Aquaculture research, tập 41, số 1, 2010

E-mail: chke@xmu.edu.cn Abstract Independent and combined e¡ects of stocking density and algal concentration on the survival, growth and metamorphosis of the Bobu Ivory shell Babylonia f

Huaiping Zheng1,2, Caihuan Ke1, Zewen Sun2, Shiqiang Zhou1& Fuxue Li1

1 State Key Laboratory of Marine Environmental Science, Department of Oceanography, Xiamen University, Xiamen, China

2 Mariculture Research Center for Subtropical Shell¢sh & Algae, Shantou University, Shantou, China

Correspondence: C Ke, State Key Laboratory of Marine Environmental Science, Department of Oceanography, Xiamen University, Xiamen 361005, China E-mail: chke@xmu.edu.cn

Abstract

Independent and combined e¡ects of stocking density

and algal concentration on the survival, growth and

metamorphosis of the Bobu Ivory shell Babylonia

for-mosae habei larvae were assessed using a 5 5

fac-torial design with densities of 0.25, 0.5, 0.75, 1.00 and

1.50 larvae mL 1and algal concentrations of 5, 10,

15, 20 and 25 104cells mL 1in the laboratory

Larval growth, survival and metamorphosis were

signi¢cantly a¡ected by both the independent e¡ects

of stocking density and algal concentration and by

their interaction The highest per cent survival

(72.5%) and metamorphosis (49.5%), fastest growth

(41.57mm day 1) and shortest time to initial

meta-morphosis (10 days) all occurred at the lowest

stock-ing density and the highest algal concentration Both

crowding and food limitation had independently

ne-gative impacts on the survival, growth and

metamor-phosis of larvae, and these negative impacts were

further strengthened by the interaction of a higher

stocking density and a lower algal concentration

Moreover, the results suggest that stocking density

and algal concentration obviously played di¡erent

roles in determining larval survival and growth To

maximize survival and growth, B formosae habei

lar-vae should be reared at a lower stoking density of

0.25 larvae mL 1and fed a higher algal

stock-IntroductionTemperature, salinity, diet and rearing density areexogenous factors a¡ecting larval growth, settle-ment and metamorphosis (for a review, see Crisp1974) Stocking density and algal concentration aretwo more important factors in£uencing the success

of hatchery seed culture for planktotrophic larvae inmarine molluscs, because they are easier to manipu-late than other environmental factors in arti¢ciallarval production systems Therefore, the e¡ects ofstocking density or food concentration on larval sur-vival, growth and metamorphosis have been welldocumented in marine molluscs (Fretter & Montgom-ery 1968; Pilkington & Fretter 1970; Perron & Turner1977; Aldana-Aranda, Lucas, Brule, Salguero &Rendon 1989; Pechenik, Eyster, Widdows & Bayne1990; Hansen 1991; His & Seaman 1992; Pechenik,Estrella & Hammer 1996; Avila, Grenier,Tamse & Ku-zirian 1997; Preece, Shepherd, Clarke & Keesing 1997;Doroudi & Southgate 2000; Pechenik, Jarrett &Rooney 2002; Powell, Bochenek, John, Klinck &Hofmann 2002; Daume, Huchette, Ryan & Day 2003;Zhao, Qiu & Qian 2003; Zheng, Ke, Zhou & Li 2005;

Liu, Dong, Tang, Zhang & Xiang 2006; Yan, Zhang &

Yang 2006; Mazo¤n-SuaŁstegui, Ru|¤z-Ru|¤z,

Parres-Haro & Saucedo 2008; Raghavan & Gopinathan

2008; Capo, Bardales, Gillette, Lara, Schmale &

Sera-fy 2009; Rico-Villa & Robert 2009) In general, larvae,

reared under a lower stocking density or higher food

concentration conditions, have higher survival,

fas-ter growth and more metamorphosed individuals In

contrast, larvae, reared crowing or lower food supply

conditions show slower growth and development,

less survival and metamorphic success, and a

smal-ler size or a lower energy content at metamorphosis

Most of these studies focused on the independent

ef-fect of stocking density or food concentration on

lar-val survilar-val, growth and metamorphosis; only a few

studies involved their combined e¡ects (Doroudi &

Southgate 2000; Powell et al 2002; Mazo¤n-SuaŁstegui

et al 2008; Capo et al 2009) In large-scale hatchery

seed culture practice, it is important to de¢ne a

rea-listic strategy for maximizing larval growth and

sur-vival under optimal stocking density and diet

concentration conditions

Members of the gastropod genus Babylonia are

found only in the Indo-Paci¢c region (Altena,

Regte-ren & Gittenberger 1981) Bobu Ivory shell B formosae

habei only exists on the SE coast of China and

gener-ally inhabits the muddy/sandy subtidal zone at

depths of 4^20 m (in summer) or 40^60 m (in

win-ter) (Liu & Xiao 1998) The conch is a large marine

gastropod (adult size 50^60 mm), whose meat is an

important source of protein and is of considerable

dietary and economic signi¢cance to the inhabitants

of the coast of southern China In recent years, the

Bobu Ivory shell has been widely cultured in the

coast of southern China Larvae are generally reared

at a density of 0.5 larvae mL 1and fed an algal

con-centration of 10^20 104

cells mL 1 (Ke, Zheng,Zhu, Zhou & Li 2001; Zheng, Zhu, Ke, Zhou & Li

2001) However, the combined e¡ects of stocking

den-sity and algal concentration on larvae have been not

studied Determination of their combined e¡ects may

be an important step in developing more e⁄cient

large-scale hatchery culture techniques for B

formo-sae habei larvae culture

Because the factorial design provides a greater

precision for assessing the interactions between

di¡erent factors and allows the interpolation of

inter-actions at intermediate levels of the factors being

tested (Doroudi & Southgate 2000), a 5 5 factorial

design with densities of 0.25, 0.50, 0.75, 1.00 and

1.50 larvae mL 1and algal concentrations of 5, 10,

15, 20 and 25 104

cells mL 1was applied in the

laboratory And the independent and combinede¡ects of larval density and algal concentration

on the survival, growth and metamorphosis of

B formosae habei larvae were investigated in thepresent study

Materials and methodsAcquisition of larvaeAdults of B formosae habei were collected from thecoast of Changle, Fujian Province (China), and main-tained between 23 and 28 1C in two 1m3aquaria andfed with the razor clam Sinonovacula constricta (L.).Egg capsules containing fertilized eggs were depos-ited after several days, after which the parents wereremoved The resulting embryos were maintainedfor development by changing the seawater dailyand by continuous aeration Most larvae escapedfrom their egg capsules after about 1 week Larvaewere isolated on a 200mm sieve and transferred to0.45mm ¢ltered seawater All larvae tested in the ex-periment were released on the same day (day 0), butnot necessarily by the same female

Experimental design and larval rearing

A 5 5 factorial design with stocking densities

of 0.25, 0.50, 0.75, 1.00 and 1.50 larvae mL 1 andalgal concentrations of 5, 10, 15, 20 and 25 104cells mL 1was used in the experiment Two replicateglass beakers were used for each treatment, and in allthe experimental groups, larvae were initially reared

in 400 mL of seawater As soon as the newly hatchingveligers (0-day-larvae) were transferred to ¢lteredseawater, they were collected in pipettes and distrib-uted into ¢fty glass beakers Each day, seawater wasreplaced with fresh 0.45mm membrane-¢ltered sea-water, food was replaced with a fresh diet and bea-kers were thoroughly cleaned with deionized waterduring water changes Larval density was main-tained as in the original beakers from the beginning

to the end of the experiment by adjusting the watervolumes with daily water change Larvae were reared

at 24^25 1C, 24^26% salinity and a constant period of 12 L:12 D using a 40 W £uorescent lamp,these being optimal for larval rearing (Ke et al 2001;Zheng et al 2001)

photo-Larvae were fed a unialgal diet of Dicrateriazhanjiangensis, this enrichment ensuring the rapidgrowth of B formosae habei with low mortalityE¡ects of density and concentration on larvae H Zheng et al Aquaculture Research, 2010, 42, 1^8

r 2010 The Authors

through metamorphosis (Zheng et al 2001) Algal

concentration was microscopically measured using

a blood cell counter plate The algae was cultured in

Walne’s medium and used when the stock cultures

were in the exponential phase To avoid the

introduc-tion of algal medium into the larval containers, the

stock cultures were centrifuged and the cells were

suspended in 0.45mm membrane-¢ltered seawater

before feeding to the larvae (Lucas & Costlow 1979)

Measurement of larval growth and survival

To determine larval growth, shell lengths (longest

dimension) of 10 larvae from each replicate were

randomly measured nondestructively every other

day using a microscope equipped with an ocular

micrometer ( 64) Larvae were pipetted onto a

microscope slide in a small volume of water and the

water was quickly removed by a pipette to immobilize

the larvae, and then shell length was measured After

this, the larvae were immediately returned to their

containers The average growth rate for each

repli-cate was measured by regressing larval shell length

over time for the ¢rst 10 days of larval life for each

larva

Larval survival was expressed by a percentage of

numbers of live 10-old-day larvae to the initial

num-bers of larvae in each replicate

De¢nition of larval metamorphosis

Metamorphosis of B formosae habei larvae as

de-scribed by Zheng et al (2005) using the de¢nition of

Pechenik (1980) involves: (1) loss of the larval velum,

(2) larval behaviour changing from free swimming to

crawling, (3) shell changing from the larval

£attened-elliptic pattern to the adult spiral pattern and (4)

si-phon being extended

The time to initial metamorphosis was denoted as

developmental duration from day 0 larvae to the day

the ¢rst juvenile occurred in each replicate

Per cent metamorphosis was the ratio of the total

numbers of metamorphosed larvae to the initial

numbers of larvae in each replicate

Statistic analyses

The independent and combined e¡ects of larval

den-sity and algal concentration on per cent survival,

growth rate, time to initial metamorphosis and per

cent metamorphosis were examined byANOVA cause data on survival and metamorphosis were pre-sented as percentages of the total amount, anarcsine-transformation was performed before analy-sis All statistical analyses were performed on aSAS

Be-System for windows (SAS8.0, SAS Institute, Cary, NC,USA), and signi¢cance for all analyses was set to

Po0.05 unless noted otherwise

ResultsLarval survivalAnalyses of variance (Table 1) demonstrated that lar-val per cent survival was signi¢cantly a¡ected by theindependent e¡ects of both stocking density and al-gal concentration and by their interaction, and esti-mation of variance components further indicatedthat the e¡ect of stocking density was more impor-tant than that of algal concentration Larvae sur-vived less with increasing stocking density at thesame algal concentration and survived more with in-creasing algal concentration at the same stockingdensity Of 10-old-day larvae, 72.5% survived at thelowest larval density and the highest algal concentra-tion combination, which only survived 4.0% at thehighest stocking density and the lowest algal concen-tration combination (Table 2)

Larval growthThe average growth rates of larvae at di¡erent combi-nations of stocking density and algal concentrationare listed in Table 3 Larvae grew more slowly asthe stocking density increased regardless of thealgal concentration and with decreasing algal con-centration regardless of the stocking density Theaverage growth rate was up to 41.57mm day 1whenlarvae were cultured at the lowest density of0.25 larvae mL 1and fed the highest algal concen-tration of 25 104cells mL 1, whereas larvae onlygrew 17.06mm daily when they were cultured at thehighest density of 1.5 larvae mL 1and fed the lowestalgal concentration of 5 104

Aquaculture Research, 2010, 42, 1^8 E¡ects of density and concentration on larvae H Zheng et al.

Time to initial metamorphosis

Table 4 presents the time to initial metamorphosis for

larvae cultured at di¡erent stocking density and algal

concentration combinations Time to initial morphosis increased with increasing stocking den-sity at the same algal concentration and reducedwith increasing algal concentration at the samestocking density The time to initial metamorphosiswas only 10 days when larvae were cultured at adensity of 0.25 larvae mL 1and fed an algal con-centration of 25 104

meta-cells mL 1, whereas the time

to initial metamorphosis was 20^26 days for larvaecultured at the lowest algal concentration of

5 104cells mL 1 Stocking density and algal con-centration exerted signi¢cantly independent andcombined e¡ects on the time to initial metamorpho-sis (Table 1)

Percentage metamorphosisPer cent metamorphosis listed in Table 5 decreasedwith increasing stocking density regardless of thealgal concentration and increased with increasingalgal concentration regardless of the stockingdensity No larvae survived to metamorphosis whenthey were cultured at the highest density of1.5 larvae mL 1and fed the lowest algal concentra-tion combination of 5 104

cells mL 1, but 49.5%larvae successfully metamorphosed to juvenileswhen they were cultured at the lowest density of0.25 larvae mL 1and fed the highest algal concen-

Table 1 Analyses of variance for traits in larval survival, growth rate, time to initial metamorphosis and per cent phosis of Babylonia formosae habei larvae

Variance components (% of total)

Per cent survival a 4 1160.765856 806.36 115.01907 (77.06%)

b 4 292.964204 203.52 28.23891 (18.92%)

a  b 16 10.575150 7.35 4.56782 (3.06%) Error 25 1.439511 1.43951 (0.96%) Growth rate a 143.984948 437.41 13.91109 (27.99%)

b 336.739468 1022.98 33.18654 (66.77%)

a  b 4.874038 14.81 2.27243 (4.57%) Error 0.329176 0.32918 (0.66%) Time to initial metamorphosis a 4 34.7026495 15.39 2.98031 (16.93%)

b 4 104.9052536 46.51 9.39207 (53.36%)

a  b 14 5.3123383 2.36 2.97286 (16.89%) Error 23 2.2554348 2.25543 (12.81%) Per cent metamorphosis a 4 535.795754 1907.14 52.32758 (65.45%)

b 4 224.698804 799.80 21.21789 (26.54%)

a  b 16 12.519950 44.56 6.11950 (7.65%) Error 25 0.280943 0.28094 (0.35%)

a and b represents the stocking density (larvae mL 1

) and the algal concentration (cells mL 1

) respectively.

 o0.05; Po0.001.

df, degrees of freedom; MS, mean squares.

Table 2 Per cent survival (%) of Babylonia formosae habei

larvae at di¡erent stocking density and algal concentration

Table 3 Average growth rate (mm day 1) of Babylonia

for-mosae habei larvae at di¡erent stocking density and algal

tration of 25 104cells mL 1 Per cent

metamor-phosis of larvae was a¡ected not only by the

indepen-dent e¡ects of stocking density and algal

concentration but also by their signi¢cantly

com-bined e¡ect (Table 1)

Discussion

The independent e¡ects of stocking density

on larval survival, growth and metamorphosis

Larval culture density is an important factor

in£uen-cing the success of hatchery seed culture of molluscs

Although larger larval density may increase the yield

of hatchery-produced spat, it may result in reduced

growth and survival In the present study, stocking

density had striking e¡ects on the survival, growth

and metamorphosis of B formosae habei larvae Under

the same algal concentration conditions, a high

stock-ing density exerted negative e¡ects on larval survival,

growth and metamorphosis Such negative e¡ects of

crowding or density stress on larvae have been

re-ported in other molluscs For milk conch Strombus

costatus larvae, the survival rate for ¢ve larval

densi-ties tested (from100 to 500 larvae L 1) was17%,18%,

10%, 9% and 5% respectively (Aldana-Aranda et al.1989) Of the ¢ve densities tested, 500 larvae L 1yielded the least growth, and the most growth wasobtained with 100 larvae L 1 In the abalone Haliotisrubra, survival at 64 days followed a negative powerfunction of density at settlement (Daume et al 2003).Larvae of the nudibranch mollusk Hermissenda crassi-cornis grew 4.30mm day 1

at the minimum larvaldensity (1larvae mL 1) and 1.60mm day at the high-est larval density (15 larvae 1), and stocking densityexhibited a clear negative e¡ect on larval growth (Avi-

la et al 1997) The survival of Crassostrea gigas larvaethrough metamorphosis declined drastically whenlarval densities were increased above 1larvae mL 1,and high density reduced the survival of larvae withlow growth e⁄ciency (Powell et al 2002) Manila clamRuditapes philippinarum larvae survived more andgrew faster at 5 larvae mL 1than those at 15 and

20 larvae mL 1(Yan et al 2006) For the Californiasea hare Aplysia californica larvae, the growth ratesdeclined from approximately 16mm day 1at the low-est stocking densities to about 9mm d 1at the high-est, time to competency increased by about two-foldand survival rates declined from about 85% to 5% orless at a density ranging from 0.5 to 4 larvae mL 1(Capo et al 2009) All results exhibited an adverse re-lationship between larval survival/growth and stock-ing density Therefore, to maximize survival andgrowth, B formosae habei larvae should be cultured

at a lower density in a large-scale hatchery culture.Basch (1996) pointed out three possible e¡ects oflarval density including: (1) larvae grazing particles

to levels where detection or feeding e⁄ciency is duced, (2) physical interactions between larvae or(3) accumulation of soluble wastes, reducing thefeeding times or rates When the stocking density islower, the density-dependent behaviour in relation

re-to competition for resources such as space and food

is slight (McShane1991; Preece et al.1997), and so vae show higher survival and faster growth Whenthe stocking density is higher, crowding conditionshave a signi¢cant impact on larval survival andgrowth through chemical interactions and mechan-ical/physical interference such as collisions betweenswimming larvae (Sprung 1984; Avila et al 1997) andbecoming entangled in mucus strings (Hansen 1991).Moreover, it is di⁄cult for larvae at a high density toaccumulate energy due to competition for food, so-cial stress, and growth inhibitors, even when food isabundant (Crump 1981) Therefore, larvae have lowersurvival and slower growth when the stocking den-sity is higher

lar-Table 5 Per cent metamorphosis (%) of Babylonia formosae

habei larvae at di¡erent stocking density and algal

Table 4 Time to initial metamorphosis (day) of Babylonia

formosae habei larvae at di¡erent stocking density and algal

^, no larvae survived to metamorphosis.

Aquaculture Research, 2010, 42, 1^8 E¡ects of density and concentration on larvae H Zheng et al.

The independent e¡ects of algal

concentration on larval survival, growth and

metamorphosis

Algal concentration is another important factor

in-£uencing the success of hatchery seed culture of

mol-luscs Particularly, planktotrophic larvae survive less

and grow slower when food is limited or scarce Such

negative e¡ects of low algal concentration on larval

survival, growth and metamorphosis were also

ob-served in the present study Dicrateria zhanjiangensis

has been used as a good unialgal diet to feed B

formo-sae habei larvae, and supported rapid growth and

high survival at a high concentration of

20 104

cells mL 1 (Zheng et al 2001, 2005)

Furthermore, no detrimental e¡ect of this alga has

been reported on mollusk larvae Therefore, the

nega-tive e¡ects of low algal concentration on larvae can

be attributed to a de¢ciency in energy gain from food

The negative e¡ects of food limitation on larvae have

been documented in other molluscs (Fretter &

Mon-tgomery 1968; Pilkington & Fretter 1970;

Aldana-Aranda et al 1989; Pechenik et al 1990, 1996; His &

Seaman 1992; Strathmann, Fenaux, Sewell &

Strath-mann 1993; Avila et al 1997; Rico-Villa & Robert

2009) It is known that planktotrophic larvae feed

on phytoplankton (and possibly other organic

mate-rial in suspension) and are dependent on a net energy

gain from such food for successful growth and

devel-opment From the point of view of the individual’s

physiological energetics, the energy and nutrients

gained by the animal (in this case the larva) from

the environment are distributed among the various

metabolic requirements of maintenance, movement

and growth (Bayne 1983) The energy and nutrients

gained by the larvae from food are not able to meet

the various normal metabolic requirements when

food is scarce, and so it is inevitable that larvae

sur-vive less and grow slower Therefore, to maximize

survival and growth, B formosae habei larvae should

be fed a higher algal concentration in a large-scale

hatchery culture

Interaction of stocking density and algal

concentration on larvae

In general, the e¡ects of stocking density or algal

con-centration on the growth and survival of

plankto-trophic larvae have been studied for each factor

separately, and reports on their combined e¡ects have

been very scarce Doroudi and Southgate (2000)

found that the interaction of algal ration and larval

density did not a¡ect the growth or the survival of

7-or 20-day-old black-lip pearl oyster Pinctada tiferia larvae However, Powell et al (2002) found that

margari-a number of di¡erent combinmargari-ations of culture tions including stocking density, algal concentration,feeding frequency and food quality generated verycomplex interactions for the growth and development

condi-of Paci¢c oyster C gigas larvae In the present study,the combined e¡ects of stocking density and algalconcentration on the growth and survival of B formo-sae habei larvae were very common Larvae survivedmore, grew faster and metamorphosed earlier at alower stocking density and a higher algal concentra-tion combination, whereas larvae survived less, grewslower and metamorphosed later at a higher stockingdensity and a lower algal concentration combination.According to the present results, larvae should be cul-tured at a density of 0.25 larvae mL 1and fed an algalconcentration of 25 104

cells mL 1 in order tomaximize the survival and growth in B formosae ha-bei hatchery culture practice

Di¡erence between the e¡ects of stockingdensity and algal concentration on larvalsurvival and growth

A more interesting and important ¢nding from thepresent study is that stocking density and algal con-centration obviously exert di¡erent impacts on larvalsurvival and growth, that is, stocking density played

a more important role than algal concentration in termining larval survival, whereas algal concentra-tion played a more important role than stockingdensity in determining larval growth This resulthas been not reported in those previous studies From

de-an ecological point of view, it is reasonable to drawsuch conclusions because larval density and algalconcentration represent two di¡erent types of factors

^ spatial and nutritional Competition for space in acrowded living space with an increase in physicalcollision or interference may be a major and directfactor resulting in the lower survival of larvae,whereas an increase in physical interference in acrowded living space may be a causal or an indirectfactor resulting in slow larval growth by reducingfeeding e⁄ciency (Rasheed & Bull 1992) This can besupported by the fact that percentage survival wasvery low under higher density conditions even in thepresence of abundant food resources The energy andnutrients gained by larvae may be reallocated tomaintain survival ¢rst rather than growth whenE¡ects of density and concentration on larvae H Zheng et al Aquaculture Research, 2010, 42, 1^8

r 2010 The Authors

food is scarce (Pechenik et al 1996) Therefore,

com-petition for nutrition during food limitation may be

the principal factor resulting in slow larval growth

This can be supported by larvae surviving longer,

combined with very slow growth, at a low food

con-centration in the present study or when fully

de-prived of food (Zheng et al 2005)

In conclusion, both stocking density and algal

con-centration independently had signi¢cant impacts on

the survival and growth of B formosae habei larvae;

their combined e¡ects were also signi¢cant Larvae

survived more, grew faster and metamorphosed

ear-lier at a lower stocking density and a higher algal

concentration combination, whereas larvae survived

less, grew slower and metamorphosed later at a

high-er stocking density and a lowhigh-er algal concentration

combination In a large-scale hatchery seed culture

of B formosae habei, larvae should be reared at a

low-er stocking density of 0.25 larvae mL 1and fed a

higher algal concentration of 25 104

cells mL 1inorder to maximize larval survival and growth

Acknowledgments

We thank Dr Haihui Ye and Donghui Guo for their

kind help in this experiment.We also thank Ms Kirsty

A Mattinson (Cantab MA) and Professor I J

Hodg-kiss for helping to revise the manuscript This work

was supported in part by the Earmarked Fund for

Modern Agro-industry Technology Research System

(No nycytx-47) and Research Project of Technical

Exploitation of Fujian Province (No.: 98 Z 8)

References

Aldana-Aranda D., Lucas A., Brule T., Salguero E & Rendon

F (1989) E¡ects of temperature, algal food, feeding rate

and density on the larval growth of the milk conch

(Strombus costatus) in Mexico Aquaculture 76, 361^371.

Altena C.O., Regteren V & Gittenberger E (1981) The genus

Babylonia (Prosobranchia, Buccinidae) Zoologische

Ver-handelingen 188, 1^57.

Avila C., Grenier S.,Tamse C.T & Kuzirian A.M (1997)

Biolo-gical factors a¡ecting larval growth in the nudibranch

mollusc Hermissenda crassicornis (Eschscholtz,1831)

Jour-nal of Experimental Marine Biology and Ecology 218, 243^

262.

Basch L.V (1996) E¡ects of algal and larval densities on

de-velopment and survival of asteroid larvae Marine Biology

126, 693^701.

Bayne B.L (1983) Physiological ecology of marine

mollus-can larvae In: The Mollusca Development 3 (ed by K.M.

Wilbur), pp 299^343 Academic Press, New York, NY, USA.

Capo T.R., Bardales A.T., Gillette P.R., Lara M.R., Schmale M.C & Serafy J.E (2009) Larval growth, development, and survival of laboratory-reared Aplysia californica: ef- fects of diet and veliger density Comparative Biochemistry and Physiology Part C 149, 215^223.

Crisp D.J (1974) Factors in£uencing the settlement of marine invertebrate larvae In: Chemoreception in Marine Organ- isms (ed by P.T Grant & A.M Mackie), pp 177^263 Aca- demic Press, NewYork, NY, USA.

Crump M.L (1981) Energy accumulation and amphibian metamorphosis Oecologia 49, 167^169.

Daume S., Huchette S., Ryan S & Day R.W (2003) Nursery culture of Haliotis rubra: the e¡ect of cultured algae and larval density on settlement and juvenile production Aquaculture 236, 221^239.

Doroudi M.S & Southgate P.C (2000) The in£uence of algal ration and larval density on growth and survival of black- lip pearl oyster Pinctada margaritifera (L.) larvae Aquacul- ture Research 31, 621^626.

Fretter V & Montgomery M.C (1968) The treatment of food by some prosobranch veligers Journal of the Marine Biological Association of the United Kingdom 48, 499^520.

Hansen B (1991) Feeding behaviour in larvae of the branch Philine aperta II Food size spectra and particle se- lectivity in relation to larval behaviour and morphology of the velar structures Marine Biology 111, 263^270 His E & Seaman M.N.L (1992) E¡ects of temporary starva- tion on the survival, and on subsequent feeding and growth, of oyster (Crassostrea gigas) larvae Marine Biol- ogy 114, 277^279.

opistho-Ke C.H., Zheng H.P, Zhu J.X., Zhou S.Q & Li F.X (2001) E¡ects

of temperature on survival, growth and metamorphosis

of larvae Babylonia formosae habei (Gastropoda: dae) Transactions of the Chinese Society of Malacology 9, 70–76 (in Chinese with English abstract).

Buccini-Liu B.Z., Dong B.,Tang B.J., Zhang T & Xiang J.H (2006) fect of stocking density on growth, settlement and survi- val of clam larvae, Meretrix meretrix Aquaculture 258, 344^349.

Ef-Liu D.J & Xiao S (1998) Preliminary study on ecology of Babylonia formosae Journal of Fisheries Science of China 5, 93^96 (in Chinese).

Lucas J.S & Costlow J.D (1979) E¡ects of various perature cycles on the larval development of the gastro- pod mollusca Crepidula fornicata Marine Biology 51, 111^117.

tem-Mazo¤n-SuaŁstegui J.M., Ru|¤z-Ru|¤z K.M., Parres-Haro A & Saucedo P.E (2008) Combined e¡ects of diet and stocking density on growth and biochemical composition of spat of the Cortez oyster Crassostrea corteziensis at the hatchery Aquaculture 284, 98^105.

McShane P.E (1991) Density-dependent mortality of recruits

of the abalone Haliotis rubra (Mollusca: Gastropoda) ine Biology 110, 385^389.

Mar-Aquaculture Research, 2010, 42, 1^8 E¡ects of density and concentration on larvae H Zheng et al.

Pechenik J.A (1980) Growth and energy balance during the

larval lives of three prosobranch gastropods Journal of

Experimental Marine Biology and Ecology 44, 1^28.

Pechenik J.A., Eyster L.S., Widdows J & Bayne B.L (1990)

The in£uence of food concentration and temperature on

growth and morphological di¡erentiation of blue mussel

Mytilus edulis L larvae Journal of Experimental Marine

Biology and Ecology 136, 47^64.

Pechenik J.A., Estrella M.S & Hammer K (1996) Food

limita-tion stimulates metamorphosis of competent larvae and

alters postmetamorphic growth rate in the marine

proso-branch gastropod Crepidula fornicata Marine Biology 127,

267^275.

Pechenik J.A., Jarrett J.N & Rooney J (2002) Relationships

between larval nutritional experience, larval growth

rates, juvenile growth rates, and juvenile feeding rates in

the prosobranch gastropod Crepidula fornicata Journal of

Experimental Marine Biology and Ecology 280, 63^78.

Perron F.E & Turner R.D (1977) Development,

metamor-phosis, and natural history of the nudibranch Doridella

obscura Verrill (Corambidae: Opisthobranchia) Journal of

Experimental Marine Biology and Ecology 27, 171^185.

Pilkington M.C & FretterV (1970) Some factors a¡ecting the

growth of prosobranch veligers Helgolnder wiss

Meere-sunters 20, 576^593.

Powell E.N., Bochenek E.A., John M., Klinck J.M & Hofmann

E.E (2002) In£uence of food quality and quantity on the

growth and development of Crassostrea gigas larvae: a

modeling approach Aquaculture 210, 1^4.

Preece P.A., Shepherd S.A., Clarke S.M & Keesing J.K (1997)

Abalone stock enhancement by larval seeding: e¡ect of

larval density on settlement and survival Journal of

Mol-luscan Research 18, 265^273.

Raghavan G & Gopinathan C.P (2008) E¡ects of diet,

stock-ing density and environmental factors on growth,

survi-val and metamorphosis of clam, Paphia malabarica (Chemnitz) larvae Aquaculture Research 39, 928^933 Rasheed M.A & Bull C.M (1992) Behaviour of the western king prawn, Penaeus latisulcatus Kishinouye: e¡ect of food dispersion and crowding Australian Journal of Marine & Freshwater research 43,745^752.

Rico-Villa B & Robert S.P (2009) In£uence of food density and temperature on ingestion, growth and settlement of Paci¢c oyster larvae, Crassostrea gigas Aquaculture 287, 395^401.

Sprung M (1984) Physiological energetics of mussel larvae (Mytilus edulis) I Shell growth and biomass Marine Ecol- ogy Progress Series 17, 283^293.

Strathmann R.R., Fenaux L., Sewell A.T & Strathmann M.F (1993) Abundance of food a¡ects relative size of larval and postlarval structures of a molluscan veliger Reference: Biological Bulletin 185, 232^239.

Yan X.W., Zhang G.F & Yang F (2006) E¡ects of diet, stocking density, and environmental factors on growth, survival, and metamorphosis of Manila clam Ruditapes philippinar-

um larvae Aquaculture 253, 350^358.

Zhao B., Qiu J.W & Qian P.Y (2003) E¡ects of food availability

on larval development in the slipper limpet Crepidula onyx (Sowerby) Journal of Experimental Marine Biology and Ecology 294, 219^233.

Zheng H.P, Zhu J.X., Ke C.H., Zhou S.Q & Li F.X (2001) E¡ects

of three microalgae on survival, growth and phosis of larvae Babylonia formosae habei (Gastropoda: Buccinidae) Transactions of the Chinese Society of Malacol- ogy 9, 77–84 (in Chinese with English abstract) Zheng H.P., Ke C.H., Zhou S.Q & Li F.X (2005) E¡ects of starvation on larval growth, survival and metamor- phosis of Ivory shell Babylonia formosae habei Altena

metamor-et al., 1981 (Neogastropoda: Buccinidae) Aquaculture

243, 357^366.

E¡ects of density and concentration on larvae H Zheng et al Aquaculture Research, 2010, 42, 1^8

r 2010 The Authors

Growth and survival of juvenile lined seahorse,

densities

Dong Zhang1,2, Yinghui Zhang2, Junda Lin2& Qiang Lin2,3

1 East China Sea Fisheries Research Institute, Chinese Academy of Fisheries Sciences, Shanghai, China

2 Vero Beach Marine Laboratory, Florida Institute of Technology,Vero Beach, FL, USA

3 Key Laboratory of Tropical Marine Environmental Dynamics, South China Sea Institute of Oceanology, Chinese Academy of Sciences, Guangzhou, China

Correspondence: D Zhang, East China Sea Fisheries Research Institute, Chinese Academy of Fisheries Sciences, Shanghai 200090, China E-mail: zd_¢t@hotmail.com

Abstract

The lined seahorse, Hippocampus erectus (Perry), is an

important species in both medicinal and aquarium

trades The aim of this study was to evaluate the

ef-fects of stocking density (1, 3 and 5 individuals L 1)

on the growth performance and survival of the

early-stage juvenile H erectus The height (HT), wet

weight, weight gain (WG) and speci¢c growth rate

(SGR) were a¡ected signi¢cantly by the stocking

den-sity during the 40-day study The HT,WG and SGR of

the seahorse at 1 and 3 juveniles L 1were

signi¢-cantly higher than that at 5 juveniles L 1 The

survi-val of juveniles at the three stocking densities was

not signi¢cantly di¡erent at day 25 (90.3 4.5%,

86.7 4.2% and 86.2 3.8% for 1, 3 and

5 juveniles L 1 respectively), but was signi¢cantly

di¡erent at day 40 (87.8 3.9%, 69.6  4.2% and

52.9 2.8% for 1, 3 and 5 juveniles L 1

respectively)

For the early-stage juvenile H erectus, we recommend a

stocking density of 3 juveniles L 1, but the density

should be reduced to1^2 juveniles L 1to avoid reduced

and variable growth and high mortality after 25 days

Keywords: Hippocampus erectus (Perry), density,

growth, survival, seahorse

Introduction

Although there have been attempts at commercial

culturing of seahorses for over 50 years (Zou 1958),

signi¢cant breakthroughs in culture techniqueshave only occurred in the last 10 years This has inpart been the result of increasing culture e¡orts sinceall 33 recognized seahorse species were listed on Ap-pendix II of the Convention on International Trade inEndangered Species of Wild Fauna and Flora (CITES2004) due to overexploitation of the wild populations

to meet the growing demand in Chinese medicineand ornamental market (Vincent 1996; Lourie, Vin-cent & Hall1999) More than10 seahorse species havebeen reared successfully in captivity, such asHippocampus abdominalis (Woods 2000a, b; 2005),Hippocampus comes (Job, Buu & Vincent 2006), Hippo-campus erectus (Correa, Chung & Manrique 1989;Scarratt 1995; Lin, Lin & Zhang 2008), Hippocampuskuda (Job, Do & Hall 2002; Lin, Lu & Gao 2006; Lin,Gao, Sheng, Chen, Zhang & Lu 2007), Hippocampusreidi (Olivotto, Avella, Sampaolesi, Piccinetti, Ruiz &Carnevali 2008), Hippocampus subelongatus (Payne &Rippingale 2000) and Hippocampus trimaculatus(Sheng, Lin, Chen, Gao, Shen & Lu 2006) Researche¡orts have focused on the e¡ects of food type andfeed regimes and various environmental factors, in-cluding temperature, salinity, light intensity andphotoperiods, on the growth and survival of juvenileseahorses to establish appropriate rearing protocols(e.g Payne & Rippingale 2000; Woods 2000b, 2005;Lin et al 2006, 2008; Olivotto et al 2008) However,the e¡ect of stocking density on the growth and sur-vival of seahorses in the early juvenile stages has re-ceived much less attention, although it is recognized

as a common factor a¡ecting growth and survival in

¢n¢sh aquaculture (e.g Wallace, Kolbeinshaven &

Reinsnes 1988; Christianssen, Svendsen & Jobling

1992; Bj˛rnsson 1994; Hossain, Beveridge & Haylor

1998; Feldlite & Milstein 1999; Irwin, O’Halloran &

FitzGerald 1999; Salas-Leiton, Anguis, Manchado &

Canavate 2008)

In commercial seahorse culture, low survival,

par-ticularly in the early juvenile stages (Scarratt 1995;

Forteath 1996), is still one of the bottlenecks a¡ecting

the economic return Although high survival of

sev-eral seahorse species in the early juvenile stages has

been reported, e.g H abdominalis (Woods 2000b),

H erectus (Lin et al 2008) and H comes (Job et al

2006), these have been obtained at stocking densities

around or lower than1individual L 1, which may be

too low for an economically viable commercial mass

production

The lined seahorse, H erectus (Perry), is a highly

valued species in both medicinal and aquarium

trades (Correa et al 1989; Scarratt 1995; Lourie et al

1999; Foster, Marsden & Vincent 2003) It is

distribu-ted from Nova Scotia along the western Atlantic

coast through the Gulf of Mexico and Caribbean to

Venezuela (Lourie et al 1999; Fritzche & Vincent

2002; Foster & Vincent 2004) It has been reared

suc-cessfully in captivity, and is recognized as a good

can-didate for commercial aquaculture (Correa et al.1989;

Lin et al 2008) In this study, we investigated the

growth performance and survival of H erectus at

dif-ferent stocking densities during the early juvenile

stage in order to further develop successful culture

protocols

Materials and methods

Broodstock seahorses

Twenty (M:F) pairs of F2 seahorses [16^20 cm in

height (HT), the vertical distance from the tip of the

coronet to the tip of the outstretched tail, with the

head held at right angles to the body (Lourie et al

1999)] were used as the broodstock and maintained

in the £ow-through tanks (90 80  60 cm) at the

Vero Beach Marine Laboratory, Vero Beach, FL, USA,

with sand-¢ltered seawater pumped directly from

the Atlantic Ocean at a rate of 0.5^0.6 L min 1 The

Salinity, temperature, light intensity and photoperiod

were 35%, 24.0 0.5 1C (mean  SD, the same

for-mat throughout this paper), 1000 lx and 14 h L/10 h

D respectively Plastic plants and corallites were used

as the holdfasts and substrate for the seahorses The

broodstock seahorses were fed twice a day (09:00and 16:00 hours) with frozen mysis (HikariTM, HikariSales USA, Hayward, CA, USA) at a daily rate of ap-proximately 15% wet body weight, and faeces anduneaten food in the tanks were siphoned out 2 h aftereach feeding Upon pregnancy being noted, broodingmales were isolated temporarily in a separate £ow-through hatching tank (50 25  30 cm) contain-ing 26 L of seawater with the same environmentalconditions as those for broodstock tanks

Experimental designThree stocking densities, 1, 3 and 5 individuals L 1,each with three replicate tanks, were tested Trans-parent £ow-through glass tanks (50 25  30 cm)were used Each tank contained 30 L seawater with

a £ow rate of 0.2 L min 1and gentle aeration day post-hatch juveniles [1.11 0.02 cm HT, meanwet weight (WW) 5 2.4 0.09 mg, n 515] from fourmale broodstock seahorses were haphazardly allo-cated to the tanks Plastic plants were used as hold-fasts for the seahorses Juveniles were fed withnewly hatched Artemia sinica (350^400mm inlength) (Bohai strain, China) at approximately

One-5 nauplii mL 1during the ¢rst week, followed by day-old Artemia (500^800mm in length) enrichedwith Chlorella sp (newly hatched Artemia at approxi-mately 30 nauplii mL 1were cultured in 50 L tankswith 50 000^60 000 cells of Chlorella sp per ml) atthe same food density Juveniles were fed at 08:00,14:00 and 20:00 hours each day to maintain the fooddensity Before each feeding, the bottom of the tankswas siphoned to remove faeces and dead food Thesalinity, temperature, light intensity and photoperiodwere 34^35%, 23 0.5 1C, 1000 lx and 14 L/10 D,respectively, throughout the experimental period.Ten seahorses were haphazardly selected from eachtank every 5 days for measuring the HT and returned

2-to their respective tanks Because HT in

5 juveniles L 1was signi¢cantly lower than that in

1 and 3 juveniles L 1at day 25 (see Results), all thejuvenile seahorses were counted and 10 seahorsesfrom each tank were also weighed The experimentcontinued and was terminated when the di¡erence

in HT was signi¢cant between the treatments of 1and 3 juveniles L 1 At the termination, all the juve-nile seahorses were counted and 10 seahorses fromeach tank were weighed The weight gain[WG 5100 (¢nal body weight initial bodyweight)/initial body weight (g)], the speci¢c growthGrowth of seahorse at di¡erent densities D Zhang et al Aquaculture Research, 2010, 42, 9^13

r 2010 The Authors

rate [SGR 5100 (ln ¢nalWW ln initialWW)/time]

and the condition factor [(CF 5100 WW(g)/HT

(cm3)] of the juveniles were calculated Dead

sea-horses, if any, were removed upon daily inspection

and recorded

Statistics

One-way analysis of variance (ANOVA)was used to

com-pare the ¢nal HT andWW,WG, SGR and CF of the

juve-niles among the stocking densities All the variables

were tested for normality and homogeneity before

one-wayANOVA If there was a signi¢cant di¡erence

among the stocking densities, a Welsch-up multiple

procedure was applied to compare the di¡erent means

among the densities The di¡erence in the percentage

survival among the stocking densities was tested

using Kruskal^WallisANOVA(Sokal & Rohlf 1995)

Results

The stocking density signi¢cantly a¡ected HT

(one-wayANOVA, F2, 8953.963, P 5 0.023), WW (one-way

ANOVA, F2, 8953.147, P 5 0.047), WG (one-wayANOVA,

F2, 8955.745, P 5 0.040) and SGR (one-wayANOVA,

F2, 857.633, P 5 0.023) of the juvenile seahorses after

25 days (Table 1) A multiple comparison test indicate

that the juveniles at 1 and 3 individuals L 1grew

sig-ni¢cantly (Po0.05) faster in HT than those at 5

in-dividuals L 1(Table 1) WG and SGR of the juveniles

at 1 and 3 individuals L 1were also signi¢cantly

higher (Po0.05) than that at 5 individuals L 1

ble 1) HT,WG and SGR were not signi¢cantly di¡erent

(Ta-(P40.05) between 1 and 3 individuals L 1(Table 1)

There was no signi¢cant di¡erence in the survival

(Kruskal^Wallis ANOVA, H 51.415, P40.05) and CF

(one-wayANOVA, F2, 895 0.179, P 5 0.837) among thestocking densities (Table 1)

At day 40, HT (one-way ANOVA, F2, 89511.520,

Po0.0001), WW (one-way ANOVA, F2, 8957.152,

Po0.001), WG (one-way ANOVA, F2, 855.790,

P 5 0.039) and SGR (one-way ANOVA, F2, 857.354,

P 5 0.024) of the juvenile seahorses were still cantly di¡erent among the treatments (Table 1).Moreover, for the four measurements, the di¡erencebetween 1 and 3 individuals L 1was also signi¢cant(Po0.05) (Table 1) Survival of juvenile seahorsesamong the treatments was signi¢cantly di¡erent(Kruskal^WallisANOVA, H 5 4.153, Po0.05) after 40days (Table 1)

signi¢-Discussion

It is well known that stocking density can a¡ectgrowth and survival in ¢sh aquaculture; generally,higher densities result in a lower ¢sh growth rate(e.g Wallace et al 1988; Haylor 1991; Christianssen

et al.1992; Bj˛rnsson1994; Hossain et al.1998; Feldlite

& Milstein 1999; Irwin et al 1999; Salas-Leiton et al.2008) The results presented here indicate the samepattern of the negative e¡ect of increasing stockingdensity on the growth of early-stage juvenile H erec-tus, i.e the growth parameters (HT, WG and SRG)

of the seahorses at the highest density (5 dividuals L 1) were signi¢cantly reduced (Table 1)

in-A similar e¡ect has also been observed in late-stagejuvenile H abdominalis (Woods 2003) Anothere¡ect frequently associated with the stockingdensity is the high size variations within the rearedgroup of ¢sh (Irwin et al 1999; Lambert & Dutil2001) Perhaps due to the short experimental time,size heterogeneity was not signi¢cant in this study(Table 1)

Table 1 Final height (HT, mm), wet weight (WW, mg), weight gain (WG, %), condition factor (CF), speci¢c growth rate (SGR,

%day 1) and survival (%) of juvenile Hippocampus erectus at stocking densities of 1, 3 and 5 individuals L 1over the 25 and

3 3.31  0.33 103.1  20.4 86  7 0.28  0.04 4.07  0.21 69.6  4.2

5 3.16  0.38 82.4  19.7 72  11 0.27  0.04 3.76  0.24 52.9  2.8

Final HT, WW, WG and SGR at 1, 3 individuals L 1

were signi¢cantly higher than that at 5 individuals L 1

An inverse relationship between survivorship

and stocking density has been reported for many ¢sh

species (e.g Wallace et al 1988; Haylor 1991;

Chris-tianssen et al 1992; Bj˛rnsson 1994; Hossain et al

1998; Feldlite & Milstein 1999; Irwin et al 1999;

Salas-Leiton et al 2008) The same e¡ect was also

found in the seahorse, H abdominalis (Woods 2003)

Probably due to the short experimental time of this

study, di¡erences in survival between the stocking

densities were not signi¢cant, although survival at

higher densities tended to be lower (Table 1) In

sea-horses, conspeci¢c grasping and wrestling with each

other is considered to a¡ect juvenile growth and

sur-vival at higher stocking densities where the

inci-dence of such behaviour increased (Woods 2003)

Similar behavioural interactions were also observed

in the early-stage juveniles in this study Although

we did not quantify this behaviour accurately, the

in-cidence of the behaviour was the highest at the

high-est stocking density

Compared with other ¢sh species, seahorses

ap-pear not to be suitable for culture in high density

This may be a consequence of their habit of using

their tails to anchor themselves, which can cause

in-terference among conspeci¢cs at higher densities To

date, instances of all the high survival in cultured

seahorses have been found at low stocking densities

of around oro1individual L 1

(e.g.Woods 2000a, b;

Job et al 2006; Lin et al 2008) Five individuals per

li-tre for an early-stage seahorse may be the

high-den-sity limit For example, low survival of 20^40% over

21 days at 5 juveniles L 1 was found for

H reidi (Olivotto et al 2008) Survival of the H erectus

at a stocking density of 6 juveniles L 1over 35 days

was only 50.67% (Correa et al 1989), similar to what

we found over the 40 days for the 5 juveniles L 1

treatment in the present study

The strategy of commercial ¢sh aquaculture is to

maximize the economic return in relation to culture

in-vestment High stocking density is a common strategy

to reduce the overall production cost and to improve

¢sh health, growth and survival A trade-o¡ between

growth/survival and stocking density is essential for

the successful mass culture of seahorses, especially for

medicinal purposes Based on the results of this study,

we recommend a stocking density of 3 juveniles L 1for

the early-stage H erectus Because of signi¢cant growth

reduction and lower survival of juvenile seahorses after

25 days at higher densities (i.e 3, 5 juveniles L 1), the

density should be reduced to 1^2 juveniles L 1 (D

Zhang & J Lin unpubl obs.) to avoid reduced and

vari-able growth and high mortality after 25 days

AcknowledgmentsThis study was supported by a special research fundfor the national nonpro¢t institutes (East China SeaFisheries Research Institute)

ReferencesBj˛rnsson B (1994) E¡ects of stocking density on growth rate of halibut (Hippoglossus hippoglossus) reared in large circular tanks for three years Aquaculture 123, 259^270 Christianssen J.S., Svendsen Y.S & Jobling M (1992) The combined e¡ects of stocking density and sustained exer- cise on the behaviour, food intake and growth of juvenile Arctic charr (Salvelinus alpinus L.) Canadian Journal of Zoology 70, 115^122.

CITES (2004) Seahorses and other members of the family nathidae (decision 12.54) Report of the Working Group AC20 Doc 17 Convention on International Trade in En- dangered Species of Wild Fauna and Flora (CITES),Twen- tieth meeting of the Animals Committee Johannesburg (South Africa), 29 March^2 April 2004.

Syng-Correa M., Chung K.S & Manrique R (1989) Cultive mental del caballito de mar, Hippocampus erectus Boletin

experi-de Instituto OceanograŁ¢co experi-deVenezuela 28,191^196 Feldlite M & Milstein A (1999) E¡ect of density on survival and growth of cyprinid ¢sh fry Aquaculture International

Foster S.J., Marsden A.D & Vincent A.C.J (2003) pus erectus IUCN 2004 2004 IUCN Red List of Threa- tened Species Available at http://www.Redlist.org (accessed 2 April 2004).

Hippocam-Fritzche R.A & Vincent A.C.J (2002) Syngnathidae In: The living marine resources of the western central Atlantic Vo- lume 2 Bony Fishes part 1 (Ascipenseridae to Grammatidae) (ed by K.E Carpenter), pp 1221–1225 Food and Agriculture Organization of the United Nations, Rome, Italy.

Haylor G.S (1991) Controlled hatchery production of Clarias gariepinus (Burchell 1822): growth and survival of fry at high density Aquaculture and Fisheries Management 22, 405^422.

Hossain M.A.R., Beveridge M.C.M & Haylor G.S (1998) The e¡ects of stocking density, light and shelter on the growth and survival of African cat¢sh (Clarias gariepinus Burch- ell, 1822) ¢ngerlings Aquaculture 160, 251^258 Irwin S., O’Halloran J.O & FitzGerald R.D (1999) Stocking density, growth and growth variation in juvenile turbot, Scophthalmus maximus (Ra¢nesque) Aquaculture 178, 77^88.

Growth of seahorse at di¡erent densities D Zhang et al Aquaculture Research, 2010, 42, 9^13

r 2010 The Authors

Job S., Buu D & Vincent A (2006) Growth and survival of

the tiger tail seahorse, Hippocampus comes Journal of

World Aquaculture Society 37, 322^327.

Job S.D., Do H.H & Hall H.J (2002) Culturing the oceanic

seahorse Hippocampus kuda Aquaculture 214, 333^341.

Lambert Y & Dutil J (2001) Food intake and growth of adult

Atlantic cod (Gadus morhua L.) reared under di¡erent

con-ditions of stocking density, feeding frequency and size

grading Aquaculture 192, 233^247.

Lin Q., Lu J.Y & Gao Y.L (2006) The e¡ect of temperature on

gonad, embryonic development and survival rate of

juve-nile seahorses, Hippocampus kuda Bleeker Aquaculture

254,701^713.

Lin Q., Gao Y.L., Sheng J.Q., Chen Q.X., Zhang B & Lu J.Y.

(2007) The e¡ect of food and the sum of e¡ective

temperature on the embryonic development of the

seahorse, Hippocampus kuda Bleeker Aquaculture 262,

481^492.

Lin Q., Lin J & Zhang D (2008) Breeding and juvenile

cul-ture of the lined seahorse, Hippocampus erectus

Aquacul-ture 277, 287^292.

Lourie S.A.,Vincent A.C & Hall H.J (1999) Seahorse: An

Iden-ti¢cation Guide to theWorld’s Species and their Conservation.

Project Seahorse, London, UK, 214pp.

Olivotto I., Avella M.A., Sampaolesi G., Piccinetti C.C., Ruiz

P.N & Carnevali O (2008) Breeding and rearing the

long-snout seahorse Hippocampus reidi: rearing and feeding

studies Aquaculture 283, 92^96.

Payne M.F & Rippingale R.J (2000) RearingWest Australian

seahorse, Hippocampus subelongatus, juveniles on

cope-pod nauplii and enriched Artemia Aquaculture 188, 353^

361.

Salas-Leiton E., Anguis V., Manchado M & Canavate J.P (2008) Growth, feeding and oxygen consumption of Se- negalese sole (Solea senegalensis) juveniles stocked at dif- ferent densities Aquaculture 285, 84^89.

Scarratt A.M (1995) Techniques for raising lined seahorses (Hippocampus erectus) Aquarium Frontier 3, 24^29 Sheng J.Q., Lin Q., Chen Q.X., GaoY.L., Shen L & Lu J.Y (2006) E¡ects of food, temperature and light intensity on the feeding behavior of three-spot juveniles, Hippocampus tri- maculatus Leach Aquaculture 256, 596^607.

Sokal R.R & Rohlf F.J (1995) Biometry W H Freeman, New York, NY, USA, 887pp.

Vincent A.C.J (1996) The International Trade in Seahorse TRAFFIC International, Cambridge, UK 163pp.

Wallace J.C., Kolbeinshaven A.G & Reinsnes T.G (1988) The e¡ects of stocking density on early growth in Artic charr, Salvelinus alpinus (L.) Aquaculture 73, 101^110.

Woods C.M.C (2000a) Preliminary observations on ing and rearing the seahorse, Hippocampus abdominalis (Teleostei: Syngnathidae) in captivity New Zealand Journal

breed-of Marine and Freshwater Research 34, 475^485.

Woods C.M.C (2000b) Improving initial survival in cultured seahorse, Hippocampus abdominalis Lesson, 1827 (Teleos- tei: Syngathidae) Aquaculture 190, 377^388.

Woods C.M.C (2003) E¡ect of stocking density and gender segregation on the seahorse Hippocampus abdominalis Aquaculture 218, 167^176.

Woods C.M.C (2005) Growth of cultured seahorses campus abdominalis) in relation to feed ration Aquaculture International 13, 305^314.

(Hippo-Zou Z.H (1958) Rearing seahorse China Fisheries, 13 (in Chinese).

Aquaculture Research, 2010, 42, 9^13 Growth of seahorse at di¡erent densities D Zhang et al.

Sedimentation and sediment characteristics in

culture ponds

Yichao Ren, Shuanglin Dong, Fang Wang, Qinfeng Gao, Xiangli Tian & Feng Liu

Key Laboratory of Mariculture, Ministry of Education, Ocean University of China, Qingdao, China

Correspondence: S Dong, Key Laboratory of Mariculture, Ministry of Education, Ocean University of China, Qingdao 266100, China E-mail: dongsl@ouc.edu.cn

Abstract

Annual sedimentation, re-suspension rates and

con-tents of particulate organic carbon (POC), particulate

organic nitrogen (PON) and total phosphorus (TP) in

the sediment were investigated in two sea cucumber

culture ponds at Rongcheng, Shandong Province,

China The results showed that the average £ux of

to-tal particulate matter in the ponds was 22.1g m2d 1

The average re-suspension rate of the sediment in the

ponds was 81.7% The re-suspension rates in spring

and autumn were higher than those in summer and

winter The mean contents of POC, PON and TP in the

sediment of the ponds were 4.4, 0.5 and 0.6 (mg g 1),

respectively, and the mean contents of Chlorophyll a

(Chl a) and pheophytin in the sediment were 8.1 and

12.1mg g 1respectively The POC, PON and TP

con-tents in the sediment of the ponds increased during

the period of sea cucumber aestivation (summer)

and hibernation (winter), while they decreased

dur-ing the feeddur-ing periods The organic matter

accumu-lation rate and the contents of POC, PON, TP, Chl a

and pheophytin in the sediment were even lower

than those in the pond without sea cucumber

(Po0.05) The results demonstrated that sea

cucum-ber culture can e¡ectively stop nutrient

accumula-tion at the bottom of the cucumber culture ponds

Keywords: sea cucumber, ponds, sedimentation,

sediment, chlorophyll, pheophytin, POC, PON,TP

Introduction

Particular organic matter (POM) sedimentation and

re-suspension play an important role in the material

and energy transformation from primary producers

to benthic consumers in aquatic ecosystems, and theprocess of pelagic^benthic coupling has a consider-able in£uence on the biotic community in inter-tidalzones and shallow seas (Danovaro, Croce, Dell’Anno,Fabiano, Marrale & Martorano 2000) Re-suspensionnutrients restores to the overlying water and has afeedback e¡ect on the vertical £ux and the ecosystemstructure in shallow sea (Sun & Zhan 2002).The quality of settling particulate matter depends

on the seasonal variations in phytoplankton biomassand community composition (Josefson & Rasmussen2000) as well as the re-suspension process, and it canin£uence, in turn, the growth of aquatic invertebrates.The £ux and characteristic of POM are di¡erent indi¡erent waters due to various bio-productivity andhydrological conditions There have been several re-ports on the £ux and sources of total particulate ma-terial (TPM) in the ocean (Honjo, Manganini & Wefer1988; Sasaki, Hattori & Nishizawa 1988; Lohrenz,Knauer, Asper & Tuel 1992; Wassmann 1993; Kawa-hata 2002), and a few studies have dealt with bio-deposition of scallop culture (Zhou, Yang, Liu, Yuan,Mao, Liu, Xu & Zhang 2006) and ¢sh cage culture(Sutherland, Martin & Levings 2001; Cromey, Nickell

& Black 2002) in coastal areas

The sea cucumber Apostichopus japonicus (Selenka)

is a commercial species in the coasts of Asia (Liao1997) and is cultured in coastal ponds in north China.Sea cucumber, being a deposit feeder, ingests largeamounts of sedimentary material, absorbing nutritionfrom it (Yingst 1976) Particular organic matter in thewater column is a potentially available food source forsea cucumber after sedimentation However, quantita-tive studies of its sedimentation and re-suspension

r 2010 The Authors

rates are scarce in ponds The sea cucumber culture

pond is a shallow, semi-closed water body, with large

deposit feeder (sea cucumber) in it Therefore, its POM

sedimentation and re-suspension characteristics

should be quite di¡erent from those of other waters

Particular organic matter in the water column is

not only the potentially available food source for sea

cucumber but it can also a¡ect sediment quality and

bottom environment; therefore, for sea cucumber

culture it is very important to know sedimentation,

re-suspension and its e¡ect on sediment The present

study was conducted to investigate mainly the

sedi-mentation £ux and the sediment characteristics of

the sea cucumber culture ponds in order to gain an

insight into the pool of organic matter (OM) and the

e¡ect of sea cucumber culture on the sediment

Materials and methods

Study ponds

The two ponds investigated (Pond 1 and Pond 2),

ap-proximately 2 ha (100 200  2 m) each, are

lo-cated at Rongcheng, Shandong Province, China The

seawater in the ponds was routinely exchanged

ing the spring tide, while the sluices were closed

dur-ing the neap tide Juvenile sea cucumbers of about

5.0 g ind 1were stocked at a density of 10 ind m 2

in April 2007 No supplemental food was provided

during the sea cucumber culture period

Sampling and measurement

Straight-sided cylindrical traps with an aspect ratio of

4 (White 1990) were used for collecting sinking

parti-cles A net, approximately 0.8 cm in mesh, was used to

cover the upper end of the trap to prevent large nektons

from entering Ten sides were selected to set up the

traps evenly in the pond At each site, vertical sediment

traps with three replicates were installed on the

sedi-ment The traps were deployed for 10 days and taken

back to a lab for analysis Overlying water of the traps

was carefully removed using a siphon after standing

for 12 h Distilled water was used to rinse the salt of

the samples The dry weight of the total sediment

de-position was obtained after drying the samples at

60 1C to a constant weight The trapped material was

analysed for nutrient contents Particulate organic

carbon (POC) and particulate organic nitrogen (PON)

were analysed using a PE-24 CHN analyzer (Heraecus,

Banau, Germany) after acidi¢cation with 0.1 N HCl to

remove carbonate (Fabiano, Danovaro & Fraschetti1995) The total phosphorous (TP) content was mea-sured according to Mudroch, Azcue and Mudroch(1997) Samples for Chlorophyll a (Chl a) and pheo-phytin were analysed according to Lorenzen and Jef-frey (1980)

Surface samples of sediment (0^1cm) were collected

at monthly intervals using aWuttke standard box corerbeside the sediment traps (three replicates in each site)

A series of sediment samples for comparison were alsocollected from an adjacent pond where no organismwas cultured The samples were processed within 3 h

of collection (Gerino, Stora, Poydenot & Bourcier1995).The methods to determine POC, PON, TP, Chl a andpheophytin of the sediment samples were the same asthose for the settling particles Sediment samples werestored at 20 1C before analytical treatment.Water samples were collected above every sedimenttrap using 2.5 L Van Doorn bottles for analysis of sali-nity, suspended particle content, Chl a, pheophytinand phytoplankton of the water column Temperatureand salinity were measured in situ A one-litre watersample was ¢ltered through a Whatman GF/C(0.45mm pore size, 25 mm diameter, weight), rinsedwith a small volume of Milli-Q water and dried at

60 1C for 24 h, and then weighed to determine thesuspended particle content Chlorophyll a and pheo-phytin were determined according to Lorenzen andJe¡rey (1980) Samples were stored at 20 1C beforeanalytical treatment The abundance of phytoplank-ton species was determined by counting the cells

in a phytoplankton-counting chamber (Popovicha &Marcovecchio 2007) using an Olympus microscope(Tokyo, Japan)

The re-suspension rates were estimated using themethod proposed by Sun and Zhan (2002)

Statistical analysisData were analysed usingSPSS13.0 for Windows sta-tistical software Di¡erences in the parameters wereanalysed using one-way analysis of variance, fol-lowed by SNK tests Di¡erences were considered to

be signi¢cant if Po0.05

ResultsWater quality of sea cucumber culture pondsThere were no signi¢cant di¡erences in the waterquality (temperature, salinity, dissolved oxygenAquaculture Research, 2010, 42, 14^21 Sedimentation in sea cucumber culture ponds Y Ren et al.

(DO), pH, Chl a and content of suspended particles)

between the two investigated ponds The annual

water temperature ranged from 0.0 1.5 to

28.7 1.4 1C, and the highest water temperature

oc-curred in August and the lowest ococ-curred in January

Water salinity of the ponds ranged from 27.6 to

31.1g L 1, and pH ranged from 7.9 to 8.2 Annual

DO contents were above 5.0 mg L 1 The contents of

suspended particulate matter in the water column

ranged from 9.5 to 18.1mg L 1; it was less in summer

due to weak wind and in winter due to ice covering

and less phytoplankton (Fig 1)

Phytoplankton in the ponds

The average Chl a content in the water ranged from

2.5 to 5.5mg L 1, with an average of 3.7mg L 1, and

the maximum Chl a content occurred in July (Fig 2)

The mean number of phytoplankton was 18.5

104cell L 1in the two sea cucumber culture ponds,

and no signi¢cant di¡erence was detected between

the two ponds (P40.05)

Diatoms were the predominant species, constituting

85.1% of the total phytoplankton species

Dino£agel-lates accounted for 6.1%; the others, including

Chloro-phyta, CryptoChloro-phyta, EuglenoChloro-phyta, Chrysophyta and

Cyanophyta, only accounted for 8.8% (Fig 3)

Vertical £uxes of particulate matters and

re-suspension

There were no signi¢cant di¡erences between the

two sea cucumber culture ponds in the £uxes of

TPM, POC, PON, Chl a and pheophytin according tothe monthly analysis (P40.05) Total particulate ma-terial £ux ranged from 10.2 to 38.3 g m 2day 1,with an average of 22.1g m 2day 1, and it washigher during spring or autumn than that duringsummer or winter The maximum TPM £ux in Pond

1 reached 43.5 g m 2day 1 and it reached40.2 g m 2day 1in Pond 2, both occurring in May.However, the minimum TPM £ux in Pond 1 was8.2 g m 2day 1and 8.0 g m 2day 1in Pond 2,both occurring in January The POC content in thesettling particles collected by traps in the two sea cu-cumber culture ponds in October reached44.1mg g 1, which was higher than that in the othermonths, and no signi¢cant di¡erences were observedamong other months (P40.05) Particulate organiccarbon £ux in the two ponds ranged from 377.0 to

water column of sea cucumber culture ponds The values

have been averaged per month Values were given as

TimeFigure 2 Annual Chl a contents in the water column ofsea cucumber culture ponds Values have been averaged

Phytoplankton group composition, % 0 20 40 60 80 100

Time

groups in sea cucumber culture ponds The values havebeen averaged per month

Sedimentation in sea cucumber culture ponds Y Ren et al Aquaculture Research, 2010, 42, 14–21

r 2010 The Authors

1201.0 mg m 2day 1, with an average of 756.1

mg m 2day 1, and it had a signi¢cant correlation

with phytoplankton biomass (r 5 0.822, P 5 0.001)

The PON contents in the settling particles in the two

ponds were higher in October and November; however,

there were no signi¢cant di¡erences among other

months (Po0.05) Particulate organic nitrogen £ux

ranged from 45.5 to 174.8 mg m 2day 1, with an

average of 92.1mg m 2day 1 Two peaks occurred

in the TP contents of settling particles collected from

the two sea cucumber culture ponds in August and

January respectively, and the maximum TP contents

in the settling particles occurred in August, which

reached 1.0 mg g 1 Total phosphorus £ux in the

two sea cucumber culture ponds ranged from 9.2 to

22.7 mg m 2day 1, with an average of 15.6 mg

m 2day 1 Chlorophyll a £ux ranged from 1.0 to

4.4 mg m 2day 1, with an average of 2.5 mg

m 2day 1 The Chl a £ux was higher in spring and

autumn than in summer and winter Pheophytin £ux

in the two sea cucumber culture ponds was higher in

August, September and October than in the other

months The pheophytin £ux of the two ponds ranged

from 2.3 to 22.7 mg m 2day 1, with an average of

8.4 mg m 2day 1 The pheophytin £ux was higher

than the Chl a £ux in the ponds Re-suspension rates

of the sediment ranged from 59.6% to 95.5%, and they

were lower in summer and winter compared with

those in spring and autumn (Table 1)

Annual C/N of the settling particles in the traps ged from 9.6 to 12.7, and it was lower in winter andhigher in autumn (Po0.05) The Chl a/pheophytin ra-tio ranged from 0.2 to 0.5, and the POC/Chl a ratio ran-ged from 131.0 to 829.9, with an average of 312.1 Themaximum POC/Chl a ratio occurred in July (Table 2)

ran-Table 1 Annual £uxes of TPM, POC, PON,TP, Chl a and pheophytin in the sea cucumber culture ponds

Time

TPM

(g m 2day 1)

POC (mg m 2day 1)

PON (mg m 2day 1)

TP (mg m 2day 1)

Chla (mg m 2day 1)

Pheophytin (mg m 2day 1)

Resuspension rate (%)

Values have been averaged and were given as means  SD (n 510).

TPM, total particulate matter; POC, particulate organic carbon; PON, particulate organic nitrogen; TP, total phosphorus; Chl a, ophyll a.

chlor-Table 2 Ratio of C/N, Chl a/pheophytin and POC/Chl a of the settling particles in the ponds

Time

C/N (a/a) POC/Chl

a (w/w)

Chl a/

pheophytin (w/w)

April 2007 10.8 312.0 0.5 May 2007 11.6 239.8 0.4 June 2007 11.4 427.4 0.4 July 2007 10.0 829.9 0.4 August 2007 10.2 131.0 0.2 September 2007 12.1 301.1 0.2 October 2007 12.7 294.1 0.2 November 2007 12.1 439.2 0.4 December 2007 9.6 387.5 0.3 January 2008 9.7 377.0 0.3 February 2008 9.8 299.1 0.4 March 2008 11.0 262.5 0.5 April 2008 11.1 312.1 0.5 Average 10.9 354.8 0.4

Values have been averaged per month (n 510).

POC, particulate organic carbon; Chl a, chlorophyll a; a/a, atom/ atom; w/w, weight/weight.

Aquaculture Research, 2010, 42, 14^21 Sedimentation in sea cucumber culture ponds Y Ren et al.

Sediment characteristics of the ponds

The ponds investigated had the same sediment

qual-ity before the sea cucumber stocking In April 2007,

the sediment contents of POC, PON and TP in Pond 1

were 2.5 0.2, 0.2  0.0 and 0.7  0.0 mg g 1

, spectively, and the POC, PON and TP contents in the

re-sediment of Pond 2 were 2.5 0.1, 0.2  0.0 and

0.7 0.0 mg g 1respectively In the pond without

sea cucumber, the sediment contents of POC, PON

and TP were 2.7 0.1, 0.2  0.1 and 0.6  0.0

mg g 1respectively During 1 year of culture, no

sig-ni¢cant di¡erences were observed in the contents of

POC, PON and TP in the sediment between Pond1and

Pond 2 by monthly analysis (P40.05) In April 2008,

when the experiment was completed, the sediment

contents of POC, PON and TP in Pond 1 were

3.4 0.2, 0.4  0.0 and 0.7  0.0 mg g 1

, tively, and in Pond 2 they were 3.4 0.1, 0.4  0.0

respec-and 0.6 0.1mg g 1

respectively Nevertheless, thesediment contents of POC, PON and TP in the pond

without sea cucumber were 10.3 0.5, 1.2  0.0

and 0.9 0.1mg g 1, respectively, which were

sig-ni¢cantly higher than those in sea cucumber culture

ponds (Po0.05) The mean POC contents in the

sedi-ment of the sea cucumber culture ponds ranged from

2.4 to 7.7 mg g 1(Fig 4), the PON contents ranged

from 0.2 to 0.9 mg g 1and the TP contents ranged

from 0.3 to 1.0 mg g 1(Fig 5) Two peaks occurred

in summer and winter for all above matters in the

two sea cucumber culture ponds The pheophytin

contents in the sediment of the sea cucumber culture

ponds were signi¢cantly higher than that of Chl a

(Po0.05) (Fig 6) Chlorophyll a and pheophytin

con-tents in the sediment of sea cucumber culture pondswere signi¢cantly lower than those in the pond with-out sea cucumber (Fig.6) The annual mean sedimentcontents of POC, PON and TP of the two sea cucum-ber culture ponds were 3.9, 0.4 and 0.6 mg g 1, re-spectively, however they were 8.0, 0.9 and0.8 mg g 1, respectively, in the pond without sea cu-cumber Thus, the annual mean POC, PON and TPcontents in the sediment of the sea cucumber cultureponds were signi¢cantly lower than those in thepond without sea cucumber (Po0.05) (Fig.7)

DiscussionPrevious studies have shown that OM deposition

is a key process in the open sea (Takahashi 1986;

tents in the sediment of sea cucumber culture ponds

Va-lues have been averaged per month The vaVa-lues were

0.0 2 4 6 8 1.0

TP

TimeFigure 5 Annual particulate organic nitrogen (PON)and total phosphorus (TP) contents in the sediment ofsea cucumber culture ponds The values have been aver-

100

Sea cucumber Without sea cucumber

Figure 6 Mean Chl a and pheophytin contents in the diment of sea cucumber culture ponds and the pond with-out sea cucumber during the investigated period Values

Sedimentation in sea cucumber culture ponds Y Ren et al Aquaculture Research, 2010, 42, 14–21

r 2010 The Authors

Danovaro et al 2000), and the quality and quantity of

OM input to the sediment from the water column

is very important for macrobenthic production

(Gre’mare, Amoroux, Charles, Dinet, Riaux-Gobin,

Baudart, edernach, Bodiou, Ve’tion, Colomines &

Albert 1997) The present study showed that in sea

cucumber culture ponds, the TPM sedimentation

rates were high, which ranged from 10.2 to

38.3 g m2day 1 There was a signi¢cant correlation

between the POC contents in TPM and the biomass

of phytoplankton (r 5 0.822, P 5 0.001) In the sea

cucumber culture ponds, the Chl a and pheophytin

sedimentation rates reached 2.5 and 8.4 mg m 2

day 1, respectively, which implied that

phytoplank-ton took a certain proportion of the settling

particu-late matter Diatoms in the water column occupied

490% of the phytoplankton during spring and

autumn that could provide good-quality food to sea

cucumber after settling to the bottom The Chl a

£ux in the two sea cucumber culture ponds was

4.0 0.8 mg m 2day 1 in October and it was

4.4 0.4 mg m 2day 1 in May, which indicated

that there were many diatoms settling to the bottom

from the water layer

The sea cucumber culture ponds were shallow;

therefore, the re-suspension of the sediment was not

a negligible process In the present study, the

re-sus-pension rates of settling particulate matter ranged

from 60.0% to 95.5%, which was similar to the values

found in Erie Lake (55^95%), the Peel-Harvey

Estu-ary (69^92%) and the coastal areas of East China

Sea (72.75^96.96%) respectively (Gabrielson &

Luka-telich 1985; Sun & Zhan 2002) The wave and tide

dri-ven by wind are the main causes of re-suspension of

sedimentation matter (Rhoads 1974) In this study,the re-suspension rates were lower in summer andwinter than those in spring and autumn due to weakwind in summer and ice covering and less phyto-plankton in winter (Table 1)

Research on shallow areas of water showed thatthe particles collected by traps were complex (Lunda-gaard & Olesen 1997), which contained both newlyborn particles and re-suspension ones Generally, alower C/N of the particles implies that there is a highportion of unstable OM in the settling particles (Nick-ell, Black, Hughes, Overnell, Brand, Nickell, Breuer &Harvey 2003) In this study, the mean C/N of the set-tling particulate matter in the trap samples rangedbetween 9.6 and 12.7, and the values in autumn(12.1^12.7) were signi¢cantly higher than those inwinter (9.6^9.8) The settling particles in autumnhad undergone more cycles of sedimentation and re-suspension processes, leaving more stable OM in thesettling particles, and might be less nutritional forsea cucumber

In this study, the average TPM £ux in the pondswas 22.1g m2day 1, comprising 756.1mg C, 92.1mg

N and 15.6 mg P, and it was 10 times lower than that

in Sungo Bay, China, where ¢lter-feeder scallop wascultured (Cai, Fang & Liang 2003) Sedimentationplays an important role in the energy transformationfrom primary producers to benthic consumers (Les-ser 2006) In a high density of sea cucumber cultureponds, only natural sedimentation might not supplysu⁄cient food for sea cucumber In practice, polycul-ture of sea cucumber with scallop or jelly¢sh is ane⁄cient way to accelerate the sedimentation of or-ganic particles and improve the productivity of seacucumber (Zhou et al 2006; Zheng, Dong,Tian,Wang,Gao & Bai 2009) The polyculture of sea cucumberwith ¢lter feeders can not only accelerate sedimenta-tion but also alleviate nutrient loadings of nearbycoasts e¡ectively

The sea cucumber grew roughly from April toearly June, after aestivation from July to September,grew again from October to December and then theywent into hibernation from January to March In thisstudy, the POC, PON and TP contents in the sediment

of sea cucumber culture ponds increased during theperiod of sea cucumber aestivation and hibernation,while in contrast, they decreased during the feedingperiods In shrimp and ¢sh culture ponds, OM accu-mulation is frequent at the bottom (Allan, Moriarty &Maguire 1995); however, OM accumulation and thecontents of POC, PON, TP, Chl a and pheophytin inthe sediment of sea cucumber culture ponds were

Figure 7 Mean particulate organic carbon (POC),

parti-culate organic nitrogen (PON) and total phosphorus (TP)

contents in the sediment of sea cucumber culture ponds

and the pond without sea cucumber during the

Aquaculture Research, 2010, 42, 14^21 Sedimentation in sea cucumber culture ponds Y Ren et al.

even lower than those in the pond without sea

cu-cumber (Po0.05) (Fig 7) The results demonstrated

that sea cucumber culture can e¡ectively stop

nutri-ent accumulation at the bottom of the culture ponds

Acknowledgments

This work was supported by National Projects

of Scienti¢c and Technical Supporting Programs of

China (No 2006BAD09A01), Hi-Tech Research and

Development Program of China (No 2006AA

10Z409) and Natural Science Fundation of China

(No 30871931) We thank Fisheries Research

Insti-tute, Enterprise Homey Group, for helping with the

experiments

References

Allan G.L., Moriarty D.J.W & Maguire G.B (1995) E¡ects of

pond preparation and feeding rate on production of

Penaeus monodon Fabricius, water quality, bacteria

and benthos in model farming ponds Aquaculture 130,

329^349.

Cai L.S., Fang J.G & Liang X.M (2003) Natural

sedimenta-tion in large-scale aquaculture areas of Sungo Bay, north

China Sea Journal of Fishery Sciences of China 10, 305^310

(in Chinese with English abstract).

Cromey C.J., Nickell T.D & Black K.D (2002)

DEPOMOD-modelling the deposition and biological e¡ects of waste

solids from marine cage farms Aquaculture 214, 211^239.

Danovaro R., Croce ND., Dell’Anno A., Fabiano M., Marrale

D & Martorano D (2000) Seasonal changes and

biochem-ical composition of the labile organic matter £ux in the

Cretan Sea Progress in Oceanography 46, 259^278.

Fabiano M., Danovaro R & Fraschetti S (1995) A three-year

time series of elemental and biochemical composition of

organic matter in subtidal sediments of the Ligurian Sea

(northwestern Mediterranean) Continental Shelf Research

15, 1453^1469.

Gabrielson J.O & Lukatelich R.J (1985) Wind-related

resus-pension of sediments in the Peel-Harvey Estuarine

sys-tem Estuarine, Coastal and Shelf Science 20, 135^145.

Gerino M., Stora G., Poydenot F & Bourcier M (1995)

Benthic fauna and bioturbation on the Mediterranean

continental slope: Toulon Canyon Continental Shelf

Research 15, 1483^1496.

Gre’mare A., Amoroux J.M., Charles F., Dinet A.,

Riaux-Go-bin C., Baudart J., edernach M.L., Bodiou J.Y.,Ve’tion G.,

Co-lomines J.C & Albert P (1997) Temporal changes in the

biochemical composition and nutritional value of the

par-ticulate organic matter available to surface

deposit-fee-ders: a two year study Marine Ecology Progress Series

150, 195^206.

Honjo S., Manganini S.J & Wefer G (1988) Annual particle

£ux and a winter outburst of sedimentation in the ern Norwegian Sea Deep-Sea Research 35, 1223^1234 Josefson A.B & Rasmussen B (2000) Nutrient retention by benthic macrofaunal biomass of Danish estuaries: impor- tance of nutrient load and residence time Estuarine, Coastal and Shelf Science 50, 205^216.

north-Kawahata H (2002) Suspended and settling particles in the Paci¢c Deep-Sea Research Part II:Topical Studies in Oceano- graphy 49, 5647^5664.

Lesser M.P (2006) Benthic^pelagic coupling on coral reefs: feeding and growth of Caribbean sponges Journal of Experimental Marine Biology and Ecology 328, 277^288 Liao Y.L (1997) Fauna Sinica, Phylum Echinodermata, Class Holothuroidea Science Press, Beijing, China, 334pp (in Chinese).

Lohrenz S.E., Knauer G.A., Asper V.L & Tuel M (1992) sonal variability in primary production and particle £ux

Sea-in the northwestern Sargasso Sea: U.S JGOFS Bermuda Atlantic time-series study Deep-Sea Research 39, 1373^1391.

Lorenzen C.J & Je¡rey S.W (1980) Determination of phyll in seawater SCORUNESCO Technical Papers in Mar- ine Science 35, 1^20.

chloro-Lundagaard C & Olesen M (1997) The origin of sedimenting detrital matter in a coastal system Limnology and Oceano- graphy 42, 1001^1005.

Mudroch A., Azcue J.M & Mudroch P (1997) Manual of Physico-Chemical Analysis of Aquatic Sediments Lewis Publishers, Boca Raton, FL, USA.

Nickell L.A., Black K.D., Hughes J.D., Overnell J., Brand T., Nickell T.D., Breuer E & Harvey S.M (2003) Bioturbation, sediment £uxes and benthic community structure around a salmon cage farm in Loch Creran, Scotland Experimental Marine Biology and Ecology 285, 221^233 Popovicha C.A & Marcovecchio J.E (2007) Spatial and tem- poral variability of phytoplankton and environmental factors in a temperate estuary of South America (Atlantic coast, Argentina) Continental Shelf Research 28, 236^244 Rhoads D.C (1974) Organism sediment relations on the muddy sea £oor Oceanography and Marine Biology: an Annual Review 12, 262^300.

Sasaki H., Hattori H & Nishizawa S (1988) Downward £ux

of particulate organic matter and vertical distribution of calanoid copepods in the Oyashio water in summer Deep-Sea Research 35, 505^515.

Sun Z.T & Zhan B.Q (2002) The resusupension rate of ments in 321N section on the east china sea Chinese Journal of Oceanology and Limnology 20, 191^196 Sutherland T.F., Martin A.J & Levings C.D (2001) Character- ization of suspended particulate matter surrounding a salmonid net-pen in the Broughton Archipelago, British Columbia ICES Journal of Marine Science 58, 404^410 Takahashi K (1986) Seasonal £uxes of pelagic diatoms in the Subarctic Paci¢c, 1982^1983 Deep-Sea Research Part

sedi-A Oceanographic Research Papers 33, 1225^1251 Sedimentation in sea cucumber culture ponds Y Ren et al Aquaculture Research, 2010, 42, 14–21

r 2010 The Authors

Wassmann P (1993) Regulation of vertical export of

particu-late organic matter from the euphotic zone by planktonic

heterotrophs in eutrophicated aquatic environments.

Marine Pollution Bulletin 26, 636^643.

White J (1990) The use of sediment traps in high-energy

en-vironments Marine Geophysical Researches 12, 145^152.

Yingst J.Y (1976) The utilization of organic matter in shallow

marine sediments by an epibenthic deposit-feeding

ho-lothurian Experimental Marine Biology and Ecology 23,

55^69.

Zheng Z.M., Dong S.L.,Tian X.L.,Wang F., Gao Q.F & Bai P.F (2009) Sediment-water £uxes of nutrients and dissolved organic carbon in extensive sea cucumber culture ponds CLEAN-Soil, Air,Water 37, 218^224.

Zhou Y., Yang H.S., Liu S.L., Yuan X.T., Mao Y.Z., Liu Y., Xu X.L & Zhang F.S (2006) Feeding and growth on bivalve biodeposits by the deposit feeder Stichopus japonicus Selenka (Echinodermata: Holothuroidea) co-cultured in lantern nets Aquaculture 256, 510^520.

Aquaculture Research, 2010, 42, 14^21 Sedimentation in sea cucumber culture ponds Y Ren et al.

Cryopreservation of sperm from natural and

sex-reversed orange-spotted grouper

Taweesin Peatpisut & Amrit N Bart

Aquaculture and Aquatic Resources Management, Asian Institute of Technology, Pathumthani,Thailand

Correspondence: A N Bart, Aquaculture and Aquatic Resources Management, Asian Institute of Technology, PO Box 4, Klong Luang, Pathumthani 12120,Thailand E-mail: bart@ait.asia

Abstract

The shortage of males and/or sperm has been an

im-pediment to the aquaculture of orange-spotted

grouper (Epinephelus coioides) This study reversed

or-ange-spotted grouper females into males using

hor-mone implants A cryopreservation protocol for

sperm was developed using normal males, and then

using similar procedures the cryopreservation of

sperm from sex-reversed males was compared

Im-mature, young and mature female ¢sh were injected

with 4 mg kg 1BW 17a methyltestosterone as

im-plants and the gonad development stage was

moni-tored over a 120-day period All treated females

converted into functional males within 120 days of

the experimental period Younger females (2Y) were

all males within 30 days, although not all were

cap-able of fertilizing fresh ova until day 60 The time

after injection to sex reversal in immature ¢sh was

50% shorter than in older females Postthaw

fertili-zation (81%, 82%) and hatching (45%, 47%) of

cryo-preserved sperm from natural males were the highest

in trehalose (15^20%) with 150 mmol NaCl

treat-ment; however, it was less than the control (89%

fer-tilization and 69% hatch) There was no di¡erence in

the postthaw fertilization and the hatch percentages

between sex-reversed male sperm (64% and 46%

re-spectively) compared with natural male sperm (59%

and 49%) The ¢ndings of this study suggest the

po-tential use of sex-reversed males and cryopreserved

sperm for commercial production of orange-spotted

grouper seed for aquaculture

Keywords: sperm cryopreservation, sex-reversed,

orange-spotted grouper, sperm quality, Epinephelus

coioides

IntroductionOrange-spotted grouper (Epinephelus coioides) is animportant commercial aquaculture species in South-east Asia (Boonyaratpalin 1997; Millamena 2002).However, its culture is constrained by an inconsistentsupply of seed to meet the increasing aquaculturedemand (Liao & Leano 2008) Because of the protogy-nous hermaphroditic nature of this species, obtain-ing an adequate number of mature males from thewild has been problematic (Quinitio, Caberoy & Reyes1997) Maintaining ¢sh in a hatchery for breedingwould increase the number of males but the timeand cost of maintaining ¢sh for 6^8 years before sexinversion would naturally make it economically pro-hibitive (Quinitio et al.1997) Moreover, the volume ofexpressible milt is low in many species during the o¡peak spawning period Some authors have success-fully increased the volume of milt by inducing malesusing luteinizing hormone releasing hormone analo-gue (LHRHa) (Barry, Castanos & Fast 1991)

One of the solutions to produce more spermiatingmales is hormonal therapy Several studies havereported a successful induction of sex reversal in pro-togynous hermaphrodites by treatment with andro-gens (Chen, Chow, Chao & Lim 1977; Chao & Chow1990; Tan-Fermin, Garcia & Castillo 1994) Sex rever-sal by hormonal treatment have been successful inorange-spotted grouper and dusky grouper (Epine-phelus marginatus; Yeh, Kuo, Ting & Chang 2003c;Sarter, Papadaki, Zanuy & Mylonas 2006) Although,

a successful sex change in 2-year-old orange-spottedgrouper was possible using a17a methyl-testosteroneimplant (1000mg kg 1BW), the quality of the spermwas not assessed (Yeh, Kuo, Ting & Chang 2003b).Sex reversal of potato grouper, Epinephelus tukula, in

r 2010 Blackwell Munksgaard

vitellogenic females was also reported (Yeh, Dai, Chu,

Kuo, Ting & Chang 2003a) While sex reversal using

androgen treatment has been successful in a number

of protogynous hermaphroditic species, these studies

examined only a single age group (i.e Yeh et al

2003b, c; Sarter et al 2006) There has not been a

study to examine reversal e⁄ciency with varying age

Cryopreserved sperm would be another solution to

the timely availability of functional sperm for

fertiliza-tion and fry/¢ngerling producfertiliza-tion Cryopreserved

sperm of orange-spotted grouper would allow

year-round seed production, reduce the number of males

needed in the hatchery and facilitate arti¢cial

propa-gation Although cryopreservation techniques have

been well established for many ¢sh species, only a

lim-ited number of studies have been carried out on

groupers Spermatozoa cryopreservation has been

achieved in greasy grouper (Epinephelus tauvina;

Withler & Lim1982), malabar grouper (Epinephelus

ma-labaricus; Chao,Tsai & Liao 1992; Gwo 1993) and k elp

grouper (Epinephelus moara; Miyaki, Nakano, Ohta &

Kurokura 2005) These studies were carried out under

laboratory conditions and motility was used as the

only indicator of success The postthaw fertilization,

hatch and survival rates of larvae from frozen-thawed

spermatozoa are often lacking Moreover, only one

study on dusky grouper reported the successful

cryo-preservation of sperm from sex-reversed males

(Cabri-ta, Engrola, Conceicao, Pousao-Ferreira & Dinis 2009)

Selection of appropriate cryoprotectants is

impor-tant to successful cryopreservation of spermatozoa

There are many cryoprotectants successfully used in

freezing protocols including dimethyl sulphoxide

(DMSO) and propylene glycol (PG; Billard, Cosson &

Crim 1993; Gwo 1993, 1994; Richardson, Crim, Yao &

Short 1995; Cabrita et al 2009) Trehalose, a novel

cryoprotectant, has been used in studying only a

lim-ited number of freshwater species (Miyaki et al 2005),

but has not been tested in marine species

Year-round availability of spermiating functional

males and/or sperm of orange-spotted grouper are

critical to the commercial production of this species

The aim of this study was to: (1) better understand

the age-related sex reversal e⁄ciency and (2) to

devel-op a simple protocol for sperm crydevel-opreservation of

sex-reversed orange-spotted grouper

Materials and methods

Experimental ¢sh and design

Thirty orange-spotted groupers cultured in cages

were purchased from several farms during the

post-spawning season based on their age and health Theexperimental ¢sh were maintained and prepared forreproduction in a 30 m3 circular concrete tank(+5 m and 1.6 m deep) connected to a recirculatingsystem in the Aquaculture and Aquatic ResourcesManagement (AARM) hatchery at the Asian Institute

of Technology (AIT),Thailand Fish were maintained

in saline water (30%) at 26^30 1C and fed to satiationevery other day with frozen whole yellow striped scad(Selaroides leptolepis) Moreover, once a week a vita-min E (400 IU) capsule (Mega, Samutprakarn, Thai-land) was inserted in the frozen ¢sh before feeding.The experiment comparing an age group e¡ect onsex-reversal was initiated at the end of August 2006,while cryopreservation experiments were initiated inearly January 2007 Fish were selected from the pool

of 30 for sex reversal and cryopreservation studies.Sampled ¢sh were anaesthetized with 50 mg L 1ofbenzocaine for 3 min or until the opercular move-ment was signi¢cantly reduced Fish were canulatedusing a £exible polyethylene tube through the genitalpore and examined microscopically to determinetheir sex stage (Fig 1) The age of the ¢sh was deter-mined based on hatchery data Fifteen 1.5^4.5-year-old females (2.41^8.73 kg) at the perinucleolus oocytestage (transparent oocyte, 30^130mm diameter) werechosen for the experiment The selected ¢sh weredivided into three age groups of ¢ve individualseach of 1.5^2.5 years (2.56 0.21kg), 2.6^3.5 years5.05 1.04 kg) and 3.6^4.5 years (7.26  0.95) andidenti¢ed as group 2, 3 and 4Y respectively Experi-mental females were individually weighed and tagged(Fish eagle PIT tags, Biomark, ID, USA) In each group,three females were implanted with methyltestoster-one (MT), and two ¢sh served as controls

Figure 1 Gonadal tissue was sampled by passing a ter through the genital pore and gently vacuuming thegonadal lining The tissue was examined microscopically

cathe-to determine the sex stage

Aquaculture Research, 2010, 42, 22^30 Cryopreservation of sperm from Epinephelus coioides T Peatpisut & A N Bart

Induced sex change

The hormone pellet for implantation was prepared by

mixing 350 mg of 17a MT, 1mL of 80% ethanol,

190 mg of cholesterol powder (mixture of 95^100%

cholesterol and 0^5% animal lard) and 10 mg of

co-coa butter (modi¢ed from Lee, Tamaru & Kelly 1986)

The mixture was passed through a press pelletizer

and the pellets were cut to achieve 10 mg per pellet

with 35 total pellets.While the smallest ¢sh (2.41kg)

received a single pellet, the largest ¢sh (8.73 kg) were

implanted with 3.5 pellets The MT implant

contain-ing 4.0 mg kg 1BW was injected intramuscularly

below the base of dorsal ¢n The control ¢sh were

im-planted with blank implants

The examination of functional males

On days 30, 60, 90 and 120 after implantation with

MT, the gonadal stages of the treated ¢sh were

deter-mined During the early morning period, gonadal

tis-sue was collected using a plastic canulation catheter

(Feeding Tube CH 6, City Medical Supply, Bangkok,

Thailand) The catheter was passed through the

gen-ital pore to remove the gonadal lining The biopsy was

examined under a microscope and photographed

using a digital camera The gonadal stage was

classi-¢ed into F1, F2, F3and I stages in females Males were

classi¢ed into M1and M2stages (Fig 2)

F1, presence of transparent oocytes (30^130mmoocyte diameters)

F2, presence of some dark oocytes (130^250mmoocyte diameters)

F3, presence of oocytes with transparent circle rounding the yolk (4400 mm diameter)

sur-I, presence of a few sperm in the biopsy tissue

M1, presence of sperm cells but not motile uponactivation

M2, presence of motility competent sperm cells

Sperm cryopreservationCollection and cryopreservation of sperm from naturaland sex-reversed males

Spermiation was induced 72 h before stripping males

by implanting pellets containing LHRHa (Suprefacts,Hoechst AG, Maine, Germany) at 10mg kg 1 The pel-lets were prepared using a similar procedure asdescribed for MT Pellets were implanted intramuscu-larly behind the dorsal ¢n From the pool of remain-ing 15 ¢sh, three males of 8 years old (10.2^12.5 kg)were selected as donors of sperm for the cryopreser-vation experiment The sperm from three males wascollected by applying gentle pressure to the abdom-inal area along the midline and aspirated with a mi-cropipette (Pipetman, Gilson, France) Caution wasexercised to prevent contamination of the sperm with

(d)

(d) I stage: presence of some recognizable sperm cells in the biopsy tissue (10), (e) M1stage: presence of non-motile sperm

Cryopreservation of sperm from Epinephelus coioides T Peatpisut & A N Bart Aquaculture Research, 2010, 42, 22–30

r 2010 Blackwell Munksgaard

urine or faecal material The pooled sperm from three

males (1.5 mL) was transferred into a 25 mL beaker in

an ice bath and transported from the hatchery to the

laboratory Only the sperm samples with motility

475% were used for cryopreservation trials

Sperm from three males was pooled in equal

amounts and diluted with an extender (150 mmol

NaCl, pH 7.5^8, osmolarity 275 mOsm kg 1

) at aratio of 1:9 (sperm:extender) Approximately, 50% of

the total sperm and extender mixture was used for

cryopreservation while the remainder was held in

the refrigerator at 4 1C to serve as a control during

fertilization

Three cryodiluents were prepared by mixing each

cryoprotectant, (DMSO, Trehalose T and PG) with

150 mmol of NaCl extender solution to achieve

concentrations of 10%, 15% and 20% Each

cryodilu-ent was mixed with the diluted sperm at 1:1 (diluted

sperm:cryodiluents) ratio From the mixture, 0.48 mL

was loaded into 0.50 mL French straws using a

mi-cropipette (Nichipet EX, Nichiryo, Japan) and sealed

with a heated hemostat Straws were subsequently

placed in the cryochamber The equilibration period

waso10 min

The straws were frozen in a

programmable-con-trolled rate freezer (CL 863) using Cryogenesis

ver-sion 4.0, for WINDOWS (Crylogic 1999) A one-step

freezing procedure (10 1C min 1from 25 to 80 1C)

was programmed and run When the ¢nal freezing

temperature reached 80 1C, the samples were

re-moved from the cryochamber and stored in a liquid

nitrogen Dewar

The sex-reversed males (M2stage) were used to

in-duce spermiation by implantation with 10mg kg 1of

LHRHa After 72 h of implantation, sperm was

col-lected in a manner similar to that described above

Milt from each male was examined to determine

sperm concentration and motility Only sperm

sam-ples with motility475% were pooled and used for

cryopreservation

The cryopreservation procedure was similar to

that described above The milt collected from three

sex-reversed males was cryopreserved using 15%

tre-halose with the NaCl extender The equilibration time

was o10 min The samples were frozen at

10 1C min 1before storing in liquid nitrogen

Sperm concentration, motility and

viability test

Stripped milt was diluted in 150 mmol NaCl The

number of sperm was estimated using a Neubauer

haemocytometer (Paul Marienfeld GmbH KG, drichsdorf, Hessen, Germany; Bart,Wolfe & Dunham1998) To estimate the relative motility, 2mL of spermwas activated on a glass slide with 20mL of sea water.The percentage of visible progressively motile cellswas estimated using a subjective scale according

Frie-to Bart et al (1998) under an Olympus microscope

at  100 magni¢cation before and after freezing

as described by Wayman and Tiersch (2000) Spermviability was assessed by staining and observing un-der a binocular microscope (Olympus, Tokyo, Japan

 400 magni¢cation) connected to a U-PMTVC with

an HAD colour video camera (M SSC-C108P, Sony,Tokyo, Japan) To test for viability, eosin-nigrosin dyewas used to stain ruptured sperm cells (Guest 1973).Stained cells were considered non-viable

Fertilization and hatching

A mature female (oocytes diameter4450 mm) washormone induced LHRHa (Suprefacts) and Metoclo-pramide (MET-SILsT.P Drug Lab, Bangkok,Thailand)were injected intramuscularly at an initial dose of 10and 2 mg kg 1respectively The resolving dose wasadministered after 24 h with 20mg of LHRHa and

2 mg of Metoclopramide (kg 1BW) After ovulation(10^12 h), eggs were stripped from a female into a

150 mL glass beaker Batches of 0.2 g of stripped eggs(500 4 oocytes) were placed in 27 Petri dishes se-parately (nine treatments, three replicates) for natur-

al males, 18 Petri dishes (six treatments, threereplicates) for three di¡erent age groups of sex-re-versed males and 12 Petri dishes (four treatments,three replicates) for sex-reversed males to comparewith natural males

After freezing in LN2 for 1^2 h, straws werethawed in a 40 1C water bath for 15 s and poured into

an eppendorf tube Using a micro pipette, sperm fromthe eppendorf tube was removed and mixed with thenewly stripped eggs The amount removed was based

on the number of sperm required to fertilize 500 eggs

in a batch The sperm-to-egg ratio was 2.8 104sperm per egg Approximately, 10 mL of 30% sea-water was poured into the Petri dish to activate thesperm and eggs After 5 min, the eggs were rinsedand transferred to a1 L beaker containing 0.5 L water

An airstone (1.0 cm diameter) was placed in the ker to mix the eggs uniformly Eggs were sampledfrom the beaker, and100 eggs were used to determinethe per cent fertilization The per cent fertilizationwas determined by hand counting under the stageAquaculture Research, 2010, 42, 22^30 Cryopreservation of sperm from Epinephelus coioides T Peatpisut & A N Bart

bea-microscope The remaining eggs in the beaker were

allowed to settle by removing the air stone The

£oat-ing eggs were siphoned o¡ and incubated in a l L

bea-ker with 0.5 L water at 100 eggs L 1 An airstone was

placed in the centre of the beaker for gentle aeration

of fertilized eggs The water temperature (29 1C) was

maintained by placing the incubation beaker in a

Styrofoam tray (60 45  30 cm) with 10 cm of

water and an aquarium heater The number of eggs

and hatched larvae were hand counted in each

incu-bation beaker 18 h after fertilization

Data analysis

Motility, viability, fertilization and hatching

percen-tages between treatments were ¢rst arcsine

trans-formed Two-way analysis of variance (ANOVA) was

performed using the statistical package (SPSS) 11.5 for

Windows to compare the di¡erences among

treat-ment means Paired comparisons between treattreat-ment

means were performed with Tukey’s test All values

were considered to be signi¢cantly di¡erent when

Po0.05

Results

Induction of sex reversal using 17a MT

All treated females converted into functional males

within120 days of the experimental period, and none

of the control females changed their sex during this

same period (Table 1) Younger females in group 2Y

were all males at the M1stage of development within

30 days No M2stage males were found until day 60

of sampling Intersex (I) was observed in three

trea-ted females in group 3 and 4Y in the 30 days of

sam-pling only Oocytes were present only in the control

and F1stage females

Cryoprotectants

The per cent fertilization was highest in the trehalose

treatment group and was similar to the control

Simi-larly, the per cent hatch was also highest in this

treat-ment group but less than in the controls (Table 2)

Propylene glycol resulted in the lowest motility,

viabi-lity, fertilization and hatch Fertilization and hatch

variance was low within treatment groups (0.3^

4.8%) Generally, the hatch rates were low (1.3^

11.3%) except in the case of 15% and 20% trehalose

Sperm treated with a high concentration of DMSO

20% (v:v) had low fertilization and hatching

percen-tages (44.1 0.3% and 11.3  3.5% respectively) In

general, a low concentration of all three

cryoprotec-tants resulted in higher fertilization and hatching

percentages except in the case of trehalose, where

both 15% and 20% trehalose solution resulted in

fer-tilization rates similar to the control Sperm treated

with 20% PG resulted in no motility, and no

fertiliza-tion Interaction was observed between

cryoprotec-tants and the concentrations used in fertilization

and hatching

Comparison of sperm cryopreservation from

three di¡erent age groups of sex-reversed

males

The volume of sperm from ¢sh 120 days after

im-planting the hormone was low ranging from 10

to 500mL There was no di¡erence in the spermnumber among sex-reversed males (n 5 9), 2Y(2.6 0.2  10 mL 1

), 3Y (1.9 0.0  10 mL 1and 4Y (1.9 0.6  10 mL 1

) In all sex-reversedmales, the percentage of motility and viabilitydecreased in cryopreserved sperm Fresh spermmotility was 88.3 1.4% and the viability was94.8 0.8% There were no di¡erence in the percen-tage with motility and the viability of sperm amongall sex-reversed ¢sh There was also no di¡erence inper cent fertilization among each sex-reversed groupusing either fresh or postthawed sperm (Table 3)

Cryopreserved sperm from sex-reversed malescompared with natural males

The per cent hatch was similar between served sperm from sex-reversed males and freshand/or cryopreserved sperm from natural males

cryopre-Table 2 E¡ect of three cryoprotectants (DMSO, trehalose and propylene glycol) at 10%, 15% and 20% concentration using a one-step freezing procedure (10 1C min 1) on frozen thawed sperm motility, viability, fertilization and hatching per cent Cryoprotectant Concentration (%) Motility (%) Viability (%) Fertilization (%) Hatching (%)

Values within column followed by di¡erent superscripts are di¡erent (P o0.05, ANOVA ).

The pooled sperm from three males was used for the cryopreservation trials Each treatment was replicated three times with a batch of eggs from a single female.

AVOVA , analysis of variance; DMSO, dimethyl sulphoxide.

Table 3 Per cent sperm motility, viability and fertilization using postthaw cryopreserved sperm from sex-reversed spotted male grouper of three age groups

Group 2Y (n 5 3) Fresh 100 96  1 a 65.7  2.3

Cryopreserved 100 90  1 ab 64.7  2.6 Group 3Y (n 5 3) Fresh 100 95  1 a

69.3  3.5 Cryopreserved 92  8 87  2 b

62.0  1.7 Group 4Y (n 5 3) Fresh 100 93  1 ab

69.7  3.2 Cryopreserved 100 92  3 ab 66.3  2.0 Values within column followed by di¡erent superscripts are di¡erent (P o0.05, ANOVA ).

The pooled sperm from three sex-reversed males and batches of eggs from a single female was used for the trials Each treatment was replicated three times.

AVOVA , analysis of variance.

Aquaculture Research, 2010, 42, 22^30 Cryopreservation of sperm from Epinephelus coioides T Peatpisut & A N Bart

(Fig 3) However, the percentage of hatch was

signi¢-cantly lower than from fresh sperm When

cryopre-served sperm from sex-reversed males was used, the

fertilization and hatching rates were 64.1 0.5%

and 45.8 5.4%; fertilization and hatching were

67.5 2.0% and 66.7  3.9% in fresh sperm groups

Fertilization and hatching between natural males

and sex-reversed males were similar

Discussion

Induced sex-reversal experiment

Using a 4 mg 17a MT kg 1BW cholesterol-based

im-plant, this study achieved 100% sex reversal within

120 days Despite the presence of the M2stage gonad,

sperm could not be stripped However, after LHRHa

implantation (72 h), it was possible to strip all

in-duced males and obtain sperm (from 10 to 500mL)

The total sperm concentrations were not di¡erent

However, the amount of expressible milt was

approxi-mately half of that found in natural males This could

be explained by the relative size di¡erence between

sex-reversed males (5.21 2.33 kg) compared with

natural males (11.47 1.17 kg) Further studies are

needed to understand the optimum implant dose of

LHRHa and the duration of exposure to increase the

total expressible milt

It took longer for the older females to reach M2

stage The youngest group (2Y) did not have an

inter-sex period as was observed in the older females

When examined on day 30, we found all three

fe-males converted to fe-males (M1) It is conceivable that

intersex is a prerequisite step to the male stage andthat this period is shorter in younger females Sex dif-ferentiation in teleosts happens as early as a fewweeks after birth while it takes several years in somespecies (Bhandari, Alam, Higa, Soyanok & Nakamura2005) Earlier studies by Glamuzina, Glavic, Skara-muca and Kozul (1998) found oocytes at the perinu-cleorous stage in 1-year-old dusky grouper It is alsopossible that younger ¢sh have more under devel-oped sex cells and di¡erentiated more quickly withthe surge of androgen from the implanted hormone.This could explain why it took longer for the older fe-males to reach the M2stage Information on the pro-portionate number of oocytes in younger femalecould help explain the reason for the di¡erence be-tween younger and older females Unfortunately,such studies are non-existent in the literature.Induced sex-reversal using 4 mg 17a MT kg 1BWimplanted in 2^year-old females was more completeand faster compared with previous studies Yeh et al.(2003c) achieved 85.7% sex-reversed male in orange-spotted grouper (2 years old) implanted with MT at alower dose (1.0 mg kg 1BW) after 90 days Theirlower rate of response may be due to age and dose dif-ferences Clearly, more detailed experiments to inves-tigate MT dose response to age and/or gonadcharacterization is needed

Yeh et al (2003a) implanted potato grouper,6 years

of age with an androgen mixture in pellets at a dose

of 1mg kg 1BW and achieved 38.9% sex reversal(M11M2) within 30 days A large number of sper-miating males (67%) was achieved 70 days after im-

b

a

aba

010

Fresh sperm from induced sex- reversed males

Cryopreserved sperm from Induced sex- reversed males

Figure 3 Comparison of the mean (SEM) percentage of fertilization and the hatch of orange-spotted grouper usingcryopreserved sperm from sex-reversed males and natural males, and their fresh sperm as control Each treatment was

variance)

Cryopreservation of sperm from Epinephelus coioides T Peatpisut & A N Bart Aquaculture Research, 2010, 42, 22–30

r 2010 Blackwell Munksgaard

plantation This and our study suggest a longer

treat-ment duration and a higher dose requiretreat-ment as the

females become older

Cryopreservation of orange-spotted

grouper sperm

The e⁄ciency of cryopreservation is enhanced if the

prefrozen milt is diluted with a suitable extender

resulting in a minimum toxic e¡ect of the

cryo-protectant (Leung 1991) Trehalose (15^20%) with

150 mmol NaCl appeared to e¡ectively cryopreserve

orange-spotted grouper sperm with the highest

moti-lity, viabimoti-lity, fertilization and hatching obtained in

samples compared with DMSO or PG Miyaki et al

(2005) also found trehalose to be less toxic to kelp

grouper Dimethyl sulfoxide has been shown to be

generally a better cryoprotectant for sperm

cryopre-servation of marine ¢sh species (Suquet, Dreanno,

Fauvel, Cosson & Billard 2000) However, in this

study, trehalose performed better Unlike DMSO and

PG, trehalose is an extracellular cryoprotectant,

re-sistant to hydrolysis and is chemically inert (Colaco,

Kampinga & Roser 1995; Sun & Davidson 1998)

Tre-halose is reported to stabilize phospholipids and

pro-teins during freezing (Crowe, Reid & Crowe 1996)

The non-toxic and stabilizing properties of trehalose

could be the reason for its ability to protect against

cryo-damage during freezing and thawing in our

study

Postthaw sperm quality (i.e motility, viability and

fertilization capacity) was not di¡erent among age

groups of sex-reversed males or between sex-reversed

and natural males In previous studies that induced

sex reversal of marine species by hormone use, only

one study described the quality of cryopreserved

sperm from sex-reversed male Cabrita et al (2009)

tested the postthaw sperm from sex-reversed dusky

grouper While they found that the motility, viability

and fertilization percentage decreased in postthawed

samples of sex-reversed males, they did not measure

hatch percentages Because hatching rates have been

shown not to correlate with motility or fertilization

(Bart et al.1998; Kwantong & Bart 2008), hatching

in-formation is an important measure of

cryopreserva-tion success The per cent hatch obtained from eggs

fertilized with cryopreserved sperm from

sex-re-versed males was the lowest and the reasons for the

lower hatch is di⁄cult to explain, and could be due to

a large number of motility incompetent sperm from

newly sex-reversed males

ConclusionsNot only is it possible to produce male orange-spottedgrouper by hormonal therapy but also it is possible tocryopreserve sperm capable of fertilizing fresh eggssimilar to that of natural males Moreover, this studydemonstrated that younger females converted intomales more quickly than the older ones, and cryopre-servation of their sperm was most successful withtrehalose as a cryoprotectant The methods devel-oped through this study could have a signi¢cant in-

£uence on grouper seed production Application ofthese methods needs to be ¢eld tested in marinehatcheries

AcknowledgmentsThis study is part of a doctoral thesis study at theAsian Institute of Technology, Thailand supported

by the Royal Thai Government fellowship and theAsian Institute of Technology

ReferencesBarry T.P., Castanos M.T & Fast A.W (1991) Induced sper- miation in the male Spotted Scat (Scatophagus argus) by long-term administration of 17a-methyltestosterone fol- lowed by LHRHa Asian Fishery Science 4, 137^145 Bart A.N., Wolfe D.F & Dunham R.A (1998) E¡ects of cryo- protectant, sperm density and straw size of cryopreserva- tion of blue cat¢sh Ictalurus furcatus, sperm Transactions

of the American Fisheries Society 127, 819^824.

Bhandari R.K., Alam M.A., Higa M., Soyanok K & mura M (2005) Evidence that estrogen regulates the sex change of honeycomb grouper (Epinephelus merra), a pro- togynous hermaphrodite ¢sh The Journal of Experimental Zoology 303A, 497^503.

Naka-Billard R., Cosson J & Crim L.W (1993) Motility of fresh and aged halibut sperm Aquatic Living Resources 6, 67–75 Boonyaratpalin M (1997) Nutrient requirements of marine food ¢sh cultured in Southeast Asia Aquaculture 151, 283^313.

Cabrita E., Engrola S., Conceicao L., Pousao-Ferreira M.P & Dinis M.T (2009) Successful cryopreservation of sperm from sex-reversed dusky grouper Epinephelus marginatus Aquaculture 287, 152^157.

Chao N.H.,Tsai H.P & Liao I.C (1992) Short^and long^term cryopreservation of sperm and sperm suspension of the grouper, Epinephelus malabaricus (Bloch and Schnider) Asian Fishery Science 5, 103^116.

Chao T.M & Chow M (1990) E¡ect of methyltestosterone on gonadal development of Epinephelus tauvina (FORSKAL) Singapore Journal of Primary Industries 18, 1^14 Aquaculture Research, 2010, 42, 22^30 Cryopreservation of sperm from Epinephelus coioides T Peatpisut & A N Bart

Chen F.Y., Chow M., Chao T.M & Lim R (1977) Arti¢cial

spawning and larval rearing of the grouper, Epinephelus

tauvina (Forskal) Singapore Journal of Primary Industries

5, 1^21.

Colaco C., Kampinga J & Roser B (1995) Amorphous

stabi-lity and trehalose Science 268, 788.

Crowe L.M., Reid D.S & Crowe J.H (1996) Is trehalose special

for preserving dry biomaterials? Biophysics Journal 71,

2087^2093.

Cryologic (1999) Freezer Control Model-863 Operating

Man-ual Cryologic, Victoria, Australia Available at http://

www.cryologic.com.

Glamuzina B., Glavic N., Skaramuca B & Kozul V (1998)

In-duced sex reversal of dusky grouper, Epinephelus

margin-atus (Lowe) Aquaculture Research 29, 563^568.

Guest W.C (1973) Spermatophore and sperm preservation of

channel cat¢sh Ictalulus punctatus Master of Science

the-sis, The School of Forestry and Wildlife Management,

Louisiana State University, USA, 92pp.

Gwo J.C (1993) Cryopreservation of black grouper

(Epinephe-lus malabaricus) spermatozoa (Teleost, Periformes,

Spari-dae) Theriogenology 39, 1331^1342.

Gwo J.C (1994) Cryopreservation of yellow¢n seabeam

(Acanthropagrus latus) spermatozoa Theriogenology 41,

989^1004.

Kwantong S & Bart A.N (2008) Fertilization e⁄ciency of

cryopreserved sperm from striped cat¢sh, Pangasius

hypophthalmus (Sauvage) Aquaculture research 40,

292^297.

Lee C.S.,Tamaru C.S & Kelly C.D (1986) Technique for

mak-ing chronic-release LHRHa and 17a methyltestosterone

pellets for intramuscular implantation in Fishes

Aquacul-ture 59, 161^168.

Leung L.K.P (1991) Principles of biological

cryopreserva-tion In: Fish Evolution and Systematics: Evidence from

sper-matozoa (ed by B.G.M Jaemison), pp 231^234 Cambridge

University Press, London, UK.

Liao I.C & Leano E.M (2008) Introduction In: The

Aquacul-ture of Groupers (ed by C.I Liao & E.M Lean˜o), 241 pp.

Asian Fishery Society, Department of Aquaculture

National Taiwan Ocean University, Keelung, Taiwan.

Millamena M (2002) Replacement of ¢sh meal by animal

by-product meals in a practical diet for grow-out

culture of grouper Epinephelus coioides Aquaculture 204,

75^84.

Miyaki K., Nakano S., Ohta H & Kurokura H (2005)

Cryo-preservation of kelp grouper Epinephehis moara sperm

using only a trehalose solution Fisheries Science 71, 457^ 465.

Quinitio G.F., Caberoy N.B & Reyes D.M Jr (1997) Induction

of sex change in female grouper Epinephelus coioides by social control The Israel Journal of Aquaculture - Bamidgeh 49,77^83.

Richardson G.F., Crim L.W.,Yao Z & Short C (1995) servation of yellowtail £ounder (Pleuronectes jerrugineus) semen In: Proceedings of the Fifth International Sympo- sium, Reproductive Physiology of Fish (ed by F.W Goetz

Cryopre-& P Thomas), 136 pp Austin, TX, USA.

Sarter K., Papadaki M., Zanuy S & Mylonas C.C (2006) manent sex inversion in 1-year-old juveniles of the proto- gynous dusky grouper (Epinephelus marginatus) using controlled-release 17a-methyltestosterone implants Aquaculture 256, 443^456.

Per-Sun W.Q & Davidson P (1998) Protein inactivation in phous sucrose and trehalose matrices: e¡ects of phase se- paration and crystallization Biochimica et Biophysica Acta

amor-1425, 235^244.

Suquet M., Dreanno C., Fauvel C., Cosson J & Billard R (2000) Cryopreservation of sperm in marine ¢sh Aqua- culture research 31, 231^243.

Tan-Fermin J.D., Garcia L.M.B & Castillo A.R (1994) tion of sex inversion in juvenile grouper, Epinephelus suillus (Valenciennes) by injections of 17 alpha-methyltestoster- one Japan Journal of Ichthyology 40, 413^420.

Induc-Wayman W.R & Tiersch T.R (2000) Research methods for cryopreservation of sperm In: Cryopreservation in Aquatic Species (ed by T.R Tiersch & P.M Mazik), pp 264^275 World aquaculture society, Baton Roue, LA, USA Withler P.C & Lim L.C (1982) Preliminary observations of chilled and deep-frozen storage of grouper (Epinephehis tauvina) sperm Aquaculture 27, 389^392.

Yeh S.-L., Dai Q.-C., Chu Y.-T., Kuo C.-M., Ting Y.-Y & Chang C.-F (2003a) Induced sex change, spawning and larvicul- ture of potato grouper, Epinephelus tukula Aquaculture

228, 371^381.

Yeh S.-L., Kuo C.-M.,Ting Y.-Y & Chang C.-F (2003b) The fects of exogenous androgens on ovarian development and sex change in female orange-spotted protogynous grouper, Epinephelus coioides Aquaculture 218,729^739 Yeh S.-L., Kuo C.-M.,Ting Y.-Y & Chang C.-F (2003c) Andro- gens stimulate sex Change in protogynous grouper, Epi- nephelus coioides: spawning performance in sex-changed males Comparative Biochemistry and Physiology Part C

ef-135, 375^382.

Cryopreservation of sperm from Epinephelus coioides T Peatpisut & A N Bart Aquaculture Research, 2010, 42, 22–30

r 2010 Blackwell Munksgaard

The effects of grading on the growth and survival of

Julia L Overton1, Svend J Steenfeldt2& Per Bovbjerg Pedersen2

1

Section for Coastal Ecology, National Institute of Aquatic Resources,The Technical University of Denmark, Charlottenlund, Denmark

2 Section for Aquaculture, National Institute of Aquatic Resources,The Technical University of Denmark, Hirtshals, Denmark

Correspondence: J L Overton, Aquapri Danmark A/S, Lergrdsvej 2, DK-6040 Egtved, Denmark E-mail: julia.overton@aquapri.dk

Abstract

A 3-month study was carried out to investigate the

e¡ects of grading on the overall production, growth

performance and survival of juvenile Dover sole

(Solea solea L.) Juvenile ¢sh (4.0^40.4 g) were sorted

into three size groups: small (4.0^15.5 g), medium

(16.0^21.5 g) and large (22.0^40.5 g) In addition, a

group of unsorted ¢sh was followed for comparison

The ¢sh from each sorted group and the unsorted

group were divided between triplicate tanks at a

stocking density of1.5 kg m 2 The ¢sh were weighed

and counted 21, 42, 63 and 92 days after stocking In

addition, 30 randomly chosen ¢sh in each tank (590

from each group) were individually tagged The

sur-vival, size distribution, growth and productivity were

calculated for small, medium, large and unsorted

groups In addition, comparisons were made between

combined sorted and unsorted ¢sh There was no

signi¢cant di¡erence between the mean weight and

distribution of sorted and unsorted ¢sh by the end

of the trial An increased overall productivity in

com-bined sorted ¢sh was observed Regular grading could

therefore still be bene¢cial for sole farming as long as

the grading interval supports maximum growth (in

this case over 90 days) Survival was not signi¢cantly

a¡ected by the grading process

Keywords: grading, growth, survival, Dover sole,

Solea solea

Introduction

Through production management, the aquaculture

sector seeks to produce a standardized product at

the maximum yield with a minimal use of resources

(Purdom 1974) An integral part of production

man-agement is size grading, which is expected to improvecontrol of feed usage, optimize feed e⁄ciency, im-prove growth and survival and facilitate postharvestprocessing Previous grading experiments have, how-ever, yielded ambiguous results Size grading duringgrow-out in European eel, Anguilla anguilla L., ex-erted no positive e¡ects on growth or survival (Kam-stra 1993) Similar results have been reported injuvenile turbot, Psetta maxima L., where grading had

no positive e¡ects on speci¢c growth rate (SGR) ormortality (Sunde, Imsland, Folkvord & StefaŁnsson1998) In Japanese £ounders, Paralichthys olivaceus(Temminck & Schlegel), size grading resulted in animprovement in the growth of smaller ¢sh The larger

¢sh, however, showed no improved growth or val when the smaller individuals were removed (Dou,Masuda, Tanaka & Tsukamoto 2004) A similar ¢nd-ing was found in Arctic charr, Salvelinus alpinus L.(Jobling & Reinsnes 1987), where the growth of small

survi-¢sh improved as a consequence of grading This wascompensated, however, by the large size fractionshowing reduced growth, resulting in no overall posi-tive e¡ect on biomass gain with grading Size grading

is also a resource-consuming procedure, which canlead to stress and physical damage (Pickering 1981;Wendelaar Bonga 1997), compromising the welfare ofthe farmed ¢sh, and therefore can only be considered

if there is a signi¢cant improvement in biomass gain

In recent years, the aquaculture industry hasshown interest in commercial rearing of Dover sole(Solea solea L.) and the closely related species, Solea se-negalensis (Kaup), mainly due to their high value andlarge market demand (Howell 1997; Imsland, Foss,Conceicao, Dinis, Delbare, Schram, Kamstra, Rema &White 2003) One of the main hindrances to soleproduction, however, is the relatively slow growthand low stocking density necessary for optimal

performance compared with other £at¢sh species

(Howell 1998; Schram, van der Heul, Kamstra &

Ver-degem 2006) In addition to this, a large disparity in

individual size is a common problem in sole

produc-tion already from the settling phase (Howell 1997) It

would therefore be of interest to clarify whether size

grading could be of bene¢t to sole producers,

improv-ing productivity and growth rates As yet, no

investi-gation has been performed on the e¡ect of size

grading on sole growth performance

The aim of the present study therefore was to

investigate the e¡ect of size grading on the growth

performance, size distribution and production of

juvenile Dover sole

Materials and methods

Production of juveniles

Weaned juvenile sole were produced at the North Sea

Research Park, Hirtshals, Denmark, from 40

brood-stock that were collected from the North Sea and

ac-climatized to spawn in captivity A random single

batch of eggs was collected on 10 June 2004 and

hatched on 11^13 June 2004 (water temperature

and salinity were 10.5 1C and 33 g L 1respectively)

Approximately 25 000 larvae were transferred to a

recirculated system (six 150 L conical tanks) The

lar-vae were reared in this system until weaning (water

temperature was increased from 10.5 to 18 1C over 4^

5 days Salinity was maintained at 33 g L 1) Larval

feeding commenced at 2^5days post hatch (dph) with

type AF Artemia nauplii (INVE, Dendemonde,

Bel-gium) At 6^8 dph, AF Artemia were replaced with

type EG Artemia enriched with Super Selco (INVE)

Feeding took place twice a day The photoperiod was

maintained at16L:8D with a light intensity of1000 lx

at the water surface during illumination From 20 to

22 dph, 11000 larvae were transferred to the

wean-ing facility (also recirculated system) and gradually

weaned over 5 days to dry feed (AgloNorse, EWOS,

West¢eld, UK, size 1^3 granulate) The temperature,

salinity and photoperiod were the same as for larval

rearing On 21 September 2004, approximately 8000

weaned juveniles (101dph) were transported to

Born-holm Salmon Hatchery, Nex, Denmark, for further

study

Experimental design

The ¢sh were divided into four groups: one group of

unsorted ¢sh that represented the control group, and

three size groups, i.e small, medium and large so thatall groups started with the same initial biomass pertank Twenty-¢ve per cent of the initial population of

¢sh was retained as the unsorted group with thesame distribution as the original population of ¢sh.These were individually weighed and equally dividedinto three replicate tanks The size ranges chosen forsmall, medium and large ¢sh groups were based on apreliminary evaluation of the total population weightdistribution (at initial sampling, the ¢sh ranged insize from 4 to 40.5 g) The simplest form of gradingwas to divide the remaining ¢sh into three sizegroups, where a third of the total ¢sh biomass was re-presented in each size group Thus, the remaining

¢sh were individually weighed and sorted into small(4.0^15.5 g), medium (16.0^21.5 g) and large (22.0^40.5 g) ¢sh (Table 1) From these three sortings, tripli-cate tanks were formed by further randomly dividingeach of the groups into three tanks This resulted inall tanks being stocked with the same initial biomass

of 1.5 kg m 2(approximately 409, 586, 866 and 669

¢sh per tank for large, medium, small and unsortedgroups respectively) This was calculated to beequivalent to 33% tank £oor coverage The weightdistribution was equal between replicates for eachweight group [analysis of variance (ANOVA), P40.05].Any deformed/damaged ¢sh were removed duringthe individual weighing and sorting process The ¢shwere allowed to acclimatize to their rearing condi-tions over one week before the start of the experi-ment At the start of the experiment, the totalweight of ¢sh per tank was registered and adjusted

to ensure that the total weight of ¢sh was dized across all tanks In addition, day 29 into the ex-periment, 30 individuals from each tank wererandomly chosen to be tagged The ¢sh were taggedusing numbered external T-bar tags (Hallprint,Victor

standar-Table 1 Number and mean wet weight (g) (including dard deviation (SD) from mean) of juvenile Dover sole (Solea solea) in each sorted group and unsorted group

Weight range (g)

Mean (SD)

No of tagged fish

Small (Ss) 1227 4.0–15.5 11.8 (6.2) 90 Medium (Sm) 1759 16.0–21.5 18.4 (1.7) 90 Large (Sl) 2599 22.0–40.5 26.9 (4.6) 90 Unsorted (Su) 2007 4.0–40.5 16.2 (6.2) 90 Combined sorted

(Ss1Sm  Sl)

5585 4.0–40.5 17.2 (6.5) 270 Combined large, medium and small ¢sh are included.

r 2010 The Authors

Harbour, Australia) that were inserted posterior to

the dorsal ¢n, a method that has been successfully

used for turbot juveniles (Stttrup & Sparrevohn

2007) The ¢sh were sedated before tagging using

MS222 (100 mg L 1) as performed by Reig, Ginovart

& Floss (2003) These external tags allowed easy to

identi¢cation of ¢sh for individual weighing, which

took place simultaneously with the total weighing of

the whole tank of ¢sh Tagging has been shown by

Reig et al (2003) not to negatively a¡ect the growth

of sole juveniles Any mortality was recorded daily

Rearing conditions

Each replicate was reared in 3 m diameter green

¢bre-glass tanks (tank £oor area 5 7.07 m2) with an

approx-imate water exchange rate equal to two tank volumes

per hour The ¢sh were fed on 1.5 mm commercial

pel-leted feed (IDL Solea, INVE,55% protein and 16% lipid)

The daily feed ration was based on a percentage of the

total wet weight of ¢sh per tank, starting at 1.5% body

-weight day 1 and decreasing to 0.8% body weight

day 1by the end of the experiment These percentages

ensured that maximal food intake was possible (some

food was observed on the tank £oor after feeding)

Feed-ing took place over 12 h from dawn until dusk usFeed-ing

spin wheel feeders to ensure an even distribution of feed

delivered over the surface area of the tank The day

be-fore weighing, no feed was administered Daylight

hours (12 L:12 D) were controlled using indirect

£uor-escent lighting, resulting in a light intensity of 100 lx at

the water surface Water temperature was maintained

constant at 20 1C The oxygen concentration was

47 mg L 1during the entire experiment Water

sali-nity was maintained at 30 g L 1using arti¢cial salt

water (a mixture of 50% aquarium salt (Red Sea, Israel)

and 50% pure production salt (99% NaCl without

anti-clumping agents, Brste, Kongs Lyngby, Denmark ) The

system water was fully recirculated with

supplemen-tary exchange of new water of around 7% of the total

water volume per week The total ammonia nitrogen

(TAN), nitrite and nitrate concentrations were tested

three times weekly (Hach water quality ¢eld kit, Hach

Lange, Berlin, Germany) and were found to be at

ac-ceptable levels (Alderson 1979; Parra & Yu¤fera 1999;

Pinto, Aragao, Soares, Dinis & Conceicao 2007)

dur-ing the entire experiment (TANo0.0001mg L 1,

NO2-No0.09 mg L 1and NO3-No150 mg L 1)

Sampling

At the start (day 1) and end (day 92) of the

experi-ment, individual wet weights were recorded for all

¢sh from each group The total wet weight (kg) wasalso recorded at the beginning and during the study,

21, 42 and 63 days after stocking the ¢sh In order toeliminate the e¡ect of density on growth, which hasbeen shown to be an important factor in the growth

of sole (Schram et al 2006), after each weighing, thebiomass in the tanks was readjusted so that all tankshad the same biomass using the smallest tank bio-mass at each weighing to calibrate the other tanks.Tagged ¢sh were weighed on day 29 (i.e day of tag-ging) and thereafter the same day as total weighing

of the ¢sh (days 42, 63 and 92 from the start of theexperiment)

Calculation of growth and productivityThe mean wet weight was calculated for each treat-ment for each sampling point by dividing the totalweight of ¢sh by the number of ¢sh per tank.The mean weight was compared among treatmentsover time

Growth expressed as the SGR was calculated foreach weighing interval using the data from individu-ally tagged ¢sh using the following equation:SGRð%D1Þ ¼ ½ExpððLnWtLnW0Þ  t1Þ  1

where LnW represents the natural log of individualbody mass, Wtis the ¢nal wet weight (g) of ¢sh, W0the initial wet weight (g) and t the number of days.The productivity (P) was calculated on tanks of

¢sh for each size group using all ¢sh in each tank asfollows:

where S represents the tank surface area (m2) Theother parameters are as described for Eq (1) The pro-ductivity for combined sorted (i.e small, medium andlarge ¢sh combined) versus unsorted ¢sh was alsocalculated for each time frame In addition to this,the overall mean productivity (P days 1^92) was cal-culated for combined sorted versus unsorted ¢sh

Statistical analysisThe normality of frequency distributions for tanks of

¢sh individually weighed at the beginning and at theend of the experiment was tested using the Kolomo-gorov^Smirnov test Where the tanks of ¢sh werefound not to be normally distributed, the comparisonbetween the weight groups for all ¢sh individually

weighed at the beginning and at the end of the

ex-periment was tested using Kruskall^WallisANOVAon

Ranks, followed by an all pairwise multiple

compari-son procedure using Dunn’s method Coe⁄cients of

variation (CV) of wet weight were calculated for all

groups at the beginning and at the end of the

experi-ment using the following equation:

CV¼ Standard deviation/mean weight ð3Þ

In order to observe the overall e¡ect of grading ¢sh,

the frequency distributions of combined individual

weights of all ¢sh from small, medium and large ¢sh

were compared with the individual weights of three

unsorted tanks for the ¢nal sampling using a t-test

The mean wet weights (g) and mortality were

tested for signi¢cant di¡erences among groups using

a one-wayANOVA, followed by a pair-wise comparison

between groups for signi¢cance using the Tukey test

(Po0.05)

Comparison of SGR between groups (using

indivi-dual growth data) was performed using a nested

two-wayANOVA(STATISTICA, version 8, StatSoft, Tulsa,

OK, USA)

Productivity was tested between groups and

between time frames sampled using two-wayANOVA

(SIGMA STAT, version 2.03, SPSS, Chicago, IL, USA)

Results

Size frequency distribution of ¢sh

A larger mean weight and wider distribution of all

groups were observed by the end of the experiment

(Fig 1) Graded groups covering three to ¢ve weight

classes at the start had formed what could be termed

as a trend towards a normal distribution by the end of

the experiment, although none of the distributions in

the ¢nal groups passed the normality test (P40.05)

The ¢nal median weights of ¢sh were signi¢cantly

high-er for large groups (52 g 3.61) and signi¢cantly lower

for small groups (26 g 3.06) compared with medium

and unsorted groups (41g 1.89 and 32 g  2.65)

Medium ¢sh were overall larger than unsorted ¢sh by

the end of the experimental growth period At the end

of the experiment, the frequency distribution (Fig 2)

and the overall mean wet weight for combined sorted

groups compared with unsorted ¢sh were not found to

be signi¢cantly di¡erent (P 5 0.99)

Survival

The overall survival of ¢sh over the experimental

per-iod was between 76 and 85% The highest mortality

was recorded in the small ¢sh group (24%), followed

by the medium ¢sh (20%), and the least mortalitywas seen in the unsorted and large ¢sh groups (16%and 15% mortality respectively) The peak of mortal-ity was observed two weeks after the start of the ex-periment

GrowthThe mean wet weight over time for each treatment(Fig.3) was the highest for large ¢sh, followed by med-ium, unsorted and small ¢sh groups, supporting theresults from the frequency distribution The meanwet weight for large ¢sh remained signi¢cantly high-

er through the entire experiment until the lastweighing, where they, due to an increase in standarddeviation, were not signi¢cantly di¡erent from med-ium-sized ¢sh Both small and unsorted ¢sh re-mained signi¢cantly smaller than the large andmedium groups

The SGR based on individually tagged ¢sh creased over time for all groups from around 0.7^0.86% body weight increase day 1in the ¢rst period

de-to 0.22^0.27% body weight increase day 1in the

¢nal month of the study There were no signi¢cantdi¡erences in SGR between treatments throughoutthe experiment (Fig 4)

Fish biomass producedSorting had no signi¢cant e¡ect on the overallamount of ¢sh biomass produced per tank through-out the experimental period The average productivitywas calculated to be 16.11 ( 3.99), 19.08 ( 2.20),18.17 ( 2.74) and 19.49 ( 0.64) g m 2day 1 forlarge, medium, small and unsorted groups respec-tively There was no signi¢cant di¡erence betweenthe productivity for combined sorted and unsorted

¢sh for each growth period (Fig 5) By the last pling date, however, the combined sorted ¢sh revealed

sam-a trend towsam-ards better productivity thsam-an unsorted

¢sh

DiscussionFrequency distributionThe ¢nal frequency distributions of individual wetweights for sorted groups of ¢sh showed a trend to-wards what could be termed normal distributions(Fig 1) This has also been observed for juvenile

r 2010 The Authors

CV = 20.08Skewness = –0.30Kurtosis = –0.95

Median = 18.00Mean = 18.40

CV = 9.08Skewness = 0.21Kurtosis = –1.14

Median = 25.50Mean = 26.90

CV = 17.03Skewness = 1.37Kurtosis = 1.73

Median = 16.00Mean = 16.20

CV = 38.36Skewness = 0.36Kurtosis = –0.38

Median = 26.50Mean = 27.30

CV = 33.80Skewness = 0.36Kurtosis = 0.42

Median = 41.00Mean = 41.90

CV = 23.92Skewness = 0.43Kurtosis = 0.46

Median = 52.00Mean = 53.10

CV = 23.82Skewness = 0.53Kurtosis = 0.45

Median = 32.00Mean = 33.90

CV = 39.45Skewness = 0.86Kurtosis = 1.97

Figure 1 Size frequency distributions at the beginning and at the end of rearing trial for juvenile Dover sole Fish weredivided into small, medium, large and unsorted groups at the beginning of the experiment K^Sdis, the goodness of ¢t to anormally distributed population (Kolmogorov^Smirnov distance) Data on mean, median, coe⁄cient of variation (CV)skewness and kurtosis for each treatment are provided

turbot 85 days post grading (Strand & iestad 1997).

The observed increased width of the frequency

distri-butions is most likely the result of individual

di¡er-ences in the growth rate (Magnuson 1962; Abbott &

Dill 1989) It has been suggested that individually

ob-served variations in growth are caused by individual

physiologically based food conversion e⁄cacies and

speci¢c growth rates (Abbott & Dill 1989; FitzGerald,

Flanagan, Brennan, Haraldsson, Clarke, Irwin, Roby,

Ni Chara & Cross 1996), while others believe that it isprimarily due to both food availability and the indivi-dual’s ability to defend this resource from others(Koebele 1985)

Growth performanceSorting was shown to have no overall signi¢cant ef-fect on the growth performance of sole (Fig 4) Thesmall ¢sh continued to show a poorer growth perfor-mance throughout the experiment than was ex-pected even with the removal of large and medium

¢sh This seems to contradict the ¢ndings of othergrading studies (e.g Abbott, Dunbrack & Orr 1985;Koebele1985; Joblin & Reinsnes 1987; Gri⁄ths & Arm-strong 2002) including those on £at¢sh (e.g Sunde

et al 1998) One reason for the slightly poorer mance in smaller ¢sh relative to large and medium

perfor-¢sh could be due to increased sensitivity to a higherstocking density Increasing the stocking density forsole (from 5 to 131% £oor coverage) has been shown

to depress growth in the smaller individuals, whereasthe growth rates of the larger ¢sh are largely unaf-fected (Howell 1998; Schram et al 2006) Another ex-planation could be due to the use of a restricted food

0-4,99 5-9,9910-14,99 15-19,99 20-24,99 25-29,99 30-34,99 35-39,99 40-44,99 45-49.99 50-54,99 55-59,99 60-64,99 65-69,99 70-74,99 75-79,99 80-84,99 85-89,99 90-94,99 95-99,99

100-104,99 105-109,99

0246810

Size group (g)Figure 2 Size frequency distributions of sorted and unsorted Dover sole at the end of the experimental period (day 93)

Days from start of experiment

a a b

b b

b

c

c

bc d

c c

c c

b b

Figure 3 Mean weight (SD) over time for graded (small,

medium, large) and unsorted groups of juvenile Dover

sole

r 2010 The Authors

particle size, which has been shown to favour the

growth of ¢sh of the appropriate size (Knights 1983),

although the particle size chosen was recommended

by the feed supplier for the size range of ¢sh used in

the experiment Although the food was well

dis-persed in the tanks and some food was observed onthe tank £oor for an extended period of time, thisfood may not be interesting to the sole after sometime (as described by Schram et al 2006); therefore,food availability may still be limited for those tanks

Period (days from start of experiment)

1.4

Large Medium SmallUnsorted

Figure 4 Speci¢c growth rate (SGR) (SD) for graded (small, medium, large) and unsorted groups of juvenile Dover sole

rate for entire monitoring period

Period (days from start of experiment)

30

SortedUnsorted

with larger numbers of ¢sh (e.g small ¢sh), and

thus increased competition between individuals

could occur

Productivity

Productivity based on tank biomass was not

signi¢-cantly di¡erent between combined sorted and

un-sorted ¢sh, although there was a tendency towards

a higher biomass gained in the combined sorted

groups from day 17 and onwards and for the overall

productivity (days 1^92) (Fig 5) Previous studies in

other ¢n¢sh species have shown that sorting does

not necessarily lead to increased biomass gain

(Jobling & Reinsnes 1987; Baardvik & Jobling 1990;

Strand & iestad 1997) The expected increase in

pro-duction as a response to the removal of the

domi-nance e¡ect of larger individuals is most likely o¡set

by the need to reestablish a hierarchy in the

remain-ing ¢sh, resultremain-ing in energy used to reestablish

hier-archy (through aggression) rather than allocating it

to growth Dover sole is a species of ¢sh characterized

by a high level of social interaction and therefore it is

not surprising that hierarchy has some in£uence on

their social structure (StefaŁnsson, FitzGerald & Cross

2002) Because of the ¢rst signs of increased

produc-tivity in combined sorted groups being observed 3

months after the start of the project, we assume that

the hierarchical resolution phase (i.e the time taken

to reestablish hierarchy in a group of ¢sh) is longer

than 90 days StefaŁnsson et al (2002) describe a

resolution phase of 185 days for juvenile turbot An

extended experiment to ¢nd this point of resolution

for sole would be useful in determining the optimal

grading frequency

Mortality according to size groups

Mortality was not signi¢cantly di¡erent between

groups; however, it was observed that the smaller ¢sh

in each category were more susceptible to mortality

Schram et al (2006) also observed that mortality in

juvenile sole in general took place in below

average-sized ¢sh The reasons for this elevated mortality in

smaller ¢sh could include the general fact that

indivi-duals with elevated metabolic rates are more

aggres-sive, due to their higher metabolic scope (Priede 1985;

Metcalfe,Taylor & Thorpe 1995), and use this

aggres-sive behaviour to dominate smaller individuals, and

that social stress can result in secondary responses

in subordinate ¢sh including depletion of glycogen

deposits due to stress handling and potential mulation of waterborne toxicants (Sloman, Baker,Wood & McDonald 2002) Compared with other stu-dies on Dover Sole, this study observed a higher mor-tality than would be expected (15^24% mortality inthis study compared with 0^18% observed bySchram et al 2006) This higher mortality occurredjust after the hand sorting and individual weighingprocedure The physical process of handling and sort-ing of ¢sh is stressful, and should be taken into con-sideration when grading is being suggested as anoption to improve the overall production

accu-ConclusionsThere was no signi¢cant bene¢t of sorting, observed

in terms of the overall growth rate or the ¢nal meanweight of ¢sh when comparing combined sorted ¢shwith unsorted ¢sh However, a trend towards in-creased overall productivity after sorting, following

an initial decline, could be achievable after 90 days.Regular grading could therefore still be bene¢cial forsole farming as long as the grading interval supportsmaximum growth The sorting procedure did notdetrimentally a¡ect the ¢sh when taking into theconsideration of the basic handling that all groupsreceived

Acknowledgments

We wish to thank Jim Hee, John Hkonsson and ClausJespersen (technicians at Bornholms Salmon Hatch-ery, Nex) and Lars- Flemming Pedersen (DTU Aqua)for their technical support in carrying out this study

We would also like to thank Dr Erik H˛glund, DTUAqua, for his invaluable comments on the manu-script The study was funded by the Financial Instru-ment for Fisheries Guidance programme and theDanish Ministry of Fisheries, under the administra-tion of the Directorate for Food and Agribusiness

ReferencesAbbott J.C & Dill L.M (1989) The relative growth of domi- nant and subordinate juvenile steelhead trout (Salmo gairdneri) fed equal rations Behaviour 108, 104^113 Abbott J.C., Dunbrack R.L & Orr C.D (1985) The interaction

of size and experience in dominance relationships of nile steelhead trout (Salmo gairdneri) Behaviour 92, 241^ 253.

juve-Alderson R (1979) The e¡ect of ammonia on the growth of juvenile Dover sole, Solea solea (L.) and Turbot, Scophthal- mus maximus (L.) Aquaculture 17, 291^309.

r 2010 The Authors

Baardvik B.M & Jobling M (1990) E¡ect of size-sorting on

biomass gain and individual growth rates in Artic charr,

Salvelinus alpinus L Aquaculture 90, 11^16.

Dou S-Z., Masuda R.,Tanaka M & Tsukamoto K (2004) Size

hierarchies a¡ecting the social interactions and growth

of juvenile Japanese £ounder, Paralichthys olivaceus.

Aquaculture 233, 237^249.

FitzGerald R.D, Flanagan M., Brennan C., Haraldsson H.,

Clarke J., Irwin S., Roby F., Ni Chara U & Cross T.F (1996)

Research on farmed turbot (Scophthalmus maximus) in

Ireland Irish Marine Science 1995, 407^422.

Gri⁄ths S.W & Armstrong J.D (2002) Kin-biased territory

overlap and food sharing among Atlantic salmon

juve-niles Journal of Animal Ecology 71, 480^486.

Howell B.R (1997) A re-appraisal of the potential of the sole,

Solea solea (L.) for commercial cultivation Aquaculture

155, 355^365.

Howell B.R (1998) The e¡ect of stocking density on the

growth and size variation in cultured turbot,

Scophthal-mus maxiScophthal-mus, and sole Solea solea ICES CM 1998/L:10,

1–10.

Imsland A.K., Foss A., Conceicao L.E.C., Dinis M.T., Delbare

D., Schram E., Kamstra A., Rema P & White P (2003) A

review of the culture potential of Solea solea and

S senegalensis Reviews in Fish Biology and Fisheries 13,

379^407.

Jobling M & Reinsnes T.G (1987) E¡ects of sorting on size

frequency distributions and growth of arctic charr,

Salve-nius alpinus L Aquaculture 60, 27^31.

Kamstra A (1993) The e¡ect of size grading on individual

growth in Eel, Anguilla anguilla, measured by individual

marking Aquaculture International 1, 90^93.

Knights B (1983) Food particle size preferences and feeding

behaviour in warm water aquaculture of the European

eel, Anguilla anguilla (L.) Aquaculture 30, 173^190.

Koebele B.P (1985) Growth and the size hierarchy e¡ect: an

experimental assessment of three proposed mechanisms;

activity di¡erences, disproportional food acquisition,

physiological stress Environmental Biology of Fishes 12,

181^188.

Magnuson J.J (1962) An analysis of aggressive behaviour

and competition for food and space in medaka (Oryzias

latipes), Pisces, Cyprinodontidae Canadian Journal of

Zool-ogy 40, 313^363.

Metcalfe N.B.,Taylor A.C & Thorpe J.E (1995) Metabolic rate,

social status and life-history strategies in Atlantic

salmon Animal Behaviour 49, 431^436.

Parra G & Yu¤fera M (1999) Tolerance response to ammonia and nitrite exposure in larvae of two marine ¢sh species (Gilthead seabream Sparus aurata L and Senegalese sole Solea senegalensis Kaup) Aquaculture Research 30, 857^863.

Pickering A.D (1981) Introduction: the concept of biological stress In: Stress and Fish (ed by A.D Pickering), pp 1^9 Academic Press, London, UK.

Pinto W., Aragao C., Soares F., Dinis M.T & Conceicao L.E.C (2007) Growth, stress and free amino acid levels in Sene- galese sole (Solea senegalensis Kaup 1858) chronically ex- posed to exogenous ammonia Aquaculture Research 38, 1198^1204.

Priede I.G (1985) Metabolic scope in ¢shes In: Fish getics: New Perspectives (ed by P Tytler & P Calow), pp 33^64 Croom Helm, London, UK.

Ener-Purdom C.E (1974) Variation in ¢sh In: Sea Fisheries search (ed by F.R Haredn Jones), pp 347^355 Elek Science, London, UK.

Re-Reig L., Ginovart M & Floss R (2003) The application and evaluation of an individual marking techniques of sole

¢ngerlings (Solea sp.) for studies of growth Journal of plied Ichthyology 19, 49^51.

Ap-Schram E.,Van der Heul J.W., Kamstra A & Verdegem M.C.J (2006) Stocking density-dependent growth of Dover sole (Solea solea) Aquaculture 252, 339^347.

Sloman K.A., Baker D.W.,Wood C.M & McDonald D.G (2002) Social interactions a¡ect physiological consequences of sublethal copper exposure in Rainbow trout, Onchor- hynchus mykiss Environmental Toxicology and Chemistry

21, 1255^1263.

StefaŁnsson M.., FitzGerald R.D & Cross T.F (2002) Growth, feed utilization and growth heterogeneity in juvenile turbot Scophthalmus maximus (Ra¢nesque) under di¡erent photoperiod regimes Aquaculture Research 33, 177^187.

Strand H.K & iestad V (1997) Growth and the e¡ect of grading of turbot in a shallow raceway system Aquacul- ture International 5, 397^406.

Stttrup J.G & Sparrevohn C.R (2007) Can stock ment enhance stocks? Journal of Sea Research 57,104^113 Sunde L.M., Imsland A.K., Folkvord A & StefaŁnsson S.O (1998) E¡ects of size grading on growth and survival of juvenile turbot at two temperatures Aquaculture Interna- tional 6, 19^32.

enhance-Wendelaar Bonga S.E (1997) The stress response in ¢sh Physiological Reviews 77, 591^625.