The north-west of the Iberian Peninsula is crucial for conservation ofMargaritifera margaritifera (L.) in Europe

S. LOISa,*, P. ONDINAa, A. OUTEIROa, R. AMAROb and E. SAN MIGUELb

aDepartment of Zoology, University of Santiago de Compostela, Faculty of Veterinary, Lugo, SpainbDepartment of Genetics, University of Santiago de Compostela, Faculty of Veterinary, Lugo, Spain

ABSTRACT

1. Extensive assessments of a species distribution, its variation in density and demographic status acrossenvironments are crucial to develop successful conservation efforts.

2. The freshwater pearl mussel (Margaritifera margaritifera) is considered one of the most endangered freshwaterbivalves with many recovery efforts continuing throughout its geographic range. The distribution and conservationstatus of M. margaritifera are not well documented in the Iberian Peninsula. Galicia (NW Iberian Peninsula)represents nearly 70% of the historical distribution of the species in the southern limit of its European range.

3. An extensive field survey was conducted in two phases at 2436 locations. The presence and density ofM. margaritifera was determined at 555 sampling points spread across 54 rivers that belong to 23 drainagebasins in Galicia. The present work has more than doubled the number of rivers known to provide habitat forthis endangered species in the Iberian Peninsula.

4. In Galicia the species is heterogeneously distributed with a highly variable density of individuals within andbetween rivers. For example, within the River Camba density ranged from 0.02 to 47.8 ind m-2. The maximumdensity detected in a sample was 332 ind m-2. Twelve rivers in Galicia are thought to have more than 5000individuals, and small individuals were found in 11 rivers.

5. High rates of decline and extinction of M. margaritifera populations are known in some areas and the mainthreat to unionoid bivalves is a lack of natural recruitment. Thus, Galician populations are important forproviding new opportunities for conservation of the species in Europe because it is vitally important to find thereasons for recruitment failure.Copyright # 2013 John Wiley & Sons, Ltd.

Received 11 October 2012; Revised 3 February 2013; Accepted 26 February 2013

KEY WORDS: river; catchment; endangered species; survey; distribution; invertebrates

INTRODUCTION

As biodiversity loss continues unabated, increasinglymore effort ismade for conservation aimed at preventing

species extinction. To maximize conservation successa sound framework is needed including extensivebaseline information (Sutherland et al., 2004).Efforts to increase quantitative data about a species

*Correspondence to: S. Lois, Department of Zoology,University of Santiago deCompostela, Faculty of Veterinary, Lugo, Spain. Email: sabela.lois@usc.es

Copyright # 2013 John Wiley & Sons, Ltd.

AQUATIC CONSERVATION: MARINE AND FRESHWATER ECOSYSTEMS

Aquatic Conserv: Mar. Freshw. Ecosyst. 24: 35–47 (2014)

Published online 16 April 2013 in Wiley Online Library(wileyonlinelibrary.com). DOI: 10.1002/aqc.2352

in its present habitats are crucial to obtain a completepicture of the species’ ecology across its range (Fortinet al., 2004). Learning more about a speciesdistribution and its ecology will help to identifyconservation units and thus attain successfulintegrative conservation efforts. Moreover, extensivefield evaluations will help identify stressors andimpairments to habitats and they are an importantprerequisite to implementing recovery strategies suchas captive breeding (Snyder et al., 1996).

Compared with other ecosystems, fresh waters andtheir species are more greatly endangered around theworld (Lydeard et al., 2004; Dudgeon et al., 2006)and one of the most rapidly declining groups are thefreshwater mussels (Vaughn and Taylor, 1999).Margaritifera margaritifera (Linnaeus, 1758) is anendangered freshwater bivalve mollusc (Unionoida:Margaritiferidae) which occurs in oligotrophic riversof the Holarctic region. It is a long-lived organismwith a complex life-cycle sensitive to changes inwater quality (Bauer, 1991; Geist, 2010). It has alarval stage parasitic on the gills of salmonid fish,and later settles on the river bed and formsaggregates of individuals (Outeiro et al., 2008). It isconsidered a key species for the conservation ofaquatic ecosystems (Geist, 2010) and it is classified asendangered by the International Union forConservation of Nature (IUCN, 2012). The specieshas been called one of the most threatenedfreshwater bivalves in the world (Machordom et al.,2003). It is included in Annex II of the EuropeanUnion Directive on Conservation of NaturalHabitats and of Wild Fauna and Flora (Directive92/43/EEC; Council of the European Communities,1992). Within Europe populations of M.margaritifera have declined by 90% during the lastcentury (Araujo and Ramos, 2001), leading to thedevelopment of captive breeding techniques forfreshwater mussels in Europe that are focused on M.margaritifera (Gum et al., 2011). Despite this earlyand extensive investment in captive breeding,knowledge about the species’ ecological requirementsis still fragmented and limited (Bauer, 2000; Geist,2010). Consequently, more research is needed toobtain information about populations across thespecies range in order to improve conservation efforts.

Although a few new populations ofM.margaritiferahave been found recently (Ostrovsky and Popov, 2011),

the most up-to-date synthesis of its distributionand status in Europe is that of Geist (2010). Thenorth-west quadrant of the Iberian Peninsulaappears to be the southern limit of the Europeandistribution of the species. Galicia is the westernmostregion of the European continent and encompasses29 574 km2 (Figure 2). It has 1660 km of coastlinebathed by the Atlantic Ocean and the CantabrianSea and it borders Portugal to the south. In Galiciathere are limited historical data on the historicaldistribution of M. margaritifera (Macho, 1878;Azpeitia, 1933) and more recent studies have focusedon only a few populations (Álvarez-Claudio et al.,2000; Ziuganov et al., 2000; Araujo and Ramos,2001; Grande et al., 2001; Machordom et al., 2003;San Miguel et al., 2004; Bouza et al., 2007; Outeiroet al., 2008). Thus, it is important to conductextensive surveys to define the contemporarydistribution of the species in the southern part ofits range. No previous study has aimed atdetermining the complete distribution of thisspecies in the extensive network of rivers in Galicia.

The objectives of the present paper are to identifythe present distribution of the species, estimatepopulation densities and search for small individuals,which may be indicative of recruitment. Presentationof these data is intended to provide the necessarybaseline for assessing the conservation status ofM. margaritifera in the Iberian Peninsula and to enabledevelopment of effective conservation efforts suchas captive breeding, reintroductions and relocations.

MATERIALS AND METHODS

The sampling methods commonly used forfreshwater mussels are both qualitative andquantitative (Bauer, 1986; Beasley and Roberts,1996; Álvarez-Claudio et al., 2000; Cosgrove et al.,2000; Hastie et al., 2000b; Young et. al, 2001b; Reis,2003; Morales et al., 2004; Rudzīte, 2005; Outeiroet al., 2008; Österling et al., 2010), using differentdesigns depending on the objectives. Galicia has anextensive river network (Rodríguez, 2001); rivers aretypically short with small drainage areas and themajority of them are oligotrophic with low calciumcontent, contain various salmonid species, and thuscould provide suitable habitat for the species(Ziuganov et al., 1994). An extensive sampling plan is

S. LOIS ET AL.36

Copyright # 2013 John Wiley & Sons, Ltd. Aquatic Conserv: Mar. Freshw. Ecosyst. 24: 35–47 (2014)

needed to search for the presence of freshwater musselsin as many rivers as possible to establish the currentspecies distribution in Galicia. In addition, other dataabout its abundance and recruitment are required toexamine the condition of populations (Geist, 2010).Thus, a combination of qualitative and quantitativemethods is the best option (Vaughn et al., 1997).

Sampling methods

Fieldwork was carried out in the north-west regionof the Iberian Peninsula within the administrativeboundaries of the Autonomous Region of Galicia.The sampling method was based on Villella andSmith (2005) for efficient estimation of freshwatermussel densities in large areas. It is a two-phasedoubly-stratified sampling method that makes itpossible to establish a relationship betweenquantitative and qualitative estimations (Villellaand Smith, 2005). This strategy is also aimed atincreasing the probability of locating recruitingareas, as juveniles and adults can share the samehabitat (Hastie et al., 2000a, 2010).

Phase I: qualitative exploratory sampling

Phase I field sampling served to search for M.margaritifera in rivers and estimate its abundance.For this phase, rivers that could potentiallycontain M. margaritifera were chosen from amongall Galician rivers with the following criteria: (1)the existence of bibliographic records of thespecies and/or (2) the presence of intermediatehosts of M. margaritifera (Salmo salar, Salmotrutta and/or Salmo trutta fario) while maximizingthe number of rivers sampled in Galicia. In total,148 rivers and tributaries were selected. For fieldsurveys each river was divided into sections 5 km

long using ArcMap 9.3 (ESRI, 2009) (Figure 1). Ineach 5 km section two sampling stretches of variablelength (between 50 and 200 m) were searched formussels using aquascopes. In addition, the width ofthe river at each sampling point was measured.

In every stretch whereM.margaritiferawas found,individuals were counted, but only those present in alimited area of the river bed. This area was called the‘bank corridor’, defined as a corridor of the river bed2 m wide and 50–200 m long adjacent to the riverbanks where mussels were expected to bemost abundant. Previous work showed a pattern ofriver-bed distribution with half-buried but spatiallysegregated groups of mussels close to the banks(Outeiro et al., 2008). From these counts, thedensity of individuals in each bank corridor wascalculated and referred to here as ‘bank density’.Bank density of M. margaritifera was subsequentlyadjusted by procedures described in Phase II.

Once all selected rivers had been searched, thesampling effort of Phase I was intensified in thoserivers where the species was found. In each of theserivers, the 5 km section with the highest bankdensity was selected. This section was divided into1 km sub-sections in which two or five samplingpoints were surveyed (Figure 1). By sampling theareas with the highest abundances more thoroughly,the possibilities of finding individuals during theseadditional assessments increased, which was necessaryto carry out Phase II properly (Smith, 2006).

Phase II: quantitative sampling

Phase II focused on estimating the density of M.margaritifera in limited transects and on locatingsmall-sized individuals that might be indicative ofrecruitment. Quantitative sampling was carried outat 25 sampling points examined in Phase I. Samplingpoints were chosen using two criteria: (1) to representareas of high and low density (Villella and Smith,2005), and (2) to ensure representation acrossGalicia.

At each sampling point a fixed-length river transect(50 m) was established; the sampling unit was 0.25 m2

delimited by a metal frame. The number of sampleunits for a transect ranged from 20 to 60 dependingon the width of the river. Two methods of samplecollection were used, depending on the transectcharacteristics. In three cases, they were narrow river

Figure 1. In Phase I rivers were divided into 5 km transects for surveys(upper). The area with highest density was further divided into 1 kmsections and additional surveys were conducted. White dots representsampling points where M. margaritifera was not detected and black

dots represent presence.

A CORE AREA FOR CONSERVATION OF MARGARITIFERA MARGARITIFERA 37

Copyright # 2013 John Wiley & Sons, Ltd. Aquatic Conserv: Mar. Freshw. Ecosyst. 24: 35–47 (2014)

channels (mill channels or natural channels in braidedriver transects) where the sampling surface wasrelatively small and the individuals ofM. margaritiferawere spatially distributed as groups of individualsalong the river bed with no clear agregation pattern.In these places, a simple random sampling wasused. For the remaining 22 cases, sampling wascarried out in river transects with a larger samplingarea where the species was mostly distributed alongone or both banks and less abundant in the middle ofthe river. In these cases stratified random samplingwith optimal allocation was used (Krebs, 1999;Strayer and Smith, 2003). Two strata were defined,one of them with high density and another with lowdensity of mussels. Thus, by following the spatialdistribution pattern of the species on the river bed, ahigh density stratum was adjacent to the riverbanks. Once the number of 0.25 m2 sample units ofeach stratum was determined, they were allocated toeach stratum. The co-ordinates of each sample wererandomly chosen and laser rangefinders were usedto locate their position in the field.

All visible individuals within a randomly locatedsample unit were counted. Subsequently, a hole15 cm deep was excavated to detect the presence ofjuveniles, as they usually remain buried (Hastie et al.,2000b). All individuals less than 65 mm long werecounted and measured, as individuals of this size aregenerally considered to be juveniles (Bauer, 1986;Hastie et al., 2000b; Young et al., 2001b).

Using the counts of mussels in each quadratsample, estimates were made of density, averagedensity by stratum, and the weighted density for thetransect following Krebs (1999). The pattern ofspecies distribution among samples was calculatedfor each stratum using Morisita’s index of dispersion(Id) (Morisita, 1959). The critical value Mu and thestandardized Morisita’s index (Id) were obtained withthe software EcoMeth (Krebs, 1999). Further detailson the indices, and the estimation of populationparameters is given by Strayer and Smith (2003) andKrebs (1999).

The relationship between the densities obtained inPhase I (bank density) and Phase II (transect density)was established for the 22 transects sampled inPhase II. This relationship between transect densityand bank density provided a way of calibrating bankdensity to indicate the density of mussels more

accurately. Average density of mussels per river wasused to estimate the total population size in each river.

RESULTS

Phase I

Of the 148 rivers and tributaries surveyed in the firstphase of sampling, M. margaritifera was not foundin 91, its presence was verified in 54 (Figure 2)whereas only shells were found in three. In total,2436 sampling points were examined, representing350 km of river length. Of these sampling points,555 (23%), had occurrences of M. margaritifera.Thirty-three previous records of M. margaritiferadistributed in 26 rivers were confirmed in thepresent study. In addition, it was found in 522 newsampling points, representing a total of 54 riversbelonging to 23 river basins.

The density of individuals observed in the bankcorridor (bank density) was calculated in the 555sampling points at which the species was present.The bank density values ranged from 0.003 ind m-2

to 31.25 ind m-2.

Phase II

The density ofM.margaritifera in the river transects,obtained in the quantitative sampling of Phase II,ranged from 0.1 ind m-2 (River Ulla) to 21.4 ind m-2

(River Eo) (Table 1). However, the highest transectdensity was found in a channel of the Río Camba:47.8 ind m-2. The maximum density observed in asingle sample also came from the channel of thisriver with 332 ind m-2. The rivers Ulla and Arnego,where several quantitative samples were taken indifferent transects, showed highly variable densityvalues, especially in the Arnego (0.3–8.9 ind m-2).The transect with the highest number of individualswas in the River Eo, where 20 135 individuals werefound in an area of 943 m2 (Table 1).

According to Morisita’s standardised index (Id),the individuals of M. margaritifera found in the highdensity stratum and three channels showed a spatialdistribution pattern with mussels occurring in groups(Id> 0, from 2.77 to 0.05) (Table 1). The spatialdistribution was uniform in the high density stratum(Id=�0.04) in just one locality (High Ulla 3). In thelow density stratum, M. margaritifera also occurred

S. LOIS ET AL.38

Copyright # 2013 John Wiley & Sons, Ltd. Aquatic Conserv: Mar. Freshw. Ecosyst. 24: 35–47 (2014)

in spatially segregated groups, although in some casesall the samples had no mussels so the density was zero(Table 1).

Sampling was also conducted to gain informationabout recruitment in Galician rivers. Individuals ofM. margaritifera less than 65 mm long were found in11 sampled transects (Table 2). Except for two rivers(Limia and Eo) they were found in high densitystrata. Only 27% of these mussels <65 mm wereburied, including two individuals less than 30 mm.No relationship was found between M. margaritiferadensity and presence of individuals <65 mm.

Combining Phase I and Phase II results

When comparing bank densities of Phase I withtransect densities of Phase II for the same localities,a linear relationship was observed (Figure 3). This

could be clearly seen in a double log transformationcalibration curve that made it possible to adjustbank density estimates to those obtained in themore intensive survey estimates of density fromPhase II. The regression of transect density on bankdensity (R2= 0.81, P< 0.001) for 22 transects in 12rivers allowed transect density to be estimated forall 555 sampling points in Phase I.

The density estimates for all sampling points wasassigned a value within one of four categories ofdensity (Figure 4). At more than 50% of thesampling points (298) M. margaritifera occurred atlow density (<0.1 ind m-2), whereas densities werehigh or very high at 110 sampling points. Fifteenrivers had areas of high density and five rivers hadareas of very high density. Mean density wascalculated for each river (Table 3). Densityestimates show high variability within and between

Figure 2. Sampling results of Phase I. The grey dots represent sampling points where M. margaritifera was not detected (n= 1881). The red dots(n= 555) represent sampling points where the species was present. Galicia is the shaded portion of the inset map of the Iberian Peninsula.

A CORE AREA FOR CONSERVATION OF MARGARITIFERA MARGARITIFERA 39

Copyright # 2013 John Wiley & Sons, Ltd. Aquatic Conserv: Mar. Freshw. Ecosyst. 24: 35–47 (2014)

Tab

le1.

Resultsof

samplingcarriedou

tin

Pha

seII;(S)surfacearea,(D)densityindm

-2,(V)Variance,

(Id)

stan

dardised

Morisitaindexan

d(s)stan

dard

deviation

Transectdensity

Highdensitystratum

Low

densitystratum

Transecttotal

River

tran

sects

S(m

2 )D

(ind

m-2)

VId

Dmax

(ind

m-2)

S(m

2 )D

(ind

m-2)

VId

D(ind

m-2)

óS(m

2 )total

Total

No

Eo

100

115.5

2223.4

0.54

256

843

10.19

214.48

0.59

21.4

5.3

943

20135

Arnego4

125

26.8

404.5

0.60

104

723

5.80

66.66

0.59

8.9

2.9

848

7541

Ouro

200

15.4

221.9

0.57

92407

2.40

17.03

0.87

6.7

2.4

607

4049

Navia

100

51.7

944.7

0.53

236

1178

0.00

0.00

0.00

4.1

1.0

1278

5185

Salas

759.6

21.6

0.51

28252

2.15

3.59

0.50

3.9

0.8

327

259

Man

deo

100

24.0

254.0

0.58

92810

0.77

0.91

0.16

3.3

1.2

909

3027

Arnego5

125

12.5

41.4

0.51

48719

0.36

0.34

�0.07

2.2

0.5

844

1820

Arnego1

200

6.9

27.3

0.55

40580

0.31

0.30

�0.04

2.0

0.7

780

1550

Masma

100

10.8

36.2

0.52

44861

0.61

1.03

0.52

1.7

0.6

961

1604

Narla

100

2.0

4.0

0.24

20550

1.39

3.95

0.55

1.5

0.6

650

965

Lim

ia60

7.0

11.7

0.16

16759

1.02

1.47

0.49

1.5

0.4

819

1195

Arnego6

100

3.6

13.3

0.56

32270

0.57

0.91

0.42

1.4

0.5

370

512

Alto_

Ulla

5100

5.9

28.8

0.53

44445

0.24

0.24

-1.3

0.4

545

699

Tam

bre

300

3.3

5.8

0.47

201905

0.57

0.70

0.23

0.9

0.3

2205

2089

Arnego2

100

4.0

7.3

0.50

20596

0.16

0.16

0.00

0.7

0.2

696

495

Alto_

Ulla

6150

1.1

1.2

0.08

12625

0.43

1.29

1.00

0.6

0.4

773

431

Alto_

Ulla

2200

2.4

4.9

0.51

12890

0.00

0.00

-0.4

0.1

1090

480

Alto_

Ulla

4125

1.9

3.2

0.50

12519

0.00

0.00

-0.4

0.1

644

239

Arnego3

100

1.0

3.0

1.00

12341

0.14

0.14

0.00

0.3

0.2

441

149

Tea

200

2.6

5.5

2.77

241414

0.00

0.00

0.00

0.3

0.1

1614

516

Alto_

Ulla

1150

1.7

1.8

0.05

8842

0.00

0.00

-0.3

0.1

992

257

Alto_

Ulla

3125

0.3

0.3

�0.04

4379

0.00

0.00

-0.1

0.1

504

38

Cha

nnel

density

Cha

nnels

S(m

2 )D

(ind

m-2)

VId

Dmax

(ind

m-2)

Total

No

Cam

ba_C

287.0

47.8

15.43

0.52

332

13719

Lan

dro_

C143.2

3.8

1.10

0.19

16554

Tea_C

187.8

2.5

1.19

0.52

24469

S. LOIS ET AL.40

Copyright # 2013 John Wiley & Sons, Ltd. Aquatic Conserv: Mar. Freshw. Ecosyst. 24: 35–47 (2014)

rivers. The number of individuals present in eachriver (Figure 5) was also estimated (Table 3). Forthe rivers in which presence of small individualswas observed, all were estimated to contain morethan 3000 individuals. The basin with the mostrivers containing M. margaritifera was the Miñodespite having low numbers of individuals. TheRiver Eo had the highest estimated number ofindividuals (>45000). It was estimated that there aremore than 188 000 individuals of M. margaritiferain the 54 rivers of Galicia.

DISCUSSION

Most populations ofM.margaritifera are known to bedecreasing in number or have become extinct in Europe(Bauer, 1986, 1988; Ziuganov et al., 1994; Beasley and

Roberts, 1996; Beasley et al., 1998; Cosgrove et al.,2000; Araujo and Ramos, 2001; Young et al., 2001a;Bespalaya et al., 2007; Moorkens et al., 2007; Geist,2010). Populations of M. margaritifera have been

Figure 3. Log-linear relationship between bank density of M. margaritifera (Phase I) and transect density (Phase II).

Figure 4. Density of M. margaritifera in 54 Galician rivers. Symbolsdenote the four categories of density: very high (>5.0 ind m-2), high(>0.5–5 indm-2), medium (0.1–0.5 indm-2) and low density (<0.1 indm-2).

Table 2. Number of individuals less than 65 mm long found in thequantitative sampling of Phase II

River

Length>50–65mm

Length30–50 mm

Length<30 mm

Total<65 mm

%Samples

Navia 24 2 0 26 24Eo 11 3 0 14 10Tea_C 8 1 0 9 17.5Limia 3 2 2 7 10Salas 2 2 0 4 10Alto_Ulla 1 4 0 0 4 8Masma 1 2 0 3 6Mandeo 1 1 0 2 5Ouro 0 2 0 2 2.5Arnego 4 2 0 0 2 2Tambre 1 0 0 1 1.7

A CORE AREA FOR CONSERVATION OF MARGARITIFERA MARGARITIFERA 41

Copyright # 2013 John Wiley & Sons, Ltd. Aquatic Conserv: Mar. Freshw. Ecosyst. 24: 35–47 (2014)

found in 54 rivers of Galicia, 27 of which had notbeen documented before and information has beenobtained on 522 new localities where this speciesis present.

This survey confirmed the presence of M.margaritifera for all previous records for Galicia(Macho, 1878; Azpeitia, 1933; Rolán andOtero-Schmitt, 1996; San Miguel, 1999; Ziuganov

Table 3. Estimated number of individuals, sampled surface area, length, number of sampling points in which the species occurred (N), average densityper river (ind m-2), standard deviation (SD), minimum density (Min), (*) presence of individuals less than 65 mm and maximum density (Max)

RiverEstimated no.of individuals Surface (m2) Length (m) N Average SD Min Max

1 Eo* 45334 42193 2565 46 1.341 3.550 0.00727 21.4002 Navia* 30555 33983 1667 34 1.290 2.740 0.072 14.1203 Masma* 19405 20256 1740 30 0.889 0.860 0.021 3.7004 Camba 15588 5695 651 24 2.430 9.510 0.019 47.7805 Arnego 14085 13654 1240 31 0.620 1.626 0.014 8.9006 Salas* 8482 28127 1130 19 0.660 1.110 0.014 3.8607 Ulla (alto)* 7982 37000 3700 43 0.270 0.348 0.008 1.4508 Landro 7703 13168 1860 26 0.567 1.051 0.011 3.8609 Mandeo* 7569 21672 1445 33 0.397 0.853 0.008 3.70010 Tambre* 7397 41530 2503 45 0.146 0.282 0.007 1.26011 Narla 7333 14638 1865 34 0.498 0.560 0.007 2.23012 Ouro* 6731 11390 1440 25 0.603 1.495 0.009 6.70013 Tea* 3468 18752 1030 21 0.365 0.421 0.018 2.50014 Limia* 3182 8244 615 21 0.347 0.421 0.018 1.50015 Avia 859 7150 350 4 0.180 0.249 0.020 0.54016 Támoga 578 4425 430 6 0.087 0.050 0.024 0.12817 Tórdea 377 6138 935 12 0.069 0.070 0.015 0.22818 Anllo 194 4160 620 9 0.043 0.038 0.012 0.12919 Mera (Miño) 176 1825 520 5 0.078 0.046 0.022 0.14320 Eume 158 1700 150 3 0.105 0.089 0.038 0.20721 Lavandeira 137 895 180 2 0.158 0.179 0.031 0.28522 Miño (Alto) 137 4750 260 5 0.027 0.013 0.018 0.04023 Ladra 98 8000 320 2 0.012 0.001 0.011 0.01324 Xallas 98 4500 300 6 0.028 0.019 0.014 0.06025 Chamoso 94 2630 490 5 0.047 0.026 0.016 0.08026 Parga 91 2415 230 3 0.054 0.023 0.020 0.06827 Suarna (Lamas) 86 7125 750 5 0.015 0.025 0.008 0.02028 Ulla (medio) 82 1735 250 6 0.043 0.021 0.022 0.06829 Furelos 75 250 50 1 0.31030 Deza 75 7500 500 2 0.011 0.005 0.007 0.01431 Zas 64 1200 300 6 0.053 0.052 0.018 0.15032 Rosende 59 1680 420 4 0.019 0.010 0.008 0.02933 Rodil 54 3000 200 1 0.01834 Asneiro 49 620 310 3 0.022 0.020 0.008 0.04435 Mera 48 1125 150 3 0.032 0.013 0.018 0.04036 Xubia 40 2325 310 3 0.017 0.013 0.010 0.03037 Lodoso 33 375 50 1 0.09038 Maior 32 1440 740 3 0.038 0.030 0.008 0.06839 Neira 26 3000 120 1 0.00940 Castro (Xubia) 25 960 240 2 0.026 0.025 0.009 0.04441 Lerez 20 1275 170 2 0.017 0.009 0.010 0.02442 Cabe 17 1350 250 2 0.013 0.008 0.007 0.01843 Arnoia 17 1250 110 1 0.01444 Toxa 16 1875 250 2 0.008 0.000 0.008 0.00945 Umia 16 900 100 1 0.01846 Requeixo 15 1500 200 1 0.01047 Bibei 13 1120 280 2 0.011 0.002 0.009 0.01248 Mero 10 600 100 1 0.01849 Mestas 9 625 250 3 0.015 0.005 0.010 0.01850 Barcés 8 900 120 1 0.00951 Lambre 8 750 100 1 0.01052 Trimaz 8 750 100 1 0.01053 Reigadas 7 480 120 1 0.02854 Ferreirías 7 280 70 1 0.02455 Pambre 4 400 100 1 0.010

S. LOIS ET AL.42

Copyright # 2013 John Wiley & Sons, Ltd. Aquatic Conserv: Mar. Freshw. Ecosyst. 24: 35–47 (2014)

et al., 2000; Araujo and Ramos, 2001; Grande et al.,2001; Machordom et al., 2003; Ondina et al., 2003;Fernández de la Cigoña, 2004; San Miguel et al.,2004; Bouza et al., 2007; Araujo, 2008; Outeiroet al., 2008;), except for that presented by Seoane(1866) for the River Sionlla, in which no livingindividuals or shell remains were found. Thepresence of M. margaritifera in all large river basinsof Galicia suggests that the species was broadlydistributed in the rivers of this area in the past. Thisis supported by one of the few historical accounts inwhich M. margaritifera is regarded as a commonspecies in the rivers of northern Galicia, from theEo to the Tambre (Macho, 1878).

The densities of M. margaritifera recorded in the25 transects of Phase II ranged from 0.1 (RiverUlla) to 47.8 ind m-2 (River Camba) (Table 1).Much of the data on densities of freshwatermussels is also based on the use of samplesdelimited by quadrats, but in most cases these

samples are intentionally located in areas with thehighest abundances, therefore their values arebiased and only represent the density of specificareas of the river bed (Cawley, 1993). Thus, thosedata would only be comparable with densitiesfound in high density strata (up to 256 ind m-2 in theRiver Eo) or in mill channels (up to 332 ind m-2 inthe River Camba) (Table 1). These are high values,similar to those found in some of the largestpopulations of the species, such as those present inRussia (194 ind m-2, in Ziuganov et al., 1994) or inScotland (398 ind m-2 in Hastie et al., 2000b).

At present,M. margaritifera is spatially distributedin a fragmented way, gathering only in specific areasof the river channel. The estimated densities fordifferent sampling points of the same river showhigh variability, as in the case of the Río Cambawhose sampling points had values close to zero(0.019 ind m-2) and up to 47.8 ind m-2 (Table 3).Strayer (2008) stated that freshwater musselpopulations and their habitats are heterogeneousand fragmented. Although distribution andabundance patterns of populations are the resultof the interaction among different causes (Hilbornand Stearns, 1982), in the case of unionoids theconnection among patches depends, above all, onthe mobility and density of host fish (Arvidssonet al., 2012). This dispersal strategy makes itpossible to link different patches, which enablesthe colonization of different areas and therecovery of patches with low densities (Strayer,2008). In Galicia, the connection among thedifferent habitat patches along the river is severedby many dams, which limit the movement ofthe host fish and, in some cases, the access ofsalmon and sea trout to the mussel populations.Habitat fragmentation by dams is thought to beresponsible for the extinction of salmon in somerivers of Galicia (Hervella and Caballero, 2002). Itis notable that five of the Galician rivers containingsmall individuals are still connected to the sea.

The genetic studies ofM. margaritifera previouslycarried out in seven Galician rivers that are includedin the present paper, revealed a high degree ofstructuring and a reduced genetic intrapopulationdiversity (Bouza et. al., 2007), which could beindicative of the low rates of gene flow. However,the discovery of additional populations and new

Figure 5. Classification of Galician rivers according to the estimatednumber of individuals of M. margaritifera. Rivers where individuals<65 mm were observed are denoted with a ‘*’. Numbers represent

different rivers and tributaries (see Table 3).

A CORE AREA FOR CONSERVATION OF MARGARITIFERA MARGARITIFERA 43

Copyright # 2013 John Wiley & Sons, Ltd. Aquatic Conserv: Mar. Freshw. Ecosyst. 24: 35–47 (2014)

sites of occupancy in Galicia will help expand geneticand ecological studies of the species and increaseknowledge on its population dynamics in Galicianrivers.

In general, the number of individuals present in ariver and the recruitment rate are criteria used toassess population status. According to some studies,a population of 5000 individuals can be consideredas ‘large’ (Bauer, 1991; Hastie and Cosgrove, 2002).In the present study, 12 rivers were found to exceedthis number (Figure 5, Table 3).

Individuals less than 65 mm long were detected inPhase I in three of the rivers sampled (Limia, Eoand Tea). However, quantitative sampling provedto be the most efficient for locating smallindividuals, whose presence was confirmed in 11rivers belonging to 10 different basins (Eo, Navia,Limia, Mandeo, Masma, Narla, Landro, Ouro,Tambre, Arnego and Ulla) (Table 2). Individualsless than 30 mm were found in the River Limia(Table 2). However, the number of rivers withjuveniles has probably been underestimated, asquantitative sampling was carried out only in alimited number of rivers. Hastie and Cosgrove(2002) and Hastie (2011) emphasized the difficultiesof detecting small individuals on the river bed,which leads to underestimation of the abundanceof juveniles in populations. Individuals less than10 mm long go unnoticed, even when using sieves(Young and Williams, 1984) and those less than20 mm are difficult to find in many cases. Thus,Hastie (2011) suggests that a population should beconsidered to have recent recruitment whenincluding individuals less than 30 mm. In Galicia,according to data obtained by Outeiro et al. (2008),only the River Eo was classified within thiscategory. The results of the present study suggestthat recruitment in the River Limia is also recent.

In general, individuals less than 65mmwere foundin areas near the river bank. Despite some indicationsthat adults and juveniles may take up differenthabitats within the river bed (Geist and Auerswald,2007), these results showed that most individualsless than 65 mm shared the same habitat as adults,consistent with Hastie et al. (2000b).

Although important, there are no general criteriafor assessing populations according to their type ofrecruitment. Thus, Cosgrove et al. (2000) regarded

populations in Scotland as ‘functional’ when theycontained at least a specimen less than 65 mm,regardless of the population size, whereas Hastie(2011) established a set of criteria for Scottishpopulations based on the percentage of individualsless than 65 mm with the aim of determiningwhether recruitment is sufficient to sustain long-termpopulations. In order to assess recruitmentobjectively, it is necessary to define ‘juvenile’ on thebasis of the species’ biology (Cosgrove et al., 2000).Galician populations of M. margaritifera have thehighest growth rates, the shortest lifespan knownand, according to the analysis of their growthpattern, they probably reach sexual maturity earlierthan other populations (San Miguel et al., 2004).Thus, to determine whether recruitment is sufficientto sustain populations, a demographic analysis isneeded based on the distribution of age frequencies.

Galicia has the largest number of populationsof M. margaritifera within the Iberian Peninsulaand the total number of mussels estimated in itsrivers exceeds that found in territories withgreater surface areas including Germany (69populations and 144 000 individuals), France(84 populations and a maximum of 100 000individuals) and Austria (29 populations and50 000 individuals) (Geist, 2010). Moreover, theapparent recruitment observed in 11 Galician rivers isin contrast to the accounts that many Europeanpopulations are not reproducing successfully (Younget al., 2001a; Geist, 2010). Thus, the populationsin Galicia are important for conservation of M.margaritifera and they provide new opportunities tostudy its ecological requirements and geneticdiversity, especially in populations that aresuccessfully producing small mussels.

Although peripheral populations usually inhabitless favourable habitats and show low andvariable densities (Brown, 1984; Gaston, 1990;Brown et al., 1995), recent studies (Channell andLomolino, 2000a, b) have indicated that thesepopulations may be very persistent over time. Onthe other hand, Hampe and Petit (2005) arguedthat populations at the periphery of a speciesrange, especially those living in lower latitudessuch as M. margaritifera in Galicia, are ofgreat relevance for the conservation of genediversity and evolutionary potential of the species.

S. LOIS ET AL.44

Copyright # 2013 John Wiley & Sons, Ltd. Aquatic Conserv: Mar. Freshw. Ecosyst. 24: 35–47 (2014)

This underscores the need for higher priorityresearch and conservation of the populations ofM. margaritifera in Galicia. The extensive baselinedata reported here will enable future researchon ecological requirements and environmentallimitations on M. margaritifera, and they providea solid foundation for conservation planning.

ACKNOWLEDGEMENTS

This study was supported financially by the Xunta deGalicia (Project No. 07MDS018261PR). We thankDr Ramón Mascato and Dr Rafael Romero fortheir great help during this project. We are gratefulto Dr David E. Cowley for the review of themanuscript and especially to the biologists FernandoR. Brea and Jesús Latas (Consellería de MedioAmbiente) for assisting with locating sampling sites.

REFERENCES

Álvarez-Claudio C, García-Rovés P, Ocharan R, Cabal J,Ocharan F, Álvarez M. 2000. A new record of thefreshwater pearl mussel Margaritifera margaritifera L.(Bivalvia, Unionoida) from the River Narcea (Asturias,north-western Spain). Aquatic Conservation: Marine andFreshwater Ecosystems 10: 93–102.

Araujo R. 2008.Mejillón de ríoM.margaritifera (L.). InAtlas delos Invertebrados Amenazados de España (especies en peligrocrítico y en peligro). Verdú JR, Galante E (eds). OrganismoAutónomo Parques Nacionales: Madrid; 246–253.

Araujo R, Ramos MA. 2001. Action plans for Margaritiferaauricularia and Margaritifera margaritifera in Europe.Convention on the Conservation of European Wildlife andNatural Habitats, Council of Europe, Nature andEnvironment 17, Strasbourg.

Arvidsson BL, Karlsson J, Österling ME. 2012. Recruitment ofthe threatened mussel Margaritifera margaritifera in relationto mussel population size, mussel density and host density.Aquatic Conservation: Marine and Freshwater Ecosystems22: 526–532.

Azpeitia F. 1933. Conchas bivalvas de agua dulce de España yPortugal. Memorias del Instituto Geológico y Minero deEspaña: Madrid.

Bauer G. 1986. The status of the freshwater pearl musselMargaritifera margaritifera L. in the south of its Europeanrange. Biological Conservation 38: 1–9.

Bauer G. 1988. Threats to the freshwater pearl musselMargaritifera margaritifera L. in central Europe. BiologicalConservation 45: 239–253.

Bauer G. 1991. Plasticity in life history traits of the freshwaterpearl mussel – consequences for the danger of extinctionand for conservation measures. In Species Conservation: a

Population Biology Approach, Seitz A, Loeschcke V (eds).Birkhäuser Verlag: Basel; 103–120.

Bauer G. 2000. Life-history variation on different taxonomiclevels of naiads. In Ecology and Evolution of the FreshwaterMussels Unionoida, Bauer G, Wächtler K (eds). Springer:New York; 83–92.

Beasley C, Roberts D. 1996. The current distribution and statusof the freshwater pearl mussel Margaritifera margaritifera L.1758 in north-west Ireland. Aquatic Conservation: Marineand Freshwater Ecosystems 6: 169–177.

Beasley C, Roberts D, Mackie T. 1998. Does the freshwaterpearl mussel, Margaritifera margaritifera L., face extinctionin Northern Ireland? Aquatic Conservation: Marine andFreshwater Ecosystems 8: 265–272.

Bespalaya YV, Bolotov IN, Makhrov AA. 2007. State of thepopulation of the European pearl mussel Margaritiferamargaritifera (L.) (Mollusca, Margaritiferidae) at thenortheastern boundary of its range (Solza River, White SeaBasin). Russian Journal of Ecology 8: 204–211.

Bouza C, Castro J, Martínez P, Amaro R, Fernández C,Ondina P, Outeiro A, San Miguel E. 2007. Threatenedfreshwater pearl mussel Margaritifera margaritifera L. inNW Spain: low and very structured genetic variation insouthern peripheral populations assessed usingmicrosatellite markers. Conservation Genetics 8: 937–948.

Brown JH. 1984. On the relationship between abundance anddistribution of species. The American Naturalist 124: 255–279.

Brown JH, Mehlman DW, Stevens GC. 1995. Spatial variationin abundance. Ecology 76: 2028–2043.

Cawley E. 1993. Sampling adequacy in population studies offreshwater mussels. In Conservation and Management ofFreshwater Mussels. Proceedings of an Upper MississippiRiver Conservation Committee (UMRCC) Symposium, St.Louis, Missouri, October 1992, Cummings KS, BuchananAC, Koch LM (eds). Upper Mississippi River ConservationCommittee: Rock Island, IL; 12–14.

Channell R, Lomolino MV. 2000a. Dynamic biogeographyand conservation of endangered species. Nature 403: 84–86.

Channell R, Lomolino MV. 2000b. Trajectories to extinction:spatial dynamics of the contraction of geographical ranges.Journal of Biogeography 27: 169–179.

Cosgrove P, Young M, Hastie L, Gaywood M, Boon P. 2000.The status of the freshwater pearl mussel Margaritiferamargaritifera Linn. in Scotland. Aquatic Conservation:Marine and Freshwater Ecosystems 10: 197–208.

Council of the European Communities. 1992. Directive 92/43/EEC of the Council of the European Communities of 21stMay 1992 on the conservation of natural habitats and ofwild fauna and flora. Official Journal of the EuropeanCommunities L206: 7–50.

Dudgeon D, Arthington A, Gessner M, Kawabata Z, KnowlerD, Lévêque C, Naiman R, Prieur-Richard A, Soto D,Stiassny M, Sullivan C. 2006. Freshwater biodiversity:importance, threats, status and conservation challenges.Biological Reviews 81: 163–182.

ESRI. 2009. ArcMap version 9.3. ESRI, Redlands, California.Fernández de la Cigoña E. 2004. A Voltas coa Fauna Galega(Traballos e Novas sobre Invertebrados, Peixes, Réptiles,Aves e Mamíferos da Nosa Terra). Asociación Galega paraa Cultura e a Ecoloxía: Vigo, Pontevedra.

Fortin MJ, Keitt T, Maurer B, Taper M, Kaufman DM,Blackburn T. 2004. Species’ geographic ranges and

A CORE AREA FOR CONSERVATION OF MARGARITIFERA MARGARITIFERA 45

Copyright # 2013 John Wiley & Sons, Ltd. Aquatic Conserv: Mar. Freshw. Ecosyst. 24: 35–47 (2014)

distributional limits: pattern analysis and statistical issues.Oikos 108: 7–17.

Gaston KJ. 1990. Patterns in the geographical ranges ofspecies. Biological Reviews 65: 105–129.

Geist J. 2010. Strategies for the conservation of endangeredfreshwater pearl mussels (Margaritifera margaritifera L.): asynthesis of conservation genetics and ecology.Hydrobiologia 644: 69–88.

Geist J, Auerswald K. 2007. Physicochemical stream bedcharacteristics and recruitment of the freshwater pearlmussel (Margaritifera margaritifera). Freshwater Biology 52:2299–2316.

Grande C, Araujo R, Ramos M. 2001. The gonads ofMargaritifera auricularia (Spengler, 1793) and M.margaritifera (Linnaeus, 1758) (Bivalvia: Unionoidea).Journal of Molluscan Studies 67: 27–35.

Gum B, Lange M, Geist J. 2011. A critical reflection on thesuccess of rearing and culturing juvenile freshwater musselswith a focus on the endangered freshwater pearl mussel(Margaritifera margaritifera L.). Aquatic Conservation:Marine and Freshwater Ecosystems 21: 743–751.

Hampe A, Petit RJ. 2005. Conserving biodiversity underclimate change: the rear edge matters. Ecology Letters 8:461–467.

Hastie LC. 2011. Are Scottish freshwater pearl mussel populationsrecruiting normally? Toxicological and Environmental Chemistry93: 1748–1763.

Hastie LC, Cosgrove PJ. 2002. Intensive searching for musselsin a fast-flowing river: an estimation of sampling bias.Journal of Conchology 37: 309–316.

Hastie LC, Boon PJ, Young MR. 2000a. Physical microhabitatrequirements of freshwater pearl mussels, Margaritiferamargaritifera (L.). Hydrobiologia 429: 59–71.

Hastie L, Young M, Boon P, Cosgrove P, Henninger B. 2000b.Sizes, densities and age structures of Scottish Margaritiferamargaritifera (L.) populations. Aquatic Conservation:Marine and Freshwater Ecosystems. 10: 229–247.

Hastie L, Tarr E, Al-Mousawi B, YoungM. 2010.Medium-termrecruitment patterns in Scottish freshwater pearl musselMargaritifera margaritifera populations. Endangered SpeciesResearch 11: 21–33.

Hervella F, Caballero P. 2002. Situación actual del SalmónAtlántico en Galicia. In El Salmón Joya de nuestros ríos.Xunta de Galicia: Santiago de Compostela; 27–37.

Hilborn R, Stearns SC. 1982. On inference in ecology andevolutionary biology: the problem of multiple causes. ActaBiotheoretica 31: 145–164.

IUCN. 2012. IUCN Red List of Threatened Species. Version2013.1. <www.iucnredlist.org>. Downloaded on 30January 2013.

Krebs CJ. 1999. Ecological Methodology, 2nd edn. Addison-WesleyEducational Publishers Inc: Menlo Park, CA.

Lydeard C, Cowie R, Ponder W, Bogan A, Bouchet P, Clark S,Cummings K, Frest T, Gargominy O, Herbert D, et al. 2004.The global decline of nonmarine mollusks. Bioscience 54: 21–330.

Macho J. 1878. Moluscos de auga dulce de Galicia, conobservaciones sobre las especies y localidades. Anales de laReal Sociedad Española de Historia Natural 7: 131–190.

Machordom A, Araujo R, Erpenbeck D, Ramos MA. 2003.Phylogeography and conservation genetics of endangeredEuropean Margaritiferidae (Bivalvia: Unionoidea).Biological Journal of the Linnean Society 78: 235–252.

Moorkens E, Killeen I, Ross E. 2007. Margaritiferamargaritifera (the freshwater pearl mussel) conservationassessment. Backing Document. A report to the NationalParks and Wildlife Service, Dublin.

Morales J, Negro A, Lizana M, Martínez A, Palacios J. 2004.Preliminary study of the endangered populations of pearlmussel Margaritifera margaritifera (L.) in the River Tera(north-west Spain): habitat analysis and managementconsiderations. Aquatic Conservation: Marine andFreshwater Ecosystems 14: 587–596.

Morisita M. 1959. Measuring of the dispersion and analysis ofdistribution patterns. Memoires of the Faculty of Science,Series E. Biology, Kyushu University 2: 215–235.

Ondina P, Fernández C, Hermida M, Amaro R, Outeiro A,San Miguel E. 2003. Datos preliminares sobre la presenciade Margaritifera margaritifera (Linnaeus, 1758) en los ríosde Galicia. XV Bienal de la Real Sociedad de HistoriaNatural: A Coruña; 26–27.

Österling M, Arvidsson B, Greenberg L. 2010. Habitatdegradation and the decline of the threatened musselMargaritifera margaritifera: influence of turbidity andsedimentation on the mussel and its host. Journal of AppliedEcology 47: 759–768.

Ostrovsky A, Popov IY. 2011. Rediscovery of the largestpopulation of the freshwater pearl mussel (Margaritiferamargaritifera) in the Leningrad oblast (north-west Russia).Aquatic Conservation: Marine and Freshwater Ecosystems 21:113–121.

Outeiro A, Ondina P, Fernández C, Amaro R, San Miguel E.2008. Population density and age structure of the freshwaterpearl mussel, Margaritifera margaritifera, in two Iberianrivers. Freshwater Biology 53: 485–496.

Reis J. 2003. The freshwater pearl mussel [Margaritiferamargaritifera (L.)] (Bivalvia, Unionoida) rediscovered inPortugal and threats to its survival. Biological Conservation114: 447–452.

Rodríguez F. 2001. Os ríos galegos. Morfoloxía e réxime.Consello da Cultura Galega: Santiago de Compostela.

Rolán E, Otero-Schmitt J. 1996. Guía dos moluscos de Galicia.Galaxia: Santiago de Compostela.

Rudzīte M. 2005. Assessment of the condition of freshwater pearlmussel Margaritifera margaritifera (Linnaeus, 1758) populationsin Latvia. Acta Universitatis Latviensis 691: 121–128.

San Miguel E. 1999. Margaritifera margaritifera en los ríos deGalicia. In Informe sobre la situación del Desmán Ibérico(Galemys pyrenaicus) en España. Seminario sobre Conservaciónde Margaritifera margaritifera y Galemys pyrenaicus en laPenínusla ibérica, Nores C (ed). INDUROT, Universidad deOviedo, Ministerio de Medio Ambiente: Oviedo; 1–4.

San Miguel E, Monserrat S, Fernández C, Amaro R, HermidaM, Ondina P, Altaba C. 2004. Growth models and longevityof freshwater pearl mussels (Margaritifera margaritifera) inSpain. Canadian Journal of Zoology 82: 1370–1379.

Seoane VL. 1866. Reseña de la historia natural de Galicia. 1996ed. Ediciós do Castro: Sada, A Coruña.

Smith DR. 2006. Survey design for detecting rare freshwatermussels. Journal of the North American Benthological Society25: 701–711.

Snyder NFR, Derrickson SR, Beissinger SR, Wiley JW, SmithTB, Toone WD, Miller B. 1996. Limitations of captivebreeding in endangered species recovery. ConservationBiology 10: 338–348.

S. LOIS ET AL.46

Copyright # 2013 John Wiley & Sons, Ltd. Aquatic Conserv: Mar. Freshw. Ecosyst. 24: 35–47 (2014)

Strayer DL. 2008. Freshwater Mussel Ecology: a MultifactorApproach to Distribution and Abundance. University ofCalifornia Press: Berkeley, CA.

Strayer DL, Smith DR. 2003. A guide to Sampling FreshwaterMussel Populations. American Fisheries Society Monograph 8,American Fisheries Society: Bethesda, Maryland.

Sutherland WJ, Pullin AS, Dolman PM, Knight, TM. 2004.The need for evidence-based conservation. Trends inEcology & Evolution 19: 305–308.

Vaughn C, Taylor C. 1999. Impoundments and the decline offreshwater mussels: a case study of an extinction gradient.Conservation Biology 13: 912–920.

Vaughn C, Taylor CM, Eberhard KJ. 1997. A comparison of theeffectiveness of timed searches vs. quadrat sampling in musselsurveys. In Conservation and Management of FreshwaterMussels II. Initiatives for the Future, Cummings KS, BuchananAC, Mayer CA, Naimo TJ (eds). Upper Mississippi RiverConservation Committee: Rock Island, IL; 157–162.

Villella RF, Smith DR. 2005. Two-phase sampling to estimateriver-wide populations of freshwater mussels. Journal of theNorth American Benthological Society 24: 357–368.

Young MR, Williams JC. 1984. The reproductive biologyof the freshwater pearl mussel Margaritifera margaritifera(Linn.) in Scotland. I. Field studies. Archiv für Hydrobiologie99: 405–422.

Young MR, Cosgrove PJ, Hastie LC. 2001a. The extentof, and causes for, the decline of a highly threatenednaiad: Margaritifera margaritifera. In Ecology andEvolution of the Freshwater Mussels Unionoida. Springer:Berlin; 337–357.

Young MR, Cosgrove PJ, Hastie LC, Henniger B. 2001b. Astandardised method for assessing the status of freshwatermussels in clear, shallow rivers. Journal of MolluscanStudies 67: 396–397.

Ziuganov V, Zotin A, Nezlin L, Tretiakov V. 1994. TheFreshwater Pearl Mussels and their Relationships withSalmonid Fish. VNIRO Publishing House: Moscow.

Ziuganov V, San Miguel E, Neves RJ, Longa A, Fernández C,Amaro R, Beletsky V, Popkovitch E, Kaliuzhin S, JohnsonT. 2000. Life span variation of the freshwater pearl shell: amodel species for testing longevity mechanisms in animals.Ambio 29: 102–105.

A CORE AREA FOR CONSERVATION OF MARGARITIFERA MARGARITIFERA 47

Copyright # 2013 John Wiley & Sons, Ltd. Aquatic Conserv: Mar. Freshw. Ecosyst. 24: 35–47 (2014)