Population structure of the Dover sole, Solea solea (L) in the Atlantic using randomly amplified polymorphic DNA (PCR-RAPD) marking

K. Erzini, L. Bentes, P.G. Lino, J. Ribeiro, R. Coelho, P. Monteiro, C. Correia and J.M.S. Gonçalves: Universidade do Algarve, CCMAR/FCMA Campus de Gambelas, 8000 Faro, Portugal [tel: +351 289 800 900, fax: +351 289 818 353, e-mail: [email protected]]

ICES CM 2001/J:16

Population structure of the Dover sole, Solea solea L., in the Atlantic using randomly amplified polymorphic DNA (PCR-RAPD) marking

Athanasios K. Exadactylos, Andreas Y. Troumbis, Audrey J. Geffen, and John P. Thorpe

The primary objective of this paper is to compare patterns of allozyme and RAPD variation in six Atlantic populations of Solea solea. Sixteen individuals were randomly selected from each of the six Atlantic populations of S. solea sampled during 1994 and 1995. In total, 96 individuals were screened from three Irish Sea, two North Sea and one Bay of Biscay populations. Alllozyme and RAPD markers produced results which were in general agreement with each other. However, RAPD markers detected more variation in the sampled sole populations than that observed by using allozyme markers. The percentage of polymorphism and genetic diversity estimates were very high. There was clearly no geographic trend to any RAPD or allozyme allele frequencies at all polymorphic loci. The chi-square contingency analyses (allele frequencies and hierarchical FST and non-hierarchical FDT) indicated a highly significant heterogeneity among populations and between groupings. This conclusion is consistent with the life history of Dover sole, the homing behaviour they exhibit and the discrete spawning areas. A north-to-south pattern of significant population differentiation in the allozyme study was replaced by a northwest-to-southeast significant variation. Natural selection for differences in spawning habitat may explain the genetic isolation between these groupings. Another reason could be the significant absence of some alleles, present in the southern European populations. Evidence in favour of the divergence between these regions is shown by the cluster analyses (UPGMA, Wagner procedure) of both allozyme and RAPD data. However, allozymes provided a slightly better fit of data to predictions about genetic distance (higher values of cophenetic correlation of clusters from the “goodness-of-fit” statistics) and better correlation between genetic and geographic distances (Mantel’s r). In both studies a strong similarity between the Bay of Biscay and the German Bight populations was evident, judging from their clustering positions. It was apparent that the trend of probable movement of migrants through the English Channel suggested by the allozyme study was supported using RAPD markers. The apparent absence of isolation-by-distance (non-significant correlation of genetic with geographic distance) in both studies suggests that S. solea may not yet be at genetic equilibrium. Overall, S. solea conforms to the type of species with either a broad variety of geographic and ecological habitats, or large and stable population sizes, or substantial gene flow at a regional scale.

A. Exadactylos: Biodiversity Conservation Laboratory, Department of Environmental Studies, University of the Aegean, Mytilene, GR81100, Hellas [tel: +0251–36247, e-mail: [email protected]]. A.Y. Troumbis: Biodiversity Conservation Laboratory, Department of Environmental Studies, University of the Aegean, Mytilene, GR81100, Hellas [tel: +0251–36230, e-mail: [email protected]]. J.P. Thorpe: Port Erin Marine Laboratory, School of Biological Sciences, The University of Liverpool, Port Erin, Isle of Man, IM9 6JA, UK [tel: +01624–831010, e-mail: J.P.Thorpe@ liverpool.ac.uk]

ICES CM 2001/J:17 – WITHDRAWN

ICES CM 2001/J:18

Automated fecundity and egg-sizing for marine species from scanned images: rapid sample processing and large data volumes

Kevin Friedland, D. Ama-Abasi, L. Clarke, D. Sigourney, and A. Stein

One of the most fundamental parameters in fisheries management and ecology is the population production of eggs. Management decisions are largely based on the dynamics of spawning stock biomass, and it is essential to know the contribution of viable gametes by age, which is affected at younger ages by growth and maturity rates and at older ages by the physiology of reproductive senescence. In addition to population modelling, recruitment dynamics are also strongly influenced by the size and quality of eggs. Egg quality can be characterized by a number of different qualitative and quantitative approaches; however, the most attractive methods developed to date rely on techniques that are efficient and lend themselves to wide application. The measurement of fecundity and egg size for fish and invertebrate fishery resources has been plagued by the cumbersome nature of the traditional approaches. We have developed an imaging-based technique that allows for the reduction of samples to electronic counts in relative little time and with a favourable costing. The technique is dependant upon high resolution scans of plated egg samples and the

7

CM 2001/J:16 The Life History, Dynamics and Exploitation of Living

Marine Resources: Advances in knowledge and methodology

1

�OT TO BE CITED WITHOUT PRIOR REFERE�CE TO THE AUTHOR

POPULATIO� STRUCTURE OF THE DOVER SOLE, SOLEA

SOLEA L., I� THE ATLA�TIC USI�G RA�DOMLY

AMPLIFIED POLYMORPHIC D�A (PCR-RAPD) MARKI�G

ATHANASIOS EXADACTYLOS1, ANDREAS Y. TROUMBIS

1, AUDREY J. GEFFEN

2,

JOHN P. THORPE2

1Biodiversity Conservation Laboratory, Department of Environmental Studies, University of the

Aegean, Mytilene, GR 81100, E-mail: [email protected].

2Port Erin Marine Laboratory, School of Biological Sciences, The University of Liverpool, Port Erin,

Isle of Man, IM9 6JA.



2

ABSTRACT

The primary objective of this work is to compare patterns of allozyme and RAPD variation in six

Atlantic populations of Solea solea. Sixteen individuals were randomly selected from each of the six

Atlantic populations of S. solea sampled during 1994 and 1995. In total, 96 individuals were

screened from three Irish Sea, two North Sea and one Bay of Biscay populations. Allozyme and

RAPD markers produced results, which were in general agreement with each other. However, RAPD

markers detected more variation in the sampled sole populations than that observed by using allozyme

marker. Percentage of polymorphism and genetic diversity estimates were very high. There was

clearly no geographic trend to any RAPD or allozyme allele frequencies at all polymorphic loci. The

chi-square contingency analyses (allele frequencies and hierarchical FST and non- hierarchical FDT)

indicated a highly significant heterogeneity among populations and between groupings. This

conclusion is consistent with the life history of Dover sole, the homing behaviour they exhibit and the

discrete spawning areas. A north to south pattern of significant population differentiation in the

allozyme study was replaced by a northwest to southeast significant variation. Natural selection for

differences in spawning habitat may explain the genetic isolation between these groupings. Another

reason could be the significant absence of some alleles, present in the southern European populations.

Evidence in favour of the divergence between these regions is shown by the cluster analyses

(UPGMA, Wagner procedure) of both allozyme and RAPD data. However, allozymes provided a

slightly better fit of data to predictions about genetic distance (higher values of cophenetic correlation

of clusters from the Goodness of fit statistics) and better correlation between genetic and geographic

distances (Mantel’s r). In both studies a strong similarity between the Bay of Biscay and the German

Bight populations was evident, judging from their clustering positions. It was apparent that the trend

of probable movement of migrants through the English Channel suggested by the allozyme study was

supported using RAPD markers. The apparent absence of isolation-by-distance (non significant

correlation of genetic with geographic distance) in both studies suggests that S. solea may not yet be

at genetic equilibrium. Overall, S. solea conforms to the type of species with either a broad variety of

geographic and ecological habitats, or large and stable population sizes, or substantial gene flow at a

regional scale.



3

1. I�TRODUCTIO�

Estimates of genetic variation are increasingly being based upon information at the DNA level

as distinct from allozyme electrophoresis. Random amplification of polymorphic DNA (RAPD) by

polymerase chain reaction (PCR), or PCR-RAPD, is a means of detecting polymorphisms rapidly for

genetic mapping and strain identification (Welsh and McClelland, 1990, 1991; Williams et al., 1990;

Hadrys et al., 1992). The method can detect extensive polymorphisms and does not require a priori

sequence information on the organism studied. Polymorphisms result from either size changes in the

amplified region or base changes that alter primer binding (Lynch and Milligan, 1994). Such

polymorphisms are inherited in a Mendelian fashion and can be used as genetic markers (Bardakci

and Skibinski, 1994). PCR-RAPD is simpler, considerably faster and less expensive than other

methods for detecting DNA sequence variation (Caetano-Anolles et al., 1991a, b). The fact that

RAPD scan numerous loci in the genome makes the method generally attractive for analysis of

genetic distance, genetic structure at a population level, breeding applications, or phylogeny

reconstruction (Clark and Lanigan, 1993).

Practical problems for RAPD applications include the difficulty in distinguishing products of

different loci with similar molecular weights (comigration) (Grosberg et al., 1996). Polymorphism is

detected as band presence versus absence after PCR amplification, since RAPD markers are usually

dominant characters (Williams et al., 1990). These phenotypically dominant markers in diploid

organisms make it impossible to distinguish between the genotypes of homozygous individuals and

those heterozygous for the dominant allele (Lynch and Milligan, 1994). This problem does not

prevent the estimation of allele frequencies necessary for population genetic analysis, but it does

reduce the accuracy of such estimation compared to analysis with codominant markers (i.e.

allozymes) (Lynch and Milligan, 1994). Each RAPD locus must be treated as a two allele system,

with the assumption that only one of the alleles per locus is amplifiable by the PCR. The existence of

multiple amplifiable alleles at a locus is reported to be relatively rare (Clark and Lanigan, 1993).

Under random mating the inbreeding index FIS equals to zero and the expected genotype frequencies

follow the Hardy-Weinberg expectations. This is another assumption that has to be considered when

carrying out population-genetic structure analyses using dominant genetic markers (Stewart and

Excoffier, 1996).



4

Little is known about the capacity of RAPD to detect variation in a broader variety of natural

populations, or how such variation compares with that of allozyme-based analysis. A comparison of

allozyme and RAPD diversity is of interest. The allozyme database, although carrying its limitations,

is quite large and for some types of questions allozymes may still provide sufficient information. For

other questions, DNA based methods may prove superior (Grosberg et al., 1996). The primary

objective of this work is to compare patterns of allozyme and RAPD variation in six Atlantic

populations of Solea solea, using some of the more recently developed statistical methods. We have

already reported a preliminary allozyme study (Exadactylos et al., 1998), and we follow here in this

work with a RAPD survey to estimate levels of genetic diversity within populations and genetic

divergence between populations based on the same material.

2. MATERIALS A�D METHODS

Sixteen individuals were randomly selected from each of six Atlantic populations of S. solea

sampled during 1994 and 1995. In total, 96 individuals were screened from three Irish Sea, two

North Sea and one Bay of Biscay populations. Frozen tissue samples of skeletal muscle stored at -75

oC were used for DNA extractions. Preliminary trials were contacted on juvenile sole to evaluate the

better of two DNA extraction techniques with phenol-chloroform (Exadactylos, 1997).

2.1. DATA ANALYSIS

The entire data set of 96 individuals was used for the analysis, including the observed OPF15-

B monomorphic marker. The computer program RapdBios (Black, 1993) provides input files suitable

for Biosys-2 (Swofford and Selander, 1989), which was used to calculate allele frequencies,

heterozygosity estimates and standard genetic analyses of population structure. Heterogeneity χ2

analyses of allele frequencies among populations and within geographic basins were performed by

hierarchical analysis using Biosys-2. The sequential Bonferroni procedure was employed to control

the probability of incorrectly rejecting null hypotheses at the 0.005 level of significance. Calculation

of Wright’s (1978) non-hierarchical FDT was carried out in Biosys-2 to facilitate comparison of this

paper’s data set with that obtained from the allozyme study (Exadactylos et al., 1998). Weir and

Cockerham’s (1984) unbiased estimates of F-statistics were calculated using the RapdFST program,

version 3.0 (Black, 1995), using the weighted mean frequency of alleles. Contemporary estimates of

gene flow (�m) were calculated using Slatkin’s (1993) approximation, assuming current equilibrium

between migration and genetic drift. Two methods of cluster analyses were performed (UPGMA and



5

Wagner procedure), in order to reveal the better fitted dendrogram, according to the Goodness of fit

statistics, using Modified Rogers Distance (DT) and Prevosti’s Distance (Wright, 1978) respectively.

The relation between matrices of genetic and linear geographic distances was examined with a

Mantel’s (1967) test of time-spacing clustering using the Ntsys program (Rohlf, 1993).

3. RESULTS

From a total of 38 loci 37 were highly polymorphic (100%) in all six populations. Table 1

reports genetic diversity estimates for all 38 markers in all samples. The mean expected genetic

diversity estimate was very high (H = 0.339 ± 0.033) but did not vary among populations. The mean

genetic diversity estimates were not significantly related to latitude (r = 0.616, F1,5 = 2.448, P =

0.193). Contingency chi-square analysis for heterogeneity of allele frequencies across all six

populations indicated significant differences at 22 loci even after the Bonferroni correction at a´ =

0.005 for multiple tests. The samples were grouped at several spatial scales, but significant

differences in allele frequencies were detected only between the three Irish Sea and the three

European Coast populations. Overall, there was extensive heterogeneity, even greater than that

observed with the allozyme data set.

Table 1: Genetic variability at 38 loci in six populations of S. solea. Standard errors in brackets.

Mean diversity (H)

Population Mean sample Mean no. of Percentage Expected** under

size per alleles per of loci Hardy-Weinberg

locus [S.E] locus [S.E] polymorphic* equilibrium [S.E]

1. CUM 16 1.7 100 0.313

[0] [0.1] [0.034]

2. IOM 16 1.8 100 0.385

[0] [0.1] [0.029]

3. GER 16 1.7 100 0.315

[0] [0.1] [0.036]

4. IRL 16 1.8 100 0.326

[0] [0.1] [0.033]

5. EA� 16 1.7 100 0.301

[0] [0.1] [0.037]

6. FRA 16 1.9 100 0.393

[0] [0.1] [0.027]

Total 0.339

[0.033]

* A locus is considered polymorphic if the frequency

of the most common allele does not exceed 0.99

** Unbiased estimate (Nei, 1978)

Among population differentiation (mean unbiased FST) estimated from RapdFST was 0.193 and

was statistically significant (χ2 = 37.056; d.f. = 5, P ≤ 10

-6), indicating that among population

variation was detectable. Weir and Cockerham’s FST estimates were significantly greater than zero for



6

21 loci, indicating a high level of population differentiation. Among them, OPF17-C was the locus

with the greatest overall genetic variation; the A allele had the greatest FST value and was therefore

the most variable between the sampling locations. It was evident that its frequency increased

gradually from north-west (Irish Sea) to south-east (European Coast). The mean estimate of Wright’s

non-hierarchical FDT calculated from Biosys-2 was 0.162 and was also significant (χ2 = 31.104; d.f. =

5, P ≤ 10-4). Gene flow estimates of approximately one immigrant into the average deme per

generation were calculated from both these FST values. Gene flow was considerably less when

estimated from RAPD (one migrant) as opposed to allozyme (11 migrants) data (Table 2).

Table 2: Estimates of among population differentiation (F-statistics), gene flow and correlation

of genetic distance with approximate linear geographic distance for RAPD and allozyme data sets.

Technique Populations Individuals FST m

1 FDT m

2 r Mantel's P

RAPD 6 96 0.193 1.045 0.162 1.293 0.37116 0.917

Allozymes 7 303 0.021 11.655 0.0262 9.292 0.54204 0.976

1 Gene flow calculated from Weir and Cockerham's FST using RapdFST (Black, 1993).

2 Gene flow calculated from Wright's (1978) FDT using Biosys-2.

The Wagner procedure tree gave a better fit according to the Goodness of fit statistics (Fig 2).

Despite the observed high cophenetic correlation of 0.887, the allozyme cophenetic correlation was

even higher 0.96, suggesting a better agreement between the original matrix and the phenogram than

with RAPD data. DT ranged from 0.232 between the Isle of Man and Eastern Irish Coast populations

to 0.342 between the two North Sea populations, indicating the small degree of genetic divergence

between these populations. Both trees produced a cluster of the geographically proximate Irish Sea

populations, which were related to the East Anglian Coast population. The Bay of Biscay population

was closely related to the German Bight population, in agreement with their clustering position in the

allozyme survey. A Mantel’s test shows that genetic distance was correlated with geographical

distance (r = 0.37; t = 1.385). However, this correlation was not statistically significant; the one-

tailed probability that the observed Z was greater than random Z in 1000 permutations was 0.917.

Correlation of geographic distance with genetic distance was also positive for the allozyme data set

yielding a higher r and t (r = 0.54; t = 1.978), but was also statistically non-significant (P = 0.976).



7

4. DISCUSSIO�

The chi-square contingency analyses (allele frequencies and hierarchical FST and non-

hierarchical FDT) indicated a highly significant heterogeneity among populations and between

groupings; results from both techniques. This conclusion is consistent with the life history of Dover

sole, the homing behaviour they exhibit (e.g. Rijnsdorp et al., 1992) and the discrete spawning areas

(e.g. Symonds and Rogers, 1995). These characteristics serve to reduce the homogenisation of

population allele frequencies between spawning stocks. A northwest to southeast significant variation

replaced a north to south pattern of significant population differentiation in the allozyme study.

Natural selection for differences in spawning habitat may explain the genetic isolation between these

groupings. Another reason could be the significant absence of some alleles, present in the southern

European populations. Therefore, the progressive decline in allozyme allelic diversity northwards

was interpreted as a possible consequence of a population bottleneck, founder effects, genetic drift, or

a combination of these. This interpretation is supported by the finding that populations sampled in

southern Europe are more variable than northern European populations; a finding consistent with the

general tendency for terrestrial populations to exhibit reduced genetic variability in areas glaciated

during the Pleistocene (e.g. Highton and Webster, 1976; Bellemin et al., 1978; Vaisanen and

Lehvasliho, 1984; Singh and Rhomberg, 1987). Such a stepping-stone pattern, that led to a loss of

genetic diversity even northern (Irish Sea) could also have been involved with colonisation of this

0 0.02 0.04 0.07 0.09 0.11 0.13

FRA

GER

IRL

EAN

CUM

IOM

Goodness of fit statistics: Farris (1972) “f” = 0.148

Prager and Wilson (1976) “F” = 4.426

% standard deviation (Fitch and Margoliash, 1976) = 5.978

Cophenetic correlation = 0.887

Figure 1: Wagner procedure demdogram using Prevosti’s Distance (Wright, 1978) and Goodness of fit statistics

between the six Atlantic Solea solea populations.



8

area by S. solea. This observed disjunction between the Irish Sea and the European Coast populations

indicates reduced exchange of migrants between these regions, in agreement with the conclusions in

the allozyme study.

Evidence in favour of the divergence between these regions is shown by the cluster analyses

(UPGMA, Wagner procedure) of both allozyme and RAPD data. However, allozymes provided a

slightly better fit of data to predictions about genetic distance (higher values of cophenetic correlation

of clusters from the Goodness of fit statistics) and better correlation between genetic and geographic

distances (Mantel’s r). In this regard, allozymes remain valuable for many types of population studies

despite the development of new genetic markers (Schaal et al., 1991). In both studies a strong

similarity between the Bay of Biscay and the German Bight populations was evident, judging from

their clustering positions. Cluster analyses also suggested a similarity between the three Irish Sea and

the East Anglian populations. It was apparent that the trend of probable movement of migrants

through the English Channel suggested by Exadactylos et al. (1998) was supported using RAPD

markers. Tagging experiments carried out by Greer-Walker and Emerson (1990) reported a seasonal

migration of soles through the Dover strait on both north-going and south-going tides. The apparent

absence of isolation-by-distance (non significant correlation of genetic with geographic distance) in

both studies suggests that S. solea may not be at genetic equilibrium (Slatkin, 1993). If populations

have not reached an equilibrium the F-statistics will underestimate the degree of differentiation

expected at equilibrium and the estimates of gene flow will overestimate the true levels (Slatkin,

1985). Estimated �m values of approximately one migrant into the average deme per generation

indicates gene flow among populations sufficient to deter population differentiation, if populations are

currently at an equilibrium between migration and genetic drift.

Several factors may explain these trends. Population fragmentation for S. solea may have been

a relatively recent event. This result is in agreement with the hypothesis reported in Exadactylos and

Thorpe (2001) that the family Soleidae and within it Solea is the most recently evolved group among

North Atlantic flatfishes (see also Lauder and Liem, 1983). Perhaps S. solea population

fragmentation has occurred sporadically throughout the species range, thus, there are no large

geographic areas that have become isolated. Furthermore, dispersal may be more restricted by

specific habitat types, rather than by distance to the next closest population. Genetic variability

depends on whether the species is a habitat specialist or a habitat generalist (Smith and Fujio, 1982).



9

Hence, adaptation for local selection pressures are possible. White (1978) proposed that effective

population size (effective number of individuals contributing to the next generation) may be related to

the extent of geographic habitat in a species undergoing substantial gene flow between its local

populations, but it can also reflect the ability of a species to occupy broad ecological habitats. S.

solea conforms to the type of species with either a broad variety of geographic and ecological

habitats, or large and stable population sizes (Kotoulas et al., 1995), or substantial gene flow at a

regional scale. Thus, FST / FDT results may be artificially significantly high from relatively poor

estimates of within-population variation. Resulting gene flow values from such FST estimates should

thus be interpreted with great caution, although they may serve as a useful relative measure with

which to compare between populations (Larcson et al., 1989). Overall, on the basis of the RAPD loci

examined, the population differentiation observed in the Atlantic Ocean is likely to be primarily the

result of random genetic drift or differential selection. Alternative means of detecting polymorphism

in S. solea including the use of single-locus minisatellite probes and mtDNA analysis may help to

further clarify the picture.

5. REFERE�CES

Bardakci, F., Skibinski, D.O.F., 1994. Application of the RAPD technique in tilapia fish: species and

subspecies identification.- Heredity 73: 117-123

Bellemin, J., Adest, G., Gorman, G. C., 1978. Genetic uniformity in northern populations of

Thamnopsis sirtalis (Serpentes: Colubridae).- Copeia 1978: 150-151

Black, B., 1993. RAPDBIOS - A FORTRAN program to convert RAPD-PCR files to BIOSYS

(DATYPE=3) files. Colorado State University, Department of Microbiology, Fort Collins, U.S.A

Black, B., 1995. RAPDFST 3.0 - A FORTRAN program to estimate Fst and effective migration rates

among subpopulations using RAPD-PCR files. Colorado State University, Department of

Microbiology, Fort Collins, U.S.A

Caetano-Anolles, G., Bassam, B.J., Gresshoff, P.M., 1991a. DNA amplification fingerprinting using

very short arbitrary oligonucleotide primers.- Biotechnology 9: 553-556

Caetano-Anolles, G., Bassam, B.J., Gresshoff, P.M., 1991b. DNA amplification fingerprinting: a

strategy for genome analysis.- Plant Mol. Biol. Rep. 9: 294-307

Clark, A.G., Lanigan, C.M.S., 1993. Prospects for estimating nucleotide divergence with RAPDs.- J.

Mol. Biol. Evol. 10: 1096-1111

Exadactylos, A., 1997. Population Genetics of the Dover sole, Solea solea L. (Linneaus, 1758)

(Teleostei: Soleidae). Ph.D. Thesis. University of Liverpool, School of Biological Sciences, Port

Erin Marine Lab.

Exadactylos, A., Geffen, A.J., Thorpe, J.P., 1998. Population structure of the Dover sole, Solea solea

L. in a background of high gene flow.- J. Sea Res. 40: 117-129



10

Exadactylos, A., Thorpe, J.P., 2001. Allozyme variation and genetic inter-relationships between seven

flatfish species of the order Pleuronectiformes. Zool. J. Linn. Soc. 132: 487-499

Farris, J.S., 1972. Estimating phylogenetic trees from distance matrices.- Amer. Nat. 106: 645-668

Fitch, W.M., Margoliash, E., 1967. Construction of phylogenetic trees.- Science 155: 279-284

Greer-Walker, M., Emerson, L., 1990. The seasonal migration of soles (Solea solea) through the

Dover strait.- Netherlands journal of Sea Research 25 (3): 417-422

Grosberg, R.K., Levitan, D.R., Cameron, B.B., 1996. Characterization of genetic structure and

genealogies using RAPD-PCR markers: A random primer for the novice and nervous. In J.D.

Ferraris, S.R. Palumbi. Molecular Zoology: Advances, Strategies and Protocols. Wiley-Liss. Inc.,

California: 67-100

Hadrys, H., Balick, M., Scierwater, B., 1992. Applications of random amplified polymorphic DNA

(RAPD) in molecular ecology.- Mol. Ecol. 1: 55-63

Highton, R., Webster, T.P., 1976. Geographic protein variation and divergence in populations of the

salamander, Plethodon cinereus.- Evolution 30: 33-45

Kotoulas, G., Bonhomme, F., Borsa, P., 1995. Genetic structure of the common sole Solea vulgaris at

different geographical scales.- Mar. Biol. 122: 361-375

Larcson, J.M., Riccardi, V.M., Calhoun, S.W., Morizot, D.C., 1989. Genetic differentiation of

bicolour damselfish (Eupomacentrus partitus) populations in the Florida Keys. - Mar. Biol. 103:

445-451

Lauder, G.V., Liem, K.F., 1983. The evolution and interrelationships of the actinopterygian fishes.-

Bulletin of the Museum of Comparative Zoology 150: 95-197

Lynch, M., Milligan, B.G., 1994. Analysis of population genetic structure with RAPD markers.-

Molecular Ecology 3: 91-99

Mantel, N., 1967. The detection of disease clustering and a generalised regression approach.- Cancer

Research 27: 209-220

Nei, M., 1987. Molecular Evolutionary Genetics. Columbia University Press, New York

Prager, E.M., Wilson, A.C., 1976. Congruency of phylogenies derived from different proteins.- J.

Mol. Evol. 9: 45-57

Rijnsdorp, A.D., Van Beek, F.A., Flatman, S., Millner, R.M., Riley, J.D., Giret, M., De Clerck, R.,

1992. Recruitment of sole stocks Solea solea L. in the Northeast Atlantic.- Netherlands Journal of

Sea Research 29 (1-3): 1173-192

Rohlf, F.J., 1993. NTSYS-pc Numerical taxonomy and multivariate analysis system, Version 1.80,

Applied Biostatistics, Setauket, NY

Schaal, B.A., O’Kane-Jr, S.L., Rogstad, S.H., 1991. DNA variation in plant populations.- Trends in

Ecology and Evolution 6: 329-336

Singh,R.S., Rhomberg, L.R., 1987. A comprehensive study of genic variation in natural populations

of Drosophila melanogaster. II. Estimates of heterozygosity and patterns of geographic variation.

Genetics 117: 255-271

Slatkin, M., 1985. Gene flow in natural populations.- A. Rev. Ecol. Syst. 16: 393-430



11

Slatkin, M., 1993. Isolation by distance in equilibrium and non-equilibrium populations.- Evolution

47: 264-279

Smith, P.J., Fujio, Y., 1982. Genetic variation in marine teleosts: high variability in habitat specialists

and low variability in habitat generalists.- Mar. Biol. 69: 7-20

Sneath, P.H.A., Sokal, R.R., 1973. Numerical Taxonomy. Freeman & Co., San Francisco

Stewart, C.N., Excoffier, L., 1996. Assessing population genetic structure and variability with RAPD

data: Application to Vaccinium macrocarpon (American Cranberry).- J. Evol. Biol. 9: 153-171

Swofford, D.L., Selander, R.B., 1989. Biosys-2: A computer program for the analysis of allelic

variation in population genetics and biochemical systematics. Champaign, Illinois 61801 USA,

D.L. Swofford, Illinois Natural History Survey

Symonds, D.J., Rogers, S.I., 1995. The influence of spawning and nursery grounds on the distribution

of sole, Solea solea (L.) in the Irish Sea, Bristol Channel and adjacent areas. - Journal of

experimental Marine Biology and Ecology 190: 243-261

Vaisanen, R.A., Lehvasliho, H., 1984. Absence of genic polymorphism in a northern population of the

house sparrow, Passer domesticus. Hereditas 100: 161-162

Weir, B.S., Cockerham, C.C., 1984. Estimating F-statistics for the analysis of population structure.-

Evolution 38 (6): 1358-1370

Welsh, J., McClelland, M., 1990. Fingerprinting genomes using PCR with arbitrary primers.- Nucleic

Acids Research 18: 7213-7218

Welsh, J., McClelland, M., 1991. Genomic fingerprinting using arbitrarily primed PCR and a matrix

of pairwise combinations or primers.- Nucleic Acids Research 19: 5275-5279

White, M, J.D., 1978. Models of speciation. W.H. Freeman and Co., San Francisco

Williams, J.G.K., Kubelik, A.R., Livak, K.J., Rafalski, J.A., Tingrey, S.V., 1990. DNA

polymorphisms amplified by arbitrary primers are useful as genetic markers.- Nucleic Acids

Research 18: 6531-6535

Wright, S., 1978. Variability within and among natural populations. Evolution and the Genetics of

populations. Vol. 4. University of Chicago Press, Chicago 580pp

Population structure of the Dover sole, Solea solea (L) in the Atlantic using randomly amplified polymorphic DNA (PCR-RAPD) marking

Documents