SELECTNE FEEMNO llJ BIVALVES: MIIPLICATIOMS FOR THE ACCUMULATIûN OF

TOXK: TRACE METAL8 BY T M FILTERIEEDIM MUSSEL (MyWus (rouulus)

A THESIS 8Ue"ED IN PARTIAL FULFILi-'UNT OF THE REQUIREMEHTS FOR THE DEOREE OF

OOCTOR OF PH(LO8OPHV

M iigM mamd. Thlr wMk nuy not k npmdwadînrrhdrorhprrSbyphobcopy

National Library Bibliothèque nationale du Canada

Acqubitioris and Acquisitions et Bibliographie Srmices sewices bibliographiques

The author has granted a non- exclusive licence allowing the National Library of Canada to reproduce, loan, distribute or sel1 copies of this thesis in microfom, paper or electronic formats.

The author retains ownership of the copyright in this thesis. Neither the thesis nor substantial extracts from it may be printed or otherwise repduced without the author's permission.

L'auteur a accordé une licence non exclusive permettant à la Bibliothèque nationale du Canada de reproduire, prêter, distribuer ou vendre des copies de cette thèse sous la forme de microfiche/film, de reproduction sur papier ou sur format électronique.

L'auteur conserve la propriété du droit d'auteur gui protège cette thèse. Ni la thèse ni des extraits substantiels de celle-ci ne doivent être imprimés ou autrement reproduits sans son autorisation.

iii

ABSTRACT

The objectives of this thesis are two fold: first to detemine the role of feeding

behavior in influencing the accumulation of cadmium (Cd) by mus!iels, and second to

incorporate the determined feeding behavior into an existing kinetic modal of metal

accumulation to allow for a more accurate pradiction of trace metal accumulation by the

b! ue mu ssel ( Mflilus Imssu!us).

The first obhctive was met by complding three laboratory expenments to

answer the following questions; 1) are mussels selectnie in their feeding? 2) if so what

is the implication of this femding behavior on Cd assimilation effciency?, end 3) is then

any cost associateâ with softing their ingested diet? The fimt laboratory expriment

showed that in responw to environmentally nlevant seston matrices, mussels were

capabîe of two feeding strategies, which are not rnutually exclusive. Under high quallty

and quantity of wston, the mussels were capable of selecting only the organic

component of seston and repcting the inorganic component (through pwudofaeces

production), Sorting etliciency was maximal at a aston quality of 40% organic matter.

The second experimnt detemined the influence of seletdive feeding on the

'%d arshilation eficbncy ('''C~-AE) by the blue mussel. The studies indicated two

patterns in ' w ~ d - ~ ~ . When mussels were exposed to a diet whem only algae had

k e n Iabeled, 'Oecd arrimilation was proportional to the organic content of the diet (r = 0.98; p < 0.05) with maximum assimilation occuning at the maximum cleannœ rates.

Howevet, when the d l was Iaôekd, 'Oecd-~~s wen independent of diet quality with

maximum values of -85% except for silt alone treatrnenb ( '%~AES = 36%). These

resuits 8ugge8t an active and pauive assimilation of '-cd from the algae and silt

components of seston, respectively

The third labotatory experimnt tested whether then was a cost associated with

the wkctive fwding procesa by the blue muswl. The study showed that them was no

signifiant diffemnœ betweem the metabolit costs associateâ with sorting during

iv setective or non-sekctive feeding. The metabolic cost associateâ with feeding was a

maximum of 0.44 J h*' and cornprised only 0.9% of the gross energy intake.

The second objective of the thesis was achkveâ by incorporating the resuîts of

above laboratory studbr into a kinetic model of metal accumulation. Under steady

state conditions. Cd bioaccumulation in mussetl tissues was better predicted when

sslective feeding was included as a pproc8ss within the kinetic-based model. Dietary Cd

uptake contributed 35 to 94% of the total Cd concentration in the mussel tissues.

Micated to My &ar w h , /na Pranoto, Ibr her patience and encouragement

and My missing bmfher, Andi Ridwan in West Java, 4 Nov 1998

ACKNOWLEDGMENTS

My graduate education and research were fundeâ by Canadian International

Development Agency (CIDA) grants through ASEAN - Canada Marine Science

Program and SFU Gnduate Fellowship, which support I gratefully acknowledge. I am

deeply grateful to Dr. Leah 1. BendeIl-Young, rny setnior supsrvisor, for her continued

encouragement and patience, and for hrr clsssroom lectures on Environmental

Toxicokgy and Ecotoxicology that provided foundations for rny work. Many thanks go

to Dr. Lawrence Dill and Dr. Brian Hartwick, who served on my Doctoral Cornmittee, for

their helpful technical suggestions on mussel sarnpling, laboratory ikills, and critical

review of the dissertation. My special appreciation goes to Dr. Paul J. Harrison for his

valuabb advice at al1 stages of thir woik and for his help in introducing me to

microalgae culture technique.

This research would not have been possible without the help of rnany people.

Christopher Gugliemo showetd me the technique on micro-bomb calorimtry. Dr. Jean-

Claude Brodovitch and W. Sharon Hope hetlped wt-up and w e n always nody to

answer problems on the gamma counter. Kimty Bennett, Murray MacDonald and Bruce

Leighton edited my earlier thesis dnfts. P. Stecko. Christine Thomas, Cnig Harris, Lori

Barjactarovic, K. Bennett, Ingrid Polkt hrlpeâ in some stage of my work. Akhmad

Fauzi. A n l Fauzi and Hen Pumomo were always mady to help me in mony wayr.

Thank you so much for al1 your help.

Terima karih yang dalam untuk dua pasang onng tua tercinta, drngan

limpahan doa dan dorongan semangat mu, okhirnya bagian dari tugas hidup ini telah

terlewati. Semoga YME mmberikan balarrn yang wtimpal buat kalian. Finally. I

acknowkdge my wife. Ina. for hot fommo8t support. na88uranm and affection have

gnatly enriched my joumey of life.

vii

TABLE O f CONTENTS

Approval Page

Abstract

Dedication

Acknowledgrnents

Table of contents

List of Tables

List of Figures

List of Appendices

CHAPTER I : GENERAL INTRODUCTION

1.1. RATIONALE FOR THE THESIS

1.2. OBJECTIVES OF THE STUDY

1.3. ORGANIZATION OF THE THESIS

1.4. REFERENCES

CHAPTER II. FEEDING RESPONSE AND CARBON ASSIMILATION BY THE BLUE MUSSE1 Myfilus trossulus EXPOSE0 TO ENVIRONMENTALLY RELEVANT SESTON MATRICES

Abstract

2.1. tNfRODUCTlON

2.2. MATERIALS AND METHOOS

2.2.1. Experimntal animals

2.2.2. Erprimcmtal apparatus

2.2.3. Seton (Surpendd paiüculate matter)

2.2.4. Se8ton C:N ratio

2.2.5. Physiological msaruremsnh

2.2.6. Statistical analyris

2.3. RESULTS

2.3.1. Quantity and qwlity of serton matricer

Page

i i

iii

v

vi

vii

~i

xiii

xvii

vüi

2.3.2. Effect of seston matrices on clearance rates

2.3.3. Effect of seston matrices on sortkig efficiency

2.3.4. Effect of seston matrices on pwudofaecas production

2.3.5. Effect of seston matrices on PO& and PlMc ingestion rate

2.3.6. Effect of wston qurlity on carbon assimilation efficiency (CAE)

2.4. DISCUSSION 2.4.1. Behavtoral response of the blue mussel to a

changing food environment 2.4.2. Assimilation of carbon from vsrying wston

matrices

2.5. REFERENCES

CHAPTER III. INFLUENCE OF A SELECTIVE FEEOING BEHAVIOR BY THE BLUE MUSSEL Mytilus tmssuIus ON THE ASSIMILATION OF 'Oecd FROM ENVlRONMENTALLY RELEVANT SESTON MATRICES

Abstract

3.1. INTRODUCTION

3.2. MAT ERlALS AND METHODS

3.2.1. Field collection of mussels

3.2.2. Seston composition

3.2.3. Pulw chase-fwding expedmnts

3.2.4. Seston qwlity (SQ) versus diet quality (DQ)

3.2.5. Estimates of 'Oecd assimilation efficiency

3.2.6. 'Oscd aCüvity in mussel tissue

3.2.7. Statistical analysis

3.3. RESULTS

3.3.1. Radionuclide biodrporition mtes

3.3.2. Cadmium msirniktion effciency ('%dA€)

3.4. DISCUSSION

3.4.1. '%&A€ in relation to a wkctive feeâing khavior

3.4.2. '%d in musseh in relation to 'Oscd -A€

3.5. REFERENCES

CHAPTER IV. METABOLIC COST O f FEEDlNG BY THE BLUE MUSSEL Mytilus t ~ u l u s OISPLAYING SELECTM AND NON-SELECTIVE FEEDING BEHAVIOR

Abstract

4.1. INTRODUCTION

4.2. MATERIALS AND METHODS

4.2.1. Caloric content of the diatom

4.2.2. Seston treatments

4.2.3. Acclimation of mussels

4.2.5. Cakuktion of swp for growth

4.2.6. Data analysis

4.3. RESULTS

4.3.1. Caloric content of diatom

4.3.2. Effact of seston matrices on respiration rates

4.3.3. Effect of seston matrices on ammonium excretion rates

4.3.4. Swpe for growth

4.4. DISCUSSION

4.5. REFERENCES

CHAPTER V. APPLICATION OF A KINETIC MOOEL OF METAL ACCUMULATION INCORPORATING MUSSEL FEEOING BEHAVIOR

Abstract

5.1. INTRODUCTION

5.2. MOOELING APPROACH

5.3. MATERIALS AND METHODS

5.3.1. ml panmeterllitkn

5.3.2. Experimental transplant

5.3.3. Cadmium analyris

5.4. RESULTS

S.S. DISCUSSION

5.6. REFERENCES

CHAPf ER VI. OVERACL CONCtUSlONS AND RECOMMENOATIONS 155

LIST OF TABLES

Table Page

2.1 Seston concentration (mg L") and mston quility (% POM) of 20 coastal easystem8. Monthly nnasummnt of seston for 12-18 months, 3 Sampling et 1-2 cm above the substmte over tidal cycles, 7 Sampling at 0.5 m above the substrate sudace over tidal cycles.

2.2 Chsracteristics of the seston (SPM) rnotricss (mean * 1 SE, n = 3). 32 SM?, = suspendeâ particukte rnatter, PO& = particulate orgrnic matter, SQ = seston quality, C:N ratio = caiboninitrogen ntio in dry weight, Tukey's multiple nnge test was done on SQ data and the significant dinennces are denoted as different ktters, 8.D.c? The silt conœntntions of 5,20 and 50 mg L" reprewnt 149,482 and 1267 x 10' particles L*' , respectively .

2.3 Behavioral and physiokgical data on mursels. CR = clearance 33 rate, PÇ, = pmudofeces production, IR, = ingestion rate. apparent and true C A € = carbon arsimibtion efficiency. Values a n Mans î ISE, n = 3. Algae : silt = (10' cells L*': mg L"). nd = no data (seston concentration was ôelow critical mass required to produce pseudofaeces). The JI concentrations of 5, 20 and 50 mg C' nprewnt 149,482 and 1267 x 10' particles 1". mspectively.

3.1 Characteristics of SPM and feeding tesponm of rnusmls under 63 diîîerent SPM matrices. Aîgae : $il= (x 10' cells L-' : mg c'), SQ = seston quality, W = diet quality, the actual food ingested by musireh after the rorting prowss (Chapter II); IR, = ingestion rate of organic matter; IRw, = ingestion rate of inorganic rnatter. tnir C- A€= Caibon arsimilation effciency. a, b, c and d ars man8 t i SE, n =3.

3.2 109 Cd and 2 4 1 ~ m in wston (SPM), preudofaec~s (PF) and faeces 72 (FE). Aîgae: siît = (X 10' ceIl L" : mg L-'), The iilt concentrationr of 5 2 0 and 50 mg L" mpmwnt 149,482 and 1267 x 10' particies L". re8pctively. Values = (man t SE. n = 3), SPM = suspendd parüculate matter (seston), PF= pmudofaecm, FE = faeces , 09 Cd-AE= cadmium aaimilation eiciency. Exp. I = expedmnt with

Iabed aigae, Exp. II = expehment with l a k b d Mt.

3.3 ûetennination of '%d assimilation efficîency ( 'Os~d-~€) b a d on 78 the dual-tnœt ratio mthod LDTR) and inge8tion rate ma8uremnt (IRM). (CdlAm)= ntb of ' Cd to " ' ~ m in food or in Ireces; (IR/FE)-= ntio of '"@cd in food ingested to '#cd in faeœs; (IRIFE),,,,,= ratio of in ingsrted food to ' "~m in f a n ; Cd-

L ) . The riît concentrations of 5, 20 and 50 mg L" siît npmsent 149,482 and 1267 x ldpaiackr L", mspsctively; values = (means I 1 SE, n = 3).

Cadmium assimilation effichncy (CdAE) in musmlr (Myfilus edulis, M. tmssulus), oyster (Cressostm vitginaca) and clams (Metcenana mercenaria, Mewma balthica, Potamcohu/a emurensis)

Selecteâ seston matrices and feeding petameten from pnvious study (Chapter II). Atgw: SM = ~10' wlls L" : mg 1". CR= ckerancc rate (mens I 1 SE. n = 3). SQ = seston quality. W = dkt quality (average values), app. GAES = apparent carbon assimilation efficiency, tnie G A € % = ttue carbon assimilation efficiency (average values), nd = no data.

Scope for growth (Eii*) of M. tmssuIus (0.164 gdw, 45 mm SL). The energy of food particles was bawd on 1 mg dry weight of Thalassbska pseudonane = 12.56 J. IR,, = ingestion rate of particulab otgantc matter, E- = gross emtgy intake, Emt = energy expndlures. For metabotic expendlums (tespiration and ammonium excmtion), the following conversions wem usd. 1 ml Ot = 20.33 J (Widdows and Johnson, 1988) and 1 mg NHcN =24.87 J (Elliot and Davison, 1975), app. CA€ = apparent wston quality, true CAES = tnir carbon assimilation etfcbncy (average values). ') (algae : silt = x 10' œlb C' : mg L"), 7 means I ISE, n = 3, nd = no data.

The definition of terms useâ in the modal.

Model paremterization for the kinetic model of Cd accumulation by fitter-feeding mussels.

Oissolved cadmium (Cd*) in Howe Sound. Briti~h Columbia, Canada (Chrétien, 1997).

Retention effiiUency (h) of '%d by the Mue mussel Myfilus tmssulus.

Sensitivtty analysir for Cd bioaccumuktion model. The quantity S ir a meawre of paramter wnriavity.

LIST OF FIGURES

Page

Mean seston (SPM) concentration and maton quality (SQ) in Fraser River estuary duringl994-1995. SPM in mg dry weight L", m a n I ISE, n = 5. SPM data from Stecko and 6endelCYoung (1999) were converted into dry weight; SQ was cakulated baseci on the ratio of organic dry weight and total =ton dry weight.

Thesis outline.

Flow through system for the study of feeding rssponnm of musseb to seston matrices. a) source of mixed algae and siII b) fiîtered wawater, c) stinsr plate, d) pristaMc pump, e) mixing chambr, 1). experimntal tanks, 0). Plexiglas chamben wen used for radiotnœr expriments in Chapter III.

Parlicîe size distribution of algae and siît (man i ISE, n = 10).

Overview of the various ôehavionl and physiological measu fements.

Ckaranœ rate (man * 1 SE, n = 3) of musseb e x p o d to dinetent seston matrices. AlgaaSil = (10' cells L": mg L"). The silt concentration of 5, 20, 150 mg L" repmwnt 49, 482, and 1267 x 10' c e ~ ~ s L" , retspctiveiy.

Cîearance rate (man I 1 SE, n = 3) of mussels exposeâ to seston tmatmnts for 4 h period. Signifiant diffsnnces a n denoted as dincmmt lettem. a.b,c,d,e above ban (Tukeyy's multiple range test, p<0.05). The E l concentrations of 5, 20 and 50 mg L" npmsent 149,482 and 1267 x 10' particks L", nsptiveîy.

Relationship ktween: (a) seston concentration8 (SPN) and cksranœ rate (CR); and (b) seston organic content (PO&) and ckannce rate (CR).

Relationship betwwn maton quslity and wrting e f f i n c y (SEF), m a n I ISE, n = 3. SEF = .~.ow(sQ)' + 7.019 (SQ) - 67.648, ? = 0.60, n = 11, open circb = SEF data h m Bayne et al. (1993).

Rdationhip beheen (a) maton qurlity and pwudofaeœr production; (b) seston conc0nhtions and prsudof#ws produaion. PFc = 4.47 (SM&) - 15.77. ? = 0.63, n = 12, 95% CI, values = means I ISE, n = 3.

Ingestion rate of mu8mk undsr diffemnt wston matrices. Significant diffemnws a n denotd as diffemnt lettem. a.b,c.d,e and f (Tuby's multipb range test, pq0.05).

Relationships between particulate organic matter (PO&) and ingestion rate of orgsnic maltsr (IR-). Open circk = ingestion rate at maximum dearanœ rate at serton matrix (20 x 106 œlh L" : 20 mg L"). IRRnr= 3.495(i3MOc) + 18.01 9,95%CI, ? = 0.50, n = 11 (open circk not includd).

Relationship between (a) wston quality and apparent carbon assimilation efçcbncy, app-CA€%= 1.204 (SQ) - 1 1.397, ? = 0.64, n = 11; (b) mston quality and tnie cribon assimilation efficiency. 3 -data with high algal concentrations. 150 IC 10' œlls L".

Relationship between wston quality and urbon assimilation rate (AR). Open circles not includd in a regremion analyrir.

Flow-chart of two types of t h experimuntal protocol: a) lakkd- algal expriment. and b) Iakledaitt expriment.

'%d activity in seawater (dpm L") during the 4 h exmrimntal exposure. (a) Seawater from experiments in which only a b w was Iabeled. (b). Seawater from experimnts in which only d l was labeled. Values = (rneans î ISE, n=3), Ratios are amounts of algae (~10' cellr L") to silt (mg L"). The rit conwntrations of 5, 20 and 50 mg silt reprisent 149, 482 and 1267 x 106 paMes L=', respectively .

109 Cd activity in wston matrix (dpm L") during the 4 h experimental expowre. (a) Seston frorn experimnts in which only abas was lakled. (b). Seston from experimentr in which onty rilt was Iakkd. Values = (man I ISE. n=3). Ratios are amountr of abas (~10' celk L") to SM (mg L"). The silt conœntmtiont of 5,20 and 50 mg 1" npmmnt 149,482 and 1267 K 10' paiticks L", re8pctively.

Biodeporition rater lœcd (a). and 2 4 ' ~ m (b) in p8eudofaeœs (PF) and fa-r (FE) by rnu88els expomd to w8ton concentrations wiM iabekd-algal component.

Bideposition mtes "cd (a). and 2 4 1 ~ m (b) in pueudofaecer (PF) and h e m 8 (FE) by m u m k sxposed to seston concentrations w M Iabekd-siit component.

Cadmium auimilation enicirncy ('Oe~d-A€) from bkkd-dgn and Iabekd-silt udng a durCtnmr ntio mthod (DTR) and ingestion rate rneawmmt (IRM). = a lignifiant dmrent htween aie two tnethodr (Shident t-test. p c 0.05).

xiv

39

l W c d - ~ € bawd on duaMncer ratio mthod (DTR) and ingestion rate measummnt (IRM) at dinerent diet qualities (W). (man I ISE, n=3). a) kbkd-algal experimnts, b) Iabed-dît experiments

Apparent and true ' w ~ d - ~ ~ b a d on the ingestion rate rneawtemnt (IRM) in relation to d# quality (DQ). Values = (man i 1 SE, n = 3). app and ttue '%d-A€ fmm Iabeled algae (a), app and ~ W ~ ' ~ C ~ - A E from labekd sil (b).

Relationship beWssn trus cadmium assimilation eWicMncy (tnnt '"@cd-A€) and diet quai$ (DQW). Values are means I SE (t = 0.98; pe0.05 for ttue ' CdAEs for the Iakkdslgal exposun venus d# qualtty). Circk denotes whem mus8els displaying maximal ckannce rates (CR) in pmvious study (Chapter II).

109 Relationship ktween Cd concentration in muusel tissues ([Cd]u,,,,,) and tme 'Oscd ss8imilation elficiency (true ' 0 9 ~ d ~ )

Rektionship beWeen Cd assimilation efkiency ('%~-AE) and true carbon assimilation efficiency ( true C-A€). Circk denotes when mussels displaying maximal ckannce rates (CR) during pnvious study (Chapt. II)

Schematic dia~tam of the changes in feeding khavior of mussels in nsponse to a btoad range of seston matrices. SQ= wston quaMy, SEF= wrting efficiency, COS= hypothetical cost of rorting, stage A= food condition when [food]<< [siît] and no sorting, stage 8= food condition when [food] s [sil] and sorting, stage C= food condition when mostly abar and no soiang, L = low, H = High.

Schematic of the expeiimntal apparatus. A, B C = mpiromoter chamben, O = food rupplioi, E = stopper with a hob. allowing W i u m to ovediow and taking ammonium umpkr , F = oxygen ekctiodo, G = rtimr, H = oxygen mter, I = chafl recorder.

Caloric content of the diatom ThaIassiosim pseudonane. A, B. C = Abal culunr hawestd during exponentirl growth pham; 0, E = abal culhirem h a w s t d durhg air wnercsnt growth phase. (man8 I 1 SE, n = 3). Signihcrnt (pg0.05, onawry ANOVA and Tukey's mulapie range teit) dWemncs8 a n denotd by dWemnt Isttrrs above barn.

Reipintion nt08 of Mytiilus ~I~DSSUIUS in mponse to diffennt =ton matrice8 (man8 t 1 SE, n = 5). Signihcrnt (p4).05. Tuîcey'r mulipk range test) diffafonces am denoteâ by dilhnnt bttem abom ban; Seston quaYi (Sa) = 0% n p m n t 8 expenmnt without naton.

4.5 Relationship ôetween oxygen conwmption and clearance rate 114 (CR). CR values were taken from previous studios (Chapter II). ln contml treatmnts, the value of CR (i.e.. number of partickt ckamd p r liter pur hour per g dry weight) war k low the limit of detection of the instrmetnt. (man8 î 1 SE, n = 5).

4.6 Ammonium excretion of Mytilus t m u l u s in responw to different 11 5 seston matrices (means î 1 SE, n = 5). Signifiant (p< 0.05, Tukey's multiple range test) diffennces are denoteû by diffemnt ktttrs abave ban.

4.7 Rdationship betweem ammonium excretion rate (pg NHcN h" gdw' 117 1 ) and tnie carbon assimilation eflicbncy (twe GA€%). Values = mean î 1 SE, n = 5, r = 0.65.

4.8 Rdationship between ammonium excretion (J h" gdw") and 119 respiration rates (J h" gdw"), Values = m a n I 1 SE. n = 5, r = 0.94.

5.1 The study's sites (sampling and tnnspkntation) of the blue mussel 1 U Mytilus tmssulus. (. . . . .) = intetiial zones.

5.2 (a) Preôicted Cd concentration in rnuusl tissues (Cd-) for a one- 147 year pefiod; wuam = Cd- under non-wbctive feding khavior, dot = Cd- under wlective feeding ôehavior, bat = L M Cd concentration (Cd*) befom (x) and ofter (y) transplantation (man I SE). (b) Dietaiy Cd uptake (%) for the Mue mussel (Mytilus tssulus) during one-year period.

LIST OF APPENDICES

A The conversion of the silt component of seston from unit of mg sitt 161 L" into particlas L".

B Sumrnary of abbreviations and physiological parameten in Chapter 164 II.

C The correction of measuied seston ftom influence of seawater. 165

D Texture of (a) pwudofaecer and (b) faeces of the blue muswl. 167 (Myfilus trossulus). bar = 0.50 mm

E Flowchart of application of kinetic model in Myilius lmssuIus 168

F Sensitivity analysis of modrkd Cd concetntration in musml tissues 169 following I 20% variation of each input panmeter. SF = sekctive fwding. nonSF = non-sektive feeâing.

1 A. RATIOWACE FOR THESIS

Fibr-feeding bivalves bdonging to the genur Mytilus are widely distributed

throughout temperate waters (Gosling 1992) and fmquently used as model organisms

for phyriological and ewtoxicological studks. T h y are used as an indiator of

changes in aquatic metal bveb, because they a n abundant, wdentary, and

nsponrive to their environment (Gmn et ai. 1985). Dominent muswl population8 in

shallow estuarine systems are able to consume more than 50% of the annual primary

production (Gonitsen et al. 1994). Hence, they play an important role in

biogeochemical cycling of nutfients and a n major factors controlling the fate of mta l

contaminant8 in surrounding waters.

Given the mussel's ability to accurnukte trace mta l contaminants. much

msearch has as L endpoint, the devekpmnt of predictive models of metal

bioaccumulation (Thomann 1981, Thornann et al. 1995, Wang and Firher 1997). Such

mode18 sewe as a management tool for protecting ecurystenw. Thus, the value of a

prsdictive mode1 ir that it can k procidive, fomaming of the potential iisks of metal

bioaccumulation in filter-feeding orglnisma.

In the prst 20 yean, musseh have been the most wmmonly uwd indicalor

specius in many monitoring program8 (e.g. Musml Watch Program). Basic to the

irnpbmentatïon of the 'Miurl Watch' pmgnm (i.e., the use of the mussel to indicste

3 changes in environmental metal ievelr) have been the assumptions that: 1) mussels

feed on whatever suspend& particles am available in their wrrounding waters (non-

mîective feeding khavior), and 2) metal k v e i in mussel tissues npmwnt a 'tirne-

integrated' value of the biologically avrilable metal in thdr surrounding environment

(Lares, 1995). The latter assumption is genenlly met when the monitoring is done on

the same species of muswis f ho violation of thb assumption, however, can occur

when monitoring programs compare metal-tissue concentrations within large

geographic amas where the dinefences in metal concentrations occur, not only

kcauw of ditfersnt pollution sources. but also ôecause the mussel specim are

different. For exampk, Loôel et al. (1990) found that blue musmls Mytilus tmssulus

had highr metal content than did Mytilus dulis sampkd from the sam location.

Therefore to ovoid this factor, 1 is important to use bioindicaton from the samo species.

In contnst to mady acceptance of the second assumption, the fimt assumption.

i.e., muswb feeâ upon wh8t0vet wston is availabk in thdr surrounding~~ is highly

contmvenial among inveitebrate biologists. Many labontory studies (e.g. Shumway et

al. 1985, MacDonald and Wafd 1894, b f o s z and Hawkins 1997) have chabnged

this assumption and have demn8tratd that the musml is crpabk of a compensatory

feeding behavior. îf this wlective feeâing behavior occun under natunl conditions. the

pmwnt application of mussels a8 a bioindicator in ecotoxiwlogical studies may k

lacking in the aspect of wkctive feeding ôehsvioi of m u s h .

Filter-feeding bivaives obtain theit nutritional requirernents (energy and

nutrients) from wspended paiticulate matter (SPM). This potential food wum. termeci

seston, indudes microalgae/phytopknkton. organic debris and inorgsnic (iilt) pirticîes.

4

The wston varies in its quaMy, from high quality dominated by organic particks to low

quality dominated by inorganic paiticles (minerab). Similarly. seston concentration

(quantity) greatly changes from 3.5 to 30.0 mg dry weight L" during tidal cycles (Berg

and Newell, 1986). In the Ftawr River estuary, the seston conwntration ranges from

4.5 to 32.0 mg dry weight (dw) L" with seston qualtty ranging beWeein 3 and 23% (Fig.

1.1). Seston concentrations are low in fall and winter and reach p a k concentrations in

mtâaummer, whik wston quality is maximal in mtd-winter. Hence, fibr-feeding

bivalves are subject to a highly dynamic environment with respect to the levels of

suspended paiticles (seston).

Seston associated with metal contaminants rnight affect the fate of metak if

high arounts of seston act as an absorbent for metsls, reducing metal dissohMd phase

(ionic forrns) in the water. Thomas and BendeCYoung (1998) found that metal (Cd. Pb

and Cu) levels in tbsue of Mecoma be(thica were best related to the concentration of

metal rssociated with the inorganic component of sediment. A recent study also

suggested that the cadmium was mainly asrociateâ with inorganic (easiîy retducibk)

suspendd particles (Stecko and Bendell-Young 1999). Hence, the adsorption of Cd to

both organic and inorganic components of wston and the proportion each component

of seston in the ingested food play an important rok in detemining Cd accumulation by

fittet-feeâing mussels.

5

Fig. 1 A . Mean seston (SPM) concentration and seston qualw (SQ) in the Fraser River estuary dumg 1994-1 995. SPM in mg dry weight L*', (mean * 1 SE. n = 5). SPM data from Stecko and BendelCYoung (1999) wem conveited into dry weight; SQ was cakulateû baseâ on the ratio of organic dry weight and total seston dry weight.

40 1

+ S Q % +ï- SPM

O 1 I 1 I 1 I I I 1 1 I I I

A S O N D J F M A M J J A

Months

6 In most ecotoxicokgical studies using bivalves (e.g., Mussd Watch Program), 1

is implicitly assumed that the filter-feeding mussels feed on whatever suspetnded

particks are avilila b k in their enviionment (non-seledive feeâing khavior) . However,

many labontory studies (e-g. Bayne et al. 1993) suggert that Mytilus have the abillty to

sort their diet (wlective feeôing). Hence, there is a gap between the knowldge about

the musseh feeâing behavior and its application in the field of ecotoxidgy as a

bioindicator species. If mussels do wkct their dbt, then the diet that b actually

ingested and the subsequent absorption of carbon frorn the diet have the potential of

k i n g entirely different frorn that which would be expected bawd on in silu

measuremen ts of seston (an assumption of non-selective feeding behavior).

Therefore, understanding the feeôing ôehavior of mussels becoms an important factor

for detemining ' t im integnted values' of metal bioavailability. This is particularîy

important if the predictive model is rxpected to provide an rarly waming of the potential

risk of metat bioaccumulation in the ecosystem.

t .P. OWECTIVES OF THE STUOY

The goal of my remarch is to provide an undemtanding of how biotic factors,

espocially feeâing khavior, inlwnce contaminant availability and metal tiswo

concentration (Cd) in Cher-feeâing muswls. By examining the khavionl and

physiolqicrl mcponses of mussah (such as clearance rate, soiting efîîciency,

p80udofaeces production, ingestion rate, and assimilation effibncy) to environmntalîy

nkvant seston conditions. a genenl model can k develom for p M i n g Cd

7 bioaccumulation in Gîter-feeding mussols. Thersfore, the specifc objectives of the

pnsent study an:

to determine the feecîing khavior of the bhie mussel (Myfilus tmssulus) in response

to environmentally relevant seston matrices;

to determine how variations in the oqanic and inorganic components of seston

matrices influence the ability of the mussel to assimilate cadmium (Cd) :

to determine if there are metabolic costs associated with the feeding (sorting)

process; and

to develop a kinetic-bawd model of Cd accumulation in such a way that the feeding

rate of Me musml and the Cd assimilation effciency can be taken into account in

detenining tissue metal burden.

By meeting these objectives, not only a n basic questions related to the ftmding of the

mussel k i ng addnssed, but also the behrvioral responses of the mussel to a dynamic

food environment a n k ing incorponteâ into a pmôictive modd of metal accumulation.

1.3. OROANUATIûII OF THE THESI8

The thesis, as outlined in Fig. 1.2, ir diviûd into 6 chapten. The main four

chapten a n written a8 standslone sections, thenby facilitating publication as rcientific

Papen. The four chapten are linkeâ by a genenl introduction (Chaptei 1) and ovenll

conckrionr (Chapter VI). The dirsdvantages in choosing thir format is that it may Iead

to m m minor rsdundancies, but mm was taken to avoid these as much 8s possible.

Organuation of the thesis is as follows: Chapter I pmvides the rationale and

objectives of the thesis. Chapter II deals with the feeding response of the mussel and

its influenœ on carbon assimilation when mus8els am expowd to environmntally

relevant seston matrices. Chapter III studies the effect of a wkctive keding strategy

on Cd assimilation efficiency. Chapter IV examines the implications of the diet sorting

process on the metabolic costs of feeding, and Chapter V presents a kinetic-based

model of Cd bioaccumulation from the food source by murwls. The modd is vedfiied

using the data of Cd in mussel tissues from field expenrnents. Finally, Chapter VI

provides geneml conclusions and recommendations based on the thesis findings. At

the beginning of each chapter, the specific background or rationale and its objectives

are provided.

Fig . 1.2. Thesis outline

CHAPT. I Rationale for the thesis 1

1 Objectives 1

t Lab. Exp.

9 Field Study

t r f f \

CHAPT. II 1 CMPT. III CHAPT. IV Feeding respnse 'Oscd assimilation Respiration mbs

C J i J and NH4 excretion J 1 1

Cdmuic = Cdm + Cd- * Mussel's transplantation L J C

Conclusions and recommendations

1.4. REFERENCES

Berg JA. Newell RIE (1986) Temporal and spatial variations in the composition of seston available to suspension feeder Cmssostrea viginr'cs. Est Coast Shen Sci 23:375-386

Defossez JM. Hawkins AJS (1 997) Seîective feeding in shellfish: size-dependent rejection of large particie within pseudofieces from Mytilus edulis, Ruditapes phiiippinranrm and Tapes decussalus. Mar Bioi 1 29: 1 39-147.

Gerritwn J, Holland AF, lrvine DE (1994). Suspension-fding bivalves and the fate of primary production: an estuarine model applied to Chesapeake Bay. Estuaries 1 7:403-416

Gosling E (1992) Systernatics and geographic distribution of Mytilus. Pp: 1-17. in The mussel Mytilus: ecology, physiology. genetics and cutture (ed. E. Gosling). Elsevier Sci. Publisher, Amsterdam, 1 he Netherîands.

Green RH. Sigh SM, BaLy RC (1985) Bivalve mollusks as response systems for modeling spatial and temporal environmental patterns. Sci Total Environ 46: 147-1 70

Lares ML (1995) Mussels as indicaton of cadmium and Iead in the marine environment. Ph. O. thesis. Oept. of Oceanognphy , Univemm of British Columbia, Canada. 153 p.

Lobei PB, Belkhode SE, Jakson SE. Longerich HP (1990) Recemt taxonornic discovehs conceming the mussel MyWus: implication for biomonitoring. Arch Environ Contam Toxiwl 19:508-512.

MacDonald BA, Ward JE (1994) Variation in food quality and particb sekctivQ in the ma scallop Plampecten megellankus (Mollusca:Bivalvia). Mar €col Prog Ser 108:251-264.

Shumway SE. Cucci TL, Newd RC, Yenthsch CM. (1985) Partick sekdion, ingestion and absorption in f i l r - f d i n g bivalves. J Exp Mar Biol Ecol 91 :77-92

Stecko JRP, Bendell-Young LI (2000) Conttasting the geochemirtry of suspnded particulate matter and deposited secliment within the Fraser River estuary. App Geochem (in press).

Thomas C, BendeIl-Young LI (1998) Linkinp the wdimnt geochrmirtry of an interadal region to m ta l bioavailability in the deposit feeâer Mecorne beIthics,. Mar €col Prog Ser 1 73: 1 97 -2 1 3.

I I

Thornrnan RV (1981) Equilibrium m o b l of fate of microcontaminants in diverse aguatic food chains. Can J Firh Aquat Sci 38280-296

Thornman RV, Mahony JD, Mudkr R (1995) Steadyatate model of biota sediment accumulation factor for metals in two marine bivalves. Env Toxicot Chem l4:t8t 1-1812.

Wang WX, Fisher NS (1 997) Modeling metal bioavailabil~ for manne mussels. Rev Environ Contam Toxicol 151 :39-65.

CEEDIWO RESPONSE AND CARBON ASSIMILATION BY THE BtUE MüSSEL YyWw frwauîua EXPOSE0 TO ENWROWWTALLY RELEVANT SESTON

MATRICES'

O) This chapter was expandecl from a pmviou8ly publiaheâ artide, F d i n g nlponse and crrbon assimilation by the blue mussel MyWus tl~ssuIus exposd to envHontwntally nbvant seston matrices, (2. AMn and Leah 1. hndelCYoung), in Mar. Ecol. Pmg. Ser. 160:241-253 (1997).

The objective of this study was to detemine the f d i n g response of the Mue

muswl Myfilus tmssuIus to environmntally relevant seston matricer. Twelve sestoton

matrices varying in quality and quantity were pmpared by mixing t h m cell densiths of

the marine diatom 7"haIassiosim pseudonena (5, 20 and 150 x 10' œlls L") and 4 silt

concenlralions (0, 5, 20 and 50 mg L") to represent an increasing seston quantity of

1.4 to 56.6 mg L*' and an increasing serton qwlity (SU) of 10 to 70% organic matter.

Ckannm nt08 (CRS). p s e u d o f ~ 8 (PF) production. wrting effikncy (SEF),

ingestion rates (RI) of particulate organic matfer (POM) and particukte inorganic

matter (PM) as well as apparent and tnie caibon auimilation efficiencier (CAES)

wen detemined for the various exporure regims.

Undsr conditions of O and 5 mg L" dl, CR$ dmased by 3-foM (14.4 to 4.9 L

h" g-' dry wt (gdw*')) und 6-fou (18.0 to 3.3 L h" g d d ) nspectively, wlh increasing

wston quanûty and quality. Under conditions of high silt loads (20 and 50 mg L"), CRS

were independent of increadng SU with maximum CRS (21.8 î 2.2 L h" gdw")

obmrveâ at the 20 x 10' colis L" and 20 mg L" dît expoaures. PF production was

dependent on serton quantity (? = 0.63, p < 0.05) with mussek prsfrnntially rejecting

the inorganic component of the serton. This SEF was maximal at a SU of 40% organic

matter. As a conMquence of th@ feeding behavior, detemined POM IR8 under high

algadhigh alt (high qwntity/quality) exporun mimes wre companbk to thoae of the

high quality Üust algae) expowmr (POM IRs of 48.0 as compamâ to 38.1 and 91.3 mg

h" gdw" for high quality and high quantitylquality, mrpctiveîy). In contmst, mu8sels

exporsd to low qualitylquantii m ton ingertd both ~8toon components (SEF s 24%),

Le. aie mumd war non-ubctive, pouiMy i ngdng both organic and inorganic wston

companents to m e t nut-nt nquimmrnts.

Apparent GA€% comîatd wiai SQ (*=0.64, p e0.05). However, mibon

amimilrtbn rxpcsswd as true CAE% was independent of SU a8 war carbon

188imMtkn rate. WnC8. nirough a dynamic int.rpky btwaen CR8 and carbon

ausimîktion effhncy, the &lue musml war able to maintain a constant rate of cribon

amimiktion, mgardbrr of the quality and quanuty of m ton to whidi it was expowâ.

Within estuanne envitonments. one of the main filter-feding bivalves is a

species compkx of blue mussel (Mytius edulis L.. M. geflopmvincieis Lamarck. and M.

t m u l u s Gould). As a resuît of its global distribution and its ability to concenttate

metal8 from its environment, the blu8 musml ha8 been used world wide for pollution

monitoring prognrns. Gken its importanw. much study has focuwd on how biological

factors such as sue. sex and growth rates influence tissue metal burdens (Cossa et al.

1979. Poulsen et al. 1982, Lobd et al. 1991 ). Mon recent studies have baen dincted

at detemining the underîytng physiological mechanisms that influence rnetal

accumulatton by filter-feeding organisms. For exampie, Wang et al. (1 995) telated the

amount of metal accumulation by bivalves to gut passage time and digestive

partitioning. Decho m d Lwma (1996) have noted that bivalves are capable of

moôifying the digestive prowssing of food to teâuw exposun to biological availabk

chtomium concentrations. Wang and Fisher (1998) suggesteâ that for son# mettais,

assimilation is nlated to the d e g m of carbon assimilation by the mursel, i.e., the metal

assimilation and hence accumulation by the mussel folkws a dietlenergy pathway. No

studies, however, have assa8wd the pomibb rok of feeding behavior in detemining

the amaunt of mtal to which the muswl i8 exposed via its diet and henm the amount

of meta1 that is potentially availobk for uptake and accumulation by fiter-feeding

musmls.

15

Cumntly, there is much controversy in the litemturs in regards to whether filter

feeâen have a sekcave feeding behavior (8.g. Bayne et al. 1993) or whether hiter-

feeding is a highly automated procws detemined by the natun of the bivalve filter

pump (Le.. non-sekctive feeding behavior (e.g. Jorgenrrsn 1990). Bayne et al. (1 993)

found that mussels fed at high concentrations of seston of relatively low organic

content (40% POM) incteasad their filtration rate, repcted a higher proportion of filtered

matehl as pseudofaeces, and increased the efficiency with which filterd matter of

higher organic content was wlecteû for ingestion. These changes in nsponso to a

changing food environmnt by the musml resulted in higher rates of carbon

assimilation than would have h n pdicted baseâ on the assumptions of a non-

wmpensating feeding-behavior. ln contrast, Clausen and Riisghd (1996), based on

their studks, concludeâ that there was no eviâenw for physiological ngulation of the

filtration to meet nutritional needs and that. as Jorgensen (1990) suggesteâ. food

uptake in natun was chamctrrizeâ by the full exploitation of the bivalve fitter pump.

îf mussels are highly wkctive as suggested by Bayne et al. (1993), then the

diet of the mussel (Le., that wmponent of the seston that the animal ir adually

ingesting) and the rubwquent absorption of carbon from the diet have the potential of

k i ng compktely diffetent fmm that which would be expecteû b s s d on in Mu

meawremnts of wston within the environmnt and what the mussels a n uipposeâly

feeding on. f hir ha8 important implications for preûictive rnodels of m ta l accumulation

by rsston-ingesting orgenisms that Unk mta l assimilation to the organic content of the

diet (e.0.. Wang and Fisher 1996). 1 mu88el8 an capsbk of a compensatory feeâing

khwior, a8 propowd by Bayne et al. (1993). kading to rektive enhancement of net

energy and nutrient acquiriîion over that which would ba predicted basd on

ofganic/inorganic compo8Wn of th8 saton matrix. then pmdktive modelr of mta l

16

contaminant u ptake, b a d on a dhtlenergy pathway , may underestimate amounts of

contaminants to which the mussels am actually exposed.

Henœ, the objective of my rtudy was to detemine aie feeding rwponse of the

mussel to environmentally relevant wston matrices. which varied in ôoth quality and

quantity. Specifnolly I am intensted, fimt, in detemining the dogme to which the

mussel could a b r its food environmnt (i.e.. war the dist that the mussel ingerted the

same as the seston matrix 10 whM it had ôeen expo8ed) given a wide nnge of

envimnmntally nkvant wrton matrices. and second, in detemining the

comrponding carbon amimiYion efficienci88 for the diet n wbctd by the musnls.

To m t these obpctives. change8 in the folbwing behavionl and phyriobgical

attributes in the mussel (My?iius tl~ssulus) expo8d to the various seston matrices were

asseswd: (1) ckrnnce rate, (2) pseudofaecer production, (3) sorthg efficiency, (4)

particulab organic and inorganic mitter ingestion rates, (5) apparent and true carbon

assimilation efiiencies, and (6) carbon assimilation rates. Seston qualtty and quantw

w e r choun to npremnt seauonal and tidal patterns in maton quality and quantity

found under natuml wndlions (i.e., within the estuarine environmnt) (brg and

Newell, 1986, Smaal et a!. 1986, Fegky et al. 1992, Galois et al. 1996, Stscko 1997).

2.2. MATERULS AND METHODS

2.2.1. Exp.iknrntil animal.

Muswls Mytilus tmssulus Gould. 1850 wers collected from an intertidal ana on

the west coast of Bntish Columbia (Horseshoe Bay), Canada. Colkcted mussels wen

placed in a cookr. and tnnsported to the labontory at the Dept. of Biological Sciences.

Simon Fraser University. Muswls (44.9 k 2.07 mm in shell kngth. SL) were acclimateâ

to sxp8"mental conditions (temperature: 13 î 1 OC, salinity: 28%) for two weeks prior to

use in each experimnt. Throughout the acclimation priod, the muss8ls were fed the

diatom Thelassiosim pseudonena daily, and the wawatet was changed on a regular

basis. Prior to the feeding enpefiments. mussels wen sepanted from their byssal

attachments to one another. brushd ckan, and kept for approximately 15 min in air.

This procedure enrured that only live mussels wenr used in the experiments and that

thoso mussels that were not viable diâ not respond to k ing submrged in wawater

following the 15 min exposum to air.

2.2.2. Expoiknrntil rppiiatur

Feeding khaviot e~periments wem conducnd under MW-through conditions in

500 mL Pkxiglas tanks (Fig. 2.1). A source of fittemd wawater was provided by

fibring supplied wawater through 5.0 and 1 .O pm cacrr(ge fifien (Labcor Inc) into an

20 L tank (Fig. 2.1a). The sestoton matrices were prepard by mixing known volumes of

aîgal culture and silt suspension into a polyethykne tank (Fig. 2.1 .b) and subsquent

dilution of the two wmponent8 of w8ton with the pnpared fibmâ wawater in a mixing

chamkr (fig. 2.1 .c).

18 Fig. 2.1. Flow-thmugh system for the study of îeding response of musseb to seston

matrices. a) source of mixd algae and si&, b) fibmd wawater, c) stimi plate. d) penstattk pump, e) mixing chamkr, f) expflmctntal tanks, g) Plexiglas charnôers wem only u s d for radiotracer expeflrnents (Chapter III).

2.2.5. Serton (rurpandrd prrtkulrdr miümr)

A nview of the literatun indicrted a wide nnge in in situ near-shon and

estuarine wston quantity (0.3 - 200 mg L") and quaMy (3 - 48% organic matter) (Table

2.1). To represent this nnge, seston matrices were manipulated so that puality rangd

fmm 10 to 50% organic matter and quantity from 1.4 to 56.6 mg L". The highest

seston quality was 71 % organic matter for the abat alone treatmnt. The sestoton matrix

was made by mixing the diatom T. peudonana with kaolinlc minenl (AS? 400-P) to

npremnt the organic and inoiganic wmponent of the seston. respectively. Twelve

seston matrims varying in puality and quantity wen pmpared by mixing thme diatom

concentrations (5, 20 and 150 x 10' d l 8 L") and four riît concentrations (0, 5. 20. and

50 mg L"). The i i t concentrations repremnting of 0, 149, 482 and 1267 x loe particies

L". respectively (Appendix A). T. pseudonane (Stmin No. B 709) was obtained from

the North East Pacific Culture Colkction (NEPCC). University of British Columbia.

Canada. Non-axenic culuns of the abae wen grown in aenteâ 4 L flasks h natunl

seawater enriched with Hanison medium (Hamron et al. 1980). at 16% and under 10 h

dark :14 h light illumination (188 pE m" s"). Twelve liten of the algae were pnpared in

each experiment. Tho kaolinitic minenl (ASP400-P) was obtatneâ from the Engelhard

Corp. The partick sue distribution of riît was detennineâ fmm 10 repliaites with a

Coulter ekctmnic particb TAI11 frttad with a 100 pm aperture tube. The partick sue

nnged fmm 2.0 pm to 25.5 pm (Fig. 2.2). Pfior to each experimnt. stock rlurry was

prepamd by suspending a pmdetermineâ weight of siît in one Iiter of fiîtered wawater

(0.45 pm). The sluny war dispefwd with an eîectric stimr and an ullrnonic bath

before mixing with the rbw in the experimntal tank. The concentrated seston wss

delivemd into a 1 .S 1 mixing chamkr by peristalüc pump. From the mixing chamber,

the wrpnsion war deliveml at a fiow rate of 110 mL min" to four 500 mL Pkxigîasm

Tabk 2.1. Seston concentration (mg L") and seston quality ($6 POM) of coratal ecosysteins. Monthty masurement of sedon for 12 - 18 months, 3 Sarnpling at 1 - 2 cm abow the wbrtnte over tidal cycles. 7 Sampling at 0.5 m above the substrat8 surface over tidal cycks.

Location

Narr mhon : Logy Bay. N.F. (Canada)

Tromse (Nomay) Great Sound, N.J. (USA)' Falm Bay (South Africa) Yaldad Bay (South of Chile)

EltIJWy: Lynher Estuary (England) Fort Boyard (France) Wadden Sea (Nethrlands) Fnmr River Estuary (Canada) The upper Chesapeake Bay (USA) Chesapeake Bay (USA)"

POM %

Navsrro & Thompson (1 995) Vahl (1 980) FegS et ai. (1992) GrMahs (1980)

Navam et al. (1 993)

Wiôdows et al. (1979) Galois et al. (1 996) Srnaal et al. (1986)

S teck0 (1 997) Berg & NeweM (1 986)

22

expedmsnntal tanks of which t h m containd musseh. The fourth remained empty and

was useci as a coritrol.

2.2.4. Santon C:N ntk

Seciton organic carbon and nitmgen content for ulculation of C:N seston ratios

ww detemined in triplkate udng a Cark Erba Ekmntal analyser (Model 1106) with

acetanilide (CaH$(HCOCk&) as a standard.

2.2.8. Phyrbîogkrl mm$unnmb

An ovenrh of the relatiinships of the vrriour experimntally detemined

behavionl and phpidogical panmetem is pmented in Fig. 2.3. A wmmary of

abbmviations useâ in quanbfying the ôehavioral and physioialogical panmeten is

prewnteâ in Appendix B. From Odokr 1995 to Febniary 1996, tweive mston

matricer wem nndomly sssignd and pspaM. Seton composition and physiological

parameters wen determind in triplkate as follow8:

S d w mWem: One I b r of each serton matnx war filtend through pre-

weigheâ 0.45 prn GFlF fiitem (Whatman) and rMwâ with seawater-kotonic ammonium

formate (0.5 M). To determine mûon quantity, filteml sampk8 wem d M to a

constant weight (60°C for gB h) and final weight mrded. Serton quantw was

expressed as the concentration of 8uspnded paiaculate matter (SPM: mg L") and was

bawd on total dry WOQM pet Mer of ruupmded matenal m v e m d h m the

experimntal control tank. To detemine r r t on quality, the dnrd rrmpbr wem fuithor

ashd in a mufh fumace (450 "C for 6 h) befon final wdghing. The seston inorganic

component (PM: mg L") was âetemineâ a8 the mnuinkig wdght of ths a- rmpb .

ri-

24

The particulate organic seston (POM: mg L") was cakulated by subtracting PIM from

SPM.

An important correction that n d e d to be made on measursd concentrations of

seston organic and inorganic concentrations was the inadverlent contribution of

seawater to the two seston component. After passtng through the 5.0 and 1.0 pm

fiîters the particle concentrations of the fiitered seawater stiH averaged 2.56 mg Li.

This particulate matter consisted of 1 .O3 mg PIM C' and 1.53 mg POM L". Vahl (1972)

noted that ctenidia of mussels are abk to retain a particle sue as small as 2.00 Pm. In

general, smaller particks (les8 than 1 .O0 pm in fittered seawater) are not retained.

Thur, although the f i b r d wawater had an average of 2.56 mg L" seston, it would not

have been biologically available to muswls. However, the seawater inorganic and

oqanic components wouM have influenced the deterrnined inorganic and organic

concentration in the seston matrices. Hence, seston composition was corrected for

influence of seawater as follows:

SPM uncomcteâ (uc) = PM + POM (2.1)

SPMcomdd (c) = (PIM - Seawater,) + (POM - Seawater,) (2.2)

= PI& + PO& (2.3)

The detaiW procedure is presenteâ in Appendix C.

CIwmnce rate ~ u I ~ ~ : Cîeannœ rate (CR) ir a masure of the

volum of water from which particles have been removed by the musels per un# t i m (

L h-' g" dry weight (gdw")). The fbw rate through experimntal tanb war maintaineâ

at 110 mL min-'. An individus1 mursel was alloweâ to acclimate to expenmental

conditions for 30 min befom the ckaranw rate was marumi. During the course of

2 5

the expriment (4 h exposure to a mson maWx), the suspended prrticîes were

cokcted from the control and treatment tanks every 15 min. The particles were

counted with a Coulter Counter TA II with a 100 pm aperture tuk . The ckannce rates

were then calculatecl as follows:

where FR was flow rate (mL min"), Ct and Cc wen the particie concentrations (number

of particks L-') in treatment and control tanks, respectively. Clearance rates were

generally independent of tirne, i.e.. the rates did not increase or decreasa over the four

hour rxposun petiod (Fig. 2.4).

Dl I rminath of p..udof.u.r producfion: Pmudofaeces (PF) production

(mg h-' gdw") is any particle that is cieared from suspemsion but rejected by the gills

and palps (Widdows et al. 1979). Pseudofaeces were nadily distinguished from

faeces and wem of a light flum texture as compared to faeces. which were 'packagedg

and msembbd a long flot string (Appendix D). All pseudofaeœs generated by the

mussl under various wston matrices wete colkctsd from expwimental tanks by

capillary pipette. The rate of pmudofaeces production for a 4 h exposun period,

conected for the contribution from rreawater, was determined as :

Fig. 2.4. Clearance rate of rnusseb (CR I 1 SE. n = 3) exposed to differsnt seston matrices. Algae:silt = (r l0%ells L": mg L-'1. The dlt concentrations of 5, 20. 150 mg L" repmssnt 49,482. 1267 x 10 cells L", respediveîy.

Faoces ploducflon and carbon conhnt Total faeces generated throug hout

the 4 h exposure period wem colkcted from the exprimental tanks by capillary piwtte.

Recovereâ faeces were drisd (in a 60°C oven for 96 h). weigheâ, and total mass of

faecem generated pur gram dry weight of mussel for the 4 h exposure period obtaineâ.

Carbon content of the faeces was detennined using a Cark Elba elemental analyzer

(Modd 11û6) with acetanilide (CsHSNHCOCH,) as a standard.

Esümcrfion of m ü n g rmcImcy: Sorting efficiency (SEF) is defineci as the

retention. discrimination and ingestion of one partick type in prefetenw to another

(MacDonald and Wafd 1994). An estimate of SEF was based on the ratio of seston

quality and pseudofaeœs qualtty calculateû as:

SEF = (1 - (PQISQ)) x 100% (2.8)

when PQ is the quality of the pwudofaecss (% organic matter). and SQ is the qualtty

of the wston (% organic matter). Values of SEF n n g d from 0% (no seleetion; organic

content of pseudofamr and seston am equal) to 100% (full seledion; no organic

componrnt nmains in the pseudofaeces) .

m~ ca~cuktlona: ~ngetion rate (IR, mg h" g ~ ' ) was calculated

as a product of t h deannce rate and the wston qusntity, minus pwudofaeces

production.

IR, = SPW n CR

IR, is ingestion rate winiout sorting process and IR, is ingestion rate after sorting

process. IR, was calculated for both organic and inorganic SPM, components (i.e.,

where IR, = IR- + IRRw* and PFc = PF- + PF-) and represents the actual diet of

the mussel, i.e., that component of the seston that is actually ingested after takng mto

account paiücle sekction and subsequent rejedion via PFc production.

S e n qw/w Seston quality was expressed as the ratio of the concentration

of otganic matter in SPM divided by total SPM concentration (SQ = (POMJSPY) x

100) (after Bayne et al. 1993) whem both PO& and SPM, are in u n h of mg L-'.

Edmdlons of urlniikblon Mckncy. Carbon assimilation efficbncy (C-A€

%) is an estimate of the fraction of ingssted maton that is incorponteâ by the musml.

Both apparent (app GA€%) and twe (tnie CA€%) assimilation etlicienciss were

calculated. Appannt CAE%s wete b o d on the carbon content of faeces collected

from treatment tanks and the carbon content of seston rnatrix collecteâ from the conttol

tank.

Trua C-A€%$ wem bamd on estimates of the carbon content of the diet which the

muswl b actuully ingesting (Le., accounting for wkction of organic versus inotganic

partides) and the arbon content of faeœs and wem detennined as:

Carbon in the die~t (mg Cac gdw-') was estimated by multiplying amounts of PO&

ingested by the mussel over a 4 h period by carbon content of algae (measured as

12% by weight using Callo Erba ekmntal anolyzer (Mode1 1106) with ratanilide

(C&NHCOCHs) as a standard). Carbon in faeces was detemind as amount of

faeces produceâ by a muscrl for a 4 h exporun period muîtiplied by the carôon

content of the producecl faeces (mg Cr, gdw").

Statistical analyrir for a l data was prfomecl with Minitab 10 software.

Baitktt's test was applisd for testing data homogerieity and whem necessary,

transfomations wen applied to meet the asumption of the ANOVA. Two- way

ANOVAs wem applied to determine: 1) diffsnnws among the various seston quality

and quantity matrices, and 2) diffemnces in the khavioral and physiological responser

of the mussels to the vanous matricer. When ANOVA showeâ signifiant effects,

Tukey Honestty Signifcant Diffennt (HSD) tests wem applkl to detemine whem the

signifiant diffemnces o c c u ~ . Regression analysis was opplid to determine the

nlationship btween; (1) CR, and seston concentrations. (2) CR. and PO& (3) SEF

and SQ, (4) PF and SQ. (5) app CA€% and SQl (6) tnie CAE% and SQ, and (7)

carbon assimilation rate (AR) and SQ. Significance of al1 tests was accepteci at p s

0.05.

2.3. RESULTS

2.3.1. Qurnüty and qurlity of wrton mitrlc88

The m a n quantity of seston rangd from 1.4 î 0.10 mg L" to 56.6 î 3.40 mg

L" of which the rnean values for organic content wem 1 .O î 0.M mg L-' and 13.6 î

0.00 mg L". respectiveiy (Table 2.2). The resuits of the ANOVA indicated significant

difierences in seston qualities (p < 0.05) among seston treatments. which contained

both algae and sil. Seston quality nngeâ from 10% (20 x 10' celb C' and 50 mg L")

to 71% (just aigae aalon). Seston C:N ratios remained nlatively constant across al1

treatments (Tabk 2.2).

2.3.2. Emct of -8ton mrtricm on ckrrrnce ratma

Results of the average of CRS after 4 h exposure to the various seston matrices

are presented in Tabk 2.3. Changes in algal concentration significantty affect mussel

CRS (two-way ANOVA. p < 0.05). Under no and intemediate sitt exposure (O and 5 mg

L" siit), CRS decteaseâ or serton quality and quantm increawd, with as much as a

60% retâuction between the low and high algae exposure under the two lowest si&

mgims (Fig. 2.4). Surpriringly, maximum CRS were obtained undesr high silt (20 mg

L" ) and intemediate elgae (20 x 10' cals L") exporuns (Fig. 2.5). Under high aîgal

concentrations (1 50 x 10' œlk L"), CRS wrm not signifîcantly diflorent from each other

regardlesr of the incmasing sitt lord. When CRS wem regmsseû against increasing

seston quantw (SPMc, mg L") (Fig. 2.6s) or organic content in serton (PO&) (Fig.

2.6b), as a conwquence of the variabk nsponse of CRS to the various food matrims,

either no nlationship (CR, us. SPN) or s weak nlationrhip (CR, vs. PO& r = 0.54, p

> 0.05) was obbetveà.

Table 2.2. Charaderistics of the seston (SPM) matrices (man î 1 SE, n = 3). SPM, = s u t p n d d particulcite matter, PO* = particulate organic matter, SQ = w8ton quality, C:N ratio = carbonlnlrogen ratio in dry weight. Tukey's multiple range test was done on SQ data and the significsnt differences are denoted as dinerent letten. '- " . The silt conce!ntntions of 5, 20 and 50 mg L" repmsent 149,482 and 1267 x 1 O' parücîes LL' respectivety.

Tabk 2.3. Bahavioral and phydoiological data on mussels. CR = clearance rate, PFc = pseudofaeces production. IR, = Ingestion rate, apparent and true CA€% = assimilation efficiency of carbon. Vdues are mans î 1 SE, n = S. na = no data seston (conccntratbns were bebw critical value to produce pseudofaeœs). Algae : silt = (x 10' cefls C' : mg L'); the siit concentrations of 5.20 and 150 mg L-' represent 149.482. 1267 x 10' partides L". respectively.

Seston CR pFc IFGtotal I ~ o B ~ IR- App.C-A€ TM C-A€ Alpae Silt (L h-' g ~ ' ) (mg h" gdwl) (mg K' gcW1) (mp h- 'qd~ ' ) (mg K'Q~w*') % %

Fig. 2.5. Clearonw rate (man I 1 SE, n = 3) of mussels expoMd to seston tmatments for 4 h pwiod. Signifint dinefences am denoted as diffemnt Ietten, a, b, c, d, and e above bars (Tukey's mulple range test, p c 0.05). The 8iît concentrations of 5. 20 and 50 mg L" repmsent 149, 482 and 1267 x 10' particks L".

Algae: 5 20 150 5 20 150 5 20 150 5 20 150(xl0' œlh L") Silt : O 5 20 50 (mg L")

Fig 2.6. Relationship betweenr(a) seston concentration8 (SPN) and ckaranœ rate (CR). and (b) maton organic content (PO&) and ckatance rate (CR).

SPM, (mg I*')

2.3.3. EWct of raton miblcor on aoiaig omtioncy

Mean SEF (i.e.. the abilw to select otganic particks over inorganic particles) of

musæls ranged from -18 and O%, when e x p o d to low guallty (SQ 5 18%) and high

quality (SQ u 70%) of mston respctively. to a maximum of 87%. when exposed to

intermediate seston quality (25 - 60%). The obsewed relationship (Fie. 2.7) suggests

that a SEF maximum exi8ts that is dependent on the quaMy of the wston to which the

mussel is exposeâ, with a maximum SEF occumng at a SQ of 40%. For comparative

purpows, data of Bayne et al. (1993) have been includeâ on Fig. 2.7.

2.3.4. EMct of -$ton mitikar on pwudotrror production

Mussels always produwd pseudofaeces when expomd to wston tmatmnts.

with the exception of one seston matrix (1.4 mg abae L") that was k l o w the thnshold

for pseudofaess production (Table 2.3). PF production was independent of seston

quality (Fig. 2.8a), but increaseâ as seston quantity incmaseâ (PFc = 4.47 (SPMc) - 15.77. ? = 0.63. p < 0.05) (Fig. 2.8b). The highest rate of PF was 339.8 mg h" gdw"

which was associated with highest seston quant i rxposun (56.6 mg L").

2.3.5. EWct of am8ton mtiicr on ?O& and PWc inglrtion nb.

IR8 a n the product of CRS and SPM, comcted for PF production and are a true

masure of the actual diet to which the mussel ir exposed. IRs wen calcukted for

both organic (POMc) and inorganic (PIN) seston componrnts. Total IR (POM. + PIN)

as well as POMe and PlMe IR$ are pmsenteâ in Fig. 2.9. Them WIS a grneml increase

Fig. 2.7. Relationship between mston puality (SQ) and sorting efficiency (SEF), m a n î 1 SE, n = 3. SEF = -0.086 (sQ)? + 7.019(SQ) - 67.648, = 60, n = Il; open circkt = SEF data fmm Bayne et al. (1993).

Fig. 2.8. Relatbnship ôetween (a) seston quality and pseudofaeces production, (b) mston concentntions and pseudofaeces production. PF, = 4.47(SP&) )- 15.77, ? = 0.63, n = 12,95% CI liner. mean8 1 1 SE, n = 3.

O 10 20 30 40 50 60 70

SPMc (mg L'')

Fig . 2.9. Ingestion rate of mussels under diffemnt seston matfices. Significant differences ers denoted as diffennt Men, a, b, c, d. e and f (lukey's multiple range test. p < 0.05).

Algae:S 20 150 5 20 150 5 20 150 5 20 Silt : O 5 20 50

in total IRs with increasing wston quantity, with maximum rates occurring at

intemdiate algal concentrations (20 x10' mils L") and high silt lords (20 and 50 mg

L-' SM). At these maxima, both PlMc and PO& wston components were k ing

selected by the mussel for ingestion. Of most importance was the observation that at

the highest silt and rlgal exposurer. mussels wem capabk of selecting for only the

organic component of the seston, i.e., the diet of the mussel was comprised only of

PO& nther than PO& and PI&, as was notd for the lower qurkty and quantity

seston exposures. Hence, mussets showed a highly wkctive feeding behavior by

ingesting only the organic componrnt of wston and excluding almo8t a l inorganic

particles (siit) when expowd to a high quantity of seston, regardkss of their quaMy

(e.g. , under algae particles of 150 x 10' calla L"). However, under kw (5 x 10' cells L-')

and intemediate (20 x 106 tells L") algal expowrw. increasing rilt loadr resulteâ in

the ingestion of both PO& and Pl& by the mussel. There was a weak mlationship

between IR- and Me organic content of the seston (PO&. mg L") (8 = 50, p = 0.05;

Fig. 2.10). However. this mlationship wa8 loat when IR- for 20 x 10' celh L" and 20

mg L" siit treatments was induded, i.e., the maximum clearance rates obwrved under

this exposum regime msulted in the gmatest rater of organic motter ingestion. even

though the wston was of low qualw.

2.3.6. EIhct of raton quiMy on carbon r u i m h t k o afkkncy (MC)

App G A € % war dependent on seston quality (Fig. 2.1 1 a. P = 0.64). However,

when the true CA€% was mgresseâ agrinst seston quality, which accountr for the

soiting process, the true C-AE% war indepident of seston quality (Fig. 2.llb).

Selon matricer with the highest aigre conwntrationr (denotecl as open squares)

genemlly displayed the lowest true CA€% and. when vieweâ independently from al1

Fig. 2.10. Relationship ôetween partitulate organic rnafier (POMc) and ingestion rate of organic matter (IR-). Open circk = ingeution rate at a maximum clearance rate at seston rnatrix (20 x 10' celb L": 20 mg L"). IR- = 3.495 (POMc) + 18.019.9596 CI lines, P = 0.50, n = 11 (open circle not includad)

O 2 4 6 8 10 12 14 16

POM, (mg L")

Fig. 2.11. Relationship between: (a) wston quality and apparent carbon assimilation effichncy, app-CA€%= 1.204 (SQ) - 11.397, P = 0.64, n = 11; (b) ssston quality and tnie ciarbon assimilation eficiency; 2 - data with high algal concentrations, i.8. 150 x 10' tells L".

O 10 20 30 40 50 60 70 80

Seston qualtty (%)

other data, for wston quality > 40%, CA€% was positively conslatd to =$ton qualw.

When expnsseâ a8 the carbon arsimilation n te (Fig. 2.12) (a mearum of the amount

of carbon absohed frorn the diet by the mussel, giwn the amount of organic matter

ingestd over the 4 h exposure priod), the rate was independent of seston quality.

Greatest rates of carbon assimilation comsponded to treatments in which maximum

clearance rates were observed and in which seston matrices were comprised of the

highest algae and silt loadr.

Fig. 2.12. Relationship between seston quality and carbon assimilation rate (AR). Open circb not hcluded in a mgnssionanalysis.

O 10 20 30 40 50 60 70 80

Seton quality (%)

Numerous studies have a d d n s d the feding behavior of fitter femding

organisms undw changing seston conditions (Winter 1973. Bayne et al. 1989, Bayne et

al. 1993. Clausen and RiisgPrd 1996. Navano et al. 1996). From my experiments.

however, I am attempting to elucidate the pomibb mk miedive feeding may have in

detemining what part of the total seston the mussel is ingesting and the conorponding

carbon assimilation effickncy from Me diut as sekcted by the mussel. Assuming that

for SOM mtals (e.g.. cadmium; Wang and Fisher 1996) assimilation follows a

dieuenergy pathway, ultimateîy this information can k used to improve existing modds

which predict bioaccumulation of metal by filter-feeding ofganisms.

2.4.1.B.hrvkirl mponw of (hm MW mwrrl to r chrn~ing bad envlionrmnt

Two paradigms with respect to the feeding behavior of the mussel cumntly

exist within the litenhrre: 1) that they have the abilRy to mspond to changes in their

food environment, und 2) mussels am m m physiological slaves to their environment.

If the f o m r applies. then it is feasibbk that, as shown by Bayne et al. (1993). the

composition of seston in the environmnt and what the mussel i8 actually ingesting

(Le., diet) am not the mm. If however, the latter paradigm ir trw, as suggested by

Jergenwn (1990) and Ckuwn and Riisghd (1996). Men the murml is not mkctive

and h m evolveâ to filter at a maximum rate, regadeu of chrnging enviionmental

conditions.

46

An important aspect of my study was that I exporad mussels to a wide fange of

seston conditions that have been measured within the natural environmnt. My

expedmntal lreatments included conditions mpresentative of the ana from which the

mussels wete sampW, as well as extnmes that have k e n noted within other systems

(Tabîe 2.1). This ronge allows my data to be compated to ptevious studks on the

behaviotal msponso of the blue mussed to a changing food environment (e-g. Bayne et

al. 1989. 1993, Clausen and Riisglrd 1996). My findings suggested that both

paradigm in regards to feeôing behrvior of the musels are tme.

As in other studies (e.g., Winter 1973, Bayne et al. 1989. Clrusen and RiiqArd

1996). I found that, under low si l mgimes (O 8ftd 5 mg L"), CRS decmaued as abal

concentrations incnawd. However Clausen and Riisghrd (1996) interpreted this nsult

as k i n g a consequence of the mussel shply closing its inhalant siphon to avoid

srnothering under lem than optimal conditions. In contmst, Bayne et al. (1993)

suggesteâ that this was a selactive nsponw of the mussel to conditions of higher food

quahty. Importantly, in my study, if mussels were simply closing down in responw to

lsss than optimal conditions as suggesteâ by Clausen and Riisghd (1996). the

maximum ckannce rates at higher iIt and aîgal lords would not have baen obwmd.

This observation, pkir the nrponse of the muasel to incmasing abal conœntrationr

suppott the finding of Bayne et al. (1993). in that musuels am capabîe of a

compensatory feeôing bshavior which rllows to maintain a constant ingerteâ organic

diet from its food environment.

However, RiisgOrd and Lamn (1995) reœntîy m v M the litenture on the

filter-feeding chancterirticr of mirine invertebnte and conckided airt fibr-fedem a n

adaptecl to pump water continuourly at rates charactefistic of the m i e s , i.e., in filer-

47

feeding bivalves, the bivalve pump ic the evolutionacy msul of the interaction khneen

the organism and its biotope. My msub support this conclusion, as well as that of

Bayne et al. (1989), in that, maximum ingestion rates for mussels used in the study

wera obsemd under seston exposure regimes that would k considered Iess than

optimal for the mussel. This maximum (at intemediate algal and intemediate silt lord

exposures). most likely corresponds to the pfeûetemined fibr-feeding characteristic of

this specks for the particular eauary. However, under increasing algal concentrations,

and under extfemly high silt loads. mussels were alw abk to move from this filtering

maximum. and adapt their feeding strategy to maintain a constant organic ingestion

rate by wkcting organic particks and rejectng inorganic particies. Such an adsptive

abilw, elher through attering clearance rates andlor incmasing SEFs, on the part of M.

tmssulus would i n d d k advantageous under rapid and continuous changing quality

and quantrty of seston, such as nota by Stecko (1 997) for the coastal area from which

the errperimental mussels were sampîed.

In thrir studies on the feeding responw of the blue mussel chalknged with diet

matrices of differing quality and quantity, Bayne et al. (1993) no td that as the oqanic

content of the d i a incmasd (43 to 81%). SEFs decreased. Similarly, I found that at

the higher quality wston matrices, SEF declined as quality incmaseâ (30 to 71%), with

my data comparing well with the data of Bayne et al. (1993). However. SEFs also

declined under poor quality se8ton rxposurer. Under the condition of a poor quality

diet, the wbction of organic venus inorganic partickr is negligibk as the mussel

rttempts to m t L energy requimmnt by incorporatirtg aîî seston particies. Similarîy,

in high quaHty seaton matricm (i.r.. jurt aîgae alone), and as noteâ by Bayne et al.

(1993), no sorting war nquimd, Le.. them is no inorganic mrterkl to m m ; hence

again SEFs a n low. Maximum SEF was obsenmd when wston quality waa

48

approximately 40%, i.r. under conditions where the algal concentrations were high

enough that, despite an incmasing silt lord, the mussel could effectively improve diet

quality by sslecting only the abal component of the seston.

2.4.2. AuknMtkn of carbon h m va ylng raton mûicrr

The rate of absorption of carôon by the blue mussel has previously been shown

to be dependent on seston quality (Bayne et al. 1987). lgksias et al. (1996) have also

no td that absorption efficbncy of carbon by the cockk Ceraslodenne edule (L.) was

dependent on the organic content of ingested diet according to an exponential

satunting function, i.e., urbon absorption poslvely increased to an organic matter

content of ingested diet of approximately 45%, at which point carbon absorption

became independent of organic content.

When expresseâ as an app CA€% versus seston quaMy, then wos a positive

relationship, i.e., carbon auimilstion by the mussd war positively comlsteâ wlh

increasing seston organic content. When expressed as a tnie GA€%, however,

cebon assimilation was indepndent of seston quaMy. Tnie GA€% for the exposure

mgimes which containetd the Iargest amount of aigae, whrn vkweâ independently from

the nmaining data, appeared to ôe dependent on incnasing wston quahty. When

expnswd as carbon auimilation rate (i.e. a masure of the actual amount of carbon

as8imilated by the musml given the amount of PO& the mussd ha8 ingested over the

4 h feding penod), the rate of carbon assimilation by the mussel war indewndent of

seston quality. In other words, through a dynamic inteplay of aibon as8imilation

(which in tum is dependent on gut midence time and digestive procemes) and

49

clearance ntes, MyWs tmssulus is able to maintain a constant rab of carbon

assimilation - on average 29.3 k 6.9 (SE) mg C h-' gdw".

My findings differ from thow of Bayne et al. (1993) who noted that the rate of

carbon absorption by the blue mussel (Mytilus W i s ) was depndent on POM. In the

studies of Bayne et al. (1993) POM concentrations and seston quality ranged only from

0.63 to 5.85 mg C' and 53 to 97%, nspctively. In contrast, PO& conœntration and

seston quaMy for rny experimnts n n g d from 1 .O to 13.6 mg L" and 10 to 71 %,

respectivety. Hena, the gnater scope of the d# exposure mgime may have nsuited

in a ddfennt re~ationship k ing obsewed. lt is possible that the mlationship notd by

Bayne et al. (1993) was a consequena of the strong nlationship ktweem mussel

filtration rate and total particulste matter (mg c') that occurred in their study. As

determined in oui study and that of Bayne et al. (1993), carbon assimilation rates are a

product of ingestion rates and carbon absorption efficiency. îngestion rater an, in tum,

dependent on filtration rates. Bayne et al. (1983) noted no consistent ditferences in

absofption etficiencies foi their various 2 d dbt exposun mgim, i.e., carban

absorption was constant over the various exposum regirnes. lt is possibk that their

obwwed mlationship between carbon amiimilation rate and WMc was driven mostly

by dinerences in mussel filtration rates among the various diet mimes as opposed to

mfbcting the interplay between carbon assimilation efficiencies and mussel filhtion

rates.

In summary, rny findings suggest that the mussel is capabk of two feeding

strategies, which am not mutually exclusiw. As concluded by Riisgld and Larsen

(1995) and Ckuwn and RiisgId (1996), rnu88elr have a fibring maximum thrt is

preôetemined by the phybgeny of the m i e s . In my study, th18 filtering maximum by

50

Myfnlus t ~ u I u s was obtained at algae and sik concentrations of 20 x 1 o6 cells L" and

20 mg L" nrpectively, concentrations which would have been considered les$ than

optimum by studbs of C laum and RiisgOrd (1996). Further. at this filtering maximum.

highest coibon assimilation rater wem observ~. In addition, as proposed by Bayne et

al. (1993), muswls have the abilw to physiologically mspond to a changing food

environment. I found that at the highest si& and aigal seston exposure. mussels were

capable of sekding only the organic component of seston (i.e., high SEF) and repcting

the inorganic component of ssston (through PF production). In the absence of rilt,

musseb decreased CR wiVi incrsrsing atgal concentrations, whik maintatning PO&

ingestion rates. Henœ, M. triossulus was genenlly able to maintain constant ingestion

and carbon assimilation rates dunng itr feeding procem.

2.5. REFERENCES

Bayne BL, Hawkins AJS, Navano E (1987) Feeding and digestion by the mussel Mytilus eûulis L. (Bivaliva:Molusca) in mixture of SM and algal cells at k w concentrations. J Exp Mar Biol €col 1 1 1 : 1-22

Bayne BL, Hawkins AIS, Navarro El Iglesias JIP (1989) Effects of seston concentration on feeding, digestion and growth in the mussel Mytilus edulis. Mar Eco1 Prog Ser 55:47-54

Bayne BL, lglesias JIP, Hawkins AJS. Navano E. Heral M. ûeslous-Paoli JM (1993) Feeding khavior of the muswl Mfl11us edulis: re8ponses to variations in quantity and organic content of the seston. J Mar Biol As8 UK 73:813-829

Berg JA. Newell RIE (1986) Temporal and spatial variations in the composition of wston availabk to suspension feeder Crassostm~ vhyinice. Est Coast Shetf Sci 23:375-386

Clausen Ib, Riisgird HU (1996) Growth, filtration and respiration in the mussel MytiIus edulis: no evidence for physiologicol regulation of the fibr-pump to nutritional needs. Mar Ecol Prog Ser 141 :3745

Cossa D, Bourget El Piuze J (1979) Sexual maturation as a souics of variation in the relationship between cadmium and body weight of Mylilus eûulis. Mar Poll aull 1O:f 74-176

k c h o AW, Luoma S (1996) Fiexibk digestion strateghs and trace rnetal assimilation in manne bivalves. LHnnol Ocsanogr 41 : 568-572

Fegky SR, McOonalds BA, Jacobsen TR (1992) Short-term variation in the quantity and quality of seston availabk to knthic suspension beden. Est Coast Shen Sci 34: 393-412

Galois R. Richard P. Fricouit B (1996) Seasonal variation in suspended particulate matter in the Memnnes-ûkron Bay, France, using lipids as biomaiken. Est Coast SheH Sci 43:335-357

Griffiths RJ (1980) Natunl food avalbility and assimilation in the bivalve Chnnwnyti/us meMEOnaIis. Mar Ecol Prog Ser 3: 1 5 1 -1 56

Hamron PJ, Waters RE, Taykr FJR (1980) A broad-spectrum artificirl seawater medium for coastal and open Oman phytoplankton. J Phycol16:28-35

Igksias JIP, Unutia MB, Navam E. Alvares-Joma P, Lambtea X, Boughr S, Henl M (1996) Variability feding pmc8sws in the cockie Cemstodema Wule (L.) in responw to changes in wston concentration and composition. J Exp Mar Biol Ecol 197:121-143

Jirgenciwn CB (19sO) Bivalve fibr fading: hydrodynamics, biwnergetics, physiology and ecology. Olsen and Oisen. Fmdenborg, 140 p.

LoW PB, Bajdik CD, Belkhonde SP, Jabon SE, Longefich HP (1991) lmprovd protocol for colkcting mussel watch sp8cimns taking into account sex, size. condition index, shell shape and chronological age. Arch Environ Contam Toxicol 1 O2:5l 3-518

MacOonald BA. Ward JE (1994) Variation in food quality and partick sekctivity in the sea scalkp PIawpecten magellanicus (Mollusca:Bivalvia). Mar Ecol Prog Ser 1 O8%1-264

Navam JM, Clasing E, Umtia G, Asenicio G, Stead R. Hensra C (1993) Biochemical composiüon and nutritive value of suapended particulete matter over a tual Rat of southem Chile. Est Coast SheH Sci 3759-73

Navano JM, lgksias JIPI Camacho AP, Labarta U (1996) The effect of diets of phytoplankton and suspended bottom material on feeding and absorption of raft mus8els (Mytdus ga/bpmvMcielis Lmk). J Exp mai Biol Ecol 198: 175-1 89.

Navarro JM, Thompwn RJ (1995) Seasonal fiuctuation in the size spectra. biochemical composlion and nutritive value of the seston availabk to a suspendon-feeding bivalve in a subaftic environment. Mar €col Prog Ser 125:95 - 10g

Poulsen E, Riisghrd HU, Mehknkrg F (1982) Accumulation of cadmium and bioenergetics in the mussel MytiIus edulis. Mar Biol68:25-29

RiisgOrd HU and L a m PS (1995) Fibr-feeding in manne rnacio-invertebrates: pump chsracteristic8, modrling and energy wst, Biol Rev 70:6?-108

Schubel JR (1971) Tidal variation of the sue distribution of suspended sedimnt at a station in the Chesapeaûe Bay tuibiôity maximum. Neth J of Sw Rus 5: 252 - 266

Smaal AC, Verhagen JHG, Coosen JI Hass HA (1986) Interaction betwwn wston quantity and qwlQ and benthic suspension feeden in the Ooserchelde, the Nethedands. Ophelia 26385-399

Stecko JRP (1997) Contnaing the geochrrnistry of suspemded particulste matter and deporited W i m n t s of the Fnwr river ertuary: implications for m ta l expowre and uptake in estuarine depod and Mer feeden. M.Sc. thwir, ûept. of Biokgical Sciences. Simon F n w r University.

Vahl O (1 972) Etficiency of paitick retention in Mytilus dulis L. Ophelia 1 0: 1 7 - 25

Vshl O (1980) Seasonal variation in seston and in the growth rate of the Iceknd scalkp, Chlemys idandka (O f . Malbr) from Babfprd 700 N. J rxp mai Biol €col 48: lSS-2M

Wang WX, Fisher NS (1996) Amimilition of trace e b m n b and uibon by the musel Mytius edulis: effects of food composition. Limnol Ocsanogr 41 : 197-207

Wang WX. Fisher NS. Luoma S (1995) Auimilation of trace ekmnts ingesteâ by the murwl Myfnus eûukio: effects of abal food abundrnœ. Mir €col Pmg Ser 129:165-176

Widdom J. Fieth P, Womll CM (1 979) Relationships between seston, avaikbk food and feeding activity in the common mussel ( M w s edulis). Mar Biol 50:195 - 207

Winter JE (t973) The filtntion rate of Mfli/us edulis and ih dependence on algal conœntntion. masumâ by a Continuous Automatic Recording Apparatus. Mar Biol Z2:3 17-328

INFLUENCE OF A 8ELECTiVE FEEDING BEHAVIOR 6Y THE BLUE MUSSEL, Myülur brorurtw, ON THE AOMYUTION OF '@@cd PROM ENVHIOWMENTALLY

*) This chapter is exprnded fmm an articb in Mar. Ecol. Pmg. Set (in prem), Influence of a mbctive feecîing khavior by the blue mumwt (Mylius tmu lus ) on the assimilation of '%d h m environmentrlly rebvant seston matrices, (2. Arilin and L. BendelCYoung) .

The objective of this portion of the study war to detemine the influence of a

selective feeding strategy on the assimilation of '%d ('%d-~€s) by the blue mussel

(MytiIus tmssuIus). Two complementary experiments which used five seston matrices

of different seeton quality (SQ) wem conducted; 1) riîgae labeW with Iwcd was mixed

with unlaklsd sitt, and 2) l a k W sin was mixed with unlabeffl algae. '"C~-AES were

detenind by a dual tracer ratio ('Os~d~''Arn) (DTR) method and baseâ on ingestion

rate of lWcd by the musml (IRM) (i.e., total amount of '''cd ingested over the fourhout

feeâing pend). As a re8ul of th8 nonconsetvative khavior of 214~m, Me DTR

underertimated mussel 'Os~d-A~s as compareâ to the IRM. Themfore, only IRM

detemined ' m ~ d - ~ ~ s wem considemd fumer. When only algae wem IaôeW, '%d-

AEs wrre proportional to the organic content of the diet (diet quality, DQ, P = 0.98. p g

0.05) with the highest ' O e c d - ~ ~ occumng at the mussel's filtedng maximum whem the

highest carbon assimilation rates have bmn obmrved. Howevrr. for the labeîed silt

exporum, '*cd-AES wen independent of W with maximum vakies of 85% occuning

at al1 diets rxcept for sitt-only exporum. 'Oscd-~€ for the silt-only exposure was 36 %,

suggesting that digestive processes that occur in d k h of mixed abae and sih wem not

operating as ef(icMnUy in the silt-only exposums. 109 CdAEs cornlated with ['%dl in mume1 tissue (r = 0.63, P < 0.05), with the

radiotracer as8imilatsd from dit kkkd matricer comisponding to the gnatest amountr

of '''cd activity within the mussel. These resultr suggest an active and parrive

assimilation of '*cd fmm the and dît components of seston, mspectivey. Active 109 Cd-AEs will be proportional to W wini possibly highest assimilation occumng at the

musnl's f i l r ing maximum. Passive 'Oe~d-A€s will k dependent on the amounb of

metal awcirted with the inorganic wmponent of seston, with digestive processes that

are adivated in the pmwnce of abae concumtly deoibing inorganic cadmium.

57

Luoma 1985. Zang et al. 19W. Absil et al. 1994 and Wang et al. 199S), organiully

coated silica kads (Decho and Luoma, 1994) and natural W i m n t (Gagnon and

Fisher, 1997).

An important aspect of the incorporation of AEs into predictive modek of

bioaccumulaüon is oMaining values that a n repressntetive of the diet that is actually

ingerted by the fi lr-fwding organirmr under natural conditions. Henin lies the

diffculty in the use of labotatory derived A€# from a single food source, such as algas,

foc use M predictive models of metal accumulation. In the environment, filter-feeding

organbms are exposed to a highîy dynamic, compkx food source whtch constantly

changes in tennr of quality (srount of organic relative to inorganic matter) and quantw

(concentrations in mg L") (Fegley et al. 1992). In n s o n w to this dynamic food

condition, many fibr-feeding organisms have developed a highly sekctive feeding

khavior that depnding on seston quality and quantity, allows for sekction of organic

over inofganic particks (undsr conditions of high quantw and quality wston) or both

organic and inorganic particks (undet conditions of low qualtty and low quanttty) for

ingestion (Bayne et al. 1993, Ward et al. 1997).

Depending on mussel feeding khavior, when seston b of high quslity and

abundant, mta l eontaminants auociateâ with only the organic fraction of seston rnay

tm ingested and avaikbk for uptake. In contnst, when the muswh are exporsd to

low qwntity and low quality of seston, metal contaminant8 aswciatd with both the

inorganic and organic components of seston rnay be ingeated and as8imilated by the

mussel. if this ability of the musml to choow sp8cific cumponents of the seston in a

quantity and quaMy dependent rnrnnet has not t w n taken into actount, then

prodidive moddr of m t a l contaminanti uptake, basd only on one component of the

58

diet. such as oqanic content, may undemstirnate the amount of metal that the mussel

is actually ingesting .

Therefon, the objedve of the second study was to detemine the influence of o

selective M i n g , that either excludeci (complete sorting) or indudod (minimal sorting)

the inarganic components of seston. on the assimilation of ' * ~ d by Vie blue musse!.

To meet th& objective, i expos8â mussels to seston matrices that 1 had previously

shown (Chapter II) either to k completely ingested with no sorting occuning (low

qwntity and quality dists) or to evoke a sekctive feeding bahavior, such that only

organic matter was ingested. Two sets of expeiimnts with the variow serton matrices

were conducteci; 1) aîgae k k W with lWcd were mixed with unlabekd sih. and 2)

lakleâ sitî were rnixeâ wlh unlakkd aîgae. Resulng CdAEs from vanous seston

matrices wem thon relatd to seston quality. Uîtimately, msutts from thiu study will be

used to prsdict Cd concentrations in the blue mussel (Chapter V).

3.2. MATERUCS AND METHOOS

3.2.1. Fkîd colkcdkn of muab

Mussels (Mytilus trossulus) wen collectecl ftom the intertidrl area along the

coast of Honrshoe Bay. British Columbia. Canada. Mussels (44.9 + 208 mm shell

kngth, 0.18 I 0.01 g dry weight) wem rcclimatizeâ to experimental conditions

(temperatum 13 i 1°C, mlinity = 28W0) for two w e e b ptior to use in each expriment

(Bayne et al. 1978). Dunng the acclimation pfiod. mussds wen fed the diatom

ThaIassicwim pseudonana daily, and the seawatet was changecl on a mgular basis.

Prior to the ïwding exprimnts, musmls were sepanted h m theit byssal-thnads

attachment to one another. brushed ckan and kept for apptoximately 15 min in air.

This proceâute ensurd that only live mursels wen uwd in the experhent as those

mursels that w m not viabk did not mspond to k i n g submerged in mawater following

the 15 min exposute to air.

3.2.2. $raton compomiüon

Five seston (SPM) matrices wen sekcteâ b a W on my pteviour rtudies (Tabk

2.2, Chapter II) which indicatsd that depnding on the quality and quantdy of the

seston, mussels have the rbility to sekt either organic particks or bath inotganic and

organic particks for ingestion. As shown in Chapter II, the wrting efficbncy of th9 blue

mu8sel nnges from -18 to 0% when exporrd to low aeston quality (SU) (< 18%) ot

high SQ (60 -70%), with the maximum sorting efficbncy occumng at an intemediate

SQ (40%). At thir maximum rorling efficiency, musaeh wes crpabk of intnating a

SQ of 40% organic mattet to a dia quality (M1) of 60 - 70% organic rnattet. Furthet, at

60

maximum tsston ingestion ntes (where mussels c k t a d the greatest numôer of

inorganic and organic particles from the water over a given period of time), which

occutmâ at a brs than optimal wston quality (SQ = 20%). the highest amounts of

carbon were assimilateâ by the mussel from its d#. This led me to hypothesize that if

cadmium assimiiktion followed a dievenergy pathway. the highest 'Ogcd assimilation

efliciency by the m u m l would k obsewed at the mussela' filtering maximum. Ba$ed

on these findings. three seston matrices were preparetd: 1) algae only. with a SQ of

60%; this matrix represented the diet after maximum sooiting (tejedion of al1 inorganic

parlieles) of seston of only 40% organic matter, 2) 2 and 10% SQ; low quality and

quantity where no sorting occun and both inorganic and organic components of the

seston am includeâ in the dia, and 3) 18 - 20% SQ; minimal sorting and where the

fibnngoptimum for thk specbs was obwnred.

3.2.3. Puhchrw M i n g rrpwhmnb

Two separate seta of expriment8 (Fig. 3.1) were conducted under fbw-thmugh

systems (Fig. 2.1). T h fimt expdments includd labekd-aîgao and unlabekd sil.

The marine centtic diatom Th8/8SSjOSim pseudonane was gtown in 4.0 L h rnkch

fiasks containing 3.0 L of fibred (0.45 pm) natutal seawater wdh the nutrîent-

enrichment solution of Hamson et al. (1980). Two days after inoculation, the algae

wem spikeâ with 185 kBq L" lacd (in 0.5 M HCI, Dupont) and 74 k8q L" ' " ~ m (in 1 .O

M HCI. Isotope Pmduct Lab.) a8 an inerl tracer. Algae were exposeâ to the

ndiotracem for 4 d and kept on a 14 h light : 10 h d8rk cycle at 16 O C . The algae were

hawested rRer t h y h d undergone log phase gmwth and w m considend unifomily

labekd. The Irbekbiîgau wsre then counted by partick counter (Coubr counter.

mode1 TA-II). and diluteâ to concentration of 20 and 1 50 x 1 ml18 L1. Ugae (20 x 1 0'

Fig. 3.1. Flow-chart of two-typs of experimental pmtocol: a) Iakkd-algal expriment. and b) labeîed-si# experiment.

Mixture of labeled-

aigae and

unlabled-silt.

'Hot' feetding for 4 h. Collect faeces and pseudofaeces.

'Cold' feeding for 24

h; mussels feed on algae.

Collect faeces . Measun 'Oecd (L 214 Am in tissues, pseudofaeces &

faeces.

b) Lakkd-rYt

Mixture of labeled-silt

I 'Hot' faeding for 4 h. Collect faeces and

- 2 pseudofaeces.

I 'Coîd' feeding for 24

h, mussels feed on

-!siBEm- aigre.

Collect faeces.

T l u i m FE PF " k m in tissues,

pseudofaeces &

Note: PF = pwudofaeces, FE = faeces Expefimnh wem done in thme repliater

62

ce& 1") wen then mixed with the unlabakd-itt component at concentntions 5, 20

and 50 mg L", to obtain a seston quality of 18.2, 20.6 and 10.3% oganic matter.

mspectively (Tabk 3.1). The SM- and algaaonly matrices contained 2 and 60%

organic matter. mspectively .

The mixtures of Iabekd-algae and unlaôeleâ-silt particbs ('hot' feding) were

given to the acclimated mussels that were placed in treatment tanks. T h m mussels

per tank were expowd to both food and dismtved 'Oscd. and one mussel per Pkniglas

chamkr was exposed to the dirsohmd 'Oscd alone. Experimnts were fun in thrw

nplicates with 13 groups of m u w l per trial. ARer 'hot' feeding for 4 hl musmls were

tnnsferrsd into a 1.5 L chamôer and fsd with unlabled rlgae (cou feding).

Compkte faecets (FE) and pseudofaems (PF) wem colkcted at 15 min. 4. 8 and 24 h

durini this coki feeding priod. Atter 24 h, the mussels were sacrificd to detemine

the amounts of accumulateâ radiotracm. Radioactivity of '%d and " ' ~ m in FE. PF

and mussel tissues wem measured using a Canberra Model 2030 gamma counter

equipped with a Na-iodide crystal detector. Gamma emissions were detected at 22 keV

foi '"cd and 60 keV for m. The marurements were conactsd for background.

and for the decay of isotopes.

The second experimnt was identical to the lakkd-algae experirnents except

that mussels were exposeci to mutturer of kbebd-sitt wlh unlaôekdilgae. Kaolinlit

minml (average diamter 4.8 pm) (Engelhard Corp., Pigments and Additives Division . Edison, NJ) was u a d for the ri4 cornponent. Thtw different sitt concentrations (5, 20

and 50 mg L") wem spikd with the mm amunt of radioisotope as t h l a k W - a b w

experimnb in 25 m l piautic ôeakrn with 0.45 pm fiîtemd mawater and sonicated to

produce unifom mixtures. A M 24 h. the kôekd sik rlurry wrr t rans fed to 4.0 L

Tabîe 3.1. Charaderistics of SPM and feeding response of mussels under ditferent SPM matrices. Aigae! : sili = (x 10' œlls L-' : mg L' ). SQ = seston quality. W = diet quality. the adual food ingerted by mussels after soiling process (Chapter II); IR, = ingestion rate of organic matter; IR,, = ingestion rate of inorganic malter. ûue C-A€ = carbon assimilation efficiency. a,b,c and d are means I ISE

Seston SPM' Quality (%) i b b IRpm" tnie C-A€

ab- Sili (mg L-') SQ W (mg h" gdw") (mg h" gdw") %

*)The se& of tests that were done by mixing 20 x 10' aîgal cells L" and 5 mg sin L-' were mîtd an average of 23.2 î 2.14 % organic matter (n = 7). Thete was no significant drfferenœ to that of SC2 = 18.2 % (Student t-test. p > 0.05)

64

kmbach flasks and diluted to 3 L with ftltered bbawater. Baseâ on the pnvious studks

by Steccko and BendeII-Young (1999) demonatrated that the majority of the lmcd

desorbs from SPM within the first 100 h of Iabeling. Hence, the l a k W silt was stived

vigorously, then alloweâ to sit for four days to allow loowly bound '"cd to desoorb from

the silt surface. The silt solution was filtered, and the recovered radiotabeled silt mixeâ

with unlaôeld algae (20 x 10' cells L") to obtain the SU'S descnbeâ for the algae

labekd experiments (Table 3.1). The same procedum for counting radioactivity of FE,

PF and mussel tissues was done as the fimt set of expenmcwits.

To ensum conristency and that the radiolabel did not desorb from the seston

matrices, both wston and watef in experimental chamben were sampW every 60 min 108 for each experimnts duiing the 4 h feeding peciod. Cd activw in water did not

increase indicating mat no '%d d e s o M h m the prepamd matrices over the coume

of the 4 h exposun (Fig. 3.2). Similady. the '"cd of the seston matrices for al1

exposures remaineâ constant over the 4 h feeding period (Fig. 3.3) indicating that

mussels were expowd to a constant amount of radioactivity via the preparetû seston

matrices during the expriment.

3.2.4. S.rton qurlity (SU) vrnur dkt quiMy (W)

Seston quality is defined as

whem POM is patticulate orgonic matter and SPM ir surpendetd particulate matter and

both are in units of mg L". Oist quality (W), which corrects SQ for sebctiw fwâing

109 Fig. 3.2. Cd activity in sea water (dpm L*') dunng the 4 h expecimntal exposure. (a) Seawater h m experiments in which only aîgae wrs Iabeled. (b) Semater from experimetnts in which only silt was lakled. Values = (man *ISE, N =3). Ratios a n amounts of algae (x 10' mWs L") to SIN (mg L"). The rit concentrations of 5, 20 and 50 mg L*' repnsetnt 149, 482 and 1267 x 10' particles L" . respectively .

Time (h)

O 1 2 3 4 5

T i m (h)

Fig. 3.3. '%d activtty in seston matrix (dpm L") during the 4 h experimntal rxposure. (a) Seston fmm experiments in which only algae was labekd. (b) Seston hom experiments in which only dît was labkd. Values = (man 11 SE, n =3). Ratios am amounts of algae (X 10' œil8 L") to SM (1 mg L"). The silt concentrations of 5, 20 and 50 mg silt mpresent 149, 482 and 1267 r 106 particles L",

O 1 2 3 4 5

T i m (h)

67

khrvior of the mussel (Le., rmpction of inorganic seston components via

pwudofaeœs production), is defined as

where IR, is arnount of ingested particulate organic matter, and IR ,,, is amount of

ingested organic plus inorganic rnatter in units of mg h" gdw".

Cadmium assimilation efficiency (CdAE) is defind as the proportion of

ingested cadmium ('Oscd) rstsined after cornpktion of digestion and gut evacuation.

Wang et al. (1 995) noted that kss than 10% of 2 4 ' ~ m was mtained in soft tissue after a

24 h depuration priod. Most unassimilated lmcd was egested within the first 17 h

after which very little '%d appeared in the faeces or was lost from tissues. Based on

109 thb teference, Cd-A€ was detemined after a 24 h depuration period.

tw Cd-AEs were detemind in two ways. Çimt, 'Oe~d-~€s were calculated

based on the dw l tracer ratio method (DTR) as described by Fisher and Reinfelder

(1991) and Luoma et al. (1 992) where,

109 CdAE = ~( '~~d?~'~rn),m - (lWcdP1~m)c, )/(loe~dP'~m),] x 100% (3.3)

and when ('Oe~dP4'~m)- it the ratio of lWcd and 24'~rn activities in seston and

('m~dpl~m)- is the ratio of 'Oscd and 2 4 1 ~ m actmties in faeces. This method

awums that '%d passe8 thmugh the digestive tract at a similar rate to ' " ~ m and the

l o u rates of lWcd and from fmœa into the medium a s companbk.

Second, 'O'C~-AES wem cornputed based on the amount of '%d ingested over

the 4 h feeding priod (IRM) as detemined in Tabk 2.3 as followr,

where IRrm is ingested 'Ogcd frorn algao (dpm h-' gdw*'), ['Oecdly, is '''cd

ndioactiwty in l a k k d abae (dpm mg") and IR,, is ingested organic particles after

correction for the sorting proces8 (mg h" gdw-'). Ingestion rater wem detemind as

the pcoduct of the ckannco rates and the seston quantity minus pmudofaeces

production. When the laôeled-sitt wmponent of serton war us8d for exposums,

equation (3.4) was modifmd to

when ['oo~d]un is 'Oscd radioactivw in labled sitt (dpm mg") and I%, is ingestd

inorganic particks after comcton for the soding proœss (mg h" gdw*').

Appomt ' O e ~ d - ~ € s (Le., assimilation of cadmium from wston, uncomcted for

sorting) were calcukted sr;

app. ' %~-AE = [( IRIw - FEIw )/ (IR[- + PFld] x 1 00% (3.6)

and

whem app . lW~d-~€ is apparent 'Oscd assimilation efiusncy. FEw is the activQ of

'%d in faeces (dpm h" gdw"). and PF[rn is the activw of mjected 'Oecd in

pwudofaeces (dpm h" gdw-').

The true lW~d-A€ from both aîgae and siit components of the diet, i.e.,

assimilation of cadmium from the diet that is adually ingested by the muswl after

sorting ptocess. wem determined as.

where IRlcdw, is ingested 'Oscd from the algal source, IRicciw h ingested 'Oecd from

the inorgantc wurœ.

3.2.6. '%d rcüvîty in imnwl (luw

To correct foi the possibk contribution of the dissolvd fom of the radiotracer to

mussel tissue '%d acüvtty (i.e.. uptake ftom wlute plus diet mther than just due to diet

alone). '''cd activrty determined for control mrwk exposeâ only to seawater w u

wbtnded fmm muasel8 exposai to both Me k h d seston plus seawater. These

correctd vakes wem u W in detemining the nlatiomhip beheen '%~-AE and

m u m i tissue 'Oscd activity.

2 . Statkticil rnrlyrir

Statistical analyses for the partitionhg of radiotracen ('Oscd, '''~rn) in faeces,

pseudofaews and '%d-A€ data won done with Systat 5.0 software (Wilkinson et al.

1992). Significance of ditferences in 'Oecd-~€s deteminetci by OTR venus IRM. and

between a p p l W c d - ~ € venus tnie- '"~d-~€, was detemined with Student 1-tests.

Comlation analyses wem uwd to detemine nYionships between lWcd-~€ and 1)

dbt quality (W), and 2) corndeci mussel tissue '-cd activi(y, and 3) tnie carbon

assimilation efficiency (C-A€).

The rates of logcd and " ' ~ m production in pseudofaews (PF) (rejected

matefia!) and faeas (FE) (egestion) npment the radionuclide biodeposition rates

(Tabb 3.2). The production rates of '%d and '''~rn indicated that different

biodeposition proasses occurred for m u s ~ l s exposed to labeld-algw vemus those

exposeâ to labeted-dît. For the labeîed-aîgal experimnts, the biodeposition rates of

'''~d and " ' ~ m in PF's wen generally lower than thom of the FE component, except

at a SPM concentration of 43.4 mg L" (Fig. 3.4a. and b). Th8 deposition rate of 'Oecd

and "'Am in FE decmased with a concurrent increase in the deposition rate of 'Oscd

and " ' ~ m in PF with incmaring SPM concentrations. In contnst. for the Iabded-silt

experiments, '*cd adivity in FE was nlrtively constant acrors tnatment exposures

(16 x IO' dprn h" muswrl), but the activdy of '%d in PF increawd with incmasing

SPM concentration (16.3 x IO' to 34.3 x IO' dprn h" mussur') (Fig. 3.k). A similar

pattern was also shown for "'Am in PF with bideposition rater increasing from 2.8 x

lo5 to 16.3 x 10' dpm h-' mussef'with inmwing SPM concentrations (Fig. 3.5b). The

active procews of repcting the siît camponent of the -ton by mu8seIs resulted in an

incnaw in 'Oscd and ' " ~ m deporition rate in PF, but. the deposition rate of the two

radiotracers in FE remained constant.

This biodepositional pattern detected with the use of radiokkkd algae and 8P

ruggetr that the musml i8 rctiveîy wfüng Me sestolon matricer (the 88bction of aîgw

ove? silt mutting in an i n c m a d W), even though a pmvious study (Fig. 2.7) has

Tobie 3.2. 'Oscd and " ' ~ r n in seston (SPM). pseudofaeces (PF) and faeces (FE). Algae : Siît = (II 10' celk L" : mg c'). The silt concentrations of 5. 20 and 50 mg ~ % î t represent 149, 482 and 1267 IC 106 particies L-'; respectively. Values = ( m a n I ISE. n = 3). SPM = wspended paiüculate matter (seston). PF = pseudofams. FE = faeces, '%~-AE = cadmium assimilation efficiency, Exp. I = experimant with labeteû algal component, Exp. II = experiment with Iabded SM

Seston SPM (IC 10~dprn mg-')

Algae Stlt ' OeCd " ' ~ m

Exp. I 20 50 0.1 Hl01 0.8I0.02

20 20 0.5îû.09 2.5I0.30

20 5 1 .OI0.08 4.5I0.43

150 O 2.4M.18 14.811.17

Elp. I O 50 2.1 Hl. 18 35.0 0.50

20 50 3.5I0.51 6 4 . W 1.74

20 20 2.510.18 54.2 k 1.87

20 5 1.3S.11 26.7 k 1.36

PF (XI O' dprn h" mussel-')

' Oecd '"Am

FE (xlo3 dpm h" mussel")

'Oecd 24'~rn

Fig. 3.4. Biodeposition rates of 'Oscd (a) and ' " ~ m (b) in pseudofaeœs (PF) and faeces (FE) by musse18 exposed to se8ton concentrations wlh Iakbd-algal component.

SPM conc. (mg L")

Fig. 3.5. Biodepotition rates of '%d (a) and " ' ~ m (b) in pseudofaeces (PF) and faeœs (FE) by murseh exposeâ to four sestoton concentrations wiai labed-sitt cornponents

algae 20 20 20 silt 5 20 50

SPM conc. (mg L")

shown uring the gnvirnetnc method that soorting of seton at a wston qurltty of less

than 20% is very low. The excuption was at the highest SPM concentrations when the

pmvious study showed that the mussels teûuce thek Citering activity.

3.3.2. Cadmium i u M l i ( k n ~(Ackncy ('%~-AE )

Estimates of "%d-A~s from algw alone baed on IRM were 6 t ims grnater

than those deterrnineâ by DTR. In contnst, the two estimates of CdAE from silt alone

wem not significantty diffemnt from one other (Fig. 3.6). For the labled-algal and

unlaôeld-si& matrices (Fig. 3.7a) and the unlaôekd-algal and Iabded-silt m a t h s (Fig.

3.7b), ' % ~ A E S baseâ on OTR wete genenly lesr than thow detemineci by the IRM.

Diffemnœs in ' O ' C ~ - A E estimations based on DTR venus IRM a n the terult of

the ' " ~ m behaving nonconwwativeîy. When '''Am functioned as an ineort tracer, i.e.,

whete the amount of '*'~rn ingested was apptoximatey q u a i to the amount of " ' ~ m in

faecus (at seston mixtures of Iabekd-aîgw and silt of 2050, 205 and 0:50, Tabk 3.3).

estimates of ' O e ~ d - ~ € b a W on the two methods won not diffennt ftom each other.

However when thir ratio war greater than one, (i.8.. when " ' ~ m w a t taken up by the

mussel and behrved non-tonservatively), '%d-~€s basecl on DTR undenstirnated

'Os~d-A€s estimted from IRM. FuiMer, for labekd abal expowns (with the exception

of the mston mixtures outlined above) the ntio of " ' ~ m ingosted foodIfaeces w8s

approxirnately 4; whik for expriment8 whem sitt was kkkd th18 ntio avengeâ 8.

Fig 3.6. Cadmium assimilation efficiency ( ' a ~ 6 ~ ~ ) ftom laôekd-algue and Iabkd-sit uJng a dual-tracer ratio mthod (OTR) and ingestion rate m88unmnt (IRM).

= a signircant dilhmnce betwwn the Iwo mthodr (Student 1-test, p < 0.05)

[ ] DTR IRM

Fig 3.7. ?X-AE baseâ on dual tracer ratio method (DTR) and ingestion rate masunment (IRM) at diffennt d Y qualitks (W). (mans f ISE, n d). a) Irbekd-algal rxperiments. b) Iakkd-rilt expriments.

DTR n IRM

1-1 DTR g IRM

79

These fimdings wggest that not only is ' " ~ m non-consenrative. but as well, the rmount

that is taken up by the oiganism will depend on type of wbsttnte that is Iabskd.

Because of Mis, '*cd-AES bawd on the DTR are not wnsidrmd for fuither analysis.

Apparent and trw 'Os~d-A€s crlcukted for both the abae and silt I a b W

exposuns were not significantly dWersnt hom euch other (Student's t-test, p > 0.05).

Given that the quality of the seston matrices for thew expriments wem wkcted

bawd on my previous studies, a SQ of ca. 20 %. sorüng shouiâ k minimal (Chapter

II), and both the apparent and ûue '%~-AE rhould be companbb to each other. The

pmwnt studies rhowed that although them w8t no statistically rignincant diffemnœ

between the two meaturemnts of ' w ~ d - ~ ~ s by muswlr. the tnie '%dd€s were

generally highr than the apparent lmcd-~€s (Fig. 3.8a and 3.8b. a and b mfer to

labekd abae and UN, iwpectiveîy). As pmvtously indicated by the 'Oscd and 241~rn

biodepositional pattern, rom wrting of the seston by the m u ~ l s war apparently still

occurring .

The twe '%~-AE deteminecl for the aîgae Iabeleâ exp8riments comhted with

diet quality (* = 0.98. Fig. 3.9), wini maximum lWcd-~€s occurring at a dW quality of

33% (25 mg L" SPM). The ' œ ~ d - A ~ for the Jh-onty tmatmnt war 36%. With the

exception of the siit-onîy exporunr, true lm~d-A€s detemineâ for the rilt taôekd

108 expriment$ wem 85% and w n independent of diet quality. Cd in mussel tissues

(aftet comcting for the posaibîe contribution of ndiotmcer taken up from sotution to

rnuwel tissue activity) war pouitiwly cumliteâ with '%AE (r = 0.63; p < 0.05, Fig.

3.10) with maximum murirl lmcd tissue concentrations comsponding to which

matrices of labebd dît.

Fig.3.8. Apparent and tme '%~-AE bawd on ingestion rate mrwnment (IRM) in relation to dia quality (W). Values = (man I 1 SE, n =3). App. and t ~ e '%d- A€ from kbsled-algae (a), App. and true lrncd-~€ from Iabded-sitt (b).

30 40 50 60 70

Diet quality (%)

Fig.3.9. Rektionship between tnie cadmium assirniiatbn effichncy (true '%d-A€) and diut quaMy (W). Values a n maans t ISE, n = 3; r = 0.98, p < 0.05, for tme 'O@C~-AEI for the Iabebcî-aîgai expauma wmur ôiet quaîity. C i t d denotes when mussels displayeâ maximal ckrnnce rates (CR) in pmvious study (C hapter II).

Fig.3.10. Rektionthip between '%d concentration in the mussel tissues (['*cd]-) and true '@cd arnimilation eflicbncy (lwcd -A€).

3.4. DISCUSSION

Given the importance of incorponting into pfedictive models of metal

accumulation how effectiwly a fifier-feeding organism assirnilates a metal of interest

(e.9. Cd) from its diet, a number of studies have mpoiled CdAEs for several fibr-

fctding oqenisrns !rom vanous single faad sourcss (Table 3.4). Most notable are the

studies of Borcharôt (1985) and Wang and Fisher (19960) who reporteci, based on

studies whem musmls were rxpo8ed to various species algae of diffennt organic

content, that cadmium asdmiktion is proportional to the assimilation of carbon by the

musml. These studius have furthered understanding of how muswls obtaind tmce

metals specifially Cd fmn a pure food source and identifii potential conditions where

uptab from food may k maxirnuecî. However. an important aspect not included in the

studies summarized in Table 3.4 is the responw of the filter-feeding organism to its

food environment.

The b lw musml i8 capabk of a highly sebctive feeding behavtor, which

depending on the qualitylquantity of L food condition, will resuît in the wkction and

ingestion of ju8t ofganic matter or the ingestion of bath inorganic and organic m t o n

components. Hence, tather than simply ingesting one food component ruch as abae

(as in pmvious studim). the musml will ingest a combination of inorganic and organic

matter, depending on wston quslity and quantity. 00th componentr of wston

themiore have the potential to contribute to the amounts of cadmium uîtimrtely

assimilateâ by the organisrn.

Tabk 3.4. Cadmium assimilation efficiency (Cd-A€) in mussels (Mytius edulis, M. tmssuIus), oy stem (Crassostrea vienaca) and dams (Mercenaria mercenana, Macoma baîîhica, P o t a m ~ c ~ ~ u l a amurensis).

SM Sire Exp. Design Foodlpartide b Partick quantrty CdAE (%) References (mmSL) STIETIMTIS8 (mp L")

M. edulis

M. edulis M. edulis

- SS/300/-i26 I. galbana -

35 SSl30/15/28 T. pseudonana

30 SS/30115128 1. pseudonana P. tnbmutum 1. maculata ChW =pVt- Dinoflagellates

30 -35 SSl30115128 T. pseudonana 30 - 35 SS/40-60118128 T. pseudonana

l. gelbana

ReinfeiderB Fisher (1 99qe

Wang et al. (1 995)'

Wang& Fisher (1 996a)'

Wang& Fisher (1 996b)' ReinfeMer et al. (1 997)g

3.4.1. '%~-AE k, nlrtkn bo a wkctlvo M i n g khrvkr

Results of the expowrs exptimnts whm the algai component of the rnatnx

war labeM indicated that 'Oscd-~€ was stmngly comlateâ with dbt qualfty (? = 0.98).

Maximum ' O s ~ d - ~ € occufred at a seston puality of 60% (just algae) and at 33% (25 mg

L" SPM) whem maximum carbon assimilation eficiencies for this s p c k s have

pmviously b e n noted. This relationship supporteci the previous findings of Wang and

Fisher (1996b) who noted a posithre relationship between carbon assimilation and

109 CdAE for the filr-feeding blue muswl.

I hypothesized further. therufore. given the findings of Wang and Fisher

(1996b). that ' O e c d - ~ ~ would be proportional to carbon assimilation effciencies and

that 1 would be at the mussel's filtering maximum that ' * C ~ A E by the mussel would be

maximized. To test this hypothesis, I mgresseâ 'Oecd-A€ against the carbon

asrimilotion efficioncies (C-A€) pmviously deteminetci in Chapter II (Fig . 3.1 1 ).

However. a strong comlation batween the two variables was not obaerved. Further.

for rxposures when silt war kkkd. 'Oecdd€ was indepndent of W (Fig. 3.9) and

C-A€ (Fig. 3.1 1); maximum assimilation of 85% occunad for al1 DQr except for the tilt

only when '%~-AE was h a l of what war obsurveâ for matrices that contuinmi organic

matter.

Thesa two distinctty diffennt patterns in 'Oecd assimilation fmm labled-iigal

and I a b d - a i l exposums suggest two proœmer. Fint, it is an active auirnilation of

the mtal from ibn. Reinfeiûer et al. (1997) have shown that AEr of cadmium in a

numbei of bivaîves wem di- mlated to the proportion of the e k m n t in the

Fig. 3.1 1. Relationship between Cd assimilation efficiency ( 'Oecd-~~) and tnie carbon assimilation eflicbncy (tiue CAE). Circb denotes whem muswls dispbyed maximum ckarancx rate (CR) in pmvious study (Chapter II)

cytoplasmic fraction of Mgesteâ phytoplankton. ln my experimnts, Cd wouM have

been incorporateâ within the algae (as well as aswciated with the algal surface);

heme, increasing the amount of rlgae wouîd incnaw the amounts of cadmium

potentially availabk for uptake by the muswl. It wouM folbw that the mon algae, the

more intense the digestion process to bmakdown the algae and hence, the gnater the

amount of cadmium avaikbk for uptake. However. and as suggested by my findings,

the amounts of cadmium assirnilated by the m u s d moy not necessarily be proportional

to amounts of assimilatd carbon, i.e., there ir no mason to amurne that the two

ekmnts follow the same physiokgicrl pathways. Rether, for purs aîgal dists, dkt

quality and carbon assimilation by the mus& a# comlated with each other, henœ Cd-

AEs detenined for pure algal diets wouM be mlateâ to both variables. For diet

mixtures, and as no td in the pnsent study, when mussels a n chalknged with algae

mixd with siît, diffemnt digestive processes are possibly evokeâ (e.g., longer gut

msidenw ttm) as compareâ to the pure rlgal diet. uncoupling the nktionship between

diet quaMy and carbon arsimilation.

Second, concurrent with the acWe breakdown and nbaw of cytoplasmic 'Oscd

and as indiuted by the m r u b of the Iiôekd silt exposums, then is passive mkaw of

cadmium associateâ with the siît surface. Owen (1966) npoited that the pH of the

digestive tract of mursels ir clors to 5.5 and such rcidic environment a n nsuW in the

passiw d ~ ~ t i o n of cadmium from the !surface of inorganic particies. Imp~ftWly, this

derorption w i r independent d d# qurldy suggesting thrt ewn at a wry low organic

content (10%) digestive p r o m that are occumng within the gut of the rnuswl in

ruffcbnt to nmove the rime amount of cadmium from the siît as comprmô to a diet

90 that contains thiw t ims the amount of organic matter. Of fumer note, in the silt-only

exposure (in the absence of organic matter), 'Oscd-A~ was still36%.

3d.2. '@@cd In mumalr in nlitkn to '@@cd-A€

Given that assimilation efficiencies are representative of the amount of

radiotracer essimilateâ by the muswl from its dbt, crlculateâ 'Oecd-A€ and the amount

of radiotracer incorporated into the mussel tissue (correctecl for uptake of the radiolabel

from solution) should be highîy comlateâ with each other. When ' O ' C ~ A E for both

algae and sitt Iabed exprimentr w o n regnrwd with '"cd in the mussels' tissues, a

weak but signifiant positive comlation war found (r = 0.63. p < 0.05). Gnatest tissue

concentrations of '%d comsponded to diet expowres w h m the s i l was labeled

nther than the aîgae. This is an important finding, in that it suggests that metal

desorôed from the inorganic component of seston tends to k more readily

incorponteâ into the animal tissues, and hance more biologically availabk, compared

to metal associated with the aîgal component of the diet.

Sevenl studies (ag., Rule and Alkn 1996, Thomas and Benddl-Young 1998)

have npoiteâ that cadmium concentrations in seston-ingesting organirms are highly

comlateâ with cadmium concentrations aswciated with the easily reâucibk inorganic

fraction of seôiment (i.e., mtals associated with the surface of rnanganese oxiâer).

Assuming that adsorption of the '%d onto sitt repmnnts simiiar mption processes

that occur on the surface of owider of rnanganew, that although mston-ingeüng

organismr n u y k ultimately seMing for the organic corngonent of the diet, the

passive uptake of Cd from the inorganic cornponent rnay ovewhelm the contribution of

9 1 Cd from the organic component of sediment alone. Henœ, in nature, when

concentrations of Cd asoociated with the easily mducible component of a sedimnt a n

conelated to cadmium levels in the a8sociated biota.

Of further note, and as yet unexplained is that the lowest '''cd tissue kvels in

mussels wen observed at the mussels' filtering maximum, whem 1 expected that I

would have obwwed maximum tissue levek (Fig. 3.10). If this value is omitted fmm

the relationship betwemn '''~d in the mussl tissue and ' w ~ d - ~ ~ from the abal Iabeled

diet exposun, there Q the expected positive nlationship behwwn the two variabks.

B a d on these three points. amounts of lmcd incorponted within the mussel tissue at

the fibring maximum shouid have b8m a. 50 dpm gdw-'.

In summary, the pmsont study indicated two distinct patterns in the assimilation

of '-cd depending on whether the 'Oscd had b e n intorpontsd into the algal

component or sorbeâ ont0 the sitt component of the diet. When mussels wen exposed

to a diet whem only algae had h n Iaôekd, '''cd as8iinilation was proportional to Me

organic content of the diet with maximum ' O e ~ d - ~ € occunSng at the filtering maximum

109 for this specks. Howevet, when the silt was Iabeled, Cô-A€s wem independent of

d# quahty, and, with t h exception of the silt-only rxposures were rnaintained at a

constant maximum value of 85%. It ir po8sibk that the addition of aîgae to the sitt

kôebd diet activates digestive processes that n w l in greater assimilation of silt-

bound '%d as compamd to dieh compriwd of ju8t silt abne. Hence, uptake of Cd

from seston will k dependent on both active (i.e.. digestion of ownic matter) and

passive (Le., dewrption fmm the surface of inorganic s i l prrticks) proœrer.

92 Within the natunl envhnmnt. seston is compowd of both inorganic and

organic components. Henw both will k important in providing a route of metal

exposute to seston ingesting organisms. Importantly, the relative importance of the

inorganic component of the seson will k the grnatest under conditions of low quant*

and quaMy of the seston when mussels are ingesting both components of the seston.

Under these conditions, maximum desoorption of the metal h m the inorganic

component of the seston is expectd to occur. ln contnst, under conditions of high

wston quality, whem the musel is capable of highly seîective foeding (i.e., excluding

al1 inorganic cornpanents over organic components). only that metal aswciated with the

organic component of the seston will k availabb for uptake. The pteviour study

(Chapter Il) indicateâ that this sorting maximum occun at a seston quality of - 40%.

Under thew conditions. Cd assimilation should k proportional to the amount of Cd

associateâ with the organic content of the diet with maximum vahies ôeing achieved at

the rnuswlst maximum filtering capability.

Absil MCP, Kroom JJ, Wobrôeek HT (1994) Avaiîability of copper from phytoplankton and water for the bivalve M8coma baItica II: Uptake and elimination from CUM- labeffl diatoms and water. Mar Biol 1 18: 129 - 135.

Bayne BL, Igksias JIP, Hawkins AJS. Navano E, Heral M. ~s lous9aoI i JM (1993) Feeding behavior of the musml Mfldus edulis: msponm to variations in quantw and organic content of the seston. J Mar Biol Ass UK 73:813-829

Bayne BL, Thomson RJ, Widdows J (1976) Physiology I. in Marine mussels: their ecology and physiology. (M. B.L. Bayne). Cambridge Univ. Press. p: 122 - 206.

Borchardt T (1983) Influence of food quanbty on the kineticr of cadmium uptske and los8 via food and mawater in Mylilus dulis. Mar Biol 76:67-76.

Borchardt T (1985) Relationship between carbon and cadmium uptake in Myti/us edulis. Mar Biol85:233-244.

Cosm. O (1989) A miew of the uw of MytiIus tpp. as quantitative indiators of cadmium and mrcury contamination in coastatal watem. Oceanol Acta 12:417- 432

ûecho AW, Luoma SN (1994) Humic and fuhric acids: sink or source in the aviilabitity of metals to the marine bivalves Mawma baîthics and P o t e m u / % amumnsis? Mar Ecol Prog Ser 108: 133-145.

Fegky SR, McDonaM BA, Jacobsen TR (1 992) Short-tenn variation in t h qwntm and quality of seston avaikbk to benthic rur~nsion feeâen. Est Coast Shen Sci 34: 393412.

Fisher NS, Reinfelder JR (1991) Assirniktbn of wbnium in the marine copepod AcaRia tonse studied with radiotracer ratio mthud. Mar €cd Prog Ser 70:157- 164.

Gagnon C, Fisher NS (1997) The bioavailabil~ty of sedimnt-ôound Cd, Co and Ag to the mussel Mytiius etiu/is. Can J Fidr Aqurt Sci 54: 147-1 56

Harrison PJ, Waters RE, Taylor FJR (1Q8O) A broad sprdrum artifiiirl mawuter medium for wastal and open oœm phytopknkton. J Phycol 16:28-35.

Harvey RW, Luomr SN (1985) Soprntion of iolute and parliculate voctors of heavy mta l uptike in conttolbâ su8pension fwding experimnts with M B C W ~ ~ baithka. Hydrobiologia 121 : 07 - 102.

Luoma SN, Johns C, Fisher NS, Steinôerg NA. Oremknd RS, Reinfeider JR (1992) Detemination of wkniurn bioavaitability to a benthic bivalve hom particulate and solute pathwayr. Environ Sci Technol26:485491

Owen G (1966) Digestion Physiology of Mollusca vol. II. Wilbur KM and Young CM (eâs.). Academic Press, New York, NY. pp: 53-96.

Reinfelder JR, Wang WX. Luoma SN, Fisher NS (1997) Assimilation elfciencies and turnover rates of trace elemnts in marine bivalves: a cornparison of oystem, clams and mussela. Mar Bial I 2 9 : U M S .

Ruk JH. Aîôen RW (1996) Cadmium bioavailsbility in thrw estuarine animais in relation to geochemical fractions of seâirnentr. Arch Environ Contam Toxicol l9:878-685.

Stecko JRP, BendelFYoung LI (1999) Uptake of '%d from sedimnts by the bivalves Macoma baithka and Rotothece staninee. Aquat. Toxicol. 1-1 3

Thomonn RV (1981) Equilibrium model of fate of microcontaminants in diverse aquatic food chahs. Can J Fish Aquat Sci 38:280-296.

Thomrnan RV, Mahony JO, Muelkr R (1995) Steady-state model of biota sediment accumulation factor for metals in two marine bivalves. Env Toxicol Chem 14:1811-1812.

Thomas CA, ûendelCYoung L (1 998) Linking the geochemistry of intertidal mgion to metal avaikbildy in Macoma bailhica. Mar Ecol Prog Ser

Wang W-X, Fisher NS (1996a). Assimilation of traw ekmnts by the muswl My?iIus edulis: effect8 of diatom chernical composition. Mar Biol 125:715-724

Wang W-X. Fisher NS (19Wb). Amimiktion of trace ekmnts and carbon by the mussel Myfilus eûuIis:eflects of food composition. Limnol Oceanog 4 1 : 197-207.

Wang W-X, Fisher NSl Luoma SN (1995) Assimilation of tnce ekmnts ingesteâ by the mussel Mylilus edulis:effects of algal food abundance. Mar Biol Eco1 Prog Sur 129: 165-1 76.

Waid JEl Levinton JS. Shurnway SE, Cucci T (1997) Site of particle wkcüon in a bivalve mollusc. Nature 390: 131 -1 32.

Wilkinson L, Hill MA, WJna JP, Biikenbeuel GK (1992) Syrtat for Windows: Statisticr. Version 5 Edition. Evanston, IL. Syrtrt Inc., 750 p

Zang GH., Hu MH. H u ~ g YP, Hamron PJ (1990) Se uptake and accumuktion in marine phytopîankton and tnnrîer of Se to the clam (Puditapes phIippnanm). Mar Env Res 30: f 79 - 190.

MEtA8OLK: COS1 OF FEEDIW BY THE BLUE MUSSEL (Mfl Iw tmsdua) DISPîAYINO SELECTNE AND NOM4ELECTNE FEEDIWO BEHAVtûR

Abstmct

The metabolic cost associated with pumping water and sorting food psrtickr by

the bkir murwl, MyWs tmssuIus, perfonning 8ekctive and non-selective fwding

ôehavion was deteminecl under laboratory conditions. The mtabolic cost arsociated

with feeding war mea8ursd from the nrpintion rate during a 1 to 2 h priod of seston

exposure. Seston matrices were prepareâ to produce: 1) non-sekctive feeding tactic

when muswls wem expowâ to 0150 and 15010 seston matrices (x106 ceIli L" 1 mg rilt

L"); and 2) sekctive feeding tadic when mussels were exposed to 20120 and 150150

seston matrices. In a control tmatmnt. m u ~ l s wem expowd to only filteml (0.45pm)

seawater to detemine standard respiration. There was no signifiant difference

between the metabolic costs associateâ wini sorting during sekctive or non-sehWive

feeding (Studrnt's t-test, p = 0.182). Moreover, the metabolic cost asociated with

feeding for a 0.20 g dry weight musml was a maximum of 0.44 J h" and compriseû

only 0.9% of the gros energy intake. Thenfore, I conclude that there is negligibb cort

auociated with wlective fwding .

The blue muss8l8 (MyUus spp.) am one of the important groups of filr-feeding

bivalves that inhabit estuarine ecoryrternr chancterized by a mlattvely high and

variabk abundance of surpnded particulate matter (SPM). Concentrations of

wspendd particks typically nnge from 4.5 mg C' to 32.0 mg 1" and am primarily (70

- 80%) cornposed of inorganic matter (Stecko 1997). Living under such conditions,

fiîter-feeding muiseh obtain their food by purnping water from which they promu and

mtain ruspended partiiles. Partides with high nutiitional value am ingeiteâ and thon

of low nutritional value an repctd n pseudofaews (see Chapter II).

Cumntty, them a n two opporing vkws as to whether the feding procesr in

fiîtei-fding mussels ir energetically costly (e.g., Bayne et al. 1988, 1903, Willows,

1992), or is mlativeîy inexpensive, with the energy swnt on water processing

repmmnting only a maN fraction of the musnh' total metabalhm (8.g. Jargensen

1990). Bayne et al. (1988, 1993) ruggert that the metabolic rate va* with the

functionrl state of the organirm, with oxygen conrumption isfkcting energetic costs

that a n k physiobgically mgulated for water pmcaming and fwding. For eximpb.

8ayne and New011 (1983) note that r~tiw hrding wa8 auocirted with a 2- to 3-lold

incmaw in oxygen comumption when compamâ with standard metabdiam (a

98

quiescent rnusml). The- authon wncludeâ that the overall expenses for water and

food procering anse fmm the cost of watet transport. filtration, digestion and

absorption.

On the other hand. Jergensen (1990) suggests that the co8t of wabr processing

in filr-feeding muswls is virtually non-existent. He concludes that the low rate of

oxygen conrumption that p m i l s in muswlr with reduceû valve g a p in particle-

depkted water dws not mfkct a physiokgical regulation, but rathet is a consequena

of an incnaring diffusional rssistanœ with a decteasing (kw thmugh th8 mantle cavity.

A non-existent feeding cost. a8 8uggestd by Jergenwn (197S), is based on the

hypothesis that fiRe? feeâing evobed according to the pmcipb of a pump systm which

allows foi continuou8 feeding at low rotes, rathet Man discontinuous feeding at

conespondingly high rates. Studiss that support thir concept indude Clausen and

Riisglrd (1996) who found that the energy cost of growth for M. edulis was 12% of

amimilated food. RiisgOrd and Larsen (1995) also note mat the cost of active filration

may account for only 0.3 - 4.0% of the total mtabolic expenditum.

The different points of view of fwâing costs of Rîtet-feding mussels may in part

k nlated to dinennt food mime8 in the exp«imrnts. Clauwn and Riisghrd (1996)

empbyeâ a seston concentrath of betwoen 0.2 and 6.7 mg L" (aruming a dry

weight of Rhodomonas b&ka ca. 81 pg per IO@ cella) with a low otganic content. i.e.,

3. 18, 25% organic content for seston mina with silt and m. 60% for algae. In

contnst. Bayne et al. (1993) used a seston concentration of b ( m r n 0.99 and 10.3 mg

L" with iow to high organk content. i.e., 41, 64 and 74% organic content. Nimw

mngm of mrton mirices, boîh in quanûty and qurüty, that w w etnpkyed by Bayne

99

et ai. (1993) and Clauwn and Riisghrd (1 996) may not be quite enough to observe the

wide nnge and the ar8ociated potential costr mat can ocair under natural condition$.

My pmious study (we Chapter II, Fig. 2.5) showd that when muswh wen

exposeci to a wide nnge of seston concentrationr (from 1.4 to 56.6 mg c') and organic

contents (frorn 10 to 71% omanic matter, equivaknt to 0.12 to 1.63 mg C L"), two

dinesnt feeding stntegies wem ernpkyed. Fimt, with high food abundunc8,

mgardkss of seston quality, muswb sorted thJi diet, choosing organiwlly fich

particka over inorganic ones, and consequently wntilated at low nter. Second, when

mursels wen exposed to low food concentrations and seston quality, they srhowed

non-mkctive f d i n g and ventilateâ at a maximum rate. The ability of rnus8eh to

switch between these Mo f d i n g behavion demonrtratsd that the feeding procesr

was dependent on food conditions.

These two feeding strategies provided the impetur for further study to test the

two diflering points of vW prewnted by Bayne et al. (1993) and Jergenwn (1 990) on

feeding costr. Theoreticalîy (Fig. 4 4 , muswb exposed to a low seston concentration

and quality (stagd) would not k tequird to sort their diet ôecruw they would

maximûe their ckrnnœ rate and fitter all paitickr avaikble for ingestion. Simikrty,

when thers ir h i ~ h serton quality, primarily c o m p o ~ of rbae (stage C), murwls

wouîâ not neeâ to wrt their diet. In conhst, under a high wnœntntiin and a low

quaMy, e.g. 40% organic matter (stage 0). musmk wouîd employ their rorting pmceu

and con8eqwntty would incream their mtabolic ar t of feeding. Henm, under the

FQ. 4.1. Schernrtic diagnm of the changes in feeâing behavior of rnuselt in nrponse to a broad range of seston matrices. SC2 = seston quality, SEF = sorîing efficbncy, COS = hypotheticrl co8t of 8oflin~. stage A = food condition when [food] [silt] and no sorling, stage B = food condition when [food] ; [sin] and sorting, stage C = food condition when m08tly algar and no sorting, L = low, H = high.

H

A

COS

L

101

latter condition, I wouîâ expect the highest mtabolic cort as8ociated with wkcaw

feeding in the blue musml. As the watr of feading invohm two rtage of activitiis

(pumping of water and sorting of food), the cos of sorting w i r detemined by

subtnding the cost of pumping (standard motaboliun) in control tmatment fmm thow

in expr imta l tmatrnent.

The objective of thir study war to detemine if thrre is a coit associateâ with

wkctm foeding. The cost of wfting was deterrnimd by chalbnging the rnumeb with

wrton matricer Mt elicited either wbdve or non-aebdve f d i n g and maruring the

rnusml's respiration rats duhg feding. To test the hypothmir that them was a cost

aswciated with wîective feeâing. an rnalyoi8 of variance war conductecl on respiration

mtes and ammonium excmtion mtes. Further, a rimpk comlation was ca- out to

test if them was a poolive wmiation ô e n 1) m8pimtion and ckrnnce rater, and

2) excmtion rate and cribon nrimiktion efficiency.

4.2. MATERU8 AND METHOOS

4.2.1. Cakrlc conbnt of the dimdom

The diatom, ThaIasSiosim p~eudonena (Stnin No. 8709) was grown in natural

seawater enricheâ with Hamson W i u m (set? Chapter II). Five batches of algal

culunr (3.0 L pet batch) were hawested by centrifugation for 10 min at 3000 rpm. and

w a r M with dirtilkd water bdom a second centrifugation. The abal ceIl8 were then

put in pomlain cnicbbr and d M in a 60°C oven for 96 h. Each brtch gave an algal

yied of approximately 200 mg dry weigM.

The caloric value of the diatom war deteminoâ uring an oxygen micro-bomb

calonmeter (Phillipron 1964). After several pmliminaiy detemiinationr, a 15 to 20 mg

pill wmpowâ of a mixture of 50% diatomr and 50% bMoic acid war chom to obtrin

a sufficiont temperature ri80 and wnwtent n r u b . Pilh wem pmpareû by thoioughly

mixing algae and bMo ic aciâ powder in 15 mL g l i u vial. Pilh wem then d M for 24

h in an oven at 60°C. The m m rmount of benzoic rcid onty wu8 umd for the

calibntion of the micro-bomb cakrimter. T h m nplicrtec of each batch wem

pnpared for calorie marummnts. The calant content of the aîgw war then

cakulateâ a8 the total calodes pmducd minus the calories producad by the knzoic

rcid in the pilk.

4.2.2.8mton tnrûmnta

Batrd on rny prwious study (Chipter II, Tibk 2.3), experimntal tnatmenta

mpmmting control, sabcth and non-wkthm 1Wing bhavion wen wîecteâ

103

(Tabk 4.1). All experimnts wem mplicated 5 t ims with thm groups of musmls per

trial.

The mussels were rxposed to combinations of O, 20 and 1 50 x l o6 abal œllt L"

and to the following silt loads: control (O mg L"), intermediate (20 mg L") and high (50

mg L"). The four mrton tnatmnts mpmmnted the food conditions as follows: fimt,

no algae and 50 mg L" silt kad; the mussels wen exposed to an extremo SM stress

wiai oniy 2% seston quality (SQ) and non-mkctive feeding was expected. Second,

150 x 10' celh L" and no silt load; the muswh were expowd to a high abundan- of

food to mimic a condition during a phytoplankton bloom with SQ = 60% and non-

wkctive f d i n g w8s again expected. Third, 150 x loe wlb L" and 50 mg L-' SP load;

the mussels wem exposecl to a wston matrix combining a high quant@ of food with a

high siit load, with a tesuling seston quality of 2496, which should resuît in selective

feeding. Fourth, 20 x 10' celh L" and 20 mg rilt L"; a wtton mahix with SQ = 21%

when the mumeh exhibited a maximum cbannw rate and km-wkctive feeding. In

the Iast troattmsn, the mussels wem exposai to onty 0.45 um fikeisd mawater, which

was a contrd tmaûmnt with no dgw or M, a W n g me to ma8un standard

mtaboli8m.

Muswh (MyWus ~~OSSUIUS) wem colbcteâ from inteWaI ana8 of Homerhoe

Bay, British Columbia, Canada during the fa1 pcior to the expenmnt. The colîecteâ

muswh were acdimated to the diatom T. pseuûmana for two weoks, and d e rupply

of semater war miintrind a1 13 I 1% and 28 ppt alinity. The muuds wem rorted

Tabk 4.1. Sekcted seston matriœs and feeding panmeten from a pmvious study (Chapter II). Abae : sitt = x 106 cells L" : mg L". CR = ckannce rate (man8 I 1 SE. n = 3), SQ = seston quality. W = diet quaMy (average valuer), app. C- A€% = rppamnt carbon assimilation efficiency, tnie CA€% = tnie carbon assirnilati~ effickncy (average vakre~). nd = no data.

Seston

Algae SM

Non-wiective Ç.

O 50

150 O

Selective F.

150 50

20 20

Control

O O

for the sam sue (44.5 mm shell kngth (SL). 0.16 gdw), and the shells wen cleand of

fouling organbms by rcnping. Pnor to the expnments, rnuswls wem fsd on wston

$0 that they would be acclimated the seston tnatmnts.

The total mtabolic cost is composed of khavionl and physiokgical costs.

Ammonium excmtion rate war used as an indicator of the physiological cost (Bayne

and Newell 1983). whereas the behavionl cost associatecl with the Mlration and softing

of food was estimatecl brrsd on the respiration rate (Widdows 1985). The respiration

rate was messurd continuously during the expefiments. and the excmtion nte was

determined at the end of the experimnts.

The respiration and the ammonium excmtion ntes wem detemined in a 500 mL

closed-chamber, temperatum-controtkd respirometen (Fig. 4.2). Airaaturated

wmater was addeâ to the respirometer chamkn and stined by man8 of a rnagnetic

stinar kneath a perforatecl Plsxigksw plate supporthg the mursels. Seston matrices

wen œntrifugd and bubbîed with purs nlrogen ga8. then injucted into the

expnmental chrmbn every 15 min. With this technique. a roughly constant food

supply was maintaind throughout the masummnts. The rate of dedine in oxygen

partial prerurs war masund uring a polirognphic oxygen rbctrode connected to

an oxygen mter (OM2W type. Camron Inatmment Co.) and a chaR recorder.

Mussels were aIlowed to open their v a b s for 10 min to msum filtedng, and oxygen

(O2) conrumption was meawmd for onr to two houn. 0xyg.n rolubility tables won

used to conveit values of oxygen partial pressure (PO p ) to oxygen concentrations in

mL O2 L", comcting for ternpntun, salinity and baromtric pm8sum during the

expriment. Oxygen wnsurnption wrs not masuureâ bslow an oxygen partial pnrrure

of ca. 100 mm Hg, as th8 rate of O2 uptake by the murwls becoms dependent on

extemd p02 et bwer oxygen tensions (Widdowr 1985). The respiration rate (Vd2) WB$

calculated as

when Cc was the oxygen concentration in control chamber wlhout mussek, was

oxygen concentration in tibatmnt chamkr at the end of rwiwnrnent; Vr was the

volum of the mspiromster (L); Vm war the volume of the mu8sels (L), and to and t1

wem the start and finish timr (h) of the measunment period.

T h m mplicate water sarnpks (10 mL per mmpb) wem collecteci from the

m8pintion charnôen, and from the chamber without mu8setls, and anaîyzed foi

ammonium concentrations using the phenoî-hypochlorite mthod (Solbcuno 1969).

The ammonium excmtion rate wa8 detemind by subtracting the ammonium

conc8ntrations in the control chamber from thom in the tmatment chamber.

4.2.8. Cakulitkn of uop, for gmwth

Measumment of physiological traits, wch as rater of ingestion (IR), aibon-

auimiîation e f f ï n c y (GA€), reqMhtion rate (RR) and excmtion rate (ER), and thdr

inteQmtiin by means of phyriobgical rneqetics, can provide insight into aie gmwai

1 O8 proass and how it rnight k disrupted by environmental stnss and pollution (Bayne

and Newell, 1983). The rrt8arunmnt of the energy avaikbk for growth, t e m d sape

for growth (SFG), provides a npid and quantitative ar~8sment of the energy status of

the muswh. Acardingly, SFG or net rnergy intaûe rate (€4 for a mussetl can be

expmsed as:

SFGor E- =€--€a,',

= (IR x A€) - (RR + ER)

whrm the a b w M energy (food), E-, is the product of t h ingestion rate (IR) end

the assimilation eflciency of the food (C-AE%). The energy expenditure. Eccu. is the

sum of the total respiration, RR (maintenance respiration and respiratory cost

associated with sorting) and the energy lost a8 exmta. ER. 00th expmrsed as rates.

Each component of SFG was ulculated as follows:

IR = (cksnnce rate (L h") x paiticulate ofganic matter (mg L")) - pseudofaeces

production (mg If1) x 12.52 J mg" (the energy value for 1 mg algal cells of T.

pseudonena).

CAES wem pmdicted b a W on a musml's feeding behavior in rerponse to itr

food condition (i.e., sekcave f d i n g , true C-A€, or non-seledive f d i n g , apparent C-

A€). The appamnt CAE%r wem cakulateâ ar carbon content in the ts8ton matrix

minus arbon content in faecus, i.e., rpp. CA€% = ((C- - C -)ICIUDn} x 100.

When the muurls diaplayed wWvs fmding behavior, true CAE%s wem srtirnatd

bawâ on the carbon content of the d# which nie musml was acturlly ingesting (i.e.,

accounting for 8ebcüon of the organic wmr the irorganic components of the wston)

minus the carbon content in faecer. Le.. tme CAEW = ((C ,,,a* - C m -)lC m 4 x 100

(see Chapter II, eq. 2.1 1 and 2.12).

The respiration rate (RR) and excmtiin rate (ER) values wen conveited to

energy unit8 wing Me factors, 1 rnL O2 = 20.33 J (Widdows and Johnson, W88) and 1

mg Nil4-N = 24.87 J (Elliot and Davison. 1975), mspectively.

4.2.6. Dlti rnr)yr&

After the cornpletion of each expdmnt, aie muuds wwe d W in a 80°C oven

for 96 h. Dry masr was deteminal to the nerrest millignm with a Cahn

ekdrobalance. The data on the cakric content of ais aigw wem anaîyzed for each

culture batch harverteâ during exponential and seneccent gtowth phases uring a one-

way ANOVA. T h data for oxygen and ammonium wem crlculated a8 mau-specific

rater, i.0. ml. O2 h" gdw" and pg N K - N h" g dw". Stati8tiul cornparisons ôetween

the tmatmentr wem canibd out by a one-way ANOVA followeâ by Tukey multipb

comprriwns (Zar 1984). Signi&rnce of a l tests wrs accuptd at p c 0.05. All

statirthl an i î ym w m done with SYSTAT 5.0 (Wilkinson et al. 1992)

4.3.1. C 8 k e contant of dirtom

The calonc content of the diatom iangd from 5.70 to 7.37 J mg" at the

senescent phase, whenas at the exponential phaw the cakric content of the atgae

almost doubled and rangeâ from 10.84 to 13.59 J mg". Them was no significant

diffeice in the caloric content between algal batches at exponential phaw, but then,

was a signifiant difference between the csloric content of the algae et the exponential

and senescent phase (one-way ANOVA. p < 0.05, Fig. 4.3). Therefore, throughout the

expriment, algae were harvesteâ at the exponential growth stage.

4.3.2. Elhct of nston nrbke8 on mplntkn

During active feeding (150i50 wston matrix), a respiration rate of mussels war

doubk that at the quiescent condition (OiO), but then wen no signifiant differenws in

nspiiation rates among mussels exposeci to seston matricer at meâiurn (20120) and

high (1 SOISO, 15010) food concentrations (Fig. 4.4, one-way ANOVA, p c 0.05).

Arong the experimental treatrnents. the standard respiration rate under the

control treatment ((010). 0.54 mL 01 h" gdw-') was not significrntîy duersnt h m that of

the high SN bad treatment ((0/50), 0.41 mL O2 h" gdw"). The maximum 0 2

conrumption rate wa8 1 .O2 m l 01 h-' gdwv' at a m8ton matri% of 15060 (SQ = 24%),

doubk the conttd rate and this inmase in mtaôolic cort was rignircant (Fig. 4.4).

However, combining rH active feeding tnrtmntr (20R0, 150150, 15010), the average

incnase of metabolic cost abow strndarâ metaboliun was only 0.28 mL 4 h" gdw".

Fig. 4.3. Calorie content of the diatom fhaIaSSjSsni pseudonana. A, B. C = aigal cultures harvesteâ dunng the exponential growth pham; 0, E = algal cuituns hawesteâ during the senescunt gmwth phase. (means I 1 SE, n = 3). Signifiant (p < 0.05, one-way ANOVA and Tuby mulapk range test) diffemnœs a n denoted by diffemnt ktten above ban.

Algal cuituns

Fig. 4.4. Respiration rates of MyCi/us tmssuIus in n s o n w to diffemnt maton matrices ( m a n i 1 1 SE. n = 5). Signficant (p < 0.05, Tukey's mulipk range test) diffemnces an denoted by diffemt letters sbove bon; Seston qualily (SQ) = 0% mpnmnts expefifnent without seston.

Seston matrices

equivaknt to 5.63 J h" gdw*'. The ovenll average value for the active mtabolic cost

assoctated with feeding was not significantly diffemnt from that of the standard

metaôolic cost (StudenYs 1-test, p > 0.05). FurMernom, cornparhg non-sekctive and

seledive fwding khavior treatmnts (i.0.. 0150, 1 SOI0 vs. 20120, 1 50/5O), although

there wae an inctease of 5.9 J h" gdw" in the seledive feeding tmatments, the

mtabolic cost (soorünng process) of selective feeding was not significantly greater than

that of non-sekctive feeding (Student's t-test, p > 0.05).

Oxygen uptake was independent of the ckaranw rate (ventilation detennined

by prrticîe depîetion) (Fig. 4.5). At a 20120 seston matrix, despite the ckamnw rate

k ing maximal, the respiration rate nmaineâ constant compareâ to the control

treatrnents (i.e., standard mtabolism).

4.3.3. Effbct of amton mMcm on ammonium rnxcntkn ntm8

The m a n excmtion rate for non-fittecing animais was 25.42 pg NH&l h" gdw"

and for active animais was between 26.28 and 46.12 vg N H A h" gdw". The

ammonium excmtion rate for mussels exposed to the high-abw and high-riît-lord

(150150) treatmnt war not signircantly diffemt from that of mus8eb exposeci to the

high-able only (1 Solo) (Fig. 4.6). The maximum exwtion ntes under 1 Sot50 seston

matrices irnplieû that the musml8 ingesteci a rimikr amount of the orgsnic component

of seston w the murseh~ expowd to the high abal concentration. Analysis of variance,

followed by a Tukey muîtipk range test, showeâ that the increaw in the rate of

exmion between the standard (non-fiîtering) and active metaboluing musse18 wrs not

signilicant at p <0.05, exœpt for a seston matfin of 150/50 (SQ= 24%).

Fig. 4.5. Relaüon8hip between oxygen conrumption and ckannce rate (CR). CR -

values wem taken fmm pmvioÜs exprin#nts (Chrpter II). In control tmatments (010). the value of CR (Le., numkr of parlicks c k a d p r litet per hour per g dry weight) war below the limit of detection of the instrument. (mans î 1 SE. n

Clearance rate (L h" gdw")

Fig. 4.6. Ammonium encrstion of MyWus tl~ssu/us in nrponw to ciillemnt seston matricer (means î 1 SE, n = 5). Signifiant (p < 0.05, Tukey'r muîtipk range test) diffenncor ais denoted by difiennt ktters above ban

Seston matrices

The ammonium excmtion rate was propoctional to the carbon assimilation

efficiency (r = 0.65, n = 5. Fig. 4.7). When etxposeâ to the 20120 serton matri%. the

mursels showeâ active feeding with maximum clearance rates and minimum

ammonium excmtion rater (i.e., the excmtion rate only increarrd 0.85 vg NHrN h"

gdw-' (0.02 J h" gdw" ) above the contiol expriment). The average of the ammonium

excretion rates, which npresents an energy 1088 from the unanimilated protein, was

lem than 0.1 J h" (Table 4.2). Themfom, the contribution of the ammonium rxcretion

rate to the total mtabolic cost war insignifiant, although the magnitude of its

contribution was proportional to respiration rate (r = 0.94, n = 5, Fig. 4.8).

4.3.4. Scopo for giowth

When the rates of respiration and excmtion wem balanced againrt the energy

intake b a W on the apparent and tnie CAEs h m my pmvious experimentr (Tabîe

4.1), the scope for growth (SFG) of a mussel nmained positive, exœpt undrr

tmatmnts with a high sitt lord and no abae (0150) vabte 4.2). The SFG for 0.16 gdw

murrd was lower when ul#ilationr wem baseâ on the apparent C A € nther than the

true C-AE. For exampie, wlh the seston mat& of 20120 whem non-seledive feeûing

was auumd, the SFG was only 20 J h", but when the musmis wem sebctively

hg-teâ their dia (i.e, the SFG estimatd bsmd on true CA€, Tabîe 4.2), the SFG

war 266 J h" .

Fig. 4.7. Relationship between ammonium excretion rate (pg NHcN h" gdw*') and true carbon aasiimilation efficiency (tnie CA€%). Values = m a n I 1 SE, n = 5, r = 0.65

" y - % $ $

Fig. 4.8. Relotionship between ammonium excntion (J h" gdw") and respiration rate (J h" gdw"). Values = m a n î 1 SE, n = 5, r = 0.94.

O 4 8 12 16 20 24

Respiration rate (J h" gdw")

The total metabolic cosb repmwnt both khavioral, such as the beating of the

gill cilia for the movernent of water and the filtration of paiacles in the mantle cavity. and

the physiologiurl cosh, ie.. üie irsimibtion of food. The cost of food acquisition

meawteâ as nspimtion rate ha$ been nported previourly with confliding resub. No

effect of nation kvel on oxygen uptake couid k found in marine bivalves Spisula

subtmmta (Mehknkrg 6 Kirrboe, 1981) or freshwater mus8elr. HymhlIle menzie

(Roper and Hickey 1995). By contrast, GrilMhs and King (1979) working on ribôed

m u s ~ l s (Aulawmya etet) and Gaffney and Diehl (1988) on Mytilus edulis noteci a

progmmive incidrw in the mtabolic rate associated wiai incmssed feeding sctivity.

GMiths and Griffiis (1987) stateâ that peak rates of oxygen conrumption couM be as

much as 4 - 5 times the qubsœnt kvel.

My study showbd that the respiration rate incmased wlh an increasing food

concentration. When murwls ernployed a sekctive feeding tactic, Le., undw a high

algrl and a high silt bad (150150), the mrpintion rate asmciuted with their feeding war

double that at the quiescent kvel (010) (Fig. 4.4). However, this rate was not

signfkantly diffennt among adiw f d i n g musseh (20î20, 150150 and 15010 seston

matrices). The gmate8t Gort of acquinng and rorling food, for muswls exposai to high

algal und high rilt lord, war 1 59 J h" mur se^' or 0.44 mW (1 mW = 3.6 J h") which

was 0.0% of the g m ~ energy intaka. Ei-. Momver, under food conditions of a

20120 serton mattix similet to the condition whem the rnumels wem colîectd in the

Md. the cost of Mnging and rorting food was onîy 0.2% of ruimilited food. or 22% of

12 l the total metabolic cost. Thersfore, my study shows that the cost of ciliary activity in

acquiring and sorting food particbs in M. trossulus ir genenlly low.

My resuh, however, differ from other worken. Using 35 mm SL Mytilus edulis

(0.21 gdw), Jergensen et al. (1986) calculated the water transport across the gills

baseci on the diffemnœ of water pressure in inhalant and exhalant cumnt. and found

that the pumping power w8s 14 pW. This value was about 1.8% of the total metabolic

rate of the mussel. Widdows and Hawkins (1989), brsed on direct calorimetry,

eaürnatad the mechanical cost of food acquisition for a 11 mm SL muswl(10 mg dw )

at 0.4 pW. which was the limit of detedion for their technique. This value war < 3% of

the rate of heat dissipation by the mumet. In contnst, Bayne and Scullad (1977)

estimitd that üte M a n h l wst of feeding in a musmi (1.33 gdw) was 3.56 mW

which was 47% of the as8imilated ration.

The cost of ci&y activitier and sorting of food in Jorgenrsn et al.'r (1986) and

Widdom and Hawkins's (1989) experimnts d# not mfbct the dynamic nature (e.g.,

compositions and concentrations) of -ton in a natunl environment. Their

exparlmnts were conducted by exposing mu88eIr to pure abal particles below

pouibk pseudofaecer production. On the other hand, in Bayne and Scullard'r (1 977)

studios, the high estimate could k d w to thdi experimntal murwls being kept

without food for mon than 19 days. Thus, in their expetimnts, a starvation factor may

have confounded the rner8ummnt of the mtabolic cost of filtration,

A numbet of studies of filter-fding bivaîves (se@ mview8 in Bayns and Newell

1983, and G M i s and G M i s 1987) rugget t h t oxygen conwmption is highly

122 comlated with the filtration rate during feeding, i.r., the respiration nte (mtabolic cost

of filtntion) incnases logarinimkally with incmaring filtration ntes. My m u l h are

contrary to these pmviour studios in that then war no evidence of a signifiant

increase in respiration rate during active f d i ng . Moreover, under optimal wrton

matrices (20120). a maximum ckannce rate war in fact accompanied by a nlatively

constant tespiration rate and a low excmtion rate. This confimis previous obsurvationr

of a douMing in the cbarance rate of rock-pool bivalves (Vene~pis comg8tus) without

a signifiant change in rates of respiration and ammonium exwtion in msponw to

increadng natunl particb concentrations and siît losd (Stenton-Dozy and Brown,

1994). This phenomenon implies that in natural environment, Riter-feeding mussels

respond by fully oprning their valves, cauring a signiliant increase in ingestd

paiticks (inorganic andior organic particbs) whik maintoining a constant mtabolic

cost of feeding.

In estuarine syatem whem the inotganic wmponent genenlly dominates the

wrton. the blue mu8sel ir genenlly kss wkctive in it8 choice of food paiticks. The

advantige of ôeing kss wktive ir that the muaml will mrintain a conrtant ingestion

rate of food wlh minimal sxpnw. Under condlkn dmilar to natunl environment. e.g.

s mston matfin (20/20), the wrt of rorting by the muasel Myflilus does not rignircantly

change the rate of respiration which onîy incnrsed 0.54 J h" (mm thit of the $tandard

metaboliun. This obwwation is wpporteâ by the fmîâ study of Stenton-Dozey and

Brown (1 994) who al80 found mat under episodk Mal cycbr (wch as duMg high tiàe)

maulting in a high quantity and quality of wspended prrtickr, bivalves do sort their d Y

and the cort of roiaig ir mîatWy bw.

123 When the energy intake ftorn ingestecl food is balanced against the total

metabolic cost of feding, the pmdicted value of rcop for growth of a 0.16 gdw mussel

ir 266 J h-'. This pmdicted value compares favonbly to the fmId values of 236 - 273 J

h" musse^' deteminecl by Taylor et al. (1992) for musml8 niW 2 m ôdow the

surtace.

In summaiy. my findings indicate that the mtabolic costs associatecl with a

seledive feeding behavior accounts for a maximum of 0.9% of the assimibted enetgy.

Under an optimal feeding condition (0.g. a 20/20 seston matrix). the cost war

significantly reduad to 0.2% of the assimiktd nation. or only 22% of the mtabolic

cost. Themfon, my rwub ruggeat that any pib- ingestion sekcaw pmwmes nkteâ

to soiting andlor mjecting unwanted seston components is done with a negligibls

energy expemditun by the Mue mussel.

4.5. REFERENCES

Bayne BL, Newell RC (1 983) Physiological energetics of manne molluscs, p. 404-51 5. In: Saleuddin AMS and Wilbur KM (ed.). The Molluscia vol. 4. Academic Press Inc, New York.

Bayne BL. Scullard C. (1977) An apparent specific dynamic action in Mytilus edulis L. J Mar Biol Asir UK 57:371-378

Bayne BL, Hawkins AIS. Navam E (1988) Feeding and digestion in sus~nsion feeding bivalve molluscs: The relevance of physiological cornpansationr. Am 2001 28: 147-1 59

Bayne BL, Igksias. JIP, Hawkins AIS. Navano E, Hecal M, Deslous-Paoli JM (1993) F d i n g behavior of the mussel, MytiSus eûulis: msponses to variations in quantity and organic content of the seston. J Mar Biol Ass UK 73: 813-829

Clausen Ib, Riisglrd RU (1996) Growth, filtration and respiration in the musml Myfilus edulis: no evdence for physiokgical regulation of the filtet-pump to nutritional needr. Mar Ewl Prog Ser 141:37 - 45.

Elliot JM, Davison U (1975) Energy quivalent of oxygen consumption in animal enorgetics ûecologia 19: 195-201 .

Gaffney PM. Diehl WJ (1 986) Gtowth, condition. specific dynamic action of the mussel MyWs edulis recovecing from starvation. Mar. BU. 93: 401 40%

Griffihs CL. King JA (1979) S o m nlationships between s in , food availability and enegy baiance in the r i b M murwl Aulecomya atw. Mar Biol51: 141 -149.

GMihs CL. Gnffihs RJ (1987) Bivalvia, p. 1-88. In Pandian TJ AND Vernberg FJ [d.] Animal energstics Bivahh through Reptilia. Academic Pm$, Inc. California.

Jergenwn, CB (1975) Comparative physiology of suspension feedin~. Ann. Rev. Ph ysiol. 37: 57-79

Jergenwn, CB (1 990) Bivalve filter feetâing : hydrodynamics, bioenergeticr. physiology and emlogy. Oisen and Olsen, Fredrnsborg. ûenmaik.

Jergenwn CB, Pet h m m . Knutenrn HS, Lafsen PS, Mehienôerg F, Riisgbrd HU (1 986) The bivaîve pump. Mar €col Piog Set 34:69-77.

Mehknôerg FI Ki@- 1 (1Q8l) Growth and energetim in Spisule subîmnceta (Da Costa) and the effect of tuspendd bottom. Ophelia 20: 79-90.

Phillipron J (1 964) A miniature bomb cabrimeter for 8mall biological rrmpkr. Oikos 15: 130 -139

Riisglrd HU. Lamn PS (1995) Fiiter-feeding in marine macro-invertebrater: pump characteristic8, modding and energy cost. Biol Rev iO:67-106

Roper OS. Hickey CW (1995) Effect of food and dlt on filtration. respiration and condition of the fmhwater mussel HyrrdeIIa mentiesi (Unionacea: Hydidw): implication for bioaccumulation. Hydrobklogia 31 2: 1 7-25.

Sol6rzano L (19û9) ûetemination of ammonium in natunl waters by the phenok

Stecko P (1997) ContWng the geochemistry of su8pend.d parüculate matter and drposited Wiments of the Fnur Riwr etuay: implications for metal expowm and uptake in estuafine depoul and fibt feedem. M.Sc. thesis, Dept. of Biobgiwl Sciences, Simon Frawr University.

Stenton-Dozey JMEl Brown CA 1994. Short-tenn change8 in the energy balance of Venenrpis comrgetus (Bivaîvia) in relation to tidal avrikôility of natunl su8pended partickr. Mar Ecol ?mg Ser 103:57-64

Taylor BE, Jambon O, Camfoot TH (1992) Mussel culun in British Columbia: Me influence of w l m n fam on growth of Mytilus edulis. Aquaculure 1 O8 :SI 66.

Wiâdows J (1985) Phyriobgicrl procedursr, pl 161-178. In Bryne BL, Brown DA. Burns K, Dixon DR, lvanovici A, Livingstone DR. Lowe DM, Moore MN, Stebbing ARDI Widdom J. [edr.] The effects of rtnr and pollution on marine animah. Praqpt Publisher.

Wiâdowr JI Johmon D (1988) Phyriological enorgetics of MylYus eûulis: sape for growth. Mir Eco1 Pmg Sei 46: 1 1 3-121.

Widdowr JI Hawkinr AIS (1989) Paiationing of rate of heat diuipation by Myt#us eûulis into maintenance, feeding and growth. Phyriol Zoo 62:764-784

Wiidom J, Donkin P (1992) Mumek and enlonmntal contaminantr: biorccwmuîation and phy8idogicrl arprdr, p. 383424. ln E. Goating [dl, The mussel Myakrs: ecokgy, phyriology, geneth and cullum. Elsevier, Amsterdam:

Wilkinson LI Hill MA, Welna JPl Birkenkuel GK (1992) Syrtrt for Window: Strtistics, Version 5 Edition. Evamton, IL. Syrtit lnc., 750 p.

126 Wilbws RI (1992) Optimal digestive inwstmnt: a mode1 for filter fWem rxperiencing

variable dietr. Limnol Oceanog 37:829-847

Zar JH (1984) Biontatistical analysir. Second eâition. Pnntice Hal Inc., Engkwood Cliffs. New Jersey.

Bioaccumulation of Cd in mu8sel tissues is the resul of a numôer of piocesses

including uptrke, transformation, stofage and dimination. This chapter pmsmts the

application of a kinetic mode1 of mta l accumulrtion to pndict Cd concentration in

mussel tissues. A unique aspect of the mode1 is the incorporation of rnusml fading

behavior. B a d on input pamtwten frorn pmvious studies (Chapter 11, III), Cd

bioaccumulation in mursel tissues (Cd-) was botter pmdicteâ when wkcüve f d i n g

was included as a process within the kinetic-baW model. When non-sekctive feeâing

behavior was consiâemâ, the Cd- was two times highrr than observud Cd values.

Cd- was sensitive! to change in seston quality (SU), mussel clearance rate (CR), Cd

dimination rate constant (k) and mussel growth rate (9). The model also showed mat

dietary Cd uptake contributed 35 to 94% of the Cd concentration in mussel tissues.

5.1. INTRODUCTION

Coastal environmsnts are the ultimate repositorbs for al1 contaminants

originating from agriculture, urban and industrial activitios. Among these contaminants,

mrcury (Hg), kad (Pb) and cadmium (Cd) are especially dangerou8 and have been

observeâ to cauw signifiant damage to aquatic organisms (Kennirh 1998, Salanki

1985). To protect these resourœs and human heaîth from hamful m ta l contaminents,

environmental managers requim tools such as accufate predictive modeb to precisely

predict the fate and effed of metal contaminants in ecogstems.

Various models developed for m a l bioaccumulation in aquatic organisms have

been b a W on biwnergetics (Norstrom et al. 1975). free-metal ion actMty (Tessier et

al. 1803) and kinetb (Walker 19W, Luoma et al. 1992). The kinetic mode1 ha8 bwn

tested on several seston-ingesting invertebntes, cg.. bivalves (Wang et al. 1996).

zooplankton (Riîterhoff & Zauke 1997) and chironomids (6endell-Young 1999). Metal

concentrations pfeâicteâ by these modela a n genenlly in the range of obwmed

values for metal in bivalves and zoopknkton. but oRm over- or under-mtimate final

tissue concentration. One pouibk mason for ruch dircrspancies is that mout rnodeh

have not taken into account the f d i n g ôehavior of the organisms in nsponw to the

wide nnge food quaMy and quantw that occun in the natunl environment. The

130 pmwnt chaptei atternpts to apply a kinetic model of rnetal accumuktion that for the

fimt time incoipontes a seledive feeding khavior of musmls.

The developmnt of the model requhs the incorpontion of the processes

examina in radier expriment8 (feeding behavior, metal assimilation and metabolic

cost d fwding) into a laiger framework that indudes the pahways through which filter-

feding musmh can accumulate metal. The uptake and exchange proasses of

cadmium betweem water column and mussels can be viewed as s two-compartmnt

p r o s s . This means that if a 8ub8tanœ (Cd) passes fmm one location (water column)

to another location (rnussl's tissues) st a meawnbk rate, then each location

constitutes a separate cornpartment for the substance (Cd) (Riggs 1963). The route of

uptake of metal contaminans is through the gills from the dissolved phase and from

ingested particles in the gut (Bryan and Langston 1QQ2, Wang et al. 1996), whereas

the route of toss is via the gilh and excretion into the fueces (Borchardt 1983). Thus,

the model of cadmium bioaccumulation by filr-feeding musmls wil focw on two

procemes, uptake of Cd from dissolwd and paiüculate phases and elhination of Cd

by the musseIr. Pndicted values of cadmium to o b ~ w e â values from the fmld are

then compad to detemine modal acwncy

Tabk 5.1 provides definitions of terms used in the kinetic model of cadmium

bioaccumulation. A fiowch8rt of the model application is p n ~ n t e d in Appendix E. The

mode1 is based on iirstorder kinetics and assumes that 1 ) mussels obtain metal (Cd)

from the uptake of water and ingested diet, 2) the concentration of Cd reached by the

mussel is balanceâ between uptake and elimination, and 3) uptake ftom the water (Ud

and ingesteâ diet (Ud) am independent and additive. The model is expmsed as:

dCd& = (U. + Uo) Cdm (k. + Q)

whete Cdm is the Cd concentration (pg gdw f') in the mussel tissue at tirno t (d), Uw is the

Cd uptake tate from the dissolved phase (pg d" gdw"), Ud is the Cd uptake rate from

the ingested dkt (pg d" gdw"). and (k, + g) ir an dimination rate constant due to

physioIogkaI loss (û,, d") and Cd dilution through growth (g = the mass-specific growth

rate constant, 6').

The uptaûe tate of cadmium fiom the water by mu8sel8 a n be described as the

ptoduct of the concentration of diuotved Cd in the wnounding water (Cd.), the mussl

clearance rate (CR) and the eflicisncy of cd2' ion tmnrfer actou the $i l surface

(retention eflicbncy , -):

Table 5.1. Definition of tenns used in the model.

---

Unit Ternis Symbol Meaning

Bioavailability

Uptake

Elimination

Retention efficiency

Assimilation eficiency

Bioconcentration

Oieltafy accumulation

Bioaccumulation

The detgree to which metrl contaminant in a potential source is fm for uptake The movement of metal into the tissues of organisrns The summation of metabolic excmtory and physicochernicsl procesws msulng in the decrease of rnetal contaminant in the organisms. The fraction of ionic metal in the water colurnn that is absotbed across the g i h and tissues. The fraction of total rnetal in ingesteâ p8iück8 that is absohed across the gut epithdium. Accumulation of metal from water (dissolveâ phase) Accumulation of metal from ingestd particks (paiticulate pham) Total accumulation of metal from both water and ingested particies.

The cadmium uptake rate from the ingested diet can be describeci as the

product of the concentration of Cd in food particles (Cd<). the ingestion rate of particies

(IR) and the eficiency of Cd assimilated from the gut contents into the mussel tissues

(CdAE). Because musseh may display a selective or a non-wkctive feeding behavior

in response to seston matrices, two possibk sœnarios of cadmium uptake from the

seston matrices (diet source) can occur. Fimt, when mussels fully display a wîective

feeding khavior. the cadmium uptake rate from the diet can be expmsseâ as:

where Cd, is the cadmium concemtration in the organic fraction. I& is the ingested

organk component and CdAE, is the as8imilation efficiency of Cd from the orgonic

fraction of the diet. Second, when the mussels display a non-wkctive feeding

behavior, cadmium uptake rate from the diet can be e~pmssed as:

where Cd,,,, is the cadmium concentration in the inoiganic fraction. IRn is the ingested

inorganic compnent and CdAEim is the assimilation efficisncy of Cd from the inorganic

fraction of the diet.

134 For fibr-feeding musmls, the ingestion rate (IR) a n k descriôed as the

product of the concentration of organic and inorganic surpended particks (SPM), the

clearance rate (CR) and the sorting efficiency (SEF) when mussels display a seledive

feeding behavior:

!& = SPM x CR x SEF (5.6)

When they dispky a non-seledive feeding behavior, the ingestion rates ((IR) can b

expmssed as:

IRm = (SPM x CR) - PF

= ((POM + PM) x CR ) - (PF, + PFh)

= ((CR x POM) - ?Fm) + ((CR x PM) - PF,) (5.7)

when POM and P M are the organic and inorganic fractions of suspnded particles.

and PF, and ?Fm am the organic and inorgank fractions of pwudofaeces production

rates, mspedively .

Themfore, d#ary cadmium uptake by fîter-feeding mu8sels dispkying wbctive (5.8)

and non-mkctive faxjing (5.9) a n be expmssed as follows:

Ua = Cdm x (SPM x CR x SEF) x CdAEm

Uc = Cdm x ((CR x POM) - PFm) x Cd-AEm +

Cdni ((CR x PIM) - PFpni) x Cd-AEim

135 Elimination of cadmium fmm mu8sels via the gill SUIfaCe. rnantk and f8eces Gan

bb simpliM as firstorder dimination. The elimination formula is baseâ on the rate of

msidue elimination following exposurs king dimctly proportional to the concentration in

the whok mussul's tissues (Cdm) and the dilution of Cd due to growth (equation S. 1).

By cornbhing tqs 5.1 - 5.9. the kinetic mode! of cadmium in filter-fctding

rnusds dirploying a non-seledive feeâing c m be wnttsn as follows:

dCdJdt = [Cd. x CR x -1 + [Cdoni x (CR K POM - PF) x Cd-A€-] +

[Cd, x (CR x POM - PF) x Cd-AEm] - Cdm [k + g] (5.10)

But. when mussek displaying a seledive feeding behavior the rnodel becornes

dCdmldt = [Cd. x CR x a*] + [Cdm x (SPM x CR x SEF) x Cd-AEm]

- Cdrn [km + 01 (5.1 1)

Under 8 steady state (88) condition whem the uptake and the elirnination rates are in

equilibrium (i.e., (Cd,,,/& = 0) or t apptoaches a (t + a), the mode1 is either

Application of a kinetic model to predict Cd bioaccumulation in mussel tissues

was completed by first. model parameterkation and second, executing the model under

selective and non-te!ective feeding strategks (Microsafi Corp. 1996). Made! accuracy

was detemined by conducting a sensitivity analysis for wch input paramter and by

comparing obwrved verws predicted values.

5.3.1. Mod.1 prmm(.rbrtbn

Input panineten u s d in the mode1 and thdr sources a n listed in Table 5.2.

C.dmium upkh nt. h m th dls8olv.d phase (U,J The route of mtaî

uptake from t h disroked phase is pnmarily a passive transport through the gillr and

mantb tissues (Simkiss and Taylor, 1989). Thme panmeten that am important in the

Cd uptake rate from the dissolved phaw a n Cd conc8ntration in the dissolveâ phsm

(G), Cd retention effcbncy (h). and the mussel's clsarance n te (CR). C, values

were obtained from ChrétMn (1997) who detemined dirsolved Cd concentrations in the

Howe Sound ana (Table 5.3). CR data wem derived from a pmvious rxperimnt

(Chapter II. Table 2.3). Cd retention efliciency war derived from expeflrnenb in which

muswls were exposeâ to 0.45 pm fiItmâ wawater without food particles for 4 h and

depunteâ for 24 h. Cd rstention efficiency was then caIcuIated as lWcd activtty in the

mussel's tissue (dpm) divided by '%d activity in the meâium (dpm). Tabk 5.4 shows

Tabk 5.2. Modal panmtefization for the kinetic modal of Cd accumulation by filter-fading musseIr.

Model parameter Symbol Values Units

Envkonnwntil prnmrtrn Seston concentration' SPM 4.1-32.0 rnggdw-' Seston quali 'u' SQ 3.1 - 23.1 % Dissolved Cd Cdw 0.0 14.14 P9 L' Cd in organic fraction of diet' Cda 1.5 W L': Cd in inorganic fraction of diet' Cdim 1.8 - 5.98 pg 0'

Mean dry weight of mussel

UpW. pamnnhrr: Clearance rate Pseudofaeces production rate Soiting efiicbncy Ingestion rate Cd retention effciency 109 Cd assimilation efficiency

Diatom Natunl wdimnt3

Elimination rate constant4 Growth rate constants

CR PF SEF IR

3.3 - 21.8 Eq. in Fig 2.8b Eq. in Fig 2.7 Eq. 5.6 6 5.7 0.21 - 1.55

gdw"

L gdw" h" g gdw" h" % g gdw-' h" %

Note: ') Stecko and Bendell-Young (20ûû). ')~hrdtien (1997), ' )~agmn and Fisher (1997). 3 the elirnination rate constant for marine bivalves e.g., oysters. clams and mussels (Reinfeder et al. 1 WB), 3~rrgenwn (1 996).

Tabb 5.3. Dissohred Cadmium (Cd4 in Howe Sound. British Columbia, Canada (Chrétien 1997)

Month Offshore Inshore dd/y h g L-') O ~ Q L")

O1 Il 711 994 0.072 0.073

O 1 11 8/1994 0.055 0.1 10

0310111994 0.069 0.240

03/2911994 0.097 0.230

û4/20/1 994 O .O80 0.053

05/1 O11 994 0.106 0.040

0611 611 994 O ,045 0.652

0712811 994 0.010 0.031

08/24/1994 0.021 0.01 7

0912 711 994 O .OS8 0.049

1 1 10211994 0.056 0.054

O 111 611 995 0.1 15 0.071

Average 0.065 O. 135

Table 5.4. Retention eMcLncy (or) of '''cd by the blue mussel Mytlus tmssulus. I

1 13.189 140 1 .O6

2 19,637 1 70 0.87

3 1 18,372 245 0.2 1

4 187,356 825 0 .U

5 190,625 1,618 0.85

6 241,6Sû 2,040 1 .18

7 177,442 2,756 1.55

8 258,127 3,058 1 .18

Average 134,044.89 1,294.67 O .82

that a, varies from 0.21 to 1.55% with an avenge of 0.82%. Wang et al. (1996) no td

that mean Cd retention efficiency (aJ is 0.31%. Thb lower m a n a, could be dur to a

smalkr range of '%d exposum concentrations and diffennt duration of Cd exposum

during their experiment.

C.dmIum upbk. hbn, ing..kd dkt (Ud Cd uptake from the ingrsteâ

partickr (diet) is controlled by Cd conœntmtion in seston partickt (CdOom.,), the

ingestion rate of the mussels (IR) and Cd assimilation effciency (CdAE). Stecko and

BendelCYoung (2000) reported for th8 Fraser River estuary that concentrations of

inotganic Cd were greatert during the winter monthr (total inorganic Cd content of SPM

= 5.98 pg g", Match) and lowest during the summctr (1.94 and 3.75 pg g" for June and

August, respectively). Furthrrmore, their study indicated that Cd in SPM wos

auociated primarily with msnganeoxides. Pollet and Bendell-Young (1999) also

indicated, in their labotatory study , that 'Oecd was mainly associateâ with ironoxide and

mangonesxide, but only trace amounts of Cd won, ncovemd ftorn the otganic

component of SPM.

Laborstory rtudies on feeding msponw (Chapter II) showed that total ingestion

rates (IR8) incmased with an incmaring SPM conœntmtion, with the maximum tatr

occumng at the intermediate diatom concentration (20 x 10' mWs c') and high $il lord

(20 and 50 patticks L"). With the exception of these seston matrias, total IRs of

mussels nnged hom 0.02 to 0.13 g h" gdw" either composed of onty prrticukte

otganic matter (POM), or of POM and particulots inorganic motter, PIM (Tabk 2.3).

Mussels ingested primarily the organic component of the seston when the seston

quality wss about 40%.

Previous studhs on Cd assimilation by marine bivalves showed that Cd-AEs of

cadmium are highly vanabk (Table 3.4). Cd-AE from ingested algae ranged from 25 to

65% (Borchafdt 1983). Wang et al. (1 995) indicated that CdAEs from ingested manne

aîgae nnged between 20 and 50%, and that the Cd-AE decreased linearly wlh

increasing ingestion rate and food concentrations. Labofatory experimnts (Chapter Ill)

showed that when rnixed diatom and s i l are given to musmls. the Cd-AE from the

organic component of SPM was ôetween 38 and 83%, whenas from the inorganic

component the Cd-A€ was 85%. However, remnt studbs by Gagnon and Fisher

(1 997) indicated that 'Oscd assimilation efici8ncy by filter-feeding bivalves fmm natural

s8dimnt was only 15%. Henœ. the values of Cd-A€ frorn oiganic and inorganic that

were used in the pnmnt model wen 38% and 15%, nspectively.

Cadmium diminafion mh fhm m u a d Wwo Metal elirnination from

musseh occun through a passive proam involving deorption ftom pemeabk

mmbnne surfaces, ag., giWs and mantbr. and egestion thmugh faoces (Raintmw

19Qû). Previous studies (8.g. Schok 1980, La Touche and Mia 1982, Cobman et al.

1986) concludeâ that rlimination of Cd from mussel tissues was a slow proass.

Furthemon, Cosu (1989) stated that the kinetics of Cd wiaiin biological tissues ir

slow, with a minimum bioîogicrl haClL (Le., the time n o d d foi the Cd concentration

to dmp by onshal that pnwnt at the ôeginning of any time interval) from laôoratory

142 expeflments of 14 drys. Borchardt (1983) ako indicated that food concentrations couM

affect the depuratton rate of Cd. with the kngest biological half-lifs of Cd being

obsewed at a maintenance bvel of food of 16 x 106 celb d" mussel-'. However, Lares

(1997) noted that depuration rates of Cd by M. californianus wem vefy fast (i.e., Cd

concentration rsduced mon than hatf after 5 d depuration period in a Iaboratory

expedment, and the rate of depuration was even fester in C i î d studies during a

phytoplankton bloom). Reinfeîder et al. (1998) stated that dimination rate constant in

four marine bivalves (oysten, cîarns and muwls) ranges from 0.01 to 0.04 dry".

Reinfelder et al. (1998) alw stateâ that growth dihition (Le., the apparent

demase in rnetal concentration msulng from organiem growth) mn h o m e e

signifiant factor in mta l bioaccumulotion in bivaives that expehnœ wasonal change

in tiswe mass relatecl to their reproductive cycîes. Wang et al. (1996) found that

preôicted mtal conwntrations based on a kinetic mode1 in which growth dilution war

negkcted wem overastimateâ within a factor of Iwo to three of the actual metal

conœntmtionr measumd in fwlâcolkctsd muswh. Thur. It is suggested that, when

the gmwth rate constant (g) is comparabb to or grnater than the elimination rate (îc.),

that growth rate k incocpomted into the kinetic mode1 to compensate for growth

dilution of mta l in the tiuuer. But. when the growtti rate constant ir much smalkr

than the elimination rate constant. gmwth dilution a n k ignored in the mode1

alculrtion. In the pnsent model, the growth rate constant (g) of murwh war included

in Me model kcauro the value of g was 5 times higher than that of k..

1.3.2. Exprrknntil tnnrplint

The experimental transplant was conducted to test whether the Cd predicted

from the kinetic model war within range of obwwed values. The transplant was

canied out from August to September 1998 (29 d). Same sue mussels (mean = 45.9

mm SL, N = 45) collected fmm Honashoe Bay weie tmnsplanted to Tsawwassen (Fig.

5.1). The site war chosen because it was within the Fraser River estuary and

suppocted an abundanœ of natunl muswl populations. Before the transplant. the

mussetls were depurated (i.0.. mussels were kept without food in fibred seawater

under labontory conditions to teduce metal concentrations from musel tissues) for 3

d. Ten of 45 mussels won analyzed individually for Cd content as control value of Cd

concentrations M o n transplantation in August. The nmaining of mussels were

transplanted in Tsawasmn (the Ferry Terminal). Affer 29 days of deploymrnt. the

mussetls wem recovrmd and depuratml for 3 d in the labontory at the Dept. of

Biologiml Sciences, Simon Fraser University.

S.3.3. Cadmium anriyak

Murseh wrre ckaneô, sepantecl, and individually dried in a 60% oven for 96 h.

and their rhell bngth and dry tiswe weight were masureû. The diy tissue was gmund

with a moitar and pestk. An average of 0.25 gdw of ground tiuue from individual

musmls war digesteâ in 5 m l of HNOI (70%. Baker lntra Analyzed) and 2 mL

deionwd wiater in a clowd v e s ~ l in a pnssurscontrolkd microwave (CEM Microwave

Technobgy, Modrl MDS 2000) for 15 min. The second $tep of digestion involveâ

adding 1 .S m l H a 2 (30%. BDH. Analyticrl Reagent) for another 15 min. The digesteâ

Fig. 5.1. The study's sites (sarnpling and transplantation) of the blue mussel, Mytilus tmssulus. (. . . . . . = intertidal area).

Honeshoe Bay C VANCOUVER e?

solution was then centrihigeâ to wpanb the debris and extract cadmium. The

cadmium was deteminecl using a Graphite Furnace Atomic Absorption

Spectrophotometer (GF-AAS) al Analytical Service Laboratodes Ltd. (ASL), Vancouver,

British Columbia.

Standard Reference Metefial (SRM 1566a) of oystet tissues from the U.S.

National lnstitute of Standards and Technology was included for quality assurana.

The m a n value of the Cd in mfennce materials of the SRM1566a(4.0 t 0.25 pg gdw-'

n = 7) was within the 95% confdence limits of certif i i value (4.2 î 0.38 pg gdw") with

a coefficient of variation of 6.3%.

Predicted Cd concemtrations in September (Cdm = 3.91 pg gdw") were 17%

higher than obsenred Cd concentration (3.33 pg gdw"). In contrast. when the sorting

proœss was not incorporateâ in the kinetic model, the Cdm (6.32 pg gdw-') was almost

doubk the actuat Cd concentrations of the tmnsplanted m u s k (Fig. 5.20).

The mode1 al80 pndicted that the relative importance of Cd uptake from the

ingested dW would Vary throughout the year (Fig. 5.2b). An annual average of 56% of

Cd- was attributai to dietary Cd uptake. Dietary Cd contributed 35 to 94 % of tissue

Cd concentration dunng most months, except in Novemkr and May when the values

were 29 and 3016, respectively .

Sensitivity analysis was p r f o m d on the baseline value model identifii the

influence of variation in input panmeten on the prsdicteâ Cd concentration in murrals.

The analysis comparsd the pemntage change in Cdm with a given prcentage

change in one of the mode1 input parameten (Appendir F). Each panmeter in Table

5.5 was v a d by i 20% from L bawline value and the masure of sensitivQ (S) was

avengeâ for the positive and negative variations. In geneml, the model did not greatly

fluctuate to vadations of input panmeten. Cdw is insensitive to changes in

envimnmntal panmtem of diuoived Cd (S = 0.42) and retention effciency of Cd (S

= 0.45). but sensitive to changes in the bioîogiccil panmten, most notably musml

147 Fig. 5.2. (a) Pdicted Cd concentration in mussel tissues (Cd,) for a one-par period;

square = Cd, under non-sekctive f d i n g khavior, dot = Cd, under a wkctive feeding behavior. bar = field Cd concentration (Cd,) befom (x) and after (y) transplantation (man + SE). (b) Dietory Cd uptake (%) foi the blue mussëls during a one-year priod.

12.0 - 10.0 -

S O N D J F M A M J J A

A S O N D J F M A M J J A

Table 5.5. Sensitivity analysis for Cd bioaccumulation modrl. The quantity S is a masure of paramter mnsiüvfty

Note: *)

- ~Giionrnenta~ panmeten

SPM SQ Cdw Cd,

Biological parameten CR SEF aw

Cd-AE, ka 9

mkrs to Cd wnc%ntration at the end of the integntion penod in the base case modsl, and ACd- is the change in the value of Cdpia brought about by varying the mode1 panmater. Similady, the denominatoi masures the variation in the panmeter of intemst dividd by its bamline value (Dowd 1997, Zheng and Bennet 1995). The d paramter was î 20% and A Cdpno was averageâ for + and - cases.

cleannw rate, CR (S = 1.0). elimination rate constant, k, (S = 1.0) and growth

constant, g (S = 1.03). Hencs. the senslivity test of the mode1 for the month of August-

Septernber showed that CR, b, and g wem important panmeters to mearum in order

to accumtely predict Cd concentration in mussel tissues.

Based on the characteristic of suspndeâ load of the Frawr River estuary, the

kinetic-baseû mode! w0s applied by incorponting feeding behavtor parameter (SEÇ)

and Cd in both organic and inorganic component of SPM as the source of Cd. Under

the diffennt seston conditions that a n occur in the Fraser River estuary, it was

predicted that Cd concentrations in musse1 tissues dunng a one-year penod would vaiy

betwwn 2.58 and 6.15 pg gdw". The Cd- war 17% higher than the actual Cd

concentrations from fmM studies (Cdab. ) aftet the musseh wem transplanted for one

month period. But. when a sekctive f d i n g khavior was not incorporateâ into the

model, the Cd, was double the obsenred concentrations.

The Cdm vadecl throughout the year. k ing low in summr and spring and

maximal in winter. This seasonal change in preâicteû Cd concentrations is generalîy in

agreement wlh pmvious fad studios. Amiard et al. (1986) and Farrington et al. (1983)

noted that maximum Cd concentrations in mussel tissues occunsd in winter and

minimum concentrations and summr. The fluctuation of Cd concentration in mussels

cou# k attributabk to a change in biologiul adivtty sssociated with food avaiiabihty

(microalgae) and growth of mussols. Furthemore. changes in land drainage kading to

changer of Cd inputs into the Fmr r River esturry couiâ alm mult in fluctuations in

the Cd concentrations in the fiter-feeding rnuurh.

151 The relative importance of Cd uptake from solution and diet differed among

previous studhs. Earlkr studies suggestd that uptake of particulate Cd contributed <

10% of t h total body burden in mussels (Borchardt 1983). Recent studies of Wang et

al. (1996) concluded that both dissolved and food sources contributecî almost the same

amount ta the total Cd concentration in mussel thsues, i.e., 45% of Cd was from

ingested food and 55% from water. In contrast, the prewnt model predided that the

dietary uptake couM contribute as much as 94%, or as littk as 29% to the Cd

concentration in mussl tissues (Cd-). Themfom, depending on the food conditions

and the response of the mussel to its food environment, the contribution of the dbtary

Cd wouîd Vary throughout the season.

ln sumrnary. the kinetic model incorporating fWing bhavior provided malistic

estirnates of Cd concentration in mussl tissues. The mode1 highlights the importance

of seledive feeûing in Cd bioaccumulstion in Ilr-feeding muamIr. The feeding

behavior (wkctive vs. non-sekctive fading) affects the uttimate source of Cd uptake.

When the mussel fully emplop a seMivet feeding stmtegy, the main soum of uptake

would k from the organic component of seston. However, undsr natunl conditions of

the Fraser River estuary (Le., Cd mostly auociatd with inorganic particles and seston

quality genenlly < 23% organic matter), sorting efficiency is bm than optimal and the

blue mussel ingests a large portion of inorganic paitickr in itu d#. As a rerult. Cd

accumulation in mussel tissues is mrinly dominateâ by ingested diet.

Amiard JC. Amiard-friquet Cl ûarthet C. Métayer C (1986) Contribution to ecotoxicological study of cadmium. lead. copper and zinc in the blue mussel Mytilus edulis. Mar Biol90:425431.

BendelCYoung LI (1999) Application of a kinetic mode1 of bioaccumulation across a pH and salinity gradient for the predicüon of cadmium uptake by the sedmnt dwelling chtronomidae. Environ Sci Technol 33: 1501 -1 508.

Borchardt T (1983) Inflwnce of food quantity on the kinetic8 of cadmium uptake and los$ via food and mawater in Mytihs edulis. Mar Biol76:67-76.

Bryan GW, Langrton WJ (1992) Bioavailability, accumulation and effects of hwvy metah in W i m n t s with special refennce to United Kingdom estuaries: A review. Environ Pollut 76:89-13 1.

Chrétien AR (1997) Geochemical khavior, fate and impact of Cu, Cd and Zn from Mine Effluent Discharges in Howe Sound. Ph.0. thesis. ûepartmnt of Earth and Oœan Sciences, The Univemty of British Columbia, Canada.

Coleman N. Mann TF, Mobky M. Hickmon N (1986) M@us edulis plenulatus: an 'integrator' of cadmium pollution? Mar Biol92: 1-5.

Cossa O (1989) A review of the use of Mytilus spp. as quantitative indicaton of cadmium and mercufy contamination in coastal waters. Oœan Act 1 2:) 1 7432.

Dowû M (1 997) On predicting the growth of cultumd bivalves. Ecol Model 104: 1 13 - 131.

Famington JW, Goldbrg ED, Rimbrough RW, Martin JH, Bowen W (1983) US 'musml watch' 1976 - 1978: an overview of the trace metal, ODE, PCB, hydrocaibon. and artificial ndionucbide data. Envir Sci Technol 17: 4904%

Gagnon C, Fisher SN (1997) The bioavailability of sediment bound Cd, Co, and Ag to the mussel MytiSus dulis. Can J Fish Aquat Sci 54147456.

Jrrgenwn CB (1996) Bivalve f i l r feeding mvisited. Mir Ecol Prog Ser 142:287-302.

Kennish MJ (1998) Trace metal-secliment dynamics in estuaries: Pollution assemment. Rev Environ Toxicol155:69-110.

Lares ML (1997) Muswls as indicaton of cadmium and kad in the marine enviionment. Ph0 Theris, ûepartment of Oeeanognphy, The Univemty of British Columbia, Canada 153p.

153 La Touche YD, Mix MC (1982) Seasonal variation of arsenic and other tnce ektwnts

in bsy musseh (MyWus dulis) Bull Environ Contam Toxicol29:665-670.

Luoma SN. John C, Fisher NS, Steinberg NA. Otemland RS. Reinfeldetr JR (1992) Detemination of Seknium bioavailability to a knthic bivalve from particulate and solute pathways. Environ Sci T echnol26:485491.

Microsoft Corp. 1995-96. Getting rssuîts wiai microsoft office for Windows: Analyze scientifk and engineering data, Part: 13. Microsoft Press, Redmond, WA US.

Norsttom RJ, W i n n o n A€, deFreitas ASW (1975) A bioenergdics-based mode1 for polkitant accumulation by fish : Simulation PCB and mthylmtcury nsidue Ievels in Ottawa River Yellow prch (Pem flavescens). J Fish Res Board Con 332469267.

Polkt 1, Bendell-Young LI (1999) Uptake of 'O 'C~ from natutal sedimnts by Me Mue mussel A#y?ilus tmsulus in relation to sediment nutritional and geochernicat composition. Arch Environ Contum Toxiwl36:288-294.

Reinfelder JR. Fisher NS, Luoms SN. Nichoh JW. Wang WX (1998) Tnce ekrnent trophic tmnsfer in aquatic organisms: a critique of the kinetic model approach. Scie Tot Env 219: 1 17-1 35

Riggs OS (1963) The mathematical approach to physiological probkms: a criacal primer. The William8 6 Wilkins Company, Baîtimom USA.

Rit't8fhoff J, Zauke GP (1997) Biorccumulation of tnco mstals in Greenland Ses copepod and iirnphipod colkcüves on boardship: verificrtion of toxiwkinetic model parameten. Aquat Toxicol40:63-78.

Salanki J (1985) Heavy m t a h in water organisms. Symposia Biokgica Hungarica. Amd4misi Kiad6, Budapest.

Scholz N (1980) Accumuîution. l o u and mokcular distdbution of cadmium in Myitlus edulis. Hebolandw Wi8s Meemsunten 3368.78.

Simkiss K. Taylor MO (1989) Metal fluxes across t h membnnes of aquatic organisms. Rev Aquat Sci 1 : 1 73-1 88.

Stecko JRP. ûendeKYoung LI (2000) Contnsting the geochmirtry of suspended particulate matter and deposited d i m u n t within Ftawr River esturry. App Geachem (in pmm).

Tessier A. Couilkrd Y, Campbsll PCG, Anclair JC (1993) Modeling Cd pwtitioning in oxic lake W i m n t s and Cd concentrations in the fmshwater bivalve Anodonta grandis. L h o l ôceanog 38: 1-1 7.

Walker CH (199ô) Kinetic rnodelr to ptedict bioaccumuWon of pollutants: w ~ y mview. Funct €col 4:29&301.

1 54 Wang W-X. Fisher NS, Luorns SN. (1 995) Assimilation of trace eiements ingested by

the murwl, Mytidus Mulis: effect of algal food abundance. Mar Ecol Prog Ser 1 29: 165-1 76.

Wang W-X, Fisher NS, Luoma SN (1996) Kinetic detemination of trace ekmnts bioaccumulation in the mursel, Myti/us dulis. Mar €col ?mg Ser 140:9l-I13

Zheng C, Bennet GD (1995) Applieâ contaminant transpoft modeling: theory and practiœ. Van Nostnnd Rdnhold, New York USA.

CHAPTER VI

OVERALL COUCLUSIONS AND RECOMMIEMDATIONS

The blue muswls of the genus Mytllus (M. edulis, M. g~llopmvinci~Iis, and M.

celilbmianus) have been widely adopted as bioindicators of mta l concentrations in

marine ecosystem monitoring progmms. The advantage of th8 analysis of m u s f

tissues instead of seawater itseîf is that the rnuswls provide a record of bioevailable

mtab integnted over a period of months, suasons or yean. A pfedictive rnodel of

metal accumulation can help to identify environmental and biologicrl factors that

principally affect the ekvation of Cd in bioindicator species. Hence, t h m labofatory

expenments were conducted to study the rok of suiedive fading khavior of fiBr-

feeding musmis in msponm to envimnmtally mkvant w o n matrices and its

implications for the accumulation of Cd by the Mue mussel (Mylilus tmssu/us), a native

mussel 8pcies of Britbh Columbia.

The fimt expriment detemined the feeding responw of the muswls. The

primay conclu~ion war that -$ton as masumd in the environment was not what the

mussel war actually consuming, and dependhg on the quality and quantity of the

seston, the mussel had the ability to wkct rither orgrnic particles or both inofganic

and organic partictes for ingestion. The wRng efficiency was maximal at a wrton

quaMy of 40% organic matter. Sorting efficiency bscrm lem as the seston quslity

(SQ) decressed (< 25%) or incmased (> 60%).

As t h blue musaet coulâ employ s wlective feeâing stratagy in ingesting itr

diet, the mcond experiment examineâ if this wîective feeâing detemineâ Me fate of

Cd in muwel. The main condusian war Mat when mussels were expouid to wrton

matricm where only aigm hid been kbekd, 'Oecd assimilation was proportional to Me

157 organic content of the diet with the maximum '''C~-AE occumng at the filtering

maximum for this specbs. However, when the silt was labekd, ' O e ~ d - ~ ~ s were

independent of the diet puality. and with the exception of the sittonly exposums wem

maintaind at a constant maximum value of 85%. The addition of algae to the sitt-

laôekd diet appamntly activated the digestive processes and resutted in a gnater

assimilation of SN-bound '%d cornparcd to dkts ~ ~ ~ p r b c d of just silt alont. Hence,

uptaûe of Cd from the ingesteâ d ia will k dependent on both adive (ia., digestion of

organic matter) and passive (i.e., deorption fmm the surface of inorganic silt particks)

plocesses.

To test whether wkctive feeding displayd by the rnuswb was costty, a third

experimnt war conductsd. The dudy indicated that the metabolic costs associatd

with a seledive feeding behavior accounted for a maximum of 0.9% of the assimikteâ

energy. Under an optimal feeding condition (e.g. a 20120 seston mrtnx), the cost was

significantly d u c d to 0.2% of the a8aimilated ration, or only 22% of the metabolic

cost. Therefon, my msub suggest that aoy pn-ingestion sekcüve pmcesses mlateâ

to roiüng sndlor mpcting unwantd rsston components are done with a negligibk

bnergy expandihim by the blue mussel. The implication of this feeâing stmtegy is that

under high algal and high dlt load. mussels will ôe exposd to cadmium from the

organic wmponent only. In contnd. when wrton quality ir < 23% organic matter,

mwmb wiY maximue their ckannce rate w R les8 mting etficiency and be exposd

to Cd from both organic and inorganic wmponents of the seston.

158 As the Fraser River estuary is mostly low in seston quality, the rnuswls

genenlly display kss mWive on their feeding behavior. Taking into account this

fssding ôehavior, the partition of Cd within cornponent of seston matrices and

assimilation eMcicHicy of Cd. o kinetic model was successfully applied to filter-feeding

musseîs. When a sekctive feeding was incorponted into the kinetic-based model, the

prsdicted Cd concentration in mussel tissues {Cdm) was within 17% variation of the

obsenisd Cd conœntration (Cd*)). In contmst. when an amumption of a non-sekctive

feetding behavior was consided, the Cdm was ovefestirnated of Cd* (i.e., Cd- was

approximately doubb Cd*). Hence. the model highlights the important aspect of

wlective feeâing behavior in pdicting Cd bioaccumulation in fibr-feeding musseIr.

The modd alw showed that the dietary Cd uptob va- throughout the year and

contributed 35 to 94% of the total Cd concentration in mussel tissues.

There are a few unartainties with the cumnt model that may resuît in over- or

under-estimates of Cd concentrations. For instance, Cd concentrations in water

(diwoîved Cd) fmm the Howe Sound ares might not mkct the concentration of Cd

experisneed by the mussels in the Fraser River estuary. Furthenore, the mode1 was

run bamd on SPM data of 1994 and 1995, the data migM influence the loww valw of

the 1998 cd, as the inter-annual of SPM concentration and qurlity could al80 v a m .

The elimination fate constant of Cd for the pmsent model w8r estimated based on data

hom difiemnt species of muswl (Mytjlus dulis), thus the dimination rate constant by

MyWs tmssuIus needs to k veMi with radiotracer expeflmnts. Finally, the pmsent

modeling rpproach wiW become more robust as mors fmlâ data (e.g. prration of Cd

within seston componrnt, growth rate of musseîs) and labontory studies on the Cd

elimination rate by Mytius tlpssuIus becorne avaikbk.

To test the prisent modal, fald studies and dabomtory expiments need to k

conducteâ. In labofatory studbs on the elimination rate of Cd and other toxic trace

metab by M. trossulus will add understanding the fate of toxic trace metals in the

Fraser River estuary. Momover, pseudofaeces which nsulted from the setledive

feeding of mussels had a nlatively higher Cd concentration than suspended particles.

Thur, study on the food chain transfer of Cd (toxic trace metals) wil increase the

understanding of the fate of Cd (toxic h c s metalr) in the benthic etcosystem.

Tran8piantation of muse(s as a tool for biomonitoring any change in toxic tram metals

IOvels in the Fraser River estuary cou# alw hrlp to venfy the modal. Mussels for

transplantation programs should k csrsfully chosen from a pristine enviionment or

from cultured mussels. becauw my pnvious studies showeâ that musmh from the

Homeshoe Bay tendeâ to have higher Cd tissue concentrations than those

tnnsplrnteâ at the Fraser River estuary .

APPENDICES

1 64

Appendix B. Surnrnaiy of abbmviations and physiological paremeten in Chapter II

Parameter definitton

Assimilation rate, the product of true C-A€ and IR- PO& Carbon assimilation efficiency Apparent carbon assimiktion efficiency . TM carbon assimilation effïckncy Clearance rate. Ingeaion rate conecfed for PF,, Ingestion rate baseâ on CR x SPMc PF, correcteâ for seawater contribution. Pseudofaeces production uncorrectecl for seawater contribution. Particukte inoqanic matter cofrected for inoqanic component of seawater. Particulste organic matter comctd for organic component of seawater. Sorting efficiency . Suspendeci particuîate rnatter uncomctd for seawater contribution. SPI& comcteâ for seawater contribution. Seston quality

unit

% % L h" qdw*' mg h- gdw" mg h" gdw" mg h" gdw" mg h" gdw"

165

App«idbr C. The correction of measursd seston from influence of mawater.

Ta detemine the influence of seawater on rneasured seston, mussels that had

been acclimated to unfittemd seawater were placed into experimental tanks (three

tanks with rnuueh and two tanks with no mussels as a control). The flow-through

system was supplied wHh fittered seawater (1 .O pm). The seawater was colkdeâ hom

each tank every 30 min for 4 h.

To detemine the particks in the fiitered seawater that may affect composition of

meawreKî seston, faeces and pseudofaeces production. two sampks of seawater were

taken, i.e., 1 .O L and 3.0 mL of seewater. For correction of measursd seston, the 1 .O L

of wawater was fiîtered onto a 0.45 Fm GFIF MiHipore, washed with 250 mL

ammonium formate 0.5 M, d h d at 60 OC for 24 h, weighed. and then ashed in a muffle

fumrœ (450°C for 6 h). Similady, the 3.0 mL of seawater was colkcted. p l a d in

cnicibks, dned and ashed for correction of masured pseudofaeces production and

faeces.

Rasulb

Muswls that had been acclimated in unfilterd seawater were expowd to

fibd reawater for 4 h. As they were k i n g acclimatùeâ to filtered seawater, they did

not produœ pseudofaec88. Le., the exprimntal medium of fiîted seawatet did not

cause pseudofaeces production. However, they produceâ faecos at an average rate of

3.22 mg h" (n = 6). This faeces production may be the msul of a pmvious acclimation

period with unfittered seawater or a bypioduct of breakdown endogenous protein.

The prrück concentration of filteml seawater was on average 2.54 mg L-' (n =

8) for control tanks and 2-59 mg L" (n = 7) for tmatment tanks and these values wen

1 66

not significantly different hom each other (Student 1-test, t0.0~~ zmls = 0.08, p = 0.93). By

combining seston concentration from tmatment and contml tanks, the overal average

particle concentration of mawater was 2.56 mg L". This condsted of 1 .O3 mg L"

particulate inorgantc matter (PM) and 1 .S3 mg L" particulate organic matter (POM)

These particulstes would affect the concentration of measureâ seston. but they would

not be available to mussels, because the particle size was too srnaIl( < 1 .O pm). Thus,

although the filtered wawater had an average of 2.56 mg particks ciL", the particles

wen not biologically availabk to murwls.

The average sab of wawater that affected faeces and pwudofaem

production (PF) was 8.48 mg h" and conristed of 7.14 mg h" PIM and 1.34 mg h"

POM. 8ased on these resuîts, the experirnental data were conected as follows;

SPM = POM + PIM

SPM -(cl = (POM - P O L ) + (PIM - PIM-)

= (POM - 1 S3) + (PM - 1.03)

= POMc + PlMc

The phyriological panmtem, i.e., ingestion rates (IR) and coibon assimilation

efficiuncy (CAE) wem then detemineû:

IR,,,, = (PO& + PlMc) x Cbatanœ rate (CR)

IR txmmd (a = IR, - PF,

Appandix O. Texture of (a) pseudofaeces and (b) faeœs of the blw mussel (Myiilus tmssulus) bar = 0.50 mm

Appendix E. Fkwchart of application of kinetic rnodsl in MyWus tmssulus.

Envimnrmnbl conditkn: SPM gmatly varieâ Low wston qualtty (SQ) Cd mostly associated with inorganic fraction of SPM

-7- lnbrtidrl Illlu88ek:

- 16 h WJ b m t g d d*' . - Specilic rowthiate constant P (g) 0.1 d' .

+ ---------, //

\

1 \ ' -SPM = POM + PIM I 1 I 1 -PFw = 4.47 (SPM) - 1 5.77 1 1 1 -IRwc = SPM x CR - PF I 1 I - I L = POM x CR PFom I I I -IR,, = PIM x CR - PF,, I I I

-SPM = POM + PIM -SEF = 0.086(SQ)' + 7.019(SQ) - 67.û48 -& = SPM x CR x SEF

Predicted Cd accumulation: Cdddt = ( C d x CR x aw) + (Cda x (IR@m + IRm) x

Cd-AE) - Cdm (k + Q)

W i n e vaîuer of % for Norr-Seladie f d i n g h b v i o r (Norr-SF) l R m lRim Cdw r Cdim AEim îm+g Uw Uim Cd- W h W h w c uql9d'd c c ug(*wgdww* 2.2 68.5 0.021 0.016 3.7s 0.15 0.12 0.67 4.W 5 .a 2.1 57.0 0.ûS8 0.016 3.92 0.15 0.12 1.86 4.47 6.32 3.6 74.7 0.061 0.018 5.50 0.15 0.12 1.95 8 . Z 10.17 5.4 58.5 0.057 0.016 6.67 0.15 0.12 1.82 6.63 8.46 10.2 52.4 0.106 0.016 5.85 0.15 0.12 2.37 6.13 $.SI 10.0 33.3 0.140 0.016 5.50 0.15 0.12 3.14 3.67 6.m

20% inamae d SPM )epu(ine vdue (SF) moiith SPN w%j CU S€F lRtd lRorn lRim

mon % mgw % mgmg mglh~"nglh9 A 6.02 3 15.0 4 6 78.2 42.0 37.2 S 4.93 4 15.0 4 4 67.7 32.4 35.3 O 7.13 5 15.0 -37 90.8 39.7 51-1 N 5 . 4 8 15.0 4 73.5 11.9 61.7 D 8.32 16 10.5 24 72.0 23.3 40.7 J 5.18 23 10.5 49 40.8 28.0 20.8 F M 14.27 13 10.5 10 101.8 14.7 87.2 A 15.36 8 10.5 -15 100.4 24.7 83.7 M 16.45 9 11.5 -10 131.4 18.6 112.9 J %.JO 6 7.5 -30 132.1 05.7 46.4 J 13.72 4 15.0 4 160.2 81.6 78.6 A 0.30 7 15.0 -25 113.7 34.8 78.8

Uw Uom Cd,, wm wmwgdw

0.67 3.18 3.W 1.06 2.40 4-32 1.95 3.02 4.47 1.82 0.90 2.73 2.37 1.n 4.14 3.14 2.13 S.=

ü w Uim Cd,,., w9w w* w9* 0.67 5.75 6-43 1.88 5.12 6-m 1.95 9.53 t1.4ô 1.82 7.63 8.a 2.37 7.05 #.U 3.14 4.13 xn

20% demesme d SPM W i n e vdue (SF) month S W Sû(%j CR F IRfat lRom lRim

WW % mfw' % mgmgmg"'om9"'9 A 4.02 3 15.0 46 58.1 28.0 30.1 S 3.29 4 15.0 -44 50.4 21.6 28.8 O 4.75 5 15.0 -37 66.8 26.5 39.4 N 3.86 8 15.0 -14 54.3 7.9 $6.4 D 8.22 16 10.5 24 53.3 15.5 37.7 J 3.68 23 10.5 49 37.8 18.6 19.2 F M 9.51 13 10.5 10 73.1 9.8 63.4 A 10.24 8 10.5 -15 77.5 10.5 61.1 M 10.97 9 11.5 -10 92.9 12.4 80.5 3 25.60 6 7.5 -30 93.3 57.2 38.2 3 9.14 4 15.0 4 112.1 4 57.6 A 6.20 7 15.0 -25 81.1 23.2 57.9

20% rirvvapw, of SPM W i n e volw (Non-SF) f f im msm 2.1 -1 .O 5.2 0.5 10.1 0.4

23.2 27.5 30.2 92.9 24.0 11.2

ü w Uim Cd- wgdrv w w WQ- 0.67 4.22 4.œ 1.86 3.81 S.67 1-85 0.91 8.m 1.82 5.63 7A8 2.37 5.22 7.H 4 3.20 6.34

20% inmeam d SQ M i n e vdw (SF) month SPM -SC) CR S€F IRW lRom lRim Cdw

mon % mgiw % npm*,nOm@ngh@ ugll A 5.û2 4 15.0 4 2 68.6 31.9 36.7 0.021 S 4.1 1 4 15.0 -39 59.0 24.2 34.8 0.056 O 5.81 6 15.0 -31 78.3 28.0 50.3 0.061 N 4.57 10 15.0 -5 63.9 3.6 60.3 0.057 0 7.V 20 10.5 37 ô2.6 29.9 32.8 0.106 J 4.57 28 10.5 61 43.3 29.2 14.1 0.110 F M 11.69 16 10.5 22 87.5 27.1 60.4 0.083 A 12.80 10 10.5 -6 93.0 8.4 84.5 0.080 M 13.71 11 11.5 O 112.2 0.0 112.1 0.106 3 32.00 7 7 .S -23 112.7 54.9 57.8 0.W J 1 i . a 5 15.0 -34 136.1 . n.1 OMO A 7.75 8 15.0 -17 97.4 20.1 n.2 0 . a

Lkv Uim Cd- w* w* w* 0.67 4.85 5.63 1.86 4.43 6 . a 1.95 8.14 10.09 1.02 8.51 8.33 2-37 5.90 an 3.14 3.45 6.S

t b z r 8 r 2 a y i f "rd*'

20% i- al CR M i n e vdue (SF) mmlh SPM 84(%) CR SEF IRk4 lRom 1Rim Cdw

mO(L % mgni % m*www9dr w A 5.02 3 18.0 -46 83.7 42.0 41.7 0.021 S 4.1 1 4 18.0 U 71.4 4 39.0 0,056 O 5.W 5 18.0 -37 96.1 39.7 56.5 0.081 N 4.57 8 18.0 4 77.6 11.9 65.7 0.057 O 7.77 16 12.6 24 78.9 23.3 55.7 0.108 3 4.57 23 12.6 1i0 52.9 28.0 25.0 0.110 F M 11.89 13 12.6 10 112.4 14.7 97.8 0.083 A 12.80 8 12.6 -15 118.8 24.7 95.1 0.080 M 13.71 9 13.8 -10 143.7 18.6 125.1 0.106 3 32.00 6 9.0 -30 160.7 85.7 75.0 O.M5 3 11.43 4 18.0 4 û 170.4 81.8 88.8 0.W0 A 7.75 7 18.0 -25 120.6 31.8 85.8 0.m

20% incrsobe of CR W i n e vdue (Non40 Cdw ugR. 0.021 0.058 0.061 0.057 0.106 o. 140

0.083 0.080 0.106 0.045 0.01 O 0.021

ü w Uim Cd, ugl* ug(* w9div 0.81 8.08 6.ôB 2.23 5.40 7.62 2.34 10.09 12.43 2.19 8.06 40.M 2.85 7.73 4 O . U 3.76 4.48 8.24

20% ckmase d CR M i n e value (SF) month SPü % CR SEF IRtat l R m lRim Cdrr

mon % mew % nOmWnOm~"gm@ ugR A 5.02 3 12.0 48 53.6 28.0 25.6 OB21 S 4.11 4 12.0 4 4 40.7 21.6 25.1 0.058 O 5.94 5 12.0 -37 60.5 26.5 34.0 0.061 N 4.57 8 12.0 -14 50.2 7.9 42.3 0.057 0 7.77 16 8.4 24 a . 3 15.5 30.8 0.108 J 4.57 23 6.4 48 33.7 18.6 15.1 0.140 F M 11.88 13 8.4 10 62.5 9.8 52.7 0.083 A 12.80 6 8.4 -15 86.1 16.5 48.6 0.080 M 13.71 9 9.2 -10 80.6 12.1 68.2 0.106 J 32.00 6 6.0 -30 84.7 57.2 7.6 0.045 3 11.43 4 12.0 4 101.8 4 47.4 0.WO A 7.75 7 12.0 -25 74.1 23.2 50.9 0.02l

PFim msm 6.5 2.5 10.3 4.3 15.9 3.0

32.5 36. O 41 -3 11 8.9 33.8 17.6

uw Uan Cd,,,, wm wgdur -

0.54 2.13 2.68 1 1.64 3.13 1.56 2.01 3 . n 1.46 0.60 2.a 1.90 1.18 3.œ 2-51 1.42 3.93

uw Uim Cd,",, w* wgdw w + h 0.54 3.89 4.43 1.48 3.53 5.02 1.58 6.35 7.M 1.46 5.21 6.67 1.90 4 6 4 2.51 2.85 5.36

20% thcame of SEF W i n e value (SF) inonm SPH SQ(%) CR M F IRtol lRom lRim Cdw aw Cdoni AEom b+g ü w an Cd-

Cdw + 0.021 0.058 0.061 0.057 0.108 0.140

ü w Uim Cd, w- w* wgdw 0.67 4.99 S M 1.W 4.47 6.32 1.95 8.22 10.17 1.82 6.63 8.a 2.37 6.13 8.S1 3.14 3.67 6.m

20% ir#ease d Cdnr badine vdw (SF) moirlh -%) 8EF lm lRom IRim Cdw

mon % mguv % ngm@,nglh@nglh@ y)ll A 5.02 3 15.0 46 68.6 35.0 33.6 0.025 S 4.1 1 4 15.0 4 4 59.0 27.0 32.0 0.070 0 5.94 5 15.0 -37 78.3 33.1 45.2 0.073 N 4.57 8 15.0 -14 63.9 9.9 51.0 0.088 D 7.n 16 10.5 24 62.6 19.4 43.2 0.127 3 4.57 23 10.5 49 43.3 23.3 20.0 0.168 F M 11.89 13 10.5 j O 87.5 12.2 75.3 0.100 A 12.80 8 10.5 -15 93.0 20.6 72.4 0.098 M 13.71 9 11.5 -10 112.2 15.5 96.7 0.127 3 32.00 6 7.5 -30 112.7 71.4 41.3 0 . W 3 11.43 4 15.0 -10 136.1 88.0 68.1 0.012 A 7.75 7 15.0 -25 97.4 29.0 68.4 0.025

f f im m 6.5 2.5 10.3 4.3 15.9 3.6

32.5 38.0 41 -3 119.9 33.8 17.6

20% clamme d CdW W i m vdue (SF) month SPY SQ(%) CR F lRtd lRom lRim Cdw

m % mgw % W@w'WmmW W'L A 5.02 3 15.0 46 68.6 35.0 33.6 0.018 S 4.11 4 15.0 -44 59.0 27.0 32.0 0.050 O 5.W 5 15.0 -37 78.3 33.1 45.2 0.053 N 4.57 8 15.0 -14 63.9 9.9 54.0 0.049 O 7.77 16 10.5 U 62.6 19.4 43.2 0.- J 4.57 23 10.5 49 43.3 23.3 20.0 0.121 F M 11.89 13 10.5 10 87.5 12.2 75.3 0.072 A 12.80 8 10.5 -15 93.0 20.6 72.4 0.069 M 13-71 9 11.5 -10 112.2 15.5 98.7 0.092 3 32.00 6 7.5 -30 112.7 71.4 41.3 0.039 3 11.43 4 15.0 4 136.1 68.0 68.1 0.009 A 7.75 7 15.0 -25 97.4 29.0 88.4 0.018

PFim mgm 6.5 2.5 10.3 4.3 15.0 3.8

32.5 38.0 41.3 119.9 33.8 17.6

cdw ue/L 0.01 7 0.046 o. m9 0.048 0.085 0.1 12

0.066 0.w 0.085 O. 036 0.00s 0.01 7

Uw Uom Cd, u9(* wgdiv Wgdw 0.58 2.68 3.24 1.61 2.05 3.68 1.69 2.51 4.H 1.58 0.75 233 2.06 1.47 3.53 2.72 1.n 4.49

20% incmaae of anr W i m vaiue (SF) month 8PY MY%) CR SEF R o t Rom Rim

W L % mgw % m $ m 9 m 9 A 5.02 3 15.0 4 68.6 35.0 33.8 S 4.11 4 15.0 4 4 59.0 27.0 32.0 O 5.M 5 15.0 -37 78.3 33.1 45.2 N 4.57 8 15.0 -14 63.9 9.9 54.0 O 7 . n 16 10.5 24 62.6 10.4 43.2 J 4.57 23 10.5 49 43.3 23.3 20.0 F M 11.89 13 10.5 10 87.5 12.2 75.3 A 12.80 8 10.5 -15 93.0 20.6 72.4 M 13.71 9 11.5 -10 112.2 15.5 96.7 J 32.00 6 7.5 -30 112.7 71.4 41.3 J 11.43 4 15.0 4 138.1 68.0 68.1 A 7.75 7 15.0 -25 97.4 29.0 68.4

PFim m9m 6.5 2.5 10.3 4.3 15.8 3.6

32.5 38.0 41 .3 11 9.9 33.8 17.6

Cdw ugn, 0.021 0.058 0.061 0.057 0.106 0.110

0.083 0.080 0.106 O. û45 0.mo 0 . m

üw üim Cd- w* w* WOdur 0.81 4.98 6-78 2.23 4.47 6.69 2 3 8 . Z 9O.m 2.19 6.83 8-82 2.85 6.13 8.W 3.76 3.67 7.43

20% ôecmm d aw badine vaw (SF) month SPH SQ%j CR SEC lRW lRom lRim

mon % mgw % m g m g m g m g h g A 5.02 3 15.0 46 68.6 35.0 33.6 S 4.1 1 4 15.0 4 4 59.0 27.0 32.0 O 5.W 5 15.0 -37 78.3 33.1 45.2 N 4.57 8 15.0 -14 63.0 9.9 54.0 0 7.n 16 10.5 62.6 19.4 43.2 J 4.57 23 10.5 19 43.3 23.3 20.0 F M 11.89 13 10.5 10 87.5 12.2 75.3 A 12.80 8 10.5 -15 93.0 20.6 72.4 M 13.71 9 11.5 -10 112.2 15.5 96.7 J 32.00 6 7.5 -30 112.7 71.4 41.3 J 11.43 4 15.0 4 136.1 68.0 08.1 A 7.75 7 15.0 -25 97.4 29.0 68.4

Cdw ugn, 0.021 0.058 0.061 0.057 o. 108 0.110

0.083 0.080 0.106 0.045 0 . ~ 0 0.02l

Lkv Uim Cd,,,, w9dw w9- w* 0.51 4.99 5.m 1 .a 4.47 5.95 1 . ~ 6 8.22 a.n 1.48 6.63 8.09 1.90 6.13 8.03 2.51 3.67 6.17

20% lncmam d Cdom badine value (SF) morrth 8PY Sq%) CR S€F lRtot lRom lRim

mpn % B g w % mOmgmoniamoms A 5.02 3 15.0 48 08.6 35.0 33.6 S 4.11 4 15.0 -44 59.0 27.0 32.0 O 5.W 5 15.0 -37 78.3 33.1 45.2 N 4.57 8 15.0 -14 83.9 9.9 54.0 D 7.77 16 10.5 62.6 19.4 43.2 3 4.57 23 10.5 19 43.3 23.3 20.0 F M 11.89 13 10.5 10 87.5 12.2 75.3 A 12.00 8 10.5 -15 93.0 20.8 72.4 M 13.71 8 11.5 -10 112.2 15.5 98.7 3 32.00 6 7.5 -30 112.7 71.4 41.3 J 11.43 4 15.0 4 136.1 68.0 68.1 A 7.75 7 15.0 -25 97.4 29.0 68.4

lRim W h 66.5 57.0 74.7 58.5 S î .4 33.3

76.0 85.2 101 -7 106.2 130.4 90.9

ü w Uim Cd,,,, w m w* w w 0.67 5.98 6.M 1.86 5.36 7.21 1-95 9.88 11-81 1.82 7.90 s.n 2.37 7.30 9.74 3.14 4.40 7.53

20% cbcmam d CdOm badine value (SF) moiit)i $PM SQ(%) CR SEF lF?W JRœn lRim Cdw

mon % mgiw % mglh9mOmgmOmg uDlL A 5.W 3 15.0 4 68.6 35.0 33.8 0.021 S 4.11 4 j5.0 -14 5Q.0 27.0 32.0 0.058 O 5.94 5 15.0 -37 78.3 33.1 45.2 0.081 N 4.57 8 15.0 4 63.9 9.9 54.0 0.057 D 7.77 16 10.5 U 62.6 19.4 43.2 0.106 J 4.57 23 10.5 4# 43.3 23.3 20.0 0.140 F M 11.89 13 10.5 10 87.5 2 2 75.3 0.083 A 12.80 8 10.5 -15 93.0 20.6 72.4 0.080 M 13.71 O 11.5 -10 112.2 15.5 96.7 0.106 J 32.00 6 7.5 -30 112.7 71.4 49.3 0 . W 3 11.43 4 15.0 4 136.1 68.0 68.1 0.010 A 7.75 7 15.0 -25 97.4 29.0 68.4 0.021

k Uim Cd- w- w* W9b

0.67 3.89 4.68 1.86 3.57 5.43 1.95 6.57 833 1.82 5.31 7.13 2.37 4.91 7.28 3.14 2.93 6.W

aw Cdim AEim c w- c

0.016 3.75 0.18 0.016 392 0.18 0.016 5.50 0.18 0.016 5.67 0.18 0.016 5.85 0.18 0.016 5.50 0.18

20% demase d AEim badins vrlw ( m S F ) aw C d h AEim c w w c

0.016 3.7s 0.12 0.016 3.92 0.12 0.016 5.W 0.12 0.016 5.67 0.12 0.016 5.85 0.12 0.016 5.50 0.12

g , ( s ~ % F s ~ % ry s o o o ô o

2 0 % ~ d k e ~ i n e v a l w ( S F ) inonO, SPM SQ(%) CR SEF IR(at lRom lRim Cdw r Cdocn Mon, L.+0 Uw &m Cdm

mon % mguv % molhg wPuVW' 'g uOll c ugl- c c W~W*oOlOdw A 5.02 3 15.0 4 68.6 35.0 33.6 0.021 0.016 1.5 0.38 0.116 0.70 2.75 3.4s

aw Cdim AEim c W9dM c

0.016 3.75 0.15 0.016 3.92 0.15 0.016 5.50 0.15 0.016 S.67 0.15 0.016 5.85 0.15 0.018 5.50 0.15

Uw Uim Cd,,,, w* ugl* wsan 0.70 5.16 S.8S 1.82 4.62 6.U 2.02 8.50 10.52 1 . 6.08 8.75 2.46 6.35 8.a 3.24 3.79 7.01

20% inmeam d g W i m value (SF) WU a%) CR SW IRta lRom lRim Cdw msn % mgw % m 9 m 9 m 9 * 5.02 3 15.0 4 6 68.6 35.0 33.6 0.û21 4.11 4 15.0 4 4 59.0 27.0 32.0 0.058 5.W 5 15.0 -37 78.3 33.1 45.2 0.061 4.57 8 15.0 -14 63.9 9.9 54.0 0.057 7 . n 18 10.5 24 62.6 19.4 43.2 0.108 4.57 23 10.5 49 43.3 23.3 20.0 0.140

ü w Um Cd, w* w* w* 0.58 4.27 4.m 1.59 3.83 SA2 1.67 7.W 8.?2 1.56 5.68 7.25 2.04 5.26 7.28 2.69 3.14 S.ô3

2û% deeinIae d g badine vdue (SF)