Journal of Heredity Advance Access originally published online on January 11, 2006
Journal of Heredity 2006 97(1):81-88; doi:10.1093/jhered/esj005
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Brief Communication |
Genetic Clines in the Bay of Biscay Provide Estimates of Migration for Sardina pilchardus
From the Ecole Pratique des Hautes Etudes—UMR CNRS 8046, Laboratoire d'Ichtyoécologie Tropicale et Mediterranéenne, Université de Perpignan, 52, Avenue Paul Alduy, 66860 Perpignan cedex, France
Address correspondence to Dr. Serge Planes at the address above, or e-mail: planes{at}univ-perp.fr.
 |
Abstract |
|---|
 Top Abstract Materials and MethodsResultsDiscussionReferences |
|---|
Nine allozymic loci in 1,635 individuals of Sardina pilchardus obtained at 33 sites ranging from the North to the South limits of the Bay of Biscay were analyzed to provide a description of the genetic structure of the sardine population. Individual body size and age were also recorded and analyzed. In the study population, weak but significant genetic differences were found, and a cline was observed between multilocus heterozygosity and longitude. The cline was predominantly driven by allelic frequencies of two loci, PGM-1* and PEP-lt*, and using a cline model, we estimated a migration rate of 103.1 km/gen (dispersal distance per generation). In addition, we observed that the cline was linked to biological data such as mean length and mean age of the fish. Two hypotheses may explain this cline: mixing of two different populations in the Bay of Biscay or a selective process. The weak genetic differences, the important dispersal distance per generation, and the link between genetic and biological data suggest that selection is likely to be the primary factor that maintains the cline.
Clines have long been of interest to evolutionary biologists because they often represent the interface between intra- and interspecific processes and thus provide an insight into divergence and speciation (Avise 2001; Gardner 1997). Latitudinal genetic or morphometric clines could arise as a result of the migration history or a selection gradient. Because the width and shape of genetic clines commonly represent an evolutionary balance between selection and dispersal, clines provide researchers with the opportunity to analyze both evolutionary forces simultaneously (Sotka and Palumbi, in press). In the context of the balance between dispersal and some form of selection, various cline forms can be distinguished (Barton and Hewitt 1985). "Exogenous" selection is adaptation driven by the local environment. In contrast, "endogenous" selection is driven by genetic factors; that is, the relative fitness of individuals across the cline is independent of the environment (Moore and Price 1993). Whether the dominance of either process leads to different cline spatial patterns remains controversial.
The selection process can greatly modify different markers and/or population characteristics such as heterozygosity (Gaffney 1990; Li 1978; Thelen and Allendorf 2001), allelic frequencies (Li 1978; Mallet and Barton 1989), or number of alleles maintained in the population (Li 1978; Slatkin and Muirhead 1999). The effect of selection on heterozygosity (e.g., overdominance, heterozygote superiority) may be observed as a positive relationship between heterozygosity and fitness measures such as growth rate (Koehn 1990; Planes and Romans 2004; Zouros et al. 1988), fecundity, or survival (Britten 1996; Thelen and Allendorf 2001) of individuals. However, the relationship between heterozygosity and fitness is not well understood (Theisen 1978; Thelen and Allendorf 2001; Zouros et al. 1988), and the relationship between environmental factors and allelic frequencies may also play a role in selection. In some marine species, genetic clines between populations have been observed recently, and these examples illustrate the key role of selection (Planes and Doherty 1997; Riginos and Cunningham 2005), hybridization, or mixing of isolated populations (Koehn 1980; Turgeon and Bernatchez 2001).
Small pelagic species (such as the sardines, Sardina pilchardus) with potentially large population sizes, shoaling behavior, and extensive migration are likely to demonstrate genetic homogeneity over a large spatial scale. In such cases, genetic drift is limited because of the potentially large effective population size, and gene flow is favored over the lifetime of the individuals as a consequence of shoaling behavior and migration. Many studies have investigated the genetic structure of fish populations, and the structure of pelagic species are among the more difficult to determine (Begg et al. 1998; Chikhi et al. 1998; Kinsey et al. 1994). However, despite fluid gene flow, some genetic differences may be observed in pelagic species and may persist as a result of natural selection (Jordan et al. 1997; Mork et al. 1985).
Attempts to differentiate sardine populations based on meristic variability date back to the 1920s (Andreu 1969). Since the review by Parrish et al. (1989), four stocks of S. pilchardus have been identified, based on meristic data, within its distribution: a septentrional Atlantic stock, distributed from the North Sea (57°N) to the Cantabric Coast of Spain (43°N); an Iberian stock, distributed from the Cantabric Coast to the Strait of Gibraltar (36°N); a Moroccan stock, distributed from Cap Spartel (36°N) to Cap Juby (28°N); and a Saharian stock, distributed from Cap Juby to Levrier Bay (21°N). However, failure to demonstrate persistent and significant phenotypic differences between individuals in these areas has led to disagreement concerning the population structure of the species.
The Bay of Biscay has been an area of confrontation between French and Spanish fishermen catching S. pilchardus in the last decade. A common perception is that each country holds a different stock, and each one should fish in their own region. Oceanographically, the two regions of the bay are sustained by the same source of water, and sardines in this area are likely to constitute a single stock. However, some local variations in salinity, temperature, and hydrology have been reported (Puillat et al. 2004) that may drive presence of genetic differences between fish populations.
In this study, we analyzed the genetic structure of the S. pilchardus population in the Bay of Biscay and investigated the relationship between geographic factors and genetic indices in an effort to identify genetic clines. Finding such clines should permit us to estimate dispersal rate, an important factor for stock management.
|
Materials and Methods |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
Sampling
A total of 33 samples of S. pilchardus (n = 1,635) were collected in the Bay of Biscay during the PELGAS 02 cruise, from May 6 to June 10, 2002, on the RV "Thalassa" (Table 1; Figure 1). This cruise was initially designed to assess small pelagic fish stocks during the spawning season using acoustic interpolation. Pelagic trawls were deployed to calibrate echo and to identify species. In each of the trawls containing S. pilchardus, we randomly collected 50 individuals. Samples were then stored on board at –30°C until dissection in the laboratory.
|
|
Biological and Geographic Data
Total length and weight were initially recorded for each specimen upon thawing. Each sardine was cut transversely behind the eyes, and the two sagittae were removed. Sagittae were cleaned and put in water to determine the age of the individuals using a binocular microscope (magnification x12) as recommended by Anonyme (2001) and ICES (1997).
The biological data (total length and age) and geographic and environmental data were first analyzed using principal components analysis to describe the repartition of the sardines in the Bay of Biscay. We used this analysis to determine which factor was most important for explaining the data variability and to identify correlations between factors. In the analysis, we considered mean length and mean age for each individual trawl sample. In addition, we computed correlations between geographic and biological data.
Genetic Analysis
Each sardine was dissected to isolate the liver and a portion of muscle. Each tissue sample was homogenized at 4°C in an equal volume of tris(hydrosymethyl) aminomethane/ethylenediaminetetraacetic/nicotinamide adenine dinucleotide phosphate (Tris/EDTA/NADP) buffer (pH 6.8). Homogenates were centrifuged at 15,000 g for 30 min at 4°C, and the supernatants were stored at –80°C. Samples were then processed by horizontal starch gel electrophoresis following the technique of Pasteur et al. (1987) using four buffers (Tris citrate 6.7, Tris citrate 8.0, Tris borate EDTA 8.6, and Tris glutamate 9.0). Twenty-seven loci were clearly scored, and the enzyme nomenclature was determined according to Shaklee et al. (1990). Further details are provided in Laurent et al. (in press).
The genotypic and allelic frequencies for polymorphic loci were obtained by determining the phenotypes from the gels and using the "GENETIX 4.05" software (Bonhomme et al. 1993; http://www.univ-montp2.fr/
genetix/genetix/genetix.htm). Observed heterozygosity values were computed for each sample and each locus from the genotypic frequencies. All loci (i.e., 27 loci including the monomorphic and those showing reduced polymorphism) were considered for multilocus analysis. The Fixation index (Fis) was calculated to provide descriptive genetic data for each sample. The multilocus deviation from the Hardy-Weinberg equilibrium was statistically tested using the Markov chain reaction implemented in GENEPOP 3.4 (Raymond and Rousset 1995). Significance levels for statistical tests were adjusted for each population separately according to the sequential method of Bonferroni (Rice 1989). Multiple regressions correlating allelic frequencies, heterozygosity, and Fis with geographic and biological data were calculated to investigate clinal variation. Significance was adjusted for multiple test analyses.
The correlation between geographic distance separating sites and genetic differentiation (multilocus Fst) was performed over all loci for each population using Mantel's test based directly on the distribution of 10,000 randomized matrices computed by permutation (Mantel 1967). Mantel's tests were computed using the MANTEL procedure in the "GENETIX" package. We followed Hutchinson and Templeton's model (1999) that directly links Fst and geographic distances for these correlations.
We computed 
, the dispersal rate per generation needed to produce a linkage disequilibrium in the center of a cline, with 
according to Szymura and Barton (1986). In this estimate, w is the cline width, D is the average linkage disequilibrium in the center of the cline, and r = 1/2 for unlinked loci. Linkage disequilibrium values (D) are constrained by the allele frequencies. D decreases in populations that show a lower number of polymorphisms. Therefore, we used the correlation coefficient R between loci, 
as the linkage disequilibrium value, which lies between –1 and +1 (Szymura and Barton 1986) and we obtained the equation 
Two methods were possible to estimate w, width of the clines. Following Szymura and Barton (1986), the width of the cline was the geographic distance including 20% and 80% of parental genotype. This method implied that we sampled the entire cline. The second method, provided by Barton and Gale (1993), proposes that width is the inverse slope of the cline if allelic frequence is shifted from 0 to 1. If not, the estimate of width is corrected by w = 
p/slope. Linkage disequilibria and correlation coefficients were estimated using the LINKDIS program (Black and Krasfur 1985) implemented in GENETIX (Bonhomme et al. 1993). It was assumed that diploid genotypes were in Hardy-Weinberg proportions to reduce the problem to an estimation of haploid gamete frequencies. Furthermore, the third and higher-order disequilibria were considered to be zero, which reduces the computations to pairwise comparisons among all interpretable loci. Linkage disequilibrium was statistically tested using the Markov chain reaction implemented in GENEPOP 3.4 (Raymond and Rousset 1995) that also gives a variance value to compute interval confidence.
|
Results |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
In this study, heterozygosity and longitude were significantly correlated (r = –.461, P = .0091). Overall, heterozygosity increased the farther west the samples were isolated. Within the nine highly polymorphic loci, PGM-1* and 6PGD*, each demonstrated a significant increase in heterozygosity with longitude. In these correlations, 30% of the variance was explained by linear regression (Table 2). Detailed analysis of the allelic frequencies of these 2 loci showed a significant change in heterozygosity, but only PGM-1* had a significant change in allelic frequency with longitude and depth. For example, the frequency of Pgm-1*-110 decreased as samples were isolated farther west (r = –.42, P = .018) (Figure 2) and at increasing depth. An opposite trend was observed for Pgm-1*-100.
|
|
The change in heterozygosity for 6PGD* was not related to any change in the allelic frequency through the longitudinal gradient (Table 2). Further investigation of other loci demonstrated that Pep-lt*-110 (evolving on a contrasting trend to Pep-lt*-100) also showed a significant decrease as samples were obtained farther west (r = –.46, P = .009) (Figure 2) and at increasing depth.
Several significant correlations were observed between longitude and topographic and biological data (Table 3). On a topographic view, longitude was significantly correlated with latitude (r = –.7535, P < .0001), with distance from coast (r = –.6593), and with depth (r = .4475, P = .0116). So depth seems to be the main factor explaining allele frequency variations in the cline of PGM-1*. From a biological perspective, longitude showed a significant correlation with mean total length of individuals in each sample (r = –.6135, P = .0002), mean ages of individuals in each sample (r = –.5113, P = .0033), and Fis values (r = .4024, P = .0248). Mean total length of individuals seems to be the major factor in explaining the correlation. Individuals collected within each sample appeared homogeneous with regard to biological parameters such as size and age. The standard error ranged from 0.78 to 2.90 for the length, about 1% of the mean value, and from 0.03 to 0.20 for age, less than 6% of the mean value (Table 1), which indicated a low deviation in the distribution. As expected, many of these parameters showed significant correlations between each other; however, we focused only on the correlations with longitude.
|
Based on the description of the cline for Pgm-1*-110-Pgm-1*-100 frequencies, we estimated the length of the transition zone or width of the cline. According to the Barton and Gale (1993) model, the width of the slope reached 1941 km. All samples were pooled within a single population, and a coefficient of correlation (R) between PGM-1* and PEP-lt* (R = .02256, P = .05) was determined. On the basis of the determined w (1941 km) and R value, we estimated the dispersal rate per generation to be
= 103.1 ± 11.0 km/gen, representing 5.3% of the cline width. Mantel's test found no correlation between genetic differentiation and geographic distance when all the samples in the analysis were considered (r = .071, P = .210).
|
Discussion |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
In this study, analysis of allozymes demonstrated a significant genetic cline in S. pilchardus populations along a continuum of about 1000 km on the French Atlantic coasts. This cline was first detected on heterozygosity values, and further detailed analysis demonstrated that it originates predominantly from the gradual change in allelic frequencies in PGM-1* and PEP-lt*. Despite the existence of the cline, no evidence of genetic structure was observed within the areas because the differences in allelic frequencies were not high enough to result in important Fst values (Fst = 0.005, P = .000), and there was no correlation between geographic distance and Fst. Weak Fst values have been previously reported for pelagic fish (Chikhi et al. 1998; Giaever and Stein 1998; Hedgecock 1994; Lundy et al. 2000; Ward et al. 1994), and the present results for S. pilchardus follow a similar trend. Despite evidence of a gradual change in two loci, the Bay of Biscay appears to be composed of a single panmictic population, similar to the other small pelagic species in the bay (e.g., mackerel and hake) previously examined (Lundy et al. 2000; Nesbo et al. 2000).
The genetic data for S. pilchardus in the Bay of Biscay showed significant genetic clines in two loci. Cline theory derives mainly from the analysis of hybrid zones (Barton and Hewitt 1985) but has been applied in recent case studies to investigate migration and selection within species showing gradual change in allelic frequencies along a geographic transect (Planes and Doherty 1997; Sotka et al. 2004; Sotka and Palumbi, in press). In the present study, the cline in multiloci heterozygosity resulted in fact from changes in allelic frequencies of two alleles, Pgm-1*-110 and Pep-lt*-110. Based on the cline model, the dispersal distance per generation for populations of sardines is 103.1 km/gen, representing 5.3% of the cline width. This proportion corroborates general trends previously calculated in marine species (Sotka and Palumbi, in press). The difficulty in interpreting factors affecting this cline in S. pilchardus was that longitude appeared also correlated with other geographic features such as latitude, coastal distance, and depth. Overall, regression analyses seem to point to depth as the explanation for most of the variance in the PGM-1* cline.
Clinal variation of allelic frequencies in genes underlying a particular trait can result from gene flow between partially isolated populations that have diverged in their genetic composition through drift or can be a result of an admixture between two or more genetically differentiated founding populations (Gallant et al. 1993; Turgeon and Bernatchez 2001). This process may also be described as a nonselective isolation-by-distance model (Gould and Johnson 1972). In the S. pilchardus model, longitude, latitude, distance to the coast, or depth could sustain differentiated populations that are mixing alleles via gene flow along the central area of the cline. Because we did not demonstrate significant genetic structure in the area and Mantel's test refuted an isolation by distance pattern, it seems unlikely that a model of nonselective isolation by distance is applicable in the case of S. pilchardus in the Bay of Biscay.
Clinal variation in genetically based traits can also provide compelling evidence for spatially varying selection along an environmental gradient (Endler 1977; Haldane 1948; Slatkin 1978). In the case of the S. pilchardus, the cline was also correlated with age and length of individuals, suggesting that biological processes may be involved in explaining the present cline in two alleles (Pgm-1*-110 and Pgm-1*-100). These underlying biological constraints support the possibility of selection as being the main factor producing differences in allelic frequencies. Adult and juvenile S. pilchardus do not live in similar habitats, and pelagic fishes are known to have differential distribution as a function of their size and the depth gradient (Smith and Brown 2002). Whereas juveniles usually occur in coastal areas (Yatsu and Kaeriyama 2005), adults tend to migrate from the feeding grounds to spawning areas (Diachok et al. 2001). The Bay of Biscay has been described as a nursery area for sardines as well as many other small pelagic species showing extensive migration behaviors (Parrish et al. 1989). In the Bay of Biscay, we propose that eggs and larvae mainly originate from near-coastal regions. While the individuals are growing, they tend to increase their foraging area westwardly. This would result in an age-size gradient mainly related to depth. In such a behavior model, sardines crossing different habitats during their life span could drive selective processes mostly observed in older individuals. Presently, the increase of heterozygosity with size likely results from the removal of homozygotes in the samples providing an example of overdominance (Kimura 1983).
Numerous studies have investigated the relationship between multilocus heterozygosity and fitness components (Britten 1996; Koehn et al. 1988; Thelen and Allendorf 2001; Zouros 1990), and growth has been shown to be significantly correlated with multilocus heterozygosity in few cases (Exadactylos et al. 1999; Gaffney 1990; Koehn et al. 1988). In S. pilchardus, the cline is mostly significant in PGM-1*, and some studies have shown that this locus can influence individual growth (Allendorf 1983; Planes and Romans 2004). However, in the present study, there were no differences in growth among samples, and differences in size resulted only from differences in age (Laurent V, unpublished data), providing little evidence for growth selection. When investigating the potential for selective pressure, we should also consider that the sardine is a species that is subject to an important fishing industry, especially nearshore. We cannot eliminate the possibility that this fishing pressure may modify the structure of the population, especially in the older year classes, even if the usual expectation is a reduction of heterozygosity (Smith et al. 1991). Moreover, the abundance of sardines in the bay fluctuates over the years and was reported to be important in 2002 (biomass evaluated to 301,023 t) (ICES 2005). So, sardines, like other pelagic fishes, expand their range in function of their abundance (Alheit and Hagen 1997) and therefore the repartition of the fish was related with their age and with older individuals surviving in deeper waters than younger fish (Yatsu and Kaeriyama 2005). It is possible that the cline mostly reflected this differentiated repartition of the individuals and not a selective process.
Finally, our assumptions may influence the validity of the estimation of dispersal rate. We used a cline model based on the admixture of two different populations, although the actual pattern appears more likely to be driven by selective processes. Despite this limitation, the values we obtained provide additional information about the distribution of sardines in the Bay of Biscay. Previous studies have used this approach to estimate dispersal rate. Turgeon and Bernatchez (2001) computed a dispersal rate of 31 km for Coregonus artedi; Sotka and Palumbi (in press) obtained 54–76 km for Crossastrea virginica and 67 km for Balanus glandula. We observed that the dispersal rate for sardines is larger than that for mollusks and freshwater fish, almost certainly because both larvae and adult fish contribute to dispersal. Because migration and selection are inversely related in cline modeling (Slatkin 1973) and because the dispersal rate is quite important for sardines, selection is likely weak on the populations, which raises some questions about cline maintenance. Overall, it is difficult to reach conclusions regarding the process driving the cline observed throughout the Bay of Biscay for S. pilchardus. However, the use of the cline theory helped provide a dispersal estimate (about 103 km/gen) that could not have been computed from the usual analysis of genetic structure. The dispersal seems to vary according to age, with the oldest individuals being the most mobile and so, this estimation is likely a mean value of the entire population.
|
Acknowledgments |
|---|
We thank the anonymous reviewers for their helpful comments and V. Messmer for helping with the English and the review. We thank J. Massé and P. Petitgas for their collaboration in the collection of samples and P. Lenfant for his help in the data analyses. This work was part of European SARDYN program (Q5RS-2002-000818). The sampling was achieved during the PELGAS cruise (IFREMER) with the RV "Thalassa." We would like to thank Dr. Paul Kretchmer (kretchmer{at}sfedit.net) at San Francisco Edit for his assistance in editing this manuscript.
|
Footnotes |
|---|
Corresponding Editor: Martin Tracey
Received April 18, 2005
Accepted October 11, 2005
|
References |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
-
Alheit J and Hagen E, 1997. Long-term climate forcing of European herring and sardine populations. Fish Oceanogr 6(2):130–139.
Allendorf FW, 1983. Isolation, gene flow, and genetic differentiation among populations. In: Genetics and conservation (Schonewald-Cox CM, Chambers SM, MacBryde B, and Thomas WL, eds). Menlo Park, CA: Benjamin/Cummings; 51–65.
Andreu B, 1969. Las branchiespinas en la caracterizacion de las poblaciones de Sardina pilchardus. Investigation y Pesqueria Barcelona 33:425–607.
Anonyme, 2001. Report of the Sardine (Sardina pilchardus) otolith workshop. FAO Fisheries Report, 685. 49 p.
Avise JC, 2001. Phylogeography: the history and formation of species. Cambridge, MA: Harvard University Press.
Barton NH and Gale KS, 1993. Genetic analysis of hybrid zones. In: Hybrid zones and the evolutionary process (Harrison R, ed.). New York: Oxford University Press; 13–45.
Barton NH and Hewitt GM, 1985. Analysis of hybrid zones. Annu Rev Ecol Syst 16:113–148.[CrossRef][Web of Science]
Begg GA, Keenan CP, and Sellin MJ, 1998. Genetic variation and stock structure of school mackerel and spotted mackerel in northern Australian waters. J Fish Biol 53:543–559.
Black WC and Krasfur ES, 1985. LINKDIS software. Iowa City, IA: Department of Entomology, University of Iowa.
Bonhomme F, Belkhir K, Borsa P, Mathieu E, and Roux M, 1993. Genetix: Logiciel d'analyse des données du groupe de génétique des populations de Montpellier. Version 0.1. Montpellier, France: Université de Montpellier II.
Britten HB, 1996. Meta-analysis of the association of multilocus heterozygosity and fitness. Evolution 50:2158–2164.[CrossRef][Web of Science]
Chikhi L, Bonhomme F, and Agnese JF, 1998. Low genetic variability in a widely distributed and abundant clupeid species, Sardinella aurita. New empirical results and interpretations. J Fish Biol 52:861–878.
Diachok O, Liorzou B, and Scalabrin C, 2001. Estimation of the number density of fish from resonance absorptivity and echo sounder data. ICES J Mar Sci 58:137–153.
Endler JA, 1977. Geographic variation, speciation and clines. Princeton, NJ: Princeton University Press.
Exadactylos A, Geffen AJ, and Thorpe JP, 1999. Growth and genetic variation in hatchery-reared larval and juvenile Dover sole, Solea solea (L.). Aquaculture 176:209–226.[CrossRef][Web of Science]
Gaffney PM, 1990. Enzyme heterozygosity, growth rate and viability in Mytilus edulis: another look. Evolution 40:204–210.
Gallant SL, Preziosi RF, and Fairbairn DJ, 1993. Clinal variation in eastern populations of the waterstrider Aquarius remigis: gradual intergradation or discontinuity? Evolution 47:957–964.[CrossRef]
Gardner JPA, 1997. Hybridization in the sea. Adv Mar Biol 31:1–78.
Giaever M and Stein J, 1998. Population genetic substructure in blue whiting based on allozyme data. J Fish Biol 52:782–795.
Gould SJ and Johnson RF, 1972. Geographic variation. Annu Rev Ecol Syst 3:457–498.[CrossRef]
Haldane JBS, 1948. The theory of a cline. J Genet 48:277–284.[Medline]
Hedgecock D, 1994. Does variance in reproductive success limit effective population sizes of marine organisms? In: Genetic and evolution of aquatic organisms (Beaumont AR, ed.). London, UK: Chapman & Hall; 122–134.
Hutchinson DW and Templeton AR, 1999. Correlation of pairwise genetic and geographic distance measures: inferring the relative influences of gene flow and drift on the distribution of genetic variability. Evolution 53(6):1898–1914.[CrossRef][Web of Science]
ICES 1997. Report of the workshop on sardine otolith age reading. ICES CM 1997. H:7:49.
ICES 2005. Report of the working group on the assessment of mackerel, horse mackerel, sardine and anchovy. ICES, September 7–16, 2004.
Jordan WC, Verspoor E, and Youngson F, 1997. The effect of natural selection on estimates of genetic divergence among populations of the Atlantic salmon. J Fish Biol 51:546–560.
Kimura M, 1983. The neutral theory of molecular evolution. Cambridge: Cambridge University Press.
Kinsey ST, Orsoy T, Bert TM, and Mahmoudi B, 1994. Population structure of the Spanish sardine Sardinella aurita: natural morphological variation in a genetically homogeneous population. Marine Biol 118:309–317.[CrossRef]
Koehn RK, 1980. Salinity related physiological and genetic differences between populations of Mytilus edulis. Biol J Linn Soc 14:319–334.
Koehn RK, 1990. Heterozygosity and growth rate in marine bivalves. Comments on the paper of Zouros, Romero-Dorey and Mallet. Evolution 44:213–216.[CrossRef]
Koehn RK, Diehl WJ, and Scott TM, 1988. The differential contribution by individual enzymes of glycolysis and protein catabolism to the relationship between heterozygosity and growth rate in the coot clam, Mulinia lateralis. Genetics 118:121–130.
Laurent V, Voisin M, and Planes S, in press. Lack of genetic structure among subpopulations of sardines, Sardina pilchardus, in the Bay of Biscay. Aquat Living Resour.
Li WH, 1978. Maintenance of genetic variability under the joint effect under selection, mutation and random drift. Genetics 90:349–382.
Lundy CJ, Rico C, and Hewitt GM, 2000. Temporal and spatial genetic variation in spawning grounds of European hake (Merluccius merluccius) in the Bay of Biscay. Mol Ecol 9:2067–2079.[CrossRef][Medline]
Mallet J and Barton N, 1989. Inference from clines stabilized by frequency-dependant selection. Genetics 122:967–976.
Mantel N, 1967. The detection of disease clustering and a generalized regression approach. Cancer Res 27:209–220.
Moore WS and Price JT, 1993. Nature of selection in the northern flicker hybrid zone and its implications for speciation theory. In: Hybrid zone and the evolutionary process (Harrison R, ed). New York: Oxford University Press; 196–225.
Mork J, Ryman N, Stahl G, Utter F, and Sundnes G, 1985. Genetic variation in Atlantic cod (Gadus morhua) throughout its range. Can J Fish Aquat Sci 42:1580–1587.
Nesbo CL, Rueness EK, Iversen SA, Skagen DW, and Jakobsen KS, 2000. Phylogeography and population history of Atlantic mackerel (Scomber scombrus L.): a genealogical approach reveals genetic structuring among the eastern Atlantic stocks. Proc R Soc Lond B 267:281–292.[Medline]
Parrish RH, Serra R, and Grant WS, 1989. The monotypic sardines, sardina and sardinops: their taxonomy, distribution, stock structure and zoogeography. Can J Fish Aquat Sci 46:2019–2036.
Pasteur N, Pasteur G, Bonhomme F, Catalan J, and Britton-Davidian J, 1987. Manuel de génétique par électrophorèses des protéines. Paris, France: Collection techniques et documentation, Lavoisier.
Planes S and Doherty PJ, 1997. Genetic and color interactions at a contact zone of Acanthochromis polyacanthus: a marine fish lacking pelagic larvae. Evolution 51(4):1232–1243.[CrossRef][Web of Science]
Planes S and Romans P, 2004. Evidence of genetic selection for growth in new recruits of a marine fish. Mol Ecol 13:2049–2060.[CrossRef][Medline]
Puillat I, Lazure P, Jegou AM, Lampert L, and Miller PI, 2004. Hydrographical variability on the French continental shelf in the Bay of Biscay, during de 1990s. Continental Shelf Res 24:1143–1163.[CrossRef]
Raymond M and Rousset F, 1995. GENEPOP (version 1.2): population genetics software for exact tests and ecumenicism. J Hered 86:248–249.
Rice WR, 1989. Analyzing tables of statistical tests. Evolution 43:223–225.[CrossRef][Web of Science]
Riginos C and Cunningham CW, 2005. Local adaptation and species segregation in two mussels (Mytilus edulis * Mytilus trossulus) hybrid zones. Mol Ecol 14:381–400.[CrossRef][Medline]
Shaklee JB, Allendorf FW, Morizot DC, and Whitt GS, 1990. Gene nomenclature for protein-coding loci in fish. Trans Am Fish Soc 119:2–15.[CrossRef]
Slatkin M, 1973. Gene flow and selection in a cline. Genetics 75:733–756.
Slatkin M, 1978. Spatial patterns in the distribution of polygenic characters. J Theor Biol 70:213–228.[CrossRef][Web of Science][Medline]
Slatkin M and Muirhead CA, 1999. Overdominant allele in a population of variable size. Genetics 152:775–781.
Smith KF and Brown JH, 2002. Patterns of diversity, depth range and body size among pelagic fishes along a gradient of depth. Global Ecol Biogeogr 11:313–322.[CrossRef]
Smith PJ, Francis RICC, and McVeagh M, 1991. Loss of genetic diversity due to fishing pressure. Fish Res 10:309–316.[CrossRef]
Sotka EE and Palumbi SR, in press. Genetic clines and dispersal distance of marine larvae. Ecology.
Sotka EE, Wares JP, Barth JA, Grosberg RK, and Palumbi SR, 2004. Strong genetic clines and geographical variation in gene flow in the rocky intertidal barnacle Balanus glandula. Mol Ecol 13:2143–2156.[CrossRef][Medline]
Szymura JM and Barton NH, 1986. Genetic analysis of a hybrid zone between the fire-bellied toads, Bombina bombina and B. variegata, near Cracow in Southern Poland. Evolution 40:1141–1159.[CrossRef][Web of Science]
Theisen BF, 1978. Allozyme clines and evidence of strong selection in three loci in Mytilus edulis L. (Bivalvia) from Danish waters. Ophelia 17:135–142.
Thelen GC and Allendorf FW, 2001. Heterozygosity-fitness correlations in rainbow trout: effects of allozyme loci or associative overdominance? Evolution 55(6):1180–1187.[CrossRef][Web of Science][Medline]
Turgeon J and Bernatchez L, 2001. Clinal variation at microsatellite loci reveals historical secondary intergradation between glacial races of Coregonus artedi (Teleostei: coregoninae). Evolution 55:2274–2286.[CrossRef][Web of Science][Medline]
Ward RD, Woodwark M, and Skibinski DOF, 1994. A comparison of genetic diversity levels in marine, freshwater and anadromous fishes. J Fish Biol 44:213–232.[CrossRef]
Yatsu A and Kaeriyama M, 2005. Linkages between coastal and open-ocean habitats and dynamics of Japanese stocks of chum salmon and Japanese sardine. Deep Sea Res II 52:727–737.[CrossRef]
Zouros E, 1990. Heterozygosity and growth rate in marine bivalves: response to Koehn's remarks. Evolution 44:216–218.[CrossRef]
Zouros E, Romero-Dorey M, and Mallet AL, 1988. Heterozygosity and growth rate in marine bivalves: further data and possible explanations. Evolution 42:1332–1341.[CrossRef][Web of Science]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  M. Chlaida, V. Laurent, S. Kifani, T. Benazzou, H. Jaziri, and S. Planes Evidence of a genetic cline for Sardina pilchardus along the Northwest African coast ICES J. Mar. Sci., March 1, 2009; 66(2): 264 - 271. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. Sanz, J.-L. Garcia-Marin, J. Vinas, M. Roldan, and C. Pla Spawning groups of European anchovy: population structure and management implications ICES J. Mar. Sci., December 1, 2008; 65(9): 1635 - 1644. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||