quaculture research, tập 42, số 11, 2011

Ingestion of higher level of wax esters 50% of the lipid cause, however, poorer lipid digestibility and growth, so that optimal utilization of wax esters in Atlantic salmon is closer to

REVIEW ARTICLE

Marine wax ester digestion in salmonid fish: a review

Andre¤ Sture Bogevik

Institute of Marine Research, Matre Aquaculture Research Station, Bergen, Norway

Correspondence: Present address: A S Bogevik, Institute of Marine Research, Austevoll Aquaculture Research Station, Bergen 5817, Norway Email: andreb@imr.no

Abstract

Alternative marine resources from lower trophic

le-vels could partly cover the rapidly increasing needs

for marine proteins and oils in the future The North

Atlantic calanoid copepod, Calanus ¢nmarchicus, has

a high level of lipids rich in n-3 fatty acids However,

these animals have wax esters as the main lipid

sto-rage component rather than triacylglycerol (TAG)

Although these esters are considered di⁄cult to

di-gest by many ¢sh, is it well known that juvenile

Atlantic salmon (Salmo salar) feed on zooplankton

species It is therefore possible that the capacity to

utilize these lipids should be well developed in

salmo-nids Nonetheless, salmon hydrolyse wax esters

slower than TAG and absorb fatty alcohols slower

than fatty acids However, salmon have several

adap-tations to digest diets rich in wax esters These

in-cludes increased feed conversion, higher production

of bile and higher activity of lipolytic enzymes in the

midgut Atlantic salmon has been shown to feed and

grow on diets with a medium amount of wax esters

(30% of the lipid) with results comparable to ¢sh

maintained on ¢sh oil diets Ingestion of higher level

of wax esters (50% of the lipid) cause, however,

poorer lipid digestibility and growth, so that optimal

utilization of wax esters in Atlantic salmon is closer

to 30% than 50% of the dietary lipid

Keywords: Digestion, fatty acids, fatty alcohols,

intestine, wax esters

Introduction

Aquaculture is the fastest growing food-producing

sector accounting for almost 50% of global human

¢sh consumption and is perceived as having the

great-est potential to meet the growing demand for aquaticfood (FAO 2006a) World aquaculture has grown ra-pidly over the past 50 years from a production ofo1million tonnes in the early 1950s to more than 59 mil-lion tonnes in 2004 (FAO 2006a) The global produc-tion of salmonids (2004) is close to 2 million tonnes(FAO 2006b) Forty per cent of the total global aqua-culture production is dependent upon the use of feedeither in the form of single dietary ingredients, home-made or industrially manufactured aquafeeds (FAO2006a) Low-cost ¢sh species such as sardines, her-rings or anchovies are commonly used as feed to pro-duce higher valued species such as salmon, cod, tunaand grouper Many of these feed ¢sh are either highlyexploited or over exploited (Naylor, Goldburg, Prima-vera, Kautsky, Beveridge, Clay, Folke, Lubchenco,Mooney & Troell 2000) Further growth in the indus-try is therefore dependent on the use of alternative andsustainable feed sources (Naylor et al 2000)

Although terrestrial plant oils can be used as stitutes for ¢sh oil in salmonid diets, they do not sup-ply the n-3 highly unsaturated fatty acids (HUFA)normally found in ¢sh oils Plant oil-derived diets willtherefore not increase the level of n-3 HUFA in salmo-nid £esh (Bell, McEvoy, Tocher, McGhee, Campbell &Sargent 2001; Bell, Henderson,Tocher, McGhee, Dick,Porter Smullen & Sargent 2002;Torstensen, Fryland

sub-& Lie 2004; Torstensen, Fryland, rnsrud sub-& Lie2004; Fonseca-madrigal, Karalazos, Campbell, Bell &Tocher 2005) A dietary supply of these fatty acids isregarded as important for animals and humans toprevent fatty acid de¢ciency and development of sev-eral diseases including coronary heart diseases (Ima-zio, Forno, Quaglia & Trinchero 2003; Ruxton, Reed,Simpson & Millington 2004) In order to maintain a

‘healthy’ angle of marketing salmonid products, it is

important to ¢nd alternative feed resources that

would maintain a high level of marine-type HUFA to

be channelled into human nutrition (Mo¡at & McGill

1993; Sargent & Tacon 1999; Pickova & Mrkre

2007) Thus, other resources than terrestrial plants

also have to be considered, including marine and

an-imal byproducts

At present, one alternative is to search for

alterna-tive and unexploited marine oil resources, such as

harvest from lower trophic levels Krill and copepods

constitute a huge biomass in the waters o¡ Norway,

with an annual production in the magnitude of

sev-eral hundred million tonnes (Dalpadado, Ellertsen,

Melle & Skjoldal 1998; Madden, Beare, Heath, Fraser

& Gallego 1999; Table 1) A harvest of only a small

fraction of this, a few million tonnes, would cover

the need for marine raw materials for the Norwegian

¢sh farming industry far into the future

Several studies have showed that many krill

spe-cies are suited as a replacement for ¢sh meal in

for-mulated diets to salmonids (Storebakken 1988;

Olsen, Suontama, Langmyhr, Mundheim, Ring,

Melle, Malde & Hemre 2006; Suontama, Karlsen,

Mo-ren, Hemre, Melle, Langmyhr, Mundheim, Ring &

Olsen 2007) Data on the use of zooplankton ¢sh-oil

replacements in diets are, however, lacking Many

species of krill (Meganyctiphanes norvegica, Euphausia

superba) have lower lipid content (o40%) and are

more suited as protein source, while small-sized

spe-cies of krill (e.g Thysanoessa inermis) and calanoid

copepods (e.g Calanus ¢nmarchicus) seem more

desirable as lipid sources because they have higher

levels of lipid (450%) during part of the season

(Kattner & Krause 1987; Sther & Mohr 1987;

Falk-petersen, Hagen, Kattner, Clarke & Sargent 2000;

Lee, Hagen & Kattner 2006) However, many of theseanimals, and a few other marine invertebrates andmesopelagic ¢sh species have wax esters as theirmain lipid storage component rather than triacylgly-cerols (TAG), the main storage lipid in most ¢sh spe-cies (Sargent, Lee & Nevenzel 1976; Falk-petersen,Sargent, Hopkins & Vaja 1982; Phleger, Nichols &Virtue 1997) Wax esters are more hydrophobic thanTAG and poorly digested in most mammals, includ-ing humans, where the fatty alcohols are accumu-lated in the intestine This causes discomfortranging from stomach cramps to rapid loose bowelmovements in the form of oily diarrhoea (kerior-rhoea) and absorption problems (steatorrhea) Otherproblems include loss of hair and skin damage (sebor-rhoea) (Hansen & Mead 1965; Berman, Harley &Spark 1981; Place 1992) However, in the marine en-vironment, wax ester-rich krill and copepods are theprincipal food for many ¢sh species including her-ring, sardines, anchovies and young salmon (Place1992) Provided that a sustainable harvesting regime

is established, marine wax esters could contribute to

a signi¢cant portion of the supply for marine HUFAfor human consumption that would otherwise be in-accessible This review provides therefore new infor-mation in the ¢eld of using marine zooplankton with

a high level of wax esters as lipid source in feeds toAtlantic salmon, as an alternative to ¢sh oils andvegetable oils

Origin of wax estersWax esters are esters of a long-chain fatty acid and

a monohydric long-chain fatty alcohol (Cowey &

Table 1 Biomass and production (wet weight) of macrozooplankton standardized to an area of 3.1 million km2, ing to the total area of Norwegian and Barents Seas and eastern parts of the Greenland and Iceland Seas (reproduced from Suontama 2006)

correspond-Species/group

Biomass (mill tones)

Production (mill tones)

Original area (mill km 2 ) Source

Euphausiids (krill) 161 242 3.1 W Melle, unpubl obs.

Calanus finmarchicus 22w 88 2.9 Aksnes & Blindheim, (1996)

Calanus ssp 30–125 120–500w 3.1 Hassel & Melle, (1999)

Calanus ssp 75 298w 3.1 Holst, Couperus, Hammer, Jacobsen, Ja´kupsstovu,

Krysov, Melle, Mork, Tangen Vilhja´lmsson and Smith (2000)

Based on P/B 51.5 (Sakshaug, Bjorge, Gulliksen, Loeng & Mehlum 1994).

wBased on P/B 5 4 (Sakshaug et al., 1994).

The area represents main distributional area for the species.

Sargent 1977; Christie 2003; Fig 1) These are neutral

lipids with low polarity and high hydrophobicity This

property can be attributed to the fatty alcohol part

that consists of a limiting number of chain lengths

(C14 22) which are often saturated or

monounsatu-rated However, the high level of unsaturated fatty

acid in marine wax esters keeps it liquid at least

down to 0 1C (Spark, 1982)

Distribution and function of wax esters

Phytoplankton contain little or no wax esters, so

marine wax esters are primarily of animal origin

Sargent et al (1976) showed the appearance of

marine wax esters as a major lipid component in

seven phyla, including Coelenerate, Ctenophora,

Chaetognatha, Mollusca, Annelida, Arthropoda and

Chordata

Wax esters in copepods provide energy reserves

during periods of starvation, for maintenance and

re-production (Lee & Puppione 1972; Sargent, Gatten &

Henderson 1981; Falk-Petersen et al 2000; Lee et al

2006).Wax esters can also have a variety of other

bio-logical functions In vertical migrating copepods,

wax esters are important constituents in osmotic

reg-ulation and buoyancy control through oxidation and

biosynthesis of wax esters (Sargent et al 1976) In

some myctophid ¢shes like the diurnal Cyclothone

atraria, Hoplostethus atlanticus and Latimeria

chalum-nae, wax esters are important constituents in the

swimbladder (Phleger 1998) In a few epipelagic ¢sh

species (e.g Trichogaster cosby), wax esters are found

in the roe as nourishment for larvae before initial

feeding (Sand, Rahn & Schlenk 1973), while some

deep-water ¢sh (e.g the deep-sea cod species Lotella

phycis and Laemonema morosum) have been reported

to have wax esters as energy stores in the muscle and

other internal organs (Nevenzel 1970; Phleger et al

1997) In some whales (e.g Berardius bairdi), short

chain TAG and wax esters are found in the head

re-gion and are involved in echo-location of

zooplank-ton (Litch¢eld, Greenberg, Eazooplank-ton & Ackman 1978)

The high content of wax esters in blubber of whales

(e.g Physeter catodon) appears to act as thermal

insu-lation of the musculature and as energy supplies

un-der undesirable conditions (Nevenzel 1970; Sargent

et al 1976)

Wax esters in copepodsCopepods are the dominant zooplankton in most seasand have stores of wax esters, if present, in eggs andoil sacs (Lee et al 2006) All of the examined speciesbelonging to the calanoid copepod families Calanidae,Euchaetidae, Lucicutiidae, Heterorabdidae and Augapti-lidae have more than 20% of their lipid as wax estersand are found either in deep water or near-surfacecold-temperature waters (Lee, Barnett & Hirota 1971;Sargent et al 1976) The wax esters of these marinecopepods consist of fatty alcohols that are mostly sa-turated (14:0 and 16:0) and monounsaturated (16:1n-

7, 18:1n-9, 20:1n-9 and 22:1n-11) The fatty acid eties have the same variety of chain lengths as fattyacids (C14  24) in TAG, with a high level of saturated(14:0 and16:0), monounsaturated (20:1n-9 and 22:1n-11) and n-3 (18:4n-3, 20:5n-3 and 22:6n-3) fatty acids(Bauermeister & Sargent 1979a; Kattner 1989) Thechain length of marine wax esters (total of alcoholplus acid) are thus in the range of C28 44, with C32,

moi-C34, C36and C38as the major components in water zooplankton while upper-water zooplanktonhave C42and C44as additional components (Sargent

deep-et al 1976) This is assumed to be caused by lower vels of polyunsaturated fatty acid (PUFA) in their nat-ural diets compared with photosynthetic algaes anddiatoms occurring in the pelagic water column ofArctic waters (Benson, Lee & Nevenzel 1973) The co-pepod, C ¢nmarchicus (Fig 2), constitutes the largestzooplankton biomass in mixed polar and Atlanticwaters and is distributed throughout the NorthAtlantic Ocean (Marshall & Orr 1955; Conover 1988).They are important prey organisms for larval stages

le-of many commercial ¢sh stocks, including cod, dock (Buckley & Lough 1987; Lynch, Lewis & Werner2001), sardines, herring, anchovy (Marshall & Orr1955; Dalpadado, Ellertsen, Melle & Dommasnes2000), red¢sh (Runge & de Lafontain 1996) and cape-lin (Astthorsson & Gislason 1997)

had-Calanus ¢nmarchicus has a predominantly 1-yearlife cycle in the Norwegian Sea (Melle, Ellertsen &Skjoldal 2004) During development from egg toadult, C ¢nmarchicus passes through six naupliarstages and ¢ve copepodid stages (I^V) All thesestages occur in the upper 100 m of the water columnduring spring and summer (Speirs, Gurney, Heath

& Wood 2005) The copepod shows exponentialFigure 1 General molecular structure of wax esters

increases in total lipid and wax ester content

throughout the development from nauplia to stage V

copepodid (Fig 2) In general, the total lipid content

increases from 8% of dry weight in copepodid I to a

maximum of 64% in copepodid V, while the wax ester

content increases from around 40% of lipid in

cope-podid I to almost 90% in copecope-podid V (Kattner &

Krause 1987; Scott, Kwasniewski, Falk-petersen &

Sargent 2000; Lee et al 2006) There is, however,

some uncertainty in the data as results seem to vary

with di¡erent studies (reviewed by Lee, Barnett et al

1971; Lee & Hirota 1973, Lee et al 2006) Kattner and

Krause (1987) showed a variation of more than10% in

the lipid level within copepodid stage between two

close locations in the North Sea sampled during

spring 1983^1984 This discrepancy could thus be

due to sampling at di¡erent time and location or

method of sampling and analysis At the end of the

spring bloom, the predominant stages are copepodid

IVor V (Lee, Nevenzel & Pa¡enh˛fer 1972) Stage

Van-imals then descend to deep waters (600^1000 m) in

late summer where they remain semi-torpid during

the winter months (Speirs et al 2005) These

indivi-duals return to the surface as adults in the late winter

season where the females spawn in conjunction with

the phytoplankton bloom (Melle et al 2004)

Biosynthesis of wax esters

Fatty alcohols of C ¢nmarchicus are produced in two

steps Firstly, fatty acids are synthesized by a de novo

pathway resulting in fatty acids (Benson & Lee 1975;

Bauermeister & Sargent 1979a; Sargent et al 1981) Inthe second step, the fatty acids are reduced to fattyalcohols with a strict preference for saturated andmonounsaturated products (fatty acids) The main al-cohols are therefore rich in 16:0 (10%), 20:1n-9 (23%)and 22:1n-11 (45%) (Fraser, Sargent & Gamble 1989;Kattner 1989; Albers, Kattner & Hagen 1996; Scott,Kwasniewski, Falk-petersen & Sargent 2002; Table2) The ability to produce wax esters is unique to ani-mals having the enzyme NADPH oxidoreductase(Sargent, Gatten & Mcintosh 1974)

Although copepods can produce fatty acids for waxesters by the de novo pathway, they will also incorporatedietary fatty acids The composition will therefore varywith type of prey available at the time For example, dia-toms (e.g Sceletonema costatum) and dino£agellates (e.g.Gymnodinium breve) are regarded as important prey for

C ¢nmarchicus (Marshall & Orr 1955) and will supplylarge amounts of 20:5n-3 and 22:6n-3, respectively, to

be utilized as essential fatty acids and deposited in depotlipids (Lee, Nevenzel & Pa¡enh˛fer 1971; Sargent, Gat-ten, Corner & Kilvington 1977; Leblond & Chapman2000; Table 2) Thus, C ¢nmarchicus feeding on thesediets stores a high level of n-3 PUFA in the polar lipidfraction (60%), while saturated and monounsaturatedfatty acids are more pronounced in TAG and wax esters(25^39%) (Fraser et al.1989)

The synthesis of wax esters is initiated when food

is in excess, and follows a seasonal pattern (Lee, venzel et al 1971; Kattner 1989) The production offatty alcohols and wax esters appears to decreaseTAG synthesis (Sargent et al 1976) The fatty alcoholscondense with fatty acyl CoA (dietary or endogenous

Ne-Stage indevelopment

Lipid(% of dry weight)

Wax esters(% of lipid)

Figure 2 Calanus

(adapted from Sars 1903)and (b) variation in lipid

Atlantic (adapted fromKattner & Krause 1987and Lee et al 2006)

origin) by a reaction catalysed by a relatively

non-speci¢c ester synthetase to produce wax esters that

could be stored (Bauermeister & Sargent 1979a;

Sar-gent et al 1976)

Calanus ¢nmarchicus in ¢sh feeds

Fish oil has been the major lipid source in

manufac-tured feed to carnivorous aquaculture The oils are

normally produced from industrial ¢sh species like

herring, sardine, sand-eel, menhaden, anchoveta

and pout caught from ¢sheries in Northern Europe

or Southern America The wax ester rich calanoid

copepods are important constituents of the marine

food chain in Northern Europe Depot lipids from ¢sh

in this area do have many similarities to what is

found in C ¢nmarchicus including a high level of

long-chained monounsaturated fatty acids found in

the alcohol fraction of the wax esters (Ratnayake &

Ackman 1979; Table 3) The coast of South America

has other copepods with TAG containing fatty acids

with a high level of polyunsaturated and saturated

fatty acids that is re£ected in the oils from the ¢sh

species in this area (Tocher 2003)

Calanus Finmarchicus have traditionally been

har-vested by trawl and block frozen shortly after capture

to prevent oxidation and loss of nutrients, which

might be a problem when the animals are thawed.However, new techniques in harvest and preserva-tion are under development for production of pro-ducts with minimal loss of nutrients Lipid are at thepresent extracted from thawed Calanus that areheated to 85^90 1C in a steam-heated vessel contain-ing water/press liquid or in a scraped surface heat ex-changer (e.g Contherm 4 6; Alfa Laval, Rdovre,Denmark) Most of the liquid phase is removed in dou-ble screw press (e.g P13; Stord Bartz, Bergen, Nor-way) The press liquid is then heated to at least 90 1C

in a scraped surface heat exchanger (e.g Contherm

4 6, Alfa Laval), and particulate matter is removed

in a 100 mm wet sieve before the oil is extracted in anoil separator (e.g SA1; Westfalia Separator AG, Oelde,Germany) Low lipid diets are prepared by extrusionusing similar standards and additives as used in com-mercial diets, followed by coating with the extractedCalanus oil (Olsen, Henderson, Sountama, Hemre,Ring, Melle & Tocher 2004; Bogevik, Tocher, Lang-myhr,Waagbo & Olsen 2009)

Harvests of C ¢nmarchicus could have a large iation in wax esters content of the lipid due to loca-tion of harvest, development stage and sex of thecopepods as mentioned earlier Feed production is of-ten based on replacing a certain percentage of the

var-¢sh oil with Calanus oil Thus, the ¢nal feed productcomposition could vary after the harvest product

Table 2 Fatty acid composition in two phytoplankton species (Skeletonema costatum and Gymnodinium breve) and in adult stages of Calanus ¢nmarchicus

Olsen et al (2004) used feed where100% of the ¢sh oil

was replaced by Calanus oil with a ¢nal product

con-taining 37.5% wax esters of the lipid, while Bogevik

et al (2009) also used feed that had a 100%

replace-ment of the ¢sh oil resulting in ¢nal products that

contained 47.7% wax esters of the lipid (Table 3)

Standarization of time and location of harvest are

thus important to avoid these di¡erences in the ¢nal

product Fish oil diets from the Northern hemisphere

have high levels of monounsaturated fatty acids (45^

50%) These are seen in the form of fatty alcohols

(80%) and to a lesser extent fatty acids (30^40%) in

Calanus diets, the major being 22:1n-11 (Table 3)

Furthermore, the Calanus diets have a higher level of

18:4n-3 and 20:5n-3, re£ected in a higher level of

PUFA compared with the ¢sh oil diets (Table 3) Thus,

the Calanus diets have su⁄cient amounts of essential

fatty acids to cover requirements

Although many ¢sh species have the ability to

uti-lize dietary wax esters, the rate of intestinal

hydroly-sis appears to be lower than for TAG (Patton,

Nevenzel & Benson 1975; Tocher & Sargent 1984) The

fact that wax ester-rich animals are normal prey for

wild Atlantic salmon (Salmo salar) (Rikardsen,

Haug-land, Bjorn, Finstad, Knudsen, Dempson, Holst,

Hvid-sten & Holm 2004) suggests an evolutionary

adaptation to e¡ectively utilize these lipid sources

The actual e⁄ciency of utilization has, however, notbeen the subject of extensive studies There is thus ageneral lack of knowledge on the utilization of waxester-rich oils by farmed salmon

Hydrolysis of wax esters in fishBile salt-dependent lipase (BSDL) is believed to be thepredominant lipase in ¢sh rather than the colipase-dependent pancreatic lipase that is predominant inmammals (Patton et al 1975; Lie & Lambertsen 1985).Bogevik, Tocher, Waagbo and Olsen (2008b) showedthat salmon lipases were more active in bile becausedesalting midgut extract reduced lipolytic activity bymore than 70% Bile salt-dependent lipase is there-fore most likely responsible for hydrolysis of TAG, aswell as wax esters and sterol esters in salmonids Thehydrolytic activity in the gut is, however, much lowerfor wax esters than for TAG (Patton et al 1975; Tocher

& Sargent 1984; Olsen & Ring 1997) Consequently,TAG are hydrolysed faster than wax esters (Patton

et al.1975) This is seen in both in vivo and in vitro dies, where BSDL has been reported to hydrolyze waxesters at rates of1^2 orders of magnitude slower thanTAG and four- to ¢vefold slower than sterol esters

stu-in anchovy (Engraulis mordax) and rastu-inbow trout

Table 3 Neutral lipid content (% of lipid classes) and fatty acid composition (wt%) of Southern hemisphere ¢sh oil [(anchovy oil (AO)] and diets with Northern hemisphere ¢sh oil (FO) or elevated level of wax esters through inclusions of copepod oil (CO) extracted from Calanus ¢nmarchicus

(Patton et al 1975; Patton, Warner & Benson1977;

Tocher & Sargent 1984) A recent study by Bogevik

et al (2008b) showed that desalted midgut extract

from salmon and rainbow trout hydrolysed wax

es-ters fourfold faster than sterol eses-ters, but

approxi-mately ¢vefold slower than TAG The higher activity

against wax esters in this study could be due to

di¡er-ent enzyme assay techniques Whether the slower

hydrolysis of wax esters is due to the greater

hydro-phobicity of wax esters, associated with lower biliary

emulsi¢cation or the speci¢city of the enzyme is still

unclear Feeding higher portions of wax esters would,

however, give slower release of lipolytic products

than from TAG, which is more or less completely

hy-drolysed to fatty acids in salmon (Oxley, Bogevik,

Henderson,Waagb,Tocher & Olsen 2009) However,

there is a di¡erence in the lipolytic activities between

salmon and rainbow trout upon activation by bile In

a previous study, Tocher and Sargent (1984) showed

that rainbow trout intestinal lipase was inhibited at

bile salts concentrations of around 10 mM (Tocher &

Sargent 1984) This was also observed for rainbow

trout midgut extract incubated with both TAG and

sterol esters of Bogevik et al (2008b) However, this

was not seen in salmon incubations Accordingly,

there appear to be di¡erences in the bile salt

activa-tion of the luminal lipid enzyme between these two

species

The rate of hydrolysis is not only dependent on the

substrate being TAG or wax esters but also on fatty

acid/fatty alcohol unsaturation and chain length

Analyses of wax esters in C ¢nmarchicus have shown

that the long-chain monounsaturated fatty alcohols

20:1n-9 and 22:1n-11 are esteri¢ed predominantly to

shorter chain fatty acids such as 14:0, whereas the

medium-chain fatty alcohol 16:0 is esteri¢ed mostly

to PUFA, particularly 18:4n-3 (Sargent & Henderson

1986) Several authors, reviewed by Olsen and Ring

(1997), have stated that PUFA appear to be released at

higher rates than monounsaturated followed by

sa-turated fatty acids Furthermore, longer chain fatty

acids seem to be hydrolysed at decreasing rates with

increasing chain length However, Bogevik, Oxley

and Olsen (2008) showed preferential hydrolysis of

16:0 compared with 20:1n-9 in ¢sh oil TAG

hydro-lysed by salmon lipases, indicating long-chained

monounsaturated fatty acids were poor substrates

for the lipase It follows that a preference of the

sal-mon digestive lipase for ester linkages of unsaturated

fatty acids would cause decreases in the proportions

of 18:4n-3 and 16:0 alcohols in faecal lipid as

ob-served in Olsen et al (2004), Bogevik et al (2009)

and Oxley et al (2009) At the same time, molecularspecies of wax esters containing 14:0 fatty acid ester-i¢ed to 20:1n-9 and 22:1n-11 alcohols would be poorsubstrates for the lipase and would have low absorp-tion rates In keeping with this, there was a substan-tial increase in the proportions of 14:0 fatty acid and22:1n-11 alcohol in the faeces lipid relative to that ofthe diet (Olsen et al 2004; Bogevik et al 2009; Oxley

et al 2009)

Wax esters have a higher melting point than TAGand are intrinsically more hydrophobic, and there-fore require proper emulsi¢cation for their hydrolysis

by lipases The high dietary wax ester:TAG ratio (4:1)

in diets with 100% replacement of ¢sh oil with nus oil (Bogevik et al 2009; Oxley et al 2009) couldhave limited the excess of the lipid for the lipases.Comparably, 100% replacement of ¢sh oil with Cala-nus oil in Olsen et al (2004) resulted in a ratio ofo2:1 between wax esters and TAG, due to a batch of

Cala-C ¢nmarchicus with less wax esters (Table 3) Thisseemed su⁄cient for proper emulsion and compar-able digestion to a ¢sh oil diet rich in TAG (Olsen

et al 2004) Thus, the hydrophobic properties of waxesters seem to be dependent on a certain level of TAG

to form an emulsion readily accessible for salmonlipases

Absorption and metabolism of productsfrom wax ester hydrolysis

Luminal fatty acids are solubilized in bile to formmixed micelles before absorption (Fig 3) These areabsorbed into the enterocytes mainly by passive dif-fusion at high luminal concentration, proceeding bycarrier-mediated mechanisms, either facilitated dif-fusion or active transport, at low concentrations(Carlier, Bernard & Caselli 1991; Tso, Nauli & Lo2004) Increased absorption with increased chainlength and increased unsaturation (and decreasedpolarity) has been seen in trout enterocytes (Perez,Rodriguez & Henderson 1999; Oxley, Tocher, Torsten-sen & Olsen 2005) This was also observed in recentstudies where absorption of 18:1n-9 was higher than16:0 using enterocyte assay (Bogevik,Tocher,Waagbo

& Olsen 2008a), and for apparent digestibility lated from the comparison between dietary and fae-cal lipid content in salmon (Bogevik et al 2009; Oxley

calcu-et al 2009)

Wax ester-fed ¢sh seem to absorb fatty acids morereadily than fatty alcohols This is supported by theelevated levels of 22:1n-11 alcohol in faeces of

herring, rainbow trout (Sargent, Mcintosh,

Bauerme-ister & Blaxter1979), cod (Lie & Lambertsen1991) and

salmon (Olsen et al 2004) fed wax esters from C

¢n-marchicus The rate-limiting step in wax ester

assimi-lation could therefore be the absorption of free fatty

alcohols, especially 22:1n-11 alcohol (Sargent et al

1979) Absorption of fatty alcohols has not been

stu-died in detail However, the e⁄ciency of absorption

seems to vary with species Rainbow trout, two-spot

goby, scad and Atlantic cod fed C ¢nmarchicus have

up to four times more fatty alcohols than fatty acids

in their faecal fraction (Sargent et al 1979; Prahl,

Eglinton, Corner & Ohara 1985; Olsen, Henderson &

Pedersen 1991) The absorption appeared to be more

e⁄cient in herring as less fatty alcohols than fatty

acids were seen in the faecal fraction, suggesting that

this species may be especially adapted to diets rich in

wax esters (Sargent et al.1979) The absorption of fatty

alcohols in Atlantic salmon has also been assumed to

be rapid because fatty acids and fatty alcohols have

been reported to be present in equal amounts in

fae-cal materials (Olsen et al 2004) However, Bogevik

et al (2008a) showed that the lipolytic products of

wax esters were absorbed di¡erently in a

dual-la-belled fatty acid^fatty alcohol metabolism assay in

sampled enterocytes For both substrates tested,

(16:0 and 18:1n-9), there was a higher uptake of fatty

acids than of fatty alcohols within a 2 h incubation

period Thus, passive absorption is favoured by highluminal fat concentration and decreased polarity(Tso et al 2004) The favoured absorption of fattyacids over the lesser polar fatty alcohols at low lumi-nal fat concentration suggests that the transport iseither facilitated or active, and to a less extent pas-sive Furthermore, substrates delivered in ethanol so-lution in this study likely had properties that weredi¡erent from fatty acids and fatty alcohols dissolved

in micelles in vivo Increased fatty acid absorptioncould also be caused by non-luminal absorption, asfatty alcohols could exclusively be absorbed on the lu-minal side However, the fatty alcohol absorption wasonly one-third of the absorption of fatty acids indicat-ing a highly favourable absorption of fatty acids.Absorbed fatty alcohols are converted to corre-sponding fatty acids through two-step oxidation byfatty alcohol dehydrogenase and aldehyde dehydro-genase (Sund & Theorell 1963; Bauermeister & Sar-gent 1978; Bauermeister & Sargent 1979b) Normally,all fatty alcohol oxidation is assumed to occur in theintestine, but as fatty alcohols have been detected inother organs, e.g liver, these may be of importancewhen the intestinal metabolic capacity is overloaded(Patton & Benson 1975) Absorption of fatty alcoholsappeared to be more rapid than their oxidation asthey accumulated in the enterocytes before conver-sion (Bogevik et al 2008a) In vitro studies of entero-

2-cytes from salmon fed the Calanus diet tended to

have a lower fatty alcohol accumulation and a larger

fatty acid accumulation compared with enterocytes

sampled from salmon fed the ¢sh oil diet (Bogevik

et al 2008a,b; Bogevik, Oxley et al 2008) This

indi-cates an up-regulation in the ability to oxidize fatty

alcohols upon wax ester feeding

Once inside the enterocytes, the metabolism of the

fatty acids (including oxidized fatty alcohols) can

pro-ceed through several pathways depending on the

me-tabolic status of the animal (Henderson 1996) In

general, the bulk of absorbed fatty acids (alcohols)

are esteri¢ed to TAG and incorporated into

lipopro-teins and thus distributed throughout the body

(Hil-ditch, 1964; Lee & Puppione 1972; Benson et al 1973;

Bauermeister & Sargent 1979b; Fig 3) Some fatty

acids may be elongated/desaturated before transport

but the extent of this modi¢cation is not known

(Henderson 1996; Tocher 2003) However, liver TAG

in salmonids have major similarities to dietary lipids

indicating that these processes are not quantitatively

important (Tocher 2003; Stubhaug,Tocher, Bell, Dick

& Torstensen 2005) Smaller portions of fatty acids

are esteri¢ed to other lipid classes or used in oxidative

processes (Oxley et al 2005) The synthesis of TAG

probably does not proceed through the

monoacylgly-cerol acyltransferase pathways as digestion of wax

esters will not produce 2-monoacylglycerol Rather,

in these cases, TAG synthesis probably proceeds

through the slower and more energetic consuming

G3P pathway (Johnston 1977; Bauermeister &

Sar-gent 1979b) Whether the small amount of glucose

necessary for glycerogenesis in intestines stems from

a small quantity of dietary glucose presented

lumin-ally to the intestine, through glyceroneogenesis from

luminal amino acids or from blood glucose is not

known However, the fatty alcohol incorporation into

TAG proceeds at the same rate irrespective of

whether the trout are maintained on wax ester or

TAG-based diets suggesting that their capacity to

me-tabolize fatty alcohol is not rate limiting

(Bauermeis-ter & Sargent 1979b) It was shown by Bogevik et al

(2008a) that chain length and saturation (16:0 and

18:1n-9) and oxidative state (fatty acid/fatty alcohol)

had only a minor in£uence on the degree of

esteri¢-cation to di¡erent lipid classes Both,16:0 and 18:1n-9

were to a minor extent chain-elongated/desaturated

and were preferentially esteri¢ed into TAG The rate

of incorporation of 18:1n-9 was higher than that of

16:0, which is in agreement with Oxley et al (2005)

There was also more 16:0 incorporated into

phospha-tidylcholine (PC) and phosphatidylserine than

18:1n-9 A minor fraction of the fatty acids were oxidized,where dietary treatment of ¢sh oil with a higher level

of monounsaturated fatty acids seem to improve dation of monounsaturated substrate compared withenterocytes from the Calanus oil-fed ¢sh (Bogevik

oxi-et al 2008a)

Factors regulating wax ester digestion

in fishThe release of su⁄cient amounts of bile and enzymes

in the intestines of ¢sh is crucial for optimal digestion(Mankura, Kayama & Saito 1984; Tuchweber, Yousef,Ferland & Perea 1996; Tocher 2003) Bogevik et al.(2009) elucidated the potential of salmon to adaptthe digestive capacity following wax ester feeding.One major ¢nding was that the gallbladder volumeincreased upon wax ester feeding (also seen in Oxley

et al (2009) for salmon in freshwater and seawater),which agreed with previous reports from rainbowtrout (Tocher & Sargent 1984) For both these studieswas, the gallbladder volume at fasting in average0.2 mL bile kg ¢sh 1larger in salmon fed diets coatedwith Calanus oil compared with diets coated with ¢shoil Salmon bile contains two major bile salts, tauro-cholate (420^450 mM) and taurochenodoxycholate(10^40 mM), in addition to phospholipids (mainlyPC) and cholesterol (Bogevik et al 2009; Oxley et al.2009) The increased bile volume in wax ester-fed sal-mon was with unchanged bile composition com-pared with salmon fed on a ¢sh oil diet in Bogevik

et al (2009), while there was a lower taurocholateconcentration for the wax ester group in Oxley et al.(2009) The reason for this discrepancy is unknown.One possible explanation could be the slightly lowerlevel of the bile precursor, cholesterol, in the Calanusdiet in Oxley et al (2009) (6.5%) compared with inBogevik et al (2009) (8.1%), or it is possible that thecholesterol production of bile is size-regulated as the

¢nal weight was larger in Bogevik et al (2009) (500 g)than in Oxley et al (2009) (300 g)

The lipolytic activity in salmon midgut was creased upon feeding the Calanus oil-rich diet com-pared with a ¢sh oil diet (Bogevik et al 2009) Thiswas especially seen by the higher hydrolysis of cho-lesterol esters and wax esters after 4 h of incubation.The increased enzyme activity was not due to in-creased bile volume as the assays were performed at

in-a constin-ant bile sin-alt concentrin-ation of tin-aurocholin-ate Inaddition to regulation of bile and lipases, some birdshave adapted a slower gastric emptying and re£ux of

digesta to the gizzard for further processing when

feeding on wax ester-rich diets (Place 1992) It is

pos-sible that salmon also have adapted a system of

de-layed gastric empting and prolonged digestion in the

pyloric caeca However, this hypothesis needs to be

veri¢ed Because high levels of dietary lipid may

cause an overload of the digestive capacity, it may

also lead to delayed gastric evacuation (Windell &

Norris 1969) allowing more time for the utilization

(Olsen & Ring 1997) In general, however, increased

dietary lipid above ‘normal levels’ tend to cause

re-duced digestibility in some ¢sh including cod (Lie,

Lied & Lambertsen 1988), cat¢sh, Ictalurus punctatus

(Andrews, Murray & Davis 1978) and Siberian

stur-geon, Acipenser baeri (Medale, Blanc & Kaushik

1991) Other ¢sh including rainbow trout (Takeuchi,

Watanabe & Ogino 1978; Choubert, Delanoue & Blanc

1991) and Atlantic halibut (Berge & Storebakken

1991) appear to have a relatively high tolerance to

in-creased dietary lipid content The lipid level in the

studies by Bogevik et al (2009) and Oxley et al

(2009) was given in equal amount within normal

le-vels, and was therefore not considered to a¡ect lipid

digestion The increased level of bile and enzyme

ac-tivity was thus a result of presence of wax esters in

the diets

Digestibility of wax ester rich diets

Apparent lipid digestibility was basically similar in

salmon fed ¢sh oil-based diets and wax ester-based

diets even when the level of wax esters exceeded

30% of the dietary lipids (Olsen et al 2004; Bogevik

et al 2009; Table 4) This implies that the increased

enzyme activity and bile content in ¢sh fed wax

es-ters presumably increased hydrolysis and

conse-quently absorption through better solubilization ofthe lipolytic products However, at higher inclusion

of wax esters close to 50% of the dietary lipids, lipiddigestibility was reduced This probably relates to im-paired solubilization of wax esters and therefore low-

er digestibility even with high levels of bile (Bogevik

et al 2009; Oxley et al 2009; Bogevik, Henderson,Mundheim, Waagbo, Tocher & Olsen 2010; Table 4).Wax esters of 20:1n-9 and 22:1n-11 alcohols esteri¢ed

to 14:0 are seen at a higher level in faeces than in thediets and are thus poorly digested and display there-fore a lower digestibility (Olsen et al 2004; Bogevik

et al 2009; Oxley et al 2009; Bogevik et al 2010) Thus,

a high fraction of unhydrolysed wax esters seems to

be the main reason for lower digestibility when ing a high level of wax esters This is further seen inthe preferential utilization of unsaturated to satu-rated fatty acids in these studies, where ¢sh utilizedPUFA more e⁄ciently than monosaturated and satu-rated fatty acids (Table 4)

feed-It appears that there is a discrepancy between terocyte absorption of fatty acids and fatty alcoholsand digestibility of these products In vitro studies ofsalmon enterocyte showed absorption that was high-

en-er for saturated fatty acids than for fatty alcohols gevik et al 2008a), while digestibility implies thatsaturated fatty alcohols are more readily taken upthan saturated fatty acids (Bogevik et al 2009; Oxley

(Bo-et al 2009; Bogevik (Bo-et al 2010) This discrepancy is scure and remains to be elucidated However, thepoor digestion and absorption of luminal long-chainsaturated fatty acids (Sigurgisladottir, Lall, Parrish &Ackman 1992; Olsen, Henderson & Ring 1998),where suggested by Olsen et al (2004) not to hold forsaturated alcohols due to substrate speci¢city of li-pases This implies that a large portion of saturatedfatty acids remains in the fraction of TAG in the lumi-

ob-Table 4 Apparent digestibility coe⁄cient (ADC, %) of dry matter, total lipid, saturated fatty acids/alcohols (SFA), turated fatty acids/alcohols (MUFA), polyunsaturated fatty acids/alcohols (PUFA) and total fatty acids/alcohols in Atlantic sal- mon at di¡erent ¢nal weight (250^1500 g), reared in either freshwater (FW) or seawater (SW) and fed di¡erent inclusion of dietary wax in their diets Adapted from Olsen et al (2004), Bogevik et al (2009) and Oxley et al (2009)

nal content and are therefore not available for

ab-sorption while saturated fatty alcohols, especially

16:0, are more readily hydrolysed and absorbed into

enterocytes and display therefore higher digestibility

In the wild, marine wax esters are not naturally

in-troduced to salmonids before entering the sea

(Ri-kardsen et al 2004) Accordingly, seawater adaptive

physiology (smolti¢cation) must not only increase

hypoosmoregulatory ability (Sheridan 1989) but may

also increase the ability to digest marine diets This is

seen through the increased lipolytic activity and

low-ered fatty acid de novo synthesis occurring before

sea-water adaptation (Sheridan 1989) The altered lipid

metabolism results in a change in fatty acid

composi-tion from high levels of saturated fatty acid in

fresh-water parr to a higher content of PUFA in seafresh-water

smolts (Sheridan 1989, 1994) Oxley et al (2009)

showed that the digestibilities of polyunsaturated

fatty acids/alcohols were una¡ected by salinity, while

the digestibility of saturated and monosaturated fatty

acids/alcohols was signi¢cantly lower in post-smolt

salmon in seawater than pre-smolt in freshwater

This was most likely caused by the increased

drink-ing activity for osmoregulation to form insoluble

soap of saturated/monosaturated fatty acids and

di-valent cations (Ca21) in the intestinal lumen of

sea-water ¢sh for increased secretion of these fatty acids

in the faeces (Olsen & Ring 1997) This could also be

the case for the lower digestibility of fatty alcohols in

sea water-reared salmon

Lipid digestibility is known to decrease with

in-creasing melting point (Olsen & Ring 1997) Thus,

saturated fatty acids with low melting point are

known to be harder to digest than PUFA;

conse-quently, by lowering the temperature, this will

further lower the availability of saturated fatty acids

(Olsen & Ring 1998) Salmon fed a Calanus oil diet

with a high level of wax esters (50% of dietary lipid)

showed signi¢cantly lower digestibility of saturatedfatty acid/alcohols at 3 1C compared with ¢sh reared

at 12 1C The lipid digestibility was furthermore

high-er in salmon fed ¢sh oil-based diets than ¢sh fed a dietwith Calanus oil at both temperatures (Bogevik et al.2010; Table 5)

Effect of wax ester rich diets on growthEarlier studies on wax ester digestion in the 1970sgave poor growth when feeding rainbow trout blockfrozen copepods compared with ¢sh fed commercialdiets (Sargent et al 1979) The main reason for thiswas suggested to be loss of vitamin or protein duringthawing (Brett 1971; Sargent et al 1979) However,diets with TAGs and wax esters in a mixture seemed

to improve incorporation into tissue lipid comparedwith ¢sh fed pure wax esters (Patton & Benson1975) This implies as mentioned above that a ratiobetween wax esters and TAG are important to enablesu⁄cient solubilization of lipids to ease the access ofthe enzyme and therefore improve digestibility andgrowth Formulated diets with similar compositionand coated with either ¢sh oil or oils from C ¢n-marchicus have shown previously not to a¡ectgrowth and lipid digestion in Atlantic salmon (Olsen

et al 2004) However, there are at least two factorsthat could be crucial for the successful use of thisoil Firstly, the level of wax esters or the ratio wax es-ters to TAG seems to be important The low wax es-ter:TAG ratio (2:1 and 1:1) in Olsen et al (2004) andBogevik et al (2009), respectively, seems to ensurehigh digestibility and good growth in salmon fed a

C ¢nmarchicus oil diet with 37.5% and 30.7% wax ter of the lipid respectively Increased ratio (4:1) signif-icantly reduced growth compared with control ¢sh(Bogevik et al 2009; Oxley et al 2009; Bogevik et al

es-Table 5 Apparent digestibility coe⁄cient (ADC, %) of dry matter, total lipid, saturated fatty acids/alcohols (SFA), turated fatty acids/alcohols (MUFA), polyunsaturated fatty acids/alcohols (PUFA) and total fatty acids/alcohols in Atlantic sal- mon grown from 450 to 900 g, reared at either 3 or12 1C and fed either high fat ¢sh oil (HFFO), high fat Calanus oil (HFCO), low fat ¢sh oil (LFFO) or low fat Calanus oil (LFCO) diets Adapted from Bogevik et al (2010)

2010; Fig 4) However, salmon fed these diets

in-creased their feed intake compared with ¢sh fed ¢sh

oil diets This could be due to lower lipid digestibility

in wax ester-fed ¢sh For ¢sh fedo40% wax esters,

this was su⁄cient to sustain growth However, for

those fed close to 50% wax esters, increased feed

in-take would not prevent lower growth This implies

that there might be an upper limit for optimal

utiliza-tion of wax esters in practical diets, with an upper

limit being between 30% and 40% of dietary lipid

Secondly, the life history and size/age of the ¢sh

can be an important factor for optimal utilization of

dietary wax esters Atlantic salmon in freshwater

feed on large amounts of insects with high levels of

TAG (Levings, Hvidsten & Johnsen 1994) They are

thus not dependent on an e⁄cient wax ester

diges-tion system.When the salmon enter the sea, they are

presented new lipids including wax esters from many

crustacean species (Rikardsen et al 2004) It would

therefore be expected that salmon post-smolts may

have some adaption to utilize wax ester-rich diets

Nevertheless, salmon post-smolt showed similar

ne-gative growth response as salmon pre-smolt upon a

48% wax ester diet (Oxley et al 2009) This suggests

that there is no pre-adaptation to wax ester

utiliza-tion in salmon post-smolts, although this remains to

be veri¢ed Furthermore, there might be an e¡ect of

size/age on utilization and growth Larger/older ¢sh

seem to grow better on wax ester-rich diets than

smaller/younger ¢sh This is possibly caused by the

higher luminal lipolytic activity in 1500 g salmon

compared with 300 g ¢sh (Bogevik et al 2008b)

Sal-mon-post smolt at 500 g in Bogevik et al (2009) given

a diet with 30.7% wax esters showed a tendency of

larger di¡erences in growth to the ¢sh oil diet than

the post-smolt at 1500 g in Olsen et al (2004) given

37.5% wax esters (Fig 3) This indicates that salmon

at larger size might tolerate a higher level of wax ters in the diet than smaller ¢sh

es-ConclusionsHydrolysis of TAG are faster than for wax esters, from1^2 magnitudes to four- to ¢vefold dependent on spe-cies and experimental techniques Thus, there is afaster release of fatty acids from TAG than fatty acidsand fatty alcohols from wax esters Consequently,upon wax ester feeding a larger proportion of wax es-ters will be present in the faeces while most of theTAG will be hydrolysed Salmon enterocytes absorbfatty acid faster than fatty alcohols Lower level of li-pids is therefore available for metabolism in entero-cytes upon wax ester feeding The ability to oxidizefatty alcohols seems to be improved upon wax esterfeeding as tendencies to lower accumulation of intra-cellular fatty alcohols are seen in enterocytes from

¢sh pre-fed a wax ester-rich diet Although the take is di¡erent between the fatty acids and the fattyalcohols, the intracellular metabolism are presum-ably the same after fatty alcohols were oxidized tofatty acids, and thereby preferentially esteri¢ed toTAG

up-Diets with a high level of wax esters have generally

a lower digestibility due to lower hydrolysis and sorption of fatty alcohols However, higher bile con-tent and lipases activity upon feeding wax esters tosalmon seem to improve hydrolysis and consequentlyabsorption through better solubilization of lipolyticproducts, even though the level of wax esters ex-ceeded 30% of the dietary lipid However, impairedsolubilization of lipids at higher inclusion of wax es-ters close to 50% of the dietary lipids lowered the di-gestibility even with high levels of bile Based on allthe literature covered in this review, there appears to

ab-be an upper limit for wax ester inclusion, closer to30% than 50% of dietary lipid, for optimal growth inpractical diets for Atlantic salmon However, theremight also be a size-/age-related e¡ect on intestinaladaptation to dietary wax esters, where the wax es-ter:TAG ratio could be a limiting factor in the utiliza-tion of lipid in post-smolts o500 g Furthermore,there does not seem to be an pre-adaptation to waxesters utilization in smolts Future use of oils from C

¢nmarchicus in diets has to be based on the inclusionlevel of wax esters and not the volume of Calanus oil.Before Calanus oil can be used on a large scale, levels

of sustainable harvest of C ¢nmarchicus must be

Figure 4 Speci¢c growth rate (SGR, %/day) of Atlantic

salmon fed elevated level of wax esters over a period of

ap-proximately 100 days with a ¢nal weight from 250 to

1500 g either in freshwater (FW) or in seawater (SW)

Adapted from Olsen et al (2004), Bogevik et al (2009) and

Oxley et al (2009)

determined, feed production techniques require

further development and the long-term e¡ects of

using diets with elevated levels of wax esters in

di¡er-ent species should be investigated

Acknowledgement

This is a review of my PhD thesis; ‘Marine wax ester

digestion in salmonid ¢sh’ The PhD study was funded

by the Research Council of Norway project ‘Marine

wax esters as feed resource for farmed ¢sh’ (Grant

no 165051) The project was supervized by senior

scientist Rolf Erik Olsen at the Institute of Marine

re-search Prof RuneWaagb at the National Institute of

Nutrition and Seafood Research/University of

Ber-gen, senior scientist Harald Mundheim and senior

scientist Eyolf Langmyhr at NOFIMA and Dr Douglas

Tocher and Dr Jim Henderson at the University of

Stir-ling are all thanked for provided technical support,

help and advice in this project

References

Aksnes D.L & Blindheim J (1996) Circulation patterns in the

North Atlantic and possible impact on population

dy-namics of Calanus ¢nmarchicus Ophelia 44, 7–28.

Albers C.S., Kattner G & Hagen W (1996) The compositions

of wax esters, triacylglycerols and phospholipids in Arctic

and Antarctic copepods: evidence of energetic

adapta-tions Marine Chemistry 55, 347^358.

Andrews J.W., Murray M.W & Davis J.M (1978) In£uence of

dietary-fat levels and environmental-temperature on

di-gestible energy and absorbability of animal fat in cat¢sh

diets Journal of Nutrition 108,749^752.

Astthorsson O.S & Gislason A (1997) On the food of capelin

in the subarctic waters north of Iceland Sarsia 82, 81^86.

Bauermeister A & Sargent J.R (1978) Puri¢cation and

prop-erties 1 of an alcohol dehydrogenase from the liver and

in-testinal caecum in rainbow trout (Salmo gairdnerii).

Biochemical SocietyTransactions 6, 222^224.

Bauermeister A & Sargent J.R (1979a) Wax esters ^ major

metabolites in the marine environment Trends in

Bio-chemical Sciences 4, 209^211.

Bauermeister A.E.M & Sargent J.R (1979b) Biosynthesis of

triacylglycerols in the intestines of rainbow trout (Salmo

gairdnerii) fed marine zooplankton rich in wax esters

Bio-chimica et Biophysica Acta 575, 358^364.

Bell J.G., McEvoy J., Tocher D.R., McGhee F., Campbell P.J &

Sargent J.R (2001) Replacement of ¢sh oil with rapeseed

oil in diets of Atlantic salmon (Salmo salar) a¡ects tissue

lipid compositions and hepatocyte fatty acid metabolism.

Benson A.A & Lee R.F (1975) Role of wax in oceanic food chains Scienti¢c American 232,77^86.

Benson A.A., Lee R.F & Nevenzel J.C (1973) Wax esters ^ jor marine metabolic energy sources In: Biochemical So- ciety Symposia (London), Vol 35, pp 175–185.

ma-Berge G.M & Storebakken T (1991) E¡ect of dietary-fat level

on weight-gain, digestibility, and ¢llet composition of Atlantic halibut Aquaculture 99, 331^338.

Berman P., Harley E.H & Spark A.A (1981) Keriorrhoea ^ the passage of oil per rectum - after ingestion of marine wax esters South African Medical Journal 59,791^792 Bogevik A.S., Tocher D.R., Waagb R & Olsen R.E (2008a) Utilization and metabolism of palmityl and oleoyl fatty acids and alcohols in caecal enterocytes of Atlantic salmon (Salmo salar L) Aquaculture Nutrition 14, 270^280.

Bogevik A.S., Tocher D.R., Waagb R & Olsen R.E (2008b) Triacylglycerol-, wax ester and sterol ester-hydrolases in midgut of Atlantic salmon (Salmo salar) Aquaculture Nu- trition 14, 93^98.

Bogevik A.S., OxleyA & Olsen R.E (2008) Hydrolysis of homogeneous and ¢sh oil triacylglycerols using desalted midgut extract from Atlantic salmon, Salmo salar Lipids

acyl-43, 655^662.

Bogevik A.S.,Tocher D.R., Langmyhr E.,Waagb R & Olsen R.E (2009) Atlantic salmon (Salmo salar) postsmolts adapt lipid digestion according to elevated dietary wax es- ters from Calanus ¢nmarchicus Aquaculture Nutrition 15, 94^103.

Bogevik A.S., Henderson R.J., Mundheim H.,Waagb cher D.R & Olsen R.E (2010) The in£uence of tempera- ture on the apparent lipid digestibility in Atlantic salmon (Salmo salar) fed Calanus ¢nmarchicus oil at two fat levels Aquaculture 309, 143–151.

R.,To-Brett J.R (1971) Growth responses of young sockeye salmon (Oncorhynchus nerka) to di¡erent diets and planes of nutri- tion Journal of the Fisheries Research Board of Canada 28, 1635^1643.

Buckley L.J & Lough R.G (1987) Recent growth, ical-composition, and prey ¢eld of larval haddock (Mela- nogrammus aegle¢nus) and Atlantic cod (Gadus morhua)

biochem-on Georges bank CanadianJournal of Fisheries and Aquatic Sciences 44, 14^25.

Carlier H., Bernard A & Caselli C (1991) Digestion and sorption of polyunsaturated fatty acids Reproduction Nu- trition Development 31, 475^500.

ab-Choubert G., Delanoue J & Blanc J.M (1991) Apparent ibility of canthaxanthin in rainbow trout ^ e¡ect of diet- ary-fat level, antibiotics and number of pyloric ceca Aquaculture 99, 323^329.

digest-Christie W.W (2003) Lipid Analysis: Isolation, Separation and

Structural Analysis of Lipids The Oil Press, Bridgewater,

UK.

Conover R.J (1988) Comparative life histories in the genera

Calanus and Neocalanus in high-latitudes of the northern

hemisphere Hydrobiologia 167, 127^142.

Cowey C.B & Sargent J.R (1977) Lipid nutrition in ¢sh

Com-parative Biochemistry and Physiology B 57, 269^273.

Dalpadado P., Ellertsen B., Melle W & Skjoldal H.R (1998)

Summer distribution patterns and biomass estimates of

macrozooplankton and micronekton in the Nordic Seas.

Sarsia 83, 103^116.

Dalpadado P., Ellertsen B., Melle W & Dommasnes A (2000)

Food and feeding conditions of Norwegian

spring-spawn-ing herrspring-spawn-ing (Clupea harengus) through its feedspring-spawn-ing

migra-tions Ices Journal of Marine Science 57, 843^857.

Falk-Petersen S., Sargent J.R., Hopkins C.C.E & Vaja B (1982)

Ecological investigations on the zooplankton community

of Balsfjorden, northern Norway ^ lipids in the

euphau-siids Thysanoessa raschi and Thysanoessa inermis during

spring Marine Biology 68, 97^102.

Falk-Petersen S., HagenW., Kattner G., Clarke A & Sargent J.

(2000) Lipids, trophic relationships, and biodiversity in

Arctic and Antarctic krill Canadian Journal of Fisheries

and Aquatic Sciences 57, 178^191.

FAO (2006a) State of world aquaculture ISBN

978-92-5-105631-8 FAO, Rome, Italy.

FAO (2006b) The State of world ¢sheries and aquaculture.

ISBN 978-92-5-105568-7 FAO, Rome, Italy.

Fonseca-Madrigal J., Karalazos V., Campbell P.J., Bell J.G &

Tocher D.R (2005) In£uence of dietary palm oil on

growth, tissue fatty acid compositions, and fatty acid

metabolism in liver and intestine in rainbow trout

(Oncorhynchus mykiss) Aquaculture Nutrition 11, 241^

250.

Fraser A.J., Sargent J.R & Gamble J.C (1989) Lipid class and

fatty acid composition of Calanus ¢nmarchicus

(Gun-nerus), Pseudocalanus sp and Temora longicornis Muller

from a nutrient enriched seawater enclosure Journal of

Experimental Marine Biology and Ecology 130, 81^92.

Hansen I.A & Mead J.F (1965) Fate of dietary wax esters in

rat Proceedings of the Society for Experimental Biology and

Medicine 120, 527^532.

Hassel A & Melle W (1999) Plankton som rsto¡kilde for

fram-stilling av formidler til bruk innen ¢skeoppdrett Ett

forpros-jekt Havforskningsinstituttet, Interne notat, 3, 30pp.

Henderson R.J (1996) Fatty acid metabolism in freshwater

¢sh with particular reference to polyunsaturated fatty

acids Archives of Animal Nutrition 49, 5^22.

Higgs D.A., Balfry S.K., Oakes J.D., Rowshandeli M., Skura

B.J & Deacon G (2006) E⁄cacy of an equal blend of

cano-la oil and poultry fat as an alternate dietary lipid source

for Atlantic salmon (Salmo salar L.) in sea water I: e¡ects

on growth performance, and whole body and ¢llet

proxi-mate and lipid composition Aquaculture Research 37,

180^191.

Hilditch T.P (1964) Oils, fats and waxes Chemistry and dustry 1229.

In-Holst J.C., Couperus B., Hammer C., Jacobsen J.A.,

JaŁkupssto-vu H., Krysov A., Melle W., Mork K.A., Tangen ., jaŁlmsson H & Smith L (2000) Report on surveys of the distribution, abundance and migrations of the Norwegian spring-spawning herring, other pelagic ¢sh and the environ- ment of the Norwegian Sea and adjacent waters in the late winter, spring and summer of 2000 ICES C.M 2000/D:3 Ref ACFM.

Vilh-Imazio M., Forno D., Quaglia C & Trinchero R (2003) ga-3 polyunsatured fatty acids role in postmyocardial in- farction therapy Panminerva Medica 45, 99^107 Johnston J.M (1977) Gastrointestinal tissue In: Lipid Meta- bolism in Mammals 1,Vol 1 (ed by F Snyder), pp 151^187 Plenum Press, NewYork, USA.

Ome-Kattner G (1989) Lipid composition of Calanus ¢nmarchicus from the North sea and the Arctic ^ a comparative study Comparative Biochemistry and Physiology B 94, 185^188.

Kattner G & Krause M (1987) Changes in lipids during the development of Calanus ¢nmarchicus S1 from copepodid I

to adult Marine Biology 96, 511^518.

Leblond J.D & Chapman P.J (2000) Lipid class distribution of highly unsaturated long chain fatty acids in marine dino-

£agellates Journal of Phycology 36, 1103^1108.

Lee R.F & Hirota J (1973) Wax esters in tropical zooplankton and nekton and geographical distribution of wax esters in marine copepods Limnology and Oceanography 18, 227^ 239.

Lee R.F & Puppione D.L (1972) Serum lipoproteins of Paci¢c sardine (Sardinops caerulea Girard) Biochimica et Biophy- sica Acta 270, 272–278.

Lee R.F., Barnett A.M & Hirota J (1971) Distribution and portance of wax esters in marine copepods and other zoo- plankton Deep-Sea Research 18, 1147^1165.

im-Lee R.F., Nevenzel J.C & Pa¡enh˛fer G.A (1971) Importance

of wax esters and other lipids in marine food chain ^ toplankton and copepods Marine Biology 9, 99^108 Lee R.F., Nevenzel J.C & Pa¡enh˛fer G.A (1972) Presence of wax esters in marine planktonic copepods Naturwis- senschaften 59, 406^411.

phy-Lee R.F., Hagen W & Kattner G (2006) Lipid storage in ine zooplankton Marine Ecology-Progress Series 307, 273^306.

mar-Levings C.D., Hvidsten N.A & Johnsen B.O (1994) Feeding

of Atlantic salmon (Salmo salar L.) postsmolts in a fjord

in central Norway Canadian Journal of Zoology 72, 834^839.

Lie  & Lambertsen G (1985) Digestive lipolytic enzymes in cod (Gadus morhua) ^ fatty acid speci¢city Comparative Biochemistry and Physiology B 80, 447^450.

Lie  & Lambertsen G (1991) Lipid digestion and absorption

in cod (Gadus morhua), comparing triacylglycerols, wax esters and diacylalkylglycerols Comparative Biochemistry and PhysiologyA 98, 159^163.

Lie ., Lied E & Lambertsen G (1988) Feed optimization in

Atlantic cod (Gadus morhua) ^ fat versus protein-content

in the feed Aquaculture 69, 333^341.

Litch¢eld C., Greenberg A.J., Eaton C.A & Ackman R.G.

(1978) Distinctive medium chain wax esters, triglycerides,

and diacyl glyceryl ethers in the head fats of the Paci¢c

beaked whale, Berardius bairdi Lipids 13, 860^866.

Lynch D.R., Lewis C.V.W & Werner F.E (2001) Can Georges

Bank larval 1 cod survive on a calanoid diet? Deep-Sea

Re-search 48, 609^630.

Madden H., Beare D., Heath M.R., Fraser J.G & Gallego A.

(1999) The sping/early summer distribution of Calanus

spp in the northern North Sea and adjacent areas

Fish-eries Oceanography 8, 138^152.

Mankura M., Kayama M & Saito S (1984) Wax ester

hydro-lysis by lipolytic enzymes in pyloric ceca of various ¢shes.

Bulletin of the Japanese Society of Scienti¢c Fisheries 50,

2127^2131.

Marshall S.M & Orr A.P (1955) On the biology of Calanus

¢n-marchicus 8 Food uptake, assimilation and excretion in

adult and stage-V Calanus Journal of the Marine Biological

Association of the United Kingdom 34, 495^529.

Medale F., Blanc D & Kaushik S.J (1991) Studies on the

nutri-tion of siberian sturgeon, Acipenser baeri 2 Utilizanutri-tion of

dietary nonprotein energy by sturgeon Aquaculture 93,

143^154.

Melle W., Ellertsen B & Skjoldal H.R (2004) Zooplankton:

the link to higher trophic levels In: The Norwegian Sea

Ecosystem (ed by H.R Skjoldal), pp 137–202 Tapir

Academic Press, Trondheim, Norway.

Mo¡at C.F & McGill A.S (1993) Variability of the

composi-tion of ¢sh oils: signi¢cance for the diet Proceedings of

the Nutrition Society 52, 441^456.

Naylor R.L., Goldburg R.J., Primavera J.H., Kautsky N.,

Bev-eridge M.C.M., Clay J., Folke C., Lubchenco J., Mooney H &

Troell M (2000) E¡ect of aquaculture on world ¢sh

sup-plies Nature 405, 1017^1024.

Nevenzel J.C (1970) Occurrence, function and

bio-synthesis of wax esters in marine organisms Lipids 5,

308^319.

Olsen R.E & Ring E (1998) The in£uence of temperature

on the apparent nutrient and fatty acid digestibility of

Arctic charr, Salvelinus alpinus L Aquaculture Research

29, 695^701.

Olsen R.E & Ring E (1997) Lipid digestibility in ¢sh: a

re-view Recent Results Development Lipids Research 1, 199^

265.

Olsen R.E., Henderson R.J & Pedersen T (1991) The

In£u-ence of dietary lipid classes on the fatty acid composition

of small cod Gadus morhua L juveniles reared in an

enclo-sure in northern Norway Journal of Experimental Marine

Biology and Ecology 148, 59^76.

Olsen R.E., Henderson R.J & Ring E (1998) The digestion

and selective absorption of dietary fatty acids in

Artic charr Salvelinus alpinus Aquaculture Nutrition 4,

13^21.

Olsen R.E., Henderson R.J., Sountama J., Hemre G., Ring E., Melle W & Tocher D.R (2004) Atlantic salmon, Salmo salar, utilizes wax ester-rich oil from Calanus ¢nmarchicus e¡ectively Aquaculture 240, 433^449.

Olsen R.E., Suontama J., Langmyhr E., Mundheim H., Ring E., Melle W., Malde M.K & Hemre G.I (2006) The replace- ment of ¢sh meal with Antarctic krill, (Euphausia superba)

in diets for Atlantic salmon, (Salmo salar) Aquaculture trition 12, 280^290.

Nu-Oxley A., Tocher D.R., Torstensen B.E & Olsen R.E (2005) Fatty acid utilisation and metabolism in caecal entero- cytes of rainbow trout (Oncorhynchus mykiss) fed dietary

¢sh or copepod oil Biochimica et Biophysica Acta 1737, 119^129.

Oxley A., Bogevik A.S., Henderson R.J., Waagb R., Tocher D.R & Olsen R.E (2009) Digestibility of Calanus ¢nmarch- icus wax esters in Atlantic salmon (Salmo salar) fresh- water pre-smolts and seawater post-smolts maintained

at constant water temperature Aquaculture Nutrition 15, 459^469.

Patton J.S & Benson A.A (1975) Comparative study of wax ester digestion in ¢sh Comparative BiochemistryAnd Phy- siology B 52, 111^116.

Patton J.S., Nevenzel J.C & Benson A.A (1975) Speci¢city of digestive lipases in hydrolysis of wax esters and triglycer- ides studied in anchovy and other selected ¢sh Lipids 10, 575^583.

Patton J.S., Warner T.G & Benson A.A (1977) Partial terization of the bile salt dependent triglycerol lipase from the leopard shark pancres Biochimica Biophysica Acta

carac-486, 322^330.

Perez J.A., Rodriguez C & Henderson R.J (1999) The uptake and esteri¢cation of radiolabelled fatty acids by entero- cytes isolated from rainbow trout (Oncorhynchus mykiss) Fish Physiology and Biochemistry 20, 125^134.

Phleger C.F (1998) Buoyancy in marine ¢shes: Direct and indirect role of lipids American Zoologist 38, 321^330.

Phleger C.F., Nichols P.D & Virtue P (1997) The lipid, fatty acid and fatty alcohol composition of the myctophid ¢sh Electrona antarctica: high level of wax esters and food chain implications Antarctic Science 9, 258^265 Pickova J & Mrkre T (2007) Alternate oils in ¢sh feeds European Journal of Lipid Science and Technology 109, 256^ 263.

Place A.R (1992) Comparative aspects of lipid digestion and absorption - physiological correlates of wax ester diges- tion American Journal of Physiology 263, 464^471 Porter C.J.H., Trevaskis N.L & Charman W.N (2007) Lipids and lipid-1 based formulations: optimizing the oral deliv- ery of lipophilic drugs Nature Reviews Drug Discovery 6, 231^248.

Prahl F.G., Eglinton G., Corner E.D.S & Ohara S.C.M (1985) Fecal lipids released by ¢sh feeding on zooplankton Journal of the Marine Biological Association of the United Kingdom 65, 547^560.

Ratnayake W.N & Ackman R.G (1979) Fatty alcohols in

ca-pelin, herring and mackerel oils and muscle lipids 1 Fatty

alcohol details linking dietary copepod fat with certain

¢sh depot fats Lipids 14,795^803.

Rigler M.W., Honkanen R.E & Patton J.S (1986)

Visualiza-tion by freeze-fracture, In vitro and In vivo, of the products

of fat digestion Journal of Lipid Research 27, 836^857.

Rikardsen A.H., Haugland M., Bjorn P.A., Finstad B.,

Knud-sen R., Dempson J.B., Holst J.C., Hvidsten N.A & Holm M.

(2004) Geographical di¡erences in marine feeding of

Atlantic salmon post-smolts in Norwegian fjords Journal

of Fish Biology 64, 1655^1679.

Runge J.A & de LafontaineY (1996) Characterization of the

pelagic ecosystem in surface waters of the northern Gulf

of St Lawrence in early summer: The larval red¢sh

Cala-nus microplankton interaction Fisheries Oceanography 5,

21^37.

Ruxton C.H.S., Reed S.C., Simpson M.J.A & Millington K.J.

(2004) The health bene¢ts of omega-3 polyunsaturated

fatty acids: a review of the evidence Journal of Human

Nu-trition and Dietetics 17, 449^459.

Sakshaug E., Bjorge A., Gulliksen B., Loeng H & Mehlum F.

(1994) Structure, biomass distribution, and energetics of

the pelagic ecosystem in the Barents Sea ^ a synopsis

Po-lar Biology 14, 405^411.

Sand D.M., Rahn C & Schlenk H (1973) Wax esters in ¢sh ^

metabolism of dietary wax esters in gourami Journal of

the American Oil Chemists Society 50, 302^302.

Sargent J.R & Henderson R.J (1986) Lipids In: The Biological

Chemistry of Marine Copepods (ed by E.D.S Corner &

S.C.M OŁHara), pp 59^164 Clarendon Press, Oxford, UK.

Sargent J.R & Tacon A.G.J (1999) Development of farmed

¢sh: a nutritionally necessary alternative to meat

Pro-ceedings of the Nutrition Society 58, 377^383.

Sargent J.R., Gatten R.R & McIntosh R (1974) Biosynthesis

of wax esters in cell-free preparations of Euchaeta

norvegi-ca Comparative Biochemistry and Physiology B 47, 217^

227.

Sargent J.R., Lee R.F & Nevenzel J.C (1976) Marine wax

es-ters In: Chemistry and Biochemistry of Nnatural Waxes

(ed by P Kolattukudy), pp 51^91 Elsvier, Amsterdam,

the Netherlands.

Sargent J.R., Gatten R.R., Corner E.D.S & Kilvington C.C.

(1977) Nutrition and metabolism of zooplankton 11

Li-pids in Calanus helgolandicus grazing Biddulphia sinensis.

Journal of the Marine Biological Association of the United

Kingdom 57, 525^533.

Sargent J.R., McIntosh R., Bauermeister A & Blaxter J.H.S.

(1979) Assimilation of the wax esters of marine

zooplank-ton by herring (Clupea harengus) and rainbow trout (Salmo

gairdnerii) Marine Biology 51, 203^207.

Sargent J.R., Gatten R.R & Henderson R.J (1981) Marine wax

esters Pure and Applied Chemistry 53, 867^871.

Sars G.O (1903) Copepoda, Calanoida In: An account of the

Crustacea of Norway with short descriptions and ¢gures of all

the species, Vol 4 (Part I & II) pp 8–13.

Scott C.L., Kwasniewski S., Falk-Petersen S & Sargent J.R (2000) Lipids and life strategies of Calanus ¢nmarchicus, Calanus glacialis and Calanus hyperboreus in late autumn, Kongsfjorden, Svalbard Polar Biology 23, 510^516 Scott C.L., Kwasniewski S., Falk-Petersen S & Sargent J.R (2002) Species di¡erences, origins and functions of fatty alcohols and fatty acids in the wax esters and phospholi- pids of Calanus hyperboreus, Calanus glacialis and Calanus

¢nmarchicus from Arctic waters Marine Ecology-Progress Series 235, 127^134.

Sheridan M.A (1989) Alterations in lipid metabolism companying smolti¢cation and seawater adaptation of salmonid ¢sh Aquaculture 82, 191^203.

ac-Sheridan M.A (1994) Regulation of lipid metabolism in kilothermic vertebrates Comparative Biochemistry and Physiology B 107, 495^508.

poi-Sigurgisladottir S., Lall S.P., Parrish C.C & Ackman R.G (1992) Cholestane as a digestibility marker in the absorp- tion of polyunsaturated fatty acid ethyl esters in Atlantic salmon Lipids 27, 418^424.

Spark A.A (1982) New light on marine waxes, fats and oils South African Journal of Science 78, 303^305.

Speirs D.C., Gurney W.S.C., Heath M.R & Wood S.N (2005) Modelling the basin-scale demography of Calanus ¢n- marchicus in the north-east Atlantic Fisheries Oceanogra- phy 14, 333^358.

Storebakken T (1988) Krill as a potential feed source for monids Aquaculture 70, 193^205.

sal-Stubhaug I.,Tocher D.R., Bell J.G., Dick J.R & Torstensen B.E (2005) Fatty acid metabolism in Atlantic salmon (Salmo salar L.) hepatocytes and in£uence of dietary vegetable oil Biochimica et Biophysica Acta 1734, 277^ 288.

Sund H & Theorell H (1963) Alcohol dehydrogenase In:The Enzymes,Vol 7 (ed by P.D Boyer, H Lardy & K Myrbck),

pp 25^83 Academic Press, NewYork, USA.

Suontama J (2006) Macrozooplankton as feed source for farmed ¢sh ^ growth, product quality and safety PhD thesis, University of Bergen, Norway.

Suontama J., Karlsen ., Moren M., Hemre G.I., Melle W., Langmyhr E., Mundheim H., Ring E & Olsen R.E (2007) Growth, feed conversion and chemical composi- tion of Atlantic salmon (Salmo salar L.) and Atlantic hali- but (Hippoglossus hippoglossus 1 L.) fed diets supplemented with krill or amphipods Aquaculture Nutrition 13, 241^ 255.

Sther O & Mohr V (1987) Chemical composition of north Atlantic krill Comparative Biochemistry and Physiology B

88, 157^164.

Takeuchi T.,Watanabe T & Ogino C (1978) Studies on tive-value of dietary lipids in ¢sh 12 Optimum ratio of protein to lipid in diets of rainbow trout Bulletin of the Ja- panese Society of Scienti¢c Fisheries 44, 683^688 Tocher D.R (2003) Metabolism and functions of lipids and fatty acids in teleost ¢sh Reviews in Fisheries Science 11, 107^184.

nutri-Tocher D.R & Sargent J.R (1984) Studies on triacylglycerol,

wax ester and sterol ester hydrolases in intestinal ceca of

rainbow trout (Salmo gairdneri) fed diets rich in

triacylgly-cerols and wax esters Comparative Biochemistry and

Phy-siology B 77, 561^571.

Torstensen B.E., Fryland L., rnsrud R & Lie  (2004)

Tai-loring of a cardioprotective muscle fatty acid composition

of Atlantic salmon (Salmo salar) fed vegetable oils Food

Chemistry 87, 567^580.

Torstensen B.E., Fryland L & Lie  (2004) Replacing

diet-ary ¢sh oil with increasing levels of rapeseed oil and olive

oil - e¡ects on Atlantic salmon (Salmo salar L.) tissue and lipoprotein lipid composition and lipogenic enzyme activ- ities Aquaculture Nutrition 10, 175^192.

Tso P., Nauli A & Lo C.M (2004) Enterocyte fatty acid uptake and intestinal fatty acid binding protein Biochemical So- cietyTransactions 32,75^78.

Tuchweber B.,Yousef I.M., Ferland G & Perea A (1996) tion and bile formation Nutrition Research 16, 1041^1080 Windell J.T & Norris D.O (1969) Gastric digestion and eva- cuation in rainbow trout Progressive Fish-Culturist 31, 20^26.

Nutri-Effect of photoperiod and feeding frequency on growth and feed utilization of fingerlings Persian sturgeon

Mehdi Zolfaghari, Mohammad Reza Imanpour & Esfandyar Naja¢

Fishery Faculty, Gorgan University of Agricultural Sciences and Natural Resources, Gorgan, Iran

Correspondence: M Zolfaghari, Fishery Faculty, Gorgan University of Agricultural Sciences and Natural Resources, Beheshti St., Gorgan, Iran E-mail: zolfaghari.mz@gmail.com

Abstract

The e¡ects of photoperiod (12L:12D and 18L:6D) and

feeding frequency (three, four and ¢ve evenly spaced

daily feedings) of 10% biomass per day during light

hours on growth and stress response of Persian

stur-geon (Acipenser persicus) ¢ngerlings were evaluated

The interaction between photoperiod and feeding

frequency was not signi¢cant (P40.05) Faster

growth was observed in ¢sh exposed to an 18L:6D

photoperiod (Po0.01) The feed conversion e⁄ciency

(FCE) was also better with an 18L:6D photoperiod

(Po0.05) Fish fed four and ¢ve meals per day grew

similarly (P40.05) and faster than when fed only

three meals (Po0.01) The FCE with four or ¢ve meals

per day was better than with three meals (Po0.05)

At harvest, the proximate composition was similar

in all treatments (P40.05) Stress indicators (cortisol,

glucose and haematocrit) did not di¡er between

photoperiods (Po0.05) The results showed that 0.9^

8.0 g Persian sturgeon ¢ngerlings should be reared

with an 18L:6D photoperiod and fed four times per

day to obtain good growth and FCE

Keywords: Acipenser persicus, feeding frequency,

growth, photoperiod

Introduction

The Persian sturgeon (Acipenser persicus) is an

ana-dromous species belonging to the Acipenseridae

family and is widely distributed along the Caspian

Sea coast of Iran (Birstein, Hanner & DeSalle 1997)

It is a critically endangered species due to over¢shing

and pollution (Jackson, Hurvitz, Din, Goldberg,

Pearl-son, Degani & Levavi-Sivan 2006) Arti¢cial tion and culture of Persian sturgeon could enlargethe supply and help to reduce ¢shing pressure

propaga-A long photoperiod has been applied successfully

to improve the growth of numerous ¢sh species was & Takeuchi 2003; Biswas, Seoka, Tanaka, Takii &Kumai 2006; Tucker, Booth, Allan, Booth & Fielder2006; Ruchin 2007; Biswas, Seoka, Ueno, Yong,Biswas, Kim,Takii & Kumai 2008; Askarian & Kuosha2009), including for larval stages (Fielder & Barsdley2002) However, varying photoperiod lengths showed

(Bis-no e¡ect on the growth of beluga ¢ngerlings barsa, Falahatkar & Banan 2009) Purchase, Boyceand Brown (2000) observed a decreased growth inyellowtail £ounder with the increasing photoperiod.Optimal feeding strategies can enhance growth andfeed conversion e⁄ciency (FCE) (Dwyer, Brown, Par-rish & Lall 2002) Feeding frequency during the dayshould target optimal feed intake and growth (Dwyer

(Bani,Ta-et al 2002; Mohseni, Pourkazemi, Bahmani, kar, Pourali & Salehpour 2008) The feeding frequencymaximizing ¢sh growth depends on various factorsincluding species and life stage (Wang, Kong, Li & Bu-reau 2007) In general, available information relates towhite sturgeon (Acipenser transmontanus) (Hung, Her-old, Gawlicka & Noue 1998; Deng, Koshio, Yokyama,Shao & Hung 2003) and Siberian sturgeon (Koksal,Rad & Kindir 2000), not to Persian sturgeon.Stress negatively in£uences growth and diseaseresistance (Askarian & Kuosha 2009) Biswas et al.(2006) demonstrated that photoperiod manipulationdid not cause signi¢cant acute or chronic stressresponse in red sea bream (Pagrus major) reared from

Falahat-20 to 100 g There is no information regarding thee¡ect of photoperiod manipulation on the stress

response in Persian sturgeon ¢ngerlings Therefore,

the aim of this study was to investigate the e¡ects of

photoperiod and feeding frequency on growth and

feed utilization of Persian sturgeon ¢ngerlings

Material and methods

Experimental design

The study was conducted at the aquaculture research

centre, Gorgan University of Agricultural Sciences

and Natural Resources, Iran A 2 3 factorial

de-sign, including the factors photoperiod (12 light:12

dark or 18 light:6 dark; hereafter noted as 12L:12D or

18L:6D) and feeding frequency (three, four and ¢ve

times daily feed and abbreviated as 3F, 4F or 5F,

re-spectively) was followed The feed was applied in

feed-ing intervals evenly spaced over the daylight period

(Table 1) Each treatment was randomly assigned to

three replicate aquaria, and the feeding trial lasted

36 days

Stocking and weight check

Persian sturgeon larvae with an average weight of

150 mg were obtained from the Shahid Marjani

Stur-geon Fishes Culture and Propagation Centre, Gorgan,

Iran, and transported carefully to the rearing

labora-tory They were kept in 400 L tanks and weaned to

arti¢cial feed for 2 weeks under natural light All

lar-vae were fed an arti¢cial diet ¢rst (Bio-Optimal Start,

Biomar Company, Bordeaux, French) followed by live

daphnia half an hour later (Shahid Marjani Sturgeon

Fishes Culture and Propagation Centre) At the end of

the weaning period, nearly half of the larvae were

randomly distribution over eighteen 65 L aquaria,

stocking 11 ¢sh per aquarium The larvae were

al-lowed to adapt to the new aquaria for 2 weeks and

fed only Bio-Optimal The individual weight(mean SD) of Persian sturgeon was 0.9  0.0 g atthe start of the experiment and equal between treat-ments (P40.05) The daily feeding ration was ad-justed after weighing all ¢sh per tank 10, 20 and 27days after starting the experiment

Feeding protocols and used facilitiesFish were fed with a commercial diet (Bio-OptimalStart, Biomar Company) The diet contains 56% pro-tein, 18% lipid and 10.5% ash Fish in all aquaria werefed at 10% body weight per day (in all feeding frequen-cies) Approximately 60% of the water was exchangedeach day and faeces were removed daily For light con-trol, a programmed time controller (Model BND-50/D2,Ningbo Bainian Electric Appliance, Ningbo, China) wasused to control the periods of light and dark The nineaquaria of each light regime were illuminated by

40 W lamps suspended above the water surface andthe light intensity was adjusted at 300 lx at the watersurface during light hours and 0 lx during dark peri-ods Each set of nine aquaria applying the samephotoperiod was isolated from the other set and en-vironment light by a thick black plastic cover Visiblelight intensity in each aquarium was measured using

a luxmeter (Lutron Lx-101, Chu City, Taiwan) Waterphysicochemical characteristics were monitored dai-

ly (before the ¢rst meal) and maintained (Model ibaU-10, Kyoto, Japan): temperature 19.2 0.9 1C;dissolved oxygen 7.5 0.1mg L 1; pH, 7.7 0.1and salinity 0.1ppt Fish were deprived of food for

Hor-24 h before weighing and anaesthetized with

120 ppm clove oil At the end of the experiment, afterweighing, ¢ve ¢sh were sampled randomly from eachaquarium and frozen at  20 1C for whole-bodyproximate composition analysis

Calculation of growth performanceGrowth performance [weight gain (WG), speci¢cgrowth rate (SGR), body weight increase (BWI), dailygrowth increase (DGI), FCE and protein retention ef-

¢ciency (PRE)] was calculated using the followingformulae:

Table 1 Feeding protocols of Persian sturgeon ¢ngerlings

reared under a 12L:12D or 18L:6D photoperiod

Feeding time (h:min)

where W1 and W2 are the average initial and ¢nal

individual weight (g), respectively, in each aquarium

FCEð%Þ ¼ 100½wet weight gain ðgÞ=

dry feed intakeðgÞ

PREð%Þ ¼ 100½(final whole body protein

initial whole body proteinÞ=

total protein intake

Blood sampling procedure

At the end of the experiment, approximately 1mL of

blood per ¢sh from ¢ve ¢sh per aquarium was

col-lected from the caudal vein using a heparinized

syr-inge The blood was put in a 1mL tube and kept on ice

until centrifugation Anaesthetizing (120 ppm clove

essence), measuring and blood withdrawal took

o3 min per aquarium

Haematocrit

Haematocrit values were determined immediately

after sampling by placing fresh blood in glass

capil-lary tubes and centrifuging for 5 min at 12 000 rpm

in a haematocrit centrifuge The remaining blood

samples were centrifuged at 3000 rpm for 15 min

and the plasma stored at 20 1C until analysis

(Bis-was et al 2006)

Cortisol

Plasma cortisol levels were determined by ELISA

ac-cording to the manufacturer’s instructions (IBL

International, Hamburg, Germany) and following

the method proposed by Biswas et al (2008) Cortisol

was extracted from100mL of plasma with ethyl ether

The organic phase was separated into a clean glass

tube and the solvent was evaporated with a stream

of N2 The residue was then dissolved in 100mL of

di-luted extraction bu¡er Ten microlitres of this

solu-tion was put into 990mL of diluted extraction bu¡er

Fifty microlitres of the sample or standard solution

was ¢rst added to the microplate in duplicates and

in-cubated at room temperature for 1h The plate was

then washed and quantitative test results were

ob-tained by measuring and comparing the absorbance

reading of the wells of the samples against the

stan-dards using a microplate reader at 650 nm

GlucosePlasma glucose levels were measured by using an ana-lytical kit (ParsAzZmon,Tehran, Iran) In brief, 20mL

of plasma or standard solution was added to 3 mL ofbu¡er solution and shaken Glucose levels were thenanalysed by spectrophotometer after incubation for

5 min at 37 1C in a water bath (Biswas et al 2006)

Body proximate compositionWater content was measured by drying at105 1C untilachieving a constant weight; ash content was deter-mined after burning until constant weight at 550 1C;crude protein (N 6.25) content was estimated bythe Kjeldhal method; and lipids by diethyl etherextraction by the Soxhlet method (Association ofO⁄cial Analytical Chemists 2005)

Statistical analysesData were expressed as the mean SD Normalitywas checked through the Kolmogorov^Smirnov testand Levene’s test was used to establish the homoge-neity of variance Results were analysed by two-wayANOVA, followed by Duncan’s Multiple RangeTest (SPSS,version 12, SPSS, Chicago, IL, USA)

ResultsTable 2 shows that the growth of the ¢sh was signi¢-cantly in£uenced by photoperiod After 36 days, ¢shexposed to 18L:6D grew faster (indicated byWG, BWI,

Table 2 Performance of ¢ngerling Persian sturgeon gerling exposed to di¡erent photoperiods for 36 days

Initial body weight (g) 0.92  0.02 a 0.92  0.02 a

Final body weight (g) 7.1  0.3 b 7.5  0.5 a

dif-Each value is a mean  SD.

BWI, body weight increase; SGR, speci¢c growth rate; DGI, daily growth increase; PRE, protein retention e⁄ciency; FCE, feed conversion e⁄ciency.

SGR and DGI) than ¢sh exposed to 12L:12D (Po0.01)

but PRE was not signi¢cantly di¡erent between

photoperiods (P40.05) Importantly, the photoperiod

in£uenced FCE A photoperiod of 18L:6D resulted in a

signi¢cantly better FCE than the 12L:12D

photoper-iod (Po0.05, Table 2) The moisture, fat, protein and

ash content were similar between photoperiods

(P40.05,Table 3)

Increasing the daily feeding frequency from three

to four times per day resulted in better growth (WG,

BWI, SGR and DGI) (Po0.01), but there was no

signif-icant di¡erence in growth feeding four and ¢ve times

daily (Table 4) The results showed that feeding

fre-quency signi¢cantly a¡ected the FCE of Persian

stur-geon ¢ngerlings (Po0.05,Table 4) The ¢sh proximate

composition was not in£uenced by feeding frequency

(P40.05) There were no signi¢cant di¡erences for

PRE between photoperiods and feeding frequencies

The stress indices showed that there were no

sig-ni¢cant di¡erences in haematocrit levels and cortisol

and glucose levels of plasma between photoperiods

(Table 5)

Discussion

Longer photoperiods have been reported to stimulate

growth in a number of ¢sh species (Boeuf & Le Bail

1999; Randall, North, Futter, Porter & Bromage

2001), including beluga sturgeon (Askarian &

Kuosha 2009) Ruchin (2007) demonstrated that

Siberian sturgeon grew faster under longer

photo-periods This is due to several factors Tucker et al

(2006) reported that newly weaned regularly fedsnapper kept under a long photoperiod adopted

an endogenous rhythm that allowed snapper tosynchronize with metabolic demands and utilizenutrients more e⁄ciently Lengthening of the photo-period might indirectly modify growth by increasingmuscle mass through exercise (Boeuf & Le Bail 1999;Tucker et al 2006) In this study, the FCE was betterwith the longer photoperiod, which concurs withreports of Biswas et al (2006) for juvenile red seabream and Biswas et al (2008) for striped knifejaw(Oplegnathus fasciatus)

Table 3 Proximate compositions (based on wet weight) of Persian sturgeon ¢ngerling reared under di¡erent photoperiods and feeding frequencies

Each value is a mean  SD derived from three replicates and each of replicates includes ¢ve ¢sh (n 5 5).

No signi¢cant interactions occurred between photoperiod and feeding frequency for any of the parameters measured.

0.92  0.02 a 0.92  0.02 a 0.92  0.01 a

Final body weight (g)

Each value is a mean  S.D.

BWI, body weight increase; SGR, speci¢c growth rate; DGI, daily growth increase; PRE, protein retention e⁄ciency; FCE, feed conversion e⁄ciency.

Boeuf and Le Bail (1999) speci¢ed that a longer

photoperiod improves growth not only through

bet-ter FCE but also food intake Biswas et al (2006)

emphasized that the better growth of newly weaned

snapper kept under a long and continuous

photoper-iod was the result of a higher food intake and FCE The

longer time interval between feeding times in ¢sh

ex-posed to18L:6D photoperiod might provide more time

to digest resulting in a better retention e⁄ciency

The results of growth in the present study con¢rm

the results obtained by Mohseni et al (2008) for

belu-ga, who reported a better feed conversion ratio in

higher feeding frequencies (Carmona, Domezian,

Garsia-Gallego, Hernando, Rodriguez & Ruiz-Rejon

2009) advised a feeding frequency for juvenile

stur-geons of minimum six meals per day Feeding

frequencies vary with ¢sh species size and age

Gen-erally, fry are fed small meals with a high frequency

while one feeding per day is usually su⁄cient for

brood stock and older ¢sh Considering the slow

feed-ing habit of sturgeon, continuous feedfeed-ing might be

the most e¡ective feeding regime (Carmona et al

2009) One possible reason for the improved FCE with

the increased feeding frequency could be that ¢sh

maintain nutrient stores and so can use more of their

nutrient and energy intake for somatic growth

(Tuck-er et al 2006) Similar results w(Tuck-ere found for gilthead

sea bream (Andrew, Holm, Kadri & Huntingford

2004) Feeding four and ¢ve meals per day (F4 andF5) probably also resulted in an increased absorption

of food nutrients, as ¢sh have access to nutrients moreoften during the day (Cho, Lim, Lee & Park 2003;Tucker et al 2006; Silva, Gomes & Brandao 2007).Our results demonstrated that photoperiod manip-ulation did not cause a signi¢cant chronic stress re-sponse in Persian sturgeon ¢ngerlings, which is inagreement with observations on other sturgeon spe-cies Askarian and Kuosha (2009) reported that ma-nipulation of the photoperiod does not have asigni¢cant e¡ect on chronic stress indices (cortisol,glucose and haematocrit) in beluga sturgeon Haema-tocrit has been shown to increase under stressful con-ditions (Trenzado, Morales & de la Higuera 2006) andmay be attributed to red blood cell recruitment fromthe spleen (Wojtaszek, Dziewulska-Szwajkowska, Lo-zin ska-Gabska, Adamowicz & Dzugaj 2002) or to theswelling of red blood cells However, haematocrit hasalso been shown to decrease under chronic stress(Biswas et al 2006) There was no signi¢cant increase

or decrease in haematocrit level among the di¡erentphotoperiods in this study, which concurs with theobservation for beluga sturgeon (Bani et al 2009).Therefore, photoperiod manipulation did not appear

to change the chronic stress level in Persian geon, and concurred with better growth and FCE

stur-Conclusions

In conclusion, the results suggest that the growthperformance of A persicus reared from 0.9 to 14 g isimproved by increasing the photoperiod from 12 to

18 h/day without increasing stress A better FCE wasobserved with the 18-h photoperiod Also, feedingfour or ¢ve times daily was better than three mealsper day Such a rearing strategy will be helpful to pro-duce ¢ngerling to restock the natural habitat of Per-sian sturgeon Further studies are necessary toevaluate the e¡ect of photoperiod on growth of largersize classes and on resistance to acute stress of Per-sian sturgeon

AcknowledgmentsThe authors are grateful to Montazeri for reading themanuscript This research has been funded by Gor-gan University of Agricultural Sciences and NaturalResources

Table 5 Stress indicators of ¢ngerling Persian sturgeon

¢ngerling reared under di¡erent photoperiods and feeding

Haematocrit (%)

Each value is a mean  SD derived from three replicates and

each of the replicates includes ¢ve ¢sh (n 5 5 for these analyses).

No signi¢cant interactions occurred between photoperiod and

feeding frequency for any of the parameters measured.

NS, no signi¢cant di¡erence.

Andrew J.E., Holm J., Kadri S & Huntingford F.A (2004) The

e¡ect of competition on the feeding e⁄ciency and feed

be-havior in gilthead sea bream (Sparus aurata L.) held in

tanks Aquaculture 232, 317^331.

Askarian F & Kuosha A (2009) The in£uence of

photoper-iod in farming Beluga sturgeon (Huso huso): evaluation by

growth and health parameters in serum Journal of

Fish-eries and Aquatic Science 4, 41^49.

Association of O⁄cial Analytical Chemists (2005) O⁄cial

Methods of Analysis Association of O⁄cial Analytical

Chemists, Gaithersburg, MD, USA.

Bani A.,Tabarsa M., Falahatkar B & Banan A (2009) E¡ects

of di¡erent photoperiods on growth, stress and

haemato-logical parameters in juvenile great sturgeon Huso huso.

Aquaculture Research 40, 1899^1907.

Birstein V.J., Hanner R & DeSalle R (1997) Phylogeny of the

Acipenseriformes: cytogenetic and molecular

ap-proaches Environmental Biology of Fishes 48, 127^155.

Biswas A.K., Seoka M.,TanakaY.,Takii K & Kumai H (2006)

E¡ect of photoperiod manipulation on the growth

perfor-mance and stress response of juvenile red sea bream

(Pa-grus major) Aquaculture 258, 350^356.

Biswas A.K., Seoka M., Ueno K., Yong A.S.K., Biswas B.K.,

KimY.S.,Takii K & Kumai H (2008) Growth performance

and physiological responses in striped knifejaw

(Opleg-nathus fasciatus) held under di¡erent photoperiods

Aqua-culture 279, 42^46.

Biswas A.K & Takeuchi T (2003) E¡ects of photoperiod and

feeding interval on food intake and growth rate of Nile

ti-lapia (Oreochromis niloticus) Fish Science 69, 1010^1016.

Boeuf G & Le Bail P (1999) Does light have an in£uence on

¢sh growth? Aquaculture 177, 129^152.

Carmona R., Domezian A., Garsia-Gallego M., Hernando

J.A., Rodriguez F & Ruiz-Rejon M (2009) Biology,

Conser-vation and Development of Sturgeons Springer Press,

Ber-lin, Germany, 467pp.

Cho S.H., LimY.S., Lee J.H & Park S (2003) E¡ects of feeding

rate and feeding frequency on, growth, and body

compo-sition of Ayu post-larvae altivelis Plecoglossus altivelis.

Journal of World Aquaculture Society 34, 85^91.

Deng D., Koshio S.,Yokyama B.S.C., Shao Q.C.Y & Hung S.S.O.

(2003) E¡ect of feeding rate on growth performance of

white sturgeon (Acipenser transmontanus) larvae

Aqua-culture 217, 589^598.

Dwyer K.S., Brown J.A., Parrish C & Lall S.P (2002) Feeding

frequency a¡ects food consumption, feeding pattern and

growth of juvenile yellowtail £ounder (Limanda

ferrugi-nea) Aquaculture 213, 279^292.

Fielder D.S & Barsdley W.J (2002) E¡ect of photoperiod on

growth and survival of snapper Pagrus auratus larvae.

Aquaculture 211, 135^150.

Hung S.S.O., Herold M.A., Gawlicka A & Noue J.D.I (1998) E¡ects of dietary lipid on growth and fatty acid composi- tion of white sturgeon (Acipenser transmontanus) larvae, abstract Aquaculture 161, 333^335.

Jackson K., Hurvitz A., Din S.Y., Goldberg D., Pearlson O., Degani G & Levavi-Sivan B (2006) Anatomical, hormo- nal and histological descriptions of captive Russian sturgeon (Acipenser gueldenstaedtii) with inter sex gonads General and Comparative Endocrinology 148, 359^367 Koksal G., Rad F & Kindir M (2000) Growth performance and feed conversion e⁄ciency of Siberian sturgeon juveniles (Acipenser baeri) reared in concrete Raceways Turkish Journal of Veterinary and Animal Sciences 24, 435^442.

Mohseni M., Pourkazemi M., Bahmani M., Falahatkar B., Pourali H.R & Salehpour M (2008) E¡ects of feeding rate and frequency on growth performance of yearling great sturgeon (Huso huso) Journal of Applied Ichthyology 22, 278^283.

Purchase C.F., Boyce D.L & Brown J.A (2000) Growth and survival of yellowtail juvenile £ounder Pleuronectes ferru- gineus (Storer) under di¡erent photoperiods Aquaculture Research 31, 547^552.

Randall C., North B., Futter W., Porter M & Bromage N (2001) Photoperiod e¡ects on reproduction and growth

in rainbow trout Trout News 32, 12^16.

Ruchin A.B (2007) E¡ect of photoperiod on growth, logical and hematological indices of juvenile siberian sturgeon (Acipenser baerii) Izvestia Akademii Nauk 6, 698^704.

physio-Silva C.R., Gomes L.C & Brandao F.R (2007) E¡ect of feeding rate and frequency on tambaqui (Colossoma macropomum) growth, production and feeding costs during the ¢rst growth phase in cages Aquaculture 264, 135^139.

Trenzado C.E., Morales A.M & de la Higuera M (2006) siological e¡ects of crowding in rainbow trout (Oncor- hynchus mykiss) selected for low and high stress responsiveness Aquaculture 258, 583^593.

Phy-Tucker B.J., Booth M.A., Allan G.L., Booth D & Fielder D.S (2006) E¡ects of photoperiod and feeding frequency on performance of newly weaned Australian snapper (Pagrus auratus) Aquaculture 258, 514^520.

Wang Y., Kong L.G., Li K & Bureau D.P (2007) E¡ects of feeding frequency and ration level on growth, feed utiliza- tion and nitrogen waste output of cuneate drum (Nibea miichthioides) reared in net pens Aquaculture 271, 350^356.

Wojtaszek J., Dziewulska-Szwajkowska D., Lozin ska-Gabska M., Adamowicz A & Dzugaj A (2002) Hematological ef- fects of high dose of cortisol on the carp (Cyprinus carpio L.): cortisol e¡ect on the carp blood General and Compara- tive Endocrinology 125, 176^183.

Physico-chemical characteristics of the improved

extensive shrimp farming system in the Mekong Delta

of Vietnam

Nguyen Tho1,Vu Ngoc Ut2& Roel Merckx3

1 Ho Chi Minh City Institute of Resources Geography,Vietnamese Academy of Science and Technology, Ho Chi Minh City, Vietnam

2 Department of Applied Hydrobiology, College of Aquaculture and Fisheries, Can Tho University, Can Tho City,Vietnam

3 Department of Earth and Environmental Sciences, Division Soil and Water Management, Faculty of Bioscience Engineering, Katholieke Universiteit Leuven, Heverlee, Belgium

Correspondence: N Tho, Ho Chi Minh City Institute of Resources Geography, Vietnamese Academy of Science and Technology, 01 Mac Dinh Chi Str., Dist 1, Ho Chi Minh City,Vietnam E-mails: ntho@vast-hcm.ac.vn, nguyentho3011@yahoo.com

Abstract

Consecutive failure of the improved extensive shrimp

farming system has deterred the economy of some

coastal areas in Vietnam To investigate pond

physico-chemical characteristics, a monitoring scheme was

performed in the Cai Nuoc district of Southern

Vietnam Results show that the system was not

opti-mal for shrimps While ponds were not contaminated

by organic loadings or major nutrients (N, P) and

sali-nity and pH were most optimal for shrimp, more than

37% of dissolved oxygen (DO) measurements were

lower than recommended In the early morning

hours, DO measurements were even much lower

(0.84^2.20 mg L 1) Sulphate (SO4) concentrations

were most within the acceptable range Total

sus-pended solids (TSS) were above the acceptable limit

(o50 mg L 1) Iron, alkalinity and hydrogen sulphide

were also higher than recommended Pond sediment

was anaerobic (redox potential  422 to  105 mV)

and contained high amounts of organic matter (9.84^

21.96%) Lethal DO levels, high TSS and anoxic

sedi-ment are the drawbacks in this system Suggested

measures to improve pond conditions are (1) allowing

sedimentation before ¢lling culture ponds, (2)

cover-ing dikes, (3) includcover-ing no-culture breaks between

shrimp crops, (4) drying pond bottom, (5) removing

sediment and (6) controlling pond’s vegetation

Keywords: improved extensive shrimp farming,

Mekong Delta, physico-chemical characteristics,

se-diment characteristics, water quality

IntroductionShrimp farming is one of the fastest growing eco-nomic activities in the Asia^Paci¢c region (Wolanski,Spagnol, Thomas, Moore, Alongi, Trott & Davidson2000; Raux & Bailly 2002; EJF 2004) and experi-enced spectacular growth over recent decades(Thornton, Shanahan & Williams 2003; EJF 2004)

In Vietnam, extensive farming of shrimp startedabout 100 years ago (Nhuong, Luu,Tu,Tam & Nguyet2002) and was characterized by low shrimp yields(100^400 kg ha 1year 1) (Binh & Lin 1995) In theMekong Delta of Vietnam, extensive shrimp farmingsystems started soon after the end of theVietnam war(ADB/MoFi 1996) Modern shrimp farming, however,started only after the 1980s when the countrylaunched the economic reform and the governmentencouraged shrimp farming (Nhuong et al 2002) In

2000, the government released the resolution 09/NQ-CP, which allowed conversion of saline and lowproductivity rice ¢elds into shrimp farms As a result,shrimp farming boomed in the area (Vuong & Lin2001; Nhuong et al 2002; Binh,Vromant, Hung, Hens

& Boon 2005)

The improved extensive shrimp farming system isthe most predominant in the Mekong Delta of Viet-nam (Nhuong et al 2002; Binh et al 2005) Currently,black tiger shrimp (Penaeus monodon) is widelyfarmed in the area Although the criteria for shrimpfarming system classi¢cations may di¡er (R˛nnbck2001; EJF 2003), shrimp farming in the area can becategorized as ‘extensive’, ‘improved extensive’ and

‘intensive’ systems In the ¢rst system, shrimps are

usually cultured in mangrove forests, relying mostly

on natural seedstocks, sometimes with

supplemen-tary stocking (1^1.5 postlarvae m 2) The improved

extensive shrimp farming system can be

character-ized by (1) shrimp reared in a monoculture system in

earthen ponds, (2) stocking of 1^7 ind m 2, (3)

sedi-ment dredging and subsequent pond liming and (4)

low survival rate (3^20%) The intensive shrimp

farming system witnesses the highest level of

intensi-¢cation (mechanical pond preparation, stocking

den-sity of 15^45 postlarvae m 2, use of industrial feed

and chemicals) (Phuong, Minh & Tuan 2004)

Despite great innovations in shrimp farming, farmers

have operated their farms based on their own

experi-ences and the knowledge transferred by their

neigh-bours (Vuong & Lin 2001; Sels 2004) This system

is subject to low and unstable shrimp yields (Hens,

Vromant, Tho & Hung 2009) While other shrimp

farming systems in the Mekong Delta are well

docu-mented (Alongi, Tirendi & Trott 1999; Vuong & Lin

2001; Brennan, Preston, Clayton & Be 2002; Clough,

Johnston, Xuan, Phillips, Pednekar, Thien, Dan &

Thong 2002; Preston & Clayton 2003), little is known

about the improved extensive shrimp farming

system While the physico-chemical condition of the

cultured ponds plays a vital role in the growth of

shrimps (Haws & Boyd 2001; Hena Abu,

Sharifuzza-man, Hishamuddin, Misri & Abdullah 2008), whether

pond water and sediment in this system is optimal for

shrimps remains unanswered An overall assessment

of the ponds’ physico-chemical characteristics is

not available

The survival and growth of black tiger shrimp

are a¡ected by a wide variety of physico-chemical

parameters Salinities between 5 and 35 g L 1are

most suitable for shrimp (Haws & Boyd 2001) and

low salinities are linked to shrimp diseases (Joseph

& Philip 2007) Water temperatures above 32 1C for

prolonged periods can stress shrimp and reduce

growth as it a¡ects shrimp metabolism and feeding

rates (Lazur 2007) Dissolved oxygen (DO) should be

higher than 4 mg L 1(Chien 1992; Lazur 2007) as

lower values can a¡ect shrimp respiration and reduce

shrimp resistance to diseases Total suspended solid

(TSS), which can deter the penetration of solar

radia-tion into the water column and reduce primary

production, is recommended from 2 to 14 mg L 1for

black tiger shrimp (Jayasinghe, Corea &

Wijegunawar-dana 1994) oro50 mg L 1

for aquaculture in general(Vietnamese Standard 5943-1995) Recommended

range for alkalinity, a measure of carbonate bases and

which can a¡ect primary production, is from 50 to

100 mg L 1 Extremely low pH can stress shrimp andcause soft shell and poor survival High shrimpmortality is observed at pH below 6 (Chien 1992).Nitrate (NO3-N) is generally not toxic in pond environ-ment (Lazur 2007) At high pH values, total ammo-nium nitrogen (TAN) can have an impact on shrimp,

as the toxic gas NH3 increases relative to NH4-N(Haws & Boyd 2001) NH3can damage gills and reducegrowth or cause mortality of shrimp, even at low con-centrations (Lazur 2007) Nitrite (NO2-N), an inter-mediate product of nitri¢cation and nitrate reduction,

is less toxic to shrimp (Chien1992) At high tions, nitrite combines with haemocyanin in shrimpblood and reduces the ability of the blood to transportoxygen (Haws & Boyd 2001) Toxicity levels of hydro-gen sulphide (H2S) to shrimp is controversial, rangingfrom non-detectable to o0.25 mg L 1 (Jayasinghe

concentra-et al 1994; Haws & Boyd 2001; Lazur 2007) ical oxygen demand (BOD) ^ an indicator of organicpollution ^ is allowed in shrimp ponds at concentra-tions ofo10 mg L 1

Biochem-(Jayasinghe et al 1994) WhenBOD exceeds 20 mg L 1, oxygen depletion is a danger

in ponds without mechanical aeration (Haws &Boyd 2001) High concentrations of Fe21can causeslow growth of shrimp (Poernomo 1989) Impacts ofsediment to shrimp are mainly stemmed from elevatedorganic matter (OM) accumulated on pond bottom.The degradation of sediment OM causes low DO condi-tion, resulting in the reduced toxic substances, likenitrite, NH3, Fe21, Mn21, H2S and CH4(Haws & Boyd2001), which can a¡ect the growth of cultured shrimp.Links of sediment parameters to shrimp can be found

in Boyd (1995)

This paper investigates the main physico-chemicalcharacteristics of the improved extensive shrimpfarming system in the coastal areas of the MekongDelta (Southern Vietnam) It was conducted in theCai Nuoc district, where shrimp pond areasexpanded rapidly from the year 2000 on (Binh et al.2005) In the district, shrimp expansion, basic pondcharacteristics and pollution caused by shrimp farm-ing have partly been discussed (Binh et al 2005; Tho,Vromant, Hung & Hens 2006, 2008; Hens et al 2009)

At the end of 2008, many improved extensive shrimpfarmers were in debt as a result of shrimp losses (Fielddata 2008) The study is the ¢rst of its kinds combin-ing water and sediment parameters in assessing thephysico-chemical characteristics of the improvedextensive shrimp farming system in the MekongDelta of Vietnam Accordingly, it provides importantinformation to the limited literature in this system

upon which the majority of local people rely for their

livelihood It lays a sound background for further

re-search, aiming at a more sustainable shrimp farming

practice in the area

Materials and methods

Description of the study site and the improved

extensive shrimp ponds

The Cai Nuoc district coastal lowlands (39514 km2) is

situated in the south-west of the Ca Mau province in

the Mekong Delta of Vietnam (Fig 1) This area at 81

north of the equator has two main seasons: (1) a rainy

season (May^October) with an average rainfall of

2100 mm, and (2) a dry season (November^April)

with an average rainfall of 200 mm (SIG 2001) A

dense network of rivers and canals connects the area

from the South China Sea to the Gulf of Thailand The

area experiences a complex tidal regime In 2008, the

improved extensive shrimp farming system

(includ-ing small ponds in garden areas) occupied470% of

the district’s total land area and 88% of all

shrimp-related areas (data from the Agriculture and Rural

Development Division of Cai Nuoc district) The

rice^shrimp farming system (one rice crop in the

rainy season, one shrimp crop in the dry season)

is also practised in a village north of the district

where a salt prevention sluice gate has still been

operated The local government would like to expand

the rice^shrimp area westwards However, not muchsuccess has been achieved due to salinity problem.Farmers construct the improved extensive shrimpponds by digging a deep trench (2^2.5 m width,0.6^0.8 m depth) around the traditional rice ¢elds.The central platform of the ¢eld, which accounts formost of the pond surface area, remains undisturbed(Fig 2) Each pond has a concrete gate near the cor-ner closest to the river The pond is connected to theriver through a settlement channel, which opens tothe river also via a concrete gate Farmers collect riv-

er water into the settlement channel by direct ing or just opening the channel gate at high tides (inareas of higher tidal amplitude) Before entering thesettlement channel, river water has to pass a screen

pump-Figure 1 Location of the Cai Nuoc district and the sampling sites

Figure 2 A typical improved extensive shrimp pond inthe Cai Nuoc district

for the exclusion of undesirable objects Some farmers

store river water for some time in the settlement

chan-nel before ¢lling their ponds Stocking density is low (1^

3 seeds m 2) and most farmers stock supplementary

shrimp larvae (amounting to 50^100% of the original

amount stocked) over the following months No feeding

of the shrimps is applied Pond water is not exchanged

unless farmers observe shrimp mortality or too much

turbidity Sediment is removed from the trench, usually

once or twice a year In some cases, due to ¢nancial

dif-¢culty, farmers leave the pond sediment behind for 2

years or even more Generally, shrimps are cultured all

year round Market-sized shrimps are harvested daily

using a large-sized net placed in the trench, usually

after 5 months of culture Major production inputs are

costs of (1) seed stocks (16 000^35000 VND per 1000

postlarvae), (2) lime (CaO, CaCO3) and fertilizers (NPK,

Urea) and (3) labour, mostly for sediment dredging and

disposal While the ¢rst one can be roughly estimated,

the last two items are hard to quantify as farmers do not

keep records of pond inputs.While in principle

consid-ered as ‘monoculture’, the ponds are usually stocked

with shrimps and several other aquaculture species,

such as ¢sh or crabs, mainly for an increased economic

output (Field data 2008) This system requires little

in-vestment and knowledge, and is widely adopted by the

low- and middle-income households The system was

slightly a¡ected by the 2008 world economic crisis

During ¢eldwork in 2008, most improved extensive

shrimp farmers argued that they could not a¡ord

ap-propriate sediment dredging and disposal due to

in-creased labour cost

Sampling

Pond water and sediment were sampled ¢ve times for

physico-chemical characteristics in seven shrimp

ponds (Fig 1) On a 2-month basis, pond water was

sampled in the trench with the hypothesis that water

quality in the trench and the platform do not di¡er for

most of the parameters Pond sediment was sampled

in both the trench and the platform at 0^5 and 5^

10 cm depths, taking into account that the quality of

the top sediment may di¡er between the locations,

and that these di¡erences can have an impact on

shrimp health Sampling was conducted while

farm-ers followed their own practices of farming

Sampling of pond water

Temperature, pH, DO, turbidity and salinity were

measured in situ using a TOA instrument

(WQC-22A, DKK-TOA Corporation, Tokyo, Japan) Sampling

of H2S was conducted at the water^sediment face using 125 mL dark glass bottles The other para-meters were sampled in mid-depth of the trench atthree locations in the ponds and mixed to form acomposite sample for analysis A 2 L plastic bottlewas ¢lled up for TSS, NO2-N, NO3-N, TAN, total N,

inter-PO4-P, total P, Cl, SO4 

and alkalinity Samples forBOD were collected in 1 L plastic bottles and kept inthe dark Samples for total Fe were collected in

125 mL dark glass bottles and ¢xed by concentratedHNO3, while those for Fe21were collected and stored

in 125 mL dark glass bottles COD samples werecollected in 125 mL bright glass bottles and ¢xed by

H2SO44 M All samples were stored at approximately

4 1C and transported to the lab the same day foranalysis

Sampling of pond sedimentThe pHw(pH of fresh sediment) and Eh (redox poten-tial) were measured in situ using a combined Eh/pHmeter (pH 62K, APEL Co Ltd, Saitama, Japan) A glasselectrode was used for pHwwhile an EMC 130 Meins-berg electrode (Sensortechnik Meinsberg GmbH,Ziegra-Knobelsdorf, Germany) was used for Eh mea-surement Pond sediment was sampled (0^5 and5^10 cm) using a stainless steel hand borer Sampleswere stored in plastic boxes and placed in the ice box.Bulk density (BD) sampling was conducted once bypressing a PVC tube of 6 cm diameter into the sedi-ment to a depth of about 20 cm After the tube waspulled out, sediment column from 1 to 4 cm depth(representative for the 0^5 cm layer) and from 6 to

9 cm depth (representative for the 5^10 cm layer)was cut The ¢nal sample for each depth had a volume

of 84.82 cm3 On a 2-month basis, the parameters Eh,

pHw, pH1/2.5 and EC1/5 were measured ¢ve timeswhile total N, total P and OM were measured fourtimes On a 4-month basis, the parameters NO2-N,

NO3-N and NH4-N were measured three times while

PO4-P, soluble SO4, total SO4, Fe21and Fe31weremeasured twice (Table 1)

Twenty-four-hour monitoring experiment ofselected parameters

To understand daily pond £uctuations, the meters pH, DO, turbidity and temperature were mea-sured in situ over a 24-hour period in three adjacentimproved extensive shrimp ponds These ponds areowned by the same farmer and completely isolated

para-Pond 1 was located in an open area of a former rice

¢eld and had no shade Pond 2 was also constructed

in a former rice ¢eld but had some shade from

coco-nut palms and other trees on the dikes Pond 3 was

located inside a fruit garden and was almost

comple-tely shaded by large trees Aquatic vegetation was

found in all of the ponds Water in these ponds had

been ¢lled in at di¡erent times from the same supply

channel connecting to the main river Water

ex-change had not been applied for at least more than 1

year before the monitoring experiment The stocking

density in these ponds was not recorded by the

own-er, but it was estimated around 1^3 ind m 2(Field

data 2008) The parameters were measured at a

depth of about 20 cm under the water surface in the

trench using a TOA instrument (WQC-22A)

Mea-surement was taken every hour from 12:00 hours

(27 December 2008) to 11:00 hours the next day

Analysis of pond water

Analysis was performed using the standard methods

described by the American Public Health Association

(APHA 1995): (1) alkalinity: titration method, methyl

orange 0.1% indicator; (2) H2S: methylene blue

meth-od; (3) BOD: 5-day BOD test; (4) COD: oxidation by

KMnO4, titration method; (5) total Fe and Fe21:

phe-nanthroline method, colorimetric method, 510 nm;

(6) TSS: weight method, dried at 105 1C; (7) NO2-N:

diazo method; (8) NO3-N: salicylate method; (9) TAN:

indophenol blue method; (10) total N: digested by

Kjeldahl digestion system, indophenol blue method;

(11) PO4-P and total P: ascorbic acid method; (12)

Cl: argentometric method; and (13) dissolved

SO4 : turbidimetric method, 420 nm

Analysis of pond sediment

To minimize sample oxidation, freshly collected

sedi-ment samples were analysed for total Fe, Fe21, NO2-N,

NO3-N and NH4-N within 2 days after sampling Theremaining samples were air-dried, ground in a plasticmortar and passed through a 0.25 mm sieve for theanalysis of pH1/2.5, EC1/5, sulphate (soluble/total),

OM, total N, PO4-P and total P The methods weresummarized as follows: (1) total Fe: extracted by

H2SO40.1 N, colorimetric method, line,l 5510 nm; (2) Fe21: used the same extractionmethod as total Fe but no Fe31reduction step; (3)

1,10-phenanthro-NO2-N: extracted by distilled water, Griss method,

l 5 543 nm; (4) NO3-N: extracted by distilled water,exclusion of Cl, colorimetric method, phenoldisun-phonic acid,l 5410 nm; (5) NH4-N: extracted by KCl

2 M, Kjeldahl method (Rowell 1994); (6) PO4-P: tracted by NaHCO30.5 M, colorimetric method usingascorbic acid,l 5660 nm (Olsen’s method) (Rowell1994); (7) pH1/2.5: pH meter, soil:water ratio 51:2.5(w/v) (Rowell 1994); (8) EC1/5: EC meter, soil:waterratio 51:5 (w/v) (Rowell 1994); (9) soluble sulphate:extracted by distilled water, turbidimetric method(BaCl2, l 5420 nm); (10) OM: loss-on-ignitionmethod (550 1C, 4 h); (11) total P: residues after OMdetermination were dissolved in HCl 1 M, colori-metric method, ammonium molybdate,l 5660 nm;(12) total sulphate: used the same extraction method

ex-as total P, turbidimetric method (BaCl2,l 5420 nm);(13) total N: extracted by heating with concentrated

H2SO4, catalysed by a CuSO4:K2SO4mixture of 1:10ratio, Kjeldahl method (Rump & Krist 1992); and (14)BD: core method (Boyd 1995)

Statistical analysisDescriptive statistics were applied to get an overview

of ponds’ physico-chemical characteristics Analysis

of variance (ANOVA) was used to clarify the e¡ects ofthe categorical variables on the dependent variables(physico-chemical parameters) For water para-meters, the categorical variable included the ‘sam-pling time’, which had ¢ve levels (i.e the ¢ve

Table 1 Sampling times and the sediment parameters of the improved extensive shrimp system in the Cai Nuoc district, kong Delta, SouthernVietnam (2008)

sampling times) The dependent variables included

all physico-chemical parameters analysed For

sedi-ment, the categorical variables included

‘depth’,‘sam-pling position’ and ‘sam‘depth’,‘sam-pling time’ The ¢rst two

variables had two levels (depth: 0^5 and 5^10 cm,

sampling position: trench and platform) Signi¢cant

di¡erences between the levels of the categorical

vari-ables were further analysed with the Tukey Honest

Signi¢cant Di¡erence test for post hoc mean’s

compar-ison All signi¢cance testing was conducted at the

0.05 level Correlations among the parameters of

in-terest were conducted using correlation matrices

Results and discussion

Physico-chemical parameters of pond water

Table 2 shows the water physico-chemical

character-istics of the improved extensive shrimp ponds

Tem-perature and pH were most optimal for shrimp

farming (Jayasinghe et al 1994; Haws & Boyd 2001;

Lazur 2007) Most TSS measurements exceeded the

limit Salinity showed a slightly wider range than commended (Haws & Boyd 2001; Lazur 2007) COD,BOD and nutrients (N, P) were low NO3-N valueswere about seven times higher than those in the for-mer Cai Nuoc district (including the current CaiNuoc and Phu Tan just to its west, see Fig 1) in2002^2004 (Tho et al 2006), implying a recent in-crease in nutrients in the ponds NO2-N, PO4-P andBOD were lower than intensive shrimp farms in East-ern Thailand (Tookwinas & Songsangjinda 1999) Up

re-to 37.14% of DO measurements were lower than therecommendation by Lazur (2007), and the minimum

DO value was similar to that in extensive shrimpfarming system in Bac Lieu province (Bao 2009) To-tal iron was much higher than recommended (Haws

& Boyd 2001), although being lower than the 2002^

2004 ¢gure in the former Cai Nuoc district (Tho et al.2006) This was mostly because of the lesser impact

of acid sulphate soils in the study area (further land, previously rice-dominated) as compared withthe former Cai Nuoc district (Phu Tan district con-nects to the sea and has been occupied by man-

in-Table 2 Water characteristics of the improved extensive shrimp system in the Cai Nuoc district, Mekong Delta, Southern Vietnam (2008) in comparison with criteria for shrimp farming

Parameter

(n 5 35) Unit Range/mean

Standard deviation

Acceptable range

pH – 6.25–8.95/7.59 – 7–9w 6.67  3.33  3.33 Temperature 1C 27.6–35.4/30.6 2.437 26–33z 20 18.00 14.00 18.00

DO mg L  1 2.38–6.68/4.46 1.129 4 37.14  8.57  8.57  2.86  5.71  11.43 Turbidity NTU 4–66/27.21 14.139

Salinity g L  1 3–39/14.56 12.069 5–35w 37.14 114.29  5.71  11.43  5.71 Alkalinity mg L  1 80–260/148.77 37.131 50–100‰ 94.29 120.00 117.14 120.00 117.14 120.00 COD mg L  1 2.80–22/8.66 3.445

H 2 S mg L  1 0–0.02/0.007 0.005 NDw 85.71 120.00 120.00 120.00 111.43 114.29

Cl  g L  1 1.66–20.42/7.38 6.109 2–20w 17.14 15.71  2.86  5.71  2.86

SO 4  g L  1 0.15–3.73/1.18 1.117 0.5–3w 57.14 114.29  17.14  14.29  11.43

Percentage of out-of-range measurements: 1, % above the range; , % below the range.

wHaws and Boyd (2001).

zJayasinghe et al (1994).

‰Lazur (2007).

zVietnamese standard 5943-1995 for aquatic cultivation.

ND, not detectable.

groves, acid sulphate soils dominated) (Fig 1) H2S

concentrations were higher than the threshold set

by Haws and Boyd (2001), but lower than those by

Jayasinghe et al (1994) and Lazur (2007) The

pre-sence of H2S in the ponds was not necessarily linked

to a sulphide reservoir in the soil pro¢les, but simply

produced under anoxic pond sediment caused by a

long-term accumulation of organic materials

Alkali-nity, which can a¡ect primary productivity, was

higher than recommended (Lazur 2007) It was also

slightly higher than that in the extensive shrimp

farming system in Bac Lieu province, most probably

because salinity in Cai Nuoc shrimp ponds was

high-er Several key parameters (DO, TSS, alkalinity and

H2S) showed values beyond the acceptable range for

shrimp farming all year round (Table 2) Indicators of

organic loadings in the ponds (COD, BOD, N and P)were low due to low inputs (low stocking density, lit-tle or no fertilizers and no feeding) in this shrimpfarming system

Dissolved oxygen values below the levels at whichshrimp can grow normally (Haws & Boyd 2001; La-zur 2007) were found at all sampling times (Table 2).The situation was even worse during the early morn-ing (06:00^08:00 hours) as revealed by the 24-hourmonitoring experiment (Table 3, Fig 3) The peak va-lues of DO were found around 11:00^12:00 hours.The values then gradually decreased to a minimumaround 06:00^08:00 hours, after which they in-creased again In Pond 1, a DO range between 2and 3 mg L 1 lasted for about 10 h In the othertwo ponds, DO values were even lower This was not

Table 3 Minimum and maximum values of the parameters under the 24-hour monitoring experiment in the improved tensive shrimp system in the Cai Nuoc district, Mekong Delta, SouthernVietnam (2008)

ex-Pond

experi-because of organic/nutrient pollution, but simply on

account of diurnal £uctuations of photosynthesis

and respiration intensity

Di¡erences in DO among the ponds are most likely

caused by the di¡erent degrees of shading, rather than

by the stocking densities or the quality of input water It

should be noted that DO levels are much lower near the

water^sediment interface (where shrimps stay) than

near the water surface (Flegel 1998) The maximum

DO values at noon in all three ponds were attributed

to elevated solar radiation, which led to a higher

photo-synthesis intensity by which a larger amount of DO

was released Elevated DO values during daytime were

the result of the dominance of photosynthesis over

re-spiration Lowest pH in the early morning was

attribu-ted to accumulation of carbon dioxide during the night

(Lazur 2007) When no solar radiation reached the

water, the photosynthesis stopped while the respiration

still continued in the ponds, which consumed oxygen

and released carbon dioxide into the water column

Correlation matrix using data from the 24-hour

monitoring experiment revealed that both DO and

pH were positively correlated (both at Po0.001) with

temperature (i.e solar radiation) In fact,

tempera-ture, pH and DO in the ponds showed similar

pat-terns (Fig 3) and had linear correlations (Fig 4)

In pond water, NO3-N concentrations were about

¢ve times higher than TAN and 58 times higher than

NO2-N on average (Table 2) This, together with the

positive correlation between NO3-N and TKN

(Po0.01), indicated that NO3-N accounted for the

majority of the inorganic nitrogen Positive

correla-tions between salinity and TKN (Po0.01) and NO3-N

(Po0.001) were most likely the result of the e¡ect of

concentration Total ammonium nitrogen was

posi-tively correlated with COD (Po0.05) but negatively

correlated with DO, pH and temperature (all at

Po0.01) This suggests that TAN was largely involved

in the photosynthesis process though it accountedfor a small percentage among the nitrogen species.While Vinatea, GaŁlvez, Browdy, Stokes,Venero, Have-man, Lewis, Lawson, Shuler and Le¥er (2010) found

no relationship between TAN and photosynthesis,Burford and Glibert (1999) demonstrated that phyto-plankton plays a key role in controlling TAN inshrimp ponds According to McCarthy (1980), ammo-nium is the preferred nitrogenous nutrient of manyspecies of marine phytoplankton During periods ofhigh photosynthesis intensity when a vast amount

of solar radiation reaches the water surface, the jority of TAN was taken up by the phytoplanktoncommunity, resulting in low TAN concentrations.During the nights (no photosynthesis) and the earlymornings (little photosynthesis), TAN may be accu-mulated in pond water Salinity and BOD was nega-tively correlated (Po0.001), suggesting that pondsmight have had higher biodegradable matter duringthe rainy season despite the dilution impact (Tho

ma-et al 2006) The situation can only be explained bythe contribution of OM from runo¡ water Alkalinitywas negatively correlated with PO4-P (Po0.01) butpositively correlated with NO3-N (Po0.01) Alkalinitywas positively correlated with salinity (Po0.01), sug-gesting that alkalinity often increased at correspond-ing high salt concentrations Fe31was much higherthan Fe21(Table 2) and of higher positive correlationwith total Fe (Po0.01 vs Po0.05), indicating pondwater in an oxidized state Total suspended solid waspositively correlated with turbidity (Po0.05), sug-gesting that suspended solids were the main compo-nent of water turbidity Besides, TSS was positivelycorrelated with BOD (Po0.05) and NO2-N (Po0.01),

Figure 4 Linear correlations between (a) dissolved oxygen (DO) and temperature and between (b) DO and pH in the hour monitoring experiment

24-suggesting that OM was largely made up by

sus-pended solids

Total suspended solid showed the highest values in

October (rainy season) while salinity, Cl, SO4 

and

NO3-N showed their highest values in April (end of

the dry season) (Table 4) This was not necessarily a

build-up of salt/OM but simply the concentration

ef-fect in the dry season A similar situation was

re-ported in the former Cai Nuoc district (Tho et al

2006) Other indicators of organic pollution (BOD,

TKN and TAN) also tended to show higher values in

December or April (dry season) The highest TSS

va-lues were due to the e¡ect of runo¡ water

Physico-chemical parameters of pond

sediment

The sediment quality of the shrimp ponds is

repre-sented by the 0^5 cm layer (Table 5) This is because

this layer in shrimp ponds is more reactive with thewater column than the deeper layers (Masuda & Boyd1994; Munsiri, Boyd & Hajek1995).While pHwrangedfrom slightly acidic to neutral (5.16^7.67), pH1/2.5

showed a wider range (3.97^7.91) This is similar toshrimp farms in Ecuador where pH1/1 showed awider range than pHwat both ends (Sonnenholzner

& Boyd 2000) In semi-intensive shrimp farms inHonduras and Bangladesh, and extensive shrimpfarms in the Mekong Delta of Vietnam, neutral pH va-lues of the sediment were reported (Munsiri, Boyd,Teichert-Coddington & Hajek 1996; Alongi et al.1999; Islam, Sarker, Yamamoto, Wahab & Tanaka2004) In Thailand, pH values of shrimp farm sedi-ment were higher (8.14^8.29) (Towatana, Voradaj &Panapitukkul 2002) The pH of the Cai Nuoc shrimppond sediment was lower than the optimal rangefor shrimp farming (7^8) (Haws & Boyd 2001) Thiswas because of the limited investment of the MekongDelta improved extensive shrimp farmers in chemical

Table 4 ANOVA testing of the water parameters in the Cai Nuoc improved extensive shrimp system, Mekong Delta, Southern Vietnam (2008)

Significance level

Means comparisons Sampling

Indices with di¡erent superscript are signi¢cantly di¡erent at the 0.05 level.

ANOVA signi¢cance level.

Po0.01; Po0.001.

wDry season.

zRainy season.

NS, not signi¢cant.

inputs to bu¡er pH in their shrimp ponds (Field

data 2008)

The sediment was anaerobic (Eh  304 mV),

which can be blamed for the production and

subse-quent release of toxic substances (e.g H2S) into pond

water (Physico-chemical parameters of pond water)

Anaerobic condition of pond bottom was attributed

to elevated OM loadings (OM 9.84^21.96%, mean

12.91%), resulting from accumulation over extended

periods (42 years in some cases) Organic matter in

Cai Nuoc shrimp pond sediment was much higher

than those (1.06^1.42%) in Southern Thailand

(To-watana et al 2002) and Bangladesh (2.14^3.28%)

(Is-lam, A(Is-lam, Rheman, Ahmed & Mazid 2004) While

the £occulent layer of recently deposited sediment

may have an OM content of 50%, the upper 2 cm

se-diment layer in shrimp ponds seldom has410% of

OM (Haws & Boyd 2001) Following Mitsch and

Gos-selink (2000), the content of organic C of the shrimp

pond sediment can be estimated at 6.46% According

to Boyd (2003), however, organic C content of shrimp

pond sediment should be between 1.5% and 2.5%

High OM in the shrimp pond sediment did not relate

to daily £uctuations of DO in the water column

While the former remained unchanged, the latter

exhibited critical £uctuations within 1 day The

decomposition rate of OM in Cai Nuoc shrimp pond

sediment was low due to suboptimal sediment pH

(5.16^7.67) The best pH range for this process should

be between 7.5 and 8.5 (Boyd 1992)

The anaerobic condition of the ponds was furtherevidenced by higher NH4-N as compared with NO3-Nand higher Fe21as compared with Fe31(Table 5) To-tal Fe (7340.54 mg kg 1) was much higher than that

in shrimp farms in Ecuador (661mg kg 1) holzner & Boyd 2000) Total P (0.013%) was similar tothat in Southern Thai shrimp ponds (Towatana et al.2002) but lower than that in intensive shrimp farms

(Sonnen-in Australia (0.069%) (Smith 1996) and Ecuador(0.09%) (Sonnenholzner & Boyd 2000) On average,total N (0.192%) was similar to that in semi-intensiveshrimp farms in Honduras (0.17^0.28%) (Munsiri

et al 1996), Bangladesh (0.11^0.18%) (Islam, Alam

et al 2004) and Ecuador (0.02^0.52%) ner & Boyd 2000), but lower than that in intensiveshrimp farms in Australia (0.23%) (Smith 1996) Inthe shrimp ponds of the former Cai Nuoc district,top sediment showed an elevated exchangeable so-dium percentage (35.89^58.66%) in the 2002^2004period (Tho et al 2008) Exchangeable form of majorcations (Ca, Mg, Na and K) (Tho et al 2008) were simi-lar to those in Thailand (Towatana et al 2002) Anae-robic pond bottom and high OM loadings provideddisadvantages for shrimp growth in these ponds, par-ticularly the toxic reduced substances and the im-pacted benthic community

(Sonnenholz-Di¡erences between sampling positions, betweendepths and between sampling times were found(Table 6) Trench sediment had higher values of

NH4-N, PO4-P, total P and Fe21 than the platform

Table 5 Descriptive statistics of sediment parameters (0^5 cm) in the improved extensive shrimp system in the Cai Nuoc district, Mekong Delta, SouthernVietnam (2008)

This was probably because (i) the trench is much

dee-per (thus subject to a larger amount of TSS and can

attract suspended matter from the platform after

re-suspension), and (ii) the trench is directly a¡ected by

the runo¡ water Total N and NO3-N, however,

re-mained higher on the platform Higher total N and

total P and lower BD in the top layer (Table 6)

sug-gested that the ¢rst few centimetres of sediment

sur-face still remained active This layer was enriched by

organic materials, derived mainly from intake water,

shrimp excretions or dead materials from the

vegeta-tive cover Increases in values of several parameters

(Fe31, total Fe, PO4-P, total P and OM) at times during

the rainy season were explained by the contribution

of eroded material from the dikes

The top layer showed higher EC1/5and soluble SO4

values than the second layer Sediment salinity (EC1/

5) was low at the end of the rainy season but

increas-ing durincreas-ing the dry season (Table 6) For the period

2002^2004, salinity values of 3.16 and 8.88 dS m 1

for, respectively, the rainy and dry season in the

for-mer Cai Nuoc district were reported (Tho et al 2008)

Salinity in the shrimp pond sediment was similar to

that in Southern Thailand (Towatana et al 2002)

Suitability of ponds’ physico-chemical

characteristics for shrimp (P monodon)

farming

The improved extensive ponds did not provide

appro-priate conditions for shrimp Many water parameters

show values beyond the acceptable limit for shrimp

farming (Table 2).Very low DO levels can lead to direct

shrimp mortality or reduce shrimp resistance to

dis-eases (Haws & Boyd 2001; Lazur 2007) In fact, shrimp

mortality in early mornings has been observed (Field

data 2008) In the rainy season, high TSS might cause

a reduction of the photosynthesis intensity, while low

salinity may lead to acute stress, causing increased

susceptibility of shrimp to white spot virus infection

(Joseph & Philip 2007) Large salinity variations may

reduce shrimp immunocompetence to Photobacterium

damselae (Wang & Chen 2006) Alkalinity and total Fe

were much higher than recommended, suggesting

unfavourable conditions for shrimp

Poor pond sediment quality can adversely a¡ect

cultured shrimp (Smith 1999; Avnimelech & Ritvo

2003; Lemonnier, Bernard, Boglio, Goarant &

Co-chard 2004) It is possible that toxic products are

being produced under the anaerobic conditions of

the shrimp pond bottom Indeed, H2S values beyond

the accepted level were found, pointing to a negativeimpact to shrimp growth and survival (Ritvo,Samocha, Lawrence & Neill 1998) Increased shrimposmotic pressure caused by low sediment pH (such as6.5^7.4) (Lemonnier et al 2004) was likely to occur.Elevated OM in the sediment may also pose impacts

to shrimp by favouring greater microbial activities(Masuda & Boyd 1994)

ConclusionThe unfavourable physico-chemical characteristics

of the improved extensive shrimp ponds are not mal for shrimp farming Poor water and sedimentquality can stress the cultured shrimp and increasetheir susceptibility to diseases Appropriate planningand modi¢cations of farming techniques are stronglyrecommended for a better shrimp production in thecoastal parts of the Mekong Delta of Vietnam To im-prove the pond’s physico-chemical condition, the fol-lowing measures should be taken (1) allowingsedimentation before ¢lling shrimp culture ponds;(2) covering dikes using endemic vegetation to reducethe e¡ect from runo¡ water; (3) including no-culturebreaks between shrimp crops; (4) drying and expos-ing pond bottom to sunlight for a reasonable period oftime after every production cycle; (5) removing all se-diment produced after two production cycles; and (6)controlling the pond’s vegetation to mitigate the con-dition of oxygen shortage

opti-Future research in the area should focus on (1) ning of a synchronous irrigation system for a betterwater management within the dense river/canal sys-tem; (2) optimal pond construction and management(e.g stocking density, management of pond water andsediment); (3) study of native organisms (such as oy-sters, crabs, ¢sh and vegetation) that can be incorpo-rated into shrimp ponds for both economic andenvironmental values; (4) building up a common andmost appropriate cropping calendar for the whole dis-trict; and (5) wise use of small land plots and the de-posited sediment for the diversi¢cation of incomesources for shrimp farmers

plan-AcknowledgmentsThe research was funded by the Belgian DirectorateGeneral for Development Cooperation through theBelgian Technical Cooperation The authors wish tothank the sta¡ of the Ho Chi Minh City Institute ofResources Geography and the College of Aquaculture

and Fisheries (Can Tho University) for their valuable

assistance during ¢eldwork and lab analysis Special

thanks are forwarded to Dr Nico Vromant and Ms

Nguyen Thi Hoai Chau who o¡ered a great deal of

help during statistical analysis and the revision of

the manuscript Thanks are also extended to the

lo-cal authorities and farmers for their collaboration

during sampling and investigations

References

ADB/MoFi (1996) Vietnam coastal aquaculture sector

re-view In: IndustrialTransformation and Shrimp Aquaculture

inThailand andVietnam: Pathways to Ecological, Social, and

Economic Sustainability? (ed by L Lebel, N.H Tri, A.

Saengnoree, S Pasong, U Buatama & L.K Thoa Ambio

31, 311^323.

Alongi D.M., Tirendi F & Trott L.A (1999) Rates and

path-ways of benthic mineralization in extensive shrimp

ponds of the Mekong Delta, Vietnam Aquaculture 175,

269^292.

APHA (1995) Standard Methods for the Examination of Water

and Wastewater, 19th edn American Public Health

Asso-ciation,Washington, DC, USA.

Avnimelech Y & Ritvo G (2003) Shrimp and ¢sh pond soils:

processes and management Aquaculture 220, 549^567.

Bao T.Q (2009) Impact of extensive shrimp farming on a

tropi-cal river ecosystem (Bac Lieu province - Viet Nam) PhD

the-sis,Vrije Universiteit Brussel, Belgium.

Binh C.T & Lin C.K (1995) Shrimp culture inVietnam.World

Aquaculture 26, 27^33.

Binh T.N.K.D., Vromant N., Hung N.T., Hens L & Boon E.K.

(2005) Land cover changes between 1968 and 2003 in

Cai Nuoc, Ca Mau Peninsula, Vietnam Environment,

De-velopment and Sustainability 7, 519^536.

Boyd C.E (1992) Shrimp pond bottom soil and sediment

management In: Proceedings of the Special Session on

Shrimp Farming (ed by J Wyban), pp 166^181 World

Aquaculture Society, Baton Rouge, LA, USA.

Boyd C.E (1995) Bottom Soils, Sediment, and Pond

Aquacul-ture Department of Fisheries and Allied Aquacultures at

Auburn University, Chapman & Hall, New York, NY, USA.

Boyd C.E (2003) Best management practices for water and soil

management in shrimp farming Workshop, 23 June to 25

June 2003 in MazatlaŁn, Me¤xico.

Brennan D., Preston N., Clayton H & Be T.T (2002) An

eva-luation of rice-shrimp farming systems in the Mekong Delta.

Report prepared under the World Bank, NACA,WWF and

FAO Consortium Program on Shrimp Farming and the

Environment, Work in Progress for Public Discussion,

Published by the Consortium, 10pp.

Burford M.A & Glibert P.M (1999) Short-term nitrogen

cy-cling in early and late growth-phase shrimp ponds In:

Modeling Nitrogen Dynamics in Intensive Shrimp Ponds:

the Role of Sediment Remineralization (ed by M.A Burford

& K Lorenzen Aquaculture (2004), 229, 129^145 ChienY.-H (1992) Water quality requirements and manage- ment for marine shrimp culture In: Proceedings of the Spe- cial Session on Shrimp Farming (ed by J Wyban), pp 144^ 156.World Aquaculture Society, Baton Rouge, LA, USA Clough B., Johnston D., Xuan T.T., Phillips M.J., Pednekar S.S.,Thien N.H., Dan T.H & Thong P.L (2002) Silvo¢shery farming systems in Ca Mau province,Vietnam Report pre- pared under the World Bank, NACA, WWF and FAO Con- sortium Program on Shrimp Farming and the Environment, Work in Progress for Public Discussion, Published by the Consortium,70pp.

EJF (2003) Risky Business: Vietnamese Shrimp Aquaculture Impacts and Improvements Environmental Justice Foun- dation, London, UK.

-EJF (2004) Farming the Sea, Costing the Earth: WhyWe Must Green the Blue Revolution Environmental Justice Founda- tion, London, UK.

Flegel T.W (1998) Advances in shrimp biotechnology ings to the Special Session on Shrimp Biotechnology,5th Asian Fisheries Forum,11^14 November, Chiengmai,Thailand Haws M.C & Boyd C.E (2001) Methods for Improving Shrimp Farming in Central America Central American University Press-UCA, Managua, Nicaragua.

Proceed-Hena Abu M.K., Sharifuzzaman S.M., Hishamuddin O.,

Mis-ri K & Abdullah F (2008) Pond health management of black tiger shrimp Penaeus monodon (Fabricius) using bac- terial products In: Diseases in Asian Aquaculture VI Fish Health Section (ed by M.G Bondad-Reantaso, C.V Mohan,

M Crumlish & R.P Subasinghe), pp 469^476 Asian eries Society, Manila, Philippines.

Fish-Hens L.,Vromant N.,Tho N & Hung N.T (2009) Salination of surface water, groundwater, and soils in the shrimp farm- ing areas of the coastal Cai Nuoc district, South Vietnam International Journal of Environmental Studies 66, 69^81 Islam M.L., Alam M.J., Rheman S., Ahmed S.U & Mazid M.A (2004) Water quality, nutrient dynamics and sediment pro-

¢le in shrimp farms of the Sundarbans mangrove forest, Bangladesh IndianJournal of Marine Sciences 33,170^176 Islam M.S., Sarker M.J.,YamamotoT.,Wahab M.A & Tanaka M (2004) Water and sediment quality, partial mass budget and e¥uent N loading in coastal brackishwater shrimp farms in Bangladesh Marine Pollution Bulletin 48, 471^485 Jayasinghe J.M.P., Corea S.L & Wijegunawardana P.K.M (1994) A¡ordable water supply and sanitation ^ deterioration

of sanitary conditions in coastal waters The 20th WEDC Conference Colombo, Sri Lanka.

Joseph A & Philip R (2007) Acute salinity stress alters the haemolymph metabolic pro¢le of Penaeus monodon and reduces immunocompetence to white spot syndrome virus infection Aquaculture 272, 87^97.

Lazur A (2007) Growout Pond and Water Quality ment University of Maryland, College Park, MD, USA Lemonnier H., Bernard E., Boglio E., Goarant C & Cochard J.C (2004) In£uence of sediment characteristics on

Manage-shrimp physiology: pH as principal e¡ect Aquaculture

240, 297^312.

Masuda K & Boyd C.E (1994) Phosphorus fractions in soil

and water of aquaculture ponds build on clayey, ultisols

at Auburn, Alabama In: Chemical and Physical Properties

of Shrimp Pond Bottom Soils in Ecuador (ed by S

Sonnen-holzner & C.E Boyd) Journal of the World Aquaculture

So-ciety 31, 358^375.

Masuda K & Boyd C.E (1994) Chemistry of sediment pore

water in aquaculture ponds built on clayey ultisols at

Aut-burn, Alabama In: In£uence of Sediment Characteristics on

Shrimp Physiology: pH as Principal E¡ect (ed by H

Lemon-nier, E Bernard, E Boglio, C Goarant & J.C Cochard).

Aquaculture 240, 297^312.

McCarthy (1980) Nitrogen and phytoplankton ecology In:

Exploring the Mechanism of Ammonium Uptake in

Phyto-plankton with an Ammonium Analogue, Methylamine (ed.

by W.M Balch) Marine Biology 92, 163^171.

Mitsch W.J & Gosselink J.G (2000) Wetlands, 3rd edn John

Wiley & Sons, New York, NY, USA.

Munsiri P., Boyd C.E & Hajek B.F (1995) Physical and

chemi-cal characteristics of bottom soil pro¢les in ponds at

Au-burn, Alabama, and a proposed system for describing

pond soil horizons In: Chemical and Physical Properties of

Shrimp Pond Bottom Soils in Ecuador (ed by S

Sonnen-holzner & C.E Boyd) Journal of the World Aquaculture

So-ciety 31, 358^375.

Munsiri P., Boyd C.E., Teichert-Coddington D & Hajek B.F.

(1996) Texture and chemical composition of soils from

shrimp ponds near Choluteca, Honduras Aquaculture

In-ternational 4, 157^168.

Nhuong T.V., Luu L.T.,Tu T.Q.,Tam P.M & Nguyet T.T.A (2002)

Vietnam shrimp farming review Individual Partner Report

for the Project: Policy Research for Sustainable Shrimp

Farming in Asia European Commission INCO-DEV Project

PORESSFA No IC4^2001-10042, CEMARE University of

Portsmouth, UK, and RIA1, Bac Ninh,Vietnam.

Phuong N.T., Minh T.H & Tuan N.A (2004) Overview of

shrimp farming systems in the Mekong Delta of Vietnam

Pa-per presented at the conference ‘Development of Coastal

Aquaculture Resources’, University of Agriculture and

For-estry, 4 August 2004.

Poernomo A (1989) Technical constraints in shrimp culture

and how to overcome them In: Proceedings of the Shrimp

Culture IndustryWorkshop (Jepara, Indonesia,1989), 25–28

September 1989, (ed by Yap W.G), pp 59–66 FAO

Fisheries and Aquaculture Department, Jepara City,

Indonesia Available at http://www.fao.org/docrep/field/

003/ac058e/AC058E03.htm#ch2.3

Preston N & Clayton H (2003) Rice-shrimp farming in the

Mekong Delta: biophysical and socioeconomic issues ACIAR

Technical Reports No.52e, ACIAR, Canberra, Australia,

170pp.

Raux P & Bailly D (2002) Literature review on world shrimp

farming Individual Partner Report for the Project: Policy

Research for Sustainable Shrimp Farming in Asia,

European Commission INCODEV Project No

IC4-2001-10042, CEMARE University of Portsmouth, UK, and CEDEM, Brest, France, 46pp.

Ritvo G., Samocha T.M., Lawrence A.L & Neill W.H (1998) Growth of Penaeus vannamei on soils from various Texas shrimp farms, under laboratory conditions In: Chemical and Physical Properties of Shrimp Pond Bottom Soils in Ecuador (ed by S Sonnenholzner & C.E Boyd) Journal of theWorld Aquaculture Society 31, 385^375.

R˛nnbck P (2001) Shrimp Aquaculture: State of the Art Swedish EIA Centre, Report 1 Swedish University of Agri- cultural Sciences (SLU), Uppsala, Sweden.

Rowell D.L (1994) Soil Science: Methods and Applications Department of Soil Science, University of Reading, Pearson Education, Prentice Hall, Harlow, UK.

Rump H.H & Krist H (1992) Laboratory Manual for the amination of Water, Waste Water and Soil, 2nd edn VCH, Weinheim, Germany.

Ex-Sels A (2004) Shrimp aquaculture and the socio-economic tuation of the Cai Nuoc district, Ca Mau Province,Vietnam: application of the POET model Master Thesis in Human Ecology, Faculty of Medicine and Pharmacology,Vrije Uni- versiteit Brussel.

si-SIG (2001) Studying the Scienti¢c Basis for the Planning Adjustment, Production Rearrangement at Cai Nuoc Dis- trict, Ca Mau Province (2000^2001) Sub-Institute of Geo- graphy, Ho Chi Minh City,Vietnam (inVietnamese) Smith P T (1996) Physical and chemical characteristics of sediments from prawn farms and mangrove habitats on the Clarence River, Australia Aquaculture 146, 47^83 Smith P T (1999) Coastal shrimp aquaculture inThailand: key issues for research ACIAR Technical Reports No 47, ACIAR, Canberra, Australia, 131pp.

Sonnenholzner S & Boyd C.E (2000) Chemical and physical properties of shrimp pond bottom soils in Ecuador Jour- nal of theWorld Aquaculture Society 31, 358^375 Tho N.,Vromant N., Hung N.T & Hens L (2006) Organic pol- lution and salt intrusion in Cai Nuoc District, Ca Mau Pro- vince,Vietnam.Water Environment Research 78,716^723 Tho N., Vromant N., Hung N.T & Hens L (2008) Soil salinity and sodicity in a shrimp farming coastal area of the Mekong Delta,Vietnam Environmental Geology 54, 1739^1746 Thornton C., Shanahan M & Williams J (2003) From wetlands

to wastelands: impacts of shrimp farming tal justice foundation SWS Bulletin 20, 48–53 Tookwinas S & Songsangjinda P (1999) Water quality and phytoplankton communities in intensive shrimp culture ponds in Kung Krabaen Bay, Eastern Thailand Journal of theWorld Aquaculture Society 30, 36^45.

Environmen-Towatana P.,Voradaj C & Panapitukkul N (2002) Changes in soil properties of abandoned shrimp ponds in southern Thailand Environmental Monitoring and Assessment 74, 45^65.

Vietnamese Standard 5943-1995 Available at http://khcn mt.gov.vn/tieuchuannganh/uploads/TCVN_5943-1995 pdf (accessed 1 February 2010) (in Vietnamese).

Vinatea L., GaŁlvez A.O., Browdy C.L., Stokes A.,Venero J.,

Have-man J., Lewis B.L., Lawson A., Shuler A & Le¥er J.W (2010)

Photosynthesis, water respiration and growth performance

of Litopenaeus vannamei in a super-intensive raceway

culture with zero water exchange: interaction of water

quality variables Aquacultural Engineering 42, 17^24.

Vuong D.Q.T & Lin C.K (2001) Rice-shrimp farming in the

sea-water intrusion zone of the Mekong Delta,Vietnam ITCZM

Monograph No 6.

Wang F.I & Chen J.C (2006) E¡ect of salinity on the immune response of tiger shrimp Penaeus monodon and its suscept- ibility to Photobacterium damselae subsp damselae Fish and Shell¢sh Immunology 20, 671^681.

Wolanski E., Spagnol S., Thomas S., Moore K., Alongi D.M., Trott L & Davidson A (2000) Modelling and visualiz- ing the fate of shrimp pond e¥uent in a mangrove- fringed tidal creek Estuarine, Coastal and Shelf Science 50, 85^97.

Partial replacement of fish meal by a mixture of

soybean meal and rapeseed meal in practical diets

effects on growth performance and in vivo digestibility

Zhi Luo1, Xiao-Dong Li2, Wei-Min Wang1, Xiao-Ying Tan1& Xu Liu2

1 Fishery College, Huazhong Agricultural University,Wuhan 430070, China

2 Postgraduate Research Base, Panjin Guanghe Fishery, Panjin, China

Correspondence: Zhi Luo, Fishery College, Huazhong Agricultural University,Wuhan 430070, China E-mail: luozhi99@yahoo.com.cn

Abstract

An 8-week experiment was conducted to examine

the e¡ect of partial replacement of ¢sh meal (FM)

by a mixture of soybean meal (SBM) and rapeseed

meal (RM) in practical diets of juvenile Chinese

mit-ten crab Eriocheir sinensis of initial body weight of

1.54 0.12 g (means  SD, n 5 90) Five

isonitrogen-ous diets were formulated to contain 35% protein and

5% lipid Soybean meal and RM mix (1:1 ratio) were

included at ¢ve levels of 0 (control), 15%, 30%, 45%

and 60%, replacing 0, 20%, 40%, 60% and 80% FM

respectively When FM was replaced by 15% of SBM

and RM, crab showed the highest growth, feed

utiliza-tion and moulting frequency (MF) Fish meal replaced

by SBM and RM did not signi¢cantly in£uence crude

protein, lipid and moisture contents of whole body

crab, but ash content was the lowest for crab fed the

diet with FM replaced by 15% of SBM and RM

Appar-ent digestibility coe⁄ciAppar-ents (ADCs) of dry matter,

crude protein and energy tended to decline with

in-creasing inclusion levels of dietary SBM and RM In

general, ADCs of lipid were high (over 90%) and

showed no signi¢cant di¡erences among the

treat-ments (P40.05) Based on these observations above,

these results indicated that about 40% of FM can be

replaced with a mixture of SBM and RM (1:1 ratio) in

the diet of E sinensis without adverse growth

perfor-mance, compared with the FM-based diet However,

20% of FM replaced by SBM and RM produced the

best growth performance and feed utilization

Keywords: Eriocheir sinensis, ¢sh meal, soybean

meal, rapeseed meal, growth performance,

appar-ent digestibility coe⁄ciappar-ents

IntroductionAlthough it is regarded as an invader causinggreat ecological and economic loss in European andAmerican countries (Rudnick, Hieb, Grimmer & Resh2003), the Chinese mitten crab Eriocheir sinensis haslong been a fashionable table delicacy in autumn inChina, Japan and other Asian countries At present,farming of the mitten crab in China has been prac-ticed throughout the country, and aquacultureproduction of this species had increased dramatic-ally from 8000 tonnes in 1991 to approximately

500 000 tonnes in 2004 (Yang & Zhang 2005) Forseveral recent years, its market price has been con-tinuously rising owing to the imbalance betweensupply and demand, which has provoked consider-able interest in the development of intensive culture.There are, however, many constraints remaining inthe further development of crab-intensive aquacul-ture, particularly the lack of knowledge of nutrientrequirements of this species, although several re-searches have been conducted to determine the opti-mal dietary protein levels (Mu, Shim & Guo 1998; Lin,Luo & Ye 2009), dietary n-3 highly unsaturated fattylevels and DHA/EPA ratios (Sui,Wille, Cheng & Sorge-loos 2007), apparent digestibility coe⁄cients (ADCs)

of some protein sources (Mu, Lam, Guo & Shim2000; Luo, Tan, Chen, Wang & Zhou 2008) and diet-ary arginine and lysine requirements (Jiang, Li,Chen, Wang, Liu & Liu 2005) for this species At pre-sent, the feeds used for current crab farming aremainly natural feeds, which do not meet the nutrientrequirements to sustain maximal growth of this spe-cies Accordingly, the development of a nutritionallyadequate and economical diet for its developmental