Ecology and impact of the exotic amphipod, Corophium curvispinum Sars, 1895 (Crustacea: Amphipoda), in the River Rhine and Meuse
S. Rajagopal, G. van der Velde, B.G.P. Paffen and A. bij de Vaate
Reports ofth e project "Ecological Rehabilitation of Rivers Rhine and Meuse" No.75-1998
Institute for Inland Water Management and Waste Water Treatment (RIZA), P.O. Box 17, 8200 AA Lelystad, The Netherlands To be referred to as:
Rajagopal, S., G. van der Velde, B.G.P. Paffen & A. bij de Vaate, 1997. Ecology and impact of exotic amphipod, Corophiumcurvispinum Sars , 1895 (Crustacea: Amphipoda), in the River Rhine and Meuse. Report (No. ) of the project "EcologicalRehabilitation of RiversRhine and Meuse" {with abstracts in Dutch, French and German). Institute for Inland Water Management and Waste Water Treatment (RIZA), P.O. Box 17, 8200 AA Lelystad, The Netherlands. Contents
Preface I
Summary III Samenvatting VII Résumé XI Zusammenfassung XV Listo ffigure s XIX List oftable s XXIII
1. Introduction 1
1.1. Distribution and range extensiono f Corophium curvispinum 1 1.2.Reason sfo r the present study 1 1.3. Objectives 3
2. Materials andmethod s 3
2.1. Study area 3 2.2. Methods 5 2.2.1. Life history andreproductiv e biology 5 2.2.2. Growth rates 6 2.2.3. Production 7 2.2.4. Distribution andimpact s of C. curvispinum onothe r macroinvertebrates 7 2.2.5. Mud-fixation 7 2.2.6. Filtration capacity 9 2.2.7. Hydrographie parameters 10 2.2.8. Statistical analysis 10
3. Results 10
3.1. Population density 10 3.2. Population structure 13 3.3.Se x ratio 13 3.4. Brooddevelopmen t 13 3.5. Breeding season 18 3.6. Broodsiz e 18 3.7. Losso fembryo sfro mth ebroo dpouc h 24 3.8. Relationship betweenbod y lengthan dweigh t 25 3.9. Growthrat e 25 3.10.Productio n 28 3.11. Mud-fixation 34 3.11.1. Onth e stones 34 3.11.2.O nexperimenta ltile s 38 3.12. Filtrationcapacit y 51 3.13. Seasonalvariation s inpopulatio n densities of macroinvertebrates inth e Lower Rhine 53 3.14. Distribution and impacto f C. curvispinum onothe r macroinvertebrates 62
4. Discussion 70
5. Conclusions and Recommendations 81
References 83 Appendix 89 Preface
This project was financially supported by the Ministry of Transport and Public Works, the Ministry of Housing, Physical Planning and the Environment, Netherlands Organisation of the Advancemento fPur eResearc h(BION )an dBeijerinck-Poppin g Foundation.Th eresearc hcommitte e ofthi sprojec tconsiste do f Dr.Ir .G.M .va nDij k(RIVM) ,Mr .A .bi jd eVaat e(RWS/RIZA ) (Chairman), Drs. F.M.J. Oosterbroek (RWS), Dr. F.W.B, van den Brink (KUN) and Prof. Dr. G. van der Velde (KUN).W e are grateful to Prof.J.M .va n Groenendael and Mr. B. Kelleher for useful discussions. Thanks are due to M.J.E. Orbons, M.G. Versteeg and students for their assistance inth efiel d and laboratory. Summary
Exoticspecie sinvasion sca nb eviewe da sspecia lcase so frang eextensio n(Hengeveld ,1989) .
Thesespecie sar eofte n ecologically andphyleticall yver ydifferen tfro mnativ etaxa .Therefore ,the y
may establish or spread due to different life history characteristics and exploiting resources
differently from nativeorganism s ina disturbe decosystem s (e.g. humanactivities) .Th e invasion of
exotic speciesca n cause drastic changes in ecosystems (Drake eta/. , 1989;Pinkste r etai, 1992;
Dick etal., 1993 ; Nichols & Hopkins, 1993; Stewart & Haynes, 1994; Fahnenstiel etat., 1995). In
recent years, large numbers of exotic species have invaded the River Rhine (Den Hartog etal.,
1992;Va nde nBrin k etai, 1993),however , Corophium curvispinumSars ,ha sbee nsuccessfu lan d
dominatesth eepilithi c communities. Inrespons et othi ssuccess ,a stud y wascarrie d outt o assess
ecology and impacto f C.curvispinum inth e River Rhinean d Meuse,fro m March 1992t o February
1994.
Inth eLowe r Rhine,th elife-cycl eo f C. curvispinumi sbase do nthre egeneration spe ryear .Th e breedingseaso no f C.curvispinum occurre dbetwee nMa yan dOctobe ran dwa sstrongl y correlated with watertemperature . Reproduction generally began inMarc h andwa swel lestablishe d by May.
The overwintering generation died during June/July, but reproduction continued until October as a resulto fbreedin gb ysumme rgeneratio nindividuals .Dat ao nth ese xrati oo fC. curvispinumindicat e thatfemale s(6 0t o80% )wer emor eabundan ttha nmales .A lo wpercentag eo fmale swa sobserve d during May/June.Th e meanbroo dsiz e (25± 5egg s(mea n± SD),rang e= 12-38a t Nijmegen and
18± 3 eggs, range = 10-24 at Lobith) of C.curvispinum inth e Lower Rhine, is one of the highest ever recorded and showed a positive correlation with chlorophyll-a. Like many other crustaceans, a linear relationship exists between the body length and brood size of C. curvispinum. The percentage egg loss from the brood pouch of C. curvispinumi s high when compared with other
Corophiumspecies .
On stone surfaces, the amount of fixed muddy material including all macroinvertebrates was
38 -1044 g m"2 (dry weight) and 7 -138 g m'2 (ash-free dry weight). On experimental tiles, the amount of mud material,excludin g macroinvertebrates, was found to be 0.3 -16.3 g m"2 (ash-free dryweigh t onmonthl y exposedtiles ) and0. 3 -101.8g m" 2(ash-fre e dryweigh to ncumulativ etiles) .
A seasonal pattern was observed inth e amount of mud material fixedo n the stones of groins and experimental tiles.Th e amounto f mudmateria lo nth estone s ofgroin san dexperimenta l tiles inth e
Lower Rhine was highest during the summer period. A correlation was established between IV population densities of C. curvispinuman d muddy material fixed on the stones and on tiles (dry weight and ash-free dry weight). For example, at Lobith the correlation formula was y = 228.27 +
0.003x (r = 0.74, P < 0.001) for the dry weight of mud in gram per square metre (x is the number of individuals per square metre). For the ash-free dry weight of the muddy tubes, the correlation formulawa sy = 30.5 6+ 0.0006 x (r = 0.84 , P< 0.001) .Th epercentag eo forgani c matter calculated from the dry weight and ash-free dry weight of mud material including macroinvertebrates on the stones rangedfro m 9%t o 23%(mea n± SD; 15.9± 3.7%) at Lobith andfro m 13%t o 20%(16. 0 ±
2.4%) at Nijmegen. No significant variation inthes e values existed between Lobith andNijmegen .
Incomparison ,th e percentage of organic matter of the mud material excluding macroinvertebrates ontile s at Weurt (River Waal) ranged from 11% to 29%(21. 4± 4.3%). Mean values of the rateo f mud fixation per individual calculated from the cumulative density of C.curvispinum an d amounts of mudfixe dcumulativel yo nth estone sa tLobit han dNijmege n indicatea th econstan t rateo fabou t
2.5 u,gindividual' 1day" 1( n= 1,700,000)afte r 400days .Almos t the samevalu e wasfoun d atWeur t
(2.3 u,gindividual" 1 day"1; n = 500,000) after 350days .
A tentative estimation of filtration by C. curvispinum indicates a possible mean filtration capacity of 5 x 106cm 3 m"2day' 1 in the River Rhine.Th e mean filtration rate of C. curvispinumi n
River Meusewa s estimated as0. 3 x 106cm 3m" 2day" 1.Th e variation inth efilterin g potential of C. curvispinumi nth e River Rhine and Meuse isdu e to the much higher population densities in River
Rhine compared to the Meuse. However, since there are no data available on the variation in filtration rate of C.curvispinum i n the literature, a thorough investigation is necessary to calculate the actual filtering impact of this filter feeder.
The population densities ofvariou s macroinvertebrates were studiedb ycollectin gstone sa t2 1 locations during September 1992 and September 1993 in the Rivers Rhine and Meuse. C. curvispinumwa s the most dominant macroinvertebrate, outnumbering all other species by many orders of magnitude. A maximum C.curvispinum densit y of 642,000 individuals m"2wa s recorded at De Steeg inth e River Ussel. C.curvispinum densities inth e River Rhine branches (River Waal,
R. Nederrijn/Lek and R. Ijssel) showeda positiv e correlationwit haverag e stream velocities. Higher population densities of C.curvispinum were recorded in the River Rhine than in the River Meuse.
The relatively higher salinity, water temperatures, stream velocities andioni c content (e.g.Sodium ) inth e River Rhine,resultin gfro m industrialdischarge san d mining activities,hav econtribute dt oth e enormous success of C.curvispinum inth e Rhine.Th e success of C.curvispinum isals o relatedt o its strategy in competing for settlement space by means of muddy tubes. The very high densities V of this opportunistic filter feeder had shown an enormous impact on other macroinvertebrates, especially Dreissenapolymorpha (Pallas) ;whic hha sdrasticall ydecline dsinc eth emas scolonisatio n of C. curvispinum.Th e possible reasons are discussed by comparing population densities of D. polymorphabefor e (September 1989) andafte r (September 1993)th e explosive population growth of C.curvispinum alon g the River Rhine.
Despite the importance of the population explosion of C.curvispinum inth e River Rhine,ver y little is known about the fundamental features of filtration rate and tube building activity of this species. The success of invasive species is mostly due to their plasticity in response to the characteristics of invaded ecosystem. For example, filter feeding exotics may succeed if they can easily adapt their feeding response and haveth e capacity to switch to alternative sources of food.
In order to predict future developments, much more knowledge must be gathered on the ecophysiology (e.g.feedin gbehaviour ,selectio no fparticl esize ,shape ,weight ,composition ,surfac e texture, adhesive properties etc) and impact (e.g.filtratio n impact, trophic structure, energy flow, species composition etc.) of C. curvispinum in the River Rhine. Samenvatting VII
Invasies door exotische soorten kunnen worden beschouwd als speciale gevallen van
gebiedsuitbreiding (Hengeveld, 1989).D ebetreffend e soortenverschille ndikwijl s inbelangrijk e mateva n
de oorspronkelijk inheemse taxa, zowel in morfologisch opzicht als in oecologische amplitude. Deze verschillen dragen in positieve zin bij aan de vestiging of de verdere verspreiding. Belangrijk is ook dat
nieuwevestigingsmogelijkhede nontstaa ndoo rmenselijk eactiviteite nwaardoo roecosysteme n(zoal so.a .
deRij ne nd eMaas )verstoor dzij ngeraakt .Omda texote nt.o.v . deinheems esoorte nander e levenscycli
dan wel een afwijkende manier kunnen hebben van het exploiteren van de voedselbronnen zijn ze in
staat deze nieuwe vestigingsmogelijkheden te benutten. Uit literatuurgegevens is bekend dat op
oecosysteemniveau de invasie van exotische soorten drastische veranderingen kan veroorzaken. Dat
geldti nzeker emat eoo kvoo rd eRijn .Gedurend ed elaatst edecenni ai sdez erivie rdoo ree ngroo taanta l
uitheemse soorten met succes gekoloniseerd. Met name de Kaspische slijkgarnaal, Corophium
curvispinum Sars, bleek zeer succesvol en domineerde binnen enkele jaren de epilithische
levensgemeenschappen van ongewervelde dieren. Deze succesvolle kolonisatie was de reden om van
maart 1992to tfebruar i 1994 een studie te verricht naar de oecologie van C.curvispinum i n de rivieren
Rijn en Maas en de invloed van de soort op bovengenoemde levensgemeenschappen.
Ind e Rijn in Nederland isd elevenscyclu s van C. curvispinumgebaseer do p drie generaties per jaar. Het voortplantingsseizoen loopt van mei tot oktober en is sterk gecorreleerd met de watertemperatuur. De eiproductie en de groei begon in het algemeen in maart en was in mei goed op gang gekomen. De overwinterende generatie was weliswaar in juni/juli niet meer in de bemonsterde populaties aantoonbaar, maar de voortplanting werd voortgezet door de zomergeneraties die inmiddels geslachtsrijpware ngeworden .Ui tgegeven sove rd esexrati ova n C.curvispinum blijk tda twijfje s talrijker waren dan mannetjes (de verhouding man/vrouw varieerde gedurende het grootste deel van het jaar tussen 0,6 en0,8) . Een zeer laag percentage mannetjes werdwaargenome n ind e periode mei/juni. De gemiddelde broedgroottes van 25,± 5 eieren (gemiddelde, + SD; range: 12-38) per wijfje bij Nijmegen en 18± 3eiere n(range : 10-24) bijLobit hva n C.curvispinum, behorento td ehoogst eooi t ind e literatuur vermeld en vertoonden een positieve correlatie met de chlorophyll-a concentratie in het water. Het percentage verlies van eieren in de broedzak van C. curvispinum is hoog vergeleken met andere
Corop/7/um-soorten. Zoals bij vele andere kreeftachtigen bestaat er een lineair verband tussen de lichaamslengte en broedgrootte van C. curvispinum.
Een belangrijke eigenschap van de Kaspische slijkgarnaal is dat deze slibdeeltjes uit de waterkolombenu tvoo rhe tvervaardige nva nwoonbuisjes ,bi jvoorkeu ro phard evoorwerpe nzoal sstene n VIII langs de voet van de kribben. Op dergelijke stenen varieerde de hoeveelheid vastgelegd slib (inclusief alle aanwezige macro-evertebraten) van3 8to t 1044g.m" 2 opbasi s van drooggewicht en7 to t 138 g.m"2 op basis van asvrij drooggewicht. Op tegels die gedurende een maand in het water werden gehangen
(locatie Weurt) varieerde de hoeveelheid slib,exclusie f macro-evertebraten,va n0, 3to t 16,3g.m" 2(asvri j drooggewicht). Bijtegel s die gedurende een langere periode in het water waren gehangen was dat 0,3 tot 101,8 g.m"2 (asvrij drooggewicht).
In de hoeveelheid slibmateriaal die op de stenen aan de voet van de kribben en op de uitgehangen tegels was vastgelegd bleek een duidelijk seizoenspatroon zichtbaar. De hoeveelheid was het hoogst gedurende de zomerperiode. Er kon tevens een duidelijke correlatie worden vastgesteld tussen de dichtheid van C. curvispinume n de hoeveelheid slibmateriaal op de stenen en de tegels. In de Rijn bij Lobith was de correlatieformule: y = 228,27 + 0,003x (r = 0,74; P < 0,001) voor het drooggewichtva n hetsli b ingramme n perm 2( xi she taanta lindividue nva n C. curvispinumpe rm 2). Voor het asvrij droogewicht was de correlatieformule y = 30,56 + 0,0006x ( r = 0,84; P < 0,001). Het percentage organische stof van hetsli b (inclusief de macro-evertebraten) opd e stenen,bereken d uitd e bepaling van hetdrooggewich t en het asvrij drooggewicht, varieerde van 9to t2 3 %(gemiddel d 15,9% ;
SD ± 3,7 %) bij Lobith en van 13 tot 20 % (16,0 %; ± 2,4 %) bij Nijmegen. Er was geen significant verschil aantoonbaar tussend ewaarde nva nbeid e locaties.He tpercentag e organisch materiaalva nhe t slib (exclusief macro-evertebraten) op de tegels bedroeg 11 tot 29 % (21,4 ± 4,3 %). Gemiddelde waarden van de slibfixatiesnelheid per individu berekend uit de verhouding tussen de cumulatieve dichtheid van C. curvispinume n de cumulatieve slibhoeveelheden die waren vastgelegd op de stenen bij Lobith en Nijmegen indiceren eenconstant e fixatiesnelheid vanongevee r 2,5 mgasvri j drooggewicht per individu per dag over een periode van 400 dagen. Bijna dezelfde waarde werd gevonden voor de uitgehangen tegels bijWeur t (2,3 mgasvri j drooggewicht per individu per dagove r een periode van 350 dagen).
Een voorlopige schatting van de filtratie door C. curvispinum, overigens gebaseerd op literatuurgegevens voor andere Coroph/'um-soorten, laat een mogelijke filtratiecapaciteit van de totale populatie zienva n gemiddeld 5m 3 per m2geschikt e habitat perda g ind e Rijn.Voo r de Maaswer ddez e geschat op0, 3 m3pe r m2pe r dag. Hetverschi li nfiltratiecapacitei ttuse n deRij ne nd e Maas iste danken aan de veel hogere populatiedichtheden in de Rijn vergeleken met die in de Maas. Aangezien er ook geen literatuurgegevens beschikbaar zijn over o.a. effecten van de deeltjessamenstelling, de dichtheid van de zwevende stof en van het seizoen op de filtratiessnelheid van C. curvispinum was het niet mogelijk bijd e berekening van defiltratiecapacitei t vandez efilter-feede r daar rekening meete houden. IX
Effecten van de kolonisatie van C.curvispinum o p de epilithische levensgemeenschappen van ongewervelde dieren werden bestudeerd door in de Rijne n de Maas gedurende de periode september
1992- septembe r 1993o p2 1plaatse nstene nte bemonsteren . C.curvispinum wa sd emees t dominante macro-evertebraat die alle andere soorten qua aantallen verre overtrof. Een maximum dichtheid van C. curvispinumva n642.00 0individue nm" 2wer dgevonde nbi jD eStee gi nd erivie r deIJssel .D edichthede n van C.curvispinum i nd e Rijntakken (Waal,Nederrijn/Le ke n IJssel) vertoonden een positieve correlatie met gemiddelde stroomsnelheden. Vergeleken met de Maaswerde n in de Rijn veel hogere dichtheden van C.curvispinum aangetroffen. Derelatie f hogere zoutconcentratie,watertemperatuur , stroomsnelheid en ionengehalte (o.a. natrium en chloride) als resultaat van industriële lozingen en mijnaktiviteiten in de
Rijn moeten in belangrijke mate hebben bijgedragen tot het kolonisatiesucces van C.curvispinum i nd e
Rijn. Ditsucce si soo kgerelateer daa nd estrategi eo mvestigingsplaatse nte bedekke n metsli bvanweg e debou wva nmodderig ewoonbuisjes .He tresultaa ti sda tdaardoo rd eoverig eepilithisch e soortenworde n weggeconcurreerd.Teven sword tdoo rd ebou wva nwoonbuisje sd evestigin gva nee nsessiel ealgenflor a onmogelijk zodat de voedselbron voor herbivore soorten niet meer tot ontwikkeling kan komen.
Eri snauwelijk s ietsbeken dove rbasal eactiviteite nva n C. curvispinumdi eva nbelan gzij ni nhe t waterbeheer, zoals de filtratiesnelheid en de manier waarop woonbuisjes worden gebouwd. Beide activiteiten zijn van belang omdat daardoor zwevend materiaal uit de waterkolom wordt verwijderd waardoor de helderheid toeneemt. Daarnaast dient men zichte realiseren dat hetsucce s van invasieve soorten meestal te danken is aan hun plasticiteit met betrekking tot de eigenschappen van het geïnvadeerde oecosysteem. Bijvoorbeeld, exoten, die filtreren zouden meer succes hebben als ze gemakkelijk hun voedingsstrategie kunnen aanpassen en de mogelijkheid hebben uit te wijken naar alternatieve voedselbronnen. Ten einde toekomstige ontwikkelingen te voorspellen is ook kennis
noodzakelijk overd eoecofysiologi e engedra g(bijv .voedingsgedrag ,selecti eva npartikelgrootte ,-vorm ,
gewicht, samenstelling, oppervlakte textuur, adhesieve eigenschappen etc.) en de invloed van hun
aktiviteit in het oecosysteem (bijv. plaats in de trofische structuur, invloed op energiestromen, effecten
op de soortensamenstelling van de gemeenschap etc). Résumé XI
Les invasions par des espèces exotiques peuvent être considérées comme des cas spéciaux
d'extension du territoire (Hengeveld, 1989). Les espèces en question diffèrent souvent dans une large
mesure des taxons indigènes originaux,tan t du point devu e morphologique que decelu i de l'amplitude
écologique. Ces différences contribuent positivement à l'établissement ou à l'élargissement ultérieur. Il
importe également qu'il se crée de nouvelles possibilités d'établissement par le biais des activités
humaines qui ont perturbé les écosystèmes (comme e.a. le Rhin et la Meuse). Etant donné que les
exotes peuvent avoir par rapport aux autres espèces indigènes d'autres cycles de vie ou une manière
différente d'exploiter les sources d'aliments, ils sont à même d'utiliser ces nouvelles possibilités
d'établissement.
L'on sait par des informations bibliographiques qu'au niveau des écosystèmes, l'invasion
d'espèces exotiques peut occasionner des changements radicaux. Cela s'applique aussi dans une certaine mesure au Rhin. Pendantce s dernières décennies,c e cours d'eau a étécolonis é avec succès
parde s espècesexotiques . Notamment lacrevett e Caspienne desvases, Corophiumcurvispinum Sars , s'est avérée être un grand succès et a dominé en quelques années les biocénoses épilithiques des animauxinvertébrés .Cett ecolonisatio nréussi ea donn élie uà l aréalisation ,d emar s199 2à févrie r1994 , d'une étude sur l'écologie de l'amphipode C.curvispinum dan s lescour s d'eau du Rhin et de la Meuse et sur l'influence de cette espèce sur les biocénoses susmentionnées.
Lecycl ed evi e de C.curvispinum dan s le Rhinau x Pays-Bas se base surtroi s générations par an. La saison de reproduction court de mai à octobre et est étroitement liée à latempératur e de l'eau.
Laproductio nd el'oeu fe tl acroissance ,qu ion tgénéralemen tcommenc ée nmars ,étaien tbie navancée s enmai .Quoiqu e lagénératio nqu ia pass él'hive r nefû tplu sdécelabl e enjuin/juille t dans lespopulation s
échantillonnées, la reproduction a été poursuivie par les générations d'été qui entre-temps étaient arrivéesà maturité .I lressor tde sdonnée ssu rl arati osexuell ed e C.curvispinum qu ele sfemelle s étaient plus nombreuses que les mâles (laproportio n mâle/femelle oscillait pendant lamajeur e partied e l'année entre 0,6 et 0,8). Un très faible pourcentage de mâles a été observé dans la période mai/juin. Les grandeurs moyennes decouvé ed e2 5± 5oeuf s (moyenne,± SD;range :12-38 )pa rfemell e à Nimègue et 18± 3 oeufs (range: 10-24) à Lobith de C.curvispinum, sont parmi les plus élevées qui aient jamais
été mentionnées dans les ouvrages et ont montré une corrélation positive avec la concentration de chlorophylle-adan s l'eau. Lepourcentag e depert ed'oeuf s dansl apoch ed'incubatio n de C.curvispinum est élevé en comparaison d'autres espèces de Corophium.I l existe un rapport linéaire, comme pour beaucoup d'autres crustacés, entre la longueur du corps et lagrandeu r de couvée de C.curvispinum. XII
Une caractéristique importante de la crevette Caspienne des vases est qu'elle utilise des
particules de vase de la colonne d'eau pour lafabricatio n de petits conduits d'habitation, de préférence
sur des objets durs comme les pierres au pied des épis. La quantité de vase fixée sur ces pierres (y compris tous lesmacro-invertébré s présents) avari éd e3 8à 1044g.m" 2su r labas ed upoid sd e matières
sèches et de7 à 138g.m" 2su r labas e dupoid s de matières sèches sanscendres .Su r lesdalle s quion t
été suspendues dans l'eau pendant une période d'un mois (site de Weurt) la quantité de vase, macro
invertébrés exclusivement, a varié de 0,3 à 16,3 g.m"2 (poids de matières sèches sans cendres). Pour
les dalles qui avaient été suspendues dans l'eau pendant une période plus longue ce pourcentage était
de 0,3 à 101,8 g.m"2 (poids de matières sèches sans cendres).
Dans la quantité de vase qui était fixée sur les pierres au pied des épis et sur les dalles
suspendues, il s'est avéré qu'un schéma saisonnier était clairement visible. La quantité était la plus
élevée pendant la période estivale.O n ae n outre puétabli r unecorrélatio n claire entre ladensit é de C.
curvispinum et laquantit é de boue setrouvan t sur les pierres et dalles. Dans le Rhin à Lobith laformul e
de corrélation était :y = 228,27 + 0,003x (r = 0,74; P < 0,001) pour le poids de matières sèches de la
boue en grammes par m2 (x est le nombre d'individus de C. curvispinum par m2). Pour le poids de
matières sèchessan scendre s laformul e decorrélatio n était :y =30,5 6 +0,0006 x (r = 0,84 ; P< 0,001).
Lepourcentag ed ematièr eorganiqu ed el avas e( ycompri sle smacro-invertébrés ) surle spierres ,calcul é
à partir de ladéterminatio n du poids de matières sèches et du poids de matières sèches sans cendres,
a varié de 9 à 23 % (en moyenne 15,9 %; SD ± 3,7 %) à Lobith et de 13à 20 % (16,0 %; ± 2,4 %) à
Nimègue. On n'a pas décelé de différence notoire entre les deux lieux. A titre de comparaison de ces
données, le pourcentage de matières organiques de lavas e (macro-invertébrés exclusivement) sur les
dalles était de 11 à2 9 % (21,4 %; ± 4,3%) .
Les valeurs moyennes de la vitesse de fixation de la vase par individu, calculées à partir du
rapport entre la densité cumulative de C. curvispinume t les quantités cumulatives de vase qui étaient
fixées sur les pierres à Lobith et Nimègue, indiquent une vitesse defixatio n constante d'environ 2,5 mg
poids de matières sèches sanscendre s par individu parjou r sur une période de40 0jours . Pratiquement
les mêmes valeurs ont été constatées pour les dalles suspendues à Weurt (2,3 mg poids de matières
sèches sans cendres par individu parjou r sur une période de 350jours) .
Uneévaluatio n provisoire de lafiltratio n par C.curvispinum, par ailleurs basée surde s données
bibliographiques pour d'autres espèces de Corophium, montre une capacité de filtration possible de la
population totale d'en moyenne 5 m3 par m2 d'habitat approprié par jour dans le Rhin. Pour la Meuse
celle-ci a été estimée à 0,3 m3 par m2pa r jour. La différence de capacité defiltratio n entre le Rhin et la
Meuse est due aux densités de populations bien plus élevées dans le Rhine ncomparaiso n avec celles XIII del aMeuse .V uqu'o nn edispos epa sno nplu sd'information s bibliographiques entreautre ssu rle seffet s del acompositio n des particules,d el adensit é des matières ensuspensio n etd e lasaiso n sur lavitess e defiltration de C.curvispinum, i l n'a pasét é possible d'en tenir compte dans le calcul de lacapacit é de filtration de ce filter-feeder.
Les effets de la colonisation de C. curvispinum sur les biocénoses épilithiques d'animaux invertébrés ont été étudiés en échantillonnant des pierres dans le Rhin et la Meuse pendant la période de septembre 1992 - septembre 1993 à 21 endroits. C. curvispinumétai t le macro-invertébré le plus dominant, qui a dépassé de loin toutes les espèces quant aux quantités. Une densité maximum de C. curvispinum de 642.000 individus m'2 a été trouvée à De Steeg dans l'IJssel. Les densités de C. curvispinumdan sle sbranche sd uRhi n(Waal ,Nederrijn/Le ke tUssel )on tmontr éun ecorrélatio n positive avec lesvitesse sd'écoulemen t moyennes.E ncomparaiso nave cl aMeuse ,o na trouv édan sl eRhi nde s densitésd e C. curvispinumbie nplu sélevées .L asalinité ,l atempératur ed el'eau ,l avitess e d'écoulement et lateneu r ionique relativement plusélevé s (e.a.e nsodiu m etchlorure ) comme résultatde s décharges industrielles et des activités minières dans le Rhin,on t dû contribué dans une large mesure au succès de la colonisation de C.curvispinum dan s le Rhin.C e succès est aussi lié à la stratégie qui consiste à recouvrir des lieux defixatio nave c del avase ,à raisond el aconstructio n depetit sconduit s d'habitation en vase. De ce fait, les autres espèces épilithiques sont éliminées par la voie de la concurrence.
Parallèlement, la construction de petits conduits d'habitation rend impossible la fixation d'une flore d'algues sessiles de sorte que la source d'aliments pour les espèces herbivores ne peut plus se développer.
Onsai ttrè spe ud echose ssu rle sactivité sbasale sd el acrevett e C. curvispinum, lesquellesson t importantes dans la gestion des eaux, telles que la vitesse de filtration et la manière dont les petits conduits d'habitation sont construits. Ces deux activités sont importantes puisqu'elles provoquent la disparition des matériaux en suspension de la colonne d'eau et qu'il en résulte une plus grande clarté de l'eau. De plus, il faut avoir conscience que le succès des espèces invasives est généralement dû à leurplasticit é relativement auxcaractéristique s del'écosystèm e envahi.A titr ed'exemple , lesexote squ i filtrent auraient davantage de succès si elles pouvaient plus facilement adapter leur stratégie d'alimentation et qu'elles eussent la possibilité de se rabattre sur des sources d'aliments alternatives.
Pour prévoir les développements futurs on a aussi besoin de connaissances sur l'écophysiologie et le comportement (parex .l ecomportemen t parrappor tau xaliments ,l asélectio n desgrandeu r etform e des particules, lepoids , lacomposition , lasurfac e detexture ,le s propriétés d'adhérence, etc.) et l'influence de leur activité dans l'écosystème (par ex. la place dans la structure trophique, l'influence sur les flux
énergétiques, les effets sur lacompositio n des espèces de la biocénose, etc.). XV Zusammenfassung
Invasionen exotischer Arten sind als Sonderfälle der Gebietserweiterung zu betrachten
(Hengeveld, 1989). Die betreffenden Arten unterscheiden sich oft weitgehend von den ursprünglichen einheimischen Taxa, sowohl in morphologischer Hinsicht als in ökologischer Amplitude. Diese
Unterschiede wirken positiv auf die Ansiedlung oder dieweiter e Verbreitung aus.Wichti g ist auch,das s neue Siedlungsmöglichkeiten durch menschliche Aktivitäten entstehen,wodurc h Ökosysteme (wie u.a. der Rhein unddi e Maas) gestörtworde n sind.Wei l Exoten in Bezugau f dieeinheimische n Arten andere
Lebenszyklen oder eine abweichende Nutzungsweise der Nahrungsquellen haben können, sind sie imstande, diese neuen Siedlungsmöglichkeiten zunutzen .
Aus Literaturangaben ist bekannt, dass inBezu g auf Ökosysteme die Invasion exotischer Arten grundlegende Veränderungen verursachen kann. Dies gilt gewissermaßen auch für den Rhein. In den letztenJahrzehnte n hat eineVielzah lexotische r Artendiese n Flusserfolgreic h besiedelt.Namentlic h der
Schlickkrebs, Corophium curvispinumSars , ergab sich als sehr erfolgreich und dominierte in einigen
Jahren die epilithischen Lebensgemeinschaften der Wirbellosen. Diese erfolgreiche Besiedlung war der
Grund zur Durchführung einer Studie im Zeitraum März 1992 bis Februar 1994 nach der Ökologie von
C.curvispinum i nde nFlüsse n Rhein undMaa sun dnac hde m Einflussdiese r Artau fdi e obenerwähnten
Lebensgemeinschaften.
Im niederländischen Rhein basiertde r Lebenszyclus von C.curvispinum au f drei Generationen pro Jahr. Die Fortpflanzungszeit erstreckt sich von Mai bis Oktober und ist stark korreliert mit der
Wassertemperatur. Die Eiproduktion und das Wachstum fingen im allgemeinen im März an und waren im Mai gut in Gang gekommen. Die überwinternde Generation war zwar in Juni/Juli nicht mehr in den
Bestandsaufnahmen nachweisbar, aber die Sommergenerationen, die mittlerweile laichreif geworden waren, setztendi e Fortpflanzungfort .Au sAngabe n zur Sexratevo n C.curvispinum ergib t sich, dassdi e
Weibchen zahlreicher als die Männchenware n (dasVerhältni s MannA/Veibvariiert e im größtenTei l des
Jahres zwischen 0,6 und 0,8). In der Periode Mai/Juni wurde ein sehr niedriger Anteil an Männchen beobachtet. Die mittleren Brutgrößen von 25,± 5 Eier (Mittelwert, ± SA; Bereich: 12-38) pro Weibchen in Nimwegen und 18,± 3 Eier (Bereich: 10-24) in Lobith von C.curvispinum gehöre n zu den höchsten je in der Literatur erwähnt. Sie wiesen eine positive Korrelation mit der Chlorophyll-a Konzentration im
Wasser auf. Die Verlustrate der Eier in der Bruttasche von C. curvispinum ist hoch im Vergleich zu anderen Coroph/um-Arten. Zwischen der Körperlänge und Brutgröße von C. curvispinum besteht eine lineare Beziehung wie bei vielen anderen Krustazeen.
Einewichtig eEigenschaf tde sSchlickkrebse sist ,das se rSchlammpartikel nau sde rWassersäul e XVI fürdi e Erstellungvo nWohnröhre n verwendet,di ee rbevorzug t auf hartenGegenstände n wieSteine na m
Fuß der Buhnen baut. Auf solchen Steinen variierte die festgelegte Schlammmenge (inklusive aller vorhandenen Makroinvertebraten) von3 8bi s 1044g.m" 2au fde r Basisvo nTrockengewich t und7 bi s13 8 g.m"2au fde rBasi sde saschenfreie nTrockengewichts .Au f Platten,di ewähren deine sMonat sin sWasse r gehängt wurden (Stelle Weurt) variierte die Schlammmenge, exklusive Makroinvertebraten, von 0,3 bis
16,3g.m" 2 (aschenfreies Trockengewicht). Bei Platten,di e länger insWasse r gehängt wurden,wa r dies
0,3 bis 101,8 g.m"2(aschenfreie s Trockengewicht).
Aus der Menge an Schlammmaterial, die auf den Steinen am Fuß der Buhnen und auf den ausgehängten Plattenfestgeleg t wurde, istein e deutlichejahreszeitlich e Abhängigkeit ersichtlich. Inde r
Sommerperiode war die größte Menge vorhanden. Außerdem ließ sich eine deutliche Korrelation zwischen der Dichte von C.curvispinum un d der Menge an Schlammmaterial auf den Steinen und den
Platten feststellen. Im Rhein bei Lobith war die Korrelationsformel: y = 228,27 + 0,003x (r = 0,74; P <
0,001) für dasTrockengewich t des Schlamms in Grammen pro m2 (x ist die Zahl der Individuen von C. curvispinumpr om 2).Fü rda saschenfrei eTrockengewich twa rdi e Korrelationsformel y= 30,5 6+ 0,0006 x
(r = 0,84; P < 0,001). Der Prozentsatz an organischem Stoff des Schlamms (inklusive der
Makroinvertebraten) auf den Steinen, der aus der Bestimmung des Trockengewichts und des aschenfreien Trockengewichts berechnet wurde, variierte von 9 bis 23 % (durchschnittlich 15,9 %; SA
± 3,7 %) bei Lobith und von 13 bis 20 % (16,0 %; + 2,4 %) bei Nimwegen. Es war kein signifikanter
Unterschied zwischen beiden Stellen nachweisbar. Zum Vergleich dieser Daten, der Prozentsatz an organischem Materialde sSchlamm s (exklusive Makroinvertebraten) auf denPlatte nberechnet e sichau f
11 bis zu 29% (21,4%; ±4,3%).
Die aus dem Verhältnis zwischen der kumulativen Bestandsdichte von C.curvispinum un d den auf den Steinen bei Lobith und Nimwegen festgelegten kumulativen Schlammmengen berechneten
Mittelwerte der Schlammfixierungsgeschwindigkeit pro Individuum indizieren eine konstante
Fixierungsgeschwindigkeit von etwa 2,5 mg aschenfreies Trockengewicht pro Individuum pro Tag in einem Zeitraum von 400 Tagen. Fast derselbe Wert wurde für die ausgehängten Platten bei Weurt (2,3 mg aschenfreies Trockengewicht pro Individuum pro Tag in einem Zeitraum von 350 Tagen) erhalten.
Einevorläufig e Schätzungde r Filtrationdurc h C.curvispinum, dieübrigen sau f Literaturangaben fürander e Coroph/um-Artenbasiert ,zeig tein emöglich eFiltrationskapazitä tde rgesamte nPopulatio nvo n durchschnittlich 5 m3pr o m2geeignete s Habitat proTag im Rhein. Für die Maaswurd e diese Kapazität auf 0,3 m3pr oTag eingeschätzt . DerUnterschie d inFiltrationskapazitä t zwischen Rheinun dMaa s istau f dievie lhöhere nBestandsdichte n imRhei ni mVergleic hz udene ni nde rMaa szurückzuführen .Wei lauc h keine Literaturangaben vorliegen über u.a. die Auswirkungen der Partikelzusammensetzung, Dichte der XVII
Schwebstoffe und Jahreszeit auf die Filtrationsgeschwindigkeit von C. curvispinum war es bei der
Berechnung der Filtrationskapazität dieses Filtrierers nicht möglich, diese zu berücksichtigen.
Auswirkungen der Besiedlung von C.curvispinum au f die epilithischen Lebensgemeinschaften vonWirbellose n wurdenanhan dvo n Probenahmenvo n Steinena n2 1Stelle n im Rhein undi nde r Maas im Zeitraum September 1992 - September 1993 untersucht. C. curvispinum war der dominierende
Makroinvertebrat, der alle anderen Arten in zahlenmäßiger Hinsicht weit überstieg. Die höchste
Bestandsdichte von C.curvispinum vo n 642.000 Individuen m"2fan d sich in De Steeg in der Ussel. Die
Bestandsdichten von C.curvispinum i nde n Rheinzweigen (Waal,Nederrijn/Le k undUssel )wiese n eine positive Korrelation mit mittleren Fließgeschwindigkeiten auf. Verglichen mit der Maas wurden im Rhein viel höhere Bestandsdichten von C. curvispinum angetroffen. Die relativ höhere Salzkonzentration,
Wassertemperatur, Fließgeschwindigkeit und lonengehalt (u.a. Natrium und Chlorid) als Ergebnis industrieller Einleitungen und Bergwerksarbeiten im Rhein müssen wesentlich zum Siedlungserfolg von
C. curvispinum in diesem Fluss beigetragen haben. Dieser Erfolg hängt auch mit der Strategie zusammen,fü rde nBa uschlammige rWohnröhre nSiedlungsstelle nmi tSchlam mzu bedecken .Die sführt zumErgebnis ,das sdi eübrige nepilithische nArte nwegkonkurrier twerden .Außerde m machtde rBa uvo n
Wohnröhren die Besiedlung einer sesshaften Algenflora unmöglich, so dass die Nahrungsquelle für
Herbivoren nicht mehr zur Entwicklung kommen kann.
Esis tseh rweni g bekannt überbasal e Aktivitätenvo n C.curvispinum, dievo n Bedeutungfü r die
Gewässerbewirtschaftung sind, wie Filtrationsgeschwindigkeit und die Art undWeise , wie Wohnröhren gebaut werden. Beide Aktivitäten sindvo n Bedeutung,wei l dadurch Schwebstoff aus der Wassersäule entfernt wird,wodurc h das Wasser ungetrübter wird. Außerdem muss man sich bewusst sein, dass der
Erfolg invasiver Arten meistens auf ihre Plastizität in Bezug auf die Eigenschaften des invadierten
Ökosystemszurückzuführe nist .De rErfol gfiltrierende rExote nbeispielsweis ehäng tdavo nab ,o bsi eihr e
Nahrungsstrategie leicht anpassen können und die Möglichkeit haben, alternative Nahrungsquellen zu nutzen. Zur Vorhersage künftiger Entwicklungen sind auch Kenntnisse der Ökophysiologie und des
Verhaltens erforderlich (z.B. Nahrungsverhalten, Auswahl der Partikelgröße, -form, Gewicht,
Zusammensetzung, Texturfläche, adhäsive Eigenschaften etc.) und des Einflusses ihrer Aktivität im
Ökosystem (z.B.Standor t inde rtrophische n Struktur, Einfluss auf Energieströme,Auswirkunge n auf die
Artenzusammensetzung der Lebensgemeinschaft etc.). XIX List offigure s
Fig.1. Map showing the sampling locations in the Rivers Rhine and River Meuse. 1a. Mean population densities of Corophium curvispinum (individuals m'2 stone surface at 1 m depth) at different sampling locations in the River Rhine and River Meuse during September 1993. 1b.Mea n population densities of Dreissena polymorpha (individuals m'2ston e surface at 1 m depth) at different sampling locations inth e River Rhine, before (September 1989) and after (September 1992)th e population explosion of C. curvispinum.
Fig.2. Seasonal variations in the population densities (mean ± SE) of C. curvispinumfro m April 1992 to March 1994 on stones (n = 3-5) from the groins at Lobith and Nijmegen in the Lower Rhine.
Fig.3. Length-frequencyhistogram so f C.curvispinum fro mMarc h199 2t oFebruar y 1994a tLobit h in the Lower Rhine (the animals above 6 mm body length contributed only <0.2% during summer months).
Fig.4. Length-frequency histograms of C. curvispinumfro m March 1992 to February 1994 at Nijmegen inth e Lower Rhine (theanimal sabov e6 m mbod y lengthcontribute d only<0.2 % during summer months).
Fig.5. Seasonal variations in the sex ratio of C.curvispinum from March 1992t o February 1994 at Lobith and Nijmegen inth e Lower Rhine.
Fig.6. Developmental stages of egg embryos inth e brood pouch of C.curvispinum.
Fig.7. a. Seasonal variations in the hydrographie parameters (temperature and chlorophyll-a) in the Lower Rhine from March 1992 to February 1994. b andc . Relative occurrence (%) of ovigerousfemales , pre-ovigerous females, males and juveniles of C. curvispinum at Lobith and Nijmegen from March 1992t o February 1994.
Fig.8. The relationship between (a) the density of ovigerous females and temperature and (b) mean brood size of C.curvispinum andchlorophyll- a at Lobith and Nijmegen inth e Lower Rhine from March 1992 to February 1994.
Fig.9. Acompariso n of streamvelocitie s atLobit han dNijmege n usingWilcoxon-Signe d ranktest .
Fig.10. The relationship between length of ovigerous females and numbers of stage Iegg s in the brood pouch of C. curvispinuma t Lobith and Nijmegen inth e Lower Rhine.
Fig.11. a. The relationship between body length (excluding ovigerous females) and ash-free dry weight (AFDW) of C. curvispinumi nth e Lower Rhine. b. The relationship between body length of ovigerous females and ash-free dry weight (AFDW) of C. curvispinumi nth e Lower Rhine. XX
Fig.12. Mean growth rate (mm) of different generations of C. curvispinum at Lobith in the Lower Rhine.
Fig.13. Mean growth rate (mm) of different generations of C.curvispinum at Nijmegen inth e Lower Rhine.
Fig.14. Mean growth rate (mm) of C.curvispinum wit h (A) water temperature and (B) chlorophyll-a inth e Lower Rhine at Lobith.
Fig.15. Population densities of Corophiumcurvispinum an d ash-free dry weight (AFDW) of mud material (including macroinvertebrates) on stones at Lobith and Nijmegen from July 1992 to February 1994.
Fig.16. Population densities of Corophiumcurvispinum and mud material (dry weight and ash-free dry weight) on stones at Lobith and Nijmegen and ontile s at Weurt (* data represents the tiles exposed in a cumulative way at Weurt).
Fig.17. Population densities of Corophium curvispinum andash-fre e dryweight s of mudmateria l in mgindividual" 1day" 1o nstone s (including macroinvertebrates) at Lobith and Nijmegenfro m July 1992t o February 1994
Fig.18. Seasonalvariation s inth epopulatio ndensitie so f Corophium curvispinuman d mudmateria l on monthly exposedtile s at Weurt from November 1992t o October 1994.
Fig.19. Seasonalvariation si nth epopulatio ndensitie so f Corophiumcurvispinum an dmu dmateria l on tiles exposed in a cumulative way at Weurt from November 1992t o October 1994 (A - Series I; B - Series II).
Fig.20. Relationship betweenpopulatio ndensitie s of Corophiumcurvispinum an dmu dmateria l(dr y weight andash-fre e dryweight ) onmonthl y exposedtile s atWeur tfro m November 1992t o October 1994.
Fig.21. Seasonalvariation s inth e populationdensitie so f Corophium curvispinuman dmu dmateria l (ash-free dry weight in mg individual"1 day'1) on monthly exposed tiles at Weurt from November 1992 to October 1994.
Fig.22.Variation s inth e populationdensitie so f Corophium curvispinuman dmu dmateria l (ash-free dry weight in mg individual"1day" 1) on cumulatively exposedtile s at Weurt from November 1992 to October 1994 (A - Series I; B - Series II).
Fig.23. Change in population density andamoun t of mud material of C.curvispinum o nth e stones at Lobith and Nijmegen. 2 2 [Change in density AN (N m" )= (N,-Nt+1); Change in weight Aw (g m" ) = (wt-wl+1)]
Fig.24. Change in population density and amount of mud material of C. curvispinum on the cumulatively exposed tiles at Weurt.
2 2 [Change in density AN (N m" ) = (Nt-Nt+1); Change in weight Aw (g m" ) = (wt-wt+1)] XXI
Fig.25. Meanvalue s of mudfixatio n rate per individualcalculate dfro mth ecumulativ e density of C. curvispinuman dcumulativ e amounts of mudfixe d on the stones at Lobith and Nijmegen.
1 1 [Mud fixation rate (AFDW in |ig individual" day" ) = [î(wt+wM)/Ê(Nt+NH)]/Ê(dt+dM)] t=0 1=0 1=0
Fig.26. Meanvalue s of mudfixatio n rate per individualcalculate dfro mth ecumulativ e density of C. curvispinum and cumulative amounts of mud fixed on the cumulatively exposed tiles at Weurt.
1 1 [Mud fixation rate (AFDW in (xgindividual' day" ) = [Ê(wt+wt.1)/î(Nt+NM)]/Z(dt+dl1)] 1=0 t=0 t=0
Fig.27. Mean body length of C. curvispinum calculated from the cumulative monthly mean length and density of C.curvispinum onth e stones at Lobith andNijmegen .
[Mean body length (mm) = (E NxL)/(I N)] t=0 t=0
Fig.28.Th e percentage of organic matter calculatedfro mth e dryweigh t andash-fre e dryweigh to f mud material at Lobith and Nijmegen.
[Organic matter (%) = (AFDW/DW) x 100]
Fig.29. A comparison of organic matter (%) in the mud material on the stones at Lobith and Nijmegen using Wilcoxon-signed rank test.
Fig.30. Filtering impacto f Corophium curvispinumo nstone sa tvariou s places in Rivers Rhinean d RiverMeus e(refe rFig .1 forsamplin glocations) .Mea npopulatio ndensit yo fC. curvispinum was calculated from the five sampling depths (1 - 5 m at 1m interval;a t each depth three to five stones were collected) at each sampling location during September 1993 (water temperature 12°C) .
Fig.31. Seasonal variations in population densities (mean ± SE) of Corophiumcurvispinum an d Dreissenapolymorpha fro mApri l 1992t oMarc h 1994o nstone s (n= 3-5 )fro mgroin s inth e Lower Rhine at Lobith andNijmegen .
Fig.32. Seasonal variations in population densities (mean ± SE) of Gammaridae and Gastropoda from April 1992t o March 1994o nstone s( n= 3-5 ) from groins inth eLowe r Rhinea t Lobith and Nijmegen.
Fig.33.Seasona lvariation s inpopulatio ndensitie s(mea n± SE)o fdifferen t specieso f Gammaridae (Gammarus tigrinus and Echinogammarus ischnus) from April 1992 to March 1994 on stones (n = 3-5) from groins inth e Lower Rhine at Lobith and Nijmegen.
Fig.34. Seasonal variations in population densities (mean ± SE) of Chironomidae and Trichoptera from April 1992t o March 1994o nstone s (n= 3-5 ) from groins inth e Lower Rhine at Lobith and Nijmegen. XXII
Fig.35. Seasonal variations in population densities (mean ± SE) of Tricladida and Hirudinea from April 1992t o March 1994o nstone s (n= 3-5 ) from groins inth e Lower Rhine at Lobith and Nijmegen.
Fig.36. Seasonal variations in population densities (mean ± SE) of Corbicula from April 1992 to March 1994 on stones (n = 3-5) from groins inth e Lower Rhine at Lobith and Nijmegen.
Fig.37. Seasonal variations in population densities (mean ± SE) of different species of Corbicula {Corbicula fluminea and Corbiculafluminalis) from April 1992t o March 1994 on stones (n = 3-5) from the groins inth e Lower Rhine at Lobith and Nijmegen.
Fig.38.Seasona lvariation s inpopulatio ndensitie s(percentage ) ofvariou s macroinvertebratesfro m April 1992t o March 1994o nstone s (n= 3-5) from groins inth e Lower Rhine at Lobith and Nijmegen.
Fig.39. Variations in population density (percentage) of different macroinvertebrates on stones (n = 3-5) from the groins atdifferen t locations inth e River Rhine during September 1992an d September 1993.
Fig.40. Variations in population density (percentage) of different macroinvertebrates on stones (n = 3-5) from the groins at different locations inth e River Meuse in September 1993.
Fig.41. Population densities of C.curvispinum (individuals m"2ston e surface) and average stream velocities( msec' 1)alon gth ethre emajo rbranche so fRive r Rhine(R .Nederrijn/Lek , R.Waa l and R. Ussel) and River Meuse during September 1993.
Fig.42. Relationship between population densities of Corophium curvispinum and Dreissena polymorphaalon gwit haverag estrea mvelocitie sa tdifferen t locationsi nth eRive rRhin ean d River Meuse (refer Fig. 1fo r sampling locations). List oftable s XX111
Table 1.Seasona l variations in the population densities (N m"2, n = 4-5) of C. curvispinumi n the Lower Rhine at Lobith and Nijmegen.Th e differences between densities at Lobith and Nijmegen were tested by Student f-tests.
Table 2. Body length and brood sizefo r Corophium curvispinuma t Lobith during 1992 -1993.
Table 3. Body length and brood size for Corophiumcurvispinum a t Nijmegen during 1992 -1993.
Table 4. Comparison of brood size (number of eggs) of embryonic stages I - IV of Corophium curvispinumshowin g egg lossfro m the broodpouc h at Lobith and Nijmegen during 1992. Multiple comparisons of differences between embryonic stages were done using one-factor ANOVA and followed by Student-Newmans-Keuls (SNK) tests.
Table 5. Mean brood size (number of eggs) of embryonic stages I - IV of Corophium curvispinum showing egg loss from the brood pouch at Lobith and Nijmegen during 1992 -1993.
Table6 .Compariso n betweenbod ylengt han dash-fre edr yweigh to fovigerou sfemale sfro mothe r individuals(pre-ovigerou sfemale san dmales )o f Corophiumcurvispinum i nth eRive r Lower Rhine. Student f-testswa s usedt o makestatistica lcompariso no f theash-fre edr yweigh t inth e ovigerous females andothe r individuals.
Table7 .Growt han dsurvivorshi po f Corophiumcurvispinum populatio nrecruite dduring Apri l199 2- February 1994 at Lobith,fo r computation of production (for details refer appendix 1).
Table8 .Growt han dsurvivorshi po f Corophium curvispinumpopulatio n recruiteddurin gApri l 1992- February 1994 at Nijmegen, for computation of production (for details refer appendix 1).
Table 9.Variation s inth e mudfixe do nth e stones (including macroinvertebrates) and experimental tiles (excluding macroinvertebrates) at Lobithan d Nijmegen duringJul y 1992- Octobe r 1994i nth e Lower Rhine.
Table 10.Populatio ndensitie so f Corophiumcurvispinum an d mudmateria l(dr yweigh t andash-fre e dry weight) on stones at Lobith and Nijmegen and on cumulatively exposed tiles at Weurt.
Table 11. The percentage of organic matter in the mud material on the stones at Lobith and Nijmegen and onth e tiles at Weurt.
Table 12. Mean population densities and population filtering impact of Corophiumcurvispinum o n stones from groins at various places in River Meuse and River Rhine.
Table 13.Variation s inth e population densities of macroinvertebrates (mean± SE) onstone s (n= 3-5) at different depths and locations inth e River Meuse in September 1993.Multipl e comparisons of differences between the densities atvariou s locations were done using Student-Newman-Keuls (SNK) tests along with one-factor ANOVA. XXIV
Table 14.Variation s inth e population densities of macroinvertebrates (mean± SE) on stones (n= 3-5) at different depths and locations in the River Rhine during September 1992 and September 1993. Multiple comparisons of differences between the densities at various locations were done using Student-Newman-Keuls (SNK) tests along with one-factor ANOVA.
Table 15.Mea n population densities (individuals m"2ston e surface at 1m depth ) of zebramussel , Dreissenapolymorphs on the different sampling locations in the River Rhine before (September 1989) and after (September 1992 and September 1993) the population explosion ofCorophium curvispinum.
Table 16.Selecte d data onth e losso f embryos (embryonic stages It o IV)fro m the broodpouc ho f different species of Corophium.
Table 17. Comparison of present results on C.curvispinum wit h those reported by Muskó (1992) and Van den Brink et al.(1993a) .
Table 18. Seasonal variations of Sodium (Na+) concentration in the River Rhine (at Lobith and Andijk) and Meuse (at Keizersveer and Remily) during 1993 (RIWA, 1993).
Table 19.Change si nth emaximu mpopulatio ndensitie so fC. curvispinumo nstone sfro mth egroin s (at a depth of 0.5 m) during the months of September - October and yearly median water quality parameters ofth e Lower Rhinea t Lobith (km 860)fro m 1986t o 1993(modifie d after Vande n Brink et al.,1993a) . 1. Introduction
1.1.Distribution and range extensionof Corophium curvispinum
Corophium curvispinumi sa smal ltubiculou samphipo doriginatin gfro mth ebrackis hwater s
(below6% osalinity ) ofth e Ponto-Caspicare ao f Eastern Europe(Behning , 1914;Romanova , 1975).
Several races exist of this species, of which forma deviumi s adapted to fresh water (Wundsch,
1915;Harris , 1991) .C. curvispinum ha sexpande dit sdistributio n since 1900fro mth e riversenterin g the Caspian and Black Seas through interconnected rivers to Western and Central Europe
(Wundsch, 1919).Thi s expansion has probably beenaide db y shippingtraffi c (Jazdzkewski, 1980;
Pygott & Douglas, 1989). The occurrence of C. curvispinumi n Central Europe has been reported from the Caspian Sea, Volga delta, River Volga at Saratow, River Don, River Oka, Black Sea, lowland courses of Danube, Dnieper at Kiev, Müggelsee, near Berlin, Lake Kummerow, rivers emptyingint oth eBalti can dtw o lakesi nth eCaucasu san dLak eBalato n(Wundsch , 1919;Behning ,
1925; Sebestyén, 1934; Crawford, 1935; Waterstraat & Kohn, 1989). In Western Europe, C. curvispinumwa sfoun di n 1977i nth eDortmund-Em scana l(Germany ) andi nth eBelgia npar to fth e
River Meuse in 1981 (Van den Brink era/., 1989). In Britain, C.curvispinum wa s first recorded by
Crawford (1935) from the River Severn in the vicinity of Tewkesbury, and has since expanded its distributional range to a number of rivers and canals, including stretches of the Grand Union,
Coventry, South Oxford, Worcester and Birmingham, Shropshire Union, Trent and Mersey, and
Leedsan dLiverpoo lcanal s(Gledhil l etal., 1976 ;Moon ,1970 ;Taylor , 1985;Taylo r &Harris ,1986a ;
Pygott & Douglas, 1989). C.curvispinum wa sfirs tobserve d inth e Middlean d Lower Rhine in 1987
(Van den Brink etal., 1989 ; Scholl, 1990) and in 1992i twa sfoun d inth e northern part ofth e delta area inTh e Netherlands (Platvoet & Pinkster, 1995). C.curvispinum i sth e only Corophiumspecie s recorded inth efreshwate r range ofth e Lower Rhine and its branches. However, inth e oligohaline partso fth e Lower Rhine estuary, C.curvispinum co-occur s withth e relatedbrackis hwate r species,
C. lacustre Vanhöffen and C. multisetosum Stock (Van den Brink etal., 1993a) .
1.2.Reasons for thepresent study
The River Rhine has long been exposed to human activities which have drastically altered 2
its physical and chemical characteristics (Van der Weijden & Middelburg, 1989; Van Urk & Smit,
1989; Van den Brink et ai, 1990; Den Hartog et ai., 1992). Canalisation measures, together with
organic, inorganic and thermal pollution, have contributed to the loss of many native
macrozoobenthicspecie s andfacilitate dth esucces so fsevera l invadingexoti cspecie sove rth elas t
century (Van den Brink et ai, 1990,1991; Van der Velde et ai, 1990; Pinkster er ai, 1992). The
euryhaline and often thermophilous nature of these invaders is well suited to the present physical
and chemical conditions of the Rhine (Den Hartog et ai, 1989,1992; Van den Brink, 1995).
Crustacea make up the largest and most successful portion of this exotic element. The
amphipods Gammarustigrinus Sexto nan d Corophium curvispinum have beenth e most successful
invaders (Vande nBrin k etai 1993a),currentl yoccurrin gi nver yhig hdensitie s inth emai nchannel' s
littoralzone .G. tigrinus i sa nintroduce dspecies ,originatin gfro m North-America,an dwa sfirs tfoun d
inth e Rhine insignifican t numbersdurin g 1983(Va n Urk &Bi jd eVaat e 1990).Currentl y it reaches
densities of thousands of individuals per m2.However , inth e lastdecade ,Ponto-Caspi c amphipods
have beenth e mostfrequen t invaders (seeVa nde n Brink etai, 1989,1993b; Bijd eVaat e & Klink,
1995).Thi s influx has been aided by an increase inth e connectivity of europe's major rivers, such
as the Danube and Rhine, by means of trade-route canals and shipping.
Through their dominance,thes e new invading species maywel lhav e played a large rolei n
restructuring the Rhine ecosystem inth e last decade. Exotic invaders can cause many disruptions
to ecosystem functioning (Drake etai, 1989),particularl y alter the species composition (Pinkster et
ai, 1992; Dick et ai, 1993),trophi c structure and energyflow s (Nichols & Hopkins, 1993; Steward
& Haynes, 1994; Fahnenstiel etai, 1995)o fth einvade decosystem .Unlik eth echang e ingammari d
composition, C. curvispinum is not an example of generic alteration, but is the first tubiculous
amphipod to occur in the Rhine (Pinkster et ai, 1992). Eventhough large numbers of euryoecious
and exotic species have invaded the River Rhine, C. curvispinumappear s to be more successful
than any other invader. The maximum mean densities of C. curvispinumincrease d from 2 (1987)
to 200,000 individuals m"2(1992 ) on the stones of groins inth e Lower Rhine (Van den Brink et ai,
1993a).Thi ssucces s hasbee nattribute dt oth eincrease dsalinity , ionicconten t (e.g. Na+)an dwate r
temperatures of the Lower Rhine caused by industrial pollution (Den Hartog etai, 1992). In recent years, C. curvispinum has been reported to have had an enormous impact on the River Rhine ecosystem bychangin g food webs and energyfluxe s (Kelleher etai, 1997;Marguillierefa/. , 1997;
Van der Velde et ai, 1997). While much information exists on its geographical distribution (Jazdzewski, 1980; Wouters, 1985; D'Udekem d'Acoz & Stroot, 1988;Va n den Brink et a/., 1989,
1991; Pygot t& Douglas ,1989 ;Scholl ,1990 ;Muskó ,1994 ;Platvoe t& Pinkster ,1995 )an dphysiolog y
(Taylor &Harris , 1986a,1986b ;Baylis s &Harris , 1988;Harri s& Bayliss, 1990;Musk ó etai, 1994),
very little is known about the ecology of C. curvispinum (see Muskó, 1992; Van den Brink etal.,
1993a). The population explosion of C. curvispinum in the River Rhine presents an excellent
opportunity tofil lthi s gap inknowledg e and,mor egenerally ,th eeffec t of exotic specieso n invaded
ecosystems.
1.3. Objectives
The present study was carried out in order to gather data on ecological aspects of C.
curvispinumi nth e Rivers Rhine and Meuse.Th e aims are asfollows :
1. to study life history and reproductive biology of C.curvispinum i nth e Lower Rhine related to hydrographie parameters
2. to assess demographic patterns (e.g. population structure), growth rate and production of C.curvispinum
3. to quantify mud-fixation by C.curvispinum o n a seasonal basis and relate to variations in population densities
4. to derive an estimate of filtration by C. curvispinumfro m the population density values at different locations in the River Rhine and Meuse
5. toinvestigat eth eimpac to fC. curvispinumo nspecie scompositio nb yrelatin git spopulatio n densities to changes in densities of other common macroinvertebrates in the River Rhine and Meuse.
2. Materials and methods
2.1. Studyarea
The present study was carried out at locations along the Rivers Rhine and Meuse, The
Netherlands (Fig. 1).Th e river bedo fth e River Rhine and its branches isgenerall y fixed bygroins , breakwaters which consist of large basalt stones (grauwacke). The river bed largely consists of RiverMoa e : Stapling tocatkns l.Ool 2Jhocfchuj2ai SJlcrfcu 4 Coijk 5.Gnvc 6.G«mle North g,, 7.Gcodereo LKdzcnvecr 9Anxr RiverRhin e: Samplin glocatio n lO.Lobith ll.N4nKf.e11 12.Tid 13.V«a U. 15.1 lo-BcffiudbKbt n.Letterkat U.Vdp ».DeSUef 20.0k) 21.Wipie
Fig.1. Map showing the sampling locations in the Rivers Rhine and River Meuse. 1a. Mean population densities of Corophium curvispinum (individuals m'2 stone surface at 1 m depth) at different sampling locations in the River Rhine and River Meuse during September 1993. 1b. Mean population densities of Dreissena polymorpha(individual s m"2ston e surface at1 m depth) at different sampling locations in the River Rhine, before (September 1989) and after (September 1992) the population explosion of C.curvispinum. 5 heavily disturbed, shifting sands. Low quantities of vegetation are only present in the calmer downstream reaches. Water levels vary annually and seasonally, but are generally lowest in
September andOctober . Ingeneral ,annua lflo w ratesan dstrea mvelocitie s arehighes t inth euppe r parts of the branches, and lowest inth e wider lower parts.
2.2. Methods
2.2.1. Life history and reproductive biology
In order to study the reproductive biology of C. curvispinum, stones from groins at Lobith andNijmege n(Fig .1 )wer esample dfro ma shi pb ymean so fa polyp-gra boperate dwit ha hydrauli c crane. Water depth (4 to 5 m) was measured at each sampling location using an Atlas DESO
25/SLS instrument. Per locality, three to five stones were taken for each month.Th e stones were carefully brushed inorde rt ocollec tvariou smacroinvertebrate san dwer e preservedi n70 %ethano l for further analysis.Th e surface area ofth e stones andpopulatio n densities of macroinvertebrates were estimated using the following equation (Bij de Vaate & Greijdanus-Klaas, 1991):
D = (IN/IA) * 10,000 where, D= the number of animals per m"2,X A =th e sum of meansurface s (incm 2) of all sides of stones, XN = the sum of animals on all sides of stones, and 10,000 = the conversion factor.
The preserved samples were transported toth e laboratory in labelled polyethylene bottles andsieve dfo rth eseparatio n ofmacroinvertebrates . Early inth ecours eo f study,a 1m mmes hwa s used for sieving samples. However, it was often found that small individuals of C. curvispinum passedthroug h the 1m mmesh .Subsequently , three meshsize sviz. ,0.50, 1an d2 mmwer e used for sieving in order to ensure that all size classes were represented.Th e 0.50 mm mesh size was chosen since the body length of newly hatched juveniles of C. curvispinumi s about 550 |xm.
C.curvispinum wa sseparate dfro mth ecollecte dmateria li norde rt ostud y itslif ehistor yan d reproductive biology.Th e number of C.curvispinum ineac hsampl e wascounte dan dcalculate dfo r the average numbers per square metrefro m threet ofiv e stone samples.Whe n samples werever y large (>1000individuals) , asu bsampl e wastake nfo r measuring length-frequency distribution.Th e body lengtho f all specimens was measured toth e nearest 0.05 mmfro m theanterio r margin of the 6
rostrum to the posterior margin of the telson.Th e sex of all specimens above a body length of 1.8
mm was recorded using the presence (male) or absence (female) of genital apophyses on the
ventral face of peraeon segment VII. Individuals which were less than 1.8 mm in length were considered as juveniles as it was impossible to determine their sex. The presence of males,
ovigerous and pre-ovigerous females and juveniles was recorded for each sample. The term
"ovigerous females" is usedt o define adult females which are carrying egg embryos in the brood
pouch.I neac hsample ,th elengt ho fovigerou sfemale swer erecorde dan degg swer e removedfro m the broodpouc ht ofollo w the seasonalvariation s inclutc h size (number of eggs per brood).Withi n the brood pouch, five different stages were recognised (stages are described in results). Newly fertilised eggs are stage I, and eggs close to hatching are stage IV. Stage V embryos were rarely
available asthe y often exitth e marsupium.Al legg s (stage It o IV) ineac hbroo dwer e countedwit h a stereomicroscopean d egg losswa scalculate dfo r eachsamplin g datefro m differences between various stages (Fish & Mills, 1979; Peer etal., 1986) .
Biomasso f C.curvispinum wa scalculate dfro m a correlation between body length (L) and ash-free dry weight (AFDW).Th e weights of different size groups of C.curvispinum wer e obtained from unpreserved samples,whic h were driedt o constant weight at 60° Can dcombuste d at 550° C for 5 hours and reweighed.
2.2.2.Growth rates
Mean growth rates of C.curvispinum wer e estimated on the basis of the growth of natural populations by using the length-frequency distribution.Th eweight/lengt hspecifi c growth rate(Gw p) was obtained using the following equation (Winberg etal., 1971) :
1 B2
Gwp = -— In -— t B,
where, t is the interval between sampling dates in days, B, isth e biomass at the beginning of the sampling interval and B2i s the biomass at the end as projected byth e growth model.
Meanlength so fdifferen tgeneration so fC. curvispinumwer eals oestimate dfro mth elength - 7
"frequenc y tables based on the increment of mean length of each cohort between subsequent
samples during the months.
2.2.3. Production
The production of C. curvispinum was calculated following a method described by Crisp
(1971):
t=n Nl + Nl+1 P = S -AW t=0 2
Where, N is the mean population density at time t and t+1 and AW is the individual mean weight
increment betweensuccessiv e sampling.I nthi smetho dth e incrementso fth e meanweigh t of each
cohort werecalculate d separately.
Toasses sproduction ,quantitativ elengt hfrequenc ytable swer econstructe dfro mth elength -
percent frequency and population density datafo r eachmonth .Quantitativ e estimationswer e made
by summing the weight numbers through all size classes observed for each month.Th e difference
in biomass between that of a particular date and that projected from the last date was considered
to be an estimate of mortality. Thisvalu e will be positive ifth e projected biomass is larger than the
actual sample biomass.
2.2.4.Distribution and impactsof C. curvispinumon other macroinvertebrates
In addition, population densities of C. curvispinum and other macroinvertebrates were studieda t2 1samplin g locationsalon gth e Rivers Rhinean dMeus e (Fig.1 )durin gSeptembe r 1992 andSeptembe r 1993.Al lmacroinvertebrate s were identifiedt ospecie s level.Samplin g stoneswer e collected from the end of the groynes at five different water depths viz., 0-1,1-2, 2-3, 3-4 and 4-5 m, by following above mentioned procedures.
2.2.5. Mud-fixation
In order to estimate the amount of fixed muddy material on stones, an extra stone was 8
sampled monthly at a constant depth of 4 - 5 m during July 1992 - October 1994 at Lobith and
Nijmegen inth e Lower Rhine.Th e stone surface was carefully brushed off andth e materials were
collected in plastic containers. Inth e laboratory, the muddy material,includin g macroinvertebrates,
from the extra stone was dried for 24 h at 105 °C, weighed for constant weight (dry weight) and
subsequently combusted for 4 ha t 550 °Ct o calculate the ash-free dry weight (AFDW).
Experimental tiles were suspended at Nijmegen (sluice at Weurt) in order to estimate the
seasonal variations in colonisation and mudfixatio n by C.curvispinum. The tiles were categorised
into short-term as well as long-term (Series I and Series II) exposures. The short-term tiles were
suspended and withdrawn after a month exposure and replaced in sequence at monthly intervals
during the whole year to follow the seasonal pattern of colonisation related to mud fixation of C.
curvispinum. Twoserie so flong-ter mtile swer eexpose dfo rstudyin gth ecumulativ e variations inth e
mud material. First series (Series I) of long-term tiles were all suspended together, but were
withdrawn at a rate of one every 30 days. Second series (Series II) of long-term tiles were
suspendedsuccessivel y at3 0da yinterval san don etil ewa sretrieve dfo reac hmont hafte ron eyea r
exposure. Due to the fact that thetile s are suspended intoth e water at a depth of about 4 m,the y
can only be colonised by drifting animals present inth e water. Each tile was studied for numerical
abundance of different macroinvertebrates and amount of muddy material. The muddy material
excluding macroinvertebrates from the tiles was driedfo r 24 ha t 105°C ,an d weighed for constant
weight (dry weight) and combusted for 4 h at 550 °Ct o calculate ash-free dry weight.
Thechange si npopulatio ndensit yan dchange si namoun to fmu dmateria lo fC. curvispinum
on the stones and tiles were calculated based on the following equations:
Change in density AN = (Nt-Nt+1)
where, Ni sth e population density (individuals m"2)a ttim e t andt+ 1 isth e population density at the
subsequent sampling date.
Change in mud weight Aw = (wt-wt+1)
where, w is the ash-free dry weight of mud material (g m'2) at time t and t+1 is the ash-free dry weight of mud material at the subsequent sampling date. 9
Meanvalue so fmu dfixatio nrat epe rindividua lwer ecalculate dfro mth ecumulativ edensitie s of C. curvispinum and amounts of mud fixed cumulatively on the stones and tiles by using the following equation:
Mud fixation rate = [(a/v,)+(wJ/(ÎNt)+(Njy(îd t)+(dt+1) 1=0 t=0 t=0
where, w isth e ash-free dry weight of mud material (gm" 2) attim e t, t+1 isth e àsh-free dry weight of mud material at the subsequent sampling date, N is the population density of C.curvispinum
(individuals m'2) at time t, t+1 is the population density at the subsequent sampling date, d is the duration of period in days at time t and t+1 is the duration of period in days between successive sampling.
The meanbod y lengtho f C. curvispinumwa scalculate dfro mth ecumulativ e monthly mean length and monthly densities of C.curvispinum base d on following equation:
Mean body length = S(Nxl_)/ÎN t=0 t=0
where, N is the total number of individuals at monthly intervals and L is the monthly mean body length of C. curvispinum.
The percentage of organic matter of the mud material was calculatedfro m the dry weight and ash-free dry weight of the mud material using the following equation:
Organic matter (%) = (AFDW/DW) x 100
where, DW isth e dry weight (g m"2)an d AFDW isth e ash-free dry weight (g m"2) of mudmaterial .
2.2.6.Filtration capacity
Theclearin gcapacit yo fan yfilte rfeedin gpopulatio ni na give necosyste mca nb ecalculate d from the length frequency distribution, density and length related filtration rate of that particular 10 population. Filtration rate can be calculated by the difference in cell concentration between the beginning and end of an experiment, taking into account of the experimental time, experimental volume andnumbe r of animals per experiment. This will bedon eaccordin gt oth eformul a givenb y
Peters (1984). Unfortunately, there are no data available onth e filtration rate of Corophium sp. in literature. However, our preliminary filtration experiments of C. curvispinum with Chlorella vulgaris showed about 5 cm3 m"2 h'1 per individual.Th e filtration capacity of C.curvispinum i n River Rhine and Meuse was theoretically calculated by using this filtration rate and mean population densities.
2.2.7. Hydrographieparameters
Datao nphysico-chemica l parametersan dstrea mvelocitie so fth eRiver s Rhinean dMeus e were obtainedfro m Rijkswaterstaat/RIZA, Lelystad andfro m RIWA,Amsterdam ,Th e Netherlands.
2.2.8.Statistical analysis
Comparison of brood size (number of eggs) at different embryonic stages (I to IV) of C. curvispinumwa sperforme dusin gone-wa yANOV Afollowe db ySN Ktest s (Zar, 1984).Th estandar d regression equation was used to relate density of ovigerous females and mean brood size of C. curvispinumwit h temperature and chlorophyll-a, respectively. Stream velocity data collected from
Lobithan dNijmege nwer ecompare db yusin gnon-parametri cWilcoxon-Signe dran ktes t(Zar ,1984) .
Datafro meac hsampl e intervala t Lobithan dNijmege nwer eanalyse dan dassigne da ran kt omak e statistical comparison, independent of seasonalvariation .Th edifferenc e inpopulatio n density of C. curvispinumbetwee n Lobith and Nijmegenwa s compared bystuden t Mests (Sokal &Rohlf , 1981).
Thedifference s betweenpopulatio ndensit yo feac hspecie sa tdifferen t locations in River Rhinean d
Meuse were analysed by one-way ANOVA followed by SNK tests (Zar, 1984).
3. Results
3.1. Population density
The highest density of C. curvispinum(mea n + SE; 168,100 + 27,495 individuals m"2, n = 11
3-5) was recorded in the Lower Rhine at Lobith inJul y 1993 (Fig.2) .Th e corresponding value for
Nijmegen is146,90 0+11,82 6 individuals m'2i nJul y 1993(Tabl e 1) .Th ehighes tdensitie s generally coincided with the breeding season of C. curvispinum (refer to breeding season section 3.5).
Followingth e initialreleas eo fyoun ganimal si nlat eMay ,densitie s increasedsteadil yfro m Mayt o -
September [Lobith:fro m 17,400+ 4,469 (June 1992)t o 67,990+ 9,054 individuals m"2(Septembe r
1992) and from 40,630 + 5,049 (May 1993) to 168,100 ± 27,495 individuals m"2 (July 1993);
Nijmegen:fro m 42,380 + 7,643 (June 1992) to 136,200 + 4,189 individuals m'2(Augus t 1992) and from 29,480 +5,13 4 (May 1993)t o 146,900 + 11,826individual s m"2(Jul y 1993)].A drastic decline
200000 I I Inhith 175000 HU Nijmegen 1 E J2 150000 - § •a '> 125000 - X3 100000 - I 'in i 75000 .2 2s o 50000 Q_ 25000 0 mill ilrlllln i A M J JASONDJFMAMJJASONDJFM L/S J A 1992 1993 1994
Fig.2. Seasonal variations in the population densities (mean ± SE) of C. curvispinumfro m April 1992 to March 1994 on stones (n = 3-5) from the groins at Lobith and Nijmegen in the Lower Rhine. 12
Table 1.Seasona l variations in the population densities (Nm" 2,n =4-5 ) of C.curvispinum in the Lower Rhine at Lobith and Nijmegen. The differences between densities at Lobith and Nijmegen were tested by Student f-tests.
Months Lobith Nijmegen Studentf-test s /•*.T -2\ (Nm' 2) (Nm' ) ±SE ±SE t df P
Apr199 2 3358±702 3214±1503 0.079 7 ns May 139±39 8464±2791 2.983 6 * Jun 17300±4469 42380±7643 2.833 6 * Jul 47650*7897 117800±16080 3.596 7 ** Aug 66040±7201 136200±4189 8.422 8 *** Sept 67990±9054 66300±9674 0.128 6 ns Oct 14980±3486 38400±10384 1.925 7 ns Nov 4021±2159 6545±1062 1.049 8 ns Dec 3844*819 2819±678 0.964 8 ns Jan199 3 2874±1071 4825±1094 1.260 7 ns Feb 10660±1036 15640±1345 2.933 8 * Mar 5386±987 9380±1683 2.047 6 ns Apr 2506±553 5201±280 4.001 7 * May 40630±5049 29480±5134 1.530 7 ns Jun 43940±6647 60210±1451 2.391 6 ns Jul 168100±27495 146900±11826 0.708 8 ns Aug 82750±14241 99170±11464 0.898 6 ns Sep 35900±4816 48150±8103 1.366 7 ns Oct 17470±1626 69510±7545 6.742 6 *** Nov 22450±3112 47090±6067 3.614 8 ** Dec 25440±3074 48770±7567 2.594 7 * Jan199 4 4960±1223 13030±4331 1.605 7 ns Feb 15650±1803 19070±2661 1.064 8 ns
Levels of significance indicated by *P<0.05; **P<0.01; ***P<0.001; ns= not significant.
in densities was observed from September to April [Lobith:fro m 67,990 + 9,054 (September 1992)
to 2,506 + 553 individuals m"2(Apri l 1993) and from 35,900 + 4,816 (September 1993) to 3,770 +
2,644 individuals m"2 (March 1994); Nijmegen: from 66,300 + 9,674 (September 1992) to 5,201 +
280 individuals m"2 (April 1993) and from 99,170 + 11,464 (September 1993) to 19,070 + 2,661
individuals m"2 (February 1994)]. 13
The population densities of C. curvispinum at Nijmegen are higher than at Lobith. A significant difference in population densities of C. curvispinum was observed between Lobith and
Nijmegenfro mMa y 1992(f-tests ,P<0.05 )t oAugus t 1992(f-tests , P<0.001)an dfro mOctobe r 1993
(f-tests, P<0.001) to December 1993 (f-tests, P<0.05) (Table 1).
3.2. Populationstructure
Length-frequency histogramssho wtha t inMarc h 1992,th epopulatio no f C.curvispinum a t
Lobithwa sdominate db yadult so fth eoverwinterin g generation,havin ga mea nlengt ho f=3.7 5m m
(Fig. 3). Individual lengths of C. curvispinumincrease d during April.Th e population was strongly bimodal in May as the first newly hatched juveniles appeared in the samples. By June, the population was dominated by individuals below 2.0 mm body length. The majority of the overwintering population had disappeared by June. A second peak of juveniles (<1.8 mm body length)wa sobserve di nJuly .A thir dan dfina lpea ko fjuvenile swa srecorde di n September/October.
The highpercentag e of individuals below 1.8 mm body length inJul yan dSeptember/Octobe r must have been due to the breeding of the young first recorded in May. This is possible mainly due to theirfas tgrowt han drapi dattainmen to fmaturity ,thereb y producinga summer-breedin ggeneration .
The period May - September recorded the largest shifts in size, indicating that growth rate and production was highest in that period.Th e patterns of change inth e C. curvispinumpopulatio n at
Nijmegen were largely similar to that of Lobith (Fig.4) .
3.3. Sexratio
These x ratio of malet ofemal e C.curvispinum longertha n 1.80m m rangedfro m 0.20 (May
1993 at Nijmegen) to 0.93 (November 1993 at Lobith) (Fig.5) .Th e lowest recordedvalue s in May indicatedth e decrease of males inth eoverwinterin g cohort.Th e maleso fth e Maycohor t appeared in June/July and represented the highest male ratio in subsequent months.
3.4. Brooddevelopment
Within the brood pouch of C.curvispinum, five clearly defined developmental stages were 14
Jul9 3 n = 471 20 Mar9 2 n=460 Nov 92 n = 220 15 10 5 Irf-nTT*-, 0 nFrü Jffl k FTm n • A 20 Apr9 2 n = 803 Dec9 2 n = 191 Aug 93 n = 298 15 10 5 0 J 'LJ l'U**- Sept9 3 n= 312 20 May 92 n = 329 Jan9 3 n = 201 15 10 5 0 jn •lliftfH Mt 20 Jun9 2 n =37 5 Feb 93 n = 170 Oct 93 n = 194 15 >, 10 o N' S 5 JU B, FUJ- J •fiHjflFh M M. fl-|i-m- £ 0 Jul9 2 n =40 4 Mar9 3 n = 268 Nov9 3 n=172 £ 20 S? 15 10 5 0 _&£ Pff)-1! JM: :S-EEhn. 20 Aug9 2 n =33 6 Apr9 3 n = 207 Dec 93 n = 134 15 10 5 ftftn^ 0 _JL L 20 Sept9 2 n= 225 May 93 n =38 2 Jan 94 n= 183 15 10 5 0 MM LL J A •fa_ Feb94 n= 176 20 Oct9 2 n = 233 Jun9 3 n = 351 15 10 5 Jlfllt FTH-I 0 J ¥^f Ufl 1 1 r 0 123456012345601234567
Body length (mm)
Ovigerous females Pre-ovigerous females Maies Juveniles
Fig.3. Length-frequencyhistogram so fC. curvispinum fro mMarc h 1992t oFebruar y 1994a tLobit h in the Lower Rhine (the animals above 6 mm body length contributed only <0.2% during summer months). 15
Nov 92 n = 155 Jul9 3 n = 331 20 Mar9 2 n = 277 15 10 5 0 —^ *- ^Tl-B^-Q-,-^ JÜ fltikb-« hm Aug 93 n = 350 20 Apr9 2 n = 241 Dec 92 n = 190 15 10 5 rHiïïf 0 J.I I i OL Jan 93 n = 157 Sept9 3 n= 24 5 20 May 92 n = 224 15 10 5 0 1 LL Feb 93 n = 212 Oct9 3 n = 199 20 Jun 92 n = 328 15 >* 10 o 5 3 er 0 _J -H.FCT . CD 20 Jul 92 n = 398 Mar9 3 n = 239 Nov9 3 n = 161 S? 15 10 h 5 rmif Tte^ 0 I 20 Aug 92 n = 305 Apr9 3 n = 207 Dec 93 n = 137 15 10 5 0 rrrf m Tftfl-fL^ May9 3 n = 255 Jan 94 n = 175 20 Sept9 2 n= 26 1 15 10 5 0 _J Iflffa-rfH Feb 94 n = 153 20 Oct 92 n = 172 Jun 93 n = 351 15 10 5 X. Mäfi 0 -«P M +-*T r P=-i r 01234560123456 01234567 Body length (mm)
Ovigerous females Pre-ovigerous females  Males Juveniles
Fig.4. Length-frequency histograms of C.curvispinum from March 1992 to February 1994 at Nijmegeni n theLowe rRhin e(th eanimal sabov e6 m m bodylengt hcontribute donl y<0.2 % duringsumme rmonths) . 16
recognised (Fig.6) :Stag e I- egg sar esmall ;cream y white incolour ; nodifferentiatio n of cells;egg s
held in a transparent sac, Stage II - eggs are relatively larger than stage I; yellowish-orange in
colour;eg gcas e istransparent ; optican d rudimentappendage sar ediscernible ,Stag e III- eg gcas e
istransparen to rbroken ;opti can drudimen tappendage sar emor edeveloped ,Stag e IV- transparen t
sacs are absent; all appendages are fully developed, and Stage V - newly hatched juveniles.
The first generation of juveniles were collected at the end of May in the Lower Rhine.
Ovigerous females of the first generation (May cohort) were observed at the beginning of July.
Hence,th etim e taken bya juvenil et o reachth esiz eo f anovigerou sfemal e (>2.2m m body length)
was about 4-5 weeks. The second generation of juveniles was recorded during the third week of
July, indicating approximately two weeks of embryonic development.
-• Lobith -•— Nijmegen 1.00
0.80 % rvy
0.60
o | S 0.40 Vi
0.20
0.00 • ' ' ' • 1 • MAMJ JASONDJFMAMJ JASONDJF 1992 1993 1994
Fig.5. Seasonal variations in the sex ratio of C. curvispinum from March 1992 to February 1994 at Lobith and Nijmegen in the Lower Rhine. 17
.C .8" 5 =3 O Ci
o o Q- "O O O
CO O
-O E CD O) cn CD co CD CD CO
ra
CD E _CoL >CD OCD
«3 18
3.5. Breedingseason
The breeding season of C.curvispinum a t Lobith began in Marchwhe nwate r temperature
increased above 10° C(Fig .7a) .Th e breeding seasoncease d inOctobe rwhe nth etemperatur e fell
below 10°C .Th edensitie so fovigerou sfemale so fC. curvispinumshowe da positiv ecorrelatio nwit h
the temperature at Lobith (r = 0.54, P <0.05 , Fig.8a ) and Nijmegen (r = 0.74, P < 0.001, Fig.8a) .
Lobith: Ovigerous females were first observed on 23 March 1992whe n 5% of identifiable
females (i.e.>1.8 0m mi nbod y length)wer ecarryin geggs .The ysteadil yincrease di nfrequenc yan d
contributed over 85% to the total female population from 21 April to 18 May (Fig. 7b). By June,
overwintering ovigerous females had started to disappear and the rapid growth of the summer
generation C. curvispinum resulted in the largest members being of reproductive size (> 2.2 mm
length) byJuly .Th e next peak infrequenc y of ovigerous females coincided withth e maturity of the
May-Junecohor t andnearl y70 %o fth efemale swer egravi do n 15July .Th ethir dpea k infrequenc y
correlated withth e maturity of June-July cohort and53 %o f thefemale s above 2.2 mm length were
ovigerous on 10 August. The last ovigerous females were noted on 9 September 1992. A similar
trend of breeding periodicity was also observed in 1993 (Fig.7b) .
Nijmegen: Ovigerousfemale so f C.curvispinum wer efirs t recorded inMarc h (for 1992an d
1993) and increased to a maximum level of 90%i n May (Fig.7c) . By June, the ovigerous female
percentage had decreased to 36%o f the total female population.Th e second peak was recorded in July (65%) andth e thirdan d last peakwa s observed inSeptembe r with 72%ovigerou s females.
As in Lobith, reproduction ceased in October.
3.6. Broodsize
Seasonal variation data on brood size and body length relationships of C. curvispinuma t
Lobith and Nijmegen are summarised in Table 2 and 3, respectively. There was a significant seasonal variation in the brood size of C. curvispinuma t both Lobith and Nijmegen (ANOVA, P <
0.001). The maximum brood size of 25.3 + 4.75 (mean + SD, n = 113, y = -11.390 + 8.507x, r2 =
0.89, P < 0.001) was observed at Nijmegen on 20 May 1992. At Lobith,th e maximum brood size of 17.5±3. 1 (n = 41, y = -12.7503 + 7.022x, r2 = 0.96, P < 0.001) was recorded on 18 May 1992. 19
M AMJJ ASONDAJ FMAMJ J A S O N D J F 1992 1993 1994 fWSftfl Ovigcrous ESiiM Prc-ovigerous II IH1 1 Males I I Juveniles females females
Fig.7. a. Seasonal variations in the hydrographie parameters (temperature and chlorophyll-a) in the Lower Rhine from March 1992 to February 1994. b and c. Relative occurrence (%) of ovigerous females, pre-ovigerous females, males and juveniles of C. curvispinum at Lobith and Nijmegen from March 1992 to February 1994. 20
<*H o O § o Cfl cfl > C CS CS u 's l-J 00
o « inooomooon ooocsmeioinin oo « vo r-~ es ei ei ino \» r-* ^ m nHioioHcinm 'I -s r~ooNOt-~ooooNotN ra o öö o ö o ö ö dödödddd b 'S o 5 U u x .* X X .55x x X X X X X X X X oo 25 CS £* ot m ON m ON 00 o •o oo O CS vo in in NO 00 NO t •3
vo c- 't vo vo -H m eiNOONONincSNONo — IN< N ^ ^ *—i »—i ÇS •—i •—< *—' 00 ei m v© O ON 't «-* "-1 't >n oo in >n •* ON ON <*
CS 0- o» es -H ei ooe s ONeiincsoooocsTt o\ —< oo•- •n O O in O in O m >n O O m o •n in s CS O) 't ei NO 00 m oo O) ON in t Ol ei =I 't •n in "O t' t' ei t' rt •n' •^ 't 't 't t' d ra P o m O in m o m O m O m O o •n O Cd -5- 't 't r- ND >n >n ei oj ON p NO Oj 3 ei ei ei ei (N * * >-J 00 * * o c >nvoovescsoor~ — moMS o\«wioo o—iooovcs-toovo 'I-S O\oooooor~vo>ncs oot-^vqr-ooooTt—i co o dódddóód "S S öööo'öööö § 8 X X X X X X X X X X X X X U u 00 CO r~ >-i ov c- >n vo ov r» Ov f» t~ Tf CS co v> oo 8gS fi •S Ov •^frvOvovO'-icoOvvo 4» , -H CS CO CO CS ~* CO OO ~* rt co es cs —i Ov r,N o\ *viN^oiin co es —i t-» TJ- co OS oCS\ OV CO <-i Tl VO r~ co ov v> V10\NHO\OHH ao OV CS t-; ON (-.vor^poot-^coco ü c CS. CO côco^"*ocôes'^*-H 'S 3'Si'î SUS »-*ovcst~o\ •* f- C~S 'S r* r-1 oo r~ «n 00 c oopco—jppoovq •* •* CS u oci^HwScsvoavuScs oó O N h H' OO vö Tf ÖO *-> es es — u II C o o o o o co es r-c o r~- O Il a c a 6 co u o oo c c e S M p~. -a u »- a o •9 •- JZ co 3 00 m 3 3 - o_ £ j>. 00 ^oo-00 p a a rô 3 3 3 Is 3 CS < i—i i—i < oo O CO 00 o S S ov < •> co IB ov\ >r> CS o f- >r> CS —f c~ ov CS OV I— °° es a r-t es es es <—' «—' CS —i l — H •—• O 13 J' 5 22 60000 • Lobith:y = -19466.3 5+ 1673.50x ( r= 0.54 ,P<0.05 ) O O Nijmegen:y = -21399.6 5+ 1892.77x ( r= 0.74 ,P<0.001 ) •ä 45000 •3 •s 1 o S 30000 • • o O o • • .§> 15000 > o • O o o • o • M L. —•- 5 10 15 20 25 Temperature(°C ) 30 • Lobith:y = 6.3 9 +0.12 x(r =0.68 ,P<0.001 ) O o Nijmegen: y= 7.00+0.2 0 x(r =0.71 ,P<0.001 ) O 20 •o • o x>g o 10 • • I o o o • • & 30 60 90 Chlorophyll-a(Hgi') Fig.8. The relationship between (a) the density of ovigerous females and temperature and(b ) meanbroo dsiz eo f C. curvispinum andchlorophyll- aa tLobit han dNijmege n inth eLowe r Rhinefro m March 1992t o February1994 . Theminimu mbroo dsiz eo f2. 7+ 0. 8( n= 12 , y= -3.804+ 1.838x ,r 2= 0.29 ,P > 0.05 )wa srecorde d on 7 October 1992 at Nijmegen (Table 3).Th e mean brood size of C.curvispinum inth e Lower 23 Rhine showed aclea r positivecorrelatio nwit hchlorophyll- a(Fig .8b) .I nan y given month,th e mean brood sizewa s higher at Nijmegen than at Lobith.Compariso n of stream velocities between Lobith (mean ± SD= 42.5 ± 5.1c m sec"1;rang e= 31-51, n= 35 ) and Nijmegen (mean ± SD =51. 3± 6.2 cm sec"1; range = 39-63,n = 35) showed a significant difference (Wilcoxon-sign, P <0.05 , Fig.9) . 37 44 51 Stream velocity (cmsec" 1) Lobith Fig.9. Acompariso n of streamvelocitie s atLobit han dNijmege n usingWilcoxon-Signe d ranktest . Todemonstrat eth erelationshi pbetwee nbroo dsiz ean dbod ylengt ho fC. curvispinum, only ovigerousfemale swit hbroo dstag eI o fembryoni cdevelopmen twer eused .Thi sminimise dth eerro r resulting from the loss of embryos from the brood pouch.A linear relationship was found between the female body length (Li n mm) andth e number of eggscarrie d (N) at Lobith (In N= -8.68 +6.1 2 In L, r2 = 0.92, P < 0.001, n = 137) and Nijmegen (In N= -7.25 + 6.89 In L, r2 = 0.90, P < 0.001, n = 114) (Fig. 10). 40 i Lobith 24 30 In(N )=-8.682 + 6.123I n(L ) ^ • (i2 =0.92 , P<0.001 , n= 137) ,-• •* *$ * 201- co oo 00 D #"" ' O JJ 10 • • «• • 4» S • •••••• • • • •• • • •• • • ••• • • •• • 3 40 — Nijmegen • ^ In(N )= -7.254+ 6.88 8I n(L ) • ••• 2 30 h (r =0.90 , P<0.001 , n= 114) »•••*•• gg 20 «MJÇ M* •>• ?• •• «•• • «•• ••«• • • • • 10- • • • • • • • • • • • •• • • • • i i i - ' 3 4 5 Female body length (mm) Fig.10. The relationship between lengtho fovigerou sfemale san dnumber so fstag e Iegg s in the brood poucho f C. curvispinum atLobit han dNijmege ni n theLowe rRhine . 3.7.Loss ofembryos from the broodpouch The number of embryos per brood at each developmental stage wasobserve d forC. curvispinuma tLobit han dNijmege ndurin gMarc h199 2t oOctobe r 1992. Broodsconsistin go fstag e 25 V embryos were not examined because hatching was asynchronous. There were significant differences betweenth ebroo dsize sa tstag eI an dstag eI Va tLobit h(ANOVA :d f= 383 , F= 61.893 , P <0.001 ) and Nijmegen (ANOVA: df =450 ,F = 91.142 ,P <001 ) (Table4) .Compariso n of brood sizes at stage Ian d stage IV showeda n egg losso f 33%a t Lobith(SN K tests, P< 0.001 ) and37 % at Nijmegen (SNK tests, P < 0.001) (Table 5). 3.8. Relationshipbetween body lengthand weight In order to estimate the biomass of C. curvispinum, a length-weight correlation was calculated by regression analysis.Males ,juvenile s (<1.80m m body length), pre-ovigerous females (females without eggs) and ovigerous females were separated and ash-free dry weights were determined for each individual. Results show that there was no significant difference in variation between the males and pre-ovigerous females (t-tests, P > 0.05). However, there is a significant variation between ovigerous females and other individuals (including males and pre-ovigerous females) (t-tests, P < 0.001) (Table 6). In Table 6, the readings were corrected for uniform body length by excluding juveniles for better comparison. The regression of body length and AFDW of ovigerous females and other individuals were calculated. The correlations between body length (mm) and AFDW (|j,g) of total individuals (Fig. 11a ) and ovigerous females (Fig. 11b) are given by the following equations: Ovigerous females: In (AFDW) = -4.242 + 0.480 In (L) (r2= 0.83, P< 0.001, n = 205) (Range: size = 2.2 -7. 1 mm) Other individuals: In(AFDW ) = -4.284 + 0.462 In(L ) (r2 = 0.97, P < 0.001, n = 265) (Range: size = 0.8 - 7.2 mm) 3.9. Growthrate Mean growth rates were obtained from shifts in the modes of the length frequency histograms (Figs. 3 and 4). Most of the animals were of about 3 mm body length in March. The population was strongly bimodal in May, as the first of the young appeared in the samples. Modes 26 4.3 9 0.6 1 4.0 0 15.0 2 15.0 0 13.8 0 16.2 4 52.0 0 4> 24.0 0 s 3D d 2 V «3 04 HH S3 $ S§ SS S S r-l H ^TfdrHVJrîoÔ^OÔ l-l 00 NlH HH S d 3 V 3 rH s' C/3 OH U 0) d •oa om^OOOOiHt-» a >> t-t-r>©©©©r-lTl; V Z H O\r5©t^0rf©0\© Oi TH 9) »H FH rH m «S rH M 1 00 S » 3 d «5 V 0H ©V>fO©©©©foS t-4 d M s* s£ rg * s s ä 3 M rj«^' co t- M © © © © oo ON viiH«n©©«5©ooi-j rH s ricidvîoôdrîrHrô »1 rH ^ N rt ri H g 4» 0) © Ofi eo es V 3 1—1 J3 M § 0H •P N © •a rrnxoeoootoo V ,3 o H yt 11 rfH©wi(^f»5Trf»i^f TH 00 CI T* I-H rH an 3 d V 0H rrt--««©©©©V)Tt t-;Trrj©©©© HJ H) OU a ## PB m m .8 3 2 B * Ä g g « h h < 4-1 C .5 •"• •"" 4» 4) o> ^ «ai Q y ™g «S 'aSi cï ao,, z z c« 27 eo e •c 00 ws •3 a ss r- in 00 00 co r- ë ^ r» —< o\ CTv P-; CO OD 8 60 in in' CS o\ » r~ U C ao ~* es co -1 CS CO .e z •5 XI 2 «5 -a c a 0 c •a O *o "n 00 cs av ^H vo a 'fl>y -i c< r- y o\ 00 \o •* o\ r- •* co 1 Ö Ö Ö Ö 8 ddod 1U u i u X X X X N es 00 00 —. CS in CA SS CO in es vo CS Ov CO Ov 00 m 00 t-; CS CgS gm «5 •* •«t Tf t- in VO lC ta •* XI u + + + + + + + + c O vo vo in v£> vo o\ O 0 vo 0 r" •* vo c <3\ CM I—1 00 5 0 PI 00 CO VO r-~ 00 VO Ov f ON •* CO O •"*' co vô «3 'S § ö co (A "V • c Il II II II u 5. 0Ü ><>< >H>- r^ en r- co VO CS Ov Ov • ÎI?I?I3I *—1 o\ CK £* •* Tf CO CS O f-; p p | v a r- es •S' S 00 r- —• m yo cN 00 «n 8 00 a Tf •* •* TT m ^o ^( vj c c P © Ö Q Ö ö ö ö Ö 0 CD U Tj- f^ CS CO +1+1 +1+1 V •g "" Q CO CO CO CO cs co —1 -5 rf •* •* 't Tt- •<* rf •<* s A'£ S Tt Tt Tf •*' 0 Ö V g PH * •2 » * m in *3- — cfl in o - cs —• 00 g. 00 00 co co H2 60 0 00 2 ^2 m| r- ci vo 0) 0 U • Table6 .Compariso n betweenbod ylengt han dash-fre e dry weighto fovigerou s females from other individuals (pre-ovigerousfemale s and males)o fCorophium curvispinum in the River Lower Rhine.Studen t l-testswa s used to make statistical comparison of the ash-free dry weight in the ovigerous females and other individuals. Ovigero us females Other individuals Body ash-free dry Body ash-free dry length weight length weight Mean 4.80 120.78 4.77 106.10 SD 1.28 17.66 1.61 13.09 SE 0.09 1.23 0.08 0.91 N 205.00 205.00 207.00 207.00 Minimum 2.20 79.00 2.20 75.00 Maximum 7.10 153.00 7.20 128.00 Median 4.90 126.00 4.80 109.00* Lower 95% CL 4.63 118.36 4.61 104.31 Upper 95% CL 4.98 123.19 4.93 107.88 f-tests P< 0.001 shiftscoul db efollowe ddurin gJune ,July ,Augus tan dSeptember . However, inJul yan d September, when the subsequent second andthird generations were born,som e overlap occurred among the three generations. After the end of October, the mean growth of the remaining juvenile animals slowed considerably until the following January. Mean lengths of different generations of C. curvispinuma t Lobith (Fig. 12) and Nijmegen (Fig. 13)wer eestimate dbase do nth eincremen to fmea nlengt ho feac hcohor tbetwee n subsequent samples during the months. The growth rate of C. curvispinumwa s observed to vary with age. Maximum growth rates (mean body lengtho f 1.3m mmonth" 1)wer e recorded inyoun ganimals .Th e animals also grew faster in the summer period when high temperatures (Fig. 14a) and high chlorophyll-a (Fig. 14b) concentrations are prevailed. 3.10.Production Production estimates were calculated using the increments of the mean length of each cohort for each month as a basis for weight calculations (Tables 7 and 8). Generally, the highest 29 200 In(AFDW )= -4.284 +0.46 2I n(L ) o (r2= 0.97,P < 0.001 , n= 265 ) S» 100 ET O 90 «^ 80 70 • A 60 t t 50 J I L 5 6 7 8 Bodylengt h(mm ) 200 In(AFDW )=-4242 +0.48 0I n(L ) O (r2= 0.83,P<0.001, n= 205 ) S & EP o 100 •• •*••!•*••* **** * it je 90 •f .W»* 80 • 70 - 60 2 3 4 5 6 7 8 Female bodylengt h(mm ) Fig.11. a. The relationship between body length (excluding ovigerous females) and ash-free dry weight (AFDW) of C. curvispinumi nth e Lower Rhine. b. The relationship between body length of ovigerous females and ash-free dry weight (AFDW) of C. curvispinumi nth e Lower Rhine. 30 Generations Sept 1991 May 1992 Jul 1992 Sept 1992 May 1993 Jul 1993 Sept 1993 1 - I st generation 2 - II nd generation 3 - III rd generation E £ 4 >. •o o .o c « o 2 FMAMJ JASONDJ FMAMJ JASONDJ FM Months (1992-1994) Fig.12. Mean growth rate (mm) of different generations of C. curvispinuma t Lobith in the Lower Rhine. Generations -©— Sapt 1991 -•— May 1992 Jul 1992 Sept 1992 -0 May 19S3 -• Jul 1993 Sept 1993 1 - I st generation 2 - Il nd generation 5 - 3 - III rd generation e S 4 at c * 3 >. O S 2 FMAMJ JASONDJ FMAMJ JASONDJ FM Months (1992-1994) Fig.13. Mean growth rate (mm) of different generations of C.curvispinum at Nijmegen inth e Lower Rhine. Body length Temperature 31 o o c MAMJJASONDJFMAMJJASOND Months (1992-1993) Body length -•— Chlorophyll-a B E E ca jD > CL O O J3 c «8 © ü MAMJ J ASONDJ FMAMJ J ASOND Months (1992-1993) Fig.14. Meangrowt h rate (mm) of C.curvispinum wit h(A ) water temperature and (B) chlorophyll-a in the Lower Rhine at Lobith. productionwa sfoun d inth efirs tgeneratio n (May-June) whencompare dwit hsecon d(July ) andthird generations (September). Incomparison , production was much higher at Nijmegen (1079 mg m"2) than at Lobith (34 mg m"2) for the period between April 1992 and February 1994. 32 a o i •a 9\H(StO>nvi(NU)3 aQONtVHO\«vfih.\fie u 1 9 "g a •B 3 - WO I9 uO 5i . 2 si a »H c* *H • *«r • n ' » t* fl H ' ' H • •9 11S oOS\ N«NO\V}enNHNN(SNHU)DV)O\0B«tfl 5,J a < £; es 1 a ooce«f)\ev>e«nNtfi« rH »*ï ea ! 1 i o\ t^ 90 0\ oo ooSKSt^ooAftooSw .000 0 00 t*« 000 0 ta _o < H rj %•» Ol »Q»nifl'n'flHfn»brt«floiflOioinv>oiflOifl 9 „ Z o J ^•«t^,*THOOorïi«vor§'*«oo«r(NopŒ^5o §" HioNShHamw'OÄnHNjtoov^ftniflO a. 6s & ! S a ^1I s e 1 HN«**«tfj9jhqo^r)hOHrt\eo»hïh Ê r*i t-f^om^-imMnoo^M^fnmooAsoooäb^f^ s Sw cro p u 2 (m g m ) S Standin g •3 (^Hn«Q9\AexoQl«rnin«osHh»««4»vo 1-H ^ >* V© PH .-H ^T«00(OHH(N FH N 2 I (m ) 'S densit y a Populatio n 2 ^inrjt-vi^^5«S'nt-r "E* individua l e o. OhaOOOQOOOWU^OOihOKIWOOhOQWTtulOnh a es NNNNNNNlflNNnNNNNNNNflRlflN Ü (A S* o»\ o> K 3 u "O a A L- HXVim0tWlNX>ÛNt"l0>*C«Ob*OO\CN M o H 33 1 1 a. •M [^. • • rt rt rt H H M H 1 % N« M rt PI ' H rt rt« OH »H • S.a. !^c oor-.25oo<3*Hc<£vor*.©(*i^wt'-i * m i> H \o «s m " 0»ttP«hrtr(NHWl' Mf^ooMfONOt^««t-4(n(QO\'*i-(r3Qp^r^««'q,o^- «si se § s i * S ! S ï t 00 r- t- r- b* oo i " % S 2 "S'S % M 00 M © l/ï «S H H H^OMt»l«ifl^«H H«HQocti«2HO\^uuie)rje(<]hVji B Hh9ÎN>ûrt(Shlflh|'tOnrt,«H«i §•J *-" t" «5 ^ «* ' nooH » »o invi H ni a © o © o o o o t» in IH * t^ (*Ï t- s*§ss wiàsîriô\©§5ooo\(^&®o\ rt(no^«o>wo»«nN«iHHNirciiN9\9\or) ooTfOH^ftrtNtinnh^riocoi-iriooDinoft »Q0r^r>>ta--t^0000Q00000»O00r^t--0000Q0Q0Q00000 £ .s 'S — oQoqaQoooQOQooQe-^asr^oQoQfS^ooooNvor--©^!^ a|* §3 ï &S 5 s g- g «fe,« 1g 3 = &S o a g -g NOM/lN»tJNIflnN9\h«)trtftlflHHh.(») C*fS^H^H©ö©© 3.11.Mud-fixation 3.11.1. Onstones Mud-material including macroinvertebrates fixed by C. curvispinumvarie d from 91 to 658 g m"2dr y weight and from 10t o 138g m'2 ash-free dry weight on the stones inth e Lower Rhine at Lobith(Fig .15) .I nNijmegen ,th eamoun to f mudfixe db y C.curvispinum o nth estone s rangedfro m 38 to 1044 g m'2 dry weight and from 9 to 134 g m'2 ash-free dry weight (Table 9). The highest valueswer egenerall y observeddurin gsumme r months (Julyt oSeptember) .Th e higher population densities of C. curvispinum were also observed during summer period when the favourable hydrographical conditions such as higher water temperatures, high availability of food etc., existed in the River Rhine.Th e lowest values coincided with lower temperatures during the winter period. A linear relationship wasfoun d between population densities of C.curvispinum an d muddy material (dry weight andash-fre e dryweight ) onth e stones (Fig. 16an dTabl e 10).Th e correlation between mud material (g m"2) and densities of C. curvispinum (individuals m'2) are given by the equations (y = dry weight and ash-free dry weight in g m"2 of mud material; x = densities of C. curvispinumi n individuals m"2): Dry weight Lobith: y = 228.27 + 0.003x (r = 0.74, P<0.001) Nijmegen: y = 160.01 + 0.004x (r = 0.86, P<0.001) Ash-free dry weight Lobith: y = 30.56 + 0.0006x (r = 0.84, P<0.001) Nijmegen: y = 29.72 + 0.0006x (r = 0.82, P<0.001) The calculated values of AFDW of the mud material fixed per individual shows that these values were higher during November to April at Lobith (Fig. 17).Th e higher AFDW of mud material per individual coincided with the non-breeding seasons and higher mean individual body length of C.curvispinum durin gthes eperiod s(autumn ,winte ran dspring ) (referbreedin gseason san dgrowt h rates of C.curvispinum i nsectio n 3.5 and3.9 ,respectively) .Th e seasonal pattern of changes inth e AFDW of mud material per individual at Nijmegen were largely similar to that of Lobith (Fig. 17). In 35 200000 160000 JASONDJ FMAMJ JASONDJ F 1992 | 1993 | 1994 Months 200000 1 o -^ JASONDJ FMAMJ JASONDJ F 1992 | 1993 | 1994 Months Density -•— AFDW Fig.15. Population densities of Corophium curvispinum and ash-free dry weight (AFDW) of mud material (includingmacroinvertebrates )o nstone sa t Lobithan dNijmege nfro mJul y 1992 to February 1994. 36 ao2!s2!D9S£:îao22S2S^'*^SNO«KrtrtîoS rtoeoooeeooeeoorteortâooeoo cSodööooöoöööööoöoööööocJö ? OC ^ 5. s; 3 3 M» rt rt rt rt rt rtrttNfJNNNTfNrt ' • • •f>i*iW'»f>«/>rl©'<»oOf*><- rt NrttN^^VOrtrt H rt ri H rt N N rt a13^ X it •o in § N o OOOONOrtrtnOOOOOOOOOO o öööööööööödoooööödö Ol E .s fa. 5 Ml j= Ol a tO:l , u o5N»»inMoinî»ntn J^rtaot>o • • • • • • • • B> s Soi-« f> »»A«ainr»w»ai wui m o u u I a Z osn a o\ mnawwnNnooon^^oiflrtoart H3 m m rt rt NHNN«o»»n»i»4N s u il"* q eu CA .y a a M (A o '*tN\0©C<©' e O» M E O o o S S S o d ? ? V ? 3 e CL, eiu eu eu eu Cu CS •4-» M v> Tf *T 00 00 a 00 00 00 l^ t~ I-. d d d '0 il II joa 11 i ! e e e © o O o °u » A •** •? Wl sV) 2 o> ta § -ï s • 1'x 5M M 'x 'x M o en Tf V» r* 00 ?H 'S £ rH 00 a 3 o o o | d d o U i + d fo <*î ^5 3 o r+» 00 »H N 9\ 00 i-i VO 175 1500 • Lobith: y (AFDW) - 30.56+0.0006x (r-0.84, P<0.001) O Lobith: y (DW) - 228.27+0.0025x (r-0.74, P<0.001) • Nijmegen: y (AFDW) - 29.72+0.0006x (r-0.82, P<0.001) S 140 O Nijmegen: y (DW) - 160.01+0.0044x (r-0.86, P<0.001) 1200 * Weurt: y (AFDW) - 18.70+0.0004x (r-0.74, P<0.001)* A Weurt: y (DW) - 87.70+0.0017x (r-0.71, P<0.001)* 6 60 900 00 O 600 300 30000 60000 00000 120000 150000 Density (N nr2) Fig.16. Population densities of Corophium curvispinuman d mud material (dry weight and ash-free dry weight) on stones at Lobith and Nijmegen and on tiles at Weurt (* data represents the tiles exposed in a cumulative way at Weurt). 3.11.2.On experimentaltiles A clear pattern was found in seasonal variations in the population densities of C. curvispinum and the amount of mud-fixed on the experimental tiles (monthly exposed as well as cumulatively exposed) at Weurt (Figs. 18 and 19). Colonisation by C.curvispinum fro m the water column appeared to take place during May-November with peaks in July-September. This is most probably due to a high reproduction rate inth e later period (refer to section 3.5 breeding seasons). On monthly exposed tiles, the ash-free dry weight of muddy material were ranged from 0.3 g m"2 39 ÜUUUUU 0.55 - 0.44 150000 co « - 0.33 •o > 100000 m c 0.22 o» o O E 50000 ~r 0.11 e < < HH i-B I-». r« n ri ^« 0.00 JASONDJ FMAMJ JASONDJ F 1992 | 1993 |199 4 Months Lobith Nijmegen AFDW AFDW (Lobith) (Nijmegen) Fig.17. Population densities of Corophium curvispinuman dash-fre e dryweight so f mudmateria l in mgindividual' 1day' 1 on stones (including macroinvertebrates)a t Lobith and Nijmegen from July 1992t o February 1994 (November 1992) to 63.9 g m'2 (September 1993). On cumulatively exposed tiles, the lowest and highest values were 0.3 g m"2(Novembe r 1992) and 101.8 g m"2(Jul y 1994), respectively. The dry weight of mud material on the monthly and cumulatively exposed tiles show a distribution pattern of high and low values similar to that found for ash-free dry weight (Figs. 18 and 19). The higher valueso fmud-fixatio nwer eobserve ddurin gsumme rperio dwit hlowe rvalue sbein grecorde ddurin g winter months. In comparison,th e cumulatively exposed tiles showed relatively higher densities of C. curvispinumdurin g the winter months (Fig. 19).Thi s can hardly be due to new colonisation as can be seen in Fig. 18, but may be possibly duet o a portion of autumn generation animals being maintaineddurin gth ewinte rmonths .A stron gpositiv ecorrelatio nwa sfoun dbetwee nth e population 40 20000 NDJ F M A M J J ASONDJ F M A M J J ASO 199? 1993 J 1994 Months Density AFDW Fig.18.Seasona lvariation si nth epopulatio ndensitie so fCorophium curvispinum an dmu d material onmonthl y exposedtile sa tWeur tfro mNovembe r 1992t oOctobe r 1994. densities of C. curvispinum (individuals m'2)an dth eamoun t of mudmateria l (ash-freedr yweigh t anddr yweight )o nth etile s( gm" 2)a tWeur t( y= dr yweigh tan dash-fre edr yweigh ti ng m" 2o fmu d material;x = densitie so f C. curvispinum inindividual sm" 2): Monthlytile s (Fig.20 ) AFDW: y = 2.578 +0.0007 x (r= 0.90 ,P<0.001 ) DW: y= 10.503+ 0.0023 x (r= 0.89 ,P<0.00 1) Cumulativetile s (Fig.16 ) AFDW: y= 18.70+ 0.0004 x (r= 0.74 ,P<0.001 ) DW: y= 87.7 0+ 0.0017 x( r= 0.71 , P<0.001) 41 240000 105 90 M^^ 180000h E 75 w *rf JZ 60 .? '© 120000h * to >. c 45 •o o o O eh. 30 • 60000h co < i!5 N D M J 1992 1993 Months 240000 105 180000 h 120000 o 60000 h N D J A M J 1993 J 1994 Months Density AFDW Fig.19.Seasona l variationsi n thepopulatio n densitieso fCorophium curvispinum an d mudmateria l on tilesexpose d ina cumulativ e waya tWeur tfro mNovembe r199 2t oOctobe r199 4 (A - SeriesI ;B -Serie s II). 42 Themonthl y exposedtile s showhighe rcorrelatio n value (AFDW: r=0.90;DW : r=0.89 ) than cumulatively exposedtile s (AFDW: r=0.74;DW : r=0.71) .Thi s may be in part duet o the deposition of mud material by other macroinvertebrates as a result of aggregation or, alternatively, higher mortality (due to shorter life span), competition for food and space, reproduction and erosion processes which make the correlation lower on cumulatively exposedtiles . • AFDW o DW 20 75 N AFDW: y - 2.578 + 0.0007x (r-0.90, P<0.001) B DW: y -10.503 +0.0023 x (r-0.89, P<0.001) 60 o CS 16 - • 60 I B ÖO bo 12 - 45 30 a) • o o I -a 15 CO < •8 -L _L _L 4000 8000 12000 16000 20000 Density (N nr2) Fig.20. Relationship between populationdensitie s of Corophium curvispinuman dmu dmateria l(dr y weight andash-fre e dryweight ) on monthly exposedtile sa tWeur tfro m November 1992t o October 1994. A clear relationship between population densities and mud material per individual of C. curvispinumo nmonthl yexpose dan dcumulativel y exposedtile swa sestablishe da tNijmege n(Figs . 21 and22) .Th e highestvalue so fAFD W per individual areobserve ddurin gJanuar y toApril . During this period,th eoverwinterin g population begant o increasefro m mean body length of 2.7 mmt o 4.9 mm (for details refer to section 3.9 growth rate).Th e period of highest values of the mud material 43 30000 >» ca •o w"* 20000 « E Z > >> "O CO c c o O 10000 o NDJ FMAMJ JASONDJ FMAMJ JASO Months (1992-1994) Büi Density o AFDW Fig.21.Seasona lvariation si n thepopulatio ndensitie so f Corophiumcurvispinum an dmu d material (ash-free dry weight in mg individual"1 day"1) on monthly exposed tiles at Weurt from November 1992t oOctobe r1994 . per individual coincidedwit h the lowest population densities of C.curvispinum at Nijmegen. It is possible that the competition for space and food is limited during this period (January - April) because of the lower population densities of C. curvispinum, or that the population is on the decreasewhil emu di sno teroded .Nevertheless ,i tmus tb epointe dou ttha tth epresen tdat anee d tob eaugmente dwit hmor edetaile dstudie so ntub ebuildin gi nrelatio nwit hth especiatio no fmudd y materials,i nvariou s regionso fth eRive rRhine . Thecalculate dvalue so fchange si nth eamoun to fmu d fixation(Aw )ove rth eperio do ftw o yearsindicat eth eclos ecorrelatio nwit hchange so fpopulatio ndensit y(AN )o fC. curvispinum (Fig. 23).Th ehighes tvalue so fincreas ei npopulatio ndensit yan damoun to fmu dmateria lwer eobserve d in July 1993. This must be related with the breeding season of C.curvispinum at Lobith and Nijmegen.Durin gth emont ho fJul y 1993,th e maximumnumbe ro fjuvenile swa s recordedo nth e stones.Th emaximu mdecreas ei nth epopulatio ndensit yan damoun to fmu dmateria lwa snotice d 44 300000 0.20 240000 - - 0.15 Ol •o M 3 E 180000 - •jo > - 0.10 JE c O) o 120000 - O E - 0.05 60000 - O 0.00 N D M A M J 1992 1993 Months 300000 0.20 240000 - N E 180000 - c 120000 o 60000 N D J M A M J 1993 | 1994 Months Density AFDW Fig.22. Variationsi n thepopulatio ndensitie so fCorophium curvispinum an d mud material(ash-fre e dryweigh ti nm g individual"1 day"1)o n cumulativelyexpose d tilesa tWeur tfro m November 1992 toOctobe r 1994( A- Serie sI ;B -Serie s II). 45 150000 75000- co c o O -75000 -150000 ASONDJ FMAMJ JASONDJ F 1992 | 1993 | 1994 Months Lobith " Nijmegen 100 E O) g> 'to S £• •o co co to < -100 ASONDJ FMAMJ JASONDJ F 1992 | 1993 | 1994 Months Lobith Nijmegen Fig.23. Change in population density and amount of mud material of C.curvispinum on the stones at Lobithan d Nijmegen. 2 2 [Change in density AN (N m" )= (Nt-Nt+1); Change in weight Aw (g m" ) = (w,-wt+1)] 46 200000 100000 - E z (0 e a> O -100000 - -200000 Nov Dec Jan Apr May Jun Aug Sep 1993 J 1994 Months Nov Dec Jan Mar Apr May Jun Aug Sep 1993 | 1994 Months Fig.24. Change in population density and amount of mud material of C. curvispinum on the cumulatively exposed tiles at Weurt. 2 2 [Change in density AN (N m" ) = (Nt-Nt+1); Change in weight Aw (g m" ) = (wt-wt+1)] 47 in August 1993. During this period,th e number of ovigerous females began to decrease rapidly in the Lower Rhine and, hence, the lower number of juveniles on the stones. There were also two minor peaks inth e increase of population density and mud material of C.curvispinum durin g May and October/November; again correlatedwit hth e breeding season of C.curvispinum a t Lobith and Nijmegen (refer section 3.5 breedingseasons) .Th echange so fdensit yan dmu dmateria lals osho w clear correlation on cumulatively exposed tiles at Weurt and again with the highest decrease of densities, as well as mud values, inAugus t (Fig.24) . 250 375 625 Number of days Lobith (AFDW In MgIntf'day' ) —Q Nijmegen (AFDW in pg ind ' day ') Fig.25. Meanvalue s of mudfixatio n ratepe r individual calculated fromth e cumulative density of C. curvispinuman d cumulative amounts of mud fixed on the stones at Lobith and Nijmegen. 1 1 [Mud fixation rate (AFDW in \ig individual" day ) = [£Wwtj£(N,+NJ]/Z(dt4 The mean values of mud fixation rate per individual based on the cumulative densities of C.curvispinum an dcumulativ e amountso f mudfixe do nth e stonesa tLobit han dNijmege n indicate 48 a constant levelo f about2. 5 ng individual"1day" 1( n= 1,700,000) after 400day s (Fig.25 ) andabou t 2.3 ng individual'1 day"1 (n = 500,000) on the tiles at Weurt after 350 days (Fig.26) . A significant difference was observed in the cumulative AFDW (u,g individual'1 day'1) of mud material on the stones between Lobithan d Nijmegen (t-tests, P<0.05;Fig .25 )an dthi s ispossibl y duet oth e higher mean body length of C. curvispinum at Lobiththa n Nijmegen (Fig.27) .Th e percentage of organic matter of the mud material including macroinvertebrateso nth e stones rangedfro m 9% (November 1992) to 23% (July 1992) with the mean value of 15.9% at Lobith (Fig. 28 and Table 11). At Nijmegen, the highest and lowest values were 13% (August 1993) and 20% (March 1993), respectively (Mean± SD; 16.0± 2.4).N osignifican tvariatio nwa sfoun di nth epercentag eo forgani c matter between Lobithan dNijmege n(Wilcoxon-signe d ranktest ,P>0.05 ) (Fig.29) .Th e percentage 200 400 Number of days Fig.26. Meanvalue s of mudfixatio n rate per individualcalculate dfro mth ecumulativ e density of C. curvispinum and cumulative amounts of mud fixed on the cumulatively exposed tiles at Weurt. 1 1 [Mud fixation rate (AFDW in jig individual" day" ) = [I(w.+w,.l)/2(Nt+N,1)]/i(dt+dM)] 1=0 1=0 1=0 49 3.70 300 400 700 Number of days Lobith —©— Nijmegen Fig.27. Mean body length of C. curvispinumcalculate d from the cumulative monthly mean length and density of C.curvispinum o n the stones at Lobith and Nijmegen. [Mean body length (mm) = (I NxL)/(E N)] t=0 t=0 25 20 i 15 S 10 i i i J I i i i i J I I I I L _L J ASONDJ FMAMJ J ASONDJ F 1992 | 1993 | 1994 Months Lobith Nijmegen Fig.28.Th e percentage of organic mattercalculate dfro mth e dryweigh t andash-fre e dryweigh t of mud material at Lobith and Nijmegen. [Organic matter (%) = (AFDW/DW) x 100] 50 Table 11. The percentage of organic matter in the mud material on the stones at Lobith and Nijmegen and onth e tiles at Weurt. Organic matter(% ; Lobith Nijmegen Weurt Mean 15.90 16.00 21.42 SD 3.70 2.38 4.28 SE 0.83 0.53 0.87 Median 16.00 16.50 22.30 Sample size 20.00 20.00 24.00 Minimum 9.00 13.00 11.40 Maximum 23.00 20.00 29.00 Upper 95%C I 14.17 14.88 19.61 Lower 95%C I 17.63 17.12 23.23 25 Wilcoxon-signed rank test, P>0.05 20 - * y • • yS • • • jr • yS • • a *- s^ • 0a>0 at V h 15 - • y/ • a u yS • • •••» (c0 • yS + • • • 3" o> 10 - / I II 5 10 15 20 25 Organicmatte r (%) Lobith Fig.29. A comparison of organic matter (%) in the mud material on the stones at Lobith and Nijmegen using Wilcoxon-signed rank test. 51 oforgani cmatte ro fth emu dmateria lexcludin gmacroinvertebrate sa tWeur tshowe dsligh tvariation s and it was significant (Wilcoxon-signed rank test, P<0.05; Mean = 21.4 ± 4.3; Range = 11-29) when compared with Lobith and Nijmegen (Table 11). 3.12. Filtrationcapacity Our theoretical estimation indicates that the C.curvispinum populatio n in the River Rhine possesses a mean filtration capacity of 4.7 x 106cm 3m" 2day" 1(Tabl e 12).Th e mean filtration rate of C.curvispinum i n River Meusewa s 0.3 x 106cm 3m' 2day" 1.Th evariatio n inth efilterin g potential of C. curvispinumi nth e Rivers Rhine and Meuse was due to the effect of much higher population densities inth e River Rhine than in the River Meuse (Fig.30) . However, a thorough investigation 1Ü IUUUUU cC^ - ^r* 'E '-o <0 CM (0 i — 75000 E 9 • 1 3 COE , > ts ç•5 CO Q. & E 6 50000 HjH j — 'i(U o> TÏ c c CD .o H— « •o 3 CD 3 25000 Q. 3 8. O c CO •8 o 0 0 1 1 I I i \ \ i \^ i \ nr i \ \ i \^ 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 Sampling locations Fig.30. Filtering impact of Corophium curvispinumo nstone s atvariou s places in Rivers Rhinean d RiverMeus e(refe r Fig. 1 forsamplin glocations) .Mea npopulatio ndensit yo f C.curvispinum was calculated from the five sampling depths (1- 5 m at 1m interval; at each depth three to five stones were collected) at each sampling location during September 1993 (water temperature 12°C) . 52 Table 12.Mea npopulatio n densities andpopulatio nfiltering impac t of Corophium curvispinum on stones from groins at various places in River Meuse and RiverRhine . Sampling International Sampling Average Population place River distance date population filtering (km) density* impactb (numbers.m2) xlO5 ±S.D (cm'-m'^d'1) River Meuse Ooi 76 22.09.1993 4±4 0.01 Broekhuizen 122 23.09.1993 92±156 0.11 Bergen 140 23.09.1993 111±127 0.13 Cuijk 162 24.09.1993 18±20 0.02 Grave 174 24.09.1993 9±8 0.01 Gewande 214 28.09.1993 563±700 0.68 Genderen 233 28.09.1993 189±277 0.22 Keizerveer 246 29.09.1993 17±14 0.02 Amer 256 29.09.1993 22269±12528 26.72 Total 23272 27.92 Mean 2586 3.10 River Rhine Lobith 859 14.09.1993 47721±9328 57.27 Nijmegen 891 14.09.1993 76747±25830 92.09 Tie! 917 14.09.1993 94633±24900 113.56 Vuren 951 17.09.1993 30506±6869 36.61 Rhenen 910 17.09.1993 40136±8510 48.16 Ravenswaay 930 17.09.1993 26510±2942 31.81 Bergambacht 971 16.09.1993 30±29 0.04 Lekkerkerk 980 16.09.1993 324±187 0.39 Total 316607 379.93 Mean 39576 47.49 a Average population density was calculated from the 5samplin g depths (1 -5 m at 1m interval). At each depth 3 to 5 stones were collected. b Filtration rate values were not available in literature (Corophium sp.). Filtering impact was calculated based on our preliminary experimental value of 5 cm3 animal"1 h'1 (Van der Velde, Unpubl.). 53 is necessary to calculate the actual filtering impact of this filter feeder since there is no such data available for C.curvispinum in the literature and it is known that several factors (e.g. silt content, temperature, salinity etc.) can influence the filtration rate of filter feeders. 3.13. Seasonal variationsin population densities of macroinvertebrates in the LowerRhine Thepatter no fseasona lvariation so f mostimportan t macroinvertebrates onstone sfro mth e groins in the Lower Rhine at Lobith and Nijmegen, are given in Figs. 31 to 38. The maximum abundance of C.curvispinum wa s observed duringth ewarmes t months (June -September ) inth e River Lower Rhine at Lobith and Nijmegen (Fig. 31). The highest density of 168,100 + 27,495 individuals m"2wa s recorded at Lobith inJul y 1993.Th e highest densities generally coincided with the breeding season of C.curvispinum (refe r section 3.5 for breeding seasons). The highest densities of D.polymorpha wer e observed duringth esumme r period at Lobith and Nijmegen (Fig.31) .Th e settlement rates were significantly lower at Nijmegen when compared to Lobith (ANOVA, P < 0.001). A maximum density of 131 + 59 individuals m"2 (mean + SE) was recorded at Lobith in July 1992. D.polymorpha requires a solid substrate for the settlement of its larvae. However, it was often observed that all stones at Lobith and Nijmegen were covered by a 1- 4 cm thick layer of muddy tubes of C. curvispinumleavin g less space for settlement for other macroinvertebrates. Thepopulatio n densities of Gammaridae,Gastropod a (Fig.32) ,Chironomidae ,Trichopter a (Fig. 34),Tricladid a and Hirudinea (Fig.35 ) were highest from Aprilt o October inth e Lower Rhine. During this period, temperature and food availability was also found to be at its maximum in the Lower Rhine (Fig. 7a). The higher temperature and relatively higher availability of food apparently influenced densities of macroinvertebrates in the Lower Rhine. The density ofth eGammarida e populationwa shighe r at Lobiththa n at Nijmegen (Fig.32) . At Lobith,ther e was also a significant difference inpopulatio ndensitie s betweenth e two species of Gammaridaeviz. , Gammarustigrinus Sexto nan dEchinogammarus ischnus (Stebbing ) (SNKtests , P < 0.001; Fig. 33). However, no difference was observed between the distribution of two Gammaridae species at Nijmegen (SNK tests, P > 0.05). The population densities of Corbicula fluminea(O.F .Müller ) and Corbicula fluminalis(O.F .Müller ) onth e stones also showed novariatio n between Lobith and Nijmegen (SNK tests, P> 0.05) (Fig.37) . 54 200 160 - 120 - AMJJASONDJFMAMJJASONDJFMA 1992 I 1993 I 1994 200 AMJ JASONDJ FMAMJ JASONDJ FMA 1992 I 1993 I 1994 tsmmm Lobith liiilal Nijmegen Fig.31. Seasonalvariation s inpopulatio n densities (mean± SE)o f Corophium curvispinum and Dreissenapolymorpha fro mApri l199 2t o March 1994o n stones( n= 3-5 )fro m groinsi nth e LowerRhin e atLobit h and Nijmegen. 55 4500 AMJ JASONDJ FMAMJ J ASONDJ FMA 1992 I 1993 I 1994 3000 AMJJASONDJFMAMJJASONDJFMA 1992 I 1993 I 1994 Lobith i J Nijmegen Fig.32. Seasonalvariation s inpopulatio ndensitie s (mean± SE)o f Gammaridaean dGastropod a fromApri l 1992t oMarc h199 4 onstone s( n= 3-5 )fro mgroin si n theLowe rRhin ea tLobit h andNijmegen . 56 3000 300 _ 2500- E z m 2000- -20 0 W 3 3 C C 'w SI O) O ~ 1500- m Hl o o >» 1000 100 ••-• «o 'm c c o o O 500 o AMJ JASONDJ FMAMJ JASONDJ Months (1992-1994) 3000 400 E 300 z (O 3 C .C O -20 0 w ui H— O 100 >. 'co c AMJ JASONDJFMAMJ JASONDJ Months (1992-1994) UB! G. tigrinus L I I E. ischnus Fig.33.Seasona lvariation si npopulatio ndensitie s(mea n± SE)o fdifferen tspecie so fGammarida e {Gammarus tigrinus and Echinogammarus ischnus) from April 1992 to March 1994 on stones (n= 3-5 ) fromgroin s inth e Lower Rhinea t Lobithan dNijmegen . 57 2000 AMJ J ASONDJ F M A M J J ASONDJ FMA 1992 I 1993 I 1994 3000 Trichoptera W c _s_ — -"M AMJJASONDJFMAMJJASONDJFMA 1992 I 1993 I 1994 Lobith Piüli Nijmegen Fig.34.Seasona lvariation si npopulatio ndensitie s(mea n ±SE )o fChironomida ean d Trichoptera from April199 2t oMarc h 1994o n stones( n = 3-5)fro m groinsi nth e LowerRhin ea t Lobith and Nijmegen. 58 2000 AMJ JASONDJ FMAMJ JASONDJ FMA 1992 I 1993 I 1994 500 AMJJASONDJFMAMJJASONDJFMA 1992 I 1993 I 1994 Lobith fflMH Nijmegen Fig.35.Seasona lvariation s inpopulatio n densities (mean± SE) ofTricladid a and Hirudineafro m April 1992t oMarc h 1994o nstone s( n= 3-5 )fro mgroin si nth eLowe rRhin ea tLobit han d Nijmegen. 59 UU T Corbicula 80 - 60 j to -- f c 40 O n 111II1 1li lR l11 1 III •II aM rl Ph1 1 iU y Ls ft J JASONDJ FMAMJ JASONDJ F 1992 I 1993 I 1994 Lobith HÉIII Nijmegen Fig.36.Seasona lvariation s inpopulatio ndensitie s (mean± SE)o f Corbicula fromApri l 1992 to March199 4 onstone s( n =3-5 )fro mgroin si nth eLowe rRhin ea tLobit han d Nijmegen. 60 35 35 Lobith 28 - - 28 E 2 «s o c 21 - - 21 E 3 2 14 - 14 ü o O « w c c o - 7 E (0 c E 3 ü O >. w c o Û J FMAMJ JASONDJ FMAMJ Months (1992-1994) lËBH C.fluminea I I C.fluminalis Fig.37. Seasonal variations in population densities (mean ± SE)o fdifferen t species ofCorbicula (Corbicula flumineaan d Corbiculafluminalis) from April 1992t o March 1994o nstone s( n = 3-5)fro m thegroin s inth eLowe r Rhine at Lobith and Nijmegen. Lobith Corophium 61 100 Gammaridae Dreissena a> Gastropoda o> *-es• c o 70 u V////A Chironomidae 5 a Trichoptera Tricladida 40 Hirudinea AMJ JASONDJ FMAMJ JASONDJ FM 1992 I 1993 11994 Nijmegen Corophium 100 Gammaridae Dreissena CD Gastropoda Ol as *•* c 70 CD Chironomidae U u CD Q. Trichoptera Tricladida 40 Hirudinea AMJJASONDJFMAMJJASONDJF 1992 I 1993 1199 4 Fig.38. Seasonalvariation s inpopulatio ndensitie s (percentage) ofvariou smacroinvertebrate sfro m April 1992t o March 1994o n stones (n= 3-5) from groins inth e Lower Rhine at Lobith and Nijmegen. 62 3.14. Distributionand impactsof C. curvispinumon othermacroinvertebrates The distributional abundance of important macroinvertebrates on stones from groins at different depths and locations inth e Rivers Meuse and Rhine, is presented inTabl e 13 and Table 14,respectively . Inth e River Rhine,th eC. curvispinumdensitie swer ehighe ra tLobit h(Rive rRhine) , Nijmegen,Tie lan dVure n (R.Waal ) andrelativel y lower at Rhenen (R.Nederrijn ) and Ravenswaay, Bergambacht andLekkerker k (R.Lek ) (Fig.39) .Th ehighe r densityo f C.curvispinum wa s generally observed inth e upstream parts of the River Rhine. Inth e River Meuse,th e highest densities of C. curvispinumwer eobserve d atAme ran dextremel y lowdensitie s were recordeda t8 othe r locations (Fig. 40). The average population density of C. curvispinumwa s about 100,000 individuals m'2 at manysamplin glocation si nRive rRhin edurin gSeptembe r 1992an dSeptembe r 1993.Th emaximu m C. curvispinumdensit y of 642,000 individuals m"2wa s recorded at De Steeg in the R. Ussel (Fig. 1a). The population densities of C.curvispinum i nth e Rhine branches viz.,R . Nederrijn/Lek (log y = -9.39 + 52.51x , r = 0.75, P<0.001; y = population densities of C. curvispinumi n individuals m'2; x = average stream velocities in m sec"1), R. Waal (log y = 3.95 - 0.84x, r= 0.87, P<0.001) and R. Ussel (log y = 2.82 + 3.50x, r = 0.93, P<0.001) showed a positive correlation with the average stream velocities (Fig. 41). The average densities of C. curvispinumi n the Rhine decreased in a downstream direction.Averag e C.curvispinum densitie s were lower inth e R. Nederrijn/Lek branch (varied from 13 individuals m"2 at Bergambacht to 47,463 individuals m'2 at Rhenen), which is dammed by weirs, than in the weir-free R. Waal (varied from 33011 individuals m"2 at Vuren to 111,915 individuals m"2a t Tiel) and R. Ussel (varied from 70,870 individuals m"2a t Olstt o 642,022 individuals m"2a t De Steeg). However, C. curvispinumnumber s (ranged from 2 individuals m'2 at Cuijk to 31,430 individuals m"2a tAmer ) showed noclea r correlation with average stream velocities (which ranged from 0.10 to 0.15 m sec'1) inth e River Meuse. In comparison, significant variations in the population densities of C. curvispinum, D. polymorpha, Corbiculasp. ,Gammaridae ,Chironomida e andTrichopter awer e recordeda t different locations in the River Meuse (ANOVA, P < 0.01) (Table 13). However, Gastropoda, Tricladida, Asellidae and Hirudinea didno tsho w any variation inth e distribution ofthei r population densities in the River Meuse (ANOVA, P> 0.05). Ephemeroptera and Cladocera showed very little difference inth e densities atdifferen t locations inth e River Meuse and Rhine (ANOVA, P< 0.05). Inth e River Rhine, significant variations in the population densities of C. curvispinum, D. polymorpha, 63 Rhine 1992 Corophium 100 1 SS Gammaridae Dreissena Tricladida Corbicula 10 11 12 13 14 15 16 17 Locations Rhine 1993 Corophium 100 Gammaridae Dreissena Gastropoda « O) al c Chironomidae a 50 o • a. I^HI Trichoptera I ••• \ Tricladida IMÏM Corbicula 10 11 12 13 14 15 16 17 tWSW other spec. Locations Fig.39. Variations in population density (percentage) of different macroinvertebrates on stones (n = 3-5) from the groins at different locations inth e River Rhine during September 1992 and September 1993. 64 t 3 a *a a aa a a Wl0 \o \v p K Si » m »a a i £ 3 £ *» «^ n n t o » ^ - Ü2 S !0 !S 35*«**' Orte« «s as cjt »* *r - ^ SOT f0 6 •j * *H ••»• H •>» •» 3F 8 ve < NÖÖ6 N » O ?] CJ 52g o« fô riTfJwî H p«» vi r4 wi < ClC *»- SC l t*.f»ï,-*,H m « v) H *££!££* NX O n OD « o« © ^ »-t Wl O rH ï9« s; a j- §38 ISS • «S'a • Hi CS *H W ii n 11 ii sags lias Hl! »ÄSS a « o, Ö ffi .7 M 41 S»âS*SB£ S«H38ss2? *"n|- K3t*r;?3 rH «^ iH DU' •S'a lîîï i- e , 4!<3. G 3 « » Si su s3 5«! 44 ^ fi *Sl3 3îîS M m t- S *1 in 1 s3ça QOS« j 8*»' Pi«« m o « 3 vc f4 *H W4 &* 3$«3 1^3 ••sa 53«! •H -H -H +1 «- 0\ * VO 1 1 « ve Sv ve ssga S" - * s <*ï v> *T w «sas » S!3 S "-1 S « S« H a 2 o r- o *-( H »H FH f sus •H -H « « 2 rt H M Cin «un» « N ro t "^ Q CS •a E I 5 u •3 •c o 'C 2 H Ü H U 65 * # * * * * * # * # * # a s Ï a * a : a s > o *s.a s sais 3g g s asqss s 3 ai 2« s s z Î3S lsv î ci*- ï© d ^ cii- 4c i ci© *HN Ô H(^ Nn H H r*n •4 3 •*m c* rrtf ti «3 vimm « vtm in H inV )3 n viv i VIH ggf sa* stp* aa» s$s •*<" 1« 3 s p isP 553? ^^is s-gs^s2|s ^ §|ss SUS SâiB SSal ä»äa g|£3 iï** **$j sgal p£as ^^^^ ll&lsRë* ' Tt^S^Î 3p| î§J5 fl$35 SUS ,200 Jg*s Un Hü a**ï Mas **" n^-niri 55 9!lS 1131sss i Hlä Haï 3 am ai ! sis? s S ggs§ s|sa 0\Ï S H rt Sff\ «1« ^SSl l^aSS B S"51*3 3*§2 3 *««* a«s 35S§ * S *a»s §sf§ SS§Ê Ov Si IQ © *H © »H «S »H ^ m ffj A H O f] N H H H *H *•* < N H H N •H -H -M -H * 41 •" «33 S # 5 S :* 3 £ £ «8 £<3 8 $3« » Ç> « m U\ vû ifl TT "> <*> OO Ü» •* t^ S« SS rtioS înS « O A«. * x 00 K. 00 »•nrtnî .K M >- 5^ ^Hcsm-'tf »-< «s I*Ï v »HMf^^t »H n m ** fHMn-* HMm* a. a •g g ? 8 o u I st •a E a. sa 66 n. s a a * * a a a aa a o o. nSon Oy ci x e) r- ir in » oo \© w ^ ••o N g « H S * - 3 « TT i-C «S 9 fe «s ~ $> i/i I o> H oó * t~ Cj C) (*J H ^ H S .-C S vp » < , « +1 -H aj§ a ss ass f> Sr 00 00 in o\ H »î q ^ -H -H S 2 3 «j 3 3 •H S * JJ «a v> S 50 ci •o a ci « f> t~ s *•* h* rH Ä a\ i g ci r- ve i.^ ••• y*A *•* oo •» >p N e M •*> •H t- Ci N H * » » N«rt N S S * 3 333 3 2 3o o c-i oo « « N » tfl fO 00 fi ^ 00 «ï N h v» n w >» i» •» t- o ri *o v» ts r» il ri < si •H -H -H -H 11 11 il il « -H -H H r> \e m »NO PH O m oo «0 4? 3 s H « N O Ä i-H TH a«*a m fï •* 00 o a o w Ä t/i a g nin» ai N N » i/i •a O 00 «::c ©«•"»fi o "3 M ** sSSs MOO'« H 'SH m •H ij j +i •H u -H -H a* s 2 e> o o i~- vi a '•tf)0 oa SSgS 00 oo •* oo se r; o !" VI 00 00 T3 "S « r » 2 3 3^ 4->t oj ci» m SPS« O ci ve o ä ÎO VI S?r53« 3 •a o. s. s •S ~ H Nn^t H M fO ^ •H (s m Tf »-H M n f t-i n ei Tr Hficirr al o E a c .1 E g" ••3 o. JS 1 •3 e •c J3 -a a •c H o 67 g g a a g no j» çj p> uj o\ r* oo 1*5 swj rj <*} * <-< ri m fi H O I») »O «H 00 t- O r» »H vo •* .H tf> 1 fl « -H o o", t- ** t- f*S i-H i-H Cl •H -H -H H J -H +î +1 -H -H Cî r* O rt •* in « te ^ i^?o• S S a I ££+<+< Ji* ü OKlSt oooo Se. 1 *4 oVo !« o i-i m tn _ in vç oo ui +4 -H -H -H J «J M 't n »i o o o ooo Vi OÙ VI ^t •Sil Î oo o O V> O ri O M C< •H o o o >H 1 1 1 J +) ü -H ü +) -H -H ü -H -H „ o « « Oleom O O f) ,4 •a ^ s- s rt Mrt^ H rt ffj ^ H NW t HNfl^ 3 a a s uo S Xi g ia f2 68 Gammaridae, Gastropoda and Tricladida were recorded at different locations (ANOVA, P < 0.01) (Table 14). The impact of C. curvispinum population expansion and its tubiculous settlement strategy onth e other previously dominant macrofauna has been notedwit h a simultaneous decrease inth e densities of D.polymorphs (Table 15).Th e population development of C.curvispinum an da nasso ciated decrease of zebra mussel, D.polymorpha populatio n was noticed in the River Rhine since 1987. In 1988, densities of D.polymorpha i nth e Rhine was found to be=100 0 individuals m'2an d itdecrease dt ohundred s in 1991(Paffe n etal., 1994 ;Va nde rVeld e etal., 1994).Th e presentstud y clearly indicatesth efurthe r decrease of the D.polymorpha populatio n inth e Lower Rhine (Fig.1b) . Meuse 1993 Corophium 100 Gammaridae WÊÊËÊ Dreissena $888881 Gastropoda o> « X c A Chironomidae o 50 o a> a. Trichoptera Hirudinea Corbicula 0 4 5 6 8 Other spec. Locations Fig.40. Variations in population density (percentage) of different macroinvertebrates on stones (n = 3-5) from the groins at different locations inth e River Meuse in September 1993. 69 70000Û » tmnm •— i 3 S s c I o a RiverMeus e log y = 5.34 - 25.52H (r =0.54 , p > 0.05) 1000 10 0.1 0.00 0.04 0.08 0.12 0.16 0.20 Stream velocity (m sec-1) Fig.41. Population densities of C.curvispinum (individuals m'2 stone surface) and average stream velocities( msec 1)alon gth ethre emajo rbranche so f RiverRhin e(R . Nederrijn/Lek,R .Waa l and R. Ussel) and River Meuse during September 1993. 70 Table 15.Mea npopulatio n densities(individual sm' 2ston esurfac e at 1 mdepth )o fzebr a mussel, Dreissena polymorpha on the different sampling locations in the River Rhine before (September 1989)an dafte r (September 1992an dSeptembe r 1993)th epopulatio n explosion of Corophium curvispinum. Sampling International Population density of Dreissena polymorpha location River distance (numbers m2) (km)a September September September 1989b 1992c 1993c Lobith 859 1700 39** 48** Nijmegen 891 2800 17** 154** Tiel 917 21000 210** 125** Vuren 951 18000 37** 237** Rhenen 910 13000 330** 2285** Ravenswaay 930 9100 1008** 27856** Bergambacht 971 1200 70** 221** Lekkerkerk 980 2500 5050** 5** Velp 885 100 30** — De Steeg 896 300 250* — * Ölst 957 95 600** — Wijhe 966 2500 1050** — a the difference between sampling locations are ±2 km in 1992an d 1993fro m 1989. b Bij de Vaate etal. 1992. c the differences between before (September 1989) and after (September 1992 and September 1993) the population explosion of Corophium curvispinum are compared by student's t-tests. **Significant levels:P<0.05 ;P<0.001 . — data not available. 4. Discussion In the Lower Rhine,th e maximum C.curvispinum densit y of 168,000 individuals m'2 (July 1993) was recorded at Lobith. The density of C. curvispinum in the Lower Rhine is one of the highest ever reported in the literature. The nearest density value of 100,000 individuals m'2 was reported by Scholl (1990) from River Rhine in Germany. The differences in population densities of C.curvispinum fro m different waters might be relatedt odifference s inth eamoun t of available food, 71 salinity,flo wvelocity , pollutionetc .Foo dcondition s inth eheavil y eutrophicated RiverRhin ema yb e relatively better because of a continuous supply of phytoplankton and suspended organic material provided by theflo w of the rivert o this filter feeding amphipod.Th e present population densities of C. curvispinum are relatively lower than those (750,000 individuals m"2) reported by Vande n Brink era/. (1993a) for the Lower Rhine. Inthei r population density study, only one stone with relatively maximum densities of C.curvispinum wa sselected . Inth e present study,th eaverag eo f 3-5 stones ata constan t deptho f 4-5 mwa s usedi norde r to assess the populationdensities . It isobviou s that the densities of C. curvispinum generally vary from stone to stone and with depth at any given location inth e Lower Rhine.So ,th eaverag evalue so f3 t o5 stone sa tconstan t depthwil l represent a better estimation of population densities of C.curvispinum i nth e Lower Rhinetha n those of Van den Brink etal. (1993a) . The highest densities of C. curvispinum were generally observed between May and September in the Lower Rhine; the lowest densities were recorded between October and March. Highwate rtemperature s (Fig.7a ) atLobit hwer e stronglycorrelate dwit hth e highestdensitie s of C. curvispinum(Fig .2) .Fo rexample ,i n1992 ,wate rtemperatur ea tLobit hsteadil yincrease dfro m 11.5 °C in April to 24.0 °C in August which was coupled with increasing densities of C.curvispinum. Temperature decreased from 19.5° C in September to 6.5 °C in February 1995,coincidin g with the decreasing population densities.Thi stren dwa s also evident inyear s 1993-1994.Th e high density of C.curvispinum i n 1993i sals ostrongl ycorrelate dwit hth ehighe rvalue so fchlorophyll- aa tLobith . The populationdensit y of C. curvispinumwa shighe ri n199 3tha n 1992a tLobit han dNijmegen .Thi s goes in handwit h chlorophyll-a values,whic h were also higher and more long lasting in 1993tha n 1992 at Lobith (Fig.7a) . The present study suggests that C. curvispinumha s an annual life cycle based on three generations at both Lobith (Fig.7b ) and Nijmegen (Fig.7c ) inth e Lower Rhine.Th e overwintering ovigerousfemale swer efirs tobserve di nMarc han dwer ewel lestablishe db yApri l(Fig .3) .Th e input of young animals was noted in May.Thes e young broods mature rapidly to reproduce in summer, andar e responsible for the subsequent generations of July and September. This conclusion is also supported by the disappearance of the overwintering population in June/July at both localities. Reproductioncease d inOctober .Th eprogen y ofth esumme ranimal s(generatio n2 ) alongwit h late autumn broods (generation 3) which do not mature by October, form the next overwintering generation. 72 C. curvispinum population in Lower Rhine was dominated by females (Fig. 5). The elimination of males from the population is possibly due to selective prédation by birds andfis h as reported by Muskó (1990an d references therein).Ther e are also other possibilities like shorter life span (Van den Brink et al.,1993a ) and higher activity (out of tube) of males than females (Muskó, 1992) and therefore a higher chance to drift away or to be preyed.Assumin g that males are more active, the possibility also exists that active males were not adequately sampled from the collected material. Ovigerous females of C.curvispinum wer eexclusivel yfoun d inth ewarmes t months (April- September) of the year at temperatures above 10 °C (Fig. 8a). This is in contrast to those of C. bonnelli M.Edwards and C.insidiosum Crawfordwhic har eals o reportedt o breeddurin gth e colder seasons (Sheader, 1978;Moore ,1981) .Unlik e C.curvispinum i nth eLowe r Rhine,othe rCorophium species,suc ha s C.volutator Pallas ,C. bonnelli, C. arenariumCrawfor dan dC. insidiosumhav eonl y two generations each year (Sheader, 1978; Fish & Mills, 1979; Moore, 1981; Möller & Rosenberg, 1982; Peer et al.,1986) . Variationo f broodsiz edat acollecte dfo rtw oyear sfro m Lobithan dNijmege n showsa clea r seasonal pattern (Tables 2 and 3).Th e brood size was always foundt o be relatively higher during May - July and lower during August - October. The greater body length at which the overwintering females attain reproductive maturity may be the reason for the differences in mean brood size betweenth eoverwinterin gan dth esumme rgenerations .However ,compariso no ffemal ebod ylengt h (mm) and number of eggs (N) dataobtaine d inthei r brood pouch between May -Jul y andAugus t - October indicates that there isa significant difference between the two sets of data (ANOVA, P< 0.001). This implies that the seasonal difference observed in the mean brood size of ovigerous females betweenoverwinterin g andsumme rgeneration s isno tdu esolel yt oth esmalle r body length of the summer generation. Brood size in amphipods is reported to be regulated by the interrelated effects of body length, metabolism,foo d availability and temperature (Fish & Mills, 1979; Van den Brink et al., 1993a). In the Lower Rhine, the maximum availability of food (chlorophyll-a) was recorded between April and June along with increasing temperatures (Fig. 7a). The observed maximum chlorophyll-avalue s coincidedwit hth e maximum broodsize so f C.curvispinum a t Lobith and Nijmegen (Fig. 8b). Furthermore, a significant difference in brood size of C. curvispinumwa s observed between Lobith andNijmege n inth e Lower Rhine (Tables2 an d3) .N odifferenc e inwate r temperaturean dchlorophyll- avalue swa srecorde dbetwee nLobit han dNijmegen .However , stream velocity was higher at Nijmegentha n Lobith (Fig.9) . Fora filte r feedingspecie s like C. curvispinum, food can be most easily obtained at sites where the flow is high, favouring a continuous supply of 73 food.Thus ,i t isreasonabl e toconclud etha tfoo davailability ,temperatur e andbod ylength ,togethe r with stream velocity, influences the brood size of C.curvispinum in the Lower Rhine. Inth e present study, the mean maximum number of eggs carried per female ranged from 4 (9 September 1992 at Lobith) to 25 (20 May 1992 at Nijmegen) (Tables 2 and 3).Thes e values are much higher than those reported for C.curvispinum from Lake Balaton (Muskó, 1992), Tihany peninsula (Entz, 1947; Muskó, 1989) and the River Volga, Oka and Don (Behning, 1925; loffe, 1954).Th e differences in egg clutch size of C. curvispinum from different waters might be related to differences in the amount of available food and the water movements. Food conditions in the heavily eutrophicated Lower Rhine may be relatively better because of a continuous supply of phytoplanktonan dsuspende dorgani cmateria lb yth eflo wo fth erive rt othi sfilte rfeedin gamphipo d (Van den Brink et ai, 1993a). The selected data on the level of egg loss from the brood pouch of different Corophium speciesfro mth e literature arecompile d inTabl e 16.Sheade r (1978) hasrecorde d egg lossa s high as 25-30%fro m the brood pouch of C.insidiosum atth e Cullercoats area on the North-East coast of England. The percentage of egg loss in C. curvispinumi s high when compared to the other Corophiumspecies , hence,th e chances of survival of C.curvispinum embryo s must be lower than those species reported in Table 16. Comparison of present results on C. curvispinumwit h those reportedfro m Lake Balaton,Hungar y (Muskó, 1992)an dth e Dutch Lower Rhine (Van den Brink et ai, 1993a)indicate stha tth eseparatio no fsexe si spossibl e in C.curvispinum whe n itha sa 1.8 mm body length (Table 17). The ovigerous females of C.curvispinum ca n also be distinguished in 2.2 mmbod ylength .I nth epresen tstudy ,th emaximu msiz eo f C. curvispinumwa s7. 2m mbod ylength . This value is much higher than those reported inth e literature. Growth rateo f C.curvispinum wa sfoun dt ob ehig hdurin gApri l- Septembe r (Fig. 14)whe n higherchlorophyll- aconcentration san dwate rtemperature swer eobserved .Durin gJul yan dOctober , separation between cohortswa s unclear from the length frequency histograms. From October until January, some growth was found to have taken place at both Lobith and Nijmegen although at a much slower rate than the summer period. Several periods with negative production values were observed at Lobith and Nijmegen (Tables 7 and 8). Peer et al.(1986 ) argued that the proportion of the biomass contributed by the young of the year did not form a smoothly increasing proportion of the population as a result of frequentdeat ho folde ranimals .Thi styp eo ferro rmigh thav ebee navoide db yusin gth edensit yan d sizedistributio ndat abase do nth esam eanimals .Usuall y biomasswa sestimate dfro mbot hdensit y andsiz edistribution .Thes etw o parameters aregenerall y believedt o havea lo to fspatia lvariability , 74 ON o\ vO 00 00 w CA f» >> >, 00 OS •o -o CU 3 3 o S S CA CA a ••-» +J eu B fi fi u PL! 'o B CA (30 O p o I uS C/3 3 CU s C 00 3 cd a CU 8 e o CU »•-* c 2. o o •o 3 S -o cj 3 E c S o S S O S .3 CA a I« 3 'o 3 *—* I I cj U > a C/5 S, eu Vi T3 .S B eu B .S O 1 P? B eu ca 3 > Cd U w eu -a > -a CO « C/3 eu CA 'ai eu ü ca ca o 11 o S? eu E CA o 1 Lowe r 43 ? rland s c/3 3 'S O - eu 0H 2 m 4e3u C 43 CA s ca J eu "S cd 6Û U Ö8 w43 43" eu Où ™ • •-H Z 33 C 43 eu .S6. *u S3 u 43 § 3 £ O :;? 43 -g Q W pu, co U U ES 55 P H 75 and hence the great variation in the production of C. curvispinum in each month at Lobith and Nijmegen. The success of C. curvispinum makes it one of the most important biotic factors in the functioning of the Rhine ecosystem. Inth e River Rhine, C.curvispinum was found both on stones inth e littoral zone and inth e sandy sediments atth e bottom.However , heaviest settlement occurs on hard substrates which are widespread in the river's littoral areas. The highest densities of C. curvispinumwer e recordedi nth e upstream partso fth e River Rhine branches [viz.,R .Waa lan d R. Ussel),whic h have higher average riverflo wan dstrea mvelocitie s (between 0.65 m sec"1a t Vuren Table 17.Compariso n of present results on C.curvispinum with those reportedb y Muskó(1992 ) and Van den Brink et al. (1993a). Lake Balaton Lower Rhine Lower Rhine Hungary The Netherlands The Netherlands Muskó, 1992 Van den Brink etal., Present study 1993a Sex differentiation Minimum length* 2.0 mm 2.5 mm 1.8 mm Ovigerous females Minimum length 2.5 mm 2.5 mm 2.2 mm Brood size Monthly mean 11.2 12 25.4 Range 1 -25 3-34 1 -42 Maximum size Body length 6.0 mm 6.3 mr 7.2 mm = Body length (from the tip of the rostrum to base of the telson); b = Number of eggs 76 Table 18.Seasona lvariation s of Sodium(Na +)concentratio n inth eRive rRhin e(a t Lobithan dAndijk ) and Meuse (at Keizersveer and Remily) during 1993 (RIWA, 1993). Sodium (mM) River Rhine River Meuse Months 1993 Lobith Andijk Keizersveer Remily January 2.70 2.65 0.87 0.30 February 3.31 3.65 1.00 0.26 March 4.35 3.44 1.57 0.39 April 3.87 3.65 1.87 0.43 May 3.87 4.22 2.17 0.39 June 4.05 4.35 2.31 0.48 July 3.48 4.74 2.61 0.57 August &04 4.18 2.52 0.61 September 3.87 4.61 2.78 0.48 October 3.09 3.91 1.35 0.61 November 3.44 3.26 1.52 0.30 December 3.31 3.35 1.22 0.30 Mean 3.53 3.84 1.82 0.43 Std. dev. 0.46 0.59 0.63 0.12 ANOVA P< 0.001 and 1.15m sec" 1a tTiel )tha nth ewide r partsan dth eweire d R. Nederrijn/Lek (between0. 1an d0.1 5 m sec"1). In contrast to the explosive range extension of C. curvispinumi n the River Rhine, the species density was much less inth e River Meuse.Th e maximum density of 31,430 individuals m"2 was observed at Amer, which is the connecting point of River Meuse with River Rhine. Excluding Amer, the population densities of C. curvispinum varied from 2 individuals m'2 at Cuijk to 731 individuals m'2 at Genderen. This is possibly due to the lower chloride values (CI) and higher pollution (e.g. Cd) in the River Meuse (For the year 1993; CI"1: mean = 53, range = 16-136; Cd: mean =0. 4 u.gI" 1,rang e = 0.1 -4.0 , RIWA, 1993)tha n River Rhine (For the year 1993; CI"1: mean = 151, range= 4 7- 236 ;Cd :mea n= <0. 1|i g I"1,rang e= <0. 1- 0.1, RIWA, 1993). Inaddition , Rhine water also has a relatively high ionicconten t (mean =3. 5 mMNa +; Range =2. 7 -4.4 ;measure da t Lobithdurin g 1993;Tabl e 18)tha n River Meuse (mean= 0. 4 mM Na+; Range= 0. 3 -0.6 ; measured at Remily during 1993),favourin g the occurrence of C.curvispinum, which needs a minimum of 0.5 77 mM Na+ (Taylor & Harris, 1986b). Moreover, stream velocities were also lower in River Meuse (maximum of0.1 5 m sec"1)tha n River Rhine (maximumo f 1.12 msec" 1).Fo r afilte rfeedin g species like C. curvispinum, food can be most easily collected at sites where the flow is strong enough for a continuous supply of food. The effect of flow rates on physiology and growth of different filter feeders hasbee nreporte db yvariou sworker s(Kerswill ,1949 ; Walne,1972 ;Perkinsd ,1974 ;Bayne , 1976). The highest values of mud-material including macroinvertebrates fixed by C.curvispinum were generally observed during summer months (Table 9). There was no significant difference observed in percentage of organic matter between mud materials including macroinvertebrates (mean percentage = 16+ 2 ,rang e= 13-20) andmu dmateria lexcludin gmacroinvertebrate s (mean percentage =21+4 , range =11- 29).Th e higher populationdensitie s of C.curvispinum wer e also observed during summer (June to September) when favourable hydrographie conditions such as higherwate rtemperature s (17- 2 0°C ,RIWA , 1993),hig havailabilit yo ffoo detc .existe di nth e River Rhine (Fig.7a) .Th e lowest values of fixed mud-material coincided with lower temperatures (6 - 9 °C,RIWA , 1993)durin gth ewinte rmonth s(Novembe rt oFebruary) .Moreover ,th edr yweigh to f mud materialpresen to nth e stoneswa shighl ycorrelate dwit hth epopulatio ndensitie s of C. curvispinum in the River Rhine (Fig. 16). The filter feeder C. curvispinum has increased dramatically in the River Rhine since its invasion in 1987. The population densities of C. curvispinumoutnumbe r other macroinvertebrate specieso nth estone so f groins inRive r Rhine.I na nenvironmen twher eth esubstratu m isa limiting factor, animals colonising it employ a wide range of competitive strategies to secure that resource (Dayton, 1971; Buss , 1981).However , localfactor s(physica lan dchemical ) maypla yver y important roles,modifyin g competitive capabilities of anorganis m sotha t a particular type of organism,whic h may becomedominan t ina particula r environment, may notd o so ina differen t environment. Inth e River Rhine, it is probable that the groins form practically the only available solid substrate for macroinvertebrates particularly C. curvispinum, D.polymorpha an dGammaridae .I nth eRive rRhine , C. curvispinumha s established such dominance of a stone surface that most of the other macro- invertebrate species are left with little space to settle. However, the scenario is very different at Ravenswaay,Bergambach tan dLekkerker k(R .Lek )an di nRive rMeus e(Ooi ,Broekhuizen , Bergen, Cuijk, Grave, Gewande, Genderen and Keizerveer), where D. polymorpha has outnumbered C. curvispinum (Fig. 42). As mentioned earlier, C. curvispinum was not able to establish due to unfavourableloca lhydrographica lcondition s (e.g.strea mvelocities ,ioni ccontent ,salinit yetc. ) inth e River Meuse. Selected physico-chemical parameters measured between 1986 and 1993a t Lobith, 78 1.25 0.00 4 5 6 8 9 10 11 12 13 14 15 16 17 Sampling locations 100 28 E E 2 21 Z E 3 .C «" -^9 - Q. "O •o o E S - 14 3 O O Q. o b Hu «*- o - 7 'to W C c 1 3 5 7 9 11 13 15 17 Sampling locations C. curvispinum D. polymorpha Fig.42. Relationship between population densities of Corophium curvispinum and Dreissena polymorphaalon gwit haverag estrea mvelocitie sa tdifferen t locationsi nth eRive rRhin ean d River Meuse (refer Fig. 1fo r sampling locations). 79 eau XI 1 matte r 3 eu (m g r ) Suspende d M XI {9 O 'C S 3 -O fO S 1-H in O VJ ^r 0\ rH 1« N , , o •M vé in u> vî in •/> 1 carbo n •8 00 (m g r ) 43 »H *, Tota l organi c CU os IM -o CM 0> 1 oxyge n Ü (m g r ) •M (°C ) es es 9 t O. 3 Temperatur e o es o. XI S eu E 9 ^ © © i-l © 9 >> i-I © © 't r« M © tN © t- « es r-( »S rH S >> 2 ( N m' ) V XI densit y •s S ^ es es Populatio n 3 « S «s " 61) •M •- 3 es Ä —" 43 'J « • ea VOt-ejOC7\©»H D. polymorpha requires a solid substrate for their settlement with its byssus threads. However, the filter-feeding C.curvispinum build s tubes made of mud particles, sand and spinning secretions. In the River Rhine, nearly all stones have been covered by a 1t o 4 cm thick layer of muddy materials laid down by C.curvispinum. Therefore,a decline of D.polymorpha ha s occurred in the River Rhine since 1989 (Bij de Vaate et al. 1992; van der Velde et al. 1994), after the explosive invasion of C.curvispinum (Tabl e 15). Similar observations were made by Scholl (1993) and Jantz (1996) from the German part of River Rhine between Emmerich (Rh-km 861) and Basel (Rh-km 168).Schol l(1993 )observe dtha tth ethic klayer so fmudd ytube so f C.curvispinum reduce d the D. polymorpha abundance and shifted the size distribution towards smaller shell length in comparison to contiguous mussel stocks without settlement of C. curvispinum.Jant z (1996) also reported that the populations of D.polymorpha covere d with the muddy layers of C. curvispinum showed a higher proportion of dead individuals and a shifted size distribution towards smaller specimens.Th emortalit y of D. polymorphacompletel ycovere dwit h muddytube s of C. curvispinum possibly resultedfro m the isolation of mussels from theflowin gwater . Several incidents of reduced shellgrowt h ratesan d numerous interruption lineso nth eshel lsurface s of D.polymorpha i n relation to presence of C. curvispinum have been reported (Jantz 1996). Eng (1977) has reported the influence of Corophiumspinicorne Stimpso n on the settlement rates of C. fluminea in the Delta- Mendotacanal ,Californi awher eth eencrustation s of C.spinicorne, anothertub e buildingamphipo d provides a habitat for the larval and early juvenile clams. The impact of C. curvispinumo n other macroinvertebrate species is under study. Despiteth eimportanc eo fth epopulatio nexplosio no fC. curvispinumi nth eRive rRhin e(va n den Brink etal. 1993a) ,ver y little isknow nabou tth efundamenta l featureso ffiltratio n ratean dtub e building activity of this species. The success of any invasive species is mostly contributed by its exceptional plasticity inth efilte rfeedin g responsean dth ecapacit y foralternativ e modeso ffeeding . The present large populations of C. curvispinum in the Rhine may have an capacity to remove seston andclarif y water. Sucha n increase inwate rclarit ywil l result ina nincreas e inth e magnitude ofth eeuphoti c zonean dthu stransfe r nutrientan denerg yt oth ebenthos ,thereb y increasing benthic production.I norde rt o predictfutur edevelopments , muchmor eknowledg e must begathere do nth e ecophysiology (e.g.feedin gbehaviour ,selectio no fparticl esize ,shape ,weight ,composition ,surfac e texture, adhesive properties etc) and impact (e.g. filtration impact, trophic structure, energy flow, species composition etc.) of C.curvispinum i n the River Rhine. 81 5. Conclusions and Recommendations 5.1. Conclusions 1. Thelife-cycl eo f C.curvispinum i nth eLowe r Rhinei sbase do nthre egeneration spe ryear .Th e breeding season of C.curvispinum wa s recordedbetwee n Mayan dOctobe r andwa s strongly correlated with water temperature. Reproduction generally began in March and was well established by May. The overwintering generation died during June/July, but reproduction continued until October as a result of breeding of summer generation individuals. 2. The sex ratio of C. curvispinumindicate dtha tfemale s (60t o 80%) were more abundant than males on the sampled stones. The lowest percentage of males (20%) was observed during May/June. 3 The mean brood size (number of eggs) of C. curvispinumi n the Lower Rhine, is one of the highest ever recorded and showed a positive correlation with chlorophyll-a. 4. The percentage egg lossfro mth ebroo d poucho f C.curvispinum i shig hwhe ncompare d with other Corophium species. 5. A linear relationship existsbetwee nbod y lengthan dbroo dsiz eo f C.curvispinum i nth e Lower Rhine as in many other crustaceans 6. The growth rates and production were highest during May to September when the highest temperature and chlorophyll-a values were recorded inth e Lower Rhine 7. Theamoun to f mudmateria lo nth estone so f groinsan dexperimenta ltile si nLowe r Rhinewa s foundt o be higher during summer period. 8. A highlysignifican tcorrelatio nwa sestablishe dbetwee npopulatio ndensitie so f C. curvispinum anddr yweigh t (DW)an dash-fre edr yweigh t (AFDW) of muddy materialpresen t onth estone s andtiles . 9. The seasonal pattern of changes in the AFDW of mud material per individual at Lobith and Nijmegen are largely similar. 10. Incomparison , the AFDW of the mud material fixed per individual onth e stones wasfoun dt o behighe ra tLobit htha nNijmege nan dthi s isstrongl y relatedwit hth ehighe r meanbod y length of C.curvispinum a t Lobith. 11. Mean values of the rate of mud fixation per individual calculated from the cumulative density of C.curvispinum an dcumulativel yamount so fmu dfixe do nth estone sa tLobit han d Nijmegen indicates the constant level of about 2.5 u.gindividual" 1 day"1 (n = 1,700,000) after 400 days. Nearly the same value wasfoun d onth e tiles atWeur t (2.3 u,gindividual" 1day" 1; n= 500,000) after 350 days. 82 12. The percentage of organic matter of the mud material did not show any significant variation between Lobith and Nijmegen. 13. Thepercentag e of organicmatte ro fth emu dmateria lexcludin gmacroinvertebrate so nth etile s at Weurt showed higher values than the values obtained from the stones including macroinvertebrates at Lobith and Nijmegen. This may be possibly due to the presence of macroinvertebrates such as shells of molluscs and skeletons on the stones at Lobith and Nijmegen. 14. The population density of C.curvispinum a tdifferen t samplinglocation swa s significantly lower inRive rMeus ewhe ncompare dt oRive r Rhine.Thi si spossibl ydu et oth elowe rchlorid e values (CI) and higher pollution (e.g. Cd) in the River Meuse (For the year 1993; CI'1:mea n = 53, range = 16-136; Cd:mea n =0. 4 u,gI" 1,rang e =0. 1- 4.0 , RIWA, 1993)tha n River Rhine (For theyea r 1993; CI'1: mean = 151, range =4 7- 236 ;Cd :mea n =<0. 1u, gI" 1,rang e =<0. 1 - 0.1, RIWA, 1993). Inaddition , Rhinewate r also hasa relatively high ionic content (mean= 3.5 mM Na+; Range = 2.7 - 4.4; measured at Lobith during 1993;Tabl e 18) than River Meuse (mean = 0.4 mM Na+; Range =0. 3 -0.6 ; measured at Remilydurin g 1993),favourin g the occurrence of C.curvispinum, which needsa minimu m of 0.5 mMNa +(Taylo r &Harris , 1986b). Moreover, stream velocities were also lower in River Meuse (maximum of 0.15 m sec'1)tha n River Rhine (maximum of 1.12m sec"1). For afilte r feeding species like C.curvispinum, food can be most easily collected at sites where flow is strong enough for a continuous supply of food. 15. The population densities of C. curvispinumi n the Rhine branches viz., R. Nederrijn/Lek (r = 0.75, P<0.001), R. Waal (r = 0.87, P<0.001) and R. Ussel (r = 0.93, P<0.001) showed a positivecorrelatio nwit hth eaverag estrea mvelocitie s (whichrange dfro m0.2 5t o 1.15m sec' 1). Theaverag e densities of C.curvispinum inth e Rhinedecrease d ina downstrea m direction.C. curvispinumdensitie s were lower in the R. Nederrijn/Lek branch, which is dammed by weirs, than in the weir-free R. Waal and R. Ussel. C. curvispinum densities showed no clear correlation with average streamvelocitie s (whichrange dfro m0.1 0t o0.1 5m sec" 1)i nth e River Meuse. 16. The population densities of C.curvispinum outnumbe r other macroinvertebratespecie s onth e stones of groins in River Rhine. The impact of C. curvispinumpopulatio n expansion and its tubiculous settlement strategy on the other previously dominant macrofauna has been noted. For example, D. polymorpha requires a solid substrate for their settlement with its byssus threads.However , inth e River Rhine,nearl y allstone s have beencovere d bya 1 to4 c mthic k layer of muddy materials laid down by C.curvispinum. Therefore, a decline of D.polymorpha has occurred in the River Rhine since 1989. The impact of C. curvispinum on other macroinvertebrate species is under study. 17. Selected physico-chemical parameters measured between 1986 and 1993 at Lobith, show a decline of total organic carbon and suspended matter over the years, which might be related to the increase in the population density of C. curvispinum. 83 5.2. Recommendations 1. The present large populations of C.curvispinum i nth e Rhine mayhav e ancapacit yt o remove seston and clarify water. Such an increase in water clarity will result in an increase in the magnitude of the euphotic zonean dthu stransfe r nutrient and energyt o the benthos,thereb y increasing benthic production.However , very littlei s known aboutth efundamenta lfeature so f filtration ratean dtub e buildingactivit yo fthi s species. Inorde rt o predictfutur e developments, much more knowledge must be gathered on the ecophysiology (e.g. feeding behaviour, selection of particle size,shape ,weight ,composition ,surfac etexture ,adhesiv e properties etc) andimpac t (e.g.filtratio n impact,trophi c structure,energ yflow ,specie scompositio n etc.) of C. curvispinumi n the River Rhine. 2. The dominance of C. curvispinum may lead to changes in the food webs in the rivers. Especially their role as food for predatory macroinvertebrates like crayfish and for fishes is poorly studied.Therefore , littleca n be saido f its impact on consumer populations, the effects of prédationupo n populationdensitie s and,ultimately ,th erol eo f C.curvispinum i ncarbo nflu x between primary producers and higher trophic levels in the Rhine ecosystem. References Bayliss D. & Harris R.R. (1988) Chloride ion regulation in the fresh water amphipod Corophium curvispinuman dacclimator yeffect s ofexterna l CI".Journal of Comparative PhysiologyB, 158, 81-90. BayneB.L . (1976) Thebiology of mussel larvae. Marine mussels: their ecology and physiology (eds . B.L. Bayne). pp. 81-120. Cambridge University press, Cambridge. Behning A. (1914) Corophium curvispinum G.O. Sars und seine geographische Verbreitung. ZoologischeJahrbücher. Abteilung für Systematik, Ökologieund Geographie der Tiere, 37,385 - 400. Behning A. (1925) Studien über die Malakostraken des Wolgabassins. Internationale Revue der gesammte Hydrobiologie,13 ,46-77 . Bij de Vaate A. & Greijdanus-Klaas M. (1991) Monitoring macroinvertebrates in the River Rhine. EcologicalRehabilitation of theRiver Rhine, Vol .27-1991 , pp. 1-39. Institute for Inland Water Management and Waste Water Treatment, Lelystad,Th e Netherlands. Bij de VaateA. , Greijdanus-Klaas M. & Smit H. (1992) Densities and biomass of zebra mussels in theDutc hpar to fth eLowe rRhine . Thezebra mussel Dreissena polymorpha: ecology, biological monitoring and first applicationsin the waterquality management(eds . D. Neumann & H.A. Jenner), pp. 67-77. Gustav Fischer Verlag, Stuttgart. 84 Bij de Vaate A. & Klink A.G.(1995 ) Dikerogammarus villosus Sowinsky (Crustacea: Gammaridae) a new immigrant in the Dutch part of the Lower Rhine. Lauterbomia,20 ,51-54 . BussL.W . (1981) Grou pliving ,competitio nan dth eevolutio no fcooperatio n ina sessil e invertebrate. Science,213, 1012-1014. CrawfordG.I .(1935 ) Corophiumcurvispinum (Sars )va rdevium (Wundsch ) inEngland .Nature, 136, 385. Crisp D.J. (1971) Energy flow measurements. Methodsfor thestudy of marine benthos, (eds . N.A. Holmes & A.D. Mclntyre). pp. 197-279. IBP handbook 16, Blackwell,Oxford . Dayton P.K. (1971) Competition, disturbance and community organization: the provision and subsequentutilizatio no fspac ei na rock yintertida lcommunity . EcologicalMonograph, 41,351- 389. Den Hartog C, Van den Brink F.W.B. & Van der Velde G. (1989) Brackish water invaders in the River Rhine. A bioindication for increased salinity level over the years.Naturwissenschaften, 76,80-81. Den HartogC , Vande n Brink F.W.B. &Va nde r Velde G.(1992 ) Whywa sth e invasion ofth e River Rhine by Corophium curvispinuman d Corbiculaspecie s so successful?. Journal of Natural History, 26,1121-1129. Dick J.T.A., Montgomery I. & Elwood R.W. (1993) Replacement of the indigenous amphipod Gammarus duebeni-celticus by introduced G. pulex: differential cannibalism and mutual prédation. Journal ofAnimal Ecology, 62,79-88 . Drake J.A., Mooney H.A., Di Castri F., Groves R.H., Kruger F.J., Rejmânek M. & Williamson M. (eds.) (1989) Biological invasions. A global perspective. J. Wiley & Sons, Chichester. D'Udekemd'Aco z C. &Stroo t P.(1988 ) Notesu r l'expansion de Corophium curvispinumSars ,189 5 en Meuse (Crustacea, Amphipoda: Corophiidae). Annales du Société Royal de Zoologie de Belgique, 118,171-175. Eng L.L. (1977) Population dynamics of the Asiatic clam, Corbicula fluminea (Müller), in the concrete-lined Delta-Mendota canal of central California. Proceedings of first Corbicula symposium,Texas , USA, pp 39-68. Entz B.(1947 ) Beitrag zurKenntni sde r Morphologie undBiologi ede s Corophium curvispinumG.O . Sars forma deviumWundsc h in Ungarn. Archiv fürHydrobiologie, 42,423-469 . Fahnenstiel Gl., LangTl. , BridgemanG.A. , McCormick M.J. & NalepaT.F . (1995) Phytoplankton productivity in Saginaw bay, Lake Huron: effects of zebra mussel (Dreissenapolymorpha) colonization. Journal of GreatLakes Research,21 , 465-475. Fish J.D. & Mills A. (1979) The reproductive biology of Corophium volutator and C.arenarium (Crustacea: Amphipoda). Journalof theMarine Biological Association of the UnitedKingdom, 59, 355-368. 85 GledhillT. , Sutcliffe D.W. &William sW.D .(1976 ) A revised keyt oth e British species of Crustacea: Malacostracaoccurrin gi nfres hwater .Freshwater Biological Association. ScientificPublication, 32,1-71. Harris R.R. (1991) Amphipod also invades Britain. Nature,354, 194. HarrisR.R . &Baylis sD .(1990 )Osmoregulatio ni nCorophium curvispinum (Crustacea :Amphipoda) , a recent colonizer of freshwater. III. Evidence for adaptive changes in sodium regulation. Journal of Comparative PhysiologyB, 160 , 85-92. Hengeveld R. (1989) Dynamics of biological invasions. Chapman & Hall, London. loffe T.U. (1954) Donnye kormovye Tsimyanskogo vodokhranilishcha v pervyi god ego sushchestvovaniya. Izv VNIORKH, 34, 78-114. Jantz B. (1996) Wachstum, Reproduktion, Populationsentwicklung und Beeinträchtigung der Zebramuschel (Dreissenapolymorpha) in einem Großgewässer, dem Rhein, pp. 1-141. Ph.D thesis, Universität zu Köln, Germany. JazdzewskiK .(1980 )Rang eextension so fsom egammaridea nspecie si nEuropea nwater scause d by human activity. CrustaceanaSupplement, 6,84-107. Kelleher B., Bergers P.J.M.,Va nde n Brink F.W.B., Giller P.S., Vande rVeld e G. & Bijd eVaat e A. (1997) Effects of exotic amphipod invasions on fish diet in the Lower Rhine. Archiv für Hydrobiologie(Submitted) . Kerswill L.J. (1949) Effect of water circulations on the growth of quahaugs and oysters. Journal of FisheryResearch Boardof Canada, 7,545-551. Marguillier S.,Dehair s F.,Va nde rVeld eG. ,Kellehe rB .& Rajagopa lS .(1997 )Trophi c relationships based on Corophium curvispinumi n the Rhine traced by stable isotopes: first results. New conceptsfor sustainable managementof riverbasins (eds . P.H.Nienhuis , R.S.E.W.Leuve n & A.M.J. Ragas). SPB Academic Publishing BV, Amsterdam (in press). Möller P. & Rosenberg R. (1982) Production and abundance of the amphipod Corophium volutator on the west coast of Sweden. NetherlandsJournal of Sea Research,16, 127-140. Moon H.P.(1970 ) Corophium curvispinum (Amphipoda) recorded again inth e British Isles.Nature, 226, 976. Moore P.G. (1981) The life histories of the amphipods Lembos websteri Bate and Corophium bonnelliMiln e Edwards inKel pholdfasts .Journal of Experimental Marine Biology and Ecology, 49,1-50. MuskóI.B .(1989 )Amphipod a (Crustacea) inth e littoralzon eo f LakeBalato n(Hungary) .Qualitativ e and quantitative studies. InternationaleRevue dergesammte Hydrobiologie,74,195-205 . Muskó I.B. (1990) Qualitative and quantitative relationships of Amphipoda (Crustacea) living on macrophytes in Lake Balaton (Hungary). Hydrobiologia,191 , 269-274. 86 Muskó I.B. (1992) Life history of Corophium curvispinum G.0. Sars (Crustacea, Amphipoda) living on macrophytes in Lake Balaton. Hydrobiologia, 243/244,197-202. Muskó I.B. (1994) Occurrence of Amphipoda in Hungary since 1853. Crustaceana, 66,144-152. Muskó I.B.,Tót hL.G .& Szâb oE .(1994 )Respiratio nan drespirator yelectro ntranspor tsyste m(ETS ) activity of two Amphipods: Corophium curvispinum G.O. Sars and Gammarusfossarum Koch. PolskieArchiwum Hydrobiologii, 42,547-558 . Nichols K.H. & Hopkins G.J. (1993) Recent changes in Lake Erie (north shore) phytoplankton: cumulative effects of phosphorous loading reductions and the zebra mussel introduction. Journal of GreatLakes Research,19 , 637-647. Paffen B.G.P., Van den Brink F.W.B., Van der Velde G. & Bij de Vaate A. (1994) The population explosion of the amphipod Corophium curvispinumi nth e Dutch Lower Rhine. WaferScience and Technology, 29,53-55 . Peer D.L., Linkletter L.E. & Hicklin P.W. (1986) Life history and reproductive biology ofCorophium volutator (Crustacea: Amphipoda) and the influence of shorebird prédation on population structure inChignect o Bay, Bayo f Fundy, Canada.Netherlands Journal of Sea Research, 20, 359-373. Perkins E.J.(1974 )Th e biologyo f estuaries andcoasta lwaters ,p p353-359 .Academi c press,Ne w York. PinksterS. ,Scheepmake r M.,Platvoe t D.& Broodbakke r N.(1992 ) Drasticchange s inth eamphipo d fauna (Crustacea) of Dutch inland waters during the last 25 years. Bijdr.Dierk., 61,193-204. Platvoet D. & Pinkster S. (1995) Changes inth e Amphipod fauna (Crustacea) of the Rhine, Meuse and Scheldt estuary due to the delta plan coastal engineering works. NetherlandsJournal of Aquatic Ecology, 29, 5-30. Pygott J.R. & Douglas S. (1989) Current distribution of Corophiumcurvispinum Sar s var devium Wundsch (Crustacea:Amphipoda ) in Britain with notes on its ecology inth e Shropshire Union canal. Naturalist, 114,15-17. RIWA (1993) Samenwerkende Rijn- en Maas- waterleidingbedrijven. Secretariaat RIWA, Postbus 8169,1005 AD Amsterdam. RomanovaN.N .(1975 )Quantitativ edistributio nan decolog yo fcorophiid so fth eCaspia nSe aregion . Byulleten' Moskovskogo ObshchestvaIspytatetei Prirody. Otdel Biologicheskii, 80,51-63 . Scholl, F. 1993. Der Schlickkrebs (Corophiumcurvispinum) un d die Augustfliege (Ephoron virgo): zwei Arten mit rezenter Massenentwicklung im Rhein. In: Ministerium für Umwelt Rheinland- Pfalz (Hg.):Di e Biozönose des Rheins imWandel-Iach s2000 ? Beiträge zur Fachtagung März 1992 in Mainz, S. pp.89-94 . Scholl F. (1990) Zur Bestandssituation von Corophium curvispinum Sars im Rheingebiet. Lauterbornia,5 , 67-70. 87 Sebestyén 0. (1934) Appearance and rapidincreas eo f Dreissenapolymorpha Pallan d Corophium curvispinum G 0 Sars forma devium Wundsch in Lake Balaton. Arbeiten des Ungarisch BiologischeForschungs Institut, 7,190-202. Sheader M.(1978 ) Distribution and reproductive biology of Corophium insidiosum(Amphipoda ) on the north-east coast of England. Journal of the Marine BiologicalAssociation of theUnited Kingdom,58 ,585-596 . Sokal R.R. & Rohlf F.J. (1981) Biometry. W.H. Freeman & Co., New York. Stewart T.W. & Haynes J.M.(1994 ) Benthic macroinvertebratecommunitie s of southwestern Lake Ontario following invasion of Dreissena. Journalof GreatLakes Research, 20,479-493 . Taylor P.M. (1985) The pattern of gill perfusion in two species of Corophium (Crustacea: Amphipoda) and itsrelatio nt oenvironmenta l salinity.Journal of Zoology (London), 205,29-38. Taylor P.M. & Harris R.R. (1986a) Osmoregulation in Corophium curvispinum (Crustacea: Amphipoda),a recentcolonize r offreshwater . I.Sodiu m ionregulation .Journal of Comparative PhysiologyB, 156 , 323-329. Taylor P.M. & Harris R.R. (1986b) Osmoregulation in Corophium curvispinum (Crustacea: Amphipoda), a recent colonizer of freshwater. II.Wate r balance andth efunctiona l anatomy of the antennary organ. Journal of Comparative PhysiologyB, 156 , 331-337. Van den Brink F.W.B. (1995) Biodiversity as affected by invader species. Econieuws,26, 17-18. Van den Brink F.W.B., Paffen B.G.P., Oosterbroek F.M.J. & Vande rVeld e G. (1993b) Immigration of Echinogammarus ischnus(Stebbing , 1906) (Crustacea: Amphipoda) into The Netherlands via the Lower Rhine. BulletinZoölogisch MuseumUniversiteit van Amsterdam, 13,167-170. Van den Brink F.W.B., Van der Velde G. & Bij de Vaate A. (1989) A note on the immigration of Corophiumcurvispinum Sars , 1895(Crustacea :Amphipoda ) intoth eNetherland s viath e River Rhine. BulletinZoölogisch MuseumUniversiteit van Amsterdam, 11 , 211-213. Van denBrin k F.W.B., Vande rVeld e G.& Bi jd eVaat eA . (1991)Amphipo dinvasio no nth eRhine . Nature,352 , 576. Van den Brink F.W.B., Van der Velde G. & Bij de Vaate A. (1993a) Ecological aspects, explosive rangeextensio n andimpac to fa mas sinvader , Corophiumcurvispinum Sars , 1895(Crustacea : Amphipoda), in the Lower Rhine (The Netherlands). Oecologia, 93,224-232 . Van den Brink F.W.B., Van der Velde G.& Cazemie r W.G. (1990) Thefaunisti c composition of the freshwater section of the river Rhine in The Netherlands: present state and changes since 1900. DieBiologie des Rheins(eds . G. Friedrich & R. Kinzelbach). Limnologieaktuell, 1,191 - 216. Van der Velde G, Paffen B.G.P., Van den Brink F.W.B., Bij de Vaate A. & Jenner H.A. (1994) Decline of Zebra mussel populations in the Rhine. Competition between two mass invaders {Dreissenapolymorpha and Corophiumcurvispinum). Naturwissenschaften, 81, 32-34. 88 Van der Velde G., Rajagopal S., Van den Brink F.W.B, Kelleher B, Paffen B.G.P, Kempers A.J. & Bijd e Vaate A. (1997) Ecological impact of anexoti c amphipod invasion in the River Rhine. New concepts for sustainable management of river basins (eds. P.H. Nienhuis, R.S.E.W. Leuven & A.M.J. Ragas). SPB Academic Publishing BV, Amsterdam (in press). Van derVeld e G.,Va n denBrin k F.W.B., Vande rGaa gM .& Bergers P.J.M. (1990) Changes inth e numbers of mobile macroinvertebratesan dfis h inth e RiverWaa l in 1987,studie d by sampling inth e cooling-water intakes of a power plant:firs t results of a Rhine biomonitoring project. Die Biologie des Rheins (eds. G. Friedrich & R. Kinzelbach). Limnologieaktuell, 1, 325-343. Van der Weijden C.H. & Middelburg J.J. (1989) Hydrogeochemistry of the River Rhine: long term and seasonal variability, elemental budgets, base levels and pollution. WaterResearch, 23 , 1247-1266. Van Urk G. & Bij de Vaate A. (1990) Ecological studies inth e lower Rhine inTh e Netherlands. Die Biologie des Rheins(eds . G. Friedrich & R. Kinzelbach). Limnologieaktuell, 1,131-145. Van Urk G. & Smit H. (1989) The Lower Rhine. Geomorphological changes. Historicalchange of large alluvial rivers, Western Europe (eds. G.E. Petts). pp. 167-182. J. Wiley & Sons, Chichester. Walne P.R. (1972)Th e influenceo fcurren tspeed ,bod ysiz ean dwate rtemperatur e onth efiltratio n rate of five species of bivalves. Journal of the Marine Biological Association of the United Kingdom,52 , 345-374. Waterstraat A. & Kohn J. (1989) Ein Beitrag zur Fauna des Kummerower Sees, Erstnachweis des Amphipoden Echinogammarus ischnus Stebbing 1899 in der DDR. Archiv Freunde der Naturgeschichtein Mecklenburg,29 , 93-105. Winberg G.G., Patalas K, Wright J.C., llkowska A.H., Cooper W.E. & Mann K.H. (1971) Methods forcalculatin g productivity. A manualon methods for the assessment of secondary productivity in fresh waters (eds. W.T. Edmondson & G.G. Winberg). pp. 296-317. IBP handbook 17, Blackwell, Oxford. Wouters K. (1985) Corophiumcurvispinum Sars , 1895 (Amphipoda) inth e River Meuse, Belgium. Crustaceana,48 , 218-220. Wundsch H.H.(1915 )Weiter e Beiträgezu r Fragede rSüsswasserfor mvo n Corophiumcurvispinum. Sitzungsberichte der BerlinischeGesellschaft Naturforschender Freunde, 3 , 56-85. Wundsch H.H. (1919) Weitere Funde der Süsswasserform von Corophium curvispinum in der Baltischen Tiefebene. Archiv für Hydrobiologie Planktonkunde, 12,693-697 . Zar J.H. (1984) Biostatistical analysis. Prentice-Hall Inc., Englewood Cliffs, New Jersey. 89 e o & >. il .3 es is CU «B OD 1 =L ON TH co O s ON ï« ei II 6 e?\ S OJD II m à «s o «s & + 00 § 00 00 'E »o CA m O+N f) .o m co • 1-So Oco '£ II 1 II ö II lH + DC ? a 00 .S X có Z TT 'S C3\ cu co 00 + »H 9\ *• O es + •«* BT CO co Ke fl c4 + II o z o X ë -.« ^ CO «o & S 5? e* S 'S f. t- T3 rH es oo e> B II M O es V) II T3 f) Z 3 < C co es "E es « T3 'S H a < z Z S l> Ui II O e 00 II Q. ^< jl *> 3 es CU II c B 'S O» s < c o u cu 9 * * 2-9 'S a * -as CS OU CU cu n es oo o .g u Ä o cu e +•» u a o ,o es (3 X! o 3 cl e a c 2f -o 9 #o "35 fl CU es O 2. SS e ä e eu OH a es CU -o es u u < tri o cu cu Q o LM c Ö 1 S £ CL o •»J < oo co 5 O) N o: c «j O TJ I -- (U —' 0) T TR o: ra -— c T3 10 O g Si* I O) ra a> ra ra S mgra •o w m ± c c .y 0) E E E?5> c ra - ra g 5 >. .o > n CM co m O 1- CM CO oo t-CM co CM oo en o CM CM CM CMCM co co co co co co co M- Lil Z (1) V) s> /^ ^ 'C . V > tu (5 eu eu •S 5 tv tu tu t eu ta tu eu •o 3 -o m "> ,- II ec •S oi en ra "- tu -a <7> ffi*.-ra 50: ra2-|= U£ to > g w _= o CU^^Q; -P °c .EÄ !S a> c So)3N» c: TlTÏ S'5 tonc;^ o •a §173 JC N eu O . ra 8 -S *• to raO> S ti ° » a 3 eu c P > 3 ra co i_ •Ölü tl is eu -= \y eu ^ là > co . _ DC 'a. ^?. Q: -c-e co ._ eu irira£ES a> eu x: > c^i(3> c au s.y oc o >_ to 5 " o ^ eu S tu > ç 2 O « N S .M S-:(0 tn (0 1- eu w ^-^= Ü o "^ tu »• c ç co -2 ^ =* tu > a> 8 ta g^ >>»> 5 ce ai eu rat"^ >g §s ^ S eu o J5 '5 ra c o o.a TJlCD: L O .c* ta"-CQ- eQu eora-g^ ra > eu c •= "°^ c^ eUf's E^? g-oû *;u eu 5 0)0 N c #A »- Q! r oCC eu 1 eu c ra-D eu I^^SoStc •S o,^ it^ r o an.S m o £> fûllSra >tnci)'> (3) co co 10 co S eoco co (O S co co >> >. s*> >. •>. O -o eBs *4 S ja o 3 .SP .£f 0H e -*•* 'S 'S .S1 II » •*-* » •«* » •M •a X •I M & f f a 1 S E Ë H IsM Ë u (S _ 01 V 01 o> T3 es B CA o> "O B S B 01 0) B 'S «Oi S O O es e OU o 5 «2 e•tani ai *H ^r FH •8 33 3 +1 -H -H +1 O •H -H fl -H •« r« ci rt S o Cl 00 u oo m »ri m m Ifl M O H i^ r- ci i-c S s»So rt ci •* Cl rt o VA 00 Ul (S »s ri ci ri ri *-H *H s H a -s CM a O o -l o m « N N M r» *