietnam to leave S. paramamosain megalopae in the culture tanks until they

have moulted to C1.

Megalopae are reared through to crab stages 4 - 7 (crablet) in flat-bottomed tanks

ranging from 8 l (Baylon and Failaman, 1999) to 2000 - 8000 l (Williams et al., 1998; Dat,

1999b). Cannibalism continues to be a problem throughout the nursery stage though it can

be partially averted by placing additional substrata and shelters into the rearing vessels

(Mann et al., 1999b; Marasigan, 1998; Quinitio et al., 2001). Coconut palm fronds,

seaweeds (e.g. Gracilaria bailinae) and a variety of plastic meshes and pipes have been

used for this purpose.

Megalopae are also reared in earthen ponds (50 m2 filled to a depth of 80 - 100 cm)

and prepared to promote a dense plankton bloom prior to stocking (Marasigan, 1998;

Rodríguez et al., 1998). Crablet growth is faster in ponds than in tanks and this is

presumably due to the presence of naturally occurring live feeds, although controlling

mortality in ponds is more difficult (Rodríguez et al., 2001). Management of nursery ponds

can be simplified by culturing megalopae and crablets in suspended cages or “hapa” nets (1

mm mesh size and 1 × 1 × 1.5 m deep) (Marasigan, 1998; Rodríguez et al., 2001).

Megalopae are stocked at rates of 1 to 150 l-1 in indoor tanks (Baylon and Failaman, 1999;

Dat, 1999b; Quinitio et al., 2001) and 0.1 to 125 m-2 in outdoor ponds (Marasigan, 1998;

Rodríguez et al., 1998). The density at which megalopae are stocked, depends on when the

crablets are to be harvested. If crablets are harvested at C1/C2, they can be stocked at

densities exceeding 300 m-2. However, survival to later crab stages is compromised at such

high densities (Rodríguez et al., 1998).

Megalopae are fed a variety of minced fresh and frozen feeds (Hamasaki et al., 2002b;

Quinitio and Parado-Estepa, 2003; Williams et al., 1999b) and particulate diets formulated

for penaeid prawn post larvae. Supplementation of inert diets with Artemia nauplii or adults

significantly increases survival to C1 (Baylon and Failaman, 1999; Heasman and Fielder,

1983; Mann et al., 1999b; Marasigan, 1998; Quinitio et al., 2001; Williams et al., 1999b).


29

5. Bottlenecks to commercial production

Larvae have been produced on a commercial scale in several countries. Rearing trials

conducted on S. serrata in China (Li et al., 1999) and more recently (2003) in the

Philippines and Australia have produced thousands of C1 in commercial shrimp hatcheries

with survival ranging from 3.2 - 10 %. In Japan 1.5 million hatchery-produced S.

paramamosain and S. serrata juveniles were released in 1999 for restocking purposes

(Hamasaki, 2002). Although survival rates have improved markedly over the last decade,

there are still several bottlenecks that prevent widespread commercial seed production.

Larvae are vulnerable to a variety of diseases and parasites. Under suboptimal

conditions, Zoothamnium, Vorticella and other sessile ciliates can attach to the larval

integument (Dat, 1999b); fungi of the genera Haliphthoros, Lagedinium and Atkinsiella can

be transmitted to newly-hatched larvae, which are highly susceptible to infection (Hamasaki

and Hatai, 1993a; Kaji et al., 1991; Quinitio et al., 2001). White spot syndrome virus

(WSSV) (Chen et al., 2000) can be introduced.

Bacterial disease is considered to be one of the most important bottlenecks to

commercial hatchery production. The best evidence for the role of bacterial pathogens in the

hatchery is that regardless of other factors, the only treatment known to significantly reduce

mortality is treating the rearing water with antibiotics (Parado-Estepa and Quinitio, 1998;

Mann, 2001). Potential pathogens including Vibrio spp. (Mann et al., 1999b), luminescent

Vibrio, filamentous bacteria (Takeuchi et al., 2000) and others have been identified in larval

cultures where they can rapidly rise to levels generally considered to be deleterious to

crustacean larvae. If not controlled, bacterial numbers can increase by 2 log units on

successive days (Quinitio et al., 2001). Information on how bacteria affect the larvae is

limited. Histological studies have thus far not determined the aetiology or mechanism of

mortality in mud crab larvae (Mann et al., 2001). Additionally there has been no consistent

correlation between larval performance and the structure of the bacterial community or the

presence or absence of particular bacterial strains. The bacterial community of larval

cultures seems to be highly volatile among culture vessels and batches as well as during a

culture cycle in a single vessel (Mann et al., 2001). Mass cultures do not show the same

improved response to hygiene protocols as laboratory scale cultures, possibly due to

differing microbial environments. The common thread is that antibiotics consistently lead to

improved production. Although regular application of antibiotics is not a desirable practice


30

for reasons of sustainability, it has frequently been used for experimentation as it is often the

only means of ensuring the survival of sufficient larvae from which data can be gathered.

Antibiotics are either used as a prophylactic in small scale experiments: e.g. 20 mg l-1

streptomycin (Takeuchi et al., 2000) or 2 mg l-1 Sodium nifurstyrenate (Hamasaki et al.,

2002b) or as a bath (100 mg l-1 Oxytetracycline) for Z1 larvae before being stocked into

pilot scale systems (Baylon and Failaman, 2001).

Alternatives to antibiotics such as formalin and probiotic preparations (bacterial

strains which inhibit known pathogens) have been applied. Mud crab larvae can tolerate up

to 25 µl l-1 of formalin (Kaji et al., 1991) and a 24 hour LC50 for S. serrata Z1 has been

estimated at 37 µl l-1 (Churchill, 2003). Regular dosing with 20 µl l-1 formalin every 2 days

is used to prevent fungal (e.g. Haliphthoros spp.) and bacterial infection of larvae in

Vietnam. Although of less potential risk than antibiotics, microbes can eventually build up

resistance to formalin especially if applied at regular, low doses. Results for larval rearing

trials using probiotic bacteria (and their products) have been inconclusive thus far.

Probiotics have been effective in enhancing seed production for other portunid species

(Nogami and Maeda, 1992) and the technology is receiving a great deal of attention (Irianto

and Austin, 2002; Verschuere et al., 2000). With continued research, probiotics and

immunostimulants may provide good alternatives to antibiotics for mud crab larval rearing

in the future (Lavilla-Pitogo et al., 2002).

A characteristic of mud crab larval culture is the highly variable survival. Successive

batches in a hatchery can produce 80 % survival to C1 or total mortality before first

metamorphosis. Several authors have suggested that variable survival is caused by

differences in egg and larval quality (e.g. Mann et al., 1999a; Millamena and Bangcaya,

2001). Differences in batch quality may also be a factor contributing to differences in results

and apparent contradictions between studies (Zeng and Li, 1999). Egg colour has been used

to estimate egg quality, but no correlation has been found between initial egg colour and the

quality of the eggs or larvae (Churchill, 2003). The quality of newly-hatched larvae has

been tested by subjecting them to stressors such as starvation (Djunaidah et al., 2003) or

high concentrations of salinity, ammonia and formalin (Churchill, 2003). Using stress tests

Churchill (2003) identified female size and DHA, EPA and Σn-3 content of the eggs as

possible determinants of egg and larval quality. Because rearing techniques have not yet

been standardized, validating the accuracy of a stress test is difficult. There is also an

apparent large variability in larval quality within batches so that practical interpretation of


31

stress test results is difficult. A reliable test for evaluating the quality of Z1 mud crab needs

to be developed. Ultimately the domestication of broodstock and the development of a

standardized broodstock diet could reduce variability in larval quality at hatch.

6. Discussion

The four Scylla species spawn readily in captivity. Millions of eggs with a high hatch

rate can be produced, year round from a relatively small number of wild-caught or pond-

reared broodstock. Scrupulous hygiene during gonad maturation, egg extrusion and egg

incubation is required to reduce infection of the eggs by pathogens which can reduce egg

viability and be transmitted to the larvae. The typical high variability in survival may be

partly due to variable egg and larval quality. Reliable criteria to determine the quality of

newly-hatched larvae should be developed. Once the life cycle can be closed reliably, egg

and larval quality could be enhanced by providing a domesticated broodstock with a

formulated diet.

Mud crab larvae are sensitive to captive conditions and high mortalities at all stages of

development are common. Causes of mortality include poor water quality, incorrect or

fluctuating environmental conditions, cannibalism, feeding and nutritional deficiencies,

parasitic and fungal infections and viral and bacterial disease. In some instances mortality

has been reduced during early larval stages, but high mortalities are still commonly recorded

towards the end of the rearing period. As the rearing process progresses, water quality

deteriorates. Accumulated faeces, uneaten food and dead larvae provide substratum for the

proliferation of pathogenic bacteria. Although strict hygiene protocols, pretreatment of

seawater and high rates of water exchange can mitigate the problem, antibiotics, formalin

and other anti-bacterial prophylactics are still the only way to obtain consistent survival.

Manipulation of the bacterial community through the application of probiotics may provide

a solution (Lavilla-Pitogo et al., 2002). The pathogens associated with mud crab larviculture

need to be identified and the mode of infection determined.

Mass mortality at metamorphosis to megalopa and C1 due to MDS has been linked to

nutritional deficiencies and excessive levels of antifungal or antibiotic agents. Even early

nutritional deficiencies can manifest in MDS at the end of the rearing period. Zoeae can be

reared on rotifers and Artemia and reducing the bacterial load, optimizing feeding densities

and improving nutritional quality through bioencapsulation could reduce MDS and improve


32

resistance to stress and disease (Merchie et al., 1997; Mourente and Rodríguez, 1997). A

formulated diet could help to identify the specific nutritional requirements of the larvae and

aid the development of immuno-stimulants. A domesticated broodstock could allow genetic

selection for resistance to disease (Bachère et al., 1995).

Although some research groups have some success in the mass production of mud

crab juveniles, there is still much scope for further research and development before seed

production for aquaculture becomes economically viable and widely adopted.

Figure 1. Life cycle of mud crab. Photo: David Mann; rearranged from Williams et al., 1999b.

CHAPTER 3

Reproductive performance of captive mud crab (Scylla paramamosain) broodstock in Vietnam

Nghia, T.T.*1, Wille, M.2 and Sorgeloos, P.2

1 College of Aquaculture and Fisheries, Can Tho University Email: [email protected] 2 Laboratory of Aquaculture & Artemia Reference Center, Ghent University, Belgium Email: [email protected]

Abstract From 1996 to 2002, reproductive performance of 786 wild S. paramamosain breeders

(355 ± 80 g), purchased from local markets, was recorded. The effect of a number of selected broodstock characteristics, as well as management and environmental conditions on reproductive success was evaluated. Eyestalk ablation improved spawning success, but did not alter the latency period between purchase and spawning. No negative effects of ablation on broodstock survival nor egg quality were found. Breeders collected from an inshore region with higher and more stable average salinity levels tended to perform slightly better than those collected from a region with lower and varying salinity. Females in the range of 300 to 500 g had the best overall reproductive efficiency and are preferred as breeders. Although detached eggs could be incubated artificially, egg incubation by the females themselves was the best practice. The period from March to July was most efficient for broodstock rearing and might correspond to the natural spawning season of S. paramamosain in South Vietnam. September to February could be the period for gonad maturation in the wild. With eyestalk ablation, the most favorable period for artificial reproduction could be extended from February to August.

It was noticed that shading the broodstock tanks was not necessary if shelters for hiding were available. Rearing broodstock in earthen ponds was more efficient; however, management of a pond proved more complicated than tank systems. In terms of complete domestication, broodstock rearing in tanks therefore provides a more practical alternative, provided larger tanks and more suitable substrate are used.

Spawning activity decreased with prolonged time in captivity. Egg quality criteria such as fertilization rate and egg diameter did however not vary in function of time in captivity.

Overall, controlled reproduction of wild mature broodstock females of S. paramamosain for research and pilot production is not problematic, especially with the practice of eyestalk ablation. Although individual females had high fecundities and fertilization rates, spawning and hatching success were however not very high. In this respect, it is hypothesized that broodstock captured offshore might be better.

Further research should be undertaken to completely domesticate the species and further document maturation and fertilization characteristics in captive conditions. Therefore dietary requirements and suitable rearing conditions should be investigated. Life history studies in the wild could provide very useful information in this respect.

CHAPTER 3 – Broodstock

34

1. Introduction

Inadequate seed supply is a major constraint for further development of mud crab

farming (Millamena and Bangcaya, 2001). Over-exploitation, environmental pollution and

mangrove destruction have led to declining wild populations (Heasman and Fielder, 1983;

Hill, 1984). As a consequence a renewed interest in mud crab farming has emerged. As in

most areas where larviculture of mud crab Scylla spp. is conducted, mature wild females are

however still relatively readily available, the source of eggs continues to rely on gonadal

maturation and spawning of wild-caught females in captivity (Mann et al., 1999a). In order

not to impose extra pressure on the already heavily exploited stocks, broodstock

domestication would be a convenient and sound alternative to wild spawners for a more

reliable supply of seed (Millamena and Bangcaya, 2001), moreover creating new

opportunities for disease prevention, selective breeding, etc.

Due to the migratory behavior of female mud crabs in the wild (Hill, 1994),

knowledge of spawning, brooding and hatching of eggs under natural conditions is limited

(Mann et al., 1999a). This probably also explains why far less research has been conducted

on broodstock management techniques than on larval rearing. Several studies have however

indicated that variable egg quality is one of the most critical factors underlying the variable

success of controlled seed production (Churchill, 2003; Davis, 2003).

The number of studies on broodstock rearing of Scylla spp. is limited and mainly

focused on the effects of diet and feeding regimen on ovarian maturation, spawning and

hatching rates (Lin et al., 1994; Millamina and Bangcaya, 2001; Zeng, 1987; Zeng et al.,

1991) and on effects of eyestalk ablation and season on the performance of eggs and larvae

(Mann, 1999a). However, many parameters or conditions for broodstock management in

captivity are still not yet understood thoroughly. Moreover reproductive characteristics

might differ significantly between the four species of the Scylla genus and need to be further

investigated. Experience and knowledge on one species of the genus Scylla could however

also prove useful for the other three species of the Scylla genus (Keenan et al., 1998;

Keenan, 1999b).

In this study, reproductive performance of captive S. paramamosain is described.

Although no real factorial experiments were conceived in this study, the reproductive

characteristics of broodstock animals used over a period of six years to produce larvae for


35

larval rearing purposes, were recorded and later on analyzed for the effect of a number of

variables.

In addition, the results of a number of side experiments are reported. Artificial

incubation of shedded eggs was attempted and the evolution of egg diameter during

incubation was documented.

2. Materials and methods

2.1. Broodstock

Broodstock source

Over the years (from 1996 to 2002) a variable number of female crabs was purchased

each month from local markets and transported to the hatchery for research on larval rearing

techniques. They had been caught from either the coast of East South Vietnam or the east

coast of West South Vietnam (or the Mekong Delta) where the monthly salinity levels range

around 30 ± 2 and 25 ± 5 g l-1 respectively (Figure 1). In total, 786 gravid crabs (355 ± 80 g,

ranging from 183 - 700 g) were collected.

Rearing systems and culture conditions

Three different types of rearing systems were used to accommodate the brood crabs. A

first system consisted of 20 individual plastic rearing containers of 70-l (0.7 × 0.4 × 0.25 m).

The tanks were placed indoors and darkened completely by black covers. The 20 tanks

were connected to a central biofilter of 700 liter (50 % of the total volume of the rearing

tanks). In a second system, the crabs were individually housed in ten 100-l (1 × 0.4 × 0.25

m) compartments of a 2 × 2 × 0.5 m cement tank. Two of such cement rearing units were

connected to a third cement tank which served as biofiter. The tanks received natural

daylight but were protected by a roof. In a last system, crabs were reared communally in a

60-m3 (40 × 3 × 0.5 m) earthen pond.

Both tank systems were operated in recirculating mode with approximately 100 %

water exchange every 2 - 3 hours. Once a month, approximately 80 % of the water was

renewed. A 5-cm sand layer was provided on the tank bottom to allow proper attachment of


36

spawned eggs to the tertiary setae of the female’s abdominal flap. In the cement tanks, a

ceramic tile was placed on the bottom of each compartment as shelter for the animals.

Rearing water for the tank systems (30 ± 1 g l-1) was diluted from brine (90 - 110 g l-1) with

tap water and chlorinated before use. Water temperature was not controlled (Table 1). The

concentration of NH4+, NO2

- and NO3+ during culture ranged between 0 - 0.3, 0 - 0.1 and 0 -

5 mg l-1, respectively. Every other day, a dose of 20 µl l-1 formalin was applied for the

whole system.

For the pond system, rearing water was renewed completely every 5 days upon

checking the breeders for spawning. A dose of 50 µl l-1 formalin was applied directly to the

pond after new seawater was filled. Water temperature in the earthen pond varied by the

season (29.6 ± 1.4 °C). Freshwater from a well was pumped to the pond to control the

salinity at 30 ± 4 g l-1. As the pond bottom was covered by a 5-cm mud layer and water level

was high, no extra shelters were provided. The concentration of NH4+, NO2

- and NO3+ of

rearing water ranged similarly as in the tank system.

Broodstock management

Prior to stocking in the hatchery, the crabs were bathed in a 100 µl l-1 formalin

solution for 1 hour to kill potential pathogens.

Depending on the need for larvae for larval rearing purposes, part of the animals was

unilaterally eyestalk ablated to enhance spawning. Before ablation, animals were

anaesthetized in a 1 - 3 g l-1 chloroform solution, with slight aeration. After 20 to 30

minutes, the crab became motionless. The animal was removed from the water and the

eyestalk was scorched by a hot pair of pincers. Then the part of eyestalk above the scorched

location was cut by a pair of scissors. This method also disinfected the wound. After

ablation, the crab was put in fresh seawater with strong aeration. The animal normally

recovered from anaesthesia after 5 - 10 minutes.

Each crab was fed a daily ration of 10 - 15 g of fresh marine squid, bivalve and shrimp

on alternate days.


37

2.2. Egg incubation

After spawning, the berried crab was bathed in 100 µl l-1 formalin solution for 1 hour

and transferred to a 70-l plastic tank connected to a biofilter for incubation. Daily

management consisted of siphoning out waste material and shedded eggs from the tank

bottom and controlling temperature (30 ± 1 °C), salinity (30 ± 1 g l-1) and ammonia and

nitrite levels (similar to broodstock rearing tanks). Every other day, the crab was bathed in a

50 µl l-1 formalin solution for 1 hour to prevent infestation of the eggs with fungi and

bacteria. One to two days prior to hatching, the female was moved to a 500-l fibreglass tank

for hatching. During egg incubation, the crabs were not fed.

For the pond system, water was drained every 5 days to check for berried females.

Incubation was then carried out as described above.

In some cases, the spawned eggs did not attach to the abdomen. In an attempt to make

use of these shedded/detached eggs, an artificial incubation experiment was carried out in

700-ml glass cones. Eggs were incubated at 50 eggs ml-1 with rather strong aeration.

Temperature was controlled at 28 °C. Two treatments in triplicate were set up, with or

without daily application of 30 µl l-1 formalin. Formalin addition was done upon the daily

90 % water exchange.

2.3. Reproductive performance

From 1996 to 2002, in total, reproductive performance of 786 brood crabs was

recorded. In order to standardise measurements, reproductive performance was followed for

only 60 days (763 females). To calculate survival time in captivity, part of the animals (182

females) where however further maintained beyond 60 days. Twenty three females spawned

after more than 60 days after purchase; these data are however not included for calculation

of reproductive parameters.

Reproductive performance was evaluated through the following parameters:

- time to spawn (days) = the latency period between stocking the female and spawning

- ablation-spawn time (days) = the latency period from eyestalk ablation and spawning of

the female


38

- survival time in captivity (days) = period from stocking to mortality of the crab

- spawning success = % of the females that spawning

- hatching success = % of the females that produced viable larvae to the total number of

spawning females

- reproductive efficiency = % of the females that produced viable larvae of the total number

of females (= spawning success × hatching success)

- fertilization rate = % of eggs exhibiting an eyespot on day 5 - 6 after spawning to the total

number of sampled eggs (average of triplicate samples of 30 eggs)

- egg diameter (µm) = average diameter of 30 eggs, measured under a microscope

- total zoea 1 production (103 Z1) = number of Z1 larvae produced by a female in a single

spawning event (estimated volumetrically)

- relative zoea 1 fecundity (103 Z1 g-1) = total number of viable Z1 per spawning event over

female weight (g).

In order to develop best management practices, reproductive parameters were later on

treated statistically to determine the effect of a number of factors related to broodstock

management and environmental conditions: eyestalk ablation (intact versus ablated

females); rearing system (plastic versus cement tanks, earthen pond versus cement tanks);

broodstock source (low- versus high-salinity region); month of the year; monsoon season

[rainy (May-October) versus dry (November-April) season]; and temperature-based season,

[spring-summer (March-August) versus autumn-winter (September-February) season]; and

individual female weight and time to spawn. Table 2 gives an overview of broodstock

acquisition in function of broodstock source, month of the year, as well as the percentage of

animals that was ablated in each period.

As mentioned earlier, for studying most of these factors no real experiments were

conceived. Depending on the factor under investigation, the complete or only part of the

data set was used (see Table 2). For investigating the effect of eyestalk ablation on

reproductive performance, data from all 763 females were used, consisting of 358 intact and

405 ablated females. For the effect of broodstock source, only the data from 2001 (241

females) were used, consisting of 116 and 125 females from the high- and low- salinity

region, respectively. To assess the influence of season (month of the year), female weight

and time to spawn again data from all 763 females were used.


39

However, also a number of planned comparisons were set up. In 2000, every month

approximately 20 animals were bought from the Mekong Delta and divided equally over a

cement and plastic tank system in order to compare reproductive performance in both

systems. Simlarly, 50 females were sourced from the high-salinty region and divided over

both systems for comparison. In total, data of 272 females were used for this study. In

another experiment, 117 females purchased from the Mekong Delta were divided over an

earthen pond (stocked at 6 - 12 m2 female-1) and cement tanks (0.4 m2 female-1) in order to

compare reproductive performance in both systems more straightforward.

2.4. Statistical analysis

One-way analysis of variance (ANOVA) was used to compare data. Homogeneity of

variance was tested with the Levene statistic (P or α value was set at 0.05). If no significant

differences were detected between the variances, the data were submitted to a one-way

ANOVA. The Tukey HSD post-hoc analysis was used to detect differences between means

and to indicate areas of significant difference. If significant differences were detected

between variances, data were transformed using the arcsine-square root (for percentage, i.e.

fertilization rate) or logarithmic transformations (for other other data, i.e. time to spawn,

ablation to spawn time, survival time in captivity, egg diameter, total zoea 1 production and

relative zoea 1 fecundity) (Sokal and Rohlf, 1995). The two-tailed Fisher exact test

(modified from the contingency table method) was used to compare ratios (expressed in

percent, i.e. spawning success, hatching success, reproductive efficiency and other

parameters). All data are presented as mean ± standard deviation when using the Tukey test

or as a ratio/percentage without standard deviation when the Fisher exact test was used. The

Pearson’s correlation coefficient was used to examine the correlation between selected

factors and reproductive characteristics. P was set at both 0.05 and 0.01. Whenever

differences are significant at P < 0.01, this is also indicated. All analyses were performed

using the statistical program STATISTICA 6.0.


40

3. Results

3.1. Effect of selected management and environmental parameters on reproductive

performance

Eyestalk ablation

Table 3 compares the reproductive performance of ablated and intact females.

Although ablated animals spawned on average 19 days after ablation, overall time to spawn

was similar between both groups (27 - 28 days). Spawning success and (due to a similar

hatching success) also reproductive efficiency were almost double in the ablated group

compared to intact females (35 versus 20 % and 11 versus 6 %) (both significantly different

at P < 0.01). Fertilization rate and relative Z1 fecundity on the other hand were unaffected.

Interestingly, also survival time in captivity was not affected by the ablation procedure.

Rearing system

- Plastic versus cement tank system

Table 3 also presents the reproductive performance of females reared in 2 different

tank systems. Both the spawning success and the reproductive efficiency in the cement

tanks were significantly higher than those in the plastic tanks (30 and 16 %, P < 0.01 and 6

and 1 %, P < 0.05 for cement and plastic tank systems respectively). Although not

statistically significant, also the hatching success was considerably higher in the cement

tank system (19 % as opposed to only 5 % in plastic tanks). Other reproductive

characteristics were not significantly different.

- Cement tank system versus earthen pond

The reproductive performance of broodstock reared in cement tanks and the earthen

pond is presented in Table 4. A higher spawning success and a lower percentage of females

that did not spawn were observed in the pond (both significantly different at P < 0.01). The


41

percentage of dead or missing females on the other hand, tended to be higher in ponds

compared to the tank system.

Where the percentage of females that successfully produced viable larvae seemed to

be slightly higher in tanks, a lower percentage of females shedding eggs during incubation

was found in the pond (not significantly different).

When combining spawning success and hatching success, the reproductive efficiency

in the pond was significantly higher (P < 0.05) compared to that of the tanks (31 and 15 %

respectively).

Broodstock source

Broodstock source (high versus low salinity regions) did not affect reproductive

performance significantly (Table 3). Females from high salinity regions however tended to

have however a higher (not significant) spawning and hatching success resulting in a higher

overall reproductive efficiency (13 versus 6 % for high and low salinity respectively). Time

to spawn and the latency period between ablation and spawning of the females from the

high-salinity region also tended to be slightly shorter than those of females collected from

low-salinity water.

Month and seasonal cycle

- Month

Tables 5 and 6 present the reproductive performance in relation to the month of

stocking of the females. In Table 5, data for intact and ablated females are presented

separately to detect interaction between both factors. As sample size for these split data is

for certain months small, data were also pooled for all females in Table 6. It should be noted

that the data were also affected by the broodstock source that varied from month to month.

Time to spawn showed no significant differences for both intact, ablated and all

females. The extreme data for the intact group in August (4 days) and December (60 days)

are attributable to the small sample size for these months (Table 5). Based on Table 6,

slightly shorter (not significant) time to spawn values were found from April through

August (on average 22 - 24 days). The shortest time to spawn was observed for females


42

collected in November (19 days); this value however only represents ablated females (see

Table 5).

With the exception of November, the latency period between ablation and spawning

(Table 6) seemed also shorter (though not significantly) from April to August. In January

and December, it took a considerably longer time to trigger the crabs to spawn (27 and 34

days, respectively). A statistically significant correlation was found between both time to

spawn and ablation-spawn time of the broodstock (y) and the average monthly temperature

(x, see Table 1) (r2 = 0.35, P < 0.05 and 0.64, P < 0.01 respectively, Table 6). For ablation-

spawn time, this relationship could best be described by the equation y (days) = 257.4 - 8.4

x (°C).

Based on the spawning success of the intact group (Table 5), it was apparent that the

spawning activity was higher in females collected from March to July (ranging from 23 to

41 %) compared to those collected during the other months (0 - 17 %). Eyestalk ablation

widened the period with high spawning success almost throughout the year (ranging from

18 to 52 % in January to November). The pooled data presented in Table 6 show that the

spawning success tended to increase from January to July (from 24 to 41 %) with a peak in

June - July (40 - 41 %) and thereafter decreased again towards the lowest value in

December (11 %).

The hatching success revealed a similar tendency as the spawning success. This was

especially apparent for the intact females, i.e. only the intact females collected in March to

August produced viable larvae (Table 5). Also from the pooled data presented in Table 6, it

is clear the hatching success tended to be higher for crabs collected in February to August

(29 - 55 %) and less or no hatching occurred in those collected from September to January

(0 - 27 %). With the exception of December, ablated females on the other hand managed to

successfully hatch eggs (18 - 50 %) year round (Table 5).

Similarly, February to August tended to result in the highest reproductive efficiency

(10 - 14 %) compared to 0 - 7 % in the remaining months) (Table 6).

Except the spawning success, most reproductive characteristics correlated

significantly (P < 0.05) with the month temperature (Table 6).


43

- Monsoon season

In Table 7, reproductive performance is regrouped according to the monsoon seasons.

None of the reproductive characteristics evaluated however differed significantly between

the rainy and dry season.

- Temperature-based season

Table 7 also groups reproductive performance of the crab females according to the

temperature-based seasons [warm (spring-summer) and cool (autumn-winter) seasons] (see

Table 1). Similarly to the effect of the month, the reproductive ouput increased significantly

in the warmer spring-summer (March to August), i.e. shorter time to spawn (P < 0.05) and

ablation-spawn time (P < 0.01); and higher spawning success and reproductive efficiency

(both with P < 0.01). Also the hatching success, the fertilization rate and the relative Z1

fecundity tended to be higher in the warmer season.

Female weight

Table 8 does not show any significant differences in reproductive performance

between the 3 weight classes. Females of 300 - 500 g tended to have the highest

reproductive efficiency, fertilization rate and total number of Z1 produced. The largest

weight class tended to result in the highest spawning success but much lower hatching

success. Also relative Z1 fecundity seemed much lower for the last group (1,800 Z1 g-1

female). Females in the weight classes 100 - 300 g and 300 - 500 g had similar relative Z1

fecundities of approximately 2,800 Z1 g-1.

The correlations between the female weight and some selected reproductive

characteristics are presented in Table 9. The egg diameter did not vary by the female weight

(r2 = 0.01, A). In the small weight class (236 - 415 g), the egg fertilization rate increased

significantly (P < 0.01) when female weight increased (r2 = 0.52, B); whereas in the larger

weight class (425 - 552 g), no correlation was found (r2 = 0.01, C). Logically, relative Z1

fecundity was correlated significantly (P < 0.01) with fertilization rate (r2 = 0.46, E). Time

to spawn on the other hand had no effect on fertilization rate (r2 = 0.00, D).


44

Time to spawn

Table 10 presents reproductive performance and egg quality criteria of females that

spawned either between 0 - 15, 16 - 30 or 31 - 60 days after stocking in the hatchery. No

significant differences were found for any of the criteria tested. 30 % of the spawning events

occurred in the first 15 days and 65 % in the first month. Only 35 % of the spawning events

took place in the second month of captivity. Egg fertilization rate did not vary in function of

time in captivity and varied around 60 %. Although not significant, the relative number of

viable Z1 produced seemed to be higher in the groups that spawned between 16 and 30 days

in captivity.

3.2. Artificial incubation of eggs and egg diameter during incubation

When incubating shed eggs artificially (data not shown in the tables), the hatching rate

in the treatment using formalin was significantly higher (P < 0.01) than for untreated eggs

(55 ± 1 and 25 ± 2 %, respectively).

In Table 11, egg diameter is presented in function of incubation time. The diameter of

viable eggs increased steadily during incubation from 287 µm immediately after spawning

to 387 µm upon hatching. The diameter of eggs that proved non-viable increased only

slightly during incubation however and from day 3 onwards, a significant difference (P <

0.01) could be noticed with the viable ones.

4. Discussion

4.1. Effect of selected management and environmental parameters on reproductive

performance

Eyestalk ablation

Worldwide commercial maturation of female penaeid shrimp relies almost exclusively

on the technique of unilateral eyestalk ablation (Browdy, 1992; Fingerman, 1997). Eyestalks

are the endocrine center for regulating many physiological mechanisms, such as moulting,

metabolism, sugar balance, heart rate, pigmentation, and gonad maturation. Therefore,


45

unilateral eyestalk ablation affects all aspects of shrimp physiology (Quackenbush, 1986).

As they belong to the crustacean group like shrimp, also for Scylla species the technique has

been applied ever since the start of artificial reproduction of the species.

Although time to spawn was not enhanced, eyestalk ablation did not result in any

negative effects on the crab broodstock (e.g. survival time in captivity) or egg quality

parameters (fertilization rate and egg hatchability). Ablation on the other hand significantly

improved the spawning success and hence, increased the reproductive efficiency. Ablated

females could also produce newly-hatched zoeae almost year-round, whereas the intact

females were only able to spawn in some months of the year (March - August). Most studies

have agreed that eyestalk ablation improves reproductive performance. Mann et al. (1999a)

observed no adverse effects of eyestalk ablation on egg and larval production in S. serrata.

These authors however found that eggs of ablated crabs were on average larger than those of

intact crabs and the proportion of non-viable eggs and larvae was lower for ablated crabs. In

our study, we also found a slightly higher average egg diameter and fertilization rate for

ablated females, but the results were not significantly different. Millamena and Bangcaya

(2001) found that ablated females were more fecund and presented higher survival,

suggesting that ablation did not bring undesirable stress to the crabs. However, the authors

observed that the intact females had better egg fertilization rates with higher total number of

zoea produced. In our study, the relative Z1 fecundity seemed marginally lower in the

ablated group. It therefore seems that, as in shrimp, eyestalk ablation makes maturation and

spawning of mud crab more predictable, but may have associated problems like

deterioration in spawn quality and quantity over time (Emmerson, 1980; Primavera, 1985;

Tsukimura and Kamemoto, 1991), or, depending on the conditions, lead to conflicting

results on spawn size, hatching success and other variables (Browdy, 1992).

Rearing system

The two main differences between the plastic and cement tank systems were the water

volume and the light conditions. A review of mud crab broodstock rearing techniques

showed that females kept either individually in small bins (60 - 300 l) or communally in

large tanks ranging from 1 to 12 m3 (Baylon et al., 2001a; Dat, 1999b; Hamasaki, 2002;

Mann et al. 1999a; Millamena and Quinitio, 2000; Williams et al., 1998) have all been

successfully used. Davis (2003) concluded that neither the size nor the type of the rearing


46

container seems to affect maturation. Therefore, the slight difference in rearing water

volume of the two container types (70 and 100 l) as such is unlikely to be the main factor

that resulted in the difference in reproductive performance. Mud crabs are nocturnal and

inhabit turbid estuarine water (Barnes et al., 2002; Dat, 1999a; Hill, 1978). Moreover they

are known to be very cannibalistic and territorial. Therefore, shelters are usually provided to

reduce stress and prevent cannibalism (Djunaidah et al., 2003; Hamasaki, 2002; Millamena

and Quinitio, 2000; Millamena and Bangcaya, 2001). The cement tanks in our study were

not shaded but a shelter was supplied for each crab. Possibly the natural daylight and

photoperiod, combined with the slightly more spacious (and hence less stressful)

accommodation in the cement tanks resulted in the significantly higher spawning success

and reproductive efficiency compared to the completely darkened plastic tanks. Davis

(2003) observed a kind of “spawning syncronicity” between breeders. He suggested that

pheromones released by spawning crabs or developing eggs might induce other crabs to

spawn. Although the rearing water was also shared between crabs in the plastic tank system

through the central biofilter, it could be that the more direct contact between crabs in the

cement tank system was another factor that positively influenced spawning success.

However, it is difficult to conclude which factor(s) have positively enhanced the

spawning in the cement tank system. To this point, it can only be concluded that covering

the broodstock containers completely is not necessary if shelters are available.

In the earthen pond, the females were subjected to less stress than in the tanks due to

the larger water volume, a more frequent supply of new seawater, a lower stocking density

and more natural rearing conditions. These favourable conditions enhanced the spawning

success (and consequently overall reproductive performance) significantly in the earthen

pond compared to cement tanks. A larger pond bottom with more suitable places for hiding

and spawning might also support better spawning with proper egg attachment to the

female’s abdominal flap (i.e. fewer females with detached eggs in the pond). However,

rearing the broodstock in ponds also had some disadvantages. This was clear from the high

number of escapees and dead crabs in the earthen ponds. Rearing the crabs communally

obviously increases cannibalism. Also the hatching success was lower in ponds than in

tanks. The higher infestation rate of eggs with parasites could be attributable for this. Also

Quinitio and Parado-Estepa (2003) observed that eggs from broodstock spawned in ponds

were occasionally heavily infested with parasites. This problem was probably aggravated by


47

the fact that spawned females from tanks were treated with prophylactics shortly after

spawning; while for ponds, some females were only collected several days after spawning.

Management of a pond is also much more complicated than tanks and does not allow tight

control of all parameters (e.g. water exchange, evaluation of feeding regime, hygiene

control). It moreover requires more labour.

In terms of complete domestication, broodstock rearing in tanks is therefore preferred,

provided larger water volumes and suitable substrate conditions are used as a compromise.

Broodstock source

Breeders from both regions proved useful to obtain viable larvae as none of the

reproductive characteristics were significantly different. However, broodstock from the

high-salinity region tended to be slightly better overall. This difference could result from the

more stable and higher average monthly salinities on the coast of East South Vietnam

compared to the east coast of the Mekong Delta that is influenced by the large Mekong

river. It is hypothesized that because of the lower salinity in the Mekong Delta compared to

East South Vietnam, crabs caught in the former region are less mature and therefore might

be less suitable as broodstock. Offshore migration of Scylla females for spawning was

reported by several authors (Arriola, 1940; Hill, 1975; Hill, 1994; Hyland et al., 1984; Le

Vay et al., 2001; Ong, 1966; Poovichiranon, 1992; Tongdee, 2001). Also in the Mekong

Delta, mature females of S. paramamosain seem to move from estuarine mangroves into the

subtidal fishery as reported by Le Vay (2001). It therefore seems possible that females

collected in low salinity waters are on average less mature as more mature ones move out

and therefore, there is less chance to catch advanced mature crabs near the coast.

Month and seasonal cycle

- Month

From May to August 2002 and 2003, abundant recruitment of small crablets (0.5 - 1

cm carapace width) on the coastal mudflat of South Vietnam was reported (VASEP, 2003).

Considering that the spawning season should date back about 1 - 2 months, the natural

spawning season for S. paramamosain should then be from March to July. This is in


48

agreement with the high spawning success of the intact females (not affected by eyestalk

ablation) collected in March - July observed in this study. There was also a tendency that the

time needed by females to spawn after induction by eyestalk ablation was shorter from April

to July. Also Dat (1999b) designated April and May as the peak spawning season due to

high spawning rates observed in that period. Overall, it could be considered that the main

spawning season of S. paramamosain mud crab in South Vietnam extends from March to

July.

However, spawning females could be obtained year-round in this study. Observations

on maturation and spawning of Scylla species show that for nearly all populations,

reproduction is continuous throughout the year, with some seasonal peaks (Le Vay, 2001).

Reproductive activity of S. serrata occurs year-round at low latitudes and seasonally at

higher latitudes (Heasman, 1980; Quinn and Kojis, 1987). In China, two spawning peaks

were reported for S. paramamosain (Li et al., 1999).

Although our data indicate readiness to spawn to be highest from March to July, Le

Vay (2001), in a study on population dynamics of S. paramamosain in Vietnam, recorded

mature females throughout the year, with a peak in September - October. Dat (1992) on the

other hand found a maturation peak in December - February. As maturation and spawning

are separate processes in crustaceans, which can be considerably shifted in time, these

observations do not necessarily contradict our results.

For aquaculture purposes, the combination of high spawning and hatching success of

females collected in February - August resulted in an overall higher reproductive efficiency

(10 - 14 %) than for the other months (0 - 7 %). Hatchery production of mud crab would

therefore be easier in this period.

- Monsoon season

Reproductive performance was not very different in the rainy and dry season. Several

papers have however indicated that reproduction of Scylla crabs peaked in the rainy season.

For example, the female to male ratio of S. paramamosain sampled inshore in the Mekong

Delta was much higher in the dry season, which would confirm the offshore migration of

females for spawning in the rainy season (Ut, 2003). Also in tropical populations of S.

serrata, a higher incidence of mature females appears to be associated with seasonal high

rainfall (Heasman et al., 1985). Offshore spawning migration of S. serrata was also found to


49

peak during the rainy season (Arriola, 1940; Brick, 1974; Hill, 1975). It has been suggested

that in tropical estuaries the period of peak spawning generally coincides with periods of

high nutrient input associated with monsoonal or cyclonic rainfall (Heasman, 1980). As

explained before, maturation is however a long process and can be shifted in time from

spawning considerably. Moreover, no accurate information exist on the duration of gonad

maturation and spawning migration and if gonad maturation mainly takes place inshore or

offshore. For species that spawn offshore and only recruit back into the estuaries as

juveniles, spawning might also be rather timed to maximise food abundance for juveniles

(Poovichiranon, 1992). The rainy season (May - October) includes months, which fall

within (May - July) and outside (August - October) the “high spawning success season” as

observed in this study, and therefore no differences were found.

- Temperature-based season

As could be expected from the positive relationship between the average monthly

temperature and reproductive efficiency, reproductive activities were higher in the warmer

spring-summer season (March - August) compared to the cooler autumn-winter season

(September - February). Actually, the peak spawning season overlaps completely with the

warm season.

A similar pattern for both spawning and maturation was observed for S. serrata in

Australia (Mann et al., 1999a). In this study summer coincided with a peak in spawning and

hence females were on average more mature. A peak of ovarian development recorded in

autumn was attributed to female crabs having undergone the maturity moult and gonadal

development during the warmer summer months (Mann et al., 1999a). Also Heasman et al.

(1985) observed a spawning peak in summer for South-African S. serrata. In autumn

spawning activity rapidly decreased and by mid-autumn spawning totally stopped. Heasman

(1980) determined that female S. serrata can over-winter in advanced states of ovarian

development. These females then apparently contribute to the early rise in spawning activity

in spring (Mann et al., 1999a).

It seems the spawning season of S. paramamosain is to a large extent related to the

average monthly temperature, while the rainy season is probably the environmental cue for

the onset of maturation.


50

In conclusion of seasonal effects, it should also be noted that rearing conditions in

captivity might differ considerably from natural conditions (e.g. feed abundance and quality,

salinity, light conditions) and therefore reproductive success in captivity should not

necessarily correspond to the natural reproductive season.

Female weight

Larger crabs were observed to have a higher gonado-somatic-index (ovary weight /

total live weight) (Quinn and Kojis, 1987) and higher total egg fecundity (Churchill, 2003).

However, the total number of viable Z1 larvae produced does not always positively

correlate with the total number of eggs produced by a female since other factors (e.g. mating

success, sperm quality) can affect the egg fertilization and hatching rate. In our study,

fertilization rate increased significantly by the increasing female weight in the range of 236

- 415 g. This was however not the case for larger females (425 - 552 g). As no male

breeders were kept in this study, a larger proportion of the smaller females had probably not

yet mated in the wild prior to acquisition. Logically, relative Z1 fecundity positively

correlated with the fertilization rate. Big females (over 500 g) had the highest spawning

success, but relative Z1 fecundities were considerably lower. Due to the large variation

between individual females, most criteria were not significantly different between the 3

weight classes however. Overall, females of the medium weight class (300 - 500 g) tended

to have the highest reproductive efficiency, fertilization rate and total number of Z1

produced, and thus should be preferred as broodstock.

Churchill (2003) also recommended medium-sized S. serrata for optimal reproductive

output. This author moreover found a significant negative correlation between crab size and

the eicosapentaenoic acid (EPA) content of the eggs, which in turn was correlated with

larval quality as measured through stress tests. The author suggested that the optimal size

for S. serrata broodstock females should range between 125 - 145 mm carapace width in

order to assure maximum output in terms of quantity and quality of the produced eggs.

Generally, in the favourable weight class (300 - 500 g), females of about 400 g are

considered to produce the highest quantity of viable Z1 based on the above-mentiond

correlations: (i) the fertilization rate increased significantly by increasing female weight in

the range of 236 - 415 g, (ii) relative Z1 fecundity positively correlated with the fertilization

rate and (iii) larger crabs had higher total egg fecundity.


51

Time to spawn

The number of females that spawn seemed to decrease with time in captivity. The

majority of the crabs that spawn did so within the first month (65 %). Only 35 % of the

spawning events took place in the second month of captivity. Given the low price of crab

breeders at the moment, it might therefore be uneconomical to maintain breeders for more

then a month under the given conditions. Egg fertilization was not affected by time to

spawn. This confirms that females preserve sperm well regardless long periods (up to 60

days) in captive conditions without decrease in the fertilization rate. It also shows that male

crabs are only required if sub-adult females would be used for broodstock since mating

occurs at the maturity moult and sperm is subsequently stored for long periods by the

females (Du Plessis, 1971).

The overall spawning success in this study (233 out of 786 females = 30 %), was in

between those reported by Hai et al. (2001) (25 %) and (Dat, 1999b) (52 %) for S.

paramamosain. In this study, the highest spawning success (69 %) was obtained in the

earthen pond. This is however considerably lower than the 85 % reported for S. serrata by

Mann et al. (1999a). A universal phenomenon in Scylla populations appears to be offshore

migration of females to spawn (Le Vay, 2001) and hatch their larvae in full strength

seawater. Keenan et al. (1998) argued that S. serrata populations are dominant in oceans

with water salinity above 34 g l-1 and in mangroves that are inundated with high salinity

water for most of the year. For this species, in some areas, conditions suitable for larval

development may occur in inshore coastal areas (Heasman et al., 1985). Conversely, S.

paramamosain populations are more abundant in seas where the salinity is generally below

33 g l-1, and are able to colonize estuarine habitats in which periods of low salinity occur

seasonally (Keenan et al., 1998). It therefore seems possible that under these conditions, S.

serrata already mature inshore, whereas for S. paramamosain females, this only happens

during/after migration. In this respect, it was recognized in this study that female crabs

caught from Central Vietnam (where inshore salinities are normally higher) or netted

offshore by fishermen in the Mekong Delta usually showed higher spawning success and

hatching success compared to the animals that were captured near the coast.


52

4.2. Artificial incubation of eggs and egg diameter in function of incubation time

Females kept in tanks without substrate often shed their eggs upon spawning. Previous

observations showed that these shedded eggs are generally fertilized and are able to hatch

when incubated artificially in glass cones. Reliable artificial incubation would be a very

useful tool for scientific purposes and could, potentially, also be helpful to further

standardise and control commercial rearing practices. Fungal infection and eggs adhering to

the sides of the incubation containers have so far made this practice impractical however

(Hai et al., 2001; Churchill, 2003). The higher hatching percentage observed in this study

when formalin was added daily, confirms that infestation with parasites and/or microbial

interaction are a key problem in artificial incubation. Bath treatment with 25 µl l-1 formalin

was found to prevent the occurrence of fungal infection on eggs and newly-hatched crab

larvae (Hamasaki and Hatai, 1993a, 1993b).

In contrast to non-viable eggs, the diameter of viable eggs increased steadily during

incubation. Based on this increase, it would be possible to distinguish and reject non-viable

eggs from day 3 onwards. As egg size between batches is subjected to considerable

variation it may however be more practical to postpone the decision until day 5 after hatch

when egg size can be more precisely determined and combined with the appearance of

eyespots in the eggs. The best fitting equation (r2 = 0.96, P < 0.01) to describe the diameter

of viable eggs (Y, in µm) in function of the incubation time (x, in days) was y = 286 + 3 x +

x2. For S. serrata, Churchill (2003) found the same relationship to fit the polynominal

equation y = 299 + 2 x + x2.

5. Conclusions and suggestions

Although more difficult to manage, pond systems resulted in the best reproductive

performance. Tank systems could be a more practical alternative if stocking densities are

kept low and a proper substrate is provided. It was noticed that darkening the broodstock

tanks (as is often done) was not necessary provided shelters for hiding are available.

Broodstock collected from an inshore region where the salinity level is higher and

more stable, tended to perform better than those collected from a region with lower and

variable salinity.


53

Females of approximately 400 (300 – 500) g produced the highest total number of

viable Z1 and are therefore preferred as breeders. Smaller females were sometimes not

fertilized, while big females (over 500 g) tended to have lower relative Z1 fecundities.

Although detached eggs could be incubated artificially, egg incubation by the females

themselves was the best practice.

The peak spawning season of S. paramamosain in South Vietnam seems to be from

March to July, which would be the easiest period to obtain spawners and hence perform

larval rearing. Gonad maturation in the wild probably takes place from September to

February. With eyestalk ablation, the optimal production period for mud crab could be

extended from February to August.

Although egg quality remained unchanged within 60 days in captivity, given the low

price of crab breeders and the decreasing spawning activity, it is recommended to keep the

broodstock for not more than 30 days.

In conclusion, year-round maturation and spawning of wild breeders of S.

paramamosain for research and pilot production can easily be achieved, especially when

uni-lateral eyestalk ablation is applied. Although individual females achieved high

fecundities and fertilization rates, overall reproductive efficiency (i.e. the percentage of

females that hatched viable larvae) was rather low (maximum 31 % in ponds). A possible

reason might be that the source of the broodstock was restricted to inshore regions.

Broodstock from offshore water might be more mature and therefore more efficient.

In order to avoid further pressure on this valuable resource the use of wild breeders

should however be discouraged and research efforts should be directed towards full

domestication of the species. In this respect, further research is warranted on all aspects of

maturation and fertilization in captive conditions, investigating dietary requirements and

suitable rearing conditions to trigger maturation and spawning in a more natural way.

Acknowledgements

This study was supported by the European Commission (INCO-DC), the Flemish

Inter-University Council (Vl.I.R.-IUC) and the International Foundation for Science (IFS).


54

Table 1 Average temperature [mean ± standard deviation (number of records)] of rearing water in the broodstock tanks grouped by month and seasons

Group Month/ Season Temperature (°C)

Jan 27.6±1.2f E (188) Feb 28.2±0.9e D (143) Mar 28.7±1.1bcd BCD (186) Apr 29.3±1.0a A (165) May 29.0±1.2ab AB (137) Jun 28.5±1.1cde CD(131) Jul 28.5±1.4cde CD (152) Aug 28.8±1.0bc ABC (135) Sep 28.7±1.0bcd BCD (142) Oct 28.5±1.0cde BCD (152) Nov 28.3±1.0de CD (144)

Month

Dec 27.2±1.5g E (158) Rainy season (May-Oct) 28.7±1.1A (850) Monsoon season Dry season (Nov-Apr) 28.2±1.3B (983) Spring-Summer (Mar-Aug) 28.8±1.2A (906) Temperature-based season Autumn-Winter (Sep-Feb) 28.1±1.2B (927)

Values in the same column for each group followed by the same superscript letter are not statistically different (P ≥ 0.05, regular letters and P ≥ 0.01, capital letters).

Table 2 Number (and percentage ablated over non-ablated) broodstock crabs purchased from East South Vietnam (high-salinity region) and the Mekong Delta (low-salinity region) per year and per month

Females from high-salinity region Female from low-salinity region Stocking

month 1996 1997 1998 1999 2000 2001 1999 2000 2001 2002 Total n

Jan 10 (50) 12 (50) 10 (50) 13 (46) 45 Feb 2 (50) 10 (50) 20 (60) 10 (40) 8 (63) 50 Mar 1 (100) 4 (0) 11 (36) 10 (50) 26 (39) 10 (50) 19 (68) 81 Apr 4 (45) 9 (45) 10 (50) 18 (56) 10 (50) 51 May 3 (33) 2 (100) 15 (33) 10 (50) 10 (60) 20 (45) 10 (50) 70 Jun 5 (20) 5 (100) 13 (62) 14 (43) 10 (50) 3 (0) 18 (61) 10 (50) 18 (100) 96 Jul 15 (47) 12 (50) 20 (50) 12 (50) 59 Aug 3 (67) 13 (46) 7 (43) 4 (50) 20 (50) 10 (50) 57 Sep 3 (100) 4 (50) 12 (58) 10 (50) 6 (0) 20 (50) 10 (50) 65 Oct 2 (100) 10 (80) 10 (60) 15 (100) 10 (50) 20 (50) 10 (50) 77 Nov 2 (100) 10 (50) 9 (67) 20 (50) 10 (50) 51 Dec 3 (100) 12 (42) 9 (100) 7 (29) 20 (45) 10 (50) 61 All groups 8 (100) 11 (36) 34 (64) 120 (48) 50 (72) 116 (48) 32 (44) 222 (50) 125 (45) 45 (77) 763


55

Table 3 Effects of the eyestalk ablation, tank rearing system and broodstock source on reproductive performance [mean ± standard deviation or percentage (number of observations)]

Factor Eyestalk ablation Tank system

Broodstock source

Parameter Intact Ablated Plastic Cement High salinity

Low salinity

Time to spawn (days)

28±17a (70)

27±14a

(140) 26±13a

(22) 25±13a (42)

27±17a (30)

29±16a

(21)

Ablation-spawn time (days) none 19±13

(140) 18±13a (16)

19±13a (30)

22±15a (18)

24±15a (14)

Survival time in captivity (days)

57±36a (92)§

60±32a (90)§

44±32a

(40) 37±24a

(42) n.d. n.d.

Spawning success (%)

20B (358)

35A (405)

16B (133)

30A

(139) 26a (116)

17a (125)

Hatching success (%)

30a (70)

33a (140)

5a (22)

19a (42)

50a (30)

38a (21)

Reproductive efficiency (%)

6B

(358) 11A

(405) 1b

(133) 6a

(139) 13a

(116) 6a

(125)

Fertilization rate (%)

56±28a

(15) 63±26a (20) n.d. n.d. n.d. n.d.

Egg diameter (µm)

286±8a

(38) 288±7a

(60) n.d. n.d. n.d. n.d.

Relative Z1 fecundity (103 Z1 g-1)

3.0±1.8a

(12) 2.5±1.6a

(33) n.d. n.d. n.d. n.d.

Values in the same row for each factor followed by the same superscript letter are not statistically different (P ≥ 0.05, regular letters or P ≥ 0.01, capital letters). § Including data of 23 females that spawned later than 60 days since stocking. N.d. = not determined. Table 4 Reproductive performance [% (number of observations)] of broodstock reared in an earthen pond or cement tanks

Reproductive performance (%) Pond Tank

Spawning success (number of females spawned / total) 69A (61) 23B (56) Number of females not spawning / total 6B (61) 65A (56) Number of dead or missing females / total 25a (61) 12a (56)

Hatching success (number of females hatching larvae / females spawning) 45a (42) 64a (13) Number of females shedding eggs / females spawning 17a (42) 36a (13) Number of females producing non-viable eggs / females spawning 38a (42) 0b (13) Reproductive efficiency (%) (= Spawning success × Hatching success) 31a (61) 15b (56) Values in the same row followed by the same superscript letter are not statistically different (P ≥ 0.05, regular letters or P ≥ 0.01, capital letters).


56

Table 5

stocking month on the reproductive performance of the split intact and ablated groups [mean ± Effect ofstandard deviation or percentage (number of observations)]

Time to spawn (days) Spawning success (%) Hatching success (%)

StockiIntac Intact Inta

(4) ) BC (23) 2)

ng month t Ablated Ablated ct Ablated Jan 23±17a 37±16a (7 17abcd A 31ab AB (2 0 (4) 43a (7) Feb 28±21a (2) 31±13a (12)

12)

) )

4) 3)

) (1) ) (1)

9bcd BC (23) 44a AB (27) 0 (2) 50a (12)Mar 34±12a (12) 29±11a (14) 28ab AB (43) 37ab AB (38) 50a A ( 36a (14) Apr 16±12a (6) 26±14a (12) 23abc ABC (26) 48a AB (25) 33ab A (6) 42a (12) May 23±18a (12) 22±11a (11) 32ab AB (37) 33ab AB (33) 42ab A (12) 18a (11) Jun 28±18a (15) 22±14a (23) 41a A (37) 39ab AB (59) 7b A (15) 43a (23) Jul 28±15a (9) 21±11a (15) 30ab AB (30 52a A (29) 56a A (9) 21a (15) Aug 4±0 (2) 29±13a (9) 7cd BC (29) 32ab AB (28 100a A (2) 44a (9) Sep 39±12a ( 36±18a (10) 12bcd ABC (3 31ab AB (32) 0 (4) 30a (10)Oct 33±20a (3) 28±11a (11) 12bcd ABC (26) 22b B (51) 0 (3) 27a (11) Nov / 19±13a (10) 0 (23) 36ab AB (28 / 20a (10) Dec 60 37±12a (6) 4d C (28 18b B (33) 0 0 (6) Values in the lumn foll me letter are y di (P ≥ 0.05, tters

Table stocking month on the reproductive performance of breeders (all females) [mean ± standard deviation

Stocki month Ablation-spawn Spawning )

Hatching )

Reproductive

same co owed by the sa superscript not statisticall fferent regular leand P ≥ 0.01, capital letters).

6 Effect ofor percentage (number of observations)]

Time to spawn ng (days) time (days) success (% success (% efficiency (%)

Jan 32±17a (11) 27±16a (7) 24ab AB (45) 27a (11) 7a (45) Feb 31±14a (14) 21±13a (12)

lation with (r2)

28ab AB (50) 43a (14) 12a (50) Mar 31±12a (26) 20±10a (14) 32ab AB (81) 42a (26) 14a (81) Apr 22±14a (18) 15±11a (12) 35ab AB (51) 39a (18) 14a (51) May 23±15a (23) 14±8a (11)

33ab AB (70) 30a (23) 10a (70)

Jun 25±16a (38) 14±10a (23) 40ab AB (96) 29a (38) 11a (96) Jul 23±13a (24) 15±12a (15) 41a A (59)

) 33a (24) 14a (59)

Aug 24±16a (11) 19±16a (9)

19bc AB (57 55a (11) 11a (57) Sep 36±16a (14) 20±16a (10) 22bc AB (65) 21a (14) 5a (65) Oct 29±12a (14) 23±10a (11) 18bc B (77) 21a (14) 4a (77) Nov 19±13a (10) 15±11a (10) 20bc AB (51) 20a (10) 4a (51) Dec 40±14a (7) 34±11a (6) 11c B (61) 0 (7) 0 (61) Corremonth temperature (-)0.35* (-)0.64** 0.25 0.37* 0.37*

Values in the same column followed by the sa pt letter are not statisticall nt (P ≥ 0.05, regular letters me superscri y differeand P ≥ 0.01, capital letters). * and ** = significant correlation (P < 0.05 and P < 0.01, respectively).


57

Table 7

f seasons on the reproductive performance of breeders [mean ± standard deviation or percentage

Monsoon season

Effects o(number of observations)]

Temperature-based season

Reproductive characteristic Rainy ct) Apr)

Sprin r

Time to spawn (days) ) ) (May-O

Dry (Nov-

g-Summer Autumn-Winte(Mar-Aug) (Sep-Feb)

26±15a (124 29±14a (86) 25±14b (140 31±15a (70)Ablation-spawn time (days)

(%) 2) ) )

17±12a (79) 20±13a (61) 16±11B (84) 22±14A (56) Spawning success (%) 29a (424) 25a (339) 34A (414) 20B (349) Hatching success (%)

y31a (124) 34a (86) 36a (140) 24a (70)

Reproductive efficienc 9a (424) 3)

9a (339) 12A (414) 5B (349) Fertilization rate (%)

61±29a (2 56±25a (1 62±28a (26 56±25a (9

Relative Z1 fecundity (103 Z1 g-1) 2.4±1.4a (24) 3.0±.19a (21) 2.9±1.6a (32) 2.0±1.7a (13)

Values in the same row of each seas ed b rscri stati (P ≥

Table

female weight on reproductive performance of breeders [mean ± standard deviation or percentage

Hatching Reproductive Fertilization Total number Relative Z1

on group follow y the same supe pt letter are not stically different0.05, regular letters or P ≥ 0.01, capital letters).

8 Effect of(number of observations)]

Weight Spawning class (g)

success (%)

success (%)

efficiency (%)

rate (%)

of Z1 (103 Z1)

fecundity ) (103 Z1 g-1

[100-300[ ) 2a (6) (6) 23a (167 31a (39) 7a (167) 34±2 770±717a 2.9±2.7a (6) [300-500[ 30a (538) 31a (163) 9a 163) 65±24a (24) 975±605a (36) 2.7±1.5a (36)[500-700[ 41a (56) 17a (23) 7a (23) 59±36a (4) 968±727a (3) 1.8±1.3a (3) All groups 30 (761) 30 (225) 9 (761) 59±27 (34) 947±616 (45) 2.6±1.7 (45) Values in the sam ollow ame pt lette tica

Table on between a number of reproductive characteristics and female weight (y = β + αx) [mean ± standard

Details of y Details of x r2 n

e column f ed by the s superscri r are not statis lly different (P ≥ 0.05).

9 Correlatideviation (min-max)]. n = number of observations

Correlation y x

A Egg diameter male weight ) .01 (µm)

Fe(g)

288±10 (258-312)

390±97 (228-700 (-)0 78

B ) 0.52** 24

ization rate ale weight

ndity

** 21

Fertilization rate (%)

Female weight (g)

62±24 (12-93)

334±54 (236-415

C Fertil(%)

ization rate

Fem(g)

e to spawn

54±33 (10-93)

475± 42 (425-552) 0.01 10

D Fertil(%)

ization rate

Tim(days)

e Z1 fecu

60±27 (10-95)

30±19 (2-74) 0.00 35

E Fertil(%)

Relativ(103 Z1 g-1)

69±23 (12-93)

2.0±1.2(1.8-4.1)

0.47

** = significant correlation (P < 0.01).


58

Table 10

eproductive performance of breeders in function of time in captivity before spawning (days) [mean ± eviation or percentage (number of observations)]

(days) event success (%) tion

rate (%) Egg diameter (µm)

Relative Z1 fecundity (103 Z1 g-1)

Rstandard d

Time to spawn % of spawning Hatching Fertiliza

0-15 30a (210) 32a (62) 58±32a (12) 286±7a (33) 2.4±1.2a (13) 16-30 35a (210)

3532a (74) 63±21a (11)

60±28286±7a

(32)

288±73.2±1.8a (18) 2.1±1.331-60 a (210) 31a (74) a (12) a (33) a (14)

Values in the same owed e superscript letter are not stat nt (

Table 11 of eggs during incubation (µm) (mean ± standard deviation). n = 42 and 65 batches of viable eggs iable eggs, respectively

Non-viable eggs

column foll by the sam istically differe P ≥ 0.05).

Diamet r and non-v

e

Day after hatch Viable eggs

0 287±10 a a288±9 1 291±11a 292±9a

2 295±11a 294±10296±10B

a 3 301±14A

297±10b4 309±15A 5 318±16A 298±11B 6 333±19A 302±18B

299±16B 7 347±19A 8 353±17A 308±9B 9 364±15A 312±1 B 4

/ 10 387±16 Values i e same y the ript letter are not statistically different (P ≥ 0.05, regular letters or P ≥ 0.01, ca letters)

n th row followed b same superscpital .


59

Figure 1. Broodstock collection sites in South Vietnam: in high-salinity region (Vung Tau) and low-salinity region (Vinh Chau). Experimental hatchery is located in Can Tho.

CHAPTER 4

Optimal feeding schedule for mud crab (Scylla paramamosain) larvae


1 College of Aquaculture and Fisheries, Can Tho University, Vietnam. Email: [email protected] 2 Laboratory of Aquaculture & Artemia Reference Center, Ghent University, Belgium. Email: [email protected]

Abstract The ability of the different zoeal stages to catch and consume Artemia nauplii was first

investigated in 20-ml vials containing individual larvae. Results showed that crab larvae could only start catching noticeable quantities of newly-hatched Artemia from the zoea 2 stage onwards. The ability of zoea 2 to catch newly-hatched Artemia was however variable between batches and individuals. From the zoea 2 stage onwards, the number of Artemia consumed increased with each larval stage. In a next step, two experiments were conducted to determine a suitable first feed for zoea 1 stage larvae of the mud crab Scylla paramamosain. Micro-algae, rotifers and artificial diets were compared as first feed. Of the 3 feeds tested, rotifers gave the best results. Micro-algae and artificial diets resulted in very low survival and growth. Although micro-algae were not the proper initial feed for early crab stages, they proved beneficial in improving the nutritional quality of rotifers, resulting in higher survival in later zoeal stages (zoea 4 and 5) and a more successful metamorphosis to the megalopa stage. In an attempt to simplify the feeding schedule, a series of experiments were carried out where rotifers were replaced by different forms of Artemia (live and heat-killed umbrella-stage Artemia and frozen or heat-killed Artemia nauplii). Live umbrella-stage Artemia were the best replacement for rotifers for feeding zoea 1 - zoea 2 larvae compared to other Artemia forms. The unselective feeding behaviour (especially at the early stages) seems promising to develop artificial diets in order to substitute live feed and to reduce the dependency on rotifers. In a last step, the optimal time to shift from rotifers to Artemia was investigated. Results showed that rotifers should be replaced by Artemia already in zoea 2 stage. A transition period to shift from one diet to another seemed not necessary. Prolonged feeding of rotifers beyond the zoea 2 stage tended to reduce survival and delay larval development. Although Artemia are more difficult to capture for zoea 2 larvae compared to rotifers, they probably enhanced crab larval performance due to their higher nutritional value compared to that of rotifers. The nutritional value of rotifers and Artemia is however not consistent and therefore optimal feeding schedules might depend on local facilities and culture zootechnics. From the zoea 3 stage onwards, crab larvae can ingest enriched Artemia meta-nauplii.

CHAPTER 4 – Feeding schedule

62

1. Introduction

Mud crab (Scylla spp.) fishing represents a valuable component of small-scaled

coastal fisheries in many countries in tropical and subtropical Asia. Since wild stocks are

reported to decrease in many countries (Angell, 1992; Keenan, 1999a), aquaculture of mud

crabs, is becoming increasingly popular. To date aquaculture already contributes a

considerable proportion to the world production of the genus (FA0, 1999). In Vietnam, the

mud crab Scylla paramamosain is the second most important cultured species (next to

shrimp) in the coastal zone.

Aquaculture operations however currently rely almost entirely on wild seed stock, for

which there has been a similar trend of increased exploitation in recent years (Angell, 1992;

Keenan, 1999a). It is therefore generally accepted that the main obstacle for the further

development of mud crab farming is the establishment of hatchery-techniques for controlled

production of seed (Keenan, 1999a; Liong, 1992; Mann et al., 2001; Rattanachote and

Dangwatanakul, 1992; Shelley and Field, 1999; Xuan, 2001).

In contrast to fish, there has been little study of the feeding processes of decapod

crustacean larvae (Harvey and Epifanio, 1997). First feeding (more specifically optimal prey

size, prey density, and its nutritional value) is however of utmost importance and forms the

basis for the development of successful hatchery techniques (Suprayudi et al., 2002a).

Mud crab zoeae are considered to be essentially zooplanktivorous and are able to feed

just minutes after hatch (Mann and Parlato, 1995). Most research to date has agreed that

rotifers (Brachionus spp.) and Artemia (Artemia spp.) are the most suitable feed for Scylla

larvae (Brick, 1974; Heasman and Fielder, 1983; Marichamy and Rajapackiam, 1992;

Zainoddin, 1992; Zeng and Li, 1992). These preys are however often used in combination,

and therefore the optimal live feed species or size for each larval stage has not been clearly

determined yet.

Differences in the quality of the live feed and the rearing systems and zootechnics

applied, also often lead to contradictory results. Quinitio et al. (2001) reported that for S.

serrata the best feeding sequence is to feed rotifers throughout the zoeal stages, to introduce

artificial diets for shrimp at the late zoea 1 stage and finally to introduce Artemia from zoea

3. Ruscoe et al. (2004) on the other hand found that rotifers are best replaced by Artemia

already in the zoea 2 stage. In this study, co-feeding of rotifers and Artemia resulted in

poorer performance. The recent re-classification of the genus Scylla into four separate


63

species (Keenan et al., 1998; Keenan, 1999b) moreover makes it necessary to investigate

differences among the different species. Suprayudi et al. (2002a) in this respect remarked

that the feeding preference of S. serrata seems to be slightly different from that of S.

paramamosain.

From the start of our study, it was obvious that rotifers could sustain a certain survival

rate of crab larvae to the megalopa stage. However, newly-hatched Artemia are preferred to

rotifers since the former live feed are commercially available and readily produced, while

rotifers require a laborious and sometimes unpredictable culture. Therefore, in commercial

applications, rotifers are replaced by newly-hatched Artemia as soon as possible. In this

respect, the earliest stage then crab larvae could capture newly-hatched Artemia was

investigated. From the experience observed in larviculture of Penaeid shrimp, the possibility

to replace rotifers as first feed for crab larvae with micro-bound diets and/or micro-algae

was tried since these feeds are commercially available and cultured more easily than

rotifers. In an other attempt not to rely entirely on rotifers in the early stages, different

alternative inert forms of Artemia were tested. Finally, feeding schedules were investigated

in order to find out the best time for shifting from rotifer to Artemia feeding.


2.1. Broodstock rearing

Fully gravid crabs were bought from local markets and transported to the hatchery.

Prior to stocking in the hatchery, the crabs were bathed in a 100 µl l-1 formalin solution for 1

hour. The crabs were housed individually in 100-l compartments of a roofed 2 × 2 × 0.5 m

cement tank, equipped with a biofilter. Rearing water (30 ± 1 g l-1) was diluted from brine

(90 - 110 g l-1) with tap water and chlorinated before use. Water temperature was not

controlled, but fluctuated slightly around 28 °C. Each crab was fed a daily ration of 10 - 15

g of fresh marine squid, bivalve or shrimp meat alternately.

After 3 - 5 days of acclimation, unilateral eyestalk ablation was applied to induce

spawning. After spawning, berried crabs were again bathed in a 100 µl l-1 formalin solution

for 1 hour and transferred to a 70-l plastic tank connected to a biofilter for incubation. Daily


bottom and controlling temperature (30 °C), salinity (30 g l-1) and ammonia and nitrite


64

levels. Every other day, the crab was bathed in a 50 µl l-1 formalin solution for 1 hour to

reduce or prevent infestation of the eggs with fungi and bacteria. During egg incubation, the

crabs were not fed.

One to two days prior to hatching, the female was moved to a 500-l fibreglass tank.

When the hatching process was completed, larvae were selected based on their photo-tactic

behavior: aeration in the hatching tank was turned off for several minutes and the larvae that

were actively swimming up to the surface were collected by gently scooping them from the

surface.

The larvae were then transferred to the rearing containers. In order to slowly acclimate

the larvae to the new rearing conditions, they were placed in a 50-l plastic mesh bucket and

slowly rinsed with water from the larval rearing containers for 20 to 30 minutes, before

releasing them.

2.2. Live feed culture and enrichment

Start cultures of the micro-algae Chaetoceros calcitrans and Chlorella vulgaris were

cultured indoor with Walne solution in seawater of 30 g l-1 at 25 °C. Large-scale production

was performed indoor in 500-l tanks under a transparent roof.

The same rotifer strain, Brachionus plicatilis L-strain with lorica length and width of

164 ± 22 and 120 ± 22 µm, respectively, was used in all experiments. Rotifers were cultured

indoor in 100-l fiberglass tanks operated in batch mode, following the procedure described

in Sorgeloos and Lavens (1996). Rotifers were initially grown on baker yeast, but one week

before use as feed for the larvae, the yeast was replaced by Culture Selco® (INVE

Aquaculture, Belgium). Temperature and salinity were controlled at 25 °C and 25 g l-1,

respectively. Rotifers were harvested on a 60 µm screen and rinsed. In some experiments

the rotifers were enriched with micro-algae or artificial enrichment media before being fed

to the crab larvae. Enrichment with Chlorella or Chaetoceros was performed at a density of

5 106 cells ml-1 for 3 hours (Dhert, 1996). A hemocytometer was used to count micro-algal

densities. In some experiments rotifers were also enriched with Dry Immune Selco® (DIS,

INVE Aquaculture, Belgium), using two separate doses of 0.05 g l-1 at a 3-hour interval.

Rotifer enrichment was performed at a density of 500 rotifers ml-1. The water in the

enrichment vessel was slowly heated to 29 - 30 °C to avoid exposing the rotifers to thermal

shock when they were added to the larval rearing tanks. Before being fed to the larvae,


65

enriched rotifers were rinsed and re-suspended in clean seawater at the same temperature of

the crab rearing tanks. Whenever the crab larvae were fed enriched rotifers, algae- or DIS-

enriched rotifers were used on alternate days.

Artemia nauplii (Vinh Chau strain) were hatched as described by Van Stappen (1996).

Both newly-hatched and enriched Artemia nauplii were used in the experiments of this

study. The nauplii were either enriched with Chaetoceros (maintained at 5 106 cells ml-1) for

12 hours or with DIS® (using two separate doses of 0.3 g l-1 at a 6-hour interval). In

experiment 6 only, Artemia were enriched with A1 Selco® (using 1 dose of 0.3 g added to 2

g cysts l-1 for 30 hours upon starting of cyst incubation). The temperature and salinity were

maintained at 30 °C and 30 g l-1, respectively during Artemia enrichment. The density of

Artemia during enrichment was 200 ml-1. Before feeding to the crab larvae, the Artemia

were rinsed with disinfected seawater and suspended at a known density in seawater.

2.3. Larval rearing: objectives, experimental design and techniques

The experimental conditions of the different experiments are summarized in Table 1.

An overview of the different feeding schedules used in experiment 2 to 8 is presented in

Table 2.

Experiment 1

From the very first experiments, rotifers proved to be able to sustain a certain survival

rate of crab larvae to the megalopa stage. However, rotifer culture is labor consuming and

often not consistent. Therefore, most hatchery owners prefer to replace rotifers by Artemia

that are commercially available as soon as possible. In this respect, experiment 1 was

planned to investigate from what stage crab larvae could capture newly-hatched Artemia. At

the same time, the number of prey consumed per larva was quantified. This test was

conducted in small rearing vials where individual crab larvae and a determined number of

Artemia were cultured together for a certain period of time as described by Zeng (1998).

Prior to the actual experiment, the experimental animals were reared together in 30-l

cylindro-conical fibreglass tanks in a recirculating system under standard rearing conditions.

Rotifers were fed to the crab larvae from DAH 0 - 6. Artemia meta-nauplii were supplied

from DAH 6 onwards. Rotifers and Artemia were alternately enriched with Chaetoceros or


66

DIS®. Rotifers and Artemia were added daily at 45 ml-1 and 5 - 10 ml-1 to the rearing tank,

respectively.

At each larval stage, 10 larvae were sampled from the stock tank and individually

housed in 20-ml glass vials stocked with 200 newly-hatched Artemia nauplii. Each time, 10

control vials, containing the same number of Artemia nauplii but without crab larva were

incubated as well. Before the start of the test, the selected larvae were kept in a beaker with

slight aeration and starved for 3 hours. The glass vials were incubated in a waterbath

controlled at 29 °C.

After 24 hours, the water in the vials was poured into a petri dish and examined by

means of a dissecting microscope. Those vials that were found to contain weak or dead crab

larvae or crab exuvia were eliminated. The animals in the remaining vials were killed with

a few drops of lugol and the remaining number of Artemia in each vial were counted and the

crab larvae staged. The difference between the number of Artemia in the experimental and

control vials was considered to be the number of prey consumed by the crab larvae.

This test was repeated in three times, each with a different batch of larvae, making

sure that in total at least 20 measurements for each larval stage were determined.

Experiment 2

From experience in larviculture of Penaeid shrimp, the possibility to replace rotifers as

first feed for crab larvae by micro-bound diets and/or micro-algae was tried since these

feeds are commercially available and cultured more easily than rotifers. For this purpose, in

the second experiment four different feeds, i.e. (i) a commercial micro-bound diet for

shrimp, (ii) micro-algae, (iii) rotifers and (iv) a combination of micro-algae and rotifers,

were compared as first feed. A starvation treatment was included as a control.

The crab larvae were reared at a density of 40 l-1 in 2-l glass bottles operated in batch

mode. Rather strong aeration was provided to reduce the settlement of the micro-bound diet

and micro-algae. Daily, larvae were siphoned out using a large-tip pipette and gently

transferred to new bottles containing fresh seawater of the same temperature and salinity.

Lanzy PZ® (a commercial diet for shrimp zoeal stages; INVE Aquaculture, Belgium)

was fed in 4 rations to give a total ration of 1 mg l-1 day-1 according to the manufacturer’s

recommendations. Micro-algae Chaetoceros were counted with a hemocytometer and added

to give a final concentration of approximately 150,000 ± 50,000 cells ml-1 in the treatments


67

“micro-algae” and “rotifers and micro-algae”. The density of rotifers was 30 and 15 ml-1 in

the “rotifers only” and “rotifers and micro-algae” treatments, respectively. Rotifers were not

enriched.

Experiment 3

Experiment 3 was a replication of the previous experiment. This time formalin was

used as a prophylactic chemical to increase overall larval survival. In this experiment, four

treatments from experiment 2, i.e. (i) starvation, (ii) micro-algae, (iii) rotifers and (iv)

micro-algae and rotifers were repeated. On DAH 9 and again on DAH 12, the surviving

larvae were pooled per treatment and the water volume adjusted to obtain similar larval

densities (20 l-1) for all treatments. On DAH 12, the treatment “rotifers and micro-algae”

was split up into 2 sub-treatments, i.e. one part continued to receive rotifers; the other half

was fed Artemia from then onwards. Micro-algae were supplemented in both sub-

treatments.

Rearing conditions were similar to those in experiment 2. In this experiment rotifers

were however supplied at a higher density (45 ml-1 compared to 15 - 30 ml-1 in experiment

2). Also 20 µl l-1 formalin was applied every other day to reduce or prevent fungi and

bacteria to develop.

Experiment 4

Through previous experiments, rotifers proved the most suitable live feed for early

crab larvae. However in practice, availability of rotifer stocks is sometimes problematic for

large-scale production as. Therefore, in an attempt not to rely entirely on rotifers in the early

stages, different alternative inert forms of Artemia were tested. Five treatments were

compared: (i) starvation, as negative control, (ii) rotifers, as positive control, (iii) heat-killed

umbrella-stage Artemia, (iv) heat-killed instar-1 Artemia, (v) live instar-1 Artemia and (vi)

frozen instar-1 Artemia.

Static 1-l PVC cones were used in this experiment. Larvae were stocked at a density

of 50 l-1. As in experiments 2 and 3, the crab larvae were manually transferred to new cones

daily. Rearing conditions were also similar. Aeration was however increased to maintain the

non-moving feed items in suspension. Newly-hatched Artemia and umbrella-stage Artemia


68

were heat-killed by dipping in water of 80 °C for 15 minutes. Instar-1 Artemia were put in a

-20 °C freezer for 2 hours to make “frozen Artemia”. Umbrella-stage Artemia were

produced by harvesting already after 12 hours incubation. The feeding density was identical

(10 ml-1) for all Artemia forms. In the positive control, 45 rotifers ml-1 were fed. Rotifers

and Artemia were not enriched in this experiment.

Experiment 5

Experiment 5 was the replication of experiment 4 but in larger cylindro-conical tanks

operated in recirculating mode in order to better keep the inert feed in suspension and to

increase overall survival. The same treatments as in the previous experiment were designed,

i.e. (i) starvation, as negative control, (ii) rotifers, as positive control, (iii) heat-killed

umbrella-stage Artemia, (iv) heat-killed instar-1 Artemia, (v) live instar-1 Artemia and (vi)

frozen instar-1 Artemia). Here the larvae were however reared in 30-l cylindro-conical

fibreglass tanks that were operated in recirculation mode. The stocking density of Z1 was

also higher than that used in experiment 4 (150 instead of 50 l-1 in the previous experiment).

General rearing techniques were similar to those described in experiment 1. In the

positive control treatment (feeding rotifers only), rotifers were enriched with Chlorella and

DIS® alternately and supplied to the larvae at a density of 45 ml-1. The different forms of

Artemia were prepared and fed (10 ml-1) to the crab larvae identically as in experiment 4.

Experiment 6

From experiments 4 and 5, heat-killed umbrella-stage Artemia shows some promise as

a (partial) replacement for rotifers. Live umbrella-stage Artemia, which stay in suspension

better (due to the alveolar layer of the cyst shell) and have a superior nurtritional quality,

might even be better. For that reason, in this experiment, live umbrella-stage Artemia were

tried as an alternative for rotifers as first feed. Live umbrella-stage Artemia were the only

feed used for the early crab larvae (Z1 - Z2) from DAH 0 - 6. From DAH 7, enriched (A1

Selco® and Chaetoceros alternately) Artemia nauplii were fed. The trial was conducted in

eight 500-l tanks and was operated in batch mode from DAH 0 - 6 and in recirculating mode

from DAH 7 onwards. It differed from the previous experiments in this way that the total


69

feed amount was fed in 2 - 3 rations (of approximately 2 umbrella-stage Artemia ml-1) per

day instead of the routine regimen using a higher density (5 - 10 Artemia ml-1) once per day.

Experiment 7

In experiment 7 the optimal time to shift from feeding rotifers to Artemia was more

precisely determined. Six feeding schedules, 1/1, 2/2, 3/3, 4/4, 5/5 and 6/4, differing in the

time when rotifers were replaced by Artemia were compared. Feeding schedule 1/1 implies

that rotifers were supplied on DAH 0 and flushed out on DAH 1 and Artemia was supplied

from DAH 1 onwards; feeding schedule 2/2: rotifers were supplied on DAH 0 and DAH 1

and flushed out on DAH 2 and Artemia was supplied from DAH 2 onwards, etc.

The larvae were reared in 30-l cylindro-conical fibreglass tanks that were operated in

recirculation mode. General rearing techniques were similar to those described in

experiment 1. Rotifers were not enriched and instar-1 Artemia were used.

Experiment 8

In this experiment, feeding schedules 6/4 (Artemia introduced in the Z2 stage), 9/7

(Artemia introduced in the Z3 stage) and 12/9 (Artemia introduced in the Z4 stage) were

compared. All three treatments had a 2-day overlap in feeding rotifers and Artemia, i.e.

feeding schedule 6/4 implies that rotifers were supplied from DAH 0 - 6 and Artemia were

supplied from DAH 4 onwards; feeding schedule 9/7: rotifers were supplied from DAH 0 -

9 and Artemia were supplied from DAH 7 onwards, etc.

The rearing containers, rearing techniques and feeding practices were similar to those

of experiment 7.

2.4. Evaluation criteria

In experiments 2, 3 and 4 (using small containers) the average survival rate was

calculated by individually counting all surviving larvae in each replicate. The survival rates

in the other experiments (using 30 to 500-l containers) were estimated by volumetric

sampling. Depending on the tank volume and the density of the surviving larvae, triplicate


70

300- to1000-ml samples were taken from each tank. Megalopae (M) were counted

individually.

Larval development was monitored every three days by identifying the average zoeal

instar stage of a sample of larvae and assigning it a value: first zoea (Z1) = 1; second zoea

(Z2) = 2, etc. To compare the larval development between treatments, an average larval

stage index (LSI) was calculated from the average LSI value of all replicate containers in

the same treatment. In experiments (1, 3 and 4), using small containers, all larvae were

staged visually upon counting daily survival. For larger containers (experiments 5, 6, 7, and

8), 5 or 10 larvae (in 30-l and 500-l tanks respectively) were sampled from each tank to

calculate the average LSI. For the latter method, the sampled larvae were staged under a

dissecting microscope.


One-way analyis of variance (ANOVA) was used to compare data. Homogeneity of


differences were detected between the variances, the data were submitted a one-way




survival rate) or logarithmic transformations (for other parameters) (Sokal and Rohlf, 1995).

The two-tailed Fisher exact test (modified from the contingency table method) was used to

compare ratios (expressed in percent) for the survival of pooled data. All data are presented

as means ± standard deviation when using the Tukey test or as ratio/percentage without

standard deviation when the Fisher exact test was used. P was set at both 0.05 and 0.01.

Whenever differences are significant at P < 0.01, this is also indicated. All analyses were

performed using the statistical program STATISTICA 6.0.

3. Results

Table 3 shows the Artemia consumption in the successive larval stages in experiment

1. In all three data sets, the number of remaining Artemia in the experimental vials is

significantly different from the controls from Z3 onwards (P < 0.01). This shows that crab


71

larvae can consume live Artemia nauplii from that stage. For the Z2 stage, results were less

clear. The number of Artemia remaining in tanks with Z2 larvae was always lower than the

number in controls but this was significant (P < 0.05) in only one of the experiments. These

results indicate that the ability of Z2 larvae to catch and ingest Artemia nauplii is still rather

limited and might be variable between different batches and individuals. Z1 seemed not

capable of consuming large numbers of live Artemia nauplii, indicating that they probably

require a different first feed. The average Artemia consumption in the different larval stages

is presented in Figure 1. From this it can be seen that consumption increased from

approximately 6 Artemia nauplii larva-1 24 h-1 in the Z2 stage to more then 100 in the

megalopa stage.

The average survival rates of the crab larvae fed artificial diets, algae, rotifers or a

combination of rotifers and algae in experiments 2 and 3 are presented in Table 4. In both

experiments, starved larvae could not survive beyond DAH 3. Larvae fed an artificial

shrimp (experiment 2) diet died gradually from DAH 3 and complete mortality occurred on

day DAH 5. In this treatment only a few larvae could moult to Z2. Also treatment “micro-

algae” resulted in very low survival on DAH 3 and 6 and the larvae could not moult to the

next stage, which shows that micro-algae are not a suitable first feed for S. paramamosain

larvae. The “rotifers only” treatment always result in the best survival. The survival rate of

the “rotifers only” treatment was generally also higher than the “algae and rotifers”

treatment (although only significantly different at P < 0.01 on DAH 3 in experiment 2). The

better performance of treatment “rotifers only” compared to treatment “algae and rotifers”

in experiment 2 could be explained by the lower rotifer density fed to the larvae in the latter

(15 compared to 30 ml-1 in treatment “rotifers only”). However, in experiment 3, both

treatments received the same amount of rotifers, and still there was quite lower survival

(although not significant) in treatment “rotifers and algae”. Therefore, supplementation of

micro-algae to a diet of rotifers does not seem to enhance the survival of early-stage crab

larvae. Table 5 shows the average survival rates at the later larval stages in experiment 3.

On DAH 9, treatment “rotifers only” performed still better than treatment “rotifers and

algae” (P < 0.05). After pooling and restocking the larvae on DAH 9 and again on DAH 12,

significantly better survival was however obtained in treatment “rotifers and algae” as

compared to “rotifers only” (P < 0.01). On DAH 15, feeding only rotifers resulted in a

significantltly lower (P < 0.01) survival compared to those of treatments with micro-algae

supplementation. Consequently, feeding only rotifers could not furthermore support


72

metamorphosis to the megalopa stage. The combination of rotifers and micro-algae on the

other hand yielded megalopae. A slightly higher metamorphosis rate was obtained at DAH

22 however when rotifers were replaced by Artemia on DAH 12.

In experiments 4 and 5, some alternative forms of Artemia were tested as a

replacement for rotifers to feed crab larvae in the Z1 stage. Table 6 summarizes survival and

larval development rate (expressed as LSI) of the larvae in the different treatments as well

as average water quality parameters. The results on DAH 3 and 6 confirm that rotifers are

the best live feed in terms of survival and larval development for Z1 and Z2 larvae (P <

0.01). Among the remaining treatments, treatment “heat-killed umbrella-stage Artemia”

tended to have the highest survival rates on DAH 3 and 6. Compared to the positive control

(rotifers), LSI was however compromised in all treatments. Live and frozen instar-1 Artemia

nauplii resulted in slightly better (not significant) growth compared to heat-killed Artemia

forms (instar-1 and umbrella-stage) on DAH 6 in experiment 5.

In experiment 6 a pilot-scale production trial (data not shown in the tables) was

performed using live umbrella-stage Artemia to feed Z1 and Z2 stages (DAH 0 - 6). These

umbrella-stage Artemia were fed in 2 or 3 rations of 2 - 3 individuals ml-1 day-1. The

survival rates on DAH 3, 6 and 9 were 79 ± 9, 36 ± 2 and 30 ± 5 %, respectively. The LSI

values on the same sampling days were 1.7 ± 0.1, 2.9 ± 0.2 and 3.9 ± 0.2, respectively,

representing a normal development rate.

In Tables 7 and 8, the survival and development rates are shown of larvae subjected to

different rotifer/Artemia feeding schedules (experiment 7 and 8, respectively). In

experiment 7, a higher survival rate on DAH 3 was observed for treatment 4/4 compared to

the other treatments (although only significantly different with treatment 4 at P < 0.05).

Where it is possible that treatments 1/1 and 2/2 gave low survival because of the early

introduction of Artemia; theoretically, the survival rates on DAH 3 of the last four

treatments (3/3, 4/4, 5/5 and 6/4) should be similar as the same live feed (i.e. rotifers) was

offered from DAH 0 - 3. On DAH 6 no difference were observed between any of the

treatments. On DAH 9 a significantly (P < 0.05) higher survival was found for treatment 4/4

compared to treatments 1/1 and 3/3. Treatments 2/2, 5/5 and 6/4 showed intermediate

results. This could indicate that rotifers should not be replaced by Artemia before DAH 3

(Z1 stage). It furthermore suggests that rotifers are best replaced by Artemia already early in

the Z2 stage and that an overlap in feeding rotifers and Artemia is not really necessary.

Although no statistical differences were found, LSI values of the larvae in treatments


73

receiving Artemia in the Z1 stage (treatments 1/1, 2/2 and 3/3) tended to be lower than those

in the other treatments where Artemia were offered only in the Z2 stage (treatments 3/3, 4/4,

5/5 and 6/4), which confirms the survival results.

In experiment 8 (Table 8), delaying the introduction of Artemia until the Z3 and Z4

stage (treatment 9/7 and 12/9, respectively) resulted in a significantly lower LSI value on

DAH 9 compared to early supplementation of Artemia from the Z2 stage (treatment 6/4) at

P < 0.01. The late Artemia feeding schedules also tended to reduce the survival and

development rate of crab larvae from DAH 9 - 15 although the data of all treatments shows

no significant difference.

4. Discussion

All publications to date have agreed that rotifers and Artemia are the most suitable live

food for Scylla larvae (Brick, 1974; Heasman and Fielder, 1983; Marichamy and

Rajapackiam, 1992; Zainoddin, 1992; Zeng, 1998). These preys are however often used in

combination, and there remains uncertainty about the optimal live feed species or size for

each larval stage.

From the results of experiment 1, it became clear that from the Z3 stage onwards, S.

paramamosain larvae can easily catch and consume Artemia nauplii. Although rotifers as a

sole feed could support crab larvae to reach the last zoeal stage, it might be beneficial to

supplement micro-algae of shift to a larger prey at a certain point in the rearing process. In

this respect it was shown in this study that micro-algae were necessary in combination with

rotifers in order for the larvae to successfully complete the first metamorphosis.

Live prey size seems however to be most crucial for early Z1 and Z2 larvae. Using

Artemia nauplii at start-feeding for S. paramamosain larvae usually results in low survival

(Li et al., 1999). S. serrata larvae can however be grown successfully on an exclusive diet

of Artemia, and the insignificant differences (compared to feeding rotifers) in larval survival

and growth when fed newly-hatched nauplii or even enriched meta-nauplii from the Z2

stage onwards indicate that live prey size within the range of 463 to 658 µm does not affect

larval performance of S. serrata (Mann et al., 2001).

As there are indications that feeding abilities might differ among the four Scylla

species, the best time to shift to Artemia still needs to be determined more precisely for


74

every species within the genus Scylla. The following discussion concentrates on the suitable

prey size for Z1-Z2 stages for S. paramamosain.

4.1. Ability of S. paramamosain zoeae to catch instar-1 Artemia

The ability of S. paramamosain zoeae to catch Artemia nauplii and the number of prey

consumed in each stage was checked by individually rearing larvae in small containers. The

results of experiment 1 showed that crab Z1 could not catch large numbers of newly-

hatched Artemia. This is probably due to the fast movement and relative big size of the

latter. Harvey and Epifanio (1997) estimated the swimming speed of rotifers, instar-1

Artemia and Z1 larvae of the mud crab Penopeus herbstii at 3, 6 and 2 mm s-1, respectively.

From this it is clear that rotifers are a much easier prey than Artemia for crab larvae. Zeng

and Li (1999) concluded that due to their relatively larger size and faster swimming

behaviour, Artemia as a sole diet for early S. paramamosain larvae generally yields low

survival rates compared to rotifers and it was observed that early larvae seem unable to

capture and ingest Artemia as effectively as rotifers.

When comparing the different species within the genus Scylla, there is evidence that

S. serrata Z1 can cope better with Artemia as first feed, resulting in higher survival

compared to S. paramamosain. For example, a survival rate of 48 % at the Z5 stage was

obtained by Baylon and Failaman (1999) when S. serrata zoeae were fed only Artemia. In

our trials with S. paramamosain, feeding only Artemia to Z1 larvae, resulted in extremely

low or zero survival. Table 9 compares the size of some popular live feeds, the carapace

width of the different larval stages of S. paramamosain and egg size of the four Scylla

species. The data of the diameter of newly-spawned eggs demonstrates a smaller egg size in

S. paramamosain (288 ± 10 µm) compared to S. serrata (329 ± 8 µm). When comparing all

four Scylla species, S. paramamosain seems to have the smallest newly-spawned eggs

among all. It is likely that the smaller egg diameter in S. paramamosain coincides with a

smaller Z1 size, which results in a lower capability to catch larger preys like Artemia,

making prey size even more crucial.

At the Z2 stage, few Artemia (7 larva-1 24 hours-1 on average) were consumed and the

capture ability seemed to vary from one batch to another and even between individual

larvae. In Table 9, it can be seen that the carapace width of Z2 was similar to the size of

newly-hatched Artemia resulting in difficulty for crab larvae to ingest the whole prey. That


75

S. paramamosain Z2 can survive under these conditions, can probably be explained by the

fact that Z2 are observed to consume only part (e.g. appendages) of its prey (Zeng and Li,

1999). Similarly for S. serrata, Baylon et al. (2004) observed that crab larvae hold the

Artemia nauplius for a while but eventually released them. The released prey had missing

body parts such as the head and appendages when they were examined under a microscope.

The authors suggested that although the early stage larvae could not consume the entire

Artemia nauplius, they managed to ingest bits and pieces of the prey.

From Z2 stage onwards, the number of consumed prey increased by each larval stage.

The average number of Artemia nauplii consumed were 7, 15, 25, 37 and 114 individuals

for Z2, Z3, Z4, Z5 and M, respectively. For comparison, Baylon et al. (2004) found for S.

serrata larvae ingestion rates increasing from approximately 2 Artemia nauplii larva-1 day-1

in the Z1 stage to almost 50 in the Z5 stage. The consumption of Artemia nauplii by the crab

larvae also seemed to be affected by their physiological status (e.g. moulting period) and by

the quality of the larvae. In this respect, it was observed that weak or

moulting/metamorphosing crab larvae consumed significantly less prey. In a comparable

experiment, Zeng (1998) observed that independent from prey density (ranging from 2 to 20

Artemia nauplii ml-1), larval feeding rate showed a similar disturbed pattern related to the

moulting cycle, which was characterized by a sharp decline in feeding close to moult or

metamorphosis to the megalopa stage and as soon as larvae completed metamorphosis, the

feeding rate immediately shot back to its maximum. Baylon et al. (2004) found that when S.

serrata larvae were cofed with Artemia and rotifers, there was an increased consumption of

Artemia nauplii on the day before moulting or metamorphosing and increased ingestion of

Brachionus on the day after larvae had developed to the next stage.

4.2. Suitable first feed

In experiments 2 and 3, Z1 larvae, starved at 29.5 - 30.6 °C (Table 1) could not

survive more than 3 days and could not moult to Z2. Similarly, Z1 larvae of S. serrata

starved for 48 to 72 hours (at temperatures from 27 to 29 °C) incurred high mortalities

(Lumasag and Quinitio, 1998). Also at lower temperature, newly-hatched zoeae which were

continuously starved were not able to moult to the second zoeal stage and the time to reach

100 % mortality was 5.5 and 5.9 days at 28 and 24 °C, respectively (Mann and Parlato,

1995). It therefore seems likely that the maximal starvation time for Scylla larvae, even at


76

low temperatures, is 6 days. This also means that, under our conditions, larvae that survived

beyond DAH 3 in the “feeding treatments” were capable to catch and ingest at least some of

the feed offered.

All larvae fed an artificial shrimp diet in experiment 2 died before DAH 6. Similarly,

Quinitio et al. (1999) found that artificial shrimp diets were not a suitable first feed for mud

crab larvae. In that study, the larvae could not survive to the Z3 stage and as a consequence

of water pollution and bacterial contamination, mortality was positively correlated with the

amount of artificial feed given.

In experiments 2 and 3, feeding only algae resulted in almost complete mortality of

the crab larvae before DAH 6. Davis (2003) reported that larvae probably merely ingest

micro-algae by chance when swallowing water. Brick (1974) also found that S. serrata

zoeae fed Chlorella did not actively ingest noticeable quantities of the algae. Although the

presence of micro-algae in the water prolonged the survival of Z1, they could not moult to

Z2 unless the diet was supplemented with zooplankton (Brick, 1974). Micro-algae therefore

are not a proper feed for early S. paramamosain larvae.

From experiments 2 and 3 it is clear that rotifers were the most suitable live feed for

Z1 compared to the other feeds tested here. Ruscoe et al. (2004) found that also for S.

serrata larvae, rotifers were necessary in the feeding regime for optimal growth and

survival. The size of Z1 larvae is much bigger compared to the size of rotifers (Table 9),

which facilitates capture and ingestion. Similarly, Baylon et al. (2004) reported that the

smaller size of Brachionus (220 - 240 µm) and their slow swimming movement makes them

easy prey for the early-stage zoeae of S. serrata to capture and to consume, compared with

the larger Artemia (460 - 500 µm).

The lower survival rate of treatment “rotifers and algae” compared to “rotifers only”

in experiment 2 could be explained by the lower density of rotifers offered in the former. In

this respect, Zeng and Li (1999) concluded that rotifers (Brachionus plicatilis) are a suitable

feed for early larvae (Z1 and Z2) of the mud crab S. serrata, but their density significantly

affected larval survival and development. Crustacean larvae are generally passive feeders

and do not pursue prey actively (Baylon et al., 2004). They only feed if there is a chance

encounter with the food organism (Heasman and Fielder, 1983). Harvey and Epifanio

(1997) reported an increased ingestion by common mud crab Panopeus herbstii larvae of

Brachionus and Artemia with increasing density. Minagawa and Murano (1993) found a


77

decrease in the survival of the red frog crab Ranina ranina with decreasing prey density

(0.05 -5.0 organisms ml-1).

In experiment 3, where the same number of rotifers were fed in both treatments,

treatment “rotifers only” still outperformed treatment “rotifers and algae” however. Also the

quality of the algae may have affected the results. The algae were often observed to die due

to the low light intensity (about 1,600 lux) in the rearing containers and in this way polluted

the water and created a greenish mucous biofilm on the sides and bottom of the rearing

containers that was observed to trap early crab zoeae. Chythanya and Savan (1999)

indicated that in some cases biofilm bacteria can also cause large scale infection and

mortality of fish and crustaceans. Also Mann (2001) noted that one pattern of early rapid

mortality, which occasionally occurs in crab hatcheries, seems associated with development

of a characteristic mucilaginous matrix on the bottom of the tank. However attempts to

isolate the causative bacteria and re-inoculate cultures have been unsuccessful. Removing

this layer daily however significantly improved the survival rates of crab larvae reared in 5-l

bowls (Williams et al., 1998, 1999a).

Towards the end of experiment 3 however, a significantly higher survival was

observed in the treatment “rotifers and algae” (79 % compared to 25 % in the treatment

“rotifers only” on DAH 12). During this period, the layer of dead algae was removed every

day instead of every few days, earlier in the experiment. On DAH 22, 48 % of megalopae

were obtained in treatment “rotifers and algae”, while the surviving larvae in treatment

“rotifers only” could not pass the first metamorphosis and started dying on DAH 21. A

positive effect of micro-algae on the survival and development of Scylla larvae in later

stages has been found in other studies. For S. serrata, the presence of Chlorella left zoeal

survival unaffected while stimulating production of megalopae (Brick, 1974). Several recent

studies pointed out the potential beneficial roles of micro-algae in aquaculture rearing

systems, such as maintaining the quality of live feed (Makridis and Olsen, 1999), stabilizing

water quality via either ammonia uptake or oxygen production and producing natural

antibiotics (Tseng et al., 1991). The exact mode of action (e.g. source of micronutrients,

source of immunostimulants, water quality conditioner, and microbial conditioner) of

“green-water” (i.e. high concentrations of selected species of micro-algae) in the

commercial larviculture of several species of marine fish remains however unclear and

therefore requires further study (Sorgeloos, 1995). Most likely, the addition of micro-algae

to the rearing system supported rotifer growth and maintained their nutritional quality, in


78

this way indirectly benefiting the larvae. It can therefore be concluded that although micro-

algae are not a proper initial feed for early crab stages and might cause certain negative

effects during early larval rearing, they seem important to improve the rotifer quality and in

this way improve growth and metamorphosis of the crab larvae.

4.3. Alternative Artemia forms as first feed

In experiments 4 and 5, the positive control (rotifers) always resulted in the best

survival and highest LSI throughout the whole rearing period, confirming that rotifers are

the best first feed.

Among the other treatments, heat-killed umbrella Artemia gave the best survival in

both experiments. Live Artemia nauplii seemed the best in terms of LSI in experiment 5.

The relative high survival and good growth of the treatment “live instar-1 Artemia” was

somehow contradictory to the results of experiment 1. In that experiment Z1 seemed unable

to catch newly-hatched Artemia. Whereas in experiment 1, no aeration was applied, in

experiments 4 and 5, the rearing medium was thoroughly aerated and mixed however.

Possibly, the water turbulence maximised the probability of Z1 to encounter prey, resulting

in a higher incidence of contact and more chance to capture the nauplii. In this respect, the

suitability of live feed and prey selection by the larvae can be explained by the relative

vulnerability of prey items (Dutil et al., 1977; Greene and Landry, 1985; Yen, 1982).

Vulnerability can be defined as the product of two terms, the probability of encounter and

the probability of capture (Pastorok, 1981). The rate of encounter varies directly with the

size and swimming speed of prey, but also with physical properties of the system, while the

efficiency of capture varies inversely with these factors (Swift and Fedorenko, 1975). This

also stresses the impact zootechnical factors such as system design can have on feeding.

Zeng and Li (1999) often found Z1 larvae holding Artemia for a long time but finally

abandoning them usually only having the head or appendages removed. This way, although

not an ideal prey, Artemia can sustain growth and survival of zoeal larvae to some extent.

However, Li et al. (1999) noticed that Z1 larvae (S. paramamosain) fed with Artemia

nauplii usually resulted in lower survival (compared to rotifers), which confirms our results.

Both heat-killed and frozen instar-1 Artemia tended to sink to the bottom of the

rearing containers rapidly and so reduced feed densities in the water column considerably.

Frozen Artemia moreover decomposed rapidly in the rearing water resulting in increased


79

NH4+ and NO2

- concentrations. This probably explains why in both experiments mortality

was highest in this treatment.

In terms of nutritional quality, fresh (live or frozen) Artemia seemed to be better than

heat-killed Artemia, resulting in higher LSI values. Heat treatment of Artemia can cause

protein denaturation, resulting in decreased specific energy, protein solubility and enzyme

activities (García-Ortega et al., 1995). According to these authors, 11-day-old larvae of the

African catfish Clarias gariepinus showed slower growth when fed treated cysts compared

to live Artemia nauplii or untreated cysts that may reflect the effect of the destruction of

enzymes required for feed digestion and absorption.

Despite the poor performance of the larvae receiving non-moving food, the

observation that not all the larvae died in these treatments and were found catching and

consuming the food, shows that zoeae are not restricted to live prey. This non-selective

feeding at early stages also shows a direction to develop artificial feed particles as suitable

alternatives for live feed for mud crab larvae especially at early stages. Levine and Sulkin

(1984b) concluded that calcium alginate microcapsules, or parts thereof, clearly can be

ingested by newly-hatched crab larvae of Eurypanopeus depressus indicating that

brachyuran larvae are capable of capturing and ingesting non-living, non-motile prey. This

suggests that they are not obligate carnivores, but are omnivores whose nutritional needs

could be satisfied by a variety of feed types. Recently, Genodepa et al. (2004) have shown

that microbound diets incorporated with 14C-labelled rotifers are readily ingested by S.

serrata zoeae and megalopae.

In experiment 6, a pilot-scale larval rearing trial was performed using live umbrella-

stage Artemia as first feed. Due to the size of the experiment, it was not possible to run and

compare different treatments. Acceptable survival and growth were obtained however,

comparable to previous trials where rotifers were used as a first feed. Live umbrella-stage

Artemia combine a number of advantageous characteristics in that they are smaller than

Artemia nauplii, are non-moving and thus easy to catch, cause little water deterioration, stay

in suspension quite well (due to the attached chorion) and are nutritionally similar or even

better than Artemia nauplii, making them an acceptable replacement for rotifers. Although

survival and growth of the crab larvae fed live umbrella-stage Artemia was not higher as

previous trials using rotifers, this feeding regimen could be used as a rotifer supplement or

replacement when rotifer production fails.


80

4.4. Feeding schedule

Rotifers and Artemia have been commonly used as the main live feeds for Scylla

larvae in most studies. In many cases, uncertainty of the right time to offer the most suitable

prey that are easily captured, digested and assimilated by the crab larvae at a certain stage

has led to the cofeeding of both prey organisms for a prolonged rearing period. Ruscoe et al.

(2004) recently noted that while the provision of various species of with differing physical

and nutritional characteristics throughout the rearing period may ensure the larvae have a

suitable prey at all times, the difficulties in terms of culture and hatchery management may

make the overall hatchery unprofitable. Moreover, feeding more than one prey type in the

same larval stage, created a competition on dissolved oxygen between the prey and the

cultured larvae and resulted in low water quality due to higher levels of toxic metabolites

(including ammonia) being produced (study on Penaeus semisulcatus, Samocha et al.,

1989). The poor quality of the culture medium in its turn may have an impact on larval

health, feeding ability, and ultimately growth and survival (Ruscoe et al., 2004). For these

reasons, determination of the feeding schedule, i.e. the timing of introduction and cessation

of a prey organism into a culture system needs to be addressed as precisely as possible.

In experiment 7, replacing rotifers by Artemia before DAH 4 (the second day of Z2

stage) clearly compromised the survival and development of the crab larvae. Therefore,

rotifers should be fed to Z1 in order to achieve higher survival.

Results furthermore indicated that rotifers should be replaced by Artemia already in

the Z2 stage. An abrupt shift from rotifers to Artemia on DAH 4 (the second day of Z2)

tended to give similar survival than when rotifers and Artemia were cofed for an extra 2

days. A transition period to shift from one diet to another therefore seemed not necessary.

When rotifer feeding was maintained beyond the zoea 2 stage, lower survival and a delay in

larval development was experienced. This is in accordance with the findings of Ruscoe et al.

(2004) in an experiment where rotifer feeding was discontinued at every larval stage. From

this experiment, the authors concluded that while rotifers are a valuable inclusion in the

feeding regime for larval S. serrata, their use should be limited to the first zoeal stage only

to maximize growth and survival.

When comparing S. paramamosain larvae fed rotifers and Artemia, Zeng (1987; cited

in Li et al., 1999) found no difference in the dry weight; and carbon, nitrogen and hydrogen

content in larvae of both treatments. This suggested that the nutritional value of rotifers can


81

meet the requirements for development of Z2 stage larvae. However, as larvae entered Z3,

those fed with Artemia had considerably higher values and the difference compared to

larvae fed rotifers grew wider as larvae developed. The author suggested that diet

replacement should take place at Z3.

For S. paramamosain, it is apparent that Z2 larvae are capable to ingest rotifers more

easily than Artemia nauplii. The beneficial impact of Artemia feeding on the performance of

Z2 probably is due to their nutritional value rather than their size compared to that of

rotifers. However, from a nutritional perspective, both rotifers and Artemia are far from

ideal and show nutritional inconsistency (Southgate and Partridge, 1998). Whether rotifers

or Artemia are used for Z2 stage larvae might therefore depend on their nutritional value and

general culture conditions and zootechnics. For example, in the above-mentioned analysis

study of Zeng (1987; cited in Li et al., 1999), rotifers did still meet the nutritional

requirements of crab larvae until the Z2 stage and shifting to Artemia feeding was suggested

to take place only in the Z3 stage. While our study and the study by Ruscoe et al. (2004)

indicated this should already be done at Z2 stage.

Based on the feeding schedule applied in experiment 1 (in the stock rearing tanks) and

6, crab larvae were found to be capable of catching and ingesting Artemia meta-nauplii

enriched with emulsions containing highly unsaturated fatty acids from the Z3 stage

onwards.


Rotifers were the most suitable live feed for Z1 of S. paramamosain. Although micro-

algae were not a suitable first feed for early crab stages, they seemed to improve the quality

of the rotifers resulting in a more successful metamorphosis to the megalopa stage.

Crab larvae could start catching instar-1 Artemia nauplii from the Z2 stage onwards

and the number of consumed prey increased for each consecutive larval stage. The ability of

Z2 to catch Artemia nauplii was slightly variable between batches and individuals.

Artemia are best introduced at the Z2 stage already in order to maximise survival and

growth. Optimal feeding schedules should however also take into account the nutritional

composition of both live foods. An overlap in feeding rotifers and Artemia seems not really

necessary.


82

Live umbrella-stage Artemia were the best replacement for rotifers for feeding Z1 - Z2

compared to other Artemia forms.

The non-selective feeding at early crab larval stages indicates there might be

possibilities to develop artificial diets as alternatives for live feed, in order to reduce the

dependency on rotifer cultures.

From the Z3 stage onwards, crab larvae can catch and ingest enriched Artemia meta-

nauplii.

Acknowledgements




83

Table 1 Overview of experimental conditions in experiment 1 to 8

Experiment Container volume (l)

Stocking density (Z1 l-1)

Number of replicates

Temperature (°C)

Salinity (g l-1 )

1 0.02 50 See text 29.0±0.7 30.0±0.5 2 2 40 6 29.5±0.7 29.9±1.0 3 2 40 3/12§ 30.6±1.4 30.0±0.5 4 1 50 3 29.1±1.2 30.0±0.5 5 30 150 5 29.2±0.4 30.5±1.4 6 500 100 8 29.4±0.7 30.0±0.6 7 30 100 4 29.4±1.1 30.1±0.5 8 30 50 5 29.2±0.5 30.1±0.4 § 12 replicates for treatment “rotifers and micro-algae”.


84

Table 2 Overview of feeding schedules used in experiment 2 to 8. DAH = day after hatch, Z = zoea, M = megalopa, C1 = first crab, Rot. = rotifers, (Micro-)algae = Chaetoceros spp., HK = heat-killed

DAH 0 1 2 3 4 5 6 7 8 9 10 11 12 … 15 … 22-30

Z1 Z3 Z5 C1 LARVAL STAGE Z2 Z4 M

EXPERIMENT 2 Starvation Artificial diet Micro-algae Rotifers only Rot. + algae

<------ Starvation ------------------------><------ Shrimp diet ----------------------><------ Micro-algae ----------------------><------ Rotifers ---------------------------><------ Rotifers+Micro-algae ---------->

EXPERIMENT 3

<------ Starvation ------------------------><------ Algae ----------------------------->

<------ Rotifers ----------------------------------------------------------------------------------------------><-----Rotifers+algae------->

Starvation Micro-algae Rotifers only Rot. + algae <------ Rotifers+Micro-algae --------------------------------------------> <-----Artemia+algae-------> EXPERIMENT 4 and 5 Starvation Rotifers HK umbrella HK nauplii Live nauplii Frozen nauplii

<------ Starvation ------------------------><------ Rotifers ---------------------------><------ HK umbrella Artemia ----------><------ HK Artemia nauplii -----------><------ Live Artemia nauplii -----------><------ Frozen Artemia nauplii -------->

EXPERIMENT 6 Live umbrella <--- Live umbrella Artemia ------------><---- Enriched Artemia ---> EXPERIMENT 7 1/1 2/2 3/3 4/4 5/5

<-Rot.-> <--- Artemia ---------------------------------------><---- Rot.----> <--- Artemia --------------------------------><---- Rot.----------> <--- Artemia --------------------------><---- Rot.-----------------> <--- Artemia -------------------><---- Rot.------------------------> <-Artemia --------------->

<--- Rot.-----------------------------> 6/4 <------ Artemia -----------------> EXPERIMENT 8

<------ Rotifers --------------------------> 6/4 <------- Artemia -------------------------------------------------->

<--- Rotifers -------------------------------------------> 9/7 <------- Artemia ------------------------------>

<--- Rotifers ----------------------------------------------------------------> 12/9 <------ Artemia ----------------->


85

Table 3 Number of Artemia nauplii remaining after incubating different larval stages of S. paramamosain for 24 hours together with 200 newly-hatched Artemia nauplii in 20-ml vials (experimental vials). Control vials were incubated with the same number of Artemia nauplii, but without crab larvae. Z = zoea, M = megalopa. Experiment 1 Larval stage Number of replicates Experimental vial Control vial TEST 1 Z1 / n.d. n.d. Z2 21 185±9a 196±5a

Z3 26 180±8B 197±3A

Z4 30 173±14B 199±3A

Z5 29 161±27B 199±2A

M 13 43±18B 194±5A

TEST 2 Z1 20 198±2a 199±2a

Z2 20 195±2b 199±2a

Z3 27 187±7B 199±2A

Z4 13 180±11B 200±1A

Z5 / n.d. n.d. M / n.d. n.d. TEST 3 Z1 21 200±1a 200±1a

Z2 34 196±3a 200±1a

Z3 42 187±6B 199±2A

Z4 19 174±11B 198±2A

Z5 38 164±20B 199±2A

M 9 149±23B 199±2A

Values in the same row followed the same superscript letter are not significantly different (P ≥ 0.05, regular letters or P ≥ 0.01, capital letters). N.d. = not determined. Table 4 Survival (%) and larval stage index (LSI) up to DAH 6 of S. paramamosain larvae fed different feeds. Experiments 2 and 3

Survival rate (%) LSI

Experiment 2 Experiment 3 Experiment 2

Experiment 3 Treatment

DAH 3 DAH 6 DAH 3 DAH 6 DAH 3 DAH 6 DAH 3 DAH 6 Starvation 0 / 2±2c C 0 / / 1.0±0.0b B / Artificial diet 1±2c C 0 / / 1.2±0.0b B / / / Micro-algae 7±6bc BC 0 29±26bc BC 1±2b A 1.0±0.0c C / 1.0±0.0b B 1.0±0.0b B

Rotifers only 39±12a A 17±18a 79±7a A 41±37a A 1.4±0.1a A 2.8±0.0a 1.5±0.0a A 2.9±0.0a A

Rotifers+algae 15±4b B 8±2a 54±14ab AB 19±8ab A 1.4±0.1a A 2.8±0.1a 1.5±0.1a A 2.8±0.1a A

Values in the same column followed the same superscript letter are not significantly different (P ≥ 0.05, regular letters and P ≥ 0.01, capital letters).


86

Table 5 Survival (%) from DAH 9 to DAH 22 of S. paramamosain larvae fed different diets. Algae = micro-algae (Chaetoceros spp.). Experiment 3

Experimental design Treatment No of replicates

DAH 9 (Z3)

DAH 12 (Z3-Z4)

DAH 15 (Z4-Z5)

DAH 22 (Megalopa)

Rotifers only 3 22±6a Initial Rotifers + algae 12 11±5b Rotifers only 2 100 25B After pooling on

DAH 9 Rotifers + algae 3 100 79A Rotifers only 1 100 39b B 0 Rotifers + algae 1 100 85a A 48aAfter pooling on

DAH 12 Artemia + algae 1 100 88a A 60a

Values in the same column followed the same superscript letter are not significantly different (P ≥ 0.05, regular letters and P ≥ 0.01, capital letters). Data of initial design are treated by Tukey test (with standard deviation). Pooled data are treated by Fisher exact test (without standard deviation). Table 6 Survival (%), larval stage index (LSI) and water quality parameters of S. paramamosain larvae fed different feeds. Experiments 4 and 5

Survival rate (%) LSI Treatment

DAH 3 DAH 6 DAH 3 DAH 6 NH4

+

(mg l-1) NO2-

(mg l-1)

EXPERIMENT 4 Starvation 2±3b B / n.d. / 0.23±0.15b B 0.03±0.03a

Rotifers 43±9a A / n.d. / 0.50±0.12b AB 0.04±0.03a

Heat-killed umbrella-stage Artemia 20±10ab A / n.d. / 0.45±0.13b AB 0.10±0.06a

Heat-killed instar-1 Artemia 12±16b A / n.d. / 0.40±0.08b B 0.09±0.05a

Live instar-1 Artemia 12±3ab A / n.d. / 0.40±0.08b B 0.06±0.03a

Frozen instar-1 Artemia 0 / n.d. / 0.86±0.23a A 0.12±0.08a

EXPERIMENT 5 Starvation 0 / / / / / Rotifers 75±24a A 37±17a A 1.5±0.1a A 2.8±0.1a A n.d. n.d. Heat-killed umbrella-stage Artemia 47±6b AB 10±1b B 1.0±0.0b B 2.0±0.2b B n.d. n.d. Heat-killed instar-1 Artemia 34±10bc B 6±3b B 1.0±0.0b B 2.0±0.2b B n.d. n.d. Live instar-1 Artemia 26±4bc B 9±3b B 1.0±0.0b B 2.2±0.3b B n.d. n.d. Frozen instar-1 Artemia 23±7c B 9±3b B 1.0±0.0bB 2.1±0.2b B n.d. n.d. Values in the same column for the same experiment followed the same superscript letter are not significantly different (P ≥ 0.05, regular letters and P ≥ 0.01, capital letters). N.d. = not determined.


87

Table 7 Survival (%) and larval stage index (LSI) of S. paramamosain larvae receiving different rotifer/Artemia feeding schedules. For treatment descriptions refer to Table 2. Experiment 7

Survival rate (%) LSI Feeding schedule

DAH 3 DAH 6 DAH 9 DAH 3 DAH 6 DAH 9 1/1 61±18b A 47±26a 34±11b A 2.0±0.1a 2.9±0.2a 4.0±0.1a

2/2 86±13ab A 57±14a 49±10ab A 2.0±0.1a 2.8±0.2a 4.0±0.1a

3/3 74±21ab A 44±17a 33±12b A 2.0±0.0a 2.9±0.1a 4.0±0.1a

4/4 97±4a A 75±25a 69±17a A 2.0±0.0a 3.0±0.0a 4.0±0.0a

5/5 88±12ab A 56±18a 41±9ab A 2.0±0.0a 3.0±0.0a 4.0±0.0a

6/4 90±14ab A 73±27a 59±23ab A 2.0±0.0a 3.0±0.0a 4.0±0.0a

Values in the same column followed the same superscript letter are not significantly different (P ≥ 0.05, regular letters and P ≥ 0.01, capital letter). Table 8 Survival (%) and larval stage index (LSI) of S. paramamosain larvae receiving different rotifer/Artemia feeding schedules. For treatment descriptions refer to Table 2. Experiment 8

Survival rate (%) LSI Feeding schedule

DAH 9 DAH 12 DAH 15 DAH 9 DAH 12 DAH 15 6/4 36±08a 21±10a 17±10a 3.9±0.3a A 4.3±0.4a 4.5±0.4a

9/7 35±10a 14±8a 11±7a 3.6±0.3b A 4.0±0.2a 4.2±0.5a

12/9 32±15a 11±6a 6±3a 3.6±0.3b A 3.8±0.3a 3.9±0.4a

Values in the same column showing the same superscript letter are not significantly different (P ≥ 0.05, regular letters and P ≥ 0.01, capital letters). Table 9 Sizes (µm) of used live feeds and those of Scylla eggs and larvae. n = number of observations Length Width Diameter n LIVE FEED Adult rotifers 164±22 120±22 50 Artemia cysts (with shells) 235±15 100 Newly-hatched Artemia 601±101 378±55(1) 15 24h-enriched Artemia 763±102 469±51 15 NEWLY-SPAWNED EGGS S. paramamosain(2) 287.6±9.9c C 2910 S. tranquebarica(3) 299.5±9.2b B 27 S. olivacea (3) 300.6±4.4b B 22 S. serrata(3) 329.1±8.3a A 27 CARAPACE OF S. paramomasain LARVAE(4) Z1 452 Z2 571 Z3 714 Z4 1058 Z5 1577 M 1280 840 Values in the same column followed the same superscript letter are not significantly different (P ≥ 0.05 and P ≥ 0.01). (1) Width of Artemia including appendages (around 150 µm excluding appendages); (2) Our data, sampled from 97 batches of eggs (30 eggs batch-1); (3) Data from Emilia T. Quinitio, Southeast Asian Fisheries Development Center, pers. com.; (4) Measured from drawings in Jones et al. (in press) and measuring methods based on Bigford (1978).


88

Mean; Box: Mean-SD, Mean+SD; Whisker: Min, Max

No

of re

mai

ning

Arte

mia

Experiment vialControl vial

Z1 Z2* Z3** Z4** Z5** M**

Stage of crab larvae

20

40

60

80

100

120

140

160

180

200

Figure 1. Number of remaining Artemia after incubating different S. paramamosain larval stages for 24 hours together with 200 Artemia nauplii in 20-ml vials (experimental vials). Control vials were incubated with the same number of Artemia nauplii, but without crab larvae. Z = zoea, M = megalopa. Mean = average of 3 tests. * and ** for P < 0.05 and P < 0.01, respectively indicate that the number of remaining Artemia in the experimental vial are significantly different with those in the control vial. Experiment 1.

CHAPTER 5

Influence of the content of highly unsaturated fatty acids in the live feed on larviculture success of mud crab

(Scylla paramamosain)

Nghia, T.T.*1, Wille, M.2, Vandendriessche, S.2 and Sorgeloos, P.2


Abstract Four experiments were carried out to investigate the effects of the level of

docosahexaenoic acid (DHA), eicosapentaenoinc acid (EPA) and arachidonic acid (ARA) in the live feed on survival, growth and metamorphosis success of mud crab Scylla paramamosain larvae. Six different lipid emulsions, varying in the level of total n-3 and n-6 highly unsaturated fatty acids (HUFA), DHA, EPA and ARA were used to manipulate the fatty acid profile of the live feed (rotifers and Artemia). Fatty acid profiles of the live feed and crab larvae at zoea 1, 3 and 5 stages were analyzed to study uptake of HUFA by the crab larvae. Larviculture success was measured through survival, larval development rate (expressed as larval stage index) and success of metamorphosis (survival and time and duration of first and second metamorphosis).

The fatty acid content of the live feed affected the fatty acid profiles of the crab larvae. In most experiments, survival rate in the zoeal stages was not statistically different among treatments. Larval development rate and metamorphosis success were however more strongly affected by the dietary treatments. In this respect, the DHA/EPA ratio in the live feed seems to be a key factor. Enrichment emulsions with very high (50 %) total HUFA content but low DHA/EPA ratio (0.6) or zero total HUFA content caused growth retardation and/or metamorphosis failure. An emulsion with moderate total HUFA (30 %) and high DHA/EPA ratio (4) was the best in terms of larval development rate during the zoeal stages and resulted in good metamorphosis. The optimal DHA/EPA ratio of live feed enrichment emulsions for early stages (Z1 - Z2) could however be lower than 4. Dietary arachidonic acid seemed to improve first metamorphosis, but its exact role needs further clarification. For the larval rearing of Scylla paramamosain, it is recommended to use enrichment media with a total n-3 HUFA content of approximately 30 %, with a DHA/EPA ratio of minimum 1. Further research needs to be performed on the total HUFA and DHA/EPA ratio requirements for each larval crab stage. The role of ARA in metamorphosis also needs to be further elucidated.

CHAPTER 5 – Feed quality

90

1. Introduction

Mud crabs, Scylla spp., are large, primarily carnivorous portunid crabs, which are

strongly associated with the mangrove areas throughout the Pacific and Indian oceans

(Keenan, 1999a). They form the basis of substantial, but mainly artisanal fishery operations

throughout their distribution. South East Asia has a long history in traditional forms of mud

crab farming (e.g. fattening), but a number of factors have triggered a renewed interest in

mud crab farming in recent years. Johnston and Keenan (1999) listed a number of benefits

of mud crab over shrimp farming as providing more reliable income and higher profit

margins and return of initial investment resulting from higher survival because of superior

adaptation to the mangrove environment, higher price per kg with little capital or food input,

high growth rate and lower disease risk. Mud crab farming however currently relies entirely

on wild seed stock and the main obstacle for the development of mud crab culture is the

availability of hatchery-reared seed (Liong, 1992; Keenan, 1999a; Mann et al., 2001;

Shelley and Field 1999; Xuan, 2001). Of the four Scylla species, Scylla paramamosain is

dominant in Vietnam (Keenan et al., 1998; Keenan, 1999b).

One of the major factors influencing the survival and growth of larvae of marine

species is the dietary HUFA composition. Sorgeloos et al. (2001) reviewed the history of

research on dietary HUFAs. In the 1980’s, most attention was dedicated to the presence of

eicosapentaenoic acid (20:5n-3, EPA) in Artemia as a guarantee for successful production of

marine fish larvae. In the late 1980’s and early 1990’s, more attention was paid to the level

of docosahexaenoic acid (22:6n-3, DHA) because good survival appeared to be correlated

with EPA, but DHA improved larval quality and growth. The importance of DHA, more

particularly the requirement for high DHA/EPA ratios in promoting growth, stress

resistance, and pigmentation was also revealed (Sorgeloos et al., 2001). Recent work

showed that besides DHA not only highly unsaturated fatty acids of the n-3 series are

important but that also arachidonic acid (20:4n-6, ARA) may play a significant role (Castell

et al., 1994; Estévez et al., 1999, Koven et al., 2000). ARA may improve larval growth and

pigmentation in several marine fish species since it provides precursors for eicosanoid

production.

Many studies on supplementation of essential fatty acids through the live feed, mainly

rotifers (Dhert et al., 2001; Olsen et al., 1993) and Artemia (Sorgeloos et al., 2001;

Watanabe, 1982) have been performed in marine and freshwater organisms, including fish


91

(e.g. Ashraf, 1993; Sargent et al., 1995; Watanabe, 1993), shrimp (e.g. Lavens and

Sorgeloos, 2000; Rees et al., 1994) and bivalves (e.g. Caers et al., 1998; Coutteau et al.

1996). Although data on the nutritional requirements of larval stages of brachyuran crabs

are rather limited, the requirement for dietary fatty acids has been demonstrated for several

species, including Scylla spp. (Hamasaki et al., 1998; Kobayashi et al., 2000; Levine and

Sulkin, 1984a; Suprayudi et al., 2002b; Takeuchi et al., 1999; Takeuchi et al., 2000).

Despite these findings, problems to develop reliable zootechnics and the often low and

inconsistent larval survival for Scylla larvae have hampered nutritional research

considerably and contradictory results have been reported. Takeuchi et al. (2000) for

example showed that mud crab larvae fed rotifers and Artemia enriched with HUFAs

presented increased survival and performance. Hamasaki et al. (2002b) on the other hand

reported that elevated levels of EPA in the live feed resulted in abnormal development and

mortality of the larvae at metamorphosis. In a study comparing different Artemia strains and

Artemia enrichment products, Mann et al. (2001) found no influence of the n-3 HUFA level

on the ability of the larvae to complete development. From this it is clear that exact dietary

requirements for n-3 HUFA of mud crab larvae have not been established yet.

In order to define more clearly the importance of n-3 HUFA in the diet of Scylla

larvae, the present study evaluates the effect of the level and ratio of specific n-3 HUFAs

(DHA and EPA) and ARA in the live feed on the fatty acid composition and culture

performance of mud crab S. paramamosain larvae.


2.1. Source of larvae

Fully gravid crabs were bought from local markets in the coastal area and transported

to the hatchery. Prior to stocking in the hatchery, the crabs were bathed in a 100 µl l-1

formalin solution for 1 hour. The crabs were housed individually in 100-l compartments of a

roofed 2 × 2 × 0.5 m cement tank, equipped with a biofilter. Rearing water (30 ± 1 g l-1) was

diluted from brine (90 - 110 g l-1) with tap water and chlorinated before use. Water

temperature was not controlled, but varied around 28 °C. Each crab was daily fed 10 - 15 g

of fresh marine squid, bivalve or shrimp meat alternately.


92



for 1 hour and transferred to a 70-l plastic tank connected to a biofilter for incubation. Daily


bottom and controlling temperature (30 °C), salinity (30 g l-1) and ammonia and nitrite

levels. Every other day, the crabs were bathed in a 50 µl l-1 formalin solution for 1 hour to


crabs were not fed.

One to two days prior to hatching, the female was moved to a 500-l fibreglass tank in

order to provide a clean and spacious environment for the hatching larvae. At 30 °C,

hatching normally occurred after 10 days of incubation.

When the hatching process was completed, larvae were selected based on their photo-

tactic behaviour. Aeration in the hatching tank was therefore turned off for several minutes

and the larvae that were actively swimming up to the surface were collected by gently

scooping them from the surface. In order to slowly acclimate the larvae to the new rearing

conditions, the larvae were then placed in a 50-l plastic mesh bucket and slowly rinsed with

water from the larval rearing containers for 20 to 30 minutes.

2.2. Larval rearing

Larval rearing systems and procedures

Two different rearing systems were applied in this study. In experiments 1 and 2,

larvae were reared in the recirculating system consisting of 30-l (experiment 1) or 100-l

(experiment 2) cylindro-conical fibreglass tanks connected to a submerged biological filter.

An upwelling system with a water renewal rate of 100 % every 3 - 4 hours was used. Water

was evacuated at the water surface and passed through a 70 or 300 µm filter screen during

the rotifer and Artemia feeding stage respectively, and thus retained larvae and live feed in

the culture tank. Gentle aeration was applied to all rearing tanks. Formalin at a

concentration of 20 µl l-1 was applied to the whole system every 2 days to prevent or reduce

fungi and bacteria development.

In experiments 3 and 4 a small-scale static rearing mode consisting of 1-l acrylic

bowls was used. The bowls were randomly distributed over a heated water bath to provide


93

identical rearing conditions. Like for the recirculating system, gentle aeration was applied to

all bowls. Each day the remaining larvae were pipetted into new bowls containing fresh

seawater. These “new” bowls were incubated beforehand in the same water bath to

equilibrate temperature. In order to exclude bacterial interference, 10 mg l-1 Oxytetracycline

was applied to the bowls daily. In experiment 3, each treatment was also repeated in two

100-l fibreglass tanks (operated in recirculation mode as in experiment 1 and 2).

Only in experiments 3 and 4 the larvae were reared through to the megalopa and first

crab stage. In experiment 3, megalopae were separated daily from the rearing containers and

redistributed by treatment in 1-l bowls at a density of 6 - 8 megalopae bowl-1. Substrate (1-

cm3 pieces of PVC sponge) as for hiding space was provided in each bowl to prevent

cannibalism. In experiment 4, megalopae were separated from the culture several times per

day and reared individually to crab stage in 100-ml cups without aeration.

Source and treatment of the rearing water for the larvae was the same as for

broodstock rearing.

Larval rearing conditions for all experiments are summarized in Table 1.

Live feed culture and enrichment

- Culture practices


164 ± 22 and 120 ± 22 µm, respectively, was used in all experiments. In experiments 1, 2

and 4, rotifers were cultured outdoor in an integrated recirculating system. The system

consisted of two 10-m3 tanks inoculated with Chlorella spp. and stocked with Tilapia (1 kg

m-3), connected to a 4-m3 fibreglass rotifer culture tank. When the algal concentration

reached 10 million cells ml-1, rotifers were stocked at 100 ml-1 and the system was

recirculated (1 complete exchange every 4 hours). Starting from the third day, part of the

rotifer population was then harvested daily as feed for the crab larvae. In experiment 3,

rotifers were cultured on baker yeast in an indoor recirculating system in 100-l fiberglass

tanks (Suantika, 2001). Water recirculation rate in this system was 100 % day-1.

Artemia (Vinh Chau strain) cysts were disinfected with chlorine, incubated and

hatched following standard methods (Sorgeloos et al., 1986).


94

- Enrichment practices

In order to manipulate the fatty acid content of the live feed, rotifers and Artemia were

enriched with different ICES (International Council for Exploration of the Sea) standard

reference emulsions (ICES, 1997).

Rotifer enrichment was performed at a density of 500 ml-1, using two separate doses

of 0.125 g l-1 each at a 3-hour interval. The temperature and salinity of the water was

maintained at 25 to 30 °C and 25 to 30 g l-1, respectively.

For Artemia enrichment, newly-hatched Artemia were concentrated to 200 ml-1 and

the oil emulsions were used at two separate doses of 0.3 g l-1 each at a 12-hour interval.

Temperature and salinity were kept at 30 °C and 30 g l-1, respectively.

In experiment 2, control Artemia were kept in an algal suspension (5 ± 1 million cells

of Chaetoceros ml-1) for the duration of the enrichment period of the other treatments, to

maintain a uniform size with the enriched Artemia meta-nauplii in the other treatments.

In experiment 3, control Artemia were starved at 30 °C for 24 hours.

In experiment 4, the control rotifers and Artemia were enriched with the commercial

product Culture Selco® (28 % n-3 HUFA with a DHA/EPA ratio of 1; INVE Aquaculture

NV, Belgium), under similar conditions (live feed density, salinity and temperature,

enrichment dose and time) as for the ICES emulsions.

Feeding

In experiment 1, rotifers were the sole live feed for all crab larval stages. Rotifers

were added daily (45 ml-1) to the rearing tanks after the old ones were flushed out almost

completely (3 - 5 ml-1 left).

In experiments 2, 3 and 4, crab larvae were fed rotifers from DAH 0 - 6 (Z1 - Z2) at

the same density as experiment 1 and then Artemia afterwards (7 - 10 ml-1 daily).

2.3. Experimental design

The design of the 4 experiments is presented in Table 2. Six different ICES emulsions

or combinations thereof were tested for rotifer and Artemia enrichment: emulsion 0/- (a

coconut emulsion, free of HUFA, mainly consisting of saturated fatty acids); emulsion


95

30/0.6 (containing approximately 30 % n-3 HUFA on total fatty acids with a DHA/EPA

ratio of 0.6); emulsion 30/4 (30 % n-3 HUFA with a DHA/EPA ratio of 4; emulsion 50/0.6

(50 % n-3 HUFA with a DHA/EPA ratio of 0.6); emulsion 30/1/ARA (having

approximately 30 % n-3 HUFA and a ratio of ARA/EPA/DHA of 1/1/1) and emulsion

30/4/ARA (a mixture of 75 % of the 30/4 emulsion and 25 % of an emulsion containing 40

% arachidonic acid).

In experiments 2 and 3, the same enrichment emulsions were used for rotifers and

Artemia enrichment. In experiment 4, one treatment was included (0.6-4) where rotifers

were enriched with the 30/0.6 emulsion and Artemia with the 30/4 emulsion. Hence,

treatments in experiment 4 were designated as treatment 0.6-0.6 (corresponding to treatment

30/0.6 in the other experiments), treatment 4-4 (i.e. treatment 30/4) and treatment 0.6-4. For

each experiment, a control treatment was included (see section “Live feed culture and

enrichment” and Table 2). Where appropriate, the control treatments are clarified as:

“Chlorella” control, “Chaetoceros” control, “starvation” control and “1-1” control for the

control treatments of experiments 1, 2, 3 and 4 respectively.


Fatty acid composition

In experiment 3, the ICES emulsions, the different enriched live feeds (rotifers and

Artemia) and the crab larvae (Z1, Z2 and Z5) were sampled for fatty acid analysis. Samples

were washed with freshwater and stored under –80°C until fatty acid methyl ester (FAME)

analysis. Fatty acid composition was determined by gas chromatography. FAMEs were

prepared via a procedure modified from Lepage and Roy (1984). The method consists of a

direct acid-catalized transesterification of dry samples (ranging from 10 - 150 mg) without

prior lipid extraction. An internal standard 20:2(n-6) or 22:2(n-6) was added prior to the

reaction. FAMEs were extracted with hexane. After evaporation of the solvents the FAMEs

were prepared for injection by re-dissolving them in iso-octane (2 mg ml-1). Quantitative

determination was done by a Chrompack CP9001 gaschromatograph equipped with an auto-

sampler and a TPOCI (Temperature programmable on-column injector). Injections (0.5 µl)

were performed on column into a polar 50 m capillary column, BPX70 (SGE Australia),

with a diameter of 0.32 mm and a layer thickness of 0.25 µm, connected to a 2.5 m methyl


96

deactivated precolumn. The carrier gas was H2, at a pressure of 100 kPa and the detection

mode was FID (flame ionization detection). The oven was programmed to rise from the

initial temperature of 85 to 150 °C at a rate of 30 °C min-1, from 150 to 152 °C at 0.1 °C

min-1, from 152 to 172 °C at 0.65 °C min-1, from 172 to 187 °C at 25 °C min-1 and to stay at

187 °C for 7 min. The injector was heated from 85 to 190 °C at 5 °C sec-1 and stayed at 190

°C for 30 min. Identification was based on standard reference mixtures (Nu-Chek-Prep, Inc.,

U.S.A.). Integration and calculations were done on computer with a software program

Maestro (Chrompack). Each sample was analyzed twice. The results are expressed as mg

FAME per gram of dry weight (mg g-1 DW or mg g-1).

Only the most important essential fatty acids and groups (ARA, EPA, DHA, total n-3,

total n-6 and total HUFA) and ratio’s thereof (DHA/EPA, ARA/EPA and Σn-3/Σn-6) are

reported in the results section. Total HUFA was defined as all fatty acids with more then 20

carbon atoms and 3 or more unsaturated carboxyl bonds, total n-3 HUFA as all n-3 fatty

acids ≥ 20:3n-3.

Larval performance

In experiments 1 and 2 (in 30 to 100-l containers, respectively), the average survival

rate in the zoeal stages was estimated by volumetric sampling. Depending on the tank

volume and the density of the surviving larvae, triplicate 300- to 1000-ml samples were

taken from each tank. In experiments 3 and 4 (in 1-l bowls) the average survival rates were

calculated by individually counting all surviving larvae in each replicate.

Zoeal development was monitored every three days by identifying the zoeal instar

stage of a sample of larvae and assigning it a value: first zoea (Z1) = 1; second zoea (Z2) =

2, etc. To compare the larval development in each treatment, an average larval stage index

(LSI) was calculated from the average LSI value of replicate tanks in the same treatment.

Five or ten larvae (in 30-l and 100-l tanks respectively) were sampled from every tank in

experiments 1, 2 and 3 to calculate the average LSI of each tank. The sampled larvae were

staged under a dissecting microscope. Only in experiment 4, larvae were staged visually

upon daily counting the surviving larvae.

Two parameters were used to evaluate the success of metamorphosis in experiments 3

and 4:


97

- Firstly the metamorphosis rate (MR, %) or the percentage survival through metamorphosis

was determined. Since there are 2 metamorphoses, the 2 metamorphosis rates were

calculated: (i) the percentage of Z5 that survives through metamorphosis to megalopa

(MR1) and (ii) the percentage of megalopa that survives through metamorphosis to crab

stage (MR2).

- Secondly also the average time needed for the larvae to go through metamorphosis and the

duration (minimum and maximum time needed) of both metamorphoses were recorded.

- Lastly the survival rates of Z1 to megalopa and crab stage (only for experiment 4) were

graphically presented based on the pooled data of all replicates in each treatment.



variance was tested with the Levene test (P or α value was set at 0.05). If no significant





survival rate) or logarithmic transformations (for other parameters) (Sokal and Rohlf, 1995).

The two-tailed Fisher exact test (modified from the contingency table method) was used to

compare ratio (expressed in percent) data of pooled treatments. The Pearson’s coefficient

was used to examine the correlation between the fatty acid composition of the live feed and

the crab larvae; and the correlation between the fatty acid composition of the live feed and

larval development (i.e. LSI values and metamorphosis rates). All data are presented as

mean ± standard deviation when using the Tukey test or as a ratio/percentage (without

standard deviation) when the Fisher exact test was used. P was set at both 0.05 and 0.01.

Whenever differences are significant at P < 0.01, this is also indicated. All analyses were

performed using the statistical program STATISTICA 6.0.


98

3. Results

3.1. Fatty acid composition of live feed and crab larvae (experiment 3)

Only in experiment 3, the fatty acid profile of the live feed (Table 4) and the larvae

(Table 5) was determined. The composition of the emulsions (the same batch was used for

all experiments) is presented in Table 3. We assume that, except for the control treatments

(which differed between experiments), similar live food enrichment levels were obtained in

the other experiments. Rotifer composition was clearly influenced by the enrichment

treatments. Total n-3 content of the rotifers increased from 4 mg g-1 in the control to 26 to

35 mg g-1 in treatments 30/0.6, 30/4 and 30/4/ARA. The highest n-3 level (49 mg g-1) was

obtained in treatment 50/0.6. Treatment 0/- resulted in a level similar to the control. As

expected, treatment 50/0.6 resulted in the highest EPA level in the rotifers (29 mg g-1).

DHA/EPA ratio of the rotifers was highest for treatments 30/4 and 30/4/ARA (2.57 and 2.1,

respectively), compared to 0.71 to 0.84 for the other treatments. Treatment 30/4/ARA

resulted in elevated ARA and total n-6 levels and hence a high ARA/EPA ratio. The control

and 0/- rotifers were especially low in ARA, EPA and DHA. Similar patterns were observed

for Artemia. Using the same enrichment medium, DHA levels and the DHA/EPA ratio

were however lower in Artemia compared to rotifers (except for treatment 30/4, where the

DHA level of Artemia was slightly higher than that of rotifers). Especially the DHA level in

the control Artemia was extremely low, giving a very low DHA/EPA ratio. Absolute levels

of most other fatty acids were however higher in Artemia. Compared to un-enriched rotifers,

the control Artemia also contained relatively high amounts of EPA, total n-3 and ARA.

Overall, the rotifer and Artemia composition reflected very well the total n-3, EPA, DHA

and ARA levels of the emulsions and thus the theoretic design of the experiments.

The fatty acid composition of the different crab larval stages in experiment 3 is given

in Table 5. The dietary fatty acid level in its turn affected the composition of the larvae.

Treatment 50/0.6 led to the highest total n-3 and EPA level in Z3 and Z5, while intermediate

results were obtained for treatments 30/0.6, 30/4 and 30/4/ARA. The DHA level and

DHA/EPA ratio was increased in all treatments except 0/-, but was highest for 30/4 and

30/4/ARA. A very low DHA/EPA ratio was observed for Z5 in the control and 0/-

treatments (0.11 and 0.16, respectively). The ARA content of Z3 and Z5 in the 30/4/ARA


99

treatment (5 and 7 mg g-1, respectively) was higher then those in the other treatments.

Treatment 30/4/ARA also resulted in the highest ARA/EPA ratio in the crab larvae.

In Table 6, the correlation coefficients between specific fatty acid levels in the crab

larvae and those of the live feed are summarized. For Z3, the EPA level, and the DHA/EPA,

ARA/EPA and n-3/n-6 ratio were significantly correlated to those of the rotifers (all at P <

0.01). The level of ARA, DHA and total n-3 were weakly correlated however (r2 = 0.36,

0.62 and 0.65, respectively). For Z5, all investigated fatty acid levels and ratios were

significantly correlated to those of the Artemia (P < 0.01, except ARA/EPA ratio at P <

0.05).

3.2. Zoeal survival

Survival during the zoeal stages in experiments 1 to 4 is presented in Table 7 to 10.

Survival was generally low in later larval stages and high variability was observed between

replicates. This makes comparison difficult and hence not many significant differences in

survival were observed between any of the treatments. In experiment 3 (Table 9) a quite low

survival (although not significantly different) was observed on DAH 15 for treatment 30/4,

resulting from a high mortality from DAH 12 to 15. In experiment 4 (Table 10), on DAH 9,

survival was lower in treatment 0.6-0.6 compared to the control (P < 0.05).

3.3. Larval development rate during the zoeal stages

Experiment 1

Larval development rates (expressed as LSI) in the different treatments are presented

in Table 7. From DAH 6 to 12, the LSI values in treatment 0/- were significantly lower than

in the “Chlorella” control and treatments 30/0.6 and 30/4 (P < 0.01, except on DAH 9 at P

< 0.05). Treatment 50/0.6 had intermediate values. By DAH 15 however, no significant

differences were observed anymore. Overall, a better growth rate was observed for

treatments 30/0.6, 30/4 and the control compared to the other treatments. Whereas the

control and treatment 30/0.6 performed better in early larval stages (from DAH 3 - 9),

treatment 30/4 became better in late larval stages (from DAH 12 - 15). Treatment 30/0.6

tended to have higher LSI values than treatment 50/0.6 on all sampling days.


100

Experiment 2

Table 8 presents the LSI values in experiment 2. On DAH 6, a significantly higher

development rate was observed for treatment 30/4 compared to the control and treatments

0/-, 30/0.6 and 50/0.6 (P < 0.01). LSI values of treatment 30/1/ARA were higher than the

control and treatments 30/0.6 and 50/0.6 (P < 0.05). On DAH 9 and 12, the LSI in

treatments 30/4 and 30/1/ARA were higher than in 0/-, 30/0.6 and 50/0.6 (P < 0.01). The

“Chaetoceros” control had intermediate results. LSI values of 30/1/ARA were slightly

lower than those of treatment 30/4. LSI values of the control treatment tended to be higher

than those of treatments 0/-, 50/0.6 and 30/0.6. In contrast with the previous experiment, the

LSI values of treatment 50/0.6 tended to be better than those of treatment 30/0.6 on most

sampling days.

Experiment 3

Larval development rates of experiment 3 are presented in Table 9. On DAH 3, the

LSI values of treatments 30/4 and 30/4/ARA were significantly higher than for all other

treatments (P < 0.01). Although not always statistically significant, this difference persisted

throughout the larval rearing period. The “starvation” control and treatment 0/- always gave

the lowest LSI values. For treatments 50/0.6 and 30/0.6, intermediate larval development

rates were found.

Experiment 4

LSI values in experiment 4 are shown in Table 10. On DAH 3, LSI values of the “1-1”

control (Culture Selco enriched rotifers and Artemia) and treatment 4-4 (or 30/4) were

significantly higher (P < 0.01, except treatment 4-4 at P < 0.05) than those in treatments

0.6-0.6 (or 30/0.6) and 0.6-4. LSI values in these treatments remained high throughout

larval development. No differences were detected between the control treatment and

treatment 4-4 on any of the sampling days.


101

3.4. Metamorphosis

In this study, larvae were only reared to the megalopa and first crab stage in

experiments 3 and 4.

Experiment 3

The metamorphosis rates of Z5 to megalopa and megalopa to first crab (MR1 and

MR2, respectively) are presented in Table 11. MR1 of treatment 30/4/ARA was

significantly higher (50 %) in the other treatments (ranging from 30 to 34 %) at P < 0.05.

For MR2, the “starvation” control and treatment 0/- had a significantly lower percentage of

megalopa reaching crab stage compared to the other treatments (P < 0.01). The treatments

with a high DHA/EPA ratio (30/4 and 30/4/ARA) gave a MR2 of over 90 %; the treatments

with a low DHA/EPA ratio (50/0.6 and 30/0.6) resulted in a metamorphosis rate of 73 %,

while in the treatments with a low total n-3 HUFA content (control and 0/-) only about 10 %

of the megalopae could metamorphose to crabs.

Also the average time needed to reach metamorphosis and the time period (minimum-

maximum) needed for all larvae within a treatment to go through the first and second

metamorphosis (Table 12) were affected by the dietary treatments. The first megalopae

appeared on DAH 15 in treatments 30/0.6 and 30/4/ARA. In the ‘starvation” control, the

first megalopae only appeared by DAH 18. For the first metamorphosis, a significant

difference (P < 0.01) in the mean metamorphosis time was found between the control and

treatments 30/0.6 and 30/4/ARA. On average, larvae in the control needed 1 to 2 days more

to metamorphose to megalopae compared to the other treatments. For the second

metamorphosis period, the average time to metamorphose to the first crab stage was

significantly shortest (P < 0.01) in treatments 30/4 and 30/4/ARA compared to treatment

50/0.6. In general, metamorphosis time, as well as its total duration was shorter for

treatments 30/4 and 30/4/ARA (24 days); intermediate values were obtained for treatment

30/0.6 (26 days) and the control and treatments 0/- and 50/0.6 had the longest

metamorphosis time (± 27 days).


102

Experiment 4

The first (MR1) and second (MR2) metamorphosis rates and its duration in

experiment 4 are shown in Tables 13 and 14 respectively. No statistical differences were

found between the treatments for both metamorphosis rates. The highest MR1 value was

found for treatment 4-4 while the lowest was in the “1-1” control treatment. Also the second

metamorphosis did not differ significantly between treatments. The MR2 value was

however highest in treatment 4-4, followed by treatments 0.6-0.6, 0.6-4 and the control. The

average first metamorphosis period of treatment 4-4 was significantly shorter (± 18 days)

compared to treatment 0.6-4 (± 20 days) (P < 0.01); while those of the “1-1” control and

treatment 0.6-0.6 were in between (± 19 days) (P < 0.01). The average time needed to

complete second metamorphosis was significantly longest for treatment 0.6-4 (± 27 days)

and shortest in the control treatment (± 25 days) (P < 0.01). Those of treatments 0.6-0.6 and

4-4 were intermediate (± 26 days).

First and second metamorphosis success are also illustrated in Figures 1 and 2. It can

be seen that first metamorphosis starts around DAH 16 in all treatments. From the normal

distribution curves it is obvious that first metamorphosis is centred around DAH 18 and 19

in treatments 4-4 and the “1-1” control, while it was protracted over more than one week in

treatments 0.6-0.6 and 0.6-4. Figure 2 shows that there were 2 peaks in the metamorphosis

to crab stage (around DAH 25 and 29) in treatment 0.6-4, while only 1 peak was found in

the other treatments.

In Figure 3, the cumulative metamorphosis rates from Z1 to megalopa and crab of the

different treatments is presented. From this it is obvious that treatment 4-4 outperforms all

other treatments (P < 0.01). The “1-1” control, was not different from treatment 0.6-4, but

significantly better then 0.6-0.6 (P < 0.05 and 0.01 for the first and second metamorphosis,

respectively).


103

3.5. Correlation between the fatty acid composition of the live feed and the crab larvae, and

larval development rate and metamorphosis success

LSI in relation to the fatty acid composition of the live feed

The significant correlation (P < 0.01) between the rotifer DHA/EPA ratios and the LSI

values on DAH 3 and 6 (r2 = 0.86 and 0.87, respectively) indicates that the DHA/EPA ratio

was the most important criterion in the dietary fatty acid composition for early-stage crab

larvae (Table 15).

The significant correlations between the DHA content (P < 0.01), total n-3 content (P

< 0.05) and DHA/EPA ratio (P < 0.01) of Artemia with the LSI values of crab larvae from

DAH 9 - 15 (r2 ranging from 0.77 to 0.98) proves that besides the DHA/EPA ratio, also the

absolute DHA and total n-3 content become more imperative for the crab larvae in the

Artemia-feeding stage.

The content of the other fatty acids (ARA and EPA) or ratios (ARA/EPA and Σn-

3/Σn-6) of the live feeds (rotifers and Artemia) were not correlated with the LSI values of

the crab larvae.

LSI in relation to the fatty acid composition of the crab larvae

Similar to the correlation between the fatty acid composition of the live feed with the

LSI, only the DHA/EPA ratio of the Z3 larvae was found significantly correlated with the

LSI values of crab larvae sampled on DAH 6 (r2 = 0.91, P < 0.01).

Besides the highly significant correlation between the DHA/EPA ratio of Z5 and the

LSI value on DAH 15 (r2 = 0.96, P < 0.01), also the larval DHA and total n-3 contents were

significantly correlated with the LSI value of the larvae (r2 = 0.94, P < 0.01 and 0.78, P <

0.05 respectively).

Metamorphosis success in relation to the fatty acid composition of the live feed and the crab

larvae

Following the same pattern as for the LSI, the DHA and total n-3 contents, and the

DHA/EPA ratio of Artemia (r2 ranging 0.8 to 0.92) and Z5 (r2 ranging from 0.69 to 0.93)


104

correlated significantly with the second metamorphosis rate (MR2). All correlations were

significant at P < 0.01, except P < 0.05 for total n-3.

For the first metamorphosis however, only the ARA content (of both Artemia and Z5)

and the ARA/EPA ratio (of Artemia) correlated significantly with the MR1 value (P < 0.05).

4. Discussion

4.1. Fatty acid composition of live feed and crab larvae

As was expected, the ICES emulsions influenced the ARA, EPA, DHA and total n-3

contents, and thus the relating DHA/EPA, ARA/EPA and Σn-3/Σn-6 ratios, of the live feed

significantly at P < 0.01 (treated by Pearson correlation from Tables 3 and 4, not shown in

the tables).

In general, the elevated HUFA content in the enriched live feed (rotifers and Artemia)

resulted in a significant increase of these fatty acids in the crab larvae. This confirms the

finding of previous studies that dietary HUFAs are readily assimilated by Scylla larvae

(Davis, 2003; Hamasaki et al., 1998; Kobayashi et al., 2000; Levine and Sulkin, 1984a;

Mann et al., 2001; Suprayudi et al., 2002b; Takeuchi et al., 1999; Takeuchi et al., 2000) and

on other aquaculture species including prawn, shrimp, fish and molluscs. For most fatty

acids however (except for the ARA/EPA ratio), a better correlation was found between the

composition of Artemia and Z5 compared to rotifers and Z3 (i.e. the ARA, DHA and total n-

3 contents were not significantly correlated for the latter). This probably indirectly results

from the different digestion and assimilation of HUFA in rotifers and Artemia. For example,

it has been well known that Artemia catabolize DHA much more than rotifers and thus the

former are considered to be much more difficult to “enrich” with this fatty acid (Bell et al.,

2001; Dhert et al., 1993; Navarro et al., 1999; Wouters et al., 1997). Furthermore, ARA,

EPA and DHA were the main HUFA’s present in both live feed and larvae; therefore any

changes in the level of each HUFA would certainly affect those of the others.

4.2. Survival in the zoeal stages

In experiments using large rearing tanks (30 and 100 l, in experiments 1 and 2,

respectively) without antibiotics, survival to Z5 was very low (1 to 19 %) and subjected to


105

large variations. In the small rearing containers (1-l bowls, in experiments 3 and 4) where

the water was exchanged daily and antibiotics were applied daily, the survival rates were

much higher (20 to 55 %) and only there differences in survival were detected. In

experiment 3, on DAH 15, a slightly lower (through not significant) survival was observed

for treatment 30/4 compared to the other treatments. In experiment 4 however, this

treatment (30/4) gave good survival on all sampling days. Similarly, in experiment 4 on

DAH 9, treatment 0.6-0.6 (or 30/0.6) had a significantly lower survival compared to the “1-

1” control. This was not however confirmed in any of the other experiments. We therefore

conclude that survival during the zoeal stages was not affected by the dietary treatments.

In experiment 3, live feed with the lowest total n-3 contents (4 - 5 and 9 - 11 mg g-1 in

rotifers and Artemia, in the “starvation” control and treatment 0/- respectively) overall

resulted in comparable survival rates in the zoeal stages compared to those in treatments

with higher total n-3 contents (26 - 49 and 45 - 65 mg g-1 in rotifers and Artemia,

respectively). Similarly on S. serrata, survival through the first metamorphosis was not

significantly different among treatments using live feed that contained a wide range of total

n-3 HUFA (9 - 18 and 2 - 54 mg g-1 in rotifers and Artemia, respectively) (Davis, 2003).

Mann et al. (2001) also found that feeding S. serrata larvae with Artemia nauplii enriched

with a commercial lipid booster did not significantly affect larval survival when the total n-3

level of the live feed in the different treatments varied from 3 to 84 mg g-1.

In this respect, it was observed in our study, that although the total n-3 level of rotifers

in the “starvation” control and treatment 0/- were very low, Z3 in both treatments could still

accumulate these n-3 HUFAs from their feed to obtain a level of 11 - 12 mg g-1, which is

only slightly lower than those in most other treatments (15 - 16 mg g-1, except for the high

level of 32 mg g-1 in treatment 50/0.6). Furthermore, the high initial total n-3 content in Z1

(22 mg g-1) together with trace amounts derived from the diet might be sufficient to

overcome deficiencies and maintain high survival up to the last zoea stage. For the red frog

crab Ranina ranina, it was found that dietary energy is utilized for survival first, molting

second and morphogenesis last (Minagawa, 1992). Similarly in fish, the effects of feeding

HUFA-deficient and HUFA-enriched Artemia nauplii are not always apparent in survival

and growth (Ashraf, 1993).

In the Artemia feeding stage however, Z5 larvae of the “starvation” control and

treatment 0/-, could not maintain their total n-3 content (9 and 13 mg g-1, respectively)

above that of the Artemia they were fed with (9 and 11 mg g-1, respectively). The same


106

observation could be made for other HUFAs. This means that, in contrast to the rotifer

feeding stages, the larvae fed with HUFA-deficient Artemia could not accumulate these

fatty acids. However, the survival rates in these later stages were, on most sampling days,

still not significantly different between treatments. This once again shows that larval

response to a nutritional factor does not come to expression immediately. Therefore,

although effects are maybe not immediately evident, all larval stages should be fed with

properly enriched live feed in order to build up sufficient HUFA reserves that are essential

for high survival beyond the zoeal stages. Takeuchi et al. (2000) similarly noticed that S.

paramamosain require both n-3 HUFA enriched rotifers and Artemia nauplii and should be

given the enriched Artemia nauplii from Z3 stage in order to attain a high survival rate of

the first crab stage.

4.3. Larval development rate during the zoeal stages

Larval development rate expressed as LSI was more affected by the dietary treatments

than survival rate. Anger et al. (1981) noticed that in brachyuran crabs, nutrition influences

development more directly than survival that is affected by a variety of other factors.

In most experiments, treatment 30/4 resulted in the highest larval development rate

among treatments. Enrichment media with a similar or higher total n-3 content, but lower

DHA/EPA ratio (treatments 30/0.6 and 50/0.6) usually performed significantly less.

Treatment 30/0.6 tended to have slightly higher LSI values than those of treatment

50/0.6 on most sampling days in experiment 1, but contradictory results were obtained in

experiments 2 and 3. This shows that high levels of HUFAs (particularly EPA as in

treatment 50/0.6) as such are neither needed nor beneficial for the larval performance. High

HUFA levels in crustacean larvae do not necessarily result in improved performance

(González-Félix et al., 2002) and the performance of S. serrata can even be compromised

when HUFA is supplied at excessive levels (Suprayudi et al., 2002b). In this study, an n-3

HUFA level of 30 % in the emulsions proved to be sufficient for live feed enrichment.

One of the suggested causes for the discrepancy in mud crab larval nutrition studies

has been the variability in quality between different batches of larvae as in S. paramamosain

(Djunaidah et al., 2003; Zeng and Li, 1999) and in S. serrata (Davis, 2003; Mann et al.,

1999a; Millamena and Bancaya, 2001). The larval quality was linked to the HUFA content

as was illustrated by Churchill (2003) in S. serrata larvae (particular EPA and total n-3).


107

Furthermore (Davis, 2003) showed that larvae containing different fatty acid profiles at

hatch may require different levels or ratios of certain essential fatty acids in the diet. Larval

quality is in turn affected by the changing egg quality relating to different HUFA contents

from different batches that has been found commonly in the wild-caught spawners

(Churchill, 2003; Djunaidah et al., 2003). Although the selected broodstock crabs were fed

the same diet, they spent different time in captivity. Hence, the nutritional status of female

mud crabs varied and consequently affected the quality of eggs and Z1 larvae (Davis, 2003).

Treatment 0/-, using a HUFA-deficient emulsion, produced low LSI values in all

experiments throughout the rearing period. This shows that live feed enrichment does not

improve larval development through merely supplying extra energy.

As the control treatment differed from one experiment to another, also the LSI values

of the control treatments in the four experiments were different. In experiment 1, where

control rotifers were grown on Chlorella, overall LSI values in the control treatment seemed

to be better than those of treatments 0/- and 50/0.6. The positive effects of micro-algae were

also evident in experiment 2. In this experiment, Artemia in the control treatment were

enriched with Chaetoceros prior to feeding to the crab larvae. Where initially, LSI values of

the “Chaetoceros” control treatment were similar to those of treatments 0/-, 50/0.6 and

30/0.6, by DAH 9 to 12 they became significantly higher than the latter and similar to those

in treatments 30/4 and 30/1/ARA. Enriching the Artemia with Chaetoceros probably

boosted the n-3 HUFA content of the Artemia and therefore probably increased larval

development rate in the later zoeal stages to a level similar to treatment 30/4. Chaetoceros

gracilis is after all known to contain high n-3 HUFA, and more specifically EPA levels

(Napolitano, 1990; Volkman et al., 1989). However, Chen and Jeng (1980) found that the

addition of Chlorella to the culture medium of S. serrata did not influence growth and

survival of the larvae. Although Brick (1974) reported a similar finding for the same mud

crab species up to the last zoeal stages, he observed an improved rate of larval

metamorphosis to the megalopa stage. This inconsistency in the micro-algal effect might be

caused by the variable content of n-3 HUFAs (mainly EPA and DHA) between algal species

and even from culture to culture within the same species (Olsen, 1989). As there was no

positive response of the S. serrata larvae to the addition of Nannochloropsis oculata and

Tahitian Isochrysis galpana, Mann et al. (2001) suspected that they did not improve the

larval nutrition. Johns et al. (1981) also found that enriching Artemia nauplii with Isochrysis

galbana did not positively influence the survival of Rhithropanopeus harrisii mud crab


108

larvae. Mann et al. (2001) suggested that the improvement may have been due to the water

conditioning effect of the micro-algae rather than nutritional influences. The algae may also

have delivered other essential nutrients (such as vitamins, polysaccharides, nucleotides,

etc.), which makes it difficult to define the exact factor involved.

In experiment 3, LSI values in the “starvation” control treatment were always lower

than those of the “HUFA-rich” treatments. In this trial rotifers were grown on baker’s yeast

and Artemia starved and thus especially deficient in HUFA.

In experiment 4, control live feed was “enriched” using the commercial product

Culture Selco, which contains considerable amounts of n-3 HUFA. The LSI values of the

“1-1” control treatment were always higher than those of treatment 0.6-0.6 (30/0.6). In early

larval stages, the “1-1” control also outperformed treatment 0.6-4, however LSI values

became similar towards the end of the rearing period. As the “1-1” control, which only has a

DHA/EPA ratio of 1, resulted in similar LSI values (and survival rates) compared to

treatment 4/4, it seems an emulsion with a total HUFA content of approximately 30 % with

a DHA/EPA value of 1 is sufficient to satisfy the requirements for larval development in the

zoeal early stages as the effects of dietary HUFAs would manifest in the metamorphosis

stages.

The importance of DHA and its ratio to EPA for larval development was also obvious

from the correlation coefficients between the LSI and the fatty acid profile of the live feed.

The LSI values on DAH 3 and 6 were only significantly correlated with the DHA/EPA level

of the rotifers, and not with the DHA or the total n-3 level. In contrast, during the Artemia

feeding stage, also the absolute DHA level and total n-3 level significantly influenced larval

development. This difference can probably be explained by the difference in enrichment

kinetics between rotifers and Artemia. Artemia is known to selectively catabolize DHA

during enrichment, resulting in relatively lower total DHA levels (Bell et al., 2001; Dhert et

al., 1993; Navarro et al., 1999; Wouters et al., 1997). This could explain why also the

absolute DHA level (and hence total n-3 level) becomes more important during Artemia

feeding.

Similarly, growth was also significantly correlated with the fatty acid composition of

the Z3 and Z5. Again growth during the rotifer feeding stage was only correlated with the

DHA/EPA ratio of the larvae, while in the Artemia-stage Z5 also the DHA and total n-3

content became important.


109

4.4. Metamorphosis

Metamorphosis rate

In experiment 3, the first metamorphosis rate (from Z5 to megalopa) of treatment

30/4/ARA was significantly higher (50 %) then in the “n-3 HUFA treatments” (ranging

from 30 to 34 %). No differences were detected between any of the other treatments. This

was confirmed by the fact that the ARA content of both Artemia and Z5 and the ARA/EPA

ratio of Artemia correlated significantly with the MR1 values.

The role of arachidonic acid in first metamorphosis remains unclear and requires

further investigation. Koven et al. (2000) suggested that besides DHA, not only highly

unsaturated fatty acids of the n-3 series are important but that also ARA may play a

significant role for the larvae of gilthead seabream (Sparus aurata). ARA may improve

larval growth and pigmentation in several marine fish species since it provides precursors

for eicosanoid production (Castell et al., 1994; Estévez et al., 1999). Turbot fed ARA as the

only HUFA yielded higher growth and survival compared to any of the DHA/ARA mixtures

or DHA alone (Castell et al., 1994). It is likely that, at specific stages in the life cycle of

fish, higher levels of ARA may be required to cope with periods of environmental stress

(Bell and Sargent, 2002). The requirement of ARA in fish, however, seems to depend on the

fish species and larval development, and needs to be dosed with extreme care since it may

act in a different way depending on the DHA concentration (Castell et al., 1994; Koven et

al., 2000). Studies on arachidonic acid in crustaceans are rather scarce. Glencross and Smith

(2001) investigated arachidonic acid requirements in Penaeus monodon. They concluded

that ARA is not really essential if linoleic acid (18:2 n-6), the natural precursor to ARA, is

present in sufficient amounts. These authors suggested that possible effects of ARA may lie

in the importance of the balance of n-3 to n-6 fatty acids in the diet.

Also in experiment 4, the MR1 values of the different n-3 HUFA enrichment

treatments were not significantly different. It therefore seems that the various total n-3

HUFA levels (28 - 50 %) and DHA/EPA ratios (0.6 - 4) tested in this experiment could

satisfy the requirements for first metamorphosis. There might however also be an interaction

between metamorphosis success, survival in the zoeal stages and the quality of the surviving

larvae. As in experiments 3 and 4, survival in the zoeal stages was affected by the dietary

treatments, it could be hypothesized that in the “bad” treatments the weaker larvae had died


110

before reaching Z5 stage and the survivors were capable to metamorphose successfully,

whereas in the assumed “better” treatments, also weaker larvae could reach the Z5 stage but

then failed to metamorphose to megalopae. This might have leveled out metamorphosis

results.

The effects of dietary HUFA on the second metamorphosis rate were more

pronounced. The effects of dietary HUFA often become obvious in later stages only when

the larvae have used up all reserves built up in earlier stages. In this respect, it has been

reported that n-3 HUFA are among the last components to be utilized (Galois, 1987; cited in

Wouters et al., 1997), probably because they play an important role as fatty acid groups of

polar lipids in cell membranes (Sargent et al., 1991; Watanabe, 1993).

In experiment 3, three groups could be distinguished on the basis of their MR2 rates:

high survival through second metamorphosis for the treatments with a high DHA/EPA ratio

(30/4 and 30/4/ARA, over 90 %), intermediate survival for treatments with a low DHA/EPA

ratio (50/0.6 and 30/0.6, approximately 70 %) and low survival for treatments with low total

n-3 (or total HUFA) content (“starvation” control and 0/-, approximately 10 %),

respectively. Following the same pattern as for larval development rate in the zoeal stages,

the DHA and total n-3 content, and the DHA/EPA ratio of the Artemia and Z5 were hence

significantly correlated with the second metamorphosis rate. Takeuchi et al. (2000) found

that S. paramamosain larvae require both n-3 HUFA enriched rotifers and Artemia and that

it is necessary to feed enriched Artemia nauplii from the Z3 stage onwards in order to attain

a high survival rate to first crab. Similarly, for S. tranquebarica, a high survival rate and

maximum carapace width of the first crab stage was obtained when the larvae were fed

DHA-enriched Artemia (Takeuchi, in press).

In experiment 4, no real negative control (low n-3 HUFA) was included. MR2 values

were hence rather similar in all treatments. Again the MR2 values of the four treatments

might have been leveled out by the survival in the preceding stages and the quality of the

surviving larvae as explained in the discussion of MR1 values.

Timing and duration of metamorphosis

Onset of metamorphosis was largely dictated by the larval development rate during

the zoeal stages. In experiment 3, the treatments using the emulsions containing a medium

total n-3 content (30/0.6, 30/4 and 30/4/ARA) reached first metamorphosis earlier and had a


111

shorter metamorphosis period than the treatments with high (50/0.6) or low (control and 0/-)

total n-3 HUFA content. No difference was observed however between high and low

DHA/EPA treatments. A similar trend was observed for the second metamorphosis.

In experiment 4, the average first metamorphosis time was shortest for treatment 4-4;

intermediate for treatments control and 0.6-0.6; and significantly longer for treatment 0.6-4.

The histograms and normal distribution curves of the first metamorphosis clearly

demonstrate that the duration of metamorphosis was prolonged in treatments 0.6-0.6 and

0.6-4. The average time to reach the second metamorphosis was longest for treatment 0.6-4,

shortest in the “1-1” control treatment and intermediate values for treatments 0.6-0.6 and 4-

4. Again, the survival rate could have affected the completion of metamorphosis processes

as the assumed “best” treatment (i.e. treatment 4-4) had many more surviving larvae, what

probably resulted in a wider variation in metamorphosis time in contrast to the assumed

“worst” treatment (i.e. treatment 0.6-0.6) that had few but “strong” survivors.

Figure 1 shows that in treatment 0.6-4 a number of megalopae appeared very late on

DAH 26 and 28. This could point out a kind of recovery after switching to an emulsion with

higher DHA/EPA ratio from Z3 onwards in this treatment. Zeng (1998) observed that crab

larvae that were put on a too low ration, prolonged the zoeal stage by developing a sixth

zoeal stage before metamorphosing into megalopa. The author suggested that these larvae

adopted a strategy of extending zoeal development to accumulate enough nutritional and

energy reserves necessary for this metamorphosis. Figure 2 also shows 2 peaks of

metamorphosis to crab stage in the treatment 0.6-4 while only 1 peak was found in the other

treatments. This could be linked to the late first metamorphosis of some megalopae as

presented in Figure 1.

4.5. Survival of Z1 to megalopa and the first crab (M/Z1 and C1/Z1 survival rates)

In our study, dietary HUFAs did not always bring an instant response on larval

performance. However, due to the accumulation of responses there usually is an inflection

point where larval survival or growth diverges between different treatments (Mann et al.,

2001). Using Penaeus stylirostris as a test organism, it appeared that the EPA and DHA

content in the zoea diet only showed a major impact on survival and growth in later stages,

when animals had already been switched to another diet (Léger et al., 1985). The effects of

early nutrition are often only manifested during the later stages of development and


112

particularly at metamorphosis to the post larval stages (megalopa and crabs) (Davis, 2003;

Harvey, 1996; Jeff et al., 1996; Ribeiro and Jones, 2000; Sulkin, 1978). Therefore, the

overall survival of Z1 to megalopa or C1 that are the product of zoeal survival and

metamorphosis rate are a good means to evaluate the final effect of treatments. These

overall survival rates are also the primary concern for hatchey managers. In this respect, the

M/Z1 and C1/Z1 survival rates are presented in Figure 3. This figure clearly shows that both

overall survival rates were significantly higher in treatment 4-4. The higher M/Z1 and

C1/Z1 survival rates in treatment 0.6-4 compared to those of treatment 0.6-0.6 proves that

the crab larvae can recover their growth at later stages when DHA-rich Artemia was offered

from the Z3 stage onwards. However, survival rates to M and C1 in treatment 0.6-4 were

still lower than those of treatment (4-4) and the “1-1” control. This proves that an emulsion

with DHA/EPA level lower than 1 should not be used for live feed enrichment for early

stages.

Several other studies used overall survival rates for evaluating the effect of live food

enrichment. Mann et al. (2001) found that feeding the S. serrata larvae with only Artemia

enriched with the commercial lipid booster (Super Selco®, INVE Aquaculture) did not affect

larval survival or growth to DAH 18 (megalopa stage). This commercial lipid booster

increased the DHA/EPA ratio of Artemia from 0 - 0.04 (in un-enriched Artemia) to 0.3 - 0.4,

which is similar to the DHA/EPA ratio of Artemia enriched with the emulsion 30/0.6 in

experiment 3 of our study (= 0.4). These authors concluded that the requirement for EPA,

DHA and other fatty acid are met in un-enriched newly-hatched Artemia nauplii. Our study

however indicates that the low DHA/EPA levels (0.3 - 0.4) in enriched Artemia in their

experiment were not sufficient to improve the larval performance when Artemia is the sole

feed for all crab larval stages. This is confirmed by a study on S. serrata by Davis et al.

(2003). In that study, the zoeal survival and development, first metamorphosis rate and

overall survival of Z1 to megalopa in treatments using enriched rotifers and enriched/un-

enriched Artemia (with Super Selco®) were significantly improved compared to the

treatments without rotifer enrichment or to treatments where only enriched/un-enriched

Artemia were fed (Davis, 2003). Boosting FAME levels (including HUFAs) in the rotifers

fed at hatch may have given the larvae a significant advantage which was carried through

the rearing process and culminated in improved survival through metamorphosis,

particularly when both rotifers and Artemia were enriched and the HUFA supply was thus

not uninterrupted (Davis, 2003).


113

4.6. Use of antibiotics for studying nutritional requirements Although we don’t want to encourage the use of antibiotics for commercial farming

practices, the application of antibiotics (experiments 3 and 4) as a prophylactic improved

overall survival considerably and made results more reproducible. After all, it should not be

overlooked that there are various interactions possible between pathogenic bateria and

nutritional treatments. For example, although the enriched live food was thouroughly rinsed

before being fed to the larvae, it can not be excluded that residual oil enters the system this

way. Moreover, the decomposition of uneaten enriched live feed might be a source of extra

nutrients and this way supporting the development of bacteria. In this respect, it was noticed

that treatments using emulsions with a high n-3 HUFA percentage (50/0.6) or high

DHA/EPA ratio (30/4) tended to give lower survival compared to treatments 0/- and 30/0.6

and the control towards the end of experiment 1. It might be that the use of emulsions with a

high n-3 HUFA level resulted in more waste material being produced higher accumulation

of fatty acids sourced from decomposed waste products and therefore deteriorated the

culture medium more rapidly. Possibly, HUFA’s with their higher number of carbon atoms

and many more double bonds might also be a more readily available substrate for bacterial

growth (Mead et al., 1986). Conversely, it is also known that short chain saturated fatty

acids (which are relatively abundant in the coconut oil-based ICES 0/-) have bacteriostatic

properties (Rickle, 2003). In experiments 1 and 2, the nutritional benefits arising from

feeding enriched live feed might thus have been leveled out by a negative microbial

interaction, where in experiment 3 and 4 the daily application of antibiotics should have

separated out the purely nutritional effects.


The ICES emulsions influenced the fatty acid profiles of the live feed and, in turn, the

fatty acid profiles of the crab larvae.

No real differences in survival in the zoeal stages were found between the different

enrichment treatments tested here. The “nutritional impact” of HUFAs on the zoeal survival

was probably obscured by other more decisive factors such as the batch quality, micro-biota,

zootechnics. The significantly lower metamorphosis success in the low HUFA treatments

proved however that HUFA-rich live food enrichment emulsions are needed to attain high


114

survival to the crab stage. Not that much the total n-3 level, but more particularly the DHA

level and the DHA/EPA ratio seem of crucial importance.

Also larval development rate was very much affected by the dietary n-3 HUFA level

and its DHA/EPA ratio. In all experiments, the LSI values of treatments 30/4 tended to be

higher than those of treatments 50/0.6 and 30/0.6 throughout the zoeal stages. This means

that a DHA/EPA ratio of 0.6 in enrichment emulsions is not sufficient to support larval

development. Based on the results of experiment 4, a DHA/EPA ratio of approximately 1

might be sufficient for early stages (Z1 - Z2).

There was also evidence that DHA/EPA requirements might change during

development. Therefore it might be better if this ratio is increased gradually in time (e.g.

emulsions with DHA/EPA ratio of 1 for Z1-Z2 stages, of 2 - 3 for Z3 - Z4 and of 4 for Z5

onwards). Exact requirements for each larval stage should therefore be investigated in

function of the diet type (rotifers, Artemia, micro-bound diets). This should however be

verified using more precisely formulated emulsions.

For the total n-3 HUFA content, no differences were found between live feed

enrichment products containing 30 or 50 % n-3 HUFA. An inclusion level of 30 % therefore

seems to be sufficient.

Supplementation of arachidonic acid had no effect on survival nor growth during the

zoeal stages. First metamorphosis rate was however improved by the addition of dietary

ARA. Further research on the suitable levels of ARA in the enrichment diet for crab larvae

is therefore worth pursuing.

Acknowledgements




115

Table 1 Overview of larval rearing conditions and water quality parameters in the 4 experiments

Container Water quality parameters Experi-ment

Culture period (days)


Volume (l) Type

Water exchange Temp.

(°C) Salinity (g l-1)

NH4+

(mg l-1) NO2

-

(mg l-1) NO3

+

(mg l-1)

1 15 50 30 Tank Recirculation 29.6 ±0.4

29.6 ±1.1

0.08 ±0.1

0.05 ±0.02

3.0 ±2.0

2 12 100 100 Tank Recirculation 31.1 ±0.7

30.1 ±0.5

0.13 ±0.1

0.26 ±0.28 -

29 50 1 Bowl Batch 3 15 50 100 Tank Recirculation 4 31 100 1 Bowl Batch

30.0 ±1.0

30.0 ±1.0 < 1.0 < 0.3 -

Table 2 Overview of the treatments used in the 4 experiments. In experiment 1, rotifers were the only live feed for all larval stages. In the remaining experiments, rotifers were fed to the crab larvae from DAH 0 - 6 (Z1 - Z2) and Artemia were replaced rotifers from the evening of DAH 6 (onset of Z3 stage)

Treatment Experi-ment Control 0/- 50/0.6 30/0.6

(0.6-0.6)§ 30/4 (4-4)§ 0.6-4 30/1/

ARA 30/4/ ARA

Number of repli-cates

1

freshly harvested un-enriched rotifers, cultured on Chlorella

rotifers enriched with emulsion 0/-

rotifers enriched with emulsion 50/0.6

rotifers enriched with emulsion 30/0.6

rotifers enriched with emulsion 30/4

4

2

un-enriched rotifers and Artemia enriched with Chaetoceros

rotifers and Artemia enriched with emulsion 0/-

rotifers and Artemia enriched with emulsion 50/0.6


rotifers and Artemia enriched with emulsion 30/4

rotifers and Artemia enriched with emulsion 30/1/ARA

3

3

un-enriched rotifers and starved Artemia

rotifers and Artemia enriched with emulsion 0/-




rotifers and Artemia enriched with emulsion 30/4/ARA

4 in bowls 2 in tanks

4

rotifers and Artemia enriched with Culture Selco



rotifers enriched with emulsion 30/0.6 and Artemia enriched with emulsion 30/4

3

§ = Treatment names used in experiment 4.


116

Table 3 Fatty acid composition of the ICES emulsions used to enrich the live feed in the 4 experiments. ARA = Arachidonic acid, EPA = Eicosapentaenoic acid, DHA = Docosahexaenoic acid, Σn-3 = Total n-3 HUFA, Σn-6 = Total n-6 HUFA, ΣHUFA = Total HUFA, FAME = Fatty acid methyl esters = Total fatty acid content ICES emulsion Fatty acid content

0/- 50/0.6 30/0.6 30/4 30/1/ARA 30/4/ARA

ARA (mg g-1) 0.00 10.40 5.25 4.71 124.70 94.51 EPA (mg g-1) 0.00 261.40 103.70 42.13 125.96 31.58 DHA (mg g-1) 0.00 163.60 63.73 161.87 122.60 121.46 Σn-3 (mg g-1) 0.00 499.80 167.43 204.00 248.56 156.19 Σn-6 (mg g-1) 2.88 59.50 36.57 50.91 207.79 159.93 ΣHUFA (mg g-1) 0.00 524.40 172.68 208.72 373.26 251.17 DHA/EPA - 0.63 0.61 3.84 0.97 3.85 ARA/EPA - 0.04 0.05 0.11 0.99 2.99 Σn-3/Σn-6 0.00 8.40 4.58 4.01 1.20 0.98 Σn-3/FAME (%) 0.00 51.41 29.29 30.48 28.40 22.34 Table 4 Fatty acid composition of rotifers and Artemia used in experiment 3. ARA = Arachidonic acid, EPA = Eicosapentaenoic acid, DHA = Docosahexaenoic acid, Σn-3 = Total n-3 HUFA, Σn-6 = Total n-6 HUFA, ΣHUFA = Total HUFA. Con. = Control

Rotifers Artemia

Fatty acid content Con. 0/- 50/0.6 30/0.6 30/4 30/4/ ARA Con. 0/- 50/0.6 30/0.6 30/4 30/4/

ARA ARA (mg g-1) 1.08 0.75 2.15 1.65 1.03 7.13 3.86 4.16 6.50 2.50 8.34 25.77 EPA (mg g-1) 2.33 2.65 28.61 18.91 7.19 11.26 8.94 10.49 40.09 29.60 23.29 27.49 DHA (mg g-1) 1.94 2.10 19.99 13.34 18.45 23.61 0.35 0.21 19.53 11.20 20.61 19.07 Σn-3 (mg g-1) 4.27 4.76 48.60 32.25 25.64 34.88 9.49 11.05 64.99 45.10 46.64 48.79 Σn-6 (mg g-1) 7.05 14.96 17.17 12.38 11.36 23.15 7.67 14.56 16.49 14.10 22.82 47.72 ΣHUFA (mg g-1) 5.35 5.51 50.75 33.90 26.67 42.01 14.00 16.02 73.51 50.60 58.00 77.96 DHA/EPA 0.84 0.79 0.70 0.71 2.57 2.10 0.04 0.02 0.49 0.38 0.88 0.69 ARA/EPA 0.46 0.28 0.08 0.09 0.14 0.63 0.43 0.40 0.16 0.08 0.36 0.94 Σn-3/Σn-6 0.61 0.32 2.83 2.61 2.26 1.51 1.24 0.76 3.94 3.20 2.04 1.02


117

Table 5 Fatty acid composition of S. paramamosain zoea 1 (= newly-hatched zoea), zoea 3 and zoea 5 fed different enriched live feed in experiment 3. ARA = Arachidonic acid, EPA = Eicosapentaenoic acid, DHA = Docosahexaenoic acid, Σn-3 = Total n-3 HUFA, Σn-6 = Total n-6 HUFA, ΣHUFA = Total HUFA, Con. = Control

Zoea 3 Zoea 5

Fatty acid content

Zoea 1

Con. 0/- 50/0.6 30/0.6 30/4 30/4/ARA Con. 0/- 50/0.6 30/0.6 30/4 30/4/

ARA ARA (mg g-1) 6.59 4.56 2.73 4.03 1.84 2.27 4.93 3.77 4.74 4.41 3.70 4.73 6.76 EPA (mg g-1) 12.13 7.43 7.29 18.78 8.76 6.33 6.84 8.17 10.92 16.79 12.12 11.44 12.66 DHA (mg g-1) 9.70 4.03 4.32 13.07 6.77 8.20 8.00 0.87 1.78 6.21 4.34 6.78 6.84 Σn-3 (mg g-1) 21.83 11.46 11.61 31.85 15.53 14.53 14.84 9.04 13.25 23.00 16.46 19.07 19.50 Σn-6 (mg g-1) 11.47 15.50 12.59 14.22 6.20 5.54 11.00 6.10 15.54 9.86 7.89 11.55 12.80 ΣHUFA (mg g-1)

28.42 16.02 14.33 35.88 17.36 16.79 19.77 12.81 19.18 27.41 20.15 24.88 26.25

DHA/EPA 0.80 0.54 0.59 0.70 0.77 1.29 1.17 0.11 0.16 0.37 0.36 0.59 0.54 ARA/EPA 0.54 0.61 0.37 0.21 0.21 0.36 0.72 0.46 0.43 0.26 0.30 0.41 0.53 Σn-3/Σn-6 1.90 0.74 0.92 2.24 2.50 2.62 1.35 1.48 0.85 2.33 2.09 1.65 1.52

Table 6. Pearson correlation coefficients (r2) between the fatty acid composition of the live feeds and those of the crab larvae. ARA = Arachidonic acid. EPA = Eicosapentaenoic acid. DHA = Docosahexaenoic acid, Σn-3 = Total n-3 HUFA, Σn-6 = Total n-6 HUFA

Fatty acid Correlation

ARA EPA DHA Σn-3 DHA/EPA ARA/EPA Σn-3/Σn-6

Rotifers - Zoea 3 0.36 0.72* 0.62 0.65 0.90** 0.96** 0.86** Artemia - Zoea 5 0.91** 0.83* 0.98** 0.91** 0.97** 0.81* 0.86** * and ** = significant correlations (P < 0.05 and P < 0.01, respectively). Table 7 Survival rates and larval stage index (LSI) values (mean ± standard deviation) of S. paramamosain larvae fed different enriched rotifers in experiment 1. For treatment descriptions refer to Tables 2 and 3. DAH = days after hatch Treatment DAH 3 DAH 6 DAH 9 DAH 12 DAH 15 Survival rate (%) Control 82±22a 67±38a 48±34a 42±30a 18±13a

0/- 99±0a 82±8a 53±40a 25±17a 19±17a

50/0.6 92±10a 59±26a 44±14a 21±6a 10±9a

30/0.6 96±8a 79±18a 71±19a 23±2a 15±6a

30/4 85±31a 74±24a 63±20a 29±17a 9±7a

LSI value Control 2.0±0.1a 3.0±0.0a A 3.8±0.3a A 4.1±0.1ab A 4.3±0.1a

0/- 2.0±0.1a 2.2±0.4b B 2.8±0.3b A 3.3±0.3c B 3.8±0.3a

50/0.6 1.8±0.3a 2.7±0.1ab AB 3.1±0.5ab A 3.8±0.3b AB 4.0±0.0a

30/0.6 2.0±0.0a 2.9±0.1a A 3.7±0.4a A 4.1±0.1ab A 4.2±0.3a

30/4 2.0±0.1a 2.9±0.2a A 3.6±0.3a A 4.3±0.1a A 4.8±0.0a

Survival rates or LSI values in the same column followed by the same superscript letter are not statistically different (P ≥ 0.05, regular letters and P ≥ 0.01, capital letters.


118

Table 8 Survival rates and larval stage index (LSI) values (mean ± standard deviation) of S. paramamosain larvae fed different enriched rotifers and Artemia in experiment 2. For treatment descriptions refer to Tables 2 and 3. DAH = days after hatch Treatment DAH 3 DAH 6 DAH 9 DAH 12 Survival rates (%) Control 66±17a 28±17a 5±4a 1±2a

0/- 64±26a 39±18a 22±10a 11±7a

50/0.6 74±28a 21±8a 15±11a 5±5a

30/0.6 62±18a 35±10a 17±1a 10±3a

30/4 70±12a 26±13a 11±10a 7±5a

30/1/ARA 79±9a 39±8a 11±9a 7±6a

LSI Control 1.7±0.2a 2.4±0.2cd BC 3.5±0.2bc BC 4.7±0.1a A

0/- 1.7±0.1a 2.5±0.1bc BC 3.2±0.1cd C 4.3±0.1b B

50/0.6 1.7±0.2a 2.4±0.2cd BC 3.1±0.1cd C 4.3±0.1b B

30/0.6 1.6±0.4a 2.2±0.2d C 3.1±0.1d C 4.1±0.1b B

30/4 1.8±0.2a 2.9±0.1a A 4.0±0.1a A 4.9±0.1a A

30/1/ARA 1.8±0.2a 2.8±0.2ab AB 3.7±0.2ab AB 4.8±0.1a A

Survival rates or LSI values in the same column followed by the same superscript letter are not statistically different (P ≥ 0.05, regular letters and P ≥ 0.01, capital letters. Table 9 Survival rates and larval stage index (LSI) values (mean ± standard deviation) of S. paramamosain larvae fed different enriched rotifers and Artemia in experiment 3. For treatment descriptions refer to Tables 2 and 3. DAH = days after hatch Treatment DAH 3 DAH 6 DAH 9 DAH 12 DAH 15 Survival rates (%) Control 96±2a 92±5a 88±4a 64±5a 28±11a

0/- 94±5a 89±8a 83±10a 59±7a 34±7a

50/0.6 95±4a 92±4a 85±8a 63±15a 35±14a

30/0.6 93±3a 86±5a 80±4a 67±7a 43±6a

30/4 96±4a 87±5a 85±6a 56±17a 20±10a

30/4/ARA 96±3a 89±6a 81±3a 65±13a 41±11a

LSI Control 1.7±0.1b B 2.2±0.1b C 3.0±0.1c C 3.5±0.1c B 4.4±0.1c C

0/- 1.6±0.0c B 2.0±0.1b C 2.9±0.1c C 3.4±0.1c B 4.4±0.1c BC

50/0.6 1.7±0.0b B 2.2±0.1b BC 3.3±0.1ab AB 3.8±0.1ab A 4.7±0.1ab A

30/0.6 1.6±0.1bc B 2.2±0.1b BC 3.2±0.1b B 3.7±0.1b A 4.6±0.1b AB

30/4 1.9±0.0a A 2.6±0.2a A 3.4±0.1ab AB 3.9±0.1a A 4.8±0.1a A

30/4/ARA 1.9±0.0a A 2.5±0.1a AB 3.3±0.1a A 3.8±0.1ab A 4.8±0.3ab A

Survival rates or LSI values in the same column followed by the same superscript letter are not statistically different (P ≥ 0.05, regular letters and P ≥ 0.01, capital letters).


119

Table 10 Survival rates and larval stage index (LSI) values (mean ± standard deviation) of S. paramamosain larvae fed different enriched rotifers and Artemia in experiment 4. For treatment descriptions refer to Tables 2 and 3. DAH = days after hatch Treatment DAH 3 DAH 6 DAH 9 DAH 12 DAH 15 Survival rate (%) Control 87±6a 80±6a 76±5a A 66±7a 55±12a

0.6-0.6 81±7a 64±17a 28±24b A 26±25a 22±24a

0.6-4 83±3a 63±13a 49±16ab A 45±15a 39±12a

4-4 83±12a 76±14a 71±16ab A 63±17a 50±18a

LSI Control 1.9±0.1a A 2.7±0.4a 3.4±0.1a 4.0±0.1a 4.4±0.2a

0.6-0.6 1.0±0.0b B 2.1±0.1a 2.3±0.5a 3.8±0.4a 4.5±0.2a

0.6-4 1.0±0.0b B 2.1±0.1a 3.2±0.2a 4.0±0.2a 4.2±0.3a

4-4 1.7±0.5a AB 2.6±0.4a 3.6±0.3a 4.1±0.2a 4.5±0.4a

Survival rates or LSI values in the same column followed by the same superscript letter are not statistically different (P ≥ 0.05, regular letters and P ≥ 0.01, capital letters). Table 11 Survival through first (M/Z5, MR1) and second (C1/M, MR2) metamorphosis rates of S. paramamosain larvae fed different enriched rotifers and Artemia in experiment 3. For treatment names refer to Tables 2 and 3. Z5 = zoea 5, M = megalopa, C1 = crab 1 Treatment Survival through first

metamorphosis (M/Z5) (%)

Survival through second metamorphosis (C1/M) (%)

Control 31b A 13b B

0/- 32b A 11b B

50/0.6 30b A 73a A

30/0.6 34b A 73a A

30/4 30b A 92a A

30/4/ARA 50a A 93a A

Values in the same column followed by the same superscript letter are not statistically different (P ≥ 0.05, regular letters and P ≥ 0.01, capital letters). Table 12 Average and minimum and maximum time (days after hatch) to reach first and second metamorphosis of S. paramamosain larvae fed different enriched live feed in experiment 3. For treatment descriptions refer to Tables 2 and 3

First metamorphosis Second metamorphosis Treatment mean ±

standard deviation minimum - maximum

mean ± standard deviation

minimum - maximum

Control 19.3±1.6a A 18-22 27.0±1.4ab AB 26-28 0/- 18.2±1.5ab AB 17-22 27.5±2.1ab AB 26-29 50/0.6 18.3±1.5ab AB 16-23 27.0±1.6a A 24-29 30/0.6 17.1±1.0b B 15-19 25.9±1.2ab AB 24-28 30/4 17.8±1.6ab AB 16-21 24.3±1.4b B 22-27 30/4/ARA 17.3±1.4b B 15-21 24.8±1.4b B 22-27 Values in the same column followed by the same superscript letter are not statistically different (P ≥ 0.05, regular letters and P ≥ 0.01, capital letters).


120

Table 13 Survival through first (M/Z5, MR1) and second (C1/M, MR2) metamorphosis of S. paramamosain larvae fed different enriched rotifers and Artemia in experiment 4. For treatment names refer to Tables 2 and 3. Z5 = zoea 5, M = megalopa, C1 = crab 1

Treatment Survival through first metamorphosis (M/Z5) (%)

Survival through second metamorphosis (C1/M) (%)

Control 48±9a 67±20a

0.6-0.6 65±14a 78±24a

0.6-4 50±16a 67±20a

4-4 66±20a 81±3a

Values in the same column followed by the same superscript letter are not statistically different (P ≥ 0.05). Table 14 Average time (days after hatch) to reach first and second metamorphosis of S. paramamosain larvae fed different enriched live feed in experiment 4 (mean ± standard deviation). For treatment descriptions refer to Tables 2 and 3 Treatment First metamorphosis Second metamorphosis Control 19.3±1.7ab AB 25.7±2.3b B

0.6-0.6 19.4±2.1ab AB 26.6±2.1ab AB

0.6-4 20.1±2.5a A 27.1±2.0a A

4-4 18.6±1.5b B 26.5±2.0ab AB

Values in the same column followed by the same superscript letter are not statistically different (P ≥ 0.05, regular letters and P ≥ 0.01, capital letters). Table 15 Pearson correlation coefficients (r2) of the fatty acid composition of the live feed (rotifers and Artemia) and crab larvae (zoea 3 and zoea 5) with larval stage index (LSI) values and metamorphosis rates (MR, %) in experiment 3. DAH = days after hatch, MR1 = survival through first metamorphosis (M/Z5), MR2 = survival through second metamorphosis (C1/M), ARA = Arachidonic acid, EPA = Eicosapentaenoic acid, DHA = Docosahexaenoic acid, ΣHUFA = Total highly unsaturated fatty acid, Σn-3 = Total n-3 HUFA, Σn-6 = Total n-6 HUFA, DAH = days after hatch Fatty acid Correlation

ARA EPA DHA Σn-3 DHA/EPA ARA/EPA Σn-3/Σn-6

Rotifers – LSI DAH 3 0.30 0.00 0.52 0.13 0.86** 0.10 0.08 Rotifers – LSI DAH 6 0.18 0.00 0.56 0.17 0.87** 0.01 0.19 Artemia – LSI DAH 9 0.20 0.58 0.97** 0.79* 0.94** 0.00 0.24 Artemia – LSI DAH 12 0.28 0.60 0.98** 0.81* 0.94** 0.02 0.21 Artemia – LSI DAH 15 0.32 0.57 0.95** 0.77* 0.93** 0.03 0.15 Zoea 3 – LSI DAH 6 0.00 0.05 0.14 0.00 0.91** 0.07 0.25 Zoea 5 – LSI DAH 15 0.28 0.39 0.98** 0.78* 0.96** 0.02 0.24 Artemia - MR1 0.83* 0.02 0.09 0.04 0.09 0.69* 0.15 Artemia - MR2 0.30 0.61 0.92** 0.80* 0.90** 0.02 0.20 Zoea 5 - MR1 0.73* 0.00 0.14 0.04 0.17 0.36 0.02 Zoea 5 - MR2 0.20 0.33 0.93** 0.69* 0.92** 0.02 0.35 * and ** = significant correlations (P < 0.05 and P < 0.01, respectively).


121

16 17 18 19 20 21 22 23 24 25 26 27 28

Day after hatch

0

5

10

15

20

25

30

35

40

45N

o of

obs

erva

tions 4-4

"1-1" control

0.6-4

0.6-0.6

"1-1" control 0.6-0.6 0.6-4 4-4

Figure 1. Histogram and normal distribution curves of the metamorphosis to megalopa of S. paramamosain larvae fed different enriched live feed in experiment 4. For treatment descriptions refer to Tables 2 and 3.


122

22 23 24 25 26 27 28 29 30 31

Day after hatch

0

2

4

6

8

10

12

14

16

18

20

No

of o

bser

vatio

ns

4-4

0.6-4

0.6-0.6

"1-1" control

"1-1" control 0.6-0.6 0.6-4 4-4

Figure 2. Histogram and normal distribution curves of metamorphosis to crab 1 of S. paramamosain larvae fed different enriched live feed in experiment 4. For treatment descriptions refer to Tables 2 and 3.


123

0

5

10

15

20

25

30

351

16 17 18 19 20 21 22 23 22 23 24 25 26 27 28 29 30 31

Day after hatch

Cum

ulat

ive

% o

f M/Z

1 an

d C

1/Z

"1-1" control0.6-0.60.6-44-4

b AB

c B

bc B

a A

b B

b BC

c C

a A

Megalopa Crab 1

Figure 3. Cumulative metamorphosis rates from zoea 1 to megalopa and first crabs of S. paramamosain larvae fed different enriched live feed in experiment 4. For treatment descriptions refer to Tables 2 and 3. Curves for each survival followed by the same letter are not statistically different (P ≥ 0.05, regular letters and P ≥ 0.01, capital letters).

CHAPTER 6 Improved larval rearing techniques for mud crab

(Scylla paramamosain)



Abstract Eight larval rearing trials were carried out with the purpose to develop optimal rearing

techniques for the mud crab (Scylla paramamosain). Based on the method of water exchange (discontinuous partial water renewal or

continuous treatment through biofiltration) and micro-algae (Chlorella or Chaetoceros) supplementation (daily supplementation with low levels of 0.1 - 0.2 million cells ml-1 or maintenance at high levels of 1 - 2 millions cells ml-1), six different types of rearing systems were tried.

The combination of a green-water batch system for early stages and a recirculating system with micro-algae supplementation for later stages resulted in the best overall performance of the crab larvae.

A stocking density of 100 Z1 l-1 combined with a rotifer density of 45 ml-1 for early stages and Artemia feeding density of 20 nauplii ml-1 appeared to produce the best performance of S. paramamosain larvae. Optimal rations for crab larvae should however be adjusted depending on various factors such as species, larval stage, larval status, prey size, rearing system and zootechnics. A practical feeding ration could be 30 - 45 rotifers ml-1 for Z1 - Z2 and 5 - 10 Artemia (meta) nauplii ml-1 from Z3 onwards.

Although not encouraged for commercial practices, antibiotics improved survival considerably, which shows bacterial disease is one of the key factors underlying the high mortality. Ozonation and probiotics as alternatives to prophylactic chemicals are worth investigating.

Cannibalism is also an important cause of high larval mortality at later larval stages and could be overcome by provision of suitable substrates/shelters and feeding larger Artemia meta-nauplii.

CHAPTER 6 – Rearing techniques

126

1. Introduction

Aquaculture of mud crabs, Scylla spp., contributes a large proportion to the world

production of the genus (FA0, 1999). Mud crab moreover represent a valuable component

of small-scaled coastal fisheries in many countries in tropical and subtropical Asia, for

which there has been a general trend of increased exploitation in recent years (Angell, 1992;

Keenan, 1999a). In Vietnam, the mud crab Scylla paramamosain is the second most

important marine species next to shrimp, being cultured widely in the coastal area. Mud

crab farming however currently relies entirely on the wild for seed stock and the main

obstacle for expansion is the unavailability of hatchery-reared seed (Liong, 1992; Keenan,

1999a; Mann et al., 2001; Rattanachote and Dangwatanakul, 1992; Shelley and Field, 1999;

Xuan, 2001).

Zootechnics, disease and nutrition are the three main areas of research, which have

supported commercial controlled production of marine fish and crustacean larvae

(Sorgeloos and Léger, 1992). These three aspects are to a large extent interconnected and

developing hatchery techniques for a “new” species is not possible unless all three are

addressed. Strictly speaking, the design of rearing systems covers purely zootechnical

aspects. Sub-optimal rearing conditions (e.g. physical stress, lack of oxygen or sub-optimal

water quality) however affect larval health and can cause mass mortality by the outbreak of

diseases. Similarly, system design influences (live) feed quality and its availability for the

predator larvae.

There has been a great deal of progress in marine larval rearing technology since its

beginning in the 1960’s (Howell et al. 1998; Shelbourne 1964). Many of the modern

technical improvements developed over the past decades could, with some modifications, be

applied for mud crab. An overview of the rearing systems currently applied for larviculture

of mud crabs is presented in Chapter 2. Although much experience and knowledge has been

obtained from these systems, there is a need to further optimize rearing techniques in order

to maximize larval survival and quality. Furthermore, techniques should be adapted for each

Scylla species in function of local conditions (seawater source, status of hatchery

management, local resources).

This paper describes and discusses the main larval rearing techniques for S.

paramamosain that have evolved in the present study in order to further improve seed

production of this species.


127


2.1. Source of larvae

Fully gravid crabs were bought from local markets and transported to the hatchery.

Prior to stocking in the hatchery, the crabs were bathed in a 100 µl l-1 formalin solution for 1

hour. The crabs were housed individually in 100-l compartments of a roofed 2 × 2 × 0.5 m

cement tank, equipped with a biofilter. Rearing water (30 ± 1 g l-1 salinity) was diluted from

brine (90 - 110 g l-1) with tap water and chlorinated before use. Water temperature was not

controlled, but fluctuated slightly around 28 °C. Every crab was fed a daily ration of 10 - 15

g of fresh marine squid, bivalve or shrimp meat alternately.



for 1 hour and transferred to a 70-l plastic tank connected to a biofilter for egg incubation.

Daily management consisted of siphoning out waste material and shedded eggs from the

tank bottom and controlling temperature (30 °C), salinity (30 g l-1) and ammonia and nitrite

levels. Every other day, the crab was bathed in a 50 µl l-1 formalin solution for 1 hour to


crabs were not fed.

One to two days prior to hatching, the female was moved to a 500-l fibreglass tank.

When the hatching process was completed, larvae were selected based on their photo-tactic

behaviour. Aeration in the hatching tank was therefore turned off for several minutes and

the larvae that were actively swimming up to the surface were collected by gently scooping

them from the surface.

The larvae were then transferred to the rearing containers. In order to slowly acclimate

the larvae to the new rearing conditions, they were placed in a 50-l plastic mesh bucket and

slowly rinsed with water from the larval rearing containers for 20 to 30 minutes, before

releasing them.


128

2.2. Food and feeding

Start cultures of the micro-algae Chaetoceros calcitrans and Chlorella vulgaris were

cultured indoor with Walne solution in seawater of 30 g l-1 at 25 °C. Large-scale production

was performed indoor in 500-l tanks under a transparent roof. A hemocytometer was used to

count micro-algal densities.

Rotifer culture and enrichment


164 ± 22 and 120 ± 22 µm, respectively, was used in all experiments. Rotifers were cultured

indoor in 100-l fiberglass tanks operated in batch mode, following the procedure described

in Sorgeloos and Lavens (1996). Rotifers were initially grown on baker yeast, but one week

before use as feed for the larvae, the yeast was replaced by Culture Selco® (INVE

Aquaculture, Belgium). Temperature and salinity were controlled at 25 °C and 25 g l-1,

respectively. They were harvested through a 60 µm screen and rinsed thoroughly.

Rotifers were enriched with micro-algae or artificial enrichment media before being

fed to the crab larvae. Enrichment with Chlorella was performed at a density of 5 106 cells

ml-1 for 3 hours (Dhert, 1996). Rotifers were also enriched with Dry Immune Selco® (DIS,

INVE Aquaculture, Belgium), using two separate doses of 0.05 g l-1 at a 3-hour interval.

Enrichment was performed at a density of 500 rotifers ml-1. The water in the enrichment

vessel was slowly heated to 29 - 30 °C to avoid exposing the rotifers to thermal shock when

they were added to the larval rearing tanks. Before being fed to the larvae, enriched rotifers

were rinsed and re-suspended in clean seawater at the same temperature of the crab rearing

tanks.

Artemia culture and enrichment

Artemia nauplii (Vinh Chau strain) were hatched as described by Van Stappen (1996).

Both newly-hatched or enriched Artemia nauplii were used in the experiments of this study.

Artemia were enriched with Chaetoceros in the same micro-algal density as for rotifer

Micro-algae culture


129

enrichment. The nauplii were also enriched with DIS® (using two separate doses of 0.3 g ml-

1 at a 6-hour interval). The temperature and salinity were maintained at 30 °C and 30 g l-1,

respectively during Artemia enrichment. The density of Artemia during enrichment was 200

ml-1. Before feeding to the crab larvae, the Artemia were rinsed with disinfected seawater

and suspended at a known density in seawater.

Feeding

Rotifers were fed to the crab larvae from DAH 0 - 6 (Z1-Z2 stages). Newly-hatched

Artemia or Artemia meta-nauplii were offered from DAH 6 (Z3 stage) onwards. Rotifers

and Artemia were added daily at 30 - 45 ml-1 and 5 - 10 ml-1 to the rearing tank, respectively

(experiment 1, 2, 3, 4, 7 and 8). For experiment 5 and 6, live feed were feed at the required

prey densities based on the planned treatments. Whenever the crab larvae were fed enriched

live feed, algae- or DIS-enriched live feed were used on alternate days.

In experiments 1, 2 and 3, the effect of different water exchange schemes and the

addition of micro-algae on larval survival and development were evaluated. In experiment 4

to 8, other culture aspects such as Z1 stocking density, live feed density and different

prophylactic treatments were investigated. Water quality management schemes tested in

experiment 1 - 3 are summarized in Table 1. An overview of the experimental design and

culture conditions of all the experiments is presented in Table 2. Cylindro-conical fibreglass

tanks of 30 - 100 l were used in experiment 1 to 6. Experiments 7 and 8 were executed in 1-l

plastic cones. The small-scale experiments (1-30 l) were carried out in a temperature-

controlled room (28 - 30 °C). The experiments in 100-l tanks were executed outdoor under a

transparent roof without temperature control (27 - 31 °C). The source and the disinfection

procedure of the seawater for larval rearing were similar to those used for broodstock

rearing. Formalin at a concentration of 20 µl l-1 was applied every other day as prophylactic

treatment in experiment 1 to 6. Further details on experimental conditions are discussed for

each test separately.

2.3. Larval rearing experiments: objectives and experimental design


130

Experiment 1

In this experiment, larval survival and growth in a clear water system with daily

partial water exchange (Clear-Batch) was compared to those in a clear water recirculating

system (Clear-Recirc). In the first rearing system, 30 - 50 % of the culture water was

manually replaced daily. In the recirculating system, all rearing tanks were connected to a

central biofilter. Water was recirculated at a rate of approximately 100 % of the tank volume

every 3 - 4 hours. Live feed and crab larvae were retained in the rearing tanks with the help

of a mesh screen of 70 and 300 µm during the rotifer and Artemia feeding stage,

respectively. Larger mesh screens (250 and 500 - 1000 µm for rotifer and Artemia stage,

respectively) and higher flow rates were used upon daily flushing out uneaten live feed and

waste.

Experiment 2

Here, the Clear-Recirc system was compared with 2 systems where micro-algae were

added. Rearing conditions for the Clear-Recirc system were similar to those described in

experiment 1. In the Algae-Recirc system, micro-algae were added daily to the recirculating

system at a low concentration ranging from 0.1 to 0.2 million cells ml-1. The operation of

the rearing tanks was similar to the Clear-Recirc treatment. In the Green-Batch treatment, a

classical “green-water” system, micro-algae concentrations in the culture tanks were kept at

a ten-fold higher level of 1 - 2 millions cell ml-1. In this system, the culture tanks were

initially only filled to 50 % of their capacity and gradually increased to 100 % by the end of

the Z2 stage by daily adding water and algae. Later on, 10 - 30 % of the rearing water was

replaced daily by clean seawater and/or algae, depending on the density of micro-algae

remaining in the rearing tanks. Upon water exchange, uneaten live feed was also flushed out

through a mesh screen (mesh sizes as described in experiment 1). The same amount of live

feed (30 - 45 rotifers ml-1 and 5 - 10 Artemia nauplii ml-1) was fed in all treatments. In the

systems using algae, Chlorella was used for Z1 - Z3 stages (which is unsuitable as food

source for Artemia); from Z4 onwards, Chlorella was gradually replaced with Chaetoceros.


131

Experiment 3

In this experiment, a Green-Batch and a Green-Recirc system were set-up in order to

further evaluate the application of micro-algae on the performance of crab larvae. The first

rearing system was a batch system with addition of high concentrations of algae as

described in experiment 2. The second system consisted of a combination of the Green-

Batch system for early crab stages (Z1 - Z2) and Algae-Recirc system for later stages (Z3

onwards).

Experiments 4, 5 and 6

In these experiments, the effect of Z1 stocking density (50, 100, 150 and 200 l-1,

experiment 4), rotifer feeding density (30, 45 and 60 ml-1) for feeding Z1-Z2 stages

(experiment 5) and Artemia feeding densities (10, 15 and 20 ml-1) for feeding from Z3

onwards (experiment 6) was evaluated. These experiments were run in a Green-Batch

(experiment 4) or a Clear-Recirc system (experiments 5 and 6) as described above.

Experiments 7 and 8

In experiment 7, the effect of prophylactic chemicals on the survival of the larvae was

investigated. Three treatments, consisting of a control (no chemicals used), daily addition of

formalin at 20 µl l-1 and daily addition of Oxytetracycline at 10 mg l-1, were run in 1-l

plastic cones. All cones were placed in a waterbath in order to maintain the rearing

temperature at 30 °C. Water was replaced almost completely everyday. Upon water

exchange, the survival was counted. When counting survival, the bottles were rinsed with

clean water but the biofilm on the walls of the bottles was not removed as a simulation of

large rearing tanks.

Experiment 8 was a replication of experiment 7. In this experiment used bottles were

everyday replaced by new disinfected ones, i.e. the biofilm on the bottle sides was removed.

The survival and development of the crab larvae were compared.


132


The survival rates in the experiments using large (30 to 100-l) containers (experiment

1 to 6) were estimated by volumetric sampling. Depending on the tank volume and the

density of the surviving larvae, triplicate 300- to 1000-ml samples were taken from each

tank. Megalopae (M) (15 - 18 days after hatch (DAH)) and first crabs (C1) (DAH 22) were

counted individually. In experiments 7 and 8 (using small cones) the ival rate

was calculated by individually counting all surviving larvae in each replicate.

Larval development was monitored every three days (every day in experiments 7 and

8) by identifying the average zoeal instar stage of a sample of larvae and assigning it a

value: first zoea (Z1) = 1; second zoea (Z2) = 2, etc. Megalopa stage was assigned a value

of 6. To compare the larval development in each treatment, an average larval stage index

(LSI) was calculated from the average LSI value of all replicate containers in the same

treatment. For large containers (experiment 1 to 6), 5 or 10 larvae (in 30-l and 100-l tanks

respectively) were sampled from each tank to calculate the average LSI. The sampled larvae

were staged under a dissecting microscope. In experiment 8 using small containers, all

larvae were staged visually upon counting daily survival.








survival rate) or logarithmic transformations (for LSI value) (Sokal and Rohlf, 1995). P was

set at both 0.05 and 0.01. Whenever differences are significant at P < 0.01, this is also

indicated. All analyses were performed using the statistical program STATISTICA 6.0.

average surv


133

3. Results

Experiment 1

Larval performance in experiment 1, comparing the Clear-Batch and Clear-Recirc

systems is presented in Table 3. Survival in the Clear-Recirc system at Z4-Z5 stage on DAH

15 and in the megalopa stage on DAH 18 were significantly higher than those in the Clear-

Batch system (both at P = 0.01). Although slightly higher in the recirculating system, larval

stage index was not significantly different between treatments. The better larval

performance in the Clear-Recirc system was accompanied by significantly lower average

ammonia levels (P < 0.01) and slightly higher nitrite levels (see Table 2).

Experiment 2

Table 4 shows the survival and larval development rate of crab larvae cultured in 3

different rearing systems. On DAH 9, larval survival in the Clear-Recirc system was

significantly lower (P < 0.01) than in both treatments with micro-algae supplementation

(Algae-Recir and Green-Batch treatments). On DAH 12, survival in the Clear-Recirc

treatment was lower (P < 0.05) than in the Algae-Recirc system whereas the Green-Batch

system had intermediate results. The LSI values on DAH 15 show a similar trend although

not significantly different: a slower growth was attained in treatment Clear-Recirc compared

to the rearing systems with micro-algae supplementation.

The average levels of ammonia and nitrite in the Clear-Recirc and Algae-Recirc

systems were significantly lower (P < 0.01) than those in the Green-Batch system (see Table

2). In the Green-Batch system, peaks of ammonia and nitrite concentrations of 3 and 1 mg l-

1, respectively were recorded at the end of the experiment.

Experiment 3

Table 5 presents the larval performance of the crab larvae cultured in 2 rearing

systems in experiment 3. The survival rates and LSI values of both rearing systems on all

sampling days were not significant different (Table 5). However, the survival rates on later

days (from DAH 12 - 22) in the Green-Recirc system tended to be higher than those in the


134

Green-Batch system. Also LSI values of the Green-Recirc system in later stages (DAH 9 -

15) tended to be higher than those in the Green-Batch system. The biofilter had a

significantly positive impact on water quality in the second part of the experiment, with

reduced ammonia (P < 0.05) and nitrite (P < 0.01) concentrations as a consequence (see

Table 2).

Experiment 4

Table 6 shows the survival and development rate of crab larvae stocked at 4 different

Z1 densities (50, 100, 150 and 200 l-1) in experiment 4. Survival rates were not significantly

different among treatments. From DAH 6 onwards however, the highest survival was

achieved at a density of 100 Z1 l-1. LSI values also showed little variation between

treatments. Only on DAH 6, larvae in treatment 50 Z1 l-1 had a significantly higher LSI than

larvae in treatment 200 Z1 l-1 (P < 0.05). Also beyond DAH 6, growth tended to be slightly

lower at 200 Z1 l-1.

Experiment 5

Table 7 shows the survival rates and the LSI values of crab larvae fed 3 different

rotifer densities in the Z1-Z2 stages. No significant differences were found for any of the

parameters. However, from DAH 6 onwards, feeding 45 rotifers ml-1 seemed marginally

better than the other prey densities in terms of larval survival. Although not significant,

feeding 30 rotifers ml-1 resulted in the lowest LSI value DAH 6 - 15 and therefore

seems not to be sufficient for optimal larval development. The LSI values of treatments 45

and 60 rotifers ml-1 were very similar.

Table 8 presents the survival and development rate of crab larvae fed Artemia nauplii

at three densities (10, 15 and 20 ml-1) from Z3 stage onwards. Although no significant

differences were observed between treatments, increasing Artemia density tended to

enhance the survival and this tendency became more prominent as crab larvae developed.

LSI values did not vary much between the larvae fed 10 or 15 Artemia ml-1; the larvae fed

s from

Experiment 6


135

the highest Artemia density (20 ml-1) tended to develop slightly faster than those in the other

treatments.

Table 10 shows the survival and development rate to the megalopa stage (DAH 22) of

larvae receiving different prophylactic treatments. Unlike in experiment 7, in this

experiment, rearing cones were daily disinfected upon water exchange. From DAH 6 (P <

0.05) or DAH 9 (P < 0.01) onwards, the survival rate of larvae in the treatment using

antibiotics was significantly higher than those in the remaining treatments. Survival rates of

the control and formalin treatments were similar on most days. From DAH 6 onwards, the

LSI values of the formalin treatment were generally higher than for the other treatments (not

always significant). On DAH 15 and 18, the antibiotic treatment resulted in lower LSI

values (P < 0.01).

4. Discussion

4.1. Rearing system

Recirculation

Experiment 7

In Table 9, survival of larvae receiving antibiotics or formalin as prophylactic

treatment is compared with a control receiving no treatment. Antibiotics clearly resulted in

the highest survival; the control treatment had the lowest survival rates. The difference in

survival among treatments was significant (P < 0.01) as early as DAH 3. Only on DAH 4,

the formalin treatment had a higher survival (P < 0.05) compared to the control. A tendency

for slightly higher survival remained however until the last sampling day.

Experiment 8

Water recirculation through a biofilter in the Clear-Recirc system positively affected

larval performance compared to manual partial water replacement in the Clear-Batch system

(experiment 1). The advantages of recirculating systems in commercial fish and crustacean


136

larval production have been proven before for other species (Workshop on the use of

recirculation systems in fish and shrimp larviculture, Ghent University, December 2003).

Research into recirculating systems has also been identified as a priority for shrimp culture

(Lawrence and Lee, 1997). In these systems, water exchange is minimized through the use

of biological, chemical, and/or mechanical filtration to maintain continuously good water

quality. As they provide less stress and constant good water quality to the larvae, these

systems are able to maintain a high biological carrying capacity in relatively little space

(Quillere et al., 1993; Twarowska et al., 1997). The results of this experiment show that

recirculating systems are also seem worthwhile to further investigate for crab larviculture in

order to decrease labour requirements and seawater consumption, at the same time

providing a more stable culture medium and thus reducing stress for the larvae. If system

design is kept simple, recirculating systems could also be suitable for large-scale

production.

Role of supplemented micro-algae

In experiment 3 in Chapter 4, the treatments “rotifer only” and “rotifers and algae” are

similar to the Clear-Batch and Algae-Batch treatments in this study. In that experiment,

micro-algae have been proven to aid the development of late zoeae and first metamorphosis.

Also in this study the positive effects of micro-algae addition were clearly noticeable.

Both Algae-Recirc and Green-Batch systems had better survival than the Clear-Recirc

system, and only the Algae-Recirc resulted in higher LSI than the Clear-Recirc system.

Micro-algae have been proven to be beneficial by various modes of action. They could

help to maintain the quality of live feed. As in the cultivation of marine fish larvae,

unconsumed rotifers may reside in the tanks for several days and their nutritional value may

become severely reduced (Makridis and Olsen, 1999). Furthermore, according to these

authors, poorly-fed rotifers were more sensitive to starvation than well-fed rotifers, as their

nitrogen content decreased at a higher rate. Starvation of rotifers may be prevented by

supplementation of micro-algae at a high level as in the Green-Batch rearing system.

It is critical to control and maintain micro-algae populations in aquaculture ponds

since the micro-algae play an important role in stabilizing pond water quality via either

ammonia uptake or oxygen production (Tseng et al., 1991). Since the Clear-Recirc system

already provided optimal water quality, it is unlike that the stabilizing effect on water


137

quality is the only factor responsible for the improved performance in the algae-

supplemented system. In batch culture systems this effect would be probably much more

pronounced. A direct comparison between a green and clear water batch system was

however not made in this study.

In a study on the effect of Chlorella on the population of luminous bacteria Vibrio

harveyi, no luminous bacteria were recovered on day 2 and day 3 in flasks with Chlorella,

while those without the micro-algae still harboured luminous bacteria at day 3 (Tendencia

and dela Pena, 2003). Also the diatom Chaetoceros has been shown to produce natural

antibiotics and high concentrations of this marine diatom will eliminate V. vulnificus and

other pathogenic bacteria, which contribute to the propagation of viruses in the shrimp

production environment (Wang, 2003).

In conclusion, micro-algae in mud crab larval rearing seem to play a role in improving

and maintaining live food quality and controlling bacteria levels. In batch culture systems

they might also play an important role in stabilizing water quality.

Choice of system

In experiment 3, the Green-Recirc system (which is a combination of a Green-Batch

system during the rotifer feeding stage and a recirculating system thereafter) seemed to be

better than the Green-Batch system. The Green-Batch system seems more appropriate for

early stages of crab larvae (Z1 - Z2) as it is easier and less stressful for the early zoeae, to

gradually fill up the tanks with fresh seawater, algae and rotifers than flushing out old

rotifers in the recirculation system. In the recirculating system, the young larvae might also

be prone to physical damage and spend a lot of energy trying to swim up against the current.

Early crab larvae are delicate due to their small size (see Chapter 4) with three long spines

on the carapace that are easily damaged when they are entrapped on the mesh screen during

flushing of uneaten feed in the recirculation system. The nutritional effect of micro-algae is

probably also more pronounced during the rotifers feeding stage than during the Artemia

feeding stage. Furthermore, it is not necessary to recirculate water these first days, as the

concentrations of ammonium and nitrite are still low. Using the Algae-Recirc sytem in later

stages is more favorable for reducing the increasing ammonium and nitrite concentrations as

more waste material is produced by the crab larvae. Moreover, as the larvae develop into

more efficient predators, feed is consumed faster, and maintenance of optimal feed quality is


138

less of an issue. Many studies successfully applied a similar combined rearing technique due

to its benefit for the larvae and convenience for management, particularly for large rearing

containers. Under green-water culture conditions water is not exchanged for the first three

days. Thereafter, water exchange is slowly increased from 10 - 20 % day-1 for Z2 - Z3 to

between 40 and 50 % day-1 at the end of the rearing cycle (Z4 - M) (Mann et al., 1999b;

Quinitio et al., 2001). In Japan a mesocosm system is used for culturing larvae in larger

tanks (> 10 m3). The tanks are partially filled with green-water at Z1 (20 - 25 % volume),

tanks are then filled up with clean seawater during the course of the Z2-Z3 stages and

during the Z4 and M stages water is exchanged on a flow-to-waste basis (Hamasaki et al.,

2002b).

4.2. Other zootechnics

Although data were not significantly different, stocking at 100 Z1 l-1 seemed to be

best in terms of survival. Treatment 50 Z1 l-1 tended to have lower survival than treatment

100 Z1 l-1, possibly due to higher concentrations of ammonia and nitrite (see Table 2)

released from uneaten feed since all treatments were supplied with an identical ration. For S.

paramamosain, Djunaidah et al. (2001a) found a tendency of increased survival to Z5 in

function of the Z1 stocking density (i.e. survival rates of 27, 39 and 63 % being obtained at

the densities of 50, 75 and 100 Z1 l-1, respectively). Baylon and Failaman (1999) also

reported higher survival and metamorphosis of S. serrata at a density of 50 Z1 l-1 compared

with lower densities of 10 and 25 Z1 l-1. The higher stocking densities tested in our study

(150 and 200 Z1 l-1) might have caused competition for feed and stress (higher risk of

cannibalism) for the larvae, resulting in lower survival (150 Z1 l-1 and 200 Z1 l-1) and

slower development (200 Z1 l-1).

Rotifer density for feeding early larval stages (Z1-Z2 stages)

In experiment 5, the rotifer density of 30 ml-1 tended to give the poorest performance.

A density of 60 rotifers ml-1 was best for larval growth, but did not lead to marked

improvements, and might moreover be economically unrealistic. Therefore, the intermediate

Z1 stocking density


139

density of 45 rotifers ml-1 was frequently used for feeding crab larvae in early stages.

Baylon et al. (2004) found that S. serrata Z1-Z2 larvae were capable to ingest 900 (640-

1200) rotifers larva-1 day-1. Based on these data, a density of 45 rotifers ml-1 would exactly

meet the daily live feed requirement of larvae stocked at 50 Z1 l-1 as in experiment 5. Other

studies also indicated that rather high rotifer densities (30 - 80 ml-1) are required for optimal

growth and survival of S. paramamosain (Djunaidah et al., 1998; Zeng and Li, 1999) and S.

serrata (Suprayudi et al., 2002a). For S. paramamosain larvae, feeding 30 and 60 rotifers

ml-1 resulted in a significantly higher survival compared to feeding only 15 rotifers ml-1

(Djunaidah et al., 2001a). These authors found that individual dry weight of Z5 fed 15

rotifers ml-1 was significantly lower than those of Z5 fed with higher rotifer densities.

Practically, feeding 30 rotifers ml-1 at Z1 and increasing gradually to 45 ml-1 at Z2 proved to

be sufficient for a stocking density of 100 Z1 l-1 in our trials in larger rearing tanks (500 -

1,000 l). Increasing the ration by larval stages in this way is compensating the increased

ingestion of crab larvae as they grow (Baylon et al., 2004). For early larvae, feed amount

can however not be reduced to their maximum ingestion potential as they are quite

inefficient predators and therefore might require a minimal density to maximize encounter.

As in our study, most studies investigating the effect of rotifer density, added the live

feed in one single ration. Under these circumstances, theoretic densities are only attained

upon feeding and gradually decrease as larvae consume the prey. Optimal live feed

quantities can however not be separated from feeding frequency. Since zoeal larvae can

consume their optimal ration within 1 hour, Genodepa et al. (2004) suggested they can be

fed once a day. Because of the severe reduction of nutritional value of rotifers with longer

retention times in rearing containers (Makridis and Olsen, 1999) and a minimum prey

density needed for the passive feeding behaviour of zoea larvae (Heasman and Fielder,

1983; Zeng and Li, 1999), the optimal ration and the number of feedings should be further

investigated.

Artemia for feeding later larval stages (from Z3 onwards)

For feeding Z3, at a density of 100 Z1 l-1, a daily feeding ration of 20 Artemia nauplii

ml-1 seemed to result in the best larval performance for S. paramamosain. Especially in later

larval stages (Z4 - Z5), the higher survival compared to lower rations tended to be more

pronounced. In this respect, it might be beneficial to increase the Artemia density by crab


140

stage from 10 to 20 ml-1 from Z3 to Z5. On the other hand, high live feed densities would

increase the chance for early larvae to encounter and capture feed organisms (Zeng and Li,

1999) and therefore would improve the larval performance (Brick, 1974; Heasman and

Fielder, 1983; Quinitio et al., 2001). Optimal rations should therefore be determined for

each larval stage separately. Studies on individual larvae are very useful to determine prey

consumption. According to our previous experiment (Chapter 4), one Z3, Z4, Z5 and

megalopa larva were capable to consume on average 15, 25, 37 and 114 newly-hatched

Artemia day-1, respectively. Therefore, at a stocking of 100 Z1 l-1, the daily feeding densities

of Artemia theoretically should be at least 1.5, 2.5, 3.7 and 11.4 ml-1, for each stage. In a

similar experiment with S. serrata, Baylon et al. (2004) found that there was an increasing

ingestion of Artemia nauplii of 80 - 160 individuals at Z2, 240 at Z3 - Z4 and 320 - 640 at

Z5. These data also show an increase of prey consumption with larval development, but the

absolute values are much higher than in our study. This difference can probably be

explained by the fact that we only counted missing Artemia as ingested; whereas Baylon et

al. (2004) included both missing and those having only missing body parts as consumed. In

the same study, these authors also investigated the effect of prey density on ingestion by

individual S. serrata larvae. For Z1, Z2 and Z3 stages, the number of Artemia nauplii

ingested by the larvae at a lower food density of 2.5 ml-1 was comparable to that at 5 ml-1;

and for Z4 - Z5, a density of 5 ml-1 was comparable to 10 ml-1 (Baylon et al., 2004). In that

study, Artemia was however co-fed with rotifers at a density of 15 - 20 ml-1. If Artemia

were the only food, the optimal Artemia ration would therefore probably be higher than 2.5

to 5 ml-1. In another study on S. serrata, a daily optimum food concentration of 10 Artemia

nauplii ml-1 was established for zoea survival (Brick, 1974).

In conclusion, the optimal Artemia ration for Z3 - Z5 observed in experiment 6 (10 -

20 Artemia nauplii ml-1) seems higher than in most other studies (2.5 - 10 Artemia nauplii

ml-1) (Baylon et al., 2004; Brick, 1974). Maybe the recirculating system used in this study

resulted in a greater loss of prey organisms (e.g., more Artemia were entrapped on the

overflow screen) than in the small batch culture systems used in other experiments. The

small nauplii size of the Artemia strain (Vinh Chau strain) used in our study could be

another cause that led to increased ingestion. In practice, (larger-sized) HUFA enriched

Artemia were normally used in order to reduce the prey amount to 5 - 10 ml-1.

For megalopae of S. paramamosain we found a three fold higher number of ingested

newly-hatched Artemia nauplii compared to Z5 (114 and 37 Artemia, respectively) for a


141

similar prey density (see experiment 1 in Chapter 4). This means megalopae are voracious

predators, capable to chase their prey actively and consume large amounts of feed in a short

time. From this, it could be beneficial when megalopae are fed frequently smaller rations in

order to optimise feed quality and reduce cannibalism. Genodepa et al., (2004) similarly

indicated that in contrast to earlier larval stages, which can be fed once per day, S. serrata

megalopae may need to be fed more often to maximize ingestion. These authors found no

significant differences in the ingestion rate of megalopae fed microbound diets at rations

ranging from 12.5 to 100 % of the standard ration (equivalent to 5 Artemia nauplii ml-1 in

one hour). Baylon et al. (2004) also found a high increase in Artemia ingestion in the first

few days of the planktonic phase of the megalopa stage. Later on, megalopae become more

benthic and prepare for the second metamorphosis to first crab; therefore, Artemia with

jerky swimming are no longer preferable, but minced shrimp or mussel meat are a more

suitable feed.

In general, the optimal Artemia ration for crab larvae should be adjusted depending on

various factors, e.g. species, larval stage, larval status, prey size, rearing system and

zootechnics. As for rotifer feeding, feeding sufficient smaller rations several times

throughout the day might be beneficial.

Prophylactic chemicals

In experiment 7, heavy fungal infection was observed in larvae of the control

treatment, but not in the other treatments. In an experiment to control fungal diseases caused

by Lagenidiales (an order including 5 virulent fungal strains: one Haliphthoros, Sirolpidium

and Atkinsiella strain and two Lagenidium strains) in eggs and larvae of the swimming crab,

Portunus trituberculatus, and the mud crab, S. serrata, Hamasaki and Hatai (1993a, 1993b)

found that a bath treatment with 25 µl l-1 formalin of newly-hatched larvae in the hatching

tank was a practical and effective method for inhibiting the occurrence of these fungal

diseases. As formalin as such was found not to be toxic to crab larvae at that concentration,

it could be considered that the higher mortality in the formalin treatment compared to the

antibiotics treatment, was probably caused by pathogenic bacteria. Laboratory cultures of

crab larvae often suffer severe mortality from disease, particularly from epibiotic bacteria

and larval mycosis (Amstrong et al., 1976; Hamasaki and Hatai, 1993a, 1993b). A study on

S. serrata indicated a significantly higher survival up to DAH 7 (over 90 %) when using


142

Oxytetracycline, whereas almost complete mortality occurred in the control treatment

(Mann, 2001). The author considered that potentially up to 80 % of the larval mortality

could be attributed to bacteriological causes.

In experiment 8, the improved hygiene conditions (i.e. new disinfected cones were

used everyday) did not improve survival and development of larvae in the control and

formalin treatments. Oxytetracycline in this experiment proved the best prophylactic

chemical for crab larvae. Antibiotics resulted in a significantly higher survival but seemed

to negatively affect larval development to some extent.

However, antibiotics have not always been used in a responsible manner in

aquaculture. A major consequence of using antibiotics has been the proliferation of resistant

bacteria and the transmission of resistance to other bacterial species (Benson, 1998). The

development of antibiotic resistance by pathogenic bacteria is considered to be one of the

most serious risks to human health at the global level (FAO, 2002). Formalin is more

acceptable than antibiotics as it shows no accumulation in animal tissues (Jung et al., 2001).

Recently however, Japan has strictly banned the use of formalin in aquaculture as it might

cause cancer in humans, reduces oxygen levels in the water and causes algae to die off

(VASEP, 2003). Moreover, in this experiment, formalin did not improve larval survival

sufficiently to make it economically rewarding. Pathogenic bacteria are considered as one of

the most serious causes for the high mortality of early crab larvae. It can be safely assumed

that all inputs (seawater, broodstock, live feed and daily management in hatcheries) into the

culture tank are potential sources of infection (Blackshaw, 2001). Strict hygiene at all steps

is always advised for hatchery activities. However, this advice is not always being followed,

especially in backyard hatcheries. Therefore, other zootechnics should be investigated as

alternatives for the use of chemicals. Ozonation and probiotics could be interesting in this

respect (see Chapter 7).

Cannnibalism

Cannibalism is another important cause of mortality that is strongly linked to system

design and rearing conditions. In a previous experiment in the Clear-Recirc system in 30-l

tanks, a total survival of Z1 to Z5/megalopa of 42 ± 9 % was obtained on DAH 15 (data not

shown). Mortality of the larvae mainly resulted from cannibalism as most of the dead larvae

had missing appendages, but no symptoms of disease or unsuccessful metamorphosis were


143

observed. Especially cannibalism among Z5 and of megalopa on Z5 seemed to be a

problem. Asynchronous moulting exacerbates the problem of cannibalism (Quinitio et al.,

2001). Moulting synchronicity can be improved by improving nutrition through live food

enrichment (Takeuchi et al., 1999). Cannibalism, particularly at the megalopa and crab

stages, is a common problem and accounts for a large percentage of the mortality (30 - 50

%) in Scylla crabs (Dat, 1999b; Heasman and Fielder, 1983; Latiff and Musa, 1995;

Quinitio et al., 2001; Suprayudi et al., 2002a; Zainoddin, 1992). Cannibalism could be

averted by transferring the metamorphosed larvae to separate containers (Davis, 2003),

nursing megalopae in spacious containers (Marasigan, 1998; Rodíguez et al., 1998),

supplying substrata and shelters (Mann et al., 1999b; Marasigan, 1998; Quinitio et al., 2001)

and feeding live HUFA enriched Artemia juveniles instead of small-sized Artemia nauplii.

In experiment 4 in Chapter 5, on average 27 % (maximum 48 %) of Z1 survived to the

first crab stage. In this experiment, megalopa were removed manually from the rearing

bowls immediately after metamorphosis. This survival rate was the best obtained so far and

would be economically viable. However, it is not practical to manually separate out

megalopae in large rearing tanks. Furthermore, S. paramamosain megalopae appear to be

more delicate than those of S. serrata and there is only a short time between first and second

metamorphosis (7 - 10 days). Therefore; it is more convenient to leave the megalopae in the

same culture tank and harvest the crablets after a short period of 10 days. The best practice

seems to be to offer sufficient substrate and HUFA-enriched Artemia juveniles. Optimal

type and amount of substrate have not been sufficiently investigated however and facilities

for culturing Artemia biomass are not always available on research scale.


The combination of a green-water batch system for early stages and a recirculating

system with micro-algae supplementation for later stages, stocking density of 100 Z1 l-1,

feeding density of 45 rotifers ml-1 for early stages and 20 Artemia nauplii ml-1 for later

stages resulted in the best performance of S. paramamosain larvae.

The optimal ration for crab larvae should however be adjusted depending on various

factors, e.g. species, larval stages, larval status, prey size, rearing system and zootechnics. A

feeding regime with frequent addition of small quantities of feed is worth investigating.


144

Practically, rations of 30 - 45 rotifers ml-1 for Z1 - Z2 and 5 - 10 Artemia (meta) nauplii ml-1

from Z3 onwards can be applied.

Antibiotics improved larval survival but there are some indications that they might

retard growth. Formalin enhanced larval development with intermediate survival. Both

products are not encouraged for commercial mud crab larviculture as they are unsafe and/or

unefficient. Ozonation and probiotics as alternatives to prophylactic chemicals are worth

investigating.

Cannibalism is also an important cause of high larval mortality at later larval stages

and is most practically overcome by supplementation of suitable substrates/shelters and

feeding ongrown Artemia meta-nauplii.

Acknowledgements




145

Table 1 Overview of larval rearing systems applied in this study based on way of water exchange and micro-algae supplementation

WATER EXCHANGE ALGAE SUPPLEMENTATION

Discontinuous manual partial water renewal

Continuous water treatment through the use of biofilter

No micro-algae supplemented (indoors) Clear-Batch system Clear-Recirc system

Micro-algae supplemented at low levels to provide extra food for live preys (indoors or outdoors)

Algae-Batch system Algae-Recirc system

Micro-algae supplemented at high concentration and self-sustainable under natural sunlight as an extra food for live prey and water conditioning (outdoors)

Green-Batch system

Green-Recirc system (Combination of Green-Batch and Algae-Recir system at early and late larval stages, respectively)

Table 2 Overview of experimental conditions and water quality parameters (mean ± standard deviation) in experiment 1 to 8. For description of rearing systems refer to Table 1

Factor Rearing systemContainer volume

cates


NH4+

(mg l-1) NO2

-

(mg l-1)

Experi- ment (l)

No of repli-

Clear-Batch 0.35±0.14A 0.14±0.11a1 Clear-Recirc 50 0.03±0.07B 0.11±0.09a

0.02±0.04b B 0.04±0.01b B

Rearing system 30 5

Clear-Recirc Al 0.07±0.08b B 0.10±0.08b B 2 Rearing system Green-Batch

100 8 100 1.72±1.30a A 0.57±0.33a A

1.54±1.50a 0.43±0.13A

gae-Recirc

Green-Batch 3 Rearing system Green-Recirc 100 4 100 0.11±0.07b 0.15±0.10B

50 1.50±1.09a 0.53±0.39a

100 1.07±0.62a 0.43±0.35a

150 0.98±0.68a 0.43±0.29a100

200 0.73±0.48a 0.34±0.26a

(30, 45 & 50 rotifers ml-1)

4 Z1 density Green-Batch 3

5 Rotifer density

Clear-Recir 30 5 50 0.06±0.05 0.06±0.06

6 Artemia density (10, 15 & 20 Artemia ml-1)

Clear-Recir 30 5 100 0.04±0.05 0.13±0.09

7 1 4 100 n.d. n.d.

8

Prophylactic treatment (control, formalin & Oxytetracycline)

Clear-Batch 1 4 100 n.d. n.d.

Values within an experiment in the same column followed by the same superscript letter are not statistically different (P ≥ 0.05, regular letters and P ≥ 0.01, capital letters). N.d. = not determined.


146

Table 3 Survival rates and larval stage index (LSI) values of S. paramamosain larvae cultured in 2 different rearing systems. For treatment descriptions refer to Table 1. Experiment 1. DAH = days after hatch Treatment DAH 3 DAH 6 DAH 9 DAH 12 DAH 15 DAH 18 Survival rate (%) Clear-Batch 85±6a 79±9a 70±6a 64±7a 42±6B 32±5B

Clear-Recirc 84±4a 78±8a 72±5a 70±5a 63±9A 47±6A

LSI

Clear-Batch 1.5±0.2a 2.7±0.1a 3.5±0.4a 4.0±0.0a 4.6±0.2a

Clear-Recirc 1.5±0.2a 2.7±0.1a 3.6±0.3a 4.2±0.3a 4.8±0.1an.d. n.d.

Survival rates or LSI values in the same column followed by the same superscript letter are not statistically different (P ≥ 0.05, regular letters or P ≥ 0.01, capital letters). N.d. = not determined.

Table 4 Survival rates and larval stage index (LSI) values of S. paramamosain larvae cultured in 3 different rearing systems. For treatment descriptions refer to Table 1. Experiment 2. DAH = days after hatch Treatment DAH 3 DAH 6 DAH 9 DAH 12 DAH 15 Survival rate (%) Clear-Recirc 74±12a 63±7a 44±6b A 26±11b A 8±7a

Algae-Recirc 74±12a 63±9a 61±7a A 43±7a A 15±8a

Green-Batch 74±11a 67±9a 58±9a A 35±9ab A 13±6a

LSI Clear-Recirc 1.9±0.1a 2.7±0.2a 3.9±0.1a 5.0±0.1a 5.1±0.1b A

Algae-Recirc 2.0±0.1a 2.8±0.3a 4.0±0.1a 5.0±0.1a 5.6±0.2a A

Green-Batch 2.0±0.1a 2.8±0.2a 4.0±0.1a 5.0±0.1a 5.1±0.1ab A


Table 5 Survival rates and larval stage index (LSI) values of S. paramamosain larvae cultured in 2 different rearing systems. For treatment descriptions refer to Table 1. Experiment 3. DAH = days after hatch Treatment DAH 3 DAH 6 DAH 9 DAH 12 DAH 15 DAH 22 Survival rate (%) Green-Batch 94±6a 88±9a 80±3a 66±15a 44±20a 9±1a

Green-Recirc 94±6a 89±8a 80±5a 68±11a 56±11a 12±3a

LSI

Green-Batch 1.4±0.3a 2.7±0.1a 3.8±0.4a 5.0±0.0a 5.2±0.2a

Green-Recirc 1.4±0.2a 2.6±0.2a 3.9±0.3a 5.0±0.0a 5.3±0.1an.d. n.d.

Survival rates or LSI values in the same column followed by the same superscript letter are not statistically different (P ≥ 0.05). N.d. = not determined.


147

Table 6 Survival rates and larval stage index (LSI) values of S. paramamosain larvae stocked at 4 different Z1 densities (Z1 l-1). Experiment 4. DAH = days after hatch Treatment DAH 3 DAH 6 DAH 9 DAH 12 DAH 15 DAH 22 Survival rate (%) 50 79±9a 56±19a 42±16a 31±17a 28±12a 4±6a

100 80±7a 74±6a 71±10a 56±11a 45±8a 5±4a

150 79±2a 57±12a 45±9a 31±12a 28±10a 5±1a

200 85±5a 53±17a 42±8a 34±3a 30±5a 5±1a

LSI 50 1.7±0.2a 3.0±0.1a A 3.9±0.1a 5.0±0.1a n.d. 100 1.8±0.1a 3.0±0.1ab A 4.0±0.0a 5.0±0.1a

150 1.8±0.2a 3.0±0.1ab A 3.9±0.2a 5.0±0.1a

200 1.8±0.2a 2.7±0.1b A 3.7±0.2a 4.8±0.1a

n.d. n.d. n.d. n.d. n.d. n.d. n.d.

Survival rates or LSI values in the same column followed by the same superscript letter are not statistically different (P ≥ 0.05, regular letters and P ≥ 0.01, capital letters). N.d. = not determined.

Table 7 Survival rates and larval stage index (LSI) values of S. paramamosain larvae fed 3 different rotifer densities (rotifers ml-1) from DAH 0 - 6. Experiment 5. DAH = days after hatch

Treatment DAH 3 DAH 6 DAH 9 DAH 12 DAH 15 Survival rate (%) 30 89±7a 53±10a 30±8a 13±7a 10±7a

45 87±6a 58±9a 35±7a 18±9a 14±8a

60 87±5a 55±7a 32±6a 16±11a 11±10a

LSI 30 1.8±0.2a 2.7±0.1a 3.6±0.2a 3.8±0.2a 4.0±0.4a

45 1.8±0.2a 2.8±0.2a 3.8±0.1a 3.9±0.1a 4.3±0.5a

60 1.8±0.2a 2.8±0.2a 3.8±0.2a 4.0±0.1a 4.4±0.5a

Survival rates or LSI values in the same column followed by the same superscript letter are not statistically different (P ≥ 0.05).

Table 8 Survival rates and larval stage index (LSI) values of S. paramamosain larvae fed 3 different instar-1 Artemia densities (Artemia ml-1) from DAH 6. Experiment 6. DAH = days after hatch Treatment DAH 9 DAH 12 DAH 15 Survival rate (%) 10 26±10a 12±5a 8±3a

15 30±6a 13±7a 10±6a

20 32±8a 19±9a 18±9a

LSI 10 3.1±0.2a 3.7±0.4a 4.3±0.5a

15 3.1±0.1a 3.7±0.2a 4.3±0.5a

20 3.2±0.1a 3.8±0.3a 4.6±0.3a

Survival rates or LSI values in the same column followed by the same superscript letter are not statistically different (P ≥ 0.05).


148

Table 9 Survival rates (%) of S. paramamosain larvae treated daily with prophylactic chemicals from DAH 1 - 9. Experiment 7. DAH = days after hatch Treatment DAH 1 DAH 2 DAH 3 DAH 4 DAH 5 DAH 6 DAH 7 DAH 8 DAH 9 Control 94±2a 79±3a 61±3b B 54±4c B 35±10b B 24±8b B 13±6b B 3±3b B 1±1b A

Formalin 94±3a 81±10a 75±12ab AB 68±11b AB 50±16b AB 38±16b AB 25±11b B 8±7b AB 4±6b A

Antibiotics 97±1a 89±3a 85±5a A 82±4a A 77±5a A 66±6a A 56±3a A 28±12a B 16±11a A

Survival rates in the same column followed by the same superscript letter are not statistically different (P ≥ 0.05, regular letters and P ≥ 0.01, capital letters).

Table 10

Survival rates and larval stage index (LSI) values of S. paramamosain larvae treated daily with prophylactic chemicals. Experiment 8. DAH = days after hatch Treatment DAH 3 DAH 6 DAH 9 DAH 12 DAH 15 DAH 18 DAH 22 Survival rates (%) Control 85±3a 64±7b A 48±8b B 34±13b B 28±11b B 17±7b B 9±5b A

Formalin 84±7a 66±8b A 47±7b B 34±12b B 26±8b B 13±10b AB 11±8ab A

Antibiotics 91±4a 80±2a A 74±4a A 66±8a A 52±6a A 34±3a A 21±5a A

LSI Control 1.8±0.1a 2.6±0.3a 3.4±0.2ab A 4.5±0.3a 5.1±0.1a AB 5.5±0.1b AB 5.8±0.2a A

Formalin 1.8±0.1a 2.8±0.1a 3.7±0.1a A 4.3±0.3a 5.1±0.1a A 5.8±0.1a A 6.0±0.1a A

Antibiotics 1.9±0.0a 2.7±0.2a 3.3±0.2b A 4.3±0.1a 4.9±0.1b B 5.3±0.2b B 5.6±0.2a A


CHAPTER 7

The earliest documented attempts for the breeding of a crustacean in captivity were

made by the Japanese scientist Motosaku Hudinaga (Hudinaga, 1942) using mature Penaeus

japonicus females collected from the wild. The demonstration of the technical feasibility of

mass production was successfully achieved in 1953 (Hudinaga and Kittaka, 1967) and this

laid the groundwork to modern Penaeid shrimp culture.

General discussion

1. Challenges and perspectives for mud crab larviculture

Feasibility of mud crab larviculture compared to other crustaceans

Before starting the larviculture of a new aquaculture species, it is important to first

understand some basic biological characteristics of the species in question. Wickins and Lee

(2002) have, based on some basic characteristics (control of breeding, fecundity, duration of

larval life, growth and tolerance to crowding) ranked culture feasibility of a number of

crustacean groups (shrimp, prawn, crayfish, crab and lobster). According to these authors,

feasibility of larviculture of crabs is only just better than of lobsters. Shrimp were regarded

to be the easiest in terms of larval production.

Compared to shrimp, the development of technologies for the larval culture of mud

crabs is much more recent. Although the first research on mud crab larviculture already

dates back from the early 60’s (Ong, 1964), progress was discontinuous and unreliable. In

recent years, a renewed interest in mud crab culture has emerged mainly due to the need of

diversification of cultured marine species in relation to the high risk of epidemic diseases in

shrimp and the increasing demand for high-value aquaculture species in the global market.

In 1991, the first international seminar focusing on mud crab culture and trade was

convened in Sura Thani, Thailand. The seminar stressed that especially attempts to develop

techniques for mud crab seed production have been very limited. The slow pace of progress

was attributed to the general lack of knowledge about certain aspects of the larval and

juvenile stages such as feeding behaviour, nutritional requirements and water quality

requirements. Survival from zoea to first crab stage had been obtained in the laboratory, but

this had not been transferred to commercial practice. Until the most recent international

CHAPTER 7 – General discussion

150

workshop on mud crab rearing, ecology and fisheries held in Can Tho, Vietnam in 2001,

also no commercial hatcheries had been established. In the meantime however, considerable

progress has been achieved and specific problems have been solved one by one, e.g.

maximum survival from newly-hatched zoeae to the first crabs of 48, 15 and 10 % in 1-l

bowls, 100-l tanks and 1000-l tanks, respectively have been achieved.

The slow progress in mud crab larviculture success can be understood from the

delicate nature of the larvae. Davis (2003) claimed that this evolves from the typical “r”

type reproduction strategy of the species. Mud crabs spawn and hatch a large number of

eggs but these larvae are subjected to high mortality. The newly-hatched Scylla larvae are

tiny (hatching from small eggs of 0.3 - 0.4 mm in diameter), are not fully developed yet

(Zeng and Li, 1999), have very little yolk reserve (Cheng and Li, 2001) and need to feed

soon after hatching (Djunaidah et al., 2003; Li et al., 1998; Li et al., 1999; Lumasag and

Quinitio, 1998). They are inefficient feeders in the zoeal stages (Heasman and Fielder,

1983) and therefore require high densities of live food (Zeng, 1998). They are also

extremely susceptible to disease (Mann, 2001) and thus require highly hygienic conditions

(Blackshaw, 2001). They are sensitive to subtle changes in their physical environment (Hill,

1974) and require stable, high quality seawater (Davis, 2003). They are also highly

cannibalistic in the megalopa and crab stages (Dat 1999b; Quinitio et al., 2001; Suprayudi et

al., 2002a) and undergo two energetically demanding and stressful metamorphoses before

reaching the juvenile stage (Hamasaki et al., 2002b). The larvae are thus extremely difficult

to rear in high density conditions and establishing hatchery technology has proven to be

more challenging than for any other commercially important decapod crustacean (Davis,

2003) such as the Penaeid shrimp, Chinese mitten crab (Eriocheir sinensis) (Naihong et al.,

1999) and other Portunidae such as the blue crab (Portunus trituberculatus) with only 4

zoeal stages (Cowan, 1984). However, mud crab larvae are not as difficult to rear as those of

the Palinurid spiny lobsters (mostly Panulirus spp.), which take an extremely long time to

develop (197 - 365 days) (Ong, 1966; Kittika, 1994; Wickins and Lee, 2002).

Challenges for mud crab larviculture

Crustacean larval development occurs within a narrow range of environmental

parameters (Sastry, 1983). The conditions that planktonic marine larvae encounter in the

wild are impossible to reproduce in the laboratory. The ocean acts as a buffer into which


151

harmful chemicals are quickly diluted and changes in temperature, salinity or pH occur

extremely slowly (Davis, 2003). Larvae are able to maintain a minimum distance from one

another and dead or diseased larvae drop out of the surface layers and do not pollute the

water. The larvae are also able to regulate their position in the water column, selecting ideal

conditions of light, salinity, turbulence (Forward et al., 1984) and food availability (Malkiel

et al., 1999). The distribution of plankton in the ocean is not homogenous (Benfield and

Downer, 2001) as oceanographic phenomena tend to concentrate plankton in “fronts” (Clark

et al., 2001) or “patches” (Natunewicz et al., 2001). Crab larvae that are incorporated into

these fronts are vulnerable to predation from larger zooplankton, but in turn have access to

an abundance of suitable-sized plankton as food (Davis, 2003).

Differences between the natural environment and that of a hatchery have been

clearly reviewed by Davis (2003). Predators are excluded from the rearing vessels but the

larvae are crowded together into small volumes, at artificially high densities, increasing

interactions and the risk of disease and cannibalism. Smaller volumes of water are also

much more subjected to changes in physical and chemical parameters. Although live feed is

always made available, the species provided are selected as much for their ease of culture as

for their suitability as feed and usually bear little resemblance to the natural food of the

larvae in the wild. Cultured live feed organisms moreover are often stressed from the rigors

of mass culture and processing, carry bacterial and other contamination and sometimes lack

essential nutrients (Støttrup and McEvoy, 2003).

Perspectives for larviculture of mud crabs

Comparing the conditions of “natural” and “man-made” hatcheries, one can not help

to have a “gloomy” picture about the artificial reproduction of mud crabs. However, “r”

type species can be beneficial in the hatchery operation due their own characteristics. For

example, due to their high fecundity, fewer broodstock animals are needed compared to

shrimp to provide the required number of larvae for a hatchery. Or the low survival rate of

the larvae could be compensated by maintaining more broodstock and stocking higher

numbers of larvae.

An r-selected approach to reproduction emphasizes producing large numbers of

offspring with minimal care given to each offspring by its parents (Andrews and Harris,

1986; Pianka, 1970). Based on this concept, larval rearing of mud crabs in hatcheries could


152

be considered as an “artificial” trade-off to K-selection, in which few offspring (the best,

most active ones are selected at the beginning) are produced, but maximal care (proper

feeding and adequate prophylactic measures) is taken to insure viability of each offspring.

In this light, artificial rearing of mud crab larvae could be commercially viable, although

their quality might be inferior compared to wild seed where a strong natural selection takes

place.

2. Broodstock management (Chapter 3)

Broodstock management is without doubt one of the least understood and researched

areas (Izquierdo et al., 2001), especially for a newly-studied species like the mud crab. The

basis of every hatchery operation is the maintenance and conditioning of a healthy group of

broodstock that could spawn year-round. In most areas, where larval rearing of mud crab,

Scylla spp., is conducted, the source of eggs relies on gonadal maturation and spawning of

wild-caught broodstock in captivity (Mann et al., 1999a). Due to the migratory behavior of

female mud crabs in the wild (Hill, 1994), the knowledge of spawning, breeding and

hatching of eggs under natural conditions is lacking. Most information on these processes

therefore comes from crabs that are held in captive conditions for the purposes of

aquaculture research and production (Mann et al., 1999a).

Fortunately, unlike shrimp (e.g. Penaeus monodon), S. paramamosain broodstock are

still easily available from the wild and they can be readily induced to spawn almost year-

round within a short time (i.e. one week for fully gravid females) after stocking in the

hatchery. Furthermore, pond-reared crabs also produce good quality eggs (Millamena and

Quinito, 2000) and the life cycle has been closed in captivity (Quinitio et al., 2001). When

fully gravid females are selected, there also does not seem to be special nutritional

requirements for maturation and spawning and the animals do not have to pass through a

special conditioning. Based on the average reproductive performance obtained in our study,

the cost for newly-hatched crab larvae is 20 to 40 times cheaper compared to Penaeus

monodon nauplii. In that respect, research on broodstock management was not the first

priority in our study. Nevertheless, data on rearing conditions and main reproductive

characteristics were recorded as a basis to further improve techniques. Chapter 3 firstly

intends to describe the practices that promoted the production of good quality larvae in this


153

study, and secondly, aims to elucidate a number of factors which are important in the

selection and management of broodstock for hatchery production purposes.

Our results indicated that egg quality remained unchanged within 60 days in captivity,

given the low price of crab breeders and the decreasing spawning activity however, it is

recommended to keep the broodstock for not more than 30 days. Males were not needed in

the hatchery as fertilization had happened in the wild and adult females caught from the

wild have usually mated and carry spermatophores for extended periods (Robertson and

Kruger, 1994). Although more difficult to manage, pond systems resulted in the best

reproductive performance. Tank systems could be a more practical alternative if stocking

densities are kept low and a proper substrate is provided. The peak spawning season of S.

paramamosain in South Vietnam seems to be from March to July, which would be the

easiest period to obtain good spawners and hence perform larval rearing. With eyestalk

ablation, the optimal production period for mud crab could be extended from February to

August. Spawning activities seemed to be related to ambient temperatures rather than the

monsoon season. Females of 300 - 500 g produced the highest total number of viable Z1 and

are therefore preferred. Smaller females were sometimes not fertilized, while big females

(over 500 g) tended to have lower relative Z1 fecundities. Although detached eggs could be

incubated artificially, egg incubation by the females themselves was far more reliable and is

therefore recommended for commercial application. Artificial incubation is however useful

for research purposes (e.g. where different treatments need to be tested on the same egg

batch).

Although individual females achieved high fecundities and fertilization rates, overall

reproductive efficiency (i.e. the percentage of females that hatched viable larvae) was rather

low (maximum 31 % in ponds). A possible reason might be that the source of the

broodstock was restricted to inshore regions. Broodstock from offshore water might be more

mature and therefore more efficient.

In conclusion, year-round maturation and spawning of wild breeders of S.

paramamosain for research and pilot production can easily be achieved, especially when

uni-lateral eyestalk ablation is applied. In order to avoid further pressure on this valuable

resource the use of wild breeders should however be discouraged and research efforts

should be directed towards full domestication of the species. In this respect, further research

is warranted on all aspects of maturation and fertilization in captive conditions, investigating


154

dietary requirements and suitable rearing conditions to trigger maturation and spawning in a

more natural way.

3. Optimal feeding for the larvae (Chapter 4)

Rotifers and Artemia as suitable live feed

It is well known that most species of brachyuran crabs have pelagic larvae that are

planktotrophic and have to obtain nutrition from external sources to survive (Lehto et al.,

1998). The supply of a readily available and digestible feed is therefore crucial for larvae at

first feeding. Copepods (Acartia tsuensis and Pseudodiaptomus spp.), which are

nutritionally superior to rotifers and Artemia (Delbare et al., 1996) have been used as live

feeds in aquaculture with some success (Toledo et al., 1998). Copepods are however

difficult to culture consistently at high densities (Delbare et al., 1996). Rotifers and Artemia

are, for the time being, the most practical live foods available to hatcheries (Støttrup and

McEvoy, 2003).

Artemia cysts are commercially available and are very convenient and easy to use.

Therefore, Artemia nauplii are preferred to rotifers, which require continuous culture and

are sometimes subject to crashes. For that reason rotifers are generally replaced by bigger

live prey as soon as possible.

Early zoeae of S. paramamosain might be more rotifer-dependent compared to other Scylla

species

All publications to date have agreed that rotifers are the best live feed for mud crab

(Scylla sp.) larvae at early stages (Z1 - Z2) and Artemia appear to be better than rotifers for

later stages. A comparative study of replacing rotifers with Artemia at every zoeal stage

showed that larvae initially fed with rotifers but then shifted to Artemia at Z2/Z3 or fed a

mixed diet at Z3 gave best over all zoeal survival (Zeng and Li, 1999). Analysing the dry

weight, carbon, nitrogen and hydrogen content of larvae fed rotifers and Artemia revealed

that rotifers can meet larval development requirements at early zoeal stages, but Artemia

should replace rotifers from Z3 onwards as to meet the increasing nutritional demand,

especially around the time of the first metamophosis (Zeng, 1987, cited in Li et al., 1999).


155

However, in a study of Davis (2003), S. serrata larvae benefited from the addition of

Artemia to the diet right from the zoea 1 stage. Larvae were also reared with good results

(up to 40 % to megalopa) on a diet of Artemia nauplii alone (Davis, 2003). This seems

however not possible for S. paramamosain. Our measurements of the diameter of newly-

spawned eggs demonstrated a considerable smaller egg size for S. paramamosain (288 ± 10

µm) than for S. serrata (313 ± 1 and 317 ± 1 µm for eggs spawned from ablated and intact

females, respectively) (Mann et al., 1999a). In comparison with other data sources (Quinitio,

pers. com.), S. paramamosain seems to have the smallest size of newly-extruded eggs

among the four Scylla species. A larger egg diameter probably coincides with an increase in

Z1 size, which results in a higher capability to catch larger prey like Artemia. From this it

becomes clear that differences between the species of the genus Scylla might exist. The right

time to shift to Artemia was therefore re-investigated.

Our results indicated that for S. paramamosain, rotifers were the most appropriate live

feed for the Z1 and Z2 stages. Crab larvae appeared capable of catching instar-1 Artemia

nauplii only from the Z2 stage onwards and the number of prey consumed increased in the

consequent larval stages. Based on our data, Artemia are best introduced at the Z2 stage

already in order to maximize survival and growth. Optimal feeding schedules should

however also take into account the nutritional composition of both live foods and the quality

of the larvae. In this respect, it was noticed that the ability of Z2 to catch instar-1 Artemia

might vary between batches of larvae. In our pilot-scale production trials (0.5 - 4 m3) it was

observed that feeding HUFA-rich rotifers for both Z1 - Z2 with is a good compromise to

make sure all larvae (irrespective of batch quality) can take up sufficient food. Moreover, a

population of crab larvae does not develop synchronously and always consists of a mixture

of stages, especially in large-scale commercial applications. Crab larvae seem to adapt to a

new food source quite easily and therefore an overlap in feeding rotifers and Artemia seems

not really necessary. Although micro-algal cells were not the proper initial feed for early

crab stages, they proved to be useful for stabilizing the quality of the rotifers and to maintain

good water quality.

Alternatives for rotifers

Despite recent advances, rotifer culture is still labour intensive and vulnerable to

periodic and unpredictable crashes (Fu et al., 1997; Suantika, 2001). Therefore, a rotifer-free


156

feeding schedule is always preferable in hatcheries. Although it resulted in lower survival

compared to treatments where rotifers were fed, the results indicated that live umbrella-

stage Artemia were the best replacement for rotifers for feeding Z1 - Z2. In this way,

umbrella-stage Artemia might be a valuable alternative feed in case of rotifer shortage.

Umbrella Artemia could e.g. be supplemented to the diet after a few days of feeding rotifers.

The non-selective feeding at early crab larval stages also indicates there might be

possibilities to develop artificial diets as alternatives (or supplement) for live feed, in order

to reduce the dependency on rotifer cultures.

4. Live feed quality (Chapter 5)

In captivity, rotifers and Artemia nauplii support growth and survival of mud crab

larvae, but this simplified diet is not ideal. The phenomenon of moult death syndrome

(MDS) - high mortality during or after the moult from Z5 to megalopa, has been

encountered repeatedly. Poor nutrition, even if confined to the early larval stages, has been

suggested as a cause for MDS (Hamasaki et al., 2002a, 2002b; Li et al., 1999; Mann et al.,

2001; Marichamy and Rajapackiam, 2001; Quinitio et al., 2001; Suprayudi et al., 2002b;

Zeng and Li, 1999). Inferior nutrition may also be a factor contributing to the highly

variable survival and the high susceptibility to disease often recorded in mud crab

larviculture. The nutritional quality of rotifers and Artemia can be improved by enriching

them with nutrients in a process known as bioencapsulation (Coutteau et al., 1997;

Kanazawa and Koshio, 1994; Rees et al., 1994; Wouters et al., 1997). The effect of

enriching the live food with essential fatty acids (EFAs) contained in formulated emulsions

on survival and growth of mud crab larvae has been tested in this study.

The fatty acid nutrition of crustaceans is unique and can be grouped according to the

following (Tacon, 1987): (i) fatty acids that can be synthesized de novo from acetate are

termed non-essential fatty acids. Animals possess a ∆-9-desaturase-enzyme system that can

Role of essential fatty acids in crustaceans

Lipids are a source of essential fatty acids, which in turn are essential for the

maintenance and integrity of cellular membranes, they are a major source of energy, and are

precursors of the prostaglandin hormones (Tacon, 1987).


157

convert saturated fatty acids into mono-unsaturated forms. The most abundant saturated and

unsaturated fatty acids are palmitic (16:0), stearic (18:0), arachidic (20:0) and oleic acid

(18:1n-9), respectively; (ii) essential fatty acids (EFA) composed of the linoleic acid (18:2n-

6) and linolenic acid (18:3n-3) series of PUFAs (poly unsaturated fatty acids). Crustaceans

cannot synthesize de novo these fatty acid (Kayama et al., 1980); and (iii) highly

unsaturated acids (HUFAs) also composed of the essential linoleic and linolenic series, but

with a chain length of more than 20 carbon atoms and more than 3 unsaturated bonds. They

can be only partially biosynthesized from linoleic acid or linolenic acid. The most common

HUFAs are docosahexaenoic acid (22:6n-3, DHA), eicosapentaenoic acid (20:5n-3; EPA)

and arachidonic acid (20:4n-6; ARA) (Kanazawa et al., 1979a).

The absence of the de novo synthesis of linoleic, linolenic, DHA and EPA was first

recognized in Penaeus japonicus (Kanazawa et al., 1977a, 1977b), P. monodon and P.

merguiensis (Kanazawa et al., 1979b). Chanmugam et al. (1983) reported that the linoleic

acid series (n-6) predominated in the freshwater prawn lipids whereas the linolenic acid

series (n-3) predominated in marine shrimp lipids. The difference between freshwater and

marine species has been widely attributed to a possible deficiency or impairment of the

enzyme ∆-5-desaturase in marine species, thereby rending them incapable of performing the

necessary conversion (Sargent et al., 1989; Drevon, 1992). DHA and EPA are derived from

dietary linoleic and linolenic acid through desaturation and carbon chain elongation

processes with presence of three enzymes (∆-6, 5, 4, desaturase) and the addition of 2

carbon atoms.

DHA and EPA are essential compounds required for cell membrane formation,

osmoregulation, the synthesis of prostaglandins and they also appear to have an activating

role in the immune system (Léger and Sorgeloos, 1992). Feeding crustacean larvae with

DHA and EPA enriched diets leads to improved survival and growth rates (Bengtson et al.,

1991; Kontara et al., 1995; Levine and Sulkin, 1984a).

Effects of DHA, EPA and ARA on mud crab larvae

No real differences in survival in the zoeal stages were found between the different

enrichment treatments tested here. The “nutritional impact” of HUFAs on the zoeal survival

was probably obscured by other more decisive factors such as the batch quality, micro-biota,

zootechnics. The significantly lower metamorphosis success in the low HUFA treatments


158

proved however that HUFA-rich emulsions are needed to attain high survival to the crab

stage. Not that much the total n-3 level, but more particularly the DHA level and the

DHA/EPA ratio seem of crucial importance.

Larval development rate was very much affected by the dietary n-3 HUFA level and

DHA/EPA ratio. During early zoeal stages, significantly faster growth was obtained when

the larvae were fed live feed enriched with an emulsion containing 30 % total n-3 HUFA.

Emulsions with a high (50 %) or low total HUFA content on the other hand, either tended to

decrease survival or resulted in low growth rates. The DHA/EPA ratio of the live feed,

rather than the total HUFA content, became increasingly important during the later zoeal

stages. Overall, an emulsion with moderate total n-3 HUFA content (30 %) and high

DHA/EPA ratio (4) resulted in the best overall performance in terms of survival and larval

development for all zoeal stages. However, the optimal DHA/EPA ratio of live food

enrichment emulsions for early stages (Z1 - Z3) might be lower than 4 and needs to be

verified. During the later stages, also metamorphosis success was strongly correlated with

the DHA/EPA ratio of the live feed and the crab larvae. It is therefore recommended that an

emulsion with approximately 30 % total n-3 HUFA and a DHA/EPA ratio of 4 should be

used to enrich live feed from the Z3 stage to attain high survival to megalopa and first crab

stage. There was also evidence that DHA/EPA requirements might change during

development. Therefore it might be better if this ratio is increased gradually in time (e.g.

emulsions with DHA/EPA ratio of 1 for Z1-Z2 stages, of 2 - 3 for Z3 - Z4 and of 4 for Z5

onwards). Further research remains necessary on the suitable n-3 HUFA level and

DHA/EPA ratio for each larval stage.

Selecting an efficient rearing system

Supplementation of arachidonic acid had no effect on survival or growth during the

zoeal stages. First metamorphosis rate was however improved by the addition of dietary

ARA. Further research on the suitable levels and ratios of ARA in the enrichment diet for

crab larvae is therefore worth pursuing.

5. Rearing systems and other zootechnics (Chapter 6)

When starting larval rearing research of a new species, the rearing system is often

designed as simple as possible in order to focus on more immediate concerns such as first


159

feeding, environmental parameters, etc. Typically, these systems are small in volume (from

a few-hundred ml vials or petri dishes to not more than 5-l containers). In these systems

clear water is normally used and larvae are counted and transferred individually by means of

pipette to clean containers daily. This type of system can be referred to as a clear-water

batch system (Clear-Batch). This system is very fit for fundamental research because the

larvae can be observed carefully and counted one by one. In this respect, a 5000-l tank is a

single unit and statistically it gives no more information than a 3-l pot (Blackshaw, 2001).

For the rearing of Scylla larvae, when antibiotics are used, this system has proven to

produce the highest survival of up to 60 - 90 % to megalopa stage (see Chapter 2) or 48 %

to first crab stage (see Chapter 5, experiment 4). The development of antibiotic resistance by

pathogenic bacteria is however considered to be one of the most serious risks to human

health at the global level (FAO, 2002). Furthermore, upscaling of this system is not practical

due to space requirements needed to treat and store seawater, and the high risk for

horizontal transfer of diseases.

Water quality has been identified to be one of the most important factors affecting

success of marine shrimp hatcheries (Bray and Lawrence, 1992). Due to its simplicity, the

clear-water batch system has been adopted well by researchers and producers for both larval

rearing and growout culture. Therefore most shrimp farmers prefer to set up their facilities

in coastal areas where high quality (oceanic) seawater is not limited. Such areas are however

not easily found anymore, except in remote areas where usually logistic problems are

manifold. Furthermore, these areas may also occasionally be influenced by run-off

containing agricultural chemicals or the presence of endemic or introduced pathogens (i.e.

bacteria and virus) that affect the health of shrimp. Viral diseases have had a significant

economic impact on the shrimp industry worldwide (Kalagayan et al., 1991; Wyban et al.,

1993). The aquaculture industry also faces growing pressure to operate under strict

environmental safety standards (e.g. on effluent discharge).

With respect to the above impediments, closed, recirculating seawater systems offer

an opportunity. The advent of these systems has been identified as a priority for shrimp

culture research (Lawrence and Lee, 1997). Water exchange is minimized in these systems

through the use of biological, chemical, and/or mechanical filtration to maintain proper

water quality. These systems are designed to maintain a high biological carrying capacity in

a relatively little space (Quillere et al., 1993; Twarowska et al., 1997).


160

Considering the limitations of good quality seawater in the Mekong delta, a simple

clear-water recirculation system (Clear-Recirc) for rearing mud crab larvae has been tested.

In this system, the water was recirculated over a central biofilter to cope with the nitrogen

waste produced by the larvae. Larval rearing results in this system tended to be better and

more reliable compared to the batch system due to improved and more stable water quality

and the reduction of stress to crab larvae, which are more sensitive than shrimp.

In tropical areas, outdoor “abandoned and green” ponds or tanks sometimes have an

unexpected higher production than carefully managed clean ones. Reasons for failure of the

latter are most of the time attributed to bad management (e.g. excess feeding or fertilizing,

too high stocking densities, chemical application, etc.). It is true however that these “green”

ponds often are “ecologically balanced”. Where modern aquaculture has rejected the

“green” techniques for a long time, in recent years a renewed interest in these systems has

emerged.

Several recent studies pointed out the potential beneficial roles of micro-algae in

aquaculture rearing systems, such as maintaining the quality of live feed, stabilizing pond

water quality via either ammonia uptake or oxygen production and producing natural

antibiotics. The exact mode of action (e.g. source of micro-nutrients, source of

immunostimulants, water quality conditioner, and microbial conditioner) of “green-water”

(i.e. high concentrations of selected species of micro-algae) in the commercial larviculture

of several species of marine fish remains however unclear and therefore requires further

study (Sorgeloos, 1995).

When applied to crab larvae, the green-water batch system (Green-Batch) obtained

quite high survival rates in some experiments. Results were however variable and

sometimes also culture crashes were experienced due to the die-off of blooming algae

resulting in an increase of ammonium and nitrite in the rearing tanks.

In an attempt to combine the beneficial features of both clear- and green-water rearing

systems, two “mixed systems” were designed. Supplementation of low concentrations of

micro-algae into the indoor clear-water batch system (Algae-Batch) normally resulted in an

algal collapse and the formation of a harmful biofilm due to the low light intensities applied.

Micro-algal supplementation into the outdoor clear-water recirculating systems (Algae-

Recir) on the other hand seemed to improve the survival rates, especially in the later larval

stages. Best results were obtained when operating a green-water batch culture system at the

Z1 and Z2 stages and then gradually start to recirculate the water in later stages. Low


161

densities of micro-algae can still be added during this last phase. This system was called a

green-water recirculation system (Green-Recir). The application of a green-water batch

mode for the early larval stages simplified the daily maintenance of the rearing tanks and

probably provides a type of “mature water” for the sensitive early stage larvae. In this

respect, there is increasing evidence that a process of microbial maturation can control the

composition of the bacterial flora of seawater (Skjermo and Vadstein, 1999). Settling of the

water for a number of days, results to a diverse bacterial flora, dominated by non-

opportunists, that act as a stable buffering system restricting the growth of opportunistic and

potentially pathogenic bacteria (Blackshaw, 2001).

In conclusion it should be mentioned that the selection of optimal rearing systems

largely depends on local conditions and practices should therefore be flexible. For example,

during 2000 - 2002, a project on mud crab S. paramamosain larviculture was carried out at

the Research Institute for Aquaculture III (RIA3) in cooperation with the Bribie Island

Aquaculture Research Centre (BIARC) in Australia and the Gondol Research Institute for

Mariculture (GRIM) in Indonesia. A pilot hatchery was located in Nha Trang province,

Central Vietnam with good access to ocean quality seawater. In this project, a clear water

batch rearing system has been applied in 500 to 4000-l tanks with similar results compared

to our work. Still other places have used flow-through systems with good larval survivals.

Other zootechnics

Even the best-designed rearing system can fail to produce high survival and good-

quality larvae if some other basic requirements are not met. Proper larval rearing densities

and feeding practice for rotifers and Artemia are discussed in Chapter 6. A larval rearing

density of 100 Z1 l-1 systematically produced good results. Feeding rotifers at 45 ml-1 the

first 6 days and Artemia nauplii at 20 ml-1 from day 6 (Z3 stage) proved to be the best

feeding schedule. However in practice, the best procedure for feeding was to divide the

daily ration into smaller portions with several feedings per day after checking the remaining

feed in the water and the ration should be increased by larval stages (i.e. 30 - 45 rotifers

ml-1 for Z1 and Z2 and 5 - 10 (meta) Artemia ml-1 for Z3 onwards) might also be more

practical.


162

Using the rearing systems developed in our study, microbial interactions (especially

for early stages) and cannibalism (especially at megalopa and crab stages) remain the most

important causes of high mortality.

Experiments 7 and 8 in Chapter 6 proved that bacteria are one of the most serious

causes for high mortality of early crab larvae. It can be safely assumed that all inputs

(seawater, broodstock, live feed and daily management in hatcheries) into the culture tank

are potential sources of infection (Blackshaw, 2001). Applying antibiotics gave the best

larval survival in both experiments, but the use of these chemicals is nowadays strictly

regulated in aquaculture. Formalin could increase survival of the larvae slightly, but was not

as effective as antibiotics. Strict hygiene at all steps of the rearing process are therefore

advised. However, this advice is not always followed, especially in backyard hatcheries

where facilities are rather primitive and management is not always optimal. Therefore,

zootechnics should be optimized to overcome these problems and avoid the need for

chemicals (antibiotics and formalin).

Aside from further improving live feed quality (e.g. HUFA enrichment) to meet all

nutritional requirements of the larvae, the biggest challenge for crab larval rearing will

probably be the development of safe and sustainable prophylactic treatments to prevent

microbial interactions. In this respect, the use of ozone and probiotics are two promising

techniques. The microbial flora will need to be controlled in future production systems and

there is evidence that this can be achieved by using a recirculating system in which an ozone

Cannibalism during the megalopa stage accounted in some experiments for as much as

50 % of the mortality. The best solution to overcome this is to supply an appropriate

substrate and to feed larger prey such as juvenile Artemia. Using larger tanks or earthen

ponds to lower the density for advanced larvae might also be useful.

6. Overall conclusions and suggestions for further research

Broodstock rearing of the mud crab S. paramamosain does not pose any special

difficulties. Further research is however needed to accomplish complete domestication of

the species. Another area that also deserves further attention is the development of proper

broodstock diets in order to optimize maturation and mating in captivity and improve larval

quality.


163

treatment is combined with the inoculation of the biological filter with selected nitrifying

and probiotic bacteria (Gatesoupe, 1991; Rombaut et al., 2001).

Mud crab larvae seem to accept inert feeds quite easily. Recently, Genodepa et al.

(2004) have shown that microbound diets incorporated with 14C-labelled rotifers are readily

ingested by S. serrata zoeae and megalopae. Developing formulated feeds to replace

partially the live feed could help to optimize nutrition and also may reduce the risk of

pathogenic infection for crab larvae.

More attention should also be paid to the megalopa and crab stages because these

stages require a high quality diet to metamophose to first crab and grow on to larger-sized

crabs that can be stocked in growout ponds. Cannibalism could be sufficiently reduced by

investigating suitable substrates, transfer of animals to larger containers and supplying

larger size preys like Artemia juveniles or adults.

Using the methodology described in different chapters of this thesis, recent large scale

trials have obtained survival to crab 1 stage of 20 - 30 %. Although several issues still need

to be solved, we may conclude that knowledge of mud crab larval rearing techniques have,

in recent years, evolved to a level comparable to that of shrimp in the early nineties, and

therefore commercial application might be envisaged within a few year time.

References

Agbayani, R.F., 2001. Production economics and marketing of mud crabs in the Philippines. Asian Fisheries Science 14 (2), 201-210.

Andrews, J.H., Harris, R.F., 1986. r- and K-selection in microbial ecology. Advances in Marine Biology 9, 99-147.

Angell, C.A. (Ed.), 1992. Report of the Seminar on Mud Crab Culture and Trade. Surat Thani, Thailand, 5-8 November 1991. Bay of Bengal Programme, Madras, India. BOBP/REP/51, 246 pp.

Anger, K., Dawirs, R.R., Anger, V., Costlow, J.D., 1981. Effects of starvation periods on zoeal development of Brachyuran crabs. Biological Bulletin 161 (2), 199-212.

Armstrong, D.A., Buchanan, D.V., Caldwell, R.S., 1976. A mycosis caused by Lagenidium spp. in laboratory reared larvae of the Dungeness crab, Cancer magister, and possible chemical treatments. Journal of Invertebrate Pathology 28, 329-336.

Arriola, F. J., 1940. A preliminary study of the life history of Scylla serrata (Forksal). Philippines Journal of Science 73 (4), 437-456.

Ashraf, M., 1993. Effects of dietary fatty acid enrichment on survival, growth and salinity-stress-test performance of inland silversides. The progress fish culturist 55, 280-283.

Bachère, E., Mialhe, E., Noel, D., Boulo, V., Morvan, A., Rodríguez, J., 1995. Knowledge and research prospects in marine mollusc and crustacean immunology. Aquaculture 132 (1-2), 17-32.

Barnes, D.K.A., Dulvy, N.K., Priestley, S.H., Darvall, W.R.T., Choisel, V., Whittinggton, M., 2002. Fisheries characteristics and abundance estimates of the mangrove crab Scylla serrata in southern Tanzania and northern Mozambique. South African Journal of Marine Science 24, 19-25.

Baylon, J., Failaman, A., Tanate, J., 2001a. Comparison of different feeding schemes and feeding density on survival and development of the mud crab Scylla serrata zoea. In: Book of Abstracts of 2001 Workshop on Mud Crab Rearing, Ecology and Fisheries. Institute for Marine Aquaculture, Can Tho University, Vietnam, 8-10 January 2001, pp. 14-15.

Baylon, J.C., Failaman, A.N., 1997. Survival of mud crab Scylla oceanica from zoea to megalopa when fed the rotifer Brachionus sp. and brine shrimp Artemia nauplii. University of the Philippines Visayas. Journal of Natural Science 2, 9-16.

Baylon, J.C., Failaman, A.N., 1999. Larval rearing of mud crab Scylla serrata in the Philippines. In: Keenan, C.P., Blackshaw, A. (Eds.). Mud crab Aquaculture and Biology. Proceedings of an International Scientific Forum. Darwin, Australia, 21-24 April 1997. ACIAR Proceedings No. 78, 141-146.

Baylon, J.C., Failaman, A.N., 2001. Broodstock management and larval rearing protocols for the mudcrab, Scylla serrata (Keenan et al. 1998) developed at the UPV hatchery. In: Book of Abstracts of 2001 Workshop on Mud Crab Rearing, Ecology and Fisheries. Institute for Marine Aquaculture, Can Tho University, Vietnam, 8-10 January 2001, p. 10.

Baylon, J.C., Failaman, A.N., Vengano, E.L., 2001b. Effect of salinity on survival and metamorphosis from zoea to megalopa of the mud crab Scylla serrata Forskål (Crustacea: Portunidae). Asian Fisheries Science 14 (2), 143-151.

Baylon, J.C., Bravo, M.E.A., Manigo, C., 2004. Ingestion of Brachionus plicatilis and Artemia salina nauplii by mud crab Scylla serrata larvae. Aquaculture Research 35, 62-70.

Bell, J.G., McEvoy, L.A., Estévez, A., Shields, R.J., Sargent, J.R., 2001. Optimizing highly unsaturated fatty acid levels in first feeding marine fish larvae. In: Hendry, C.I., Van Stappen G., Wille, M., Sorgeloos, P. (Eds.). Larvi’01: Fish and Shellfish Larviculture Symposium. European Aquaculture Society. Special Publication No 30, Oostende, Belgium, pp. 54-57.

Bell, J.G., Sargent, J.R., 2002. Arachidonic acid in aquaculture feeds: current status and future opportunities. Aquaculture 2002 (in press), 1-9.

References

166

Benfield, M.C. and Downer, R.G. 2001. Spatial and temporal variability in the nearshore distributions of postlarval Farfantepenaeus aztecus along Galveston Island, Texas. Estuarine Coastal and Shelf Science 52 (4), 445-456.

Blackshaw, A.W., 2001. Larval culture of Scylla serrata: Maintenance of hygiene and concepts of experimental design. Asian Fisheries Science 14 (2), 239-242.

Chaitanawisuti, N., Krittsanapuntu A., 1998. Fishery status of Portunid crabs in Thailand 1988-1995. In: Extended Abstracts of International Forum on the Culture of Portunid Crabs. Boracay, Philippines, 1-4 December 1998, p. 17.

Chong, L.P., 1993. The culture and fattening of mud crabs. Infofish International 3, 46-49.

Bengtson, D.A., Léger, P., Sorgeloos, P., 1991. Use of Artemiu as a feed source for aquaculture. In: Browne, R.A., Sorgeloos., P., Trotman, C.N.A. (Eds.). Artemia Biology, pp. 255-285.

Benson, S.M., 1998. Reservoirs of resistence. Today’s Life Science 10 (2).

Bigford, T.E., 1978. Effect of several diets on survival, development time, and growth of laboratory-reared spider crab, Libinia emarginata, larvae. Fishery Bulletin 76 (1), 59-64.

Bray, W.A., Lawrence, A.L., 1992. Reproduction of Penaeus species in captivity. In: Fast, A., Lester, L.J. (Eds.). Culture of Marine Shrimp: Principles and Practices. Elsevier Scientific Publishing Co., Amsterdam, Netherlands, pp. 93-170.

Brick R.W., 1974. Effect of water quality, antibiotics, phytoplankton and food on survival and development of larvae of Scylla serrata (Crustacea: Portunidae). Aquaculture 3, 231-244.

Browdy, C.L., 1992. A review of the reproductive biology of Penaeus species: perspectives on controlled shrimp maturation systems for high quality nauplii production. In: Wyban, J. (Ed.). Proceedings of the Special Session on Shrimp Farming. World Aquaculture Society, Baton Rouge, LA, USA, pp. 22–51.

Caers, M., Coutteau, P., Lombeida, P., Sorgeloos, P., 1998. The effect of lipid supplementation on the growth and fatty acid composition of Tapes philippinarum (L.) spat. Aquaculture, Amsterdam 162, 287-299.

Camacho, A.S., Aypa, S.M., 2001. Research needs and data on production of Portunid crabs in the Philippines. Asian Fisheries Science 14 (2), 243-245.

Carpenter, K.E., Niem, V.H., 1998. The living marine resources of the Western Central Pacific. Volume 2, Cephalopods, crustaceans, holuthurians and sharks. FAO Species Identification Guide for Fishery Purposes. FAO, Rome.

Castell, J.D., Bell, J.G., Tocher, D.R., Sargent, J.R., 1994. Effects of purified diets containing different combinations of arachidonic and docosahexaenoic acid on survival, growth and fatty acid composition of juvenile turbot (Scophthalmus maximus). Aquaculture 128, 315-333.

Chanmugam, P., Donovan, J., Wheeler, C.J., Hwang, D.H., 1983. Differences in the lipid composition of fresh water prawn (Macrobrachium rosenbergii). Comparative Biochemistry and Physiology 48, 1440-1443.

Chen, L.L., Lo, C.F., Chiu, Y.L., Chang, C.F., Kou, G.H., 2000. Natural and experimental infection of white spot syndrome virus (WSSV) in benthic larvae of mud crab Scylla serrata. Diseases of Aquatic Organisms 40 (2), 157-161.

Cheng,Y.X., Li, S.J., 2001.The variations of lipid composition of the crab Scylla serrata during the embryonic development. Acta Oceanologica Sinica 23, 79-90 (in Chinese).

Chen, H.C., Jeng, K.H., 1980. Study on the larval rearing of mud crab Scylla serrata. China Fisheries Montly 329, 3-8.

Cholik, F., 1999. Review of mud crab culture research in Indonesia. In: Keenan, C.P., Blackshaw, A. (Eds.), Mud Crab Aquaculture and Biology. Proceedings of an International Scientific Forum. Darwin, Australia, 21-24 April 1997. ACIAR Proceedings No 78, 14-20.

Cholik, F., Hanafi, A., 1992. A review of the status of mud crab (Scylla sp.) fishery culture in Indonesia. In: Angell, C.A. (Ed.). Report of the seminar on the mud crab culture and trade. Surat Thani, Thailand, 5-8 November 1991. Bay of Bengal Programme, Madras, India. BOBP/REP/51, pp. 13-27.

Churchill, G.J., 2003. An investigation into the captive spawning, egg characteristics and egg quality of the mud crab (Scylla serrata) in South Africa. MSc thesis. Department of Ichthyology and Fisheries

References

167

Science. Rhodes University, Grahamstown, South Africa, 111 pp.

Chythanya, R., Savan, R., 1999. Biofilm in fisheries. INFOFISH International 5/99, Sep/Oct, pp. 42-44.

Clark, D.R., Aazem, K.V., Hays, G.C., 2001. Zooplankton abundance and community structure over a 4000 km transect in the north-east Atlantic. Journal of Plankton Research 23 (4), 365-372.

Coutteau, P., Castell, J.D., Ackman, R.G., Sorgeloos, P., 1996. The use of lipid emulsions as carriers for essential fatty acids in bivalves: a test case with juvenile Placopecten magellanicus. Journal of Shellfish Resources 15, 259-264.

Coutteau, P., Geurden, I., Camara, M.R., Bergot, P., Sorgeloos, P., 1997. Review on the dietary effects of phospholipids in fish and crustacean larviculture. Aquaculture 155, 149-164.

Cowan, L., 1984. Crab farming in Japan, Taiwan and the Philippines. Queensland Information Series (Department of Primary Industries, Queensland, Australia), 85 pp.

Dat, H.D, 1992. Biology and culture of mud crab. A training course on aquaculture in the Mekong delta, Vietnam, 20 pp. (in Vietnamese).

Dat, H.D., 1999a. Description of mud crab (Scylla spp.) culture methods in Vietnam. In: Keenan, C.P., Blackshaw, A. (Eds.). Mud crab Aquaculture and Biology. Proceedings of an International Scientific Forum. Darwin, Australia, 21-24 April 1997. ACIAR Proceedings No. 78, 67-71.

Dat, H.D., 1999b. Preliminary studies on rearing of the larvae of the mud crab (Scylla paramamosain) in South Vietnam. In: Keenan, C.P., Blackshaw, A. (Eds.). Mud crab Aquaculture and Biology. Proceedings of an International Scientific Forum. Darwin, Australia, 21-24 April 1997. ACIAR Proceedings No. 78, 147-152.

Dau, D.V., 1998. The culture of Scylla species in Vietnam. In: Extended Abstracts. International Forum on the Culture of Portunid Crabs. Boracay, Philippines, 1-4 December 1998, p. 12.

Davis, J.A., 2003. Development of hatchery techniques for the mud crab Scylla serrata (Forskal) in South Afica. Ph.D. thesis. Faculty of Agricultural and Applied Biological Sciences, Ghent University, Belgium, 163 pp.

Davis, J.A., Churchill, G.J, Hecht, T., 2003. Spawning characteristics of the South African mud crab Scylla serrata (Forskål) in captivity. Journal of the World Aquaculture Society, in press.

Davis, J.A., Dermaux, V., Wille, M., 2001. Larval rearing protocols for the South African mudcrab, Scylla serrata as a basis for nutritional studies. In: Book of Abstracts of 2001 Workshop on Mud Crab Rearing, Ecology and Fisheries. Institute for Marine Aquaculture, Can Tho University, Vietnam, 8-10 January 2001, p. 16.

Delbare, D., Dhert, P., Lavens, P., 1996. Zooplankton. In: Lavens, P., Sorgeloos, P. (Eds.). Manual on the production and use of live food for aquaculture. FAO technical paper. Food and Agricultural Organization, Rome, pp. 321-360.

Dhert, P., 1996. Rotifers. In: Lavens, P., Sorgeloos, P. (Eds.). Manual on the production and use of live food for aquaculture. FAO technical paper No. 361, 61-100.

Dhert, P., Rombaut, G., Suantika, G., Sorgeloos, P., 2001. Advancement of rotifer culture and manipulation techniques in Europe. Aquaculture 200, 129-146.

Dhert, P., Sorgeloos, P., Devresse, P., 1993. Contribution towards a specific DHA enrichment in the live food Brachionus plicatilisi and Artemia sp. In: Reinertsen, H., Dahle, L.A., Jorgensen, L., Tvinnereim, K. (Eds). Fish Farming Technology. Balkema, Rotterdam, Netherlands, pp. 109-115, 482 pp.

Djunaidah, I.S., Mardjono, M., Lavens, P., Wille, M., 1998. Effects of light and feeding regimen on culture performance of mud crab (Scylla spp.) larvae. In: Extended Abstracts of International Forum on the Culture of Portunid Crabs. Boracay, Philippines, 1-4 December 1998, p. 26.

Djunaidah, I.S., Mardjono, M., Wille, M., Kontara, E.K., Sorgeloos, P., 2001a. Investigations on standard rearing techniques for mass production of mud crab Scylla spp. seed. A research review. In: Book of Abstracts of 2001 Workshop on Mud Crab Rearing, Ecology and Fisheries. Institute for Marine Aquaculture, Can Tho University, Vietnam, 8-10 January 2001, p. 11.

Djunaidah, I.S., Suwoyo, D., Wille, M. Kontara, E.K., Sorgeloos, P., 2001b. Investigations on the reproductive performance of mud crab Scylla spp. broodstock: A research review. In: Book of Abstracts of 2001

References

168

Workshop on Mud Crab Rearing, Ecology and Fisheries. Institute for Marine Aquaculture, Can Tho University, Vietnam, 8-10 January 2001, pp. 6-7.

Djunaidah, I.S., Wille, M., Kontara, E.K., Sorgeloos, P., 2003. Reproductive performance and offspring quality in mud crab (Scylla paramamosain) broodstock fed different diets. Aquaculture International 11 (1-2), 3-15.

Dorairaj, K., Roy, S.D., 1996. Mudcrab culture in the Andamans. In: Krishnamoorthi, B., Krishnamoorthy, K.N., Meenakshisundaram, P.T., Nayar, K.N. (Eds.). Proceedings of the Seminar on Fisheries. Madras, India, 17-19 August 1995, p. 136.

Drevon, C.A., 1992. Marine oils and their effects. Nutrition Reviews 50, 38-45.

Du Plessis, A., 1971. A preliminary investigation into the morphological characteristics, feeding, growth, reproduction and larval rearing of Scylla serrata Forskål (Decapoda: Portunidae) held in captivity. Fisheries Development Corporation of South Africa. Unpublished report, 24 pp.

Dutil, J.D., Munro, J., Peloquin, M., 1977. Laboratory study of the influence of the prey size on the vulnerability to cannibalism in snow crab (Chionoecetes opilio O. Fabricius, 1780). Journal of Experimental Marine Biology and Ecology 21, 81–94.

Emmerson, W.D., 1980. Induced maturation of prawn Penaeus indicus. Marine Ecology Progress Series 2, 121–131.

FAO, 1999. Aquaculture production statistics 1988-1997. FAO fisheries Circular No. 815, Revision 11, Rome, 203 pp.

FAO, 2002. The state of world fisheries and aquaculture. See http://www.fao.org/docrep/

Estampador, E.P., 1949. Studies on Scylla (Crustacea: Portunidae), I. Revision of the genus. Philippine Journal of Science 78 (1), 95-109.

Estévez, A., McEvoy, L.A., Bell, J.G., Sargent, J.R., 1999. Growth, survival, lipid composition and pigmentation of turbot larvae fed live-prey enriched in arachidonic (ARA) and eicosapentaenoic (EPA) acids. Aquaculture 180, 321–343.

FAO, 1998. Aquaculture production statistics 1987-1996. FAO fisheries Circular No. 815, Revision 10, Rome, 197 pp.

Felix, T.S., Gajendran, M., Subramanian, S, 1995. Aquaculture of mud crab. Seafood Export Journal, Cochin 26 (6), 5-6.

Fingerman, M., 1997. Crustacean endocrinology: a retrospective, prospective, and introspective analysis. Physiological Zoolology 70 (3), 257–269.

Fortes, R.D., 1999. Mud crab research and development in the Philippines: An overview. In: Keenan, C.P., Blackshaw, A. (Eds.). Mud crab Aquaculture and Biology. Proceedings of an International Scientific Forum. Darwin, Australia, 21-24 April 1997. ACIAR Proceedings No 78, 27-32.

Forward, R.B., Cronin, T.W., Stearns, D.E., 1984. Control of dial vertical migration: Photoresponses of a larval crustacean. Linology and Oceanography 29 (1), 146-154.

Fu, Y., Hada, H., Yamashita, T., Toshida, Y., Hino, A., 1997. Development of continuous culture system for stable mass production of the marine rotifer Brachionus plicatilis. Hydrobiologia 358, 145-151.

Fukunaga K., Uzumaki K., 1982. Production of hatchery-raised juvenile crab of Scylla tranquebarica. Saibaigiken 11 (1), 45-53 (in Japanese).

Galois, R., 1987. Les lipides neutres chez les crustacés decapodes: metabolisme et besoins. Océanis 13, 197-215.

García-Ortega, A., Van Hoornyck, A., Segner, H., Coutteau, P., Verreth, J., 1995. Effect of heat treatment on the nutritional quality of decapsulated Artemia cysts as food for African catfish Clarias gariepinus larvae. In: Lavens, P., Jaspers, E., Roelants, I. (Eds.). Larvi’95, Gent. European Aquaculture Society, Special Publication No. 24, 281-284

Gatesoupe, F.J., 1991. The effect of three strains of lactic bacteria on the production rate of rotifers, Brachionus plicatilis, and their dietary value for larval turbot Scopthalmus maximus. Aquaculture 96, 335-342.

References

169

Genodepa, J., Southgate, P.C., Zeng, C., 2004. Diet particle size preference and optimal ration for mud crab, Scylla serrata, larvae fed microbound diets. Aquaculture 230, 493-505.

Glencross, B.D., Smith, D.M., 2001. A study of the arachidonic acid requirements of the giant tiger prawn, Penaues monodon. Aquaculture Nutrition 7 (1), 59-69.

González-Félix, M.L., Gatlin, D.M., Lawrence, A.L., Perez-Velazquez, M., 2002. Effect of various dietary lipid levels on quantitative essential fatty acid requirements of juvenile Pacific white shrimp Litopenaeus vannamei. Journal of the World Aquaculture Society 33(3), 330-340.

Hai, T.N., Hassan, A., Law, A.T., Shazili, V.N.A., 2001. Some aspects on maturation and spawning performance of mud crabs (Scylla spp.) in captive conditions. In: Book of Abstracts of 2001 Workshop on Mud Crab Rearing, Ecology and Fisheries. Institute for Marine Aquaculture, Can Tho University, Vietnam, 8-10 January 2001, p. 8.

Hamasaki, K., Suprayudi, M.A., Takeuchi, T., 2002b. Mass mortality during metamorphosis to megalops in the seed production of mud crab Scylla serrata (Crustacea, Decapoda, Portunidae). Fisheries Science 68 (6), 1226-1231.

Hasegawa Y., 1989. Mud crab. 1987 Annual Report of the Kochi Prefectural fisheries experiment station (in Japanese).

Heasman, M.P., Fielder, D.R., 1983. Laboratory spawning and mass larval rearing of the mangrove crab Scylla serrata (Forskal), from first zoea to first crab stage. Aquaculture 34, 303-316.

Hill, B.J., 1974. Salinity and temperature tolerance of zoea of the Portunid crab Scylla serrata. Marine Biology 25, 21-24.

Hill, B.J., 1978. Activity, track and speed of movement of the crab Scylla serrata (Forskål) in an estuary. Marine Biology 47, 135-141.

Greene, C.H., Landry, M.R., 1985. Patterns of prey selection in the cruising calanoid predator Euchaeta elongate. Ecology 66, 1408-1416.

Hamasaki, K., Hatai, K., 1993a. Experimental infection in the eggs and larvae of the swimming crab Portunus triturbiculatus and the mud crab Scylla serrata with seven fungal strains belonging to Lagenidiales. Bulletin of the Japanese Society of Scientific Fisheries 59 (6), 1059-1066.

Hamasaki K., Hatai, K., 1993b. Prevention of fungal infection in the eggs and larvae of the swimming crab Portunus trituberculatus and the mud crab Scylla serrata by bath treatment with formalin. Bulletin of the Japanese Society of Scientific Fisheries 59 (6), 1067-1072.

Hamasaki, K., Takeuchi, T., Sekiya, S., 1998. Dietary value for larval swimming crab Portunus trituberculatus of marine rotifer Brachionus rotundiformis cultured with several feeds. Bulletin of the Japanese Society of Scientific Fisheries 64 (5), 841-846.

Hamasaki, K., 2002. Effects of temperature on the survival, spawning and egg incubation period of overwintering mud crab broodstock, Scylla paramamosain (Brachyura: Portunidae). Suisanzoshoku 50 (3), 301-308.

Hamasaki, K., Suprayudi, M.A., Takeuchi, T., 2002a. Effects of dietary n-3HUFA on larval morphogenesis and metamorphosis to megalops in the seed production of the mud crab, Scylla serrata (Brachyura: Portunidae). Suisanzoshoku 50 (3), 333-340.

Harvey, A.W., 1996. Delayed metamorphosis in Florida hermit crabs: Multiple cues and constraints (Crustacea: Decapoda: Paguridae and Diogenidae). Marine Ecology Progress Series 141 (1-3), 27-36.

Harvey, E.A., Epifanio, C.E., 1997. Prey selection by larvae of the common mud crab Panopeus herbstii Milne-Edwards. Journal of Experimental Marine Biology and Ecology 217, 79-91.

Heaseman, M.P., 1980. Aspects of the general biology and fishery of the mud crab Scylla serrata (Forskål) in Monreton Bay. Ph.D. thesis, University of Queenland, 506 pp.

Heasman, M.P., Fielder, D.R., Shepherd, R.K., 1985. Mating and spawning in the mud crab Scylla serrata (Forskal). Australian Journal of Marine and Freshwater Research 36, 773-783.

Hill, B.J., 1975. Abundance, breeding and growth of the crab Scylla serrata in two South African estuaries. Marine Biology 32, 119-126.

References

170

Hill, B.J., 1979. Biology of the crab Scylla serrata (Forskål) in the St Lucia system. Transactions of the Royal Society of South Africa 44 (1), 55-62.

Hill, B.J., 1984. Aquaculture of the mudcrab - The potential for aquaculture in the Queenland. Queenland Department of Primary Industries Publcation, pp. 29-45.

Hong, P.N., San, H.T., 1993. Mangroves of Vietnam. IUCN Bangkok, Thailand.

Hudinaga, M., Kittaka, J., 1967. The large scale production of the young kuruma prawn, Penaus japonicus Bate. Information Bulletin of Planktology. Japan, Commemoration number of Dr. Y. Matsue, 35-46.

Huy, B.Q., 2004. Some solutions for sustainable development in aquaculture in the Mekong Delta,Vietnam (in Vietnamese). See www.baocantho.com.vn

Hill, B.J., 1994. Offshore spawning by the portunid crab Scylla serrata (Crustacea: Decapoda). Marine Biology (Berlin) 120 (3), 379-384.

Horiuti, T., Yamamoto Y., 1987. Production of hatchery-raised juvenile crab of Scylla tranquebarica. 1986 Annual Report of Shizuoka Prefectural Thermal Effluent Utilization Research Center, pp. 63-69 (in Japanese).

Howell, B.R., Day, O.J., Ellis, T., Baynes, S.M., 1998. Early stages of farmed fish. In: Black, K.D., Pickering, A.D. (Eds.). The biology of farmed fish. Sheffield Academic Press Ltd. Sheffield, UK, 415 pp.

Hudinaga, M., 1942. Reproduction, development and rearing of Penaeus japonicus Bate. Japanese Journal of Zoology 10, 305-393.

, accessed in March 2004.

ICES, 1997. Report of the Working Group on Marine Fish Culture for the International Council for the Exploration of the Sea. Murcia, Spain 23-26 June 1997. ICES CM 1997/F, p. 7.

Kaji, S., Kanematsu, M., Tezuka, N., Fushimi, H., Hatai, K., 1991. Effects of formalin bath for Haliphthoros infection on ova and larvae of the mangrove crab Scylla serrata. Bulletin of the Japanese Society of Scientific Fisheries 57 (1), 55-51.

Hyland, S.J., Hill, B.J., Lee, C.P., 1984. Movement within and between different habitats by the portunid crab Scylla serrata. Marine Biology 80, 57-61.

IFEP, 1996. Vietnam’s Fisheries: A National Overview of Fisheries and Aquaculture. The First Donor Coordination Meeting for Fisheries. Hanoi, Vietnam, 9 December 1996. Institute for Fisheries Economics and Planning, Ministry of Fisheries, Vietnam.

Irianto, A., Austin, B., 2002. Probiotics in aquaculture. Journal of Fish Diseases 25 (11), 633-642.

Izquierdo, M.S., Fernandez-Palacios, H., Tacon, A.G.J., 2001. Effect of broodstock nutrition on reproductive performance of fish. Aquaculture 197, 25-42.

Jeffs, A.G., Willmott, M.E., Wells, R.M.G., 1999. The use of energy stores in the puerulus of the spiny lobster Jasus edwardsii across the continental shelf of New Zealand. Comparative Biochemistry and Physiology. A-molecular and Integrative Physiology 123 (4), 351-357.

Joel, D.R., Sanjeevaraj, P.J., 1986. Food and feeding of the two species of Scylla (de Haan) (Portunidae: Brachyura). Journal of Marine Biological Association India 28, 178-183.

Johns, D.M., Berry, W.J., McLean, S., 1981. International study on Artemia: XXI. Investigations into why some strains of Artemia are better food sources than others. Further nutritional work with larvae of the mud crab, Rhithropanopeus harrisii. Journal of World Mariculture Society 12 (1), 303-314.

Johnston, D., Keenan, P., 1999. Mud crab culture in the Minh Hai province, South Vietnam. In: Keenan, C.P., Blackshaw, A. (Eds.). Mud Crab Aquaculture and Biology. Proceedings of an International Scientific Forum. Darwin, Australia, 21-24 April 1997. ACIAR Proceedings No 78, 95-98.

Jones, D.A., Le Vay, L., Ut, V.N., Davis, J., Nghia, T.T., O’Kelley, A.G., (in press). Systematics of the larvae of some Scylla species with a key to S. paramamosain and S. serrata larval stage.

Jung, S.H., Kim, J.W., Jeon, I.G., Lee, Y.H., 2001. Formaldehyde residues in formalin-treated olive flounder (Paralichthys olivaceus), black rockfish (Sebastes schlegeli), and seawater.

Jyamanna, S.C., Jinadasa, J., 1993. Size at maturity and spawning periodicity of the mud crab Scylla serrata (Forskål) in the Negombo estuary. Journal of the Natural Scientific Council of Sri Lanka 21(10), 141-152.

References

171

Kalagayan, H., Godin, D., Kanna, R., Hagino, G., Sweeney, J., Wyban, J., Brock, J., 1991. IHHN virus as an etiological factor in runt-deformity syndrome of juvenile Penaeus vannamei culture in Hawaii. Journal of the World Aquaculture Society 22, 235–243.

Kanazawa, A., Teshima, S., Tokiwa, S., 1977a. Nutritional requirements of prawn. VII. Effect of dietary lipids on growth. Bulletin of the Japanese Society of Scientific Fisheries 43, 849-856.

Kanazawa, A., Tokiwa, S., Kayama, M., Hirata, M., 1977b. Essential fatty acids in the diet of prawn. I. Effects of linoleic and linolenic acids on growth. Bulletin of the Japanese Society of Scientific Fisheries 43, 1111-1114.

Keenan, C.P., Davie, P.J.F., Mann, D.L., 1998. A revision of the genus Scylla de Haan, 1833 (Crustacea: Decapoda: Brachchyura: Portunidae). The Raffles Bulletin of Zoology 46 (1), 217-245.

Kittika, J., 1994. Culture of phyllosomas of spiny lobster and its application to studies of larval recruitment and aquaculture. Crustaceana 66 (3), 258-270.

Kontara, E.K., Lavens, P., Sorgeloos, P., 1995. Dietary effects of DHA/EPA on culture performance and fatty acid composition of Penaeus monodon post larvae. In: Lavens, P., Jaspers, E., Roelants, I. (Eds.), Larvi’95 - Fish and Shellfish Larviculture Symposium. Gent, Belgium. European Aquaculture Society, Special Publication No. 24, 204-208.

Latiff, F.A., Musa, C.U.C., 1995. The biology of mud crab and its hatchery production. Bulletin Perikanan. Department of Fisheries, Malaysia, 20 pp.

Kanazawa, A., Teshima, S., Ono, K., 1979a. Relationship between essential fatty acid requirements of aquatic animals and capacity bioconversion linolenic acid to highly unsaturated fatty acid. Comparative Biochemistry and Physiology 63, 295-298.

Kanazawa, A., Teshima, S., Ono, K., Chalayodeja, K., 1979b. Biosynthesis of fatty acids from acetate in the prawn Penaeus monodon and Penaeus merguiensis. Memoirs of the Faculty of Fisheries, Kagoshima University 28, 21-26.

Kanazawa, A., Koshio, S., 1994. Lipid nutrition of the spiny lobster Panulirus japonicus (Decapoda, Palinuridae): a review. Crustaceana 67 (2), 226-232.

Kayama, M., Hirata, M., Kanazawa, A., Tokiwa, S., Sito, M., 1980. Essential fatty acids in the diet of the prawn. III. Lipid metabolism and fatty acid composition. Bulletin of the Japanese Society of Scientific Fisheries 46, 483-488.

Keenan, C.P., 1999a. Aquaculture of the mud crab, genus Scylla - past, present and future. In: Keenan, C.P., Blackshaw, A. (Eds.). Mud crab Aquaculture and Biology. Proceedings of an International Scientific Forum. Darwin, Australia, 21-24 April 1997. ACIAR Proceedings No 78, 9-13.

Keenan, C.P., 1999b. The fourth species of Scylla. In: Keenan, C.P., Blackshaw, A. (Eds.). Mud crab Aquaculture and Biology. Proceedings of an International Scientific Forum. Darwin, Australia, 21-24 April 1997. ACIAR Proceedings No 78, 48-58.

Kobayashi, T., Takeuchi, T., Arai, D., Sekiya, S., 2000. Suitable dietary levels of EPA and DHA for larval mud crab during Artemia feeding period. Bulletin of the Japanese Society of Scientific Fisheries 66 (6), 1006-1013.

Koven, W., Barr, Y., Lutzky, S., Ben-Atia, I., Harel, M., Behrens, P., Weiss, R., Tandler, A., 2000. The effect of dietary arachidonic acid 20:4n-6 on growth and survival prior to and following handling stress in the larvae of gilthead seabream (Sparus aurata). Abstracts of contributions. International Conference Aqua 2000. European Aquaculture Society, Special Publication No. 28, Ostende, Belgium, p. 346.

Lavens, P., Sorgeloos, P., 2000. Experiences on importance of diet for shrimp postlarval quality. Aquaculture 191, 169-176.

Lavilla-Pitogo, C.R., Catedral, D.D., Pedrajas, S.A.G., de la Peña, L.D., 2002. Selection of probiotics for shrimp and crab hatcheries. In: Inui, Y., Cruz-Lacierda, E.R. (Eds.). Disease Control in Fish and Shrimp Aquaculture in Southeast Asia - Diagnosis and Husbandry Techniques. Proceedings of the SEAFDEC-OIE Seminar-Workshop. Iloilo City, Philippines. SEAFDEC Aquaculture Department, Tigbauan, Iloilo, Philippines, pp. 136-150.

Lavilla-Pitogo, C.R., Marcial, H.S., Pedrajas, S.A.G., Quinitio, E.T., Millamena, O.M., 2001. Problems associated with tank-held mud crab (Scylla spp.) broodstock. Asian Fisheries Science 14 (2), 217-224.

References

172

Lawrence, A.L., Lee, P.G., 1997. Research in the Americas. In: D’Abramo, L., Conklin, D., Akiyama, D. (Eds.). Crustacean Nutrition. World Aquaculture Society, Baton Rouge, Louisiana, pp. 566-587.

Le Vay, L., 2001. Ecology and management of mud crab Scylla spp. Asian Fisheries Science 14 (2), 101-111.

Le Vay, L., Ut, V.N., Jones, D.A., 2001. Seasonal abundance and recruitment in an estuarine population of mud crabs, Scylla paramamosain, in the Mekong Delta, Vietnam. Hydrobiologia 449 (1-3), 231-239.

Léger, P., Bieber, G.F., Sorgeloos, P., 1985. International study on Artemia. XXXIII. Promising results in larval rearing of Penaeus stylirostris using a prepared diet as algal substitute and for Artemia enrichment. Journal of the World Aquaculture Society 16, 354-367.

Léger, P., Sorgeloos, P., 1992. Optimized feeding regimes in shrimp hatcheries. In: Fast, A.W., Lester, L.J. (Eds.). Shrimp culture: Principles and Practices. New York, Elsevier Science Publisher, pp. 225-244.

Lehto, J., Sulkin, S., Strom, S., Johnson, D., 1998. Protists and detrital particles as prey for the first larval stage of the brachyuran crab, Hemigrapsus oregonensis. Journal of Experimental Marine Biology and Ecology 230, 213–224.

Lepage, G., Roy, C.C., 1984. Improved recovery of fatty acids through direct transesterification without prior extraction or purification. Journal of Lipid Research 16, 593–600.

Levine, D.M., Sulkin, S.D., 1984a. Nutritional significance of long-chain polyunsaturated fatty acids to the zoeal development of the brachyuran crab, Eurypanopeus depressus (Smith). Journal of Experimental Marine Biology and Ecology 81, 211-223.

Li, S., Zeng, C., Tang, H, Li, F., Wang, G., Cheng, Y., Lin, Q., 1998. Nutritive requirements and metabolic mechanisms during larval development of mud crab Scylla serrata. Journal of Oceanography in Taiwan Strait 17, 9-17.

Levine, D.M., Sulkin, S.D., 1984b. Ingestion and assimilation of microencapsulated diets by brachyuran crab larvae. Marine Biology Letters 5, 147-155.

Li, S., Zeng, C., Tang, H., Wang, G., Lin, Q., 1999. Investigations into the reproductive and larval culture biology of the mud crab, Scylla paramamosain: a research overview. In: Keenan, C.P., Blackshaw, A. (Eds.). Mud crab Aquaculture and Biology. Proceedings of an International Scientific Forum. Darwin, Australia, 21-24 April 1997. ACIAR Proceedings No. 78, 121-124.

Lin, Q., Li, S., Zeng, C., Wang, G., 1994. Experimental studies on the domestication of mud crab spawner. Fujian Fishery, 1991(1), 45-50 (in Chinese).

Liong, P.C., 1992. The fattening and culture of mud crab (Scylla serrata) in Malaysia. In: Angell, C.A. (Ed.). Report of the Seminar on the Mud Crab Culture and Trade. Surat Thani, Thailand, 5-8 November, 1991. Bay of Bengal Programme, Madras, India. BOBP/REP/51, pp. 185-190.

Lumasag, G., Quinitio, E.T., 1998. Ingestion rate of mud crab (Scylla serrata) larvae after delayed feeding. In: Extended Abstracts of International forum on the culture of Portunid Crabs held in Boracay, Philippines, 1-4 December 1998, p. 29.

Makridis, P., Olsen, Y., 1999. Protein depletion of the rotifer Brachionus plicatilis during starvation. Aquaculture 174, 343–353

Malkiel, E., Alquaddoomi, O. and Katz, J., 1999. Measurements of plankton distribution in the ocean using submersible holography. Measurement Science and Technology, 10 (12), 1142-1152.

Mann, D., 2001. The influence of microbiology on the success of mud crab larval culture. In. Book of Abstracts. 2001 Workshop on Mud Crab rearing, Ecology and Fisheries. Can Tho University, Vietnam. 8-10 January 2001, pp. 18-21.

Mann, D., Asakawa, T., Blackshaw, A., 1999a. Performance of mud crab Scylla serrata broodstock held at Bribie Island Aquaculture Research Center. In: Keenan, C.P., Blackshaw, A. (Eds.). Mud crab Aquaculture and Biology. Proceedings of an International Scientific Forum held in Darwin, Australia, 21-24 April 1997. ACIAR Proceedings No. 78, 101-105.

Mann, D., Asakawa, T., Pizzuto, M., 1999b. Development of a hatchery system for larvae of the mud crab Scylla serrata at the Bribie Island Aquaculture Research Center. In: Keenan, C.P., Blackshaw, A. (Eds.). Mud crab Aquaculture and Biology. Proceedings of an International Scientific Forum. Darwin, Australia, 21-24 April 1997. ACIAR Proceedings No. 78, 153-158.

References

173

Mann, D., Parlato, D., 1995. Influence of starvation on larval growth and survival of the mud crab Scylla serrata (Forskål). In: Proceedings of Mud Crab Workshop. Aquatic Research Unit, Muresk Institute of Agriculture, Curtin University of Technology. Broome, Western Australia, 27 October 1995, pp. 41-48.

Mann, D.L., Asakawa, T., Pizzuto, M., Keenan, C.P., Brock, I.J., 2001. Investigation of an Artemia-based diet for larvae of the mud crab Scylla serrata. Asian Fisheries Science 14 (2), 175-184.

Marasigan, E.T., 1998. Mud crab (Scylla serrata) megalopae nursery in fertilized hapa nets. In: Extended Abstracts of International Forum on the Culture of Portunid Crabs. Boracay, Philippines, 1-4 December 1998, pp. 41-42.

Marichamy, R., Rajapackiam, S., 1992. Experiments on larval rearing and seed production of the mud crab Scylla serrata (Forskål). In: Angell, C.A. (Ed.). Report of the Seminar on the Mud Crab Culture and Trade. Surat Thani, Thailand, November 5-8, 1991. Bay of Bengal Programme, Madras, India. BOBP/REP/51, pp. 135-142.

Marichamy, R., Rajapackiam, S., 2001. The aquaculture of Scylla species in India. Asian Fisheries Science 14 (2), 231-238.

Mead, J.F, Slater, R.B.A., Howton, D.R., George, P., 1986. Lipids - Chemistry, Biochemistry and Nutrition. Prenum Press. New York and London. 486 pp.

Merchie, G., Lavens, P., Sorgeloos, P., 1997. Optimization of dietary vitamin C in fish and crustacean larvae: a review. Aquaculture 155 (1-4), 165-181.

Millamena, O.M., Bangcaya, J.P., 2001. Reproductive performance and larval quality of pond-raised Scylla serrata females fed various broodstock diets. Asian Fisheries Science 14 (2), 153-159.

Minagawa, M., 1992. Effects of salinity on survival, feeding and development of larvae of the red frog crab, Ranina ranina (Crustacea, Decapoda, Raninidae). Bulletin of the Japanese Society of Scientific Fisheries 58, 1855-1860.

Minagawa, M., Murano, M., 1993. Effects of prey density on the survival, feeding rate and development of the red frog crab Ranina ranina (Crustacea, Decapoda, Raninidae). Aquaculture 113, 91-100.

MOFI, 2003. Status of seafood export of Vietnam in 2002 (in Vietnamese). Ministry of Fisheries of Vietnam. See www.mofi.gov.vn

Millamena, O.M., Quinitio, E.T., 2000. The effects of diets on the reproductive performance of ablated and intact mud crab Scylla serrata. Aquaculture 181, 81-90.

, accessed in April 2003.

Mourente, G., Rodríguez, A., 1997. Effects of salinity and dietary DHA (22: 6(n-3)) content on lipid composition and performance of Penaeus kerathurus postlarvae. Marine Biology 128 (2), 289-298.

Naihong, X., Juju, S., Guizheng, Z., Sorgeloos, P.,1999. The effect of Artemia sources on growth survival and salinity stress of crab (Eriocheir sinensis) larvae. Asian Fisheries Science 12, 201-205.

Napolitano, G.E., 1990. Fatty acid composition of three cultured algal species (Isochrisis galbana, Chaetoceros gracilis and Chaetoceros calcitrans) used as food for bivalve larvae. Journal of the World Aquaculture Society 21, 122-130.

Natunewicz, C.C., Epifanio, C.E., Garvine, R.W., 2001. Transport of crab larval patches in the coastal ocean. Marine Ecology Progress Series 222, 143-154.

Navarro, J.C., Henderson, R.J., McEvoy, L.A., Bell, M.V., Amat, F., 1999. Lipid conversions during enrichment of Artemia. Aquaculture 174, 155-166.

Ng, P.K.L., 1998. Crabs. In: Carpenter, K.E., Niem, V.H. (Eds.).The living resources of the western central Pacific. FAO UN, Rome, volume 2, 1046-1134.

Nghia, T.T., Wille, M., Sorgeloos, P., 2001a. Effects of light, eyestalk ablation and seasonal cycle on the reproductive performance of captive mud crab (Scylla paramamosain) broodstock in the Mekong Delta, Vietnam. In: Book of Abstracts of 2001 Workshop on Mud Crab Rearing, Ecology and Fisheries. Institute for Marine Aquaculture, Can Tho University, Vietnam, 8-10 January 2001, p. 4.

Nghia, T.T., Wille, M., Sorgeloos, P., 2001b. Overview of larval rearing techniques for mud crab (Scylla paramamosain) with special attention to the nutritional aspects in the Mekong Delta, Vietnam. In: Books of Abstracts of 2001 Workshop on Mud Crab Rearing, Ecology and Fisheries. 8-10 January 2001. Institute for Marine Aquaculture, Can Tho University, Vietnam, p. 13.

References

174

Nogami, K., Maeda, M., 1992. Bacteria as control agents for rearing larvae of the crab Portunus triturbiculatus. Canadian Journal of Fisheries and Aquatic Sciences 49, 2373-2376.

Quinitio, E.T., Parado-Estepa, F., Alava, V., 1999. Development of hatchery techniques for the mud crab Scylla serrata (Forskål): Comparison of feeding schemes. In: Keenan, C.P., Blackshaw, A. (Eds.). Mud crab Aquaculture and Biology. Proceedings of an International Scientific Forum. Darwin, Australia, 21-24 April 1997. ACIAR Proceedings No. 78, 125-130.

Olsen, Y., 1989. Cultivated microalgae as a source of omega-3 fatty acids. In: Fish, fats and your health, proceedings of the international conference on fish lipids and their influence on human health, pp. 51-61

Olsen, Y., Reitan, K.I., Vadstein, O., 1993. Dependence of temperature on loss rates of rotifers, lipids, and w3 fatty acids in starved Brachionus plicatilis cultures. Hydrobiologia 255-256, 13-20.

Ong, K.S., 1964. Early development stages of Scylla serrata Forskål (Crustacea: Portunidae) reared in the laboratory. Proceedings of the Indo-Pacific Fisheries Council 11 (2), 135-146.

Ong, K.S., 1966. Observations of the postlarval life history of Scylla serrata Forskål reared in the laboratory. Malaysian Agriculture Journal 45 (4), 429-443.

Parado-Estepa, F.D., Quinitio, E., 1998. Use of different water treatment schemes during larval rearing of Scylla serrata: effect of chlorination, aging and antibiotics. In: Extended Abstracts of International Forum on the Culture of Portunid Crabs. Boracay, Philippines, 1-4 December 1998, p. 62.

Pastorok, R.A., 1981. Prey vulnerability and size selection by Chaoborus larvae. Ecology 62, 1311-1324.

Paterson, A.W., Whitfield, A.K., 1997. A stable carbon isotope study of the food web in a freshwater-deprived South African estuary, with particular emphasis on the ichthyofauna. Estuarine, Coastal and Shelf Science 45, 705-715.

Pianka, E.R., 1970. On r- and K-selection. American Naturalist, 104, 592-597, 952-956.

Poovichiranon, S., 1992. Biological studies of the mud crab Scylla serrata (Forskal) of the mangrove ecosystem in the Andaman Sea, In: Angell, C.A. (Ed.). Report of the Seminar on Mud Crab Culture and Trade. Surat Thani, Thailand, 5-8 November 1991. Bay of Bengal Programme, Madras. BOBP/REP/51, pp. 49-59.

Primavera, H.J., 1985. A review of maturation and reproduction in closed thelycum penaeids. In: Taki, Y., Primavera, J.H., Llobrera, J.A. (Eds.). Proceedings of the first International Conference on the Culture of Penaeid Prawns/Shrimp. Aquaculture Department, SEAFDEC, Iloilo, Philippines, pp. 47-64.

Quackenbush, L.S., 1986. Crustacean endocrinology: a review. Canadian Journal of Fisheries and Aquatic Sciences 43, 2271-2282.

Quillere, I., Marie, D., Roux, L., Gosse, F., Morot-Gaudry, J.F., 1993. An artificial productive ecosystem based on a fish:bacteria:plant association. 1. Design and management. Agriculture Ecosystems and Environment 47, 13-30.

Quinitio, E.T., Parado-Estepa, F.D., 2000. Transport of Scylla serrata megalopae at various densities and durations. Aquaculture 185, 63-71.

Quinitio, E.T., Parado-Estepa, F.D., 2001. Simulated transport of Scylla serrata zoeae at various loading densities. Asian Fisheries Science 14 (2), 225-230.

Quinitio, E.T., Parado-Estepa, F., Millamena, O.M., Rodriguez, E., Borlongan, E., 2001. Seed production of mud crab Scylla serrata juveniles. Asian Fisheries Science 14 (2), 161-174.

Quinitio, E.T., Parado-Estepa, F.D., 2003. Biology and seed production of mud crab. SEAFDEC/AQD extension manual. In press.

Quinn, N.J., Kojis, B.L., 1987. Reproductive biology of Scylla spp. (Crustacea: Portunidae) from the Labu estuary in Papua New Guinea. Bulletin of Marine Science 41 (2), 234-241.

Rattanachote, A., Dangwatanakul, R., 1992. Mud crab (Scylla serrata Forskal) fattening in Surat Thani province. In: Angell, C.A. (Ed.). Report of the Seminar on Mud Crab Culture and Trade. Bay of Bengal Programme, Madras, India. BOBP/REP/51, 171-177.

Ribeiro, F.A.L.T., Jones, D.A., 2000. Growth and ontogenetic change in activities of digestive enzymes in

References

175

Fenneropenaeus indicus postlarvae. Aquaculture Nutrition. 6 (1), 53-64.

Rees, J.F., Curé, K., Piyatiratitivorakul, S., Sorgeloos, P., Menasveta, P., 1994. Highly unsaturated fatty acid requirements of Penaeus monodon postlarvae: an experimental approach on Artemia enrichment. Aquaculture 122, 193-207.

Rickle, S.C., 2003. Perspectives on the use of organic acids and short chain fatty acids as antimicrobials.. Poultry Science 82 (4), 632-639.

Robertson, W.D., 1987. Biology of the mangrove crab, Scylla serrata. Poster presented at the sixth national oceanographic symposium. University of Stellenbosch, South Africa, 6-10 July, 1987.

Sastry, A.N. 1983. Pelagic larval development. In: Bliss, D.E. (Ed. in chief). The biology of crustacea volume 7: behaviour and ecology. Vernberg, F.J., Vernberg, W.B. (Eds.). Academic Press, New York, USA, pp. 215-282.

Robertson, W.D., Kruger, A., 1994. Size at maturity, mating and spawning in the portunid crab Scylla serrata (Forskål) in Natal, South Africa. Estuarine, Coastal and Shelf Science 39, 185-200.

Rodríguez, E.M., Quinitio, F.D., Parado-Estepa, F.D., Millamena, O.M., 1998. Culture of mud crab megalopae in brackish water ponds. In: Extended Abstracts of International Forum on the Culture of Portunid Crabs. Boracay, Philippines, 1-4 December 1998, pp. 39-40.

Rodríguez, E.M., Quinitio, F.D., Parado-Estepa, F.D., Millamena, O.M., 2001. Culture of Scylla serrata megalops in brackishwater ponds. Asian Fisheries Science 14 (2), 185-189.

Rodríguez, E.M., Triño, A., Minagawa, M., 2003. Diet and harvesting regimen for the production of mud crab Scylla olivacea in brackish water ponds. Fisheries Science 69, 37-42.

Rombaut, G., Suantika, G., Boon, N., Maertens, S., Dhert, P., Top, E., Sorgeloos, P., Verstraete, W., 2001. Monitoring of the evolving diversity of the microbial community present in rotifer culture. Aquaculture 198, 237-252.

Ruppert, E.E., Barnes, R.D. 1994. Invertebrate Zoology. Sixth Edition. Saunders College Publishing, UK, 1056 pp.

Ruscoe, I.M., Williams, G.R., Shelley, C.C., 2004. Limiting the use of rotifers to the first zoeal stage in mud crab (Scylla serrata Forskål) larval rearing. Aquaculture 231, 517-527.

Rusdi, I., Ahmad, T., Makatutu, D., 1994. Preliminary study of the spawning rate of mangrove crab, Scylla serrata on different substrates. Journal of Coastal Aquaculture Research 10, 2-4 (Translated from Indonesian).

Samocha, T.M., Uzeil, N., Browdy, C.L., 1989. The effect of feeding two prey organisms, nauplii of Artemia and rotifers, Brachionus plicatilis (Muller), upon survival and growth of larval marine shrimp, Penaeus semisulcatus (de Haan). Aquaculture 77, 11-19.

Sargent, J.R., Bell, J.C., Bell, M.V., Henderson, R.J., Tocher, D.R., 1995. Dietary origins and functions of long-chain (n-3) polyunsaturated fatty acids in marine fish. Journal of Marine Biotechnology 3, 26-28.

Sargent, J.R., Bell, J.G., Bell, M.V., Henderson, R.J., Tocher, D.R., 1991. The metabolism of phospholipids and polyunsaturated fatty acids in fish. In: Proceedings of European Society Comparative Physiology and Biochemistry. Experimental Aspects of Aquaculture, pp. 1-22.

Sargent, J.R., Henderson, R.J., Tocher, D.R., 1989. The lipids. In: Halver, J.E. (Ed.). Fish Nutrition, pp. 152-219.

Serene, R., 1952. Les espèces du genre Scylla à Nhatrang (Vietnam). Proceedings of the Indo-Pacific Fisheries Council. Academic Press, New York, pp. 113-137.

Shelbourne, J.E., 1964. The artificial propagation of marine fish. Advances in Marine Biology 2, 1-83.

Shelley, C., Field, D., 1999. The culture of Portunid crabs. Infofish International 4, 46-49.

Shen, Y., Lai, Q., 1994. Present status of mangrove crab (Scylla serrata (Forskål)) culture in China. Naga, the ICLARM Quarterly January 1994, 28-29.

Skjerkmo, J., Vadstein, O., 1999. Techniques for microbial control in the rearing of marine larvae. Aquaculture 177, 333-343.

References

176

Sokal, R.R., Rohlf, F.J., 1995. Biometry: the principles and practice of statistics in biological research. W.H. Freeman and Company, New York, USA, third edition, 887 pp.

Sorgeloos, P., 1995. Bioengineering of hatcheries for marine fish and shellfish. Journal of Marine Biotechnology 3, 42-45.

Sorgeloos, P., Dhert, P., Candreva, P., 2001. Use of the brine shrimp, Artemia spp., in marine fish larviculture. Aquaculture 200, 147-159.

Sorgeloos, P., Lavens, P., 1996. Manual on the production and use of live food for aquaculture. Fisheries Technical Paper, volume 361, Food and Agriculture Organization of the United Nation, Rome, pp. 9-100.

Sorgeloos, P., Lavens, P., Léger, P., Tackaert, W., Versichele, D., 1986. Manual for the culture and use of brine shrimp Artemia in aquaculture. Artemia Reference Center, State University of Gent, Belgium. 319 pp.

Sorgeloos, P., Léger, P., 1992. Improved larviculture outputs of marine fish, shrimp and prawn. Journal of the World Aquaculture Society 23 (4), 251-264.

Southgate, P.C., Partridge, G.J., 1998. Development of artificial diets for marine finfish larvae: problems and prospects. In: De Silva, S.S. (Ed.). Tropical Mariculture. Academic Press, London, pp. 151-170.

Srinivasagam, S., M. Kathirvel, Kulasekarapandian, S., 2000. Captive broodstock development, induced breeding and larval stages of mud crabs (Scylla spp.). Bulletin No. 12. March 2000. Central Institute of Brackishwater Aquaculture (Indian Council of Agricultural Research), Chennai, India.

Suprayudi, M.A., Takeuchi, T., Hamasaki, K., Hirokawa, J., 2002b. The effect of n-3HUFA content in rotifers on the development and survival of mud crab, Scylla serrata, larvae. Suisanzoshoku 50 (2), 205-212.

Swift, M.C., Fedorenco, A.Y., 1975. Some aspects of prey capture by Chadborus larvae. Limnology Oceanography 20, 418-425.

Takeuchi, T., Satoh, N., Sekiya, S., Shimizu, T., Watanabe, T., 1999. The effect of dietary EPA and DHA on the molting rate of larval swimming crab Portunus trituberculatus. Nippon Suisan Gakkaishi, 65 (6): 998-1004.

Støttrup, J.G., McEvoy, L.A., 2003. Live feeds in marine aquaculture. Blackwell Science Ltd. Oxford, UK, 318 pp.

Suantika, G., 2001. Development a recirculation system for the mass culturing of the rotifer (Brachionus plicatilis). Ph.D. thesis. Ghent University, Ghent, Belgium.

Sulkin, S.D., 1978. Nutritional requirements during larval development of the portunid crab Callinectes sapidus Rathbun. Journal of Experimental Marine Biology and Ecology 34, 29-41.

Suprayudi, M., Takeuchi, T., Hamasaki, K., Hirokawa, J., 2002a. Effect of Artemia feeding schedule and density on the survival and development of larval mud crab Scylla serrata. Fisheries Science 68 (6), 1295-1308.

Tacon, A.G.J., 1987. The nutrition and feeding of farmed fish and shrimp - A training manual. 1. The essential nutrients. FAO Field document, FAO, Brasilia, Brasil.

Takeuchi, T., Kobayashi, T., Shimizu, T., Sekiya, S., 2000. The necessity and suitable feeding schedule of Artemia nauplii for larval mud crab. Bulletin of the Japanese Society of Scientific Fisheries 66 (6), 984-992.

Takeuchi, T., (in press). A review of studies on the effect of dietary n-3 highly unsaturated fatty acids on larval swimming crab Portunus trituberculatus and mud crab Scylla tranquebarica.

Tan, E.S.P., 1999. Malaysian crab research. In: Keenan, C.P., Blackshaw, A. W. (Eds.). Mud Crab Aquaculture and Biology. Proceedings of an International Scientific Forum. Darwin, Australia, 21-24 April 1997. ACIAR Proceedings No. 78, 25-26.

Tendencia, E.A., dela Pena, M., 2003. Investigation of some components of the greenwater system which makes it effective in the initial control of luminous bacteria. Aquaculture 218 (1-4), 115-119.

Toledo, J. D., Quinitio, E. T., Pedajas, S.A., 1998. Feeding performance of early mud crab Scylla serrata larvae reared on mixed zooplankton. In: Extended Abstracts of International Forum on the Culture of

References

177

Portunid Crabs. Boracay, Philippines, 1-4 December 1998, p. 61.

Tongdee, N., 2001. Size distribution, sex ratio and size at maturity of mud cran (Scylla spp.) in Ranong province, Thailand. Asian Fisheries Science 14 (2), 113-120.

Triño, AT., Rodriguez, E.M., 2002. Pen culture of mud crab Scylla serrata in tidal flats reforested with mangrove trees. Aquaculture 211, 125-134.

Tseng, K.F., Huang, J.S., Liao, I.C., 1991. Species control of microalgae in an aquaculture pond. Water Research 25 (11), 1431-1437.

Tsukimura, B., Kamemoto, F.I., 1991. In vitro stimulation of oocytes by presumptive mandibular organ secretions in the shrimp, Penaeus vannamei. Aquaculture 92, 59–66.

Tuan, N.A., Hai, T.N., Hien, T.T.T., Ninh, L.Q., 1996. Culture of mud crabs in the Mekong Delta, Vietnam. Manuscript, College of Agriculture, Can Tho University, 8 pp.

Tung, H., 1995. Farmers turning back on shrimp and calling for mud crab culture. Saigon Liberation Newspaper, 7 October 1995, p. 4 (in Vietnamese).

Twarowska, J.G., Westerman, P.W., Losordo, T.M., 1997. Water treatment and waste characterization evaluation of intensive recirculating fish production system. Aquaculture Engineering 16, 133–147.

Ut, V.N., 2003. Assessment of the feasibility of stock enhancement of mud crabs Scylla paramamosain in the Mekong Delta. Ph.D. thesis. University of Wales, Bangor, UK, 188 pp.

Van Stappen, G., 1996. Artemia. In: Lavens, P.,Sorgeloos, P. (Eds.). Manual on the' production and use of live food for aquaculture. FAO technical paper No. 361. Food and Agricultural Organization, Rome, pp. 101-170.

Vandendriessche, S., 2003. Fatty acid requirements of the larvae of the mangrove crab Scylla paramamosain. M.Sc. thesis. Laboratory of Aquaculture and Artemia Reference Center, Ghent University, Belgium.

VASEP, 2003. Appearance of natural crablings in the Mekong Delta (in Vietnamese). Magazine of Vietnam Association of Seafood Exporters and Producers, August 2003.

Verschuere, L., Rombaut, G., Sorgeloos, P., Verstraete, W., 2000. Probiotic bacteria as biological control agents in aquaculture. Microbiology and Molecular Biology Reviews 64 (4), 655-671.

Volkman, J.K., Jeffrey, Nichols, P.D., Rogers, G.I., Garland, C.D., 1989. Fatty acid and lipid composition of 10 species of microalgae used in mariculture. Journal of Experimental Marine Biology and Ecology, 219-240.

Wang, J.K, 2003. Conceptual design of a microalgae-based recirculating oyster and shrimp system. Aquacultural Engineering 28, 37-46

Warner, G.F., 1977. Feed and feeding. In: The biology of crabs. Gresham Press, Old Working, Surrey, U.K., pp. 85-94.

Watanabe, T. 1982. Improvement of dietary value of brine shrimp Artemia salina for fish larvae by feeding them on ω3 highly unsaturated fatty acids. Bulletin of the Japanese Society of Scientific Fisheries 48 (12), 1775-1782

Watanabe, T., 1993. Importance of docosahexaenoic acid in marine larval fish. Journal of the World Aquacultural Society 24, 152–161.

Wickins, J.F., Lee, D.O’C., 2002. Crustacean farming ranching and culture. Second edition. Blackwell Science, Oxford, England, 446 pp.

Williams, G.R., Wood, J., Dalliston, B., 1998. A reliable small scale culture method for the production of mud crab (Scylla serrata Forskal) megalopae. In: Extended Abstracts of International Forum on the Culture of Portunid Crabs. Boracay, Philippines, 1-4 December 1998, pp. 31-32.

Williams, G.R., Wood, J., Shelley, C., 1999a. Survival rates of mud crab (Scylla serrata, Forskal) larvae from zoea 1 to megalopa. In: Book of Abstracts of World Aquaculture 1999, p. 689.

Williams, G.R., Wood, J., Dallison, B., Shelley, C.C., Kuo, C. M., 1999b. Mud crab (Scylla serrata) megalopae larvae exhibit high survival rates on Artemia based diets. In: Keenan, C.P., Blackshaw, A. (Eds.). Mud crab Aquaculture and Biology. Proceedings of an International Scientific forum. Darwin, Australia, 21-24 April 1997. ACIAR proceedings No. 78, 131-140.

References

178

Williams, M.J., Primavera, J.H., 2001. Choosing tropical portunid species for culture, domestication and stock enhancement in the Indo-Pacific. Asian Fisheries Science 14 (2), 121-142.

Williams, G.R., Ruscoe, I.M., Naylor, R., Moir, C., Shelley, C.C., 2002. The effects of pre-conditioning culture water on mud crab Scylla serrata larval survival. In: Book of abstracts of World Aquaculture 2002. International Aquaculture Conference and Exposition. Beijing, China, 23-27 April 2002, pp. 821

Wouters, R., Van Hauwaert, A., Naessens, E., Ramos, X., Pedrazzoli, A., Lavens P., 1997. The effect of dietary (n-3) HUFA and 22:6n-3/20:5n-3 ratio on white shrimp larvae and postlarvae. Aquaculture International 5 (2), 113-126.

Wyban, J.A., Swingle, J.S., Sweeney, J.N., Pruder, G.D., 1993. Specific-pathogen-free Penaeus vannamei. World Aquaculture 24, 39–45.

Xuan, T.T., 2001. The status and the prospect of mudcrab culture in Ca Mau province, Vietnam. In: Book of Abstracts of 2001 Workshop on Mud Crab Rearing, Ecology and Fisheries. Can Tho University, Vietnam, 8-10 January 2001, p. 28.

Yen, J., 1982. Sources of variability in attack rates of Euchaeta elongata (Esterly), a carnivorous marine copepod. Journal of Experimental Marine Biology and Ecology 63, 105-117.

Zainoddin, B.J., 1992. Preliminary studies on rearing the larvae of the mud crab (Scylla serrata) in Malaysia. In: Angell, C.A. (Ed.). Report of the Seminar on the Mud crab Culture and Trade. Surat Thani, Thailand, 5-8 November 1991. Bay of Bengal Programme, Madras, India. BOBP/REP/51, pp. 143-147.

Zeng, C., Li, S., 1992. Effects of temperature on survival and development of the larvae of Scylla serrata. Journal of Fisheries of China 16 (3), 214-221.

Zeng, C., 1987. Studies on inducing spawning, embryonic development and larval experimental ecology and mud crab Scylla serrata. M.Sc. thesis (in Chinese), Xiamen University, 189 pp.

Zeng, C., 1998. Effects of diet density on feeding rates of larvae of the mud crab Scylla sp. from hatching to metamorphosis. In: Extended Abstracts of the International Forum on the Culture of Portunid Crabs. Boracay, Philippines, 1-4 December 1998, pp. 27-28.

Zeng, C., Li, S., 1999. Effects of density and different combinations of diets on survival, development, dry weight and chemical composition of larvae of the mud crab Scylla paramamosain. In: Keenan, C.P., Blackshaw, A. (Eds.). Mud crab Aquaculture and Biology. Proceedings of an International Scientific Forum held in Darwin, Australia, 21-24 April 1997. ACIAR Proceedings No. 78, 159-166.

Zeng, C., Wang, G., Li, S., 1991. Observation on embryonic development and effects of temperature on developmental rate of embryonic stages in mud crab, Scylla serrata. Fujian Fisheries 1, 45-40 (in Chinese).

Summary

Mud crab (Scylla spp.) fishing and culture represent a valuable income component of

rural fishery communities in many countries in tropical and sub-tropical Asia. Since wild

stocks are reported to decrease in many countries due to increased exploitation in recent

years, mud crab farming is becoming increasingly popular. In Vietnam, the mud crab Scylla

paramamosain is the second most important cultured species (next to shrimp) in the coastal

zone. Mud crab culture however currently relies almost entirely on wild seed stock. The

main obstacle for the further development of mud crab farming is the establishment of

hatchery-techniques for controlled production of seed.

Overall, the need for development of technologies for mud crab larviculture in

Vietnam is justified for the following reasons: (i) mud crab is a high-value species with an

increasing demand on the global market, (ii) it is a robust species that forms an alternative

for shrimp culture which is facing serious disease problems, (iii) availability of wild mud

crab seed is declining due to overfishing and habitat destruction (iv) mud crab can be

cultured using relatively simple traditional techniques requiring low initial investment, but

generating considerable profit and (v) after the reclassification of the genus Scylla into four

species, S. paramamosain can be considered a “new species” on which very little

information exists (Chapter 1 - Introduction).

Mud crab aquaculture is still an emerging industry, particularly the hatchery phase of

production; therefore very little peer-reviewed literature exists. As a collaborative effort

between some of the main research groups working on larviculture, a paper that describes

the “state of the art” of mud crab larviculture technology has been prepared. This paper

(Chapter 2 - Current status of mud crab Scylla spp. hatchery technology) reviews the

various rearing techniques and conditions currently employed in (mainly experimental)

larviculture of the species to serve as a basis for further reseach on this topic.

Broodstock availability and management are the first concerns for those who wish to

develop a new species for aquaculture. The environmental conditions and broodstock

management techniques used in this study were recorded and evaluated for their effect on

Summary

180

reproductive performance as a basis for further improving culture techniques and ultimately

fully control domestication and seed production of this species (Chapter 3 - Reproductive

performance of captive mud crab Scylla paramamosain broodstock in Vietnam). From 1996

to 2002, reproductive performance of 786 wild S. paramamosain breeders was recorded.

The period from March to July was most efficient for broodstock rearing and might

correspond to the natural spawning season of S. paramamosain in South Vietnam.

September to February could be the period for gonad maturation in the wild. With eyestalk

ablation, the most favorable period for artificial reproduction could be extended from

February to August. Like marine shrimp, eyestalk ablation in mud crabs improved spawning

success, but did not alter the latency period between purchase and spawning. No negative

effects of ablation on broodstock survival or egg quality were found. Breeders collected

from an inshore region with higher and more stable average salinity levels tended to

perform slightly better than those collected from a region with lower and varying salinity.

Females in the range of 300 to 500 g had the best overall reproductive efficiency and are

preferred as breeders. Rearing broodstock in earthen ponds was more efficient; however,

management of a pond proved more complicated than tank systems. In terms of complete

domestication, broodstock rearing in tanks therefore provides a more practical alternative,

provided larger tanks and more suitable substrate are used. It was noticed that shading the

broodstock tanks was not necessary if shelters for hiding were available. Spawning activity

decreased with prolonged time in captivity. Egg quality criteria such as fertilization rate and

egg diameter did however not vary in function of time in captivity. Although detached eggs

could be incubated artificially, egg incubation by the females themselves was the best

practice. Overall, controlled reproduction of wild mature broodstock females of S.

paramamosain for research and pilot production is not problematic, especially with the

practice of eyestalk ablation. Although individual females had high fecundities and

fertilization rates; spawning and hatching success were however not very high. In this

respect, broodstock captured offshore might be better. Further research should be

undertaken to completely domesticate the species and further document maturation and

fertilization in captive conditions. Therefore dietary requirements and suitable rearing

conditions should be investigated. Life history studies in the wild could provide very useful

information in this respect.

Summary

181

Chapter 4 (Optimal feeding schedule for mud crab Scylla paramamosain larvae)

identifies the most suitable prey for the early stages of mud crabs. Of the 3 feeds (micro-

algae, rotifers and artificial diets) tested as first feed, rotifers gave the best survival and

growth. The earliest appropriate time to shift from rotifers to Artemia was investigated since

rotifer culture is very laborious and not well mastered by most of hatchery managers in

Vietnam. Crab larvae could start catching newly-hatched Artemia from the zoea 2 stage

onwards and the number of Artemia consumed increased with each larval stage. The ability

of zoea 2 to catch newly-hatched Artemia seemed however to depend on the quality of that

particular batch and varied even between individual larvae. Although micro-algae were not

a proper initial feed for early crab stages, they proved beneficial in improving the nutritional

quality of rotifers, resulting in higher survival in later zoeal stages and a more successful

metamorphosis to the megalopa stage. In an attempt to simplify the feeding schedule, a

series of experiments were carried out where rotifers were replaced by different forms of

Artemia (live and heat-killed umbrella-stage Artemia and frozen or heat-killed Artemia

nauplii). Live umbrella-stage Artemia were the best replacement for rotifers for feeding zoea

1 - zoea 2 larvae compared to other Artemia forms. The unselective feeding behaviour

seems promising to develop artificial diets in order to substitute live feed and to reduce the

dependency on rotifers at the early stages. In a last step, the optimal time to shift from

rotifers to Artemia was investigated. Results showed that rotifers should be replaced by

Artemia already in zoea 2 stage. A transition period to shift from one diet to another seemed

not necessary. Prolonged feeding of rotifers beyond the zoea 2 stage tended to reduce

survival and delay the larval development. Although Artemia are more difficult to capture

for zoea 2 larvae than rotifers, they probably enhance crab larval performance due to their

higher nutritional value. The nutritional value of rotifers and Artemia is however not

consistent and therefore optimal feeding schedules might also depend local conditions and

culture techniques. From the zoea 3 stage onwards, crab larvae can ingest enriched Artemia

meta-nauplii.

In captivity, rotifers and Artemia nauplii support growth and survival of mud crab

larvae, but this simplified diet is not ideal. Inferior nutrition may be a factor contributing to

highly variable survival and the high susceptibility to disease often recorded in mud crab

larviculture. Enriching live feed with highly unsaturated fatty acids (HUFA) has been a

common way to improve the quality of live prey for other species. Chapter 5 (Influence of

Summary

182

the content of highly unsaturated fatty acids in the live feed on larviculture success of mud

crab Scylla paramamosain) describes the effects of different standard live feed enrichment

emulsions on survival and growth of crab larvae. In most experiments, survival rate in the

zoeal stages was not statistically different among treatments. Larval development rate and

metamorphosis success were however more strongly affected by the dietary treatments. In

this respect, the DHA/EPA ratio in the live feed seems to be a key factor. Enrichment

emulsions with a high (50 %) total (n-3) HUFA content but low DHA/EPA ratio (0.6) or

zero total HUFA content caused growth retardation and/or metamorphosis failure. An

emulsion with moderate total HUFA content (30 %) and high DHA/EPA ratio (4) was the

best in terms of larval development rate during the zoeal stages and resulted in good

metamorphosis. The optimal DHA/EPA ratio of live feed enrichment emulsions for early

stages (Z1 - Z2) could however be lower than 4. Dietary arachidonic acid (ARA) seemed to

improve first metamorphosis, but its exact role needs further clarification. For the larval

rearing of Scylla paramamosain, it is recommended to use enrichment media with a total n-

3 HUFA content of approximately 30 %, with a DHA/EPA ratio of minimum 1. Further

research needs to be performed on the total (n-3) HUFA and DHA/EPA ratio requirements

for each larval crab stage. The role of ARA in metamorphosis also needs to be verified.

In addition to nutritional requirements, establishing proper zootechnics (Chapter 6 -

Improved larval rearing techniques for mud crab Scylla paramamosain) is another crucial

aspect of developing larval rearing technology. Based on the method of water exchange

(discontinuous partial water renewal or continuous treatment through biofiltration) and the

level of micro-algae (Chlorella or Chaetoceros) supplementation (daily supplementation

with low levels of 0.1 - 0.2 million cells ml-1 or maintenance at high levels of 1 - 2 millions

cells ml-1), six different types of rearing systems were tried. The combination of a green-

water batch system for early stages and a recirculating system with micro-algae

supplementation resulted in the best overall performance of the crab larvae. A stocking

density of 100 Z1 l-1 combined with a rotifer density of 45 ml-1 for early stages and Artemia

feeding density of 20 Artemia nauplii ml-1 appeared to produce the best performance of S.

paramamosain larvae. Optimal rations for crab larvae should however be adjusted

depending on various factors such as species, larval stage, larval status, prey size, rearing

system and zootechniques. A practical feeding ration could be 30 - 45 rotifers ml-1 for Z1 -

Z2 and 5 - 10 Artemia (meta) nauplii ml-1 from Z3 onwards. However, antibiotics are not

Summary

183

encouraged for mud crab larviculture as they are unsafe. Ozonation and probiotics as

alternatives for prophylactic chemicals are worthwhile to investigate. Cannibalism is also an

important cause of high larval mortality at later larval stages and could be overcome by

provision of suitable substrates/shelters and feeding larger Artemia meta-nauplii.

Although there is still ample room for further improvements, we may conclude that,

on an international level, much progress has been achieved in recent years. Knowledge on

larval rearing technology of mud crab seems to be of a similar level as for shrimp in the

early nineties. This makes us hope that commercial mud crab hatcheries could become

reality in the next few years (Chapter 7 - General discussion).

Summary

184

Samenvatting Mangrove krab (Scylla spp.) visserij en kweek is een waardevolle inkomenscomponent van

rurale vissersgemeenschappen in vele tropische en subtropische landen in Azië. Aangezien

gerapporteerd wordt dat de natuurlijke stocks in vele landen dalen door de recentelijk

toegenomen exploitatie, wordt kweek meer en meer populair. In Vietnam is de mangrove

krab Scylla paramamosain de tweede meest gekweekte soort (naast garnalen) in de

kustzone. De kweek van mangrove krab steunt echter bijna volledig op het gebruik van wild

zaad. Het belangrijkste probleem voor de verdere ontwikkeling van de kweek van mangrove

krab is de ontwikkeling van broedhuistechnieken voor de gecontroleerde productie van

zaad.

Samenvattend kan het belang van ontwikkeling van broedhuistechnieken voor mangrove

krab gemotiveerd worden door de volgende punten: (i) mangrove krab is een soort met een

hoge waarde en een groeiende vraag op de internationale markt, (ii) het is een sterke soort

die een alternatief vormt voor de kweek van garnalen die geplaagd wordt door

ziekteproblematiek, (iii) beschikbaarheid van wild zaad neemt af door overbevissing en

vernietigen van habitat, (iv) mangrove krab kan gekweekt worden gebruik makende van

relatief eenvoudige traditionale technieken die slechts lage investering vragen, maar toch

aanzienlijke winst opleveren en (v) na de reclassificatie van het genus Scylla in vier species,

kan Scylla paramamosain beschouwd worden als een nieuwe soort waarover erg weinig

informatie bestaat (Hoofdstuk 1 - Introduction).

Mangrove krab aquaculture is nog een industrie in ontwikkeling, zeker de broedhuis-fase

van de productie; daardoor bestaat er erg weinig wetenschappelijke literatuur. Als een

gezamelijk initiatief van enkele van de belangrijkste onderzoeksgroepen werkzaam op

broedhuistechnieken werd daarom een artiekel geschreven die de “state of the art” van

mangrovekrab larvicultuur beschrijft. Dit artiekel (Hoofdstuk 2 - Current stutus of mud

crab Scylla spp. hatchery technology) vat de verschillende kweektechnieken en condities

samen die tegenwoordig gebruikt worden voor de (voornamelijk experimentele) larvicultuur

van deze soort als basis voor verder onderzoek.

Summary

185

Beschikbaarheid en management van ouderdieren zijn de eerste bekommernissen voor wie

een nieuwe soort wil ontwikkelen voor de aquacultuur. De omgevingscondities en

kweektechnieken gebruikt voor de ouderdieren in deze studie werden beschreven en hun

effect op reproductie geevalueerd als basis voor de verdere verbetering van de

kweektechnieken en uiteindelijk domesticatie en zaadproductie van deze soort volledig te

controleren (Hoofdstuk 3 - Reproductive performance of captive mud crab Scylla

paramamosain broodstock in Vietnam). Van 1996 tot 2002, werden de reproductie

parameters van 786 wilde S. paramamosain ouderdieren gevolgd. De periode van maart tot

juli bleek het meest efficient voor het kweken van ouderdieren en komt waarschijnlijk

overeen met de natuurlijke voortplantingsperiode in Zuid-Vietnam. September tot februari is

mogelijks de periode voor de ontwikkeling van de gonaden in het wild. Door gebruik te

maken van de techniek van oogsteelverwijdering kan de periode voor kunstmatige

voortplanting verlengd worden van februari tot augustus. Zoals bij mariene garnalen,

verhoogt oogsteelverwijdering het aantal dieren dat eitjes aflegt, maar het had geen invloed

op de tijd nodig om eiafleg te bekomen. Er werd geen negatieve invloed vastgesteld van

oogsteelverwijdering op overleving van de ouderdieren of eikwaliteit. Ouderdieren

afkomstig van kustzones met een hogere en meer stabiele saliniteit bleken lichtjes beter dan

deze afkomstig van zones waar de saliniteit lager en meer variabel is. Vrouwtjes van 300 tot

500 gram hadden algemeen de beste voortplantingsefficientie en zijn daarom verkiesbaar als

ouderdieren. Het kweken van ouderdieren in vijvers bleek het meest efficient, maar het

management was moeilijker dan voor tanksystemen. Om complete domesticatie te bereiken

is het kweken van ouderdieren in tanks daarom aangewezen, op voorwaarde dat grotere

tanks met meer substraat gebruikt worden. Er werd ook vastgesteld dat het niet nodig was

de tanks te verduisteren indien schuilplaatsen voorzien werden. Eiafleg nam af naarmate de

tijd in gevangenschap toenam. Criteria voor eikwaliteit zoals bevruchtingsgraad en

eidiameter varieerden echter niet in functie van de tijd in gevangenschap. Hoewel

losgekomen eitjes kunstmatig konden geincubeerd worden, bleek incubatie door de

vrouwtjes zelf de beste techniek. Algemeen kan gesteld worden dat gecontroleerde

reproductie van Scylla paramamosain voor onderzoek en semi-commerciele schaal geen

bijzondere problemen stelt, zeker wanneer gebrijk gemaakt wordt van oogsteelverwijdering.

Hoewel individuele vrouwtjes hoge fecunditeiten en bevruchtingspercentages haddden, was

de algemene eiafleg- en ontluikigsefficientie echter niet hoog. Het gebruik van dieren

gevangen in open zee kan hier misschien een oplossing bieden. Verder onderzoek is nodig

Summary

186

om deze soort volledig te domesticeren en maturatie en bevruchting in gevangenschap te

documenteren. Hiervoor moeten de nutitionele vereisten en optimale kweekomstandigheden

verder onderzocht worden. De studie van de levenscyclus in het wild zou hiervoor nuttige

informatie kunnen aanbrengen.

Hoofdstuk 4 (Optimal feeding schedule for mud crab Scylla paramamosain larvae)

identificeerd de meest geschikte prooi voor de vroege larvale stadia van mangrove krab.

Van de drie voeders die als eerste voedsel getest werden (micro-algen, rotiferen en

artificiele voeders), resulteerde rotiferen in de beste groei en overleving. Het vroegste

tijdsstip om van rotiferen naar Artemia om te schakelen werd onderzocht omdat de kweek

van rotiferen erg arbeidsintensief is en voor de meeste broedhuizen erg moeilijk is. Krab

larven konden vanaf het tweede zoea stadium pas-ontloken Artemia vangen en het totaal

aantal dat geconsumeerd werd nam elk stadium toe. De mogelijkheid van zoea 2 larven om

Artemia te vangen bleek echter variabel tussen verschillende groepen en zelfs individuele

larven. Hoewel micro-algen niet geschikt waren als eerste voedsel, bleken ze gunstig voor

het verbeteren van de nutritionele kwaliteit van de rotiferen en resulteerden ze zo in een

hogere overleving in latere larvale stadia en een meer succesvolle metamorphose tot

megalopa. In een poging om het voederschema te vereenvoudigen werd een serie

experimenten uitgevoerd waar rotiferen vervangen werden door verschillende vormen van

Artemia (levende en door middel van warmte afgedode umbrella-Artemia en ingevroren en

door middel van warmte afgedode Artemia nauplii). Levende umbrella-Artemia bleken het

beste vervangvoeder voor rotiferen voor zoea 1 en 2 stadia in vergelijking met de andere

Artemia vormen. Het niet-selectieve voedingsgedrag van de larven toont aan dat mogelijks

artificiele voeders kunnen ontwikkeld worden om levend voedsel te vervangen en de

afhankelijkheid van rotiferen in de eerste stadia te verminderen. In een laatste stap, werd de

optimale tijd om over te schakelen van rotiferen op Artemia nader onderzocht. De resultaten

toonden aan dat rotiferen al moeten vervangen worden door Artemia in het zoea 2 stadium.

Een overgangsperiode om van het ene voeder op het andere over te gaan bleek niet nodig.

Het voeder van rotiferen voorbij het zoea 2 stadium leek de overleving te verminderen en de

larvale ontwikkeling te vertragen. Hoewel Artemia moeilijker te vangen zijn voor zoea 2

dan rotiferen, verbeteren ze waarschijnlijk de productie door hun hogere nutritionele

kwaliteit. De nutritionele kwaliteit van rotiferen en Artemia is echter niet constant en

optimale voederschemas kunnen daardoor afhangen van de lokale condities en de gebruikte

Summary

187

kweektechnieken. Vanaf het zoea 2 stadium konden krab larven ook aangerijkte Artemia

meta-nauplii opnemen.

In gevangenschap zijn rotiferen voldoende voor de groei en overleving van mangrove krab

larven. Dit gesimplifieerd dieet is echter niet ideaal. Sub-optimale nutritie kan een faktor

zijn die bijdraagt tot de zeer variabele overleving en hoge vatbaarheid voor ziektes die vaak

vastgesteld wordt in de larvicultuur van mangrove krab. Aanrijking van het levend voedsel

met hoog onverzadigde vetzuren (HUFA) is een algemeen gebruikte techniek om de

kwaliteit van het levend voedsel te verbeteren voor andere soorten. Hoofdstuk 5 (Influence

of the content of highly unsaturated fatty acids in the live feed on larviculture success of

mud crab Scylla paramamosain) beschrijft het effect van verschillende standaard

aanrijkingsemulsies voor levend voedsel op groei en overleving van de krab larven. In de

meeste experimenten werd de overleving in de zoea stadia niet statistisch significant

beinvloed. Larvale ontwikkeling en metamorphose succes werden echter sterker beinvloed

door de behandelingen. De DHA/EPA (docosahexaenoic acid/eicosapentaenoic acid)

verhouding in het levend voedsel bleek hierbij doorslaggevend. Aanrijkingsemulsies met

een hoog (50%) total n-3 HUFA gehalte maar een lage DHA/EPAverhouding (0.6) of totaal

deficient in n-3 HUFA veroorzaakten groei achterstand en/of slechte metamorphose. Een

emulsie met een middelmatig totaal HUFA gehalte (30%) en hoge DHA/EPA verhouding

(4) resulteerde in de beste groei in de zoea stadia en gaf een goede metamorphose. De

optimale DHA/EPA verhouding voor aanrijking van levend voedsel voor de vroege larvale

stadia kan echter lager zijn. Arachidon zuur (ARA) in het voer leek de eerste metamorphose

te begunstigen, maar de exacte rol moet verder uitgeklaard worden. Voor de larvicultuur van

Scylla paramamosain wordt aangeraden aanrijkingsmedia te gebruiken met een total n-3

HUFA gehalte van 30 %, met een minimale DHA/EPA verhouding van 1. Verder onderzoek

is echter nodig om de vereisten voor totaal n-3 HUFA gehalte en DHA/EPA verhouding van

elk larvaal stadium te bepalen. De rol van ARA in de metamorphose moet ook verder

bestudeerd worden.

Naast het bepalen van nutritionele vereisten, is het onwikkelen van goede kweeksystemen

(Hoofdstuk 6 - Improved larval rearing techniques for mud crab Scylla paramamosain) een

ander cruciaal aspect in het ontwikkelen van larvale kweektechnieken. Op basis van de

methode van wateruitwisseling (discontinue gedeeltelijke waterverversing of continue

Summary

188

behandeling door biofiltratie) en de hoeveelheid gesupplementeerde micro-algen (Chlorella

of Chaetoceros; dagelijks toegevoegd aan lage concentraties van 0.1 - 0.2 miljoen cellen per

ml of op peil gehouden aan hoge concentraties van 1 - 2 miljoen cellen per ml), werden 6

verschillende kweeksystemen getest. De combinatie van een “groen-water batch” syteem

voor de vroege larvale stadia en een recirculatie-systeem met supplementatie van micr-algen

voor de latere stadia gaf algemeen het beste resultaat. Een stockeringsdensiteit van 100 Z1

per liter gecombineerd met het voederen van 45 rotiferen per ml voor de vroege stadia en 20

Artemia nauplii per ml gaf het beste productieresultaat. Optimale voederniveaus dienen

echter aangepast te worden aan verschillende factoren zoals soort, larvaal stadium,

afmetingen van de prooi, kweeksystem en kweekcondities. Een practisch voederniveau kan

zijn 30 - 45 rotiferen per ml voor Z1 - 2 en 5 - 10 Artemia (meta-) nauplii per ml vanaf Z3.

Het prophylactisch gebruik van antibiotica verbeterde de overleving. Het gebruik van

antibiotica wordt echter ontraden voor mangrove krab larvicultuur angezien dit niet veilig is.

Het gebruik van ozon en probiotica zijn alternatieven die onderzocht kunnen worden.

Kannibalisme is een andere belangrijke oorzaak van mortaliteit bij latere larvale stadia en

kan mogelijks verholpen worden door het voorzien van substraat/schuilpaatsen en het

voederen van grotere Artemia meta-nauplii.

Alhoewel er nog veel ruimte is voor verbeteringen, kunnen we concluderen dat, op

internationaal vlak, er heel veel vooruitgang geboekt is. De kennis van de technologie voor

de kweek van mangrove krab larven lijkt op een zelfde niveau als voor garnalen in de

vroege jaren negentig. Dit laat ons hopen dat commerciele mangrove krab broedhuizen

realiteit worden in de komende jaren (Hoofdstuk 7- General discussion).

Curriculum vitae

Personal data First name NGHIA Family name TRUONG TRONG Date of birth 10 September 1956 Place of birth Can Tho, Vietnam Nationality Vietnamese Married status married, 2 children Address 192, street 30 April, Can Tho City, Vietnam Email [email protected] Education Since 2000 Registered for Ph.D., Ghent University, Belgium. Major subject:

Aquaculture. 1993-1995 M.Sc. in Aquaculture, Laboratory of Aquaculture & Artemia Reference

Center, Ghent University, Belgium. 1976-1980 Aquaculture Engineer studies at College of Aquaculture and Fisheries (CAF),

Can Tho University, Vietnam 1974-1976 B.Sc. studies on Physics, Chemistry and Natural Science at University of

Saigon, Vietnam Professional record Since 2002 Vice Dean of College of Aquaculture and Fisheries (CAF), Can Tho

University, Vietnam 2001-2002 Director of Aquaculture and Fisheries Sciences Institute (AFSI), College of

Agriculture, Can Tho University, Vietnam 1999-2001 Director of Institute for Marine Aquaculture (IMA), College of Agriculture,

Can Tho University, Vietnam 1996-1999 Vice Director of Shrimp Artemia R&D Institute (SARDI), Can Tho

University, Vietnam 1989-1993 Researcher of Artemia-Shrimp R&D Center (ASRDC), Can Tho University,

Vietnam Since 1981 Lecturer of College of Aquaculture and Fisheries, Can Tho University,

Vietnam Scientific activities International cooperation 2001-2007 Coordinator of VL.I.R. (Vlaase Interuniversitaire Raad) IUC: subcomponent

“Distant education: aquaculture as a test case” 2000-2004 Coordinator of EU INCO-DC (International Cooperation with Developing

Countries) program “Culture and Management of Scylla species” 1998-2000 Coordinator of BRITISH COUNCIL LINK “Tropical coastal ecosystems and

sustainable aquatic resource management”

CV

190

1997-2000 Coordinator of EU INCO-DC (International Cooperation with Developing Countries) program “Sustainable production of mud crab Scylla species through stock enhancement in mangroves”

1997-2000 Coordinator of VL.I.R. OWN INITIATIVE “R&D on Diversification of Marine Aquaculture in the Mekong Delta”

1997-1998 Aquaculture project specialist of GEC Global Environmental Consultants Ltd. "Vietnam coastal wetlands protection and development", World Bank project Technical assistant of Euro consult in the framework of the project “Rehabilitation of Mangrove Forests-Mekong Delta, FIPI No 2”

1996-2000 Coordinator of VVOB (Belgium) “Extension and Building Capacity on Aquaculture for Local Staff”

1990-2001 Coordinator of NOVIB (Netherlands Organization for International Development Cooperation) “Support for socio-economical development of the poor ethnic community in the coastal area of Vinh Chau district, Soc Trang province, Vietnam”

1989-1993 Coordinator of CANADA FUND “Initiatives for improving the poor community in Soc Trang and Bac Lieu provinces in the Mekong Delta, Viet Nam “Introduction of shrimp Penaeus monodon and P. merguiensis culture in the coastal area with support of technologies and credits”

Local involvement Since 1987 Planning projects and pilot models of aquaculture farming and hatchery of

freshwater fish, prawn, shrimp and mud crab for districts and provinces in the Mekong Delta, Vietnam

1980-1986 Artificial preproduction of freshwater catfish (Pangasius sp.) and sand goby (Oxyeleotris marmoratus)

1980-1983 Master planning for aquaculture in the Mekong Delta, Vietnam Conference/workshop/training participation Sept. 2002 - Jan. 2003 Training course in aquaculture, Laboratory of Aquaculture &

Artemia Reference Center, Ghent University, Ghent, Belgium 6-9 May 2002 Workshop on Integrated management of coastal area, Malaysia 27-30 Apr. 2002 Challenges in aquaculture, Beijing, China 23-27 Apr. 2002 International conference on fisheries and aquaculture in Beijing,

China 1 Sept. - 30 Nov. 2001 Training course in aquaculture, Laboratory of Aquaculture &

Artemia Reference Center, Ghent University, Ghent, Belgium 3-6 Sept. 2001 Larvi’01, held by European Aquaculture Society, Ghent, Belgium 23-26 Aug. 2001 Second workshop Vietnam-Hungary on Small animal production,

Godollo-Szarvas, Hungary 8-10 Jan. 2001 Workshop on mud crab rearing, ecology and fisheries, Can Tho

University, Vietnam 31 Oct. - 3 Nov. 2000 Third World Fisheries Congress, China Society of Fisheries,

Beijing, China 1-4 Dec. 1998 International Forum on the Culture of Portunid Crabs, Boracay, the

Philippines

CV

191

11-14 Nov. 1998 The Fifth Asian Fisheries Forum, held by Asian Fisheries Society, Chiangmai, Thailand.

3-7 Sept. 1995 Larvi’95, Gent University, Gent, Belgium Aug. - Oct. 1992 Training course in aquaculture at Lab of Aquaculture and Artemia

Reference Center, Gent University, Gent, Belgium 26-30 Oct. 1992 The third Asian Fisheries Forum, held by Asian Fisheries Society,

Singapore Publications/presentations/reports 1. Nghia, T.T., 1980. Culture for maturation and artificially spawning inducement of the fresh-water catfish

Pagasius micronemus (in Vietnamese). Aquaculture engineer thesis, Can Tho University, 55 pp.

2. Nghia, T.T., Kiem, N., Lai, B., 1985. Artificial reproduction of freshwater catfish Pangasius micronemus. Final report of research period 1980-1984 (in Vietnamese). Faculty of Fisheries, Can Tho University, 33 pp.

3. Nghia, T.T., 1990. Larviculture techniques of the marble sand goby Oxyeleotris marmoratus, Bleeker. Final report of research period 1985-1989 (in Vietnamese). Faculty of Fisheries, Cantho University, 125 pp.

4. Nghia, T.T., 1991. Larviculture techniques and economics of small-scaled Macrobrachium rosenbergii hatcheries in the Mekong Delta, Vietnam. In: Larvi’91. Symposium on Fish and Crustacean Larviculture, 27-30 August 1991. Lavens, P., Sorgeloos, P., Jaspers, E., Ollevier, F. (Eds.). European Aquaculture Society, Special Publication No 15.

5. Nghia, T.T., Quynh, V.D., Quang, N.K., 1992. Introduction of Penaeus merguiensis and Penaeus monodon culture in evaporation ponds of coastal salterns in southern Vietnam. The Third Asian Fisheries Forum, Asian Fisheries Society.

6. Nghia, T.T., Quang, N.K., Thanh, T., Danh, T.C., 1993. Evaluation of using Artemia biomass in the larviculture of Macrobrachium rosenbergii (in Vietnamese). The First Symposium on Artemia culture in Vietnam, Cantho, 16-18 April 1993.

7. Nghia, T.T., Ut, V.N., Quang, N.K., Rothuis, A.J., 1994. Improvement of traditional shrimp culture in the Mekong Delta. NAGA, The Iclarm Quartly, April 1994, pp. 20-22.

8. Ut, V.N., Quang, N.K., Nghia, T.T., Bosteels, T., Rothuis, A.J., 1995. Expansion of improved-extensive shrimp culture in the Mekong Delta. NAGA, The Iclarm Quartly, April 1995, pp. 22-23.

9. Nghia, T.T., 1995. Contribution to the speciation of genus Artemia with special emphasis to Asian populations (M.Sc. Thesis, Gent University, Belgium), 92 pp.

10. Nghia, T.T., Binh, T.V., 1996. Co-operation between research and industry in aquaculture plans. In: IFS (1998) Aquaculture research and sustainable development in inland and coastal regions in South-East Asia. Proceedings of an IFS/EU Workshop. Can Tho, Vietnam 18-22 March 1996, pp. 285-288.

11. Nghia, T.T., Quynh, V.D., Quang, N.K., 1997. Trials of nursing and culturing white shrimp (Penaeus merguiensis) and tiger shrimp (Penaeus monodon) in evaporation ponds of salt fields in the South Vietnam coast (in Vietnamese). In: Book of Technology Research 1993-1997, Can Tho University, pp. 44-50.

12. Nghia, T.T., Ut, V.N., Quang, N.K., 1997. Inprovement of shrimp traditional culture and expansion of improved-extensive shrimp culture model in the Mekong Delta. In: Book of Technology Research 1993-1997, Can Tho University, pp. 51-55.

13. Nghia, T.T., Binh, T.C., Phong, L.N., 1997. Some techniques to improve the seed quality of shrimp Penaeus monodon (in Vietnamese). In: Book of Technology Research 1993-1997, Can Tho University, pp. 57-64.

14. Nghia, T.T., Dat, N.M., 1997. Preliminary results of mud crab (Scylla paramamosain) seed production in the Mekong Delta (in Vietnamese). In: Book of Technology Research 1993-1997, Can Tho University, pp. 65-70.

15. Nghia, T.T., 1997. Improvement on larviculture of the mud crab (Scylla paramamosain) in the Mekong

CV

192

Delta. Final report of IFS research grant No. A/2505-1, period September 1996 - September 1997.

16. Nghia, T.T., Dat, N.M., Quynh, V.D., Lavens, P., 1998. Larviculture of the mud crab (Scylla serrata) in the Mekong Delta, Vietnam under controlled conditions. In Books of Abstract of International Conference on Fisheries and food security beyond the year 2000. The Fifth Asian Fisheries Forum, 11-14 November 1998 in Chiangmai, Thailand, p.155.

17. Nghia, T.T., Lavens, P., Wille, M., 1998. Zootechnical and nutritional aspects of mud crab Scylla spp. Larviculture. In: Books of Abstract of International Conference on Fisheries and Food Security beyond the year 2000. The Fifth Asian Fisheries Forum, 11-14 November 1998 in Chiangmai, Thailand, p. 425.

18. Nghia, T.T., Loc, N.H., Quynh. V.D., Lavens, P., Wille, M., 1998. Comparison of a batch and recirculation system for the larviculture of mud crab (Scylla paramamosain) in the Mekong Delta, Vietnam. In: Programm and extended abstracts of International Forum on the Culture of Portunid Crabs, 1-4 December 1998, Boracay, Philippines.

19. Le Vay. L., Ut, V.N., Nghia, T.T., Jones, D.A., 1999. Ecology, fisheries and culture of mud crabs in the Mekong Delta. 7th Coll. Crust. Dec. Med. Lisbon.

27. Hai, T.N., Nghia, T.T., 2004. Effects of rearing densities on development and survival of mud crab (Scylla paramamosain) larvae in green-water system (in Vietnamese). In: Scientific magazine of Can Tho University - Aquaculture section, pp. 187-192.

20. Le Vay. L., Ut, V.N., Nghia, T.T., Jones, D.A., 2000. Sustainable aquaculture and fisheries production of mud crabs (Scylla spp.). In: Responsible aquaculture in the new millennium. Abstracts of contributions presented at the International Conference AQUA 2000. Nice, France, 2-6 May 2000. European Aquaculture Society. Special Publication No 28, Oostende, Belgium, March 2000, p. 393.

21. Nghia, T.T., Wille, M., Sorgeloos, P., 2001. Effects of light, eyestalk ablation and seasonal cycle on the reproductive performance of captive mud crab (Scylla paramamosain) broodstock in the Mekong Delta, Vietnam. In: Book of Abstracts of 2001 Workshop on Mud Crab Rearing, Ecology and Fisheries. Institute for Marine Aquaculture, Can Tho University, Vietnam, 8-10 January 2001, p. 4.

22. Nghia, T.T., Wille, M., Sorgeloos, P., 2001. Overview of larval rearing techniques for mud crab (Scylla paramamosain) with special attention to the nutritional aspects in the Mekong Delta, Vietnam. In: Books of Abstracts of 2001 Workshop on Mud Crab Rearing, Ecology and Fisheries. 8-10 January 2001. Institute for Marine Aquaculture, Can Tho University, Vietnam, p. 13.

23. Nghia, T.T., Wille, M., Sorgeloos, P., 2001. Influence of the content and ratio of essential HUFA’s in the live food on larviculture success of the mud crab (Scylla paramamosain) in the Mekong Delta (Vietnam). In: Hendry, C.I., Van Stappen, G., Wille, M., Sorgeloos, P. (Eds), Larvi’01 - Fish and Shellfish Larviculture Symposium, European Aquaculture Society. Special Publication No 30, Oostende, Belgium, pp. 430-433.

24. Ut, V.N., Le Vay, L., Nghia, T.T., Hanh, T.T.H., Caldwell, B.S., 2001. Effect of substrate and diet in the nursery phase of mud crab (Sctlla paramamosain) production. In: Hendry, C.I., Van Stappen G., Wille, M., Sorgeloos, P. (Eds), Larvi’01 - Fish and Shellfish Larviculture Symposium, European Aquaculture Society. Special Publication, vol. 30, Oostende, Belgium, pp. 610-613.

25. O'Kelly, Le Vay, L., Mardjon M., Nghia, T.T., Jones, D.A., 2001. Larval development of Scylla paramamosain cultured in the laboratory. In: Hendry, C.I., Van Stappen G., Wille, M., Sorgeloos, P., (Eds.). Larvi’01 - Fish and Shellfish Larviculture Symposium European Aquaculture Society. Special Publication 30, Oostende, Belgium.

26. Thao, N.T.T., Nghia, T.T., 2003. Effects of different salinity levels on filtering rate of feed, growth, survival and stress resistance of blood cockle (Anadara granosa Linaeus, 1758) (in Vietnamese). In: Dien, N.H., Chinh, N., Thu, N.T.X., Phung, N.H., Nho, N.T. (Eds.). Books of selected scientific reports - Proceedings of the second national workshop on marine molluscs. Nha Trang, 3-4 August 2001, pp. 137-142.

ISBN 90-5989-036-1

Documents

FACULTEIT LANDBOUWKUNDIGE WETENSCHAPPEN Academiejaar 2004- · PDF fileWETENSCHAPPEN . Academiejaar 2004-2005 . ... Prof. Koen Dewettinck and ... and the Belgian MSc student Stijn Vandendriessche