SEARCH

SEARCH BY CITATION

Keywords:

  • Fst;
  • gene flow;
  • gill-rakers;
  • microsatellites;
  • niche;
  • polymorphism;
  • Qst;
  • speciation

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Material and methods
  5. Study area and sampling
  6. Morphology
  7. Life-history
  8. Genetic markers
  9. Data analysis
  10. Results
  11. Morphological divergence of populations in Lake Femund
  12. Life-history divergence of populations in Lake Femund
  13. Genetic differentiation of populations in Lake Femund
  14. Comparison of morphological and genetic divergence in Lake Femund
  15. Morphological relationships of Lake Femund vs. nearby lakes
  16. Genetic relationships of Lake Femund vs. nearby lakes
  17. Discussion
  18. Trait differentiation in Lake Femund compared with other coregonid systems
  19. Are phenotypic traits in Lake Femund morphs associated with use of niches?
  20. Sympatric origin of the divergent morphs in Lake Femund?
  21. Could traits separating Lake Femund morphs be altered by natural selection?
  22. Conclusion
  23. Acknowledgments
  24. References
  25. Appendices

Combining morphological and genetic analysis, we compared patterns of diversification within and between morphs among sympatric European whitefish (Coregonus lavaretus L.) populations in Lake Femund, Norway. Seven external populations, from potential colonization routes into Lake Femund were included. We found that deep-, shallow-, river- and bay spawning populations are distinct morphs in Lake Femund. Within morphs, populations range from being similar genetically (Fst = 0–0.005) among deep-spawning populations to being highly differentiated (Fst = 0.153) between bay-spawning populations. Between morphs, genetic differences ranged from a low (Fst = 0.008–0.022) between deep- and shallow-spawning populations to high difference (Fst = 0.125–0.143) between shallow- and bay-spawning populations. A higher proportion of molecular variance was seen among (3.9%) than within morphs (2.8%). The adaptive gene combinations behind the four morphs seem to have originated within the lake, although the lake could have been colonized from more than one source population.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Material and methods
  5. Study area and sampling
  6. Morphology
  7. Life-history
  8. Genetic markers
  9. Data analysis
  10. Results
  11. Morphological divergence of populations in Lake Femund
  12. Life-history divergence of populations in Lake Femund
  13. Genetic differentiation of populations in Lake Femund
  14. Comparison of morphological and genetic divergence in Lake Femund
  15. Morphological relationships of Lake Femund vs. nearby lakes
  16. Genetic relationships of Lake Femund vs. nearby lakes
  17. Discussion
  18. Trait differentiation in Lake Femund compared with other coregonid systems
  19. Are phenotypic traits in Lake Femund morphs associated with use of niches?
  20. Sympatric origin of the divergent morphs in Lake Femund?
  21. Could traits separating Lake Femund morphs be altered by natural selection?
  22. Conclusion
  23. Acknowledgments
  24. References
  25. Appendices

Understanding the mechanisms of species formation is a central theme in evolutionary biology, partitioning the roles of historical contingency and deterministic interactions of evolutionary processes (Losos et al., 1998; Taylor & McPhail, 2000; Mayr, 2001; Turelli et al., 2001; Hendry et al., 2002). Species complexes within the genera Coregonus, Gasterosteus, Oncorhynchus, Osmerus, Prosopium, Salmo, Salvelinus and Lepomis frequently display intra-lacustrine diversification in morphology, life history, habitat and diet (Svärdson, 1979; Robinson & Wilson, 1994; Taylor, 1999; Jonsson & Jonsson, 2001). For some of these species complexes, both scenarios of allopatric and sympatric speciation have been proposed (Pigeon et al., 1997; Taylor, 1999; Taylor & McPhail, 2000). The occurrence of such sympatric populations represents a challenge to taxonomists, population geneticists and management authorities as it may be imperative to sort out the current status of populations and their evolutionary origin prior to implementation of conservation plans (Bernatchez, 1995; Nielsen & Powers, 1995; Schluter, 2000, 2004).

The European whitefish species complex (Coregonus lavaretus L.) illustrates the full range of problems associated with the interpretation of morphological diversity. Here, the traditional species designations suggested that numerous whitefish species inhabit European waters (Linnaeus, 1758; Steinmann, 1950a, b, 1951; Svärdson, 1957, 1998; Berg, 1962; Resethnikov, 1968; Himberg & Lehtonen, 1995; Kottelat, 1997). However, the phenotypic species classification poses two fundamental problems (see Felsenstein & Sober, 1986; Humphries & Parenti, 1999). First, a classification may be limited by the available number of traits to assess the phylogenetic relationship between any closely related taxa, and in addition problems arise when assessing the relative importance of these traits. Secondly, comparisons based on traits that are subjected to natural selection could erroneously result in the clustering into a monophyletic group when parallelism is more likely. Alternatively, the use of molecular markers such as microsatellite DNA could provide an unbiased record of recent population divergence, while maternally inherited mitochondrial DNA (mtDNA) in nonhybridizing populations reflects evolutionary history, thus contrasting potential selection with the neutral genetical history (Li, 1997; Goldstein & Schlötterer, 1999).

For a given polymorphic whitefish system, at least three nonexclusive, evolutionary scenarios may explain diversity among sympatric populations (Smith & Todd, 1984). First, intra-lacustrine radiation and sympatric speciation may result from an adaptive diversification of a gene-pool into divergent niches (Schliewen et al., 1994, 2001; Pigeon et al., 1997; Lu & Bernatchez, 1999; Schluter, 2000). Secondly, colonizations by already differentiated genetic lineages could result in persistence of gene-pools, the emergence of new genetic adaptive variants, or introgression and subsequent breakdown of the adaptive gene combinations (Lu et al., 2001). Thirdly, phenotypic plasticity in a single gene-pool could stem from environmental induction by e.g. differential habitat preference or by developmental responses to diet (Tåning, 1958; Skúlason & Smith, 1995; Day & McPhail, 1996; Robinson & Pearson, 2002).

Adaptive divergence may evolve with or without gene flow (Haldane, 1948; Slatkin, 1973; Rice & Hostert, 1993; Liou & Price, 1994; Campbell & Bernatchez, 2004). In a divergence with gene flow scenario, adaptive differentiation results from natural selection outweighing the homogenizing effect of gene flow. In contrast, a single population could very rapidly split into two reproductively isolated populations if a chance dispersal into a new spawning area was successful, and homing to spawning sites was perfect. By comparing the quantitative trait divergence with neutral allele frequencies it is possible to indicate which traits that are likely driven by natural selection and which traits that likely have no adaptive value and hence are structured by genetic drift or reflect phenotypic plasticity (Wright, 1951; Spitze, 1993; Schluter, 2000). Given the short evolutionary time in northern lakes since the Weichselian glaciation, populations may still be evolving towards the habitat-specific optimal phenotype (Hendry et al., 2002).

In Lake Femund, Norway, local fishermen have long recognized several whitefish forms (Næsje et al., 1992). The taxonomy of the polymorphic whitefish in this lake is, however, not clear. Huitfeldt-Kaas (1918) suggested that a large bodied and a small bodied whitefish race co-occurred in the lake, whereas Svärdson (1957) suggested based on gill-raker numbers, that four whitefish species inhabited the lake. Comparing with other Scandinavian whitefish species, he also suggested that two of these species could have introgressed. More recent surveys suggested the presence of at least three whitefish morphs based on gill-raker numbers and growth differences (Sandlund & Næsje, 1989). Næsje et al. (1992) observed concordance between groups being classified according to gill-raker counts and the genetic distances, based on allozymes, among 11 spawning populations sampled from rivers, and shallow and deep waters. However, genetic heterogeneity were found within some of these gill-raker groups, implying the presence of further population sub-structuring.

On this background, we performed a combined morphological and genetic study of the whitefish in Lake Femund to quantify divergence and to study the evolutionary origin of populations and morphs. We sampled 11 known spawning populations that are separated by habitat and time of spawning. We describe patterns of differentiation in morphometric, meristic, and life-history traits, and compare these patterns with the distribution of genetic variation using microsatellites. Based on a combination of measures, we suggest the best grouping of populations into morphs inhabiting Lake Femund. In search of the traits most likely structured by natural selection, we compared the level of differentiation based on phenotypic variance of quantitative traits with that of molecular differentiation between the morphs, and between populations within morphs. Finally, to elucidate the most probable origin of whitefish populations and morphs in Lake Femund, we compared morphological and genetical characteristics with the seven externally sampled populations in the four major water drainages that potentially could have supplied founders to Lake Femund.

Study area and sampling

  1. Top of page
  2. Abstract
  3. Introduction
  4. Material and methods
  5. Study area and sampling
  6. Morphology
  7. Life-history
  8. Genetic markers
  9. Data analysis
  10. Results
  11. Morphological divergence of populations in Lake Femund
  12. Life-history divergence of populations in Lake Femund
  13. Genetic differentiation of populations in Lake Femund
  14. Comparison of morphological and genetic divergence in Lake Femund
  15. Morphological relationships of Lake Femund vs. nearby lakes
  16. Genetic relationships of Lake Femund vs. nearby lakes
  17. Discussion
  18. Trait differentiation in Lake Femund compared with other coregonid systems
  19. Are phenotypic traits in Lake Femund morphs associated with use of niches?
  20. Sympatric origin of the divergent morphs in Lake Femund?
  21. Could traits separating Lake Femund morphs be altered by natural selection?
  22. Conclusion
  23. Acknowledgments
  24. References
  25. Appendices

Lake Femund, 662 m a.s.l. and 204 km2, is located in the southern part of Norway (62°00′N, 11°55′E) (Fig. 1, Table 1). The lake has two basins with maximum depths of 134 and 90 m, respectively. More than half the lake have depths <20 m mainly in large, shallow bays. The lake is ultra-oligotrophic (Løvik & Kjellberg, 1982), and forms the headwater of the catchment area of River Trysilelva/Klarälven, which drains southeast into Lake Vänern in southern Sweden, and further through River Götaälv to the North Sea. The fish fauna comprises European whitefish, Arctic charr (Salvelinus alpinus), brown trout (Salmo trutta), pike (Esox lucius), perch (Perca fluviatilis), burbot (Lota lota), grayling (Thymallus thymallus) and European minnow (Phoxinus phoxinus). In September to December 1997, we sampled 11 spawning populations in Lake Femund, which comprised four groups according to spawning time and site (Fig. 1, Table 2).

image

Figure 1. Geographical information of the 18 studied whitefish populations. (a) Map of Fennoscandia with the four main drainages, (b) location of the eleven spawning populations in Lake Femund, with depth contours (<20 m), (c) the total seven study lakes.

Download figure to PowerPoint

Table 1.  Geographical, morphometric and biological data of study lakes (population numbers refer to map code in Fig. 1).
Population-LakeMain rivernAltitude (m.a.s.l) Area (km2)Maximum depth (m)StTtLlPfElPpSaRrCpCa
  1. n, Sample size of whitefish.

  2. Species: St, Brown trout (Salmo trutta); Tt, Arctic grayling (Thymallus thymallus); Ll, Burbot (Lota lota); Pf, Eurasian perch (Perca fluviatilis); El, Northern pike (Esox lucius); Pp, Minnow (Phoxinus phoxinus); Sa, Arctic charr (Salvelinus alpinus); Rr, Roach (Rutilus rutilus); Cp, Siberian bullhead (Cottus poecilopus); Ca, Vendace (Coregonus albula).

1–11. FemundTrysil-Klarälven509662204.2134xxxxxxx
12–13. IsterenTrysil-Klarälven6064528.920xxxxxxx
14. EngerenTrysil-Klarälven4547214.085xxxxxxx
15. DrevsjønÖsterdalälven456680.9811xxxxxxxx
16. VurrusjønÖsterdalälven456634.9720xxxxxxxx
17. StorsjøenGlomma3025148.6309xxxxxxxxx
18. LossnenLjusnan2954036085xxxxxxx
Table 2.  Summary data of the 11 spawning populations from Lake Femund with spawning habitat, depth and nominal spawning time.
Map codePopulationSpawningn
HabitatDepth (m)Time
  1. n, The number of analysed fish.

1StorvikaDeep35November–December45
2HulletDeep60November–December36
3VestfjordenDeep30November–December56
4JonesetDeep35November–December45
5HallsteinvikaShallow4–5October–November49
6FemundsendenShallow2–5October–November42
7TufsingaRiver1–2September–October45
8SorkelvaRiver1–2September–October56
9TjønnanRiver2–4September–October45
10KvernvikaBay2–4November–December45
11GløtfossenBay2–5November–December45

We also sampled seven whitefish populations (subsequently termed ‘external populations’) from six lakes situated in the four watercourses that might have been potential post-glacial immigration routes for fish to Lake Femund (Fig. 1, Table 1). Lakes Isteren and Engeren are downstream in the Trysilelva watercourse. Lake Storsjøen is in River Glomma, Lakes Vurrusjøen and Drevsjøen are in River Dalälven, and Lake Lossnen is in River Ljusnan. The Trysilelva and Glomma watercourses drain to the North Sea, whereas the rivers Dalälven and Ljusnan drain to the Baltic Sea. Lake Isteren harbours two whitefish morphs separated in body size, whereas the other lakes, to our knowledge, have monomorphic whitefish. The composition of the fish fauna of the lakes is shown in Table 1. All the fish species are natural colonizers (Huitfeldt-Kaas, 1918), and stocking of whitefish had not occurred in these lakes.

Multi-meshed bottom gill nets were used to collect all samples. In Lake Femund and Lake Storsjøen, all fish were sampled on the spawning grounds as sexually mature individuals. In the other locations, fish was sampled in early fall and represent a mixture of maturity stages. The number of fish in each sampled population ranged between 29 and 56, adding up to a total material of 756 fish (Fig. 1, Tables 1 and 2). All fish were frozen individually in plastic bags at −20 °C prior to conducting all the analysis.

Morphology

  1. Top of page
  2. Abstract
  3. Introduction
  4. Material and methods
  5. Study area and sampling
  6. Morphology
  7. Life-history
  8. Genetic markers
  9. Data analysis
  10. Results
  11. Morphological divergence of populations in Lake Femund
  12. Life-history divergence of populations in Lake Femund
  13. Genetic differentiation of populations in Lake Femund
  14. Comparison of morphological and genetic divergence in Lake Femund
  15. Morphological relationships of Lake Femund vs. nearby lakes
  16. Genetic relationships of Lake Femund vs. nearby lakes
  17. Discussion
  18. Trait differentiation in Lake Femund compared with other coregonid systems
  19. Are phenotypic traits in Lake Femund morphs associated with use of niches?
  20. Sympatric origin of the divergent morphs in Lake Femund?
  21. Could traits separating Lake Femund morphs be altered by natural selection?
  22. Conclusion
  23. Acknowledgments
  24. References
  25. Appendices

Scale length was selected as the best measurement of the body length as natural length and fork length were less reliable because of shearing of fins (Fig. 2a). Using a digital caliper, we measured 17 morphometric traits (Fig. 2b–f); head length (HEL), head height (HEH), preorbital length (PRL), orbital length (ORL), orbital height (ORH), post-orbital length (POL), upper jaw length (UPL), lower jaw length (LOL), inter-orbital breadth (INB), snout width (SNW), snout height (SNH), upper jaw width (UPW), upper gill arch length (UGL), lower gill arch length (LGL), length of gill-raker in the angle between upper and the lower gill arch (GIL), and length of intestine (INL). To measure upper and lower mandibula and calculate the difference between upper and lower bite (ULO), we used a mm-scaled cone (60°) inserted into the mouth of the fish (Fig. 2d). Three meristic traits were counted on the left side only (Fig. 2a,e); number of lateral line scales (LAS), number of gill-rakers on upper gill arch (UPG), and number of gill-rakers on lower gill arch (LOG).

image

Figure 2. Scored traits using a digital caliper. In (a), scale length (SCL) was used as body length measure, and lateral line scales (LAS) was counted along lateral line, in (b) upper jaw width (UPW) was measured on the middle of the upper mandibula, in (c) orbital height (ORH) was measured between inner and outer eye socket, whereas orbital length (ORL) included the preorbital skin-fold until end of socket, in (d) a cone of 60° was inserted into the mouth measuring the bite length for the upper and lower mandibula giving upper and lower jaw length (ULO), in (e) upper gill arch length (UGL) and lower gill arch length (LGL) extend from base of gill-raker length (GIL) until end of gill-arch rim, and GIL from base to top of raker, in (f) the measure of intestine length (INL) was taken on a fully thawed dissected and stretched intestine (from the mouth to the anal pore).

Download figure to PowerPoint

Genetic markers

  1. Top of page
  2. Abstract
  3. Introduction
  4. Material and methods
  5. Study area and sampling
  6. Morphology
  7. Life-history
  8. Genetic markers
  9. Data analysis
  10. Results
  11. Morphological divergence of populations in Lake Femund
  12. Life-history divergence of populations in Lake Femund
  13. Genetic differentiation of populations in Lake Femund
  14. Comparison of morphological and genetic divergence in Lake Femund
  15. Morphological relationships of Lake Femund vs. nearby lakes
  16. Genetic relationships of Lake Femund vs. nearby lakes
  17. Discussion
  18. Trait differentiation in Lake Femund compared with other coregonid systems
  19. Are phenotypic traits in Lake Femund morphs associated with use of niches?
  20. Sympatric origin of the divergent morphs in Lake Femund?
  21. Could traits separating Lake Femund morphs be altered by natural selection?
  22. Conclusion
  23. Acknowledgments
  24. References
  25. Appendices

DNA was extracted from pectoral fins or body muscle using the Phenol/Chloroform procedure (Sambrook et al., 1989). Six microsatellite loci were analysed; SsBgIIIM.26 (Goodier & Davidson; Genbank acc. no. U10051), Cocl-23 (Rogers et al., 2004), C2-157 (Turgeon et al., 1999), Bwf1 (Patton et al., 1997), Bwf2 (Patton et al., 1997), and C1-g (Turgeon et al., 1999). Unpublished mapping data confirm that these loci are on separate linkage groups (S. Rogers, Université Laval, Québec, Canada, unpublished data). The PCR amplifications of each microsatellite locus were performed in a 12 μL reaction volume of 0.3 units of Taq polymerase, 1.25 μL reaction buffer [19 mm Tris-HCL (pH 9.0), 0.1% TritonX-100, 50 mm KCL], 0.6 μL of 1.5 mm MgCl2, 750 μmol of dNTPs, with 0.03 pmol of each two primers, using 25–50 ng template DNA (2 μl). All loci were run in separate PCR's on a Perkin-Elmer 9600 thermocycler (version 2.01; Perkin-Elmer, Boston, MA, USA) according to the following programme: initial denaturation for 2 min at 95 °C, followed by 30 cycles [1 min at 94 °C, 45 s at locus-specific annealing temperature °C (see original references), 30 s at 72 °C], and 10 min at 72 °C. All loci were genotyped on the ABI 377 sequencer (Perkin Elmer) using the fluorescent dye detection, where one primer of each locus was 5′-labelled with one of three fluorescent dyes, allowing six loci to be scored simultaneously as two loci with the same dye had nonoverlapping allele size. Allele sizes were scored using GENESCAN 2.1 (AppliedBiosystems, 1996a) using an internal size standard (TamraTM 350 bp), and were compared with a standard sample in each gel. The final scoring of the allelic sizes was performed three times for each individual using GENOTYPER 2.0 (AppliedBiosystems, 1996b).

Data analysis

  1. Top of page
  2. Abstract
  3. Introduction
  4. Material and methods
  5. Study area and sampling
  6. Morphology
  7. Life-history
  8. Genetic markers
  9. Data analysis
  10. Results
  11. Morphological divergence of populations in Lake Femund
  12. Life-history divergence of populations in Lake Femund
  13. Genetic differentiation of populations in Lake Femund
  14. Comparison of morphological and genetic divergence in Lake Femund
  15. Morphological relationships of Lake Femund vs. nearby lakes
  16. Genetic relationships of Lake Femund vs. nearby lakes
  17. Discussion
  18. Trait differentiation in Lake Femund compared with other coregonid systems
  19. Are phenotypic traits in Lake Femund morphs associated with use of niches?
  20. Sympatric origin of the divergent morphs in Lake Femund?
  21. Could traits separating Lake Femund morphs be altered by natural selection?
  22. Conclusion
  23. Acknowledgments
  24. References
  25. Appendices

As most morphometric and meristic traits were significantly related to individual size, traits were allometrically scaled prior to a statistical analysis using scale length (SCL: two measures), head length (HEL: 14 measures), lower gill arch length (LGL: two measures), and upper gill arch length (UGL: one measure) for standardization (Reist, 1985). First, we log10 transformed traits in order to homogenize variance. Then, we estimated the common within-group regression slope between each log10-trait with log10-SCL, log10-HEL, log10-LGL, or log10-UGL using an ancova without interaction. In this step, and the following, all 18 populations were used in the allometric scaling. Slopes were used to correct each fish to the overall comparative standardized mean fish size using the mean values of the log10-transformed SCL (1.47), HEL (1.79), LGL (1.36) or UGL (1.41) as incorporated into the following formula (see Hendry et al., 2002):

  • image

where M is the trait size, L is trait used for scaling, b is the ancova slope with the interaction term removed, and subscripts std, o and x refer to standardized, observed and mean values of traits. For the meristic traits, three populations had a significant correlation between LAS × SCL, and four populations had a correlation between UPG × SCL and LOG × SCL, concordantly we size-corrected traits by SCL, LGL and UGL.

Morphological variation was assessed by a canonical linear discriminant analysis of allometrically scaled metric and meristic traits using JMP 5.0 (SAS, 2002). This analysis uses Mahalanobis distance to estimate each individual's distance from the group multivariate means. A stepwise forward variable selection, entering one trait at time, was performed where trait entering order represents discriminating power. This analysis was used to describe pattern of divergence in the 11 Lake Femund populations, and to find the most likely colonizer of Lake Femund, forcing each fish from Lake Femund into the seven external populations from the likely immigration routes using K-means clustering in JMP 5.0 (SAS, 2002).

Life-history variation was documented modelling the population specific growth curves (vonBertalanffy's growth model, Bagenal & Tesch, 1978) for the 11 Lake Femund populations. Populations with nonoverlapping confidence intervals (95% CI) for the asymptotic length (L) were considered as significantly different.

Genetic variation was described as microsatellite allele frequency, observed heterozygosity (HO), expected heterozygosity (HE), and by F-statistics using Genepop 3.3 (Raymond & Rousset, 1995). Allelic richness, i.e. allele number corrected for sample size (El Mousadik & Petit, 1996), was quantified using FSTAT 2.9.3 (Goudet, 1995). Deviation from Hardy–Weinberg equilibrium was tested by the exact (probability) test for each locus and population (Guo & Thompson, 1992), and P-values were corrected using the sequential Bonferroni method for each locus (α = 0.05; k = 18).

Population differentiation was estimated using the log-likelihood based exact test on contingency tables of genotypes (rather than on the alleles as deviation from H-W equilibrium was found in several cases). Tests were made across populations, and between pairs of populations, using Genepop 3.3, and combined across loci using Fisher's combined probabilities. Genetic relationships within Lake Femund were also studied using assignment tests where each individual was assigned to the closest population based on its genetic distance (DA; Nei et al., 1983) to all populations in GeneClass 1.0 (Cornuet et al., 1999). This distance method can be used regardless of deviations from Hardy–Weinberg equilibrium (Cornuet et al., 1999).

The distribution of allelic variation was quantified by intergroup components of total variance by φ statistics using a hierarchical analysis of molecular variance, amova, in Arlequin 2.0 (Excoffier et al., 1992; Schneider et al., 2000). In this way, variance components among morphs, among populations within morphs, and within populations were estimated in Lake Femund.

Genetic relationships between Lake Femund whitefish spawning populations and the external populations were assessed by two methods. First, we used an assignment test to find the most likely origin of the eleven Lake Femund populations, by fitting each individual into one of the surrounding reference populations using the DA distance. Secondly, we performed a principal component analysis in PCA-GEN 1.2 (Goudet, 1999; http://www2.unil.ch/popgen/softwares/pcagen.htm) to visualize genetic relationships between the Lake Femund whitefish populations and the seven external populations. Significance of the PC1 and PC2 axis was estimated by 1000 randomizations of individuals among populations.

In order to explore the possible role of natural selection in shaping morphological divergence between and within the suggested Lake Femund morphs, we compared the extent of morphological divergence (Qst) with that based on molecular divergence (Fst) using methods from Spitze (1993). These measures should be equal if the underlying loci are selectively neutral, and if the quantitative traits have an additive genetic basis (Wright, 1951; Schluter, 2000; Lópes-Fanjul et al., 2003). We used phenotypic variance as a surrogate for the additive genetic variance for phenotypic traits (Meriläet al., 1997; Merilä & Crnokrak, 2001), and estimated the phenotypic variance components following procedures in Bernatchez (2004). Qst for each trait was estimated between spawning populations, and also between the suggested morphs. The Qst and Fst values were significantly different if their 95% confidence intervals did not overlap. Confidence intervals for all pairwise Fst values were calculated from the standard error of Fst among all eleven populations, using FSTAT 2.9.3 (Goudet, 1995). We are fully aware that Qst estimates without quantitative genetics information is only tentative (e.g. Lópes-Fanjul et al., 2003), but we use this method merely to suggest traits most likely targeted by selection. In this particular case also, comparing morphs within Lake Femund will reduce environmental effects as compared with studies across lakes (Bernatchez, 2004).

Morphological divergence of populations in Lake Femund

  1. Top of page
  2. Abstract
  3. Introduction
  4. Material and methods
  5. Study area and sampling
  6. Morphology
  7. Life-history
  8. Genetic markers
  9. Data analysis
  10. Results
  11. Morphological divergence of populations in Lake Femund
  12. Life-history divergence of populations in Lake Femund
  13. Genetic differentiation of populations in Lake Femund
  14. Comparison of morphological and genetic divergence in Lake Femund
  15. Morphological relationships of Lake Femund vs. nearby lakes
  16. Genetic relationships of Lake Femund vs. nearby lakes
  17. Discussion
  18. Trait differentiation in Lake Femund compared with other coregonid systems
  19. Are phenotypic traits in Lake Femund morphs associated with use of niches?
  20. Sympatric origin of the divergent morphs in Lake Femund?
  21. Could traits separating Lake Femund morphs be altered by natural selection?
  22. Conclusion
  23. Acknowledgments
  24. References
  25. Appendices

There were significant differences for all the 20 studied traits when analysed over all the 18 populations (Appendix 1). In Lake Femund, all 20 traits also differed significantly among eleven spawning populations (anova, R2 = 0.06–0.85, P < 0.05). The five most differentiated traits, based on R2 values, were the number of lower gill rakers (LOG, R2 = 0.85, range of population means: 18.0–28.7), upper gill rakers (UPG, R2 = 0.83, 10.4–13.4), gill-raker length (GIL, R2 = 0.76, 5.29–9.33), lower jaw length (LOL, R2 = 0.49, 22.5–25.4), and upper jaw length (UPL, R2 = 0.29, 16.4–18.3).

The canonical discrimant analysis of the eleven Lake Femund populations was significant with 20 traits combined (Pillai's trace = 2.87, P < 0.0001), with only three traits not contributing significantly to this model (Table 3, Fig. 3). Based on the forward stepwise variable selection, the highest discriminating power was found in LOG, GIL, ULO, UGL and SNH (see entering order in Table 3). The first canonical axis (CA1) explained 79.5% of the total variation in the material, where +LOG loaded with −LGL, −ULO, +GIL and −SNH. Thus, separation of populations along CA1 was due to fish with many gill rakers on the gill arch, long gill rakers, small snout height and terminal mouth being separated from fish with fewer rakers on the gill arch, short gill rakers, high snout heights and a subterminal mouth.

Table 3.  Eigenvectors in canonical discriminant analysis based on metric and meristic traits for the 11 spawning populations in Lake Femund.
TraitAbbr.Corr.Population
OrderCA1CA2CA3
  1. Traits (Corr.) used for size correction (Abbr.) in allometric scaling are given. Order refers to entering order in the stepwise variable selection.

  2. *The traits that had a nonsignificant contribution in the model (P > 0.05).

1. Head lengthHELSCL17−4.63−3.83−20.62
2. Head heightHEHHEL20*1.5916.38−8.07
3. Preorbital lengthPRLHEL163.25−16.765.43
4. Orbital lengthORLHEL18*3.08−14.0911.33
5. Orbital heightORHHEL7−10.7226.905.57
6. Postorbital lengthPOLHEL1111.63−2.7525.06
7. Upper jaw lengthUPLHEL14−1.613.57−25.79
8. Lower jaw lengthLOLHEL612.785.83−26.09
9. Interorbital breadthINBHEL10−1.8925.052.67
10. Snout widthSNWHEL910.336.5329.81
11. Snout heightSNHHEL5−5.42−7.540.63
12. Upper jaw widthUPWHEL84.49−27.43−20.97
13. Upper gill arch lengthUGLHEL41.62−5.683.22
14. Lower gill arch lengthLGLHEL15−16.504.9316.32
15. Gill-raker lengthGILLGL28.131.89−2.75
16. Intestine lengthINLSCL134.5514.755.83
17. U.-L. jaw length 60 °ULOHEL3−2.04−1.772.46
18. Upper gill arch rakersUPGUGL1211.696.750.90
19. Lower gill arch rakersLOGLGL118.04−8.766.00
20. Lateral line scalesLASSCL19*5.42−5.971.17
Eigenvalues11.490.850.68
Cumulative per cent explained79.585.490.1
image

Figure 3. Canonical discriminant analysis for the 11 Lake Femund populations based on all 20 scored morphological traits (upper section). The multivariate mean with 95% confidence limits is given. Groups that are significantly different tend to have nonintersecting circles. In the lower section, the most important traits separating the four main groups are displayed with mean values (see trait definitions and abbreviations in Table 3, Fig. 2).

Download figure to PowerPoint

A graphical interpretation (and mean values) of the main trait differences between the populations is given in Fig. 3, where four groups are resolved in Lake Femund: (I) Joneset, Vestfjorden, Hullet and Storvika, all being deep-spawning populations (see Table 2), (II) Hallsteinvika and Femundsenden, both spawning in shallow locations, (III) Tufsinga, Tjønnan and Sorkelva, spawning in rivers including a river mouth and (IV) Kvernvika and Gløtfossen, spawning in semi-isolated bays. The deep spawning populations were significantly different from the shallow spawning populations, whereas these two were both significantly different from river spawners, and the two bay spawning populations. Finally, the river spawners were also significantly different from the bay spawners.

Life-history divergence of populations in Lake Femund

  1. Top of page
  2. Abstract
  3. Introduction
  4. Material and methods
  5. Study area and sampling
  6. Morphology
  7. Life-history
  8. Genetic markers
  9. Data analysis
  10. Results
  11. Morphological divergence of populations in Lake Femund
  12. Life-history divergence of populations in Lake Femund
  13. Genetic differentiation of populations in Lake Femund
  14. Comparison of morphological and genetic divergence in Lake Femund
  15. Morphological relationships of Lake Femund vs. nearby lakes
  16. Genetic relationships of Lake Femund vs. nearby lakes
  17. Discussion
  18. Trait differentiation in Lake Femund compared with other coregonid systems
  19. Are phenotypic traits in Lake Femund morphs associated with use of niches?
  20. Sympatric origin of the divergent morphs in Lake Femund?
  21. Could traits separating Lake Femund morphs be altered by natural selection?
  22. Conclusion
  23. Acknowledgments
  24. References
  25. Appendices

All estimated length-at-age growth models for the 11 populations were significant (P < 0.05) (Fig. 4). Asymptotic body-length (L) in the populations varied from 28.9 cm (Storvika) to 38.8 cm (Kvernvika). The four deep-spawning populations (Joneset, Vestfjorden, Hullet and Storvika) did not differ from each other in L, but were significantly different from all other populations. A second group (based on overlapping confidence intervals for L) consisted of the two shallow-spawning populations (Hallsteinvika and Femundsenden), and the three river-spawning populations (Tufsinga, Tjønnan and Sorkelva). They were different, however, from a third group of populations with the largest L, the bay-spawning populations (Kvernvika and Gløtfossen). The mean age of spawners in each population was 8.5–9.5 years in the bay morph (Kvernvika and Gløtfossen), 11–12 years in the river morph (Tufsinga, Tjønnan and Sorkelva), 14–15 years in the shallow morph (Hallsteinvika and Femundsenden) and 12.5–16.5 years in the deep morph (Joneset, Vestfjorden, Hullet and Storvika) (see Fig. 4).

image

Figure 4. Life-history parameters based on vonBertalanffy growth model in the 11 Lake Femund populations. Length-at-age is given for each population. Triangles denote the mean population age. Scale length (Fig. 2; SCL) is used as a measure for body-length.

Download figure to PowerPoint

Genetic differentiation of populations in Lake Femund

  1. Top of page
  2. Abstract
  3. Introduction
  4. Material and methods
  5. Study area and sampling
  6. Morphology
  7. Life-history
  8. Genetic markers
  9. Data analysis
  10. Results
  11. Morphological divergence of populations in Lake Femund
  12. Life-history divergence of populations in Lake Femund
  13. Genetic differentiation of populations in Lake Femund
  14. Comparison of morphological and genetic divergence in Lake Femund
  15. Morphological relationships of Lake Femund vs. nearby lakes
  16. Genetic relationships of Lake Femund vs. nearby lakes
  17. Discussion
  18. Trait differentiation in Lake Femund compared with other coregonid systems
  19. Are phenotypic traits in Lake Femund morphs associated with use of niches?
  20. Sympatric origin of the divergent morphs in Lake Femund?
  21. Could traits separating Lake Femund morphs be altered by natural selection?
  22. Conclusion
  23. Acknowledgments
  24. References
  25. Appendices

In all but one sample, the expected heterozygosity was higher than the observed heterozygosity (i.e. Fis > 0). Consequently, significant deviations from Hardy–Weinberg proportions were observed in a number of cases (Appendix 2). Most deviations were associated with the locus C1-g. This condition was not caused by a significant difference between cohorts (tested among age groups in a combined sample of all deep-water whitefish), nor by one specific homozygote being dominant in all the populations. Rather, different homozygotes tended to be over-represented in different populations. We therefore believe that a null allele in C1-g is the most likely cause for deviations from Hardy–Weinberg proportions. Alternatively, Wahlund effects cannot entirely be ruled out as fish on spawning grounds could have strayed from another population.

The six microsatellites were variable in all populations, with 10–29 alleles per locus in Lake Femund (Appendix 2). Allelic richness, the mean number of alleles per locus and population adjusted to n = 27, varied little among populations. The lowest mean values (6.0–6.3) were found in Gløtfossen and Kvernvika, and the highest value (7.6) in Hallsteinvika and Vestfjorden. The average observed heterozygosity within populations varied from 0.481 (Kvernvika) to 0.580 (Tufsinga), whereas the average gene diversity estimate varied from 0.561 (Storvika) to 0.655 (Tufsinga).

Five genetically distinct gene pools were resolved within Lake Femund; the four distinct morphs, as well as a distinction between the two populations of the bay morph (Table 4). Thus, pairwise tests of population heterogeneity showed that the populations of Hullet, Storvika, Vestfjorden and Joneset were not significantly differentiated (Table 4). The Tufsinga river population was not significantly different from Sorkelva or Tjønnan, although the latter two were significantly different from each other. Hallsteinvika differed neither from Femundsenden nor from Storvika, Vestfjorden or Joneset. Kvernvika and Gløtfossen populations were significantly different from all the other populations, as well as from each other. Pairwise Fst values among populations ranged from 0.005 or less between the four deep-spawning populations (Storvika, Hullet, Vestfjorden and Joneset), and between the two shallow-spawning populations (Femundsenden and Hallsteinvika), to 0.153 between the bay-spawning populations (Kvernvika and Gløtfossen). The three river-spawning populations (Tufsinga, Tjønnan and Sorkelva) had Fst values of approximately 0.01.

Table 4.  The summary statistics of the pairwise population heterogenity tests (P-values, above diagonal), and the genetic divergence as found by Fst (below diagonal).
Population1. Stor2. Hull3. Vest4. Jone5. Hall6. Femu7. Tufs8. Sork9. Tjøn10. Kver11. Gløt
  1. *Not significant after multiple alpha-correction.

 1. Storvika 0.215*0.104*0.472*0.003*0.0010.0010.0010.0010.0010.001
 2. Hullet0.000 0.019*0.085*0.0010.0010.0010.0010.0010.0010.001
 3. Vestjorden0.0040.005 0.395*0.007*0.0010.0010.0010.0010.0010.001
 4. Joneset0.0030.0040.000 0.009*0.0010.0010.0010.0010.0010.001
 5. Hallsteinvika0.0160.0130.0080.015 0.290*0.0010.0010.0010.0010.001
 6. Femundsenden0.0220.0150.0160.0190.001 0.0010.0010.0010.0010.001
 7. Tufsinga0.0460.0370.0420.0340.0290.036 0.007*0.145*0.0010.001
 8. Sørkelva0.0220.0290.0300.0230.0330.0390.011 0.0010.0010.001
 9. Tjønnan0.0440.0370.0430.0320.0480.0570.0060.012 0.0010.001
10. Kvernvika0.1200.1210.1240.1050.1250.1430.1320.1300.121 0.001
11. Gløtfossen0.1300.1290.1240.1000.1360.1410.0930.0950.0880.153 

A hierarchical analysis of molecular variance (amova; all P < 0.0001) showed that 3.9% of the genetic variation was found between the four morphs, 2.8% was found among populations within morphs and 93.3% within populations. Excluding the two highly genetically divergent bay morph populations (Kvernvika and Gløtfossen) from the analysis (amova; all P < 0.0001) caused the between morphs component to increase relative to the among populations within morphs component (2.6% vs. 0.6%, respectively), whereas the majority of variation was found within populations (96.8%).

Taking into consideration spawning site, spawning time, morphology, life history and genetic structure, the eleven spawning populations in Lake Femund were naturally grouped into; (I) Deep morph (Storvika, Hullet, Vestfjorden and Joneset), (II) Shallow morph (Hallsteinvika and Femundsenden), (III) River morph (Tufsinga, Tjønnan and Sorkelva) and (IV) Bay morph (Kvernvika and Gløtfossen). The two bay morph populations of Kvernvika and Gløtfossen were genetically different, yet similar in morphology and life-history traits. In the following, they are viewed as two genetically differentiated populations of the same morph.

Comparison of morphological and genetic divergence in Lake Femund

  1. Top of page
  2. Abstract
  3. Introduction
  4. Material and methods
  5. Study area and sampling
  6. Morphology
  7. Life-history
  8. Genetic markers
  9. Data analysis
  10. Results
  11. Morphological divergence of populations in Lake Femund
  12. Life-history divergence of populations in Lake Femund
  13. Genetic differentiation of populations in Lake Femund
  14. Comparison of morphological and genetic divergence in Lake Femund
  15. Morphological relationships of Lake Femund vs. nearby lakes
  16. Genetic relationships of Lake Femund vs. nearby lakes
  17. Discussion
  18. Trait differentiation in Lake Femund compared with other coregonid systems
  19. Are phenotypic traits in Lake Femund morphs associated with use of niches?
  20. Sympatric origin of the divergent morphs in Lake Femund?
  21. Could traits separating Lake Femund morphs be altered by natural selection?
  22. Conclusion
  23. Acknowledgments
  24. References
  25. Appendices

Between populations within the same morph, there was little evidence that the genetic variation underlying morphological traits differed from the pattern of variation in microsatellites (Fig. 5). The four deep-spawning populations (panel A) did not differ significantly in their Qst–Fst values as the 95% CI for Fst encompassed all the Qst values. The same general result held true for the three river-spawning populations (panel C), and most likely also for the two shallow-spawning populations (panel B) where all Qst values were small and close to Fst. The two bay populations were harder to evaluate, as they had a high Fst and featured traits having both higher, and lower Qst value (panel D). Also, we could not evaluate whether the Qst values are significantly different from Fst as we only have one pairwise comparison for each trait in panels B and D. Yet, it is noteworthy that 15 of 20 traits have a smaller Qst value than Fst in the bay morph populations.

image

Figure 5. Qst–Fst values for all the within- and between-morph comparisons with means and the 95% confidence intervals for Qst (bars), and 95% CI for Fst values (stippled lines). Zero is denoted by a grey dotted line.

Download figure to PowerPoint

In contrast to within-morph comparisons, in all comparisons of populations belonging to different morphs, we found traits showing significantly higher Qst values than Fst (Fig. 5e–j). The number of gill rakers (UPG, LOG) showed the highest Qst/Fst ratio and were significantly different in five (LOG) or all (UPG) of the six inter-morph comparisons. Gill-raker length had the second highest Qst/Fst ratio being different in four of six comparisons. Post-orbital length showed significantly higher Qst in all comparisons except between populations in the deep and shallow morph. Other head morphology traits, especially jaw and snout measurements, showed significantly higher Qst than Fst in some of the comparisons.

In the comparisons involving the bay-spawning populations, the lower Fst confidence limit was well above zero [Fig. 5g (0.094), h (0.110) and j (0.084)]. Hence, for these comparisons it is possible to test whether some Qst values were significantly smaller than the neutral expectation. The deep and bay morph showed significantly small Qst values for orbital length and lower gill-arch length (panel G), the shallow and bay morph showed significantly small Qst for several head measurements and for intestine length (panel H), whereas the river and bay morphs showed a significantly small Qst for three head measurements (panel J).

Morphological relationships of Lake Femund vs. nearby lakes

  1. Top of page
  2. Abstract
  3. Introduction
  4. Material and methods
  5. Study area and sampling
  6. Morphology
  7. Life-history
  8. Genetic markers
  9. Data analysis
  10. Results
  11. Morphological divergence of populations in Lake Femund
  12. Life-history divergence of populations in Lake Femund
  13. Genetic differentiation of populations in Lake Femund
  14. Comparison of morphological and genetic divergence in Lake Femund
  15. Morphological relationships of Lake Femund vs. nearby lakes
  16. Genetic relationships of Lake Femund vs. nearby lakes
  17. Discussion
  18. Trait differentiation in Lake Femund compared with other coregonid systems
  19. Are phenotypic traits in Lake Femund morphs associated with use of niches?
  20. Sympatric origin of the divergent morphs in Lake Femund?
  21. Could traits separating Lake Femund morphs be altered by natural selection?
  22. Conclusion
  23. Acknowledgments
  24. References
  25. Appendices

When we applied the canonical discriminant analysis (Pillai's trace 3.89, P < 0.001) on all phenotypic traits, extracting 59.8% of the total variation on CA1 and 24.8% on CA2, the four Lake Femund morphs clustered together with the normal morph from Lake Isteren and Lake Storsjøen (Fig. 6). Lake Engeren also appeared to be quite close morphologically, whereas Lossnen, Vurrusjøen, Drevsjøen and the Lake Isteren dwarfs were clearly much more distant. The closely localized Lake Femund and Lake Isteren, both in the Trysil-Klarälven water drainage, separated only by a short stretch of river, encompass all the variation found in all the 18 populations.

image

Figure 6. Canonical discriminant analysis for the four Lake Femund morphs (the bay morph has been partitioned into Gløtfossen and Kvernvika populations) and seven external populations using 20 traits. Multivariate means with 95% confidence limit are given.

Download figure to PowerPoint

When treating each of the 11 Lake Femund populations as unknown samples forced into the seven external populations using the K-means clustering, the assignment was highest to Storsjøen. With only one exception (Hullet), all the Lake Femund populations were closest to Lake Storsjøen, then Engeren and Lake Isteren-normal. Very few individuals were assigned to Lake Lossnen or Drevsjøen, and none was assigned to Lake Isteren-dwarf. Moreover, the three Lake Femund populations with the highest assignment to Lake Storsjøen (a river spawning population) were the three Lake Femund river spawning populations of Tufsinga, Tjønnan and Sorkelva.

Genetic relationships of Lake Femund vs. nearby lakes

  1. Top of page
  2. Abstract
  3. Introduction
  4. Material and methods
  5. Study area and sampling
  6. Morphology
  7. Life-history
  8. Genetic markers
  9. Data analysis
  10. Results
  11. Morphological divergence of populations in Lake Femund
  12. Life-history divergence of populations in Lake Femund
  13. Genetic differentiation of populations in Lake Femund
  14. Comparison of morphological and genetic divergence in Lake Femund
  15. Morphological relationships of Lake Femund vs. nearby lakes
  16. Genetic relationships of Lake Femund vs. nearby lakes
  17. Discussion
  18. Trait differentiation in Lake Femund compared with other coregonid systems
  19. Are phenotypic traits in Lake Femund morphs associated with use of niches?
  20. Sympatric origin of the divergent morphs in Lake Femund?
  21. Could traits separating Lake Femund morphs be altered by natural selection?
  22. Conclusion
  23. Acknowledgments
  24. References
  25. Appendices

Among the seven external populations, we found that the six microsatellites ranged between 3 and 26 alleles per locus (Appendix 2). Adjusted for sample size (n = 27), allelic richness varied from 5.7 (Isteren-normal) to 12.7 (Vurrusjøen). The average observed heterozygosity within populations varied between 0.528 (Lake Storsjøen) and 0.793 (Vurrusjøen), whereas the average expected heterozygosity (or gene diversity) ranged from 0.599 (Isteren-normal) to 0.856 (Lake Vurrusjøen, Appendix 2). Most values were therefore on the high side, or above, those found in the Lake Femund populations.

Using pairwise tests of population heterogeneity, all the seven external populations were significantly different from each other, as well as from the eleven Lake Femund populations (i.e. P < 0.05 after Bonferroni correction). Pairwise Fst values ranged from 0.031 (Vestfjorden vs. Isteren-normal) to 0.356 (Storvika vs. Lake Storsjøen). The sympatric dwarf and normal morphs in Lake Isteren had Fst value of 0.085 (P < 0.001).

The relationships between Lake Femund whitefish and the external populations were also assessed by assigning individuals from Lake Femund to one of the external populations. For eight of the 11 Lake Femund populations (all deep-water populations, two river populations, both shallow-water populations), more than 95% of the fish were assigned to Isteren or Engeren. Moreover, assignment to Isteren was to both the dwarf and normal morph, irrespective of Lake Femund morph. For the remaining three populations in Lake Femund, assignment was in addition to Lossnen (<26% for Gløtfossen, Tjønnan, Kvernvika) and to Lake Drevsjøen (<3% for Tjønnan, Kvernvika). No individuals were assigned to Lake Vurrusjøen or Storsjøen. Thus, for each of the 11 Lake Femund populations at least 71% were assigned to jointly Lake Isteren (dwarf + normal) and Lake Engeren in the Trysil-Klarälven river.

A principal component analysis explained 39.7 and 17.4% of variation for PC1 axis (P = 0.002) and PC2 axis (P = 0.34), respectively (Fig. 7). All Lake Femund samples (seen as morphs in Fig. 7) clustered with the morphs from Lake Isteren (normal and dwarf), and with Lake Engeren. Thus, the external populations bearing the closest genetic relationship to Lake Femund populations come from other lakes in the same river system.

image

Figure 7. Principal component analysis on multilocus microsatellites from four Lake Femund morphs (with the bay morph partitioned into Gløtfossen and Kvernvika populations) compared with the seven external populations.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Material and methods
  5. Study area and sampling
  6. Morphology
  7. Life-history
  8. Genetic markers
  9. Data analysis
  10. Results
  11. Morphological divergence of populations in Lake Femund
  12. Life-history divergence of populations in Lake Femund
  13. Genetic differentiation of populations in Lake Femund
  14. Comparison of morphological and genetic divergence in Lake Femund
  15. Morphological relationships of Lake Femund vs. nearby lakes
  16. Genetic relationships of Lake Femund vs. nearby lakes
  17. Discussion
  18. Trait differentiation in Lake Femund compared with other coregonid systems
  19. Are phenotypic traits in Lake Femund morphs associated with use of niches?
  20. Sympatric origin of the divergent morphs in Lake Femund?
  21. Could traits separating Lake Femund morphs be altered by natural selection?
  22. Conclusion
  23. Acknowledgments
  24. References
  25. Appendices

Our results showed that the four whitefish morphs (deep, shallow, river and bay) inhabiting Lake Femund constitute discrete units based on morphological, life-history and genetic traits, and spawn at different times in segregated environments. In contrast, populations within these morphs were remarkably similar, as opposed to divergence in traits across morph categories, except for the two bay morphs which are highly genetically differentiated, but very similar in terms of morphology and life-history. Potential determinants behind intralacustrine divergence are discussed below.

Trait differentiation in Lake Femund compared with other coregonid systems

  1. Top of page
  2. Abstract
  3. Introduction
  4. Material and methods
  5. Study area and sampling
  6. Morphology
  7. Life-history
  8. Genetic markers
  9. Data analysis
  10. Results
  11. Morphological divergence of populations in Lake Femund
  12. Life-history divergence of populations in Lake Femund
  13. Genetic differentiation of populations in Lake Femund
  14. Comparison of morphological and genetic divergence in Lake Femund
  15. Morphological relationships of Lake Femund vs. nearby lakes
  16. Genetic relationships of Lake Femund vs. nearby lakes
  17. Discussion
  18. Trait differentiation in Lake Femund compared with other coregonid systems
  19. Are phenotypic traits in Lake Femund morphs associated with use of niches?
  20. Sympatric origin of the divergent morphs in Lake Femund?
  21. Could traits separating Lake Femund morphs be altered by natural selection?
  22. Conclusion
  23. Acknowledgments
  24. References
  25. Appendices

In Lake Femund, the four morphs were mainly separated by the number and length of gill rakers, lower jaw length and mouth position. Differences at these traits are commonly found to separate European whitefish morphs (Rufli, 1978; Kirchhofer, 1990; Amundsen et al., 2004), as well as other coregonids, e.g. Lake Baikal Omul (C. autumnalis migratorius) (Smirnov, 1992; Bronte et al., 1999), North American lake whitefish (C. clupeaformis) (Bodaly, 1979; Lu & Bernatchez, 1999), North American ciscoes (Coregonus spp.) (Todd & Smith, 1992; Turgeon et al., 1999), and European vendace (C. albula) (Svärdson, 1979; Schulz & Freyhof, 2003). In laboratory experiments, variation in such traits has been shown to influence prey-handling ability (Malmquist, 1992; Day & McPhail, 1996; Adams & Huntingford, 2004).

Life-history traits of the four Lake Femund morphs differed with respect to juvenile growth rate, age at sexual maturation, growth trajectories and longevity (see Sandlund et al., 1995). Jensen (1985) found that these traits are related in the lake whitefish, and that changes in age at maturity and survival of immature fish had a large effect on the net reproductive rate of whitefish populations. Such life history differentiation is commonly seen in whitefish systems both in Europe and in North America (Bodaly, 1979; Kirkpatrick & Selander, 1979; Svarvar & Müller, 1982; Kirchhofer & Tschumi, 1986; Lehtonen & Niemelä, 1998; Svärdson, 1998; Lu & Bernatchez, 1999; Kahilainen & Lehtonen, 2002). It seems to be a general trend that sympatric whitefish populations differing in gill-raker numbers also display different growth trajectories. In North America, however, a much less pronounced divergence in gillraker numbers is associated with similar body size difference as seen in Europe.

The Fst values between pairs of the four Lake Femund morphs ranged from 0.008 to 0.141. This covers a large part of the range of within-lake genetic differentiation reported from Coregonus spp. and other salmonids (Hindar, 1994; Hendry & Stearns, 2004). In the European whitefish, Douglas et al. (1999) found an average Fst of 0.049 between sympatric morphs in eight Alpine lakes. In lake whitefish, Fst values between sympatric pairs within the same mtDNA lineage ranged from 0.01 to 0.084, whereas comparisons between pairs representing different mtDNA lineages gave Fst values from 0.020 to 0.256 (Kirkpatrick & Selander, 1979; Bodaly et al., 1992; Lu & Bernatchez, 1999). In four morphologically-defined cisco taxa (Coregonus spp.) in Lake Nipigon, Turgeon et al. (1999) found Fst values from 0.001 to 0.029. Thus, the maximum Fst values between the Lake Femund morphs are very high, and close to values in contact zones between divergent mtDNA lineages in lake whitefish (Lu & Bernatchez, 1999; Lu et al., 2001).

Apparently, sympatric populations of Coregonus are generally (but not always) characterized by rather small Fst values despite a pronounced divergence in morphology, life-history and ecology. Furthermore, the generality of trait differentiation between sympatric populations suggests that a similar suite of evolutionary forces may be acting in promoting divergence across the different species and environments.

Are phenotypic traits in Lake Femund morphs associated with use of niches?

  1. Top of page
  2. Abstract
  3. Introduction
  4. Material and methods
  5. Study area and sampling
  6. Morphology
  7. Life-history
  8. Genetic markers
  9. Data analysis
  10. Results
  11. Morphological divergence of populations in Lake Femund
  12. Life-history divergence of populations in Lake Femund
  13. Genetic differentiation of populations in Lake Femund
  14. Comparison of morphological and genetic divergence in Lake Femund
  15. Morphological relationships of Lake Femund vs. nearby lakes
  16. Genetic relationships of Lake Femund vs. nearby lakes
  17. Discussion
  18. Trait differentiation in Lake Femund compared with other coregonid systems
  19. Are phenotypic traits in Lake Femund morphs associated with use of niches?
  20. Sympatric origin of the divergent morphs in Lake Femund?
  21. Could traits separating Lake Femund morphs be altered by natural selection?
  22. Conclusion
  23. Acknowledgments
  24. References
  25. Appendices

If trait differences between the Lake Femund morphs reflect foraging specialization, we would expect an association between trait value and habitat use, prey selection, and behaviour. It is commonly seen that whitefish in monomorphic stocks have a wider niche than individual morphs living in sympatry (e.g. Sandlund et al., 1995; Amundsen et al., 2004). In dimorphic stocks, the most densely-rakered morph usually utilizes the pelagic habitat feeding on zooplankton, whereas the low-rakered morph to a larger extent feeds epibenthically on zoobenthos (Amundsen et al., 2004). Differentiation along a similar gradient is seen in stocks with three (Kahilainen et al., 2004) or four morphs (Bergstrand, 1982). In light of these observations, habitat use and diet of the whitefish morphs in Lake Femund are somewhat unusual. Here, adults of the densely-rakered morphs utilize the pelagic zone to a lesser extent than the low-rakered morphs, feeding on zooplankton in late summer and autumn (Næsje et al., 1998; Saksgård et al., 2002). The deep and shallow morphs are distributed in all epibenthic depth ranges (0–60 m) as well as the pelagic zone during summer, whereas the river and bay morphs are more restricted to epibenthic shallow waters (<20 m). However, in line with other studies, it seems that deep- and shallow-morphs have a wider niche than river- and bay-morphs, suggesting a more opportunistic foraging behaviour of low-rakered morphs in Lake Femund. In this ultra-oligotrophic lake, zooplankton densities are very low, even during peak densities in August–October (Løvik & Kjellberg, 1982). Thus, improved ability to use zooplankton may not be of great adaptive value to adults. Diet-related selection pressures may rather be related to the epibenthic resource in juvenile stages.

Less data are available in the literature with regard to a potential association between trait and foraging behaviour. Experiments suggest that trait variation is associated with prey-handling abilities (e.g. Day & McPhail, 1996). Thus, it seems reasonable to expect adaptive behavioural differences in foraging between morphs. Three lines of argumentation suggest that behavioural differences in foraging acquisition between morphs are likely and adaptive. First, behaviour modification will be a more rapid event associated with the use of a new niche than modification of structural traits (Dill, 1983). Secondly, there are probably functional constraints with regard to optimal foraging on a given prey item, with densely-rakered fish being more effective feeding on small prey than sparsely-rakered fish (e.g. Larson, 1976; Bentzen & McPhail, 1984). Thirdly, behaviour may be heritable (Klemetsen et al., 2002). For example, whitefish hybrids seem to be intermediate in diet choice (Svärdson, 1957; Voloshenko, 1973). Also, Rogers et al. (2002) recently documented a genetically based phenotype–environment association for swimming behaviour in two morphs of the lake whitefish. Finally, in Lake Muddusjärvi, three whitefish morphs show different foraging behaviour and prey efficiencies when experimentally consuming chironomid larvae (K. Kahilainen, personal communication, Department of Biological and Environmental Sciences, University of Helsinki, Finland). Although information on feeding behaviour is lacking in Lake Femund, it is reasonable to assume an association between adaptive traits and foraging behaviour. Also, similarity of morphs in these northern oligotrophic lakes suggests a potential for convergence in adaptive behavioural traits.

Sympatric origin of the divergent morphs in Lake Femund?

  1. Top of page
  2. Abstract
  3. Introduction
  4. Material and methods
  5. Study area and sampling
  6. Morphology
  7. Life-history
  8. Genetic markers
  9. Data analysis
  10. Results
  11. Morphological divergence of populations in Lake Femund
  12. Life-history divergence of populations in Lake Femund
  13. Genetic differentiation of populations in Lake Femund
  14. Comparison of morphological and genetic divergence in Lake Femund
  15. Morphological relationships of Lake Femund vs. nearby lakes
  16. Genetic relationships of Lake Femund vs. nearby lakes
  17. Discussion
  18. Trait differentiation in Lake Femund compared with other coregonid systems
  19. Are phenotypic traits in Lake Femund morphs associated with use of niches?
  20. Sympatric origin of the divergent morphs in Lake Femund?
  21. Could traits separating Lake Femund morphs be altered by natural selection?
  22. Conclusion
  23. Acknowledgments
  24. References
  25. Appendices

Our results suggest that an intra-lacustrine origin of the four whitefish morphs in Lake Femund is a more likely event than allopatric origin. Huitfeldt-Kaas (1918) assumed that all whitefish colonized Lake Femund via the Swedish River Österdalälven through Lake Vurrusjøen-Drevsjøen, and suggested that these lakes were connected with the southern end of Lake Femund by a historical river course. Huitfeldt-Kaas (1918) assumed that the Trysil-Klarälven River rapids were too steep to allow upstream migration. In contrast, our study suggests that colonization of Lake Femund is likely to have occurred along Trysil-Klarälven River. First, microsatellite data suggest a closer genetic relationship with other populations in this river system than with any other population in neighbouring river systems. Secondly, the fish fauna (Table 1) and also mtDNA data (Bodaly et al., 1991; Bernatchez & Dodson, 1994; K. Østbye, unpublished data) suggest a historical relationship between Lake Femund and other lakes in the Trysil-Klarälven River. In particular, immigration through Lake Vurrusjøen-Lake Drevsjøen seems unlikely, as these lakes harbour fish species, mtDNA haplotypes, and microsatellite alleles not seen in Lake Femund.

The morphological groups found in Lake Femund differed from the nearby Lake Isteren normal and dwarf morph. Lake Isteren normal and dwarf morphs are highly different in body size (R2 = 0.98, P < 0.0001, n = 60, with mean ± SD being 32.6 ± 2.1 vs. 14.5 ± 1.2 cm), and not very different in gill-raker counts (R2 = 0.29, P < 0.0001, n = 60, mean ± SD of 25.5 ± 1.9 vs. 27.9 ± 1.8 gill-rakers). The fact that Lake Femund and Lake Isteren populations differ significantly from each other suggests that the current whitefish morphs in Lake Femund is not a subset (or combination) of populations in other lakes. This is further supported by the fact that all the Lake Femund populations are most phenotypically similar to, but genetically divergent from, the geographically remote Lake Storsjøen. Apparently, the morphological resemblance originates without genetic relationship, implying either parallel selection or phenotypic plasticity.

We cannot exclude that more than one colonization of Lake Femund took place, as suggested by assignment of some individuals of the bay morph to populations in other drainages. However, genetic drift in the numerically small bay populations (Sandlund & Næsje, 1989) could also be responsible for the genetically divergent bay morphs. At any rate, it is likely that the adaptive gene combinations that have produced the four Lake Femund morphs have originated within this lake, although the genes themselves could have arrived from more than one source population. This conclusion corroborates the results of Douglas et al. (1999) who suggested that sympatric whitefish morphs in the central European Alps, being divergent in morphology, spawning site, and spawning time, likely had an intra-lacustrine origin.

Could traits separating Lake Femund morphs be altered by natural selection?

  1. Top of page
  2. Abstract
  3. Introduction
  4. Material and methods
  5. Study area and sampling
  6. Morphology
  7. Life-history
  8. Genetic markers
  9. Data analysis
  10. Results
  11. Morphological divergence of populations in Lake Femund
  12. Life-history divergence of populations in Lake Femund
  13. Genetic differentiation of populations in Lake Femund
  14. Comparison of morphological and genetic divergence in Lake Femund
  15. Morphological relationships of Lake Femund vs. nearby lakes
  16. Genetic relationships of Lake Femund vs. nearby lakes
  17. Discussion
  18. Trait differentiation in Lake Femund compared with other coregonid systems
  19. Are phenotypic traits in Lake Femund morphs associated with use of niches?
  20. Sympatric origin of the divergent morphs in Lake Femund?
  21. Could traits separating Lake Femund morphs be altered by natural selection?
  22. Conclusion
  23. Acknowledgments
  24. References
  25. Appendices

The observed differences of traits that may have an adaptive basis between the four morphs in Lake Femund suggest that their divergence may have been driven by natural selection. Alternatively, differences could reflect (non)adaptive phenotypic plasticity where trait expression is controlled by cues from a given environment (Via & Lande, 1985; Robinson & Pearson, 2002; Price et al., 2003). Analysing several geographical samples of spawning populations within morphs gave us the possibility to compare traits and suggest responsible mechanisms.

High heritability (h2) values for meristic traits, and lower values for metric and life-history traits have commonly been reported (Mousseau & Roff, 1987). Crossing experiments in European whitefish and other fish suggest a high additive genetic variance component of gill-raker numbers with h2-estimates between 0.19 and 0.84, and for metric traits, similar to the ones scored in Lake Femund, with h2-values between 0.21 and 0.84 (Lindroth, 1957; Svärdson, 1957, 1979; Hagen, 1973; Lavin & McPhail, 1987; Leary et al., 1992; Hatfield, 1997; Foote et al., 1999; Hermida et al., 2002). In a study of adaptive head shape differences in cichlids, Albertson et al. (2003) suggested that head traits may collectively be inherited together because of pleiotropy in the genetic architecture. Further, Schluter et al. (2004) observed parallel inheritance of genetic differences in adaptive traits in two independent lineages of threspine sticklebacks, suggesting that an ancestral trait in close relatives would follow the same developmental pathway when exposed to similar selection pressures. These results suggest an additive genetic basis for the analysed traits in Lake Femund implying they could be structured by selection.

Indeed, the Qst–Fst results suggest that natural selection have shaped divergence of morphs in Lake Femund, especially in traits related to foraging. Using a Qst/Fst ratio, we found that the between-morph comparison of upper and lower gill-raker numbers were 5.7–21.1 and 7.6–23.7, respectively. These ratios are much higher than those reported by Bernatchez (2004) for North American lake whitefish sympatric pairs, which suggests more pronounced selective pressures exerted on gill-rakers in Lake Femund. Thus, if natural selection has been important in driving the differentiation of morphs, we should observe a higher degree of differentiation between spawning morphs than within morphs reflecting directional selection. In addition, we should observe stabilizing selection within populations of the same morph as a manifestation of phenotypic optimum. This pattern seems in general to hold true, especially for traits related to the gill-raker apparatus which were much more different among morphs than within. Similar observations were made in comparative analysis of sympatric morphs in smelt (Osmerus mordax) (Saint-Laurent et al., 2003). The most distinct morph with regard to these traits are the bay morph which have the highest number of gill-rakers, longest rakers and the most dense gill-raker apparatus. These are also the two populations (Gløtfossen and Kvernvika), when compared within morphs, that have the highest number of Qst trait values below Fst, suggesting stabilizing selection may be involved in maintaining similarity at these traits between populations of the same morph.

As an alternative to a selection-mediated divergence, phenotypic plasticity could be induced by environmental cues such as temperature (Tåning, 1958). However, such effects on gill-raker numbers is at best small (Todd, 1998). On body-proportion traits, Svärdson (1950) demonstrated an effect of phenotypic plasticity through growth-rate related change in metric values. In the threespined stickleback, Day et al. (1994) showed that plasticity accounted for 0% (gill-raker number) up to 58% (head depth) when comparing five foraging-related traits in two morphs that were fed the natural diet of the other morph. Such phenotypic placticity induced by feeding experience has also been reported in other fish (Meyer, 1987; Wimberger, 1992; Skúlason et al., 1999; Alexander & Adams, 2004).

However, the most divergent traits between the four morphs in Lake Femund, gill-raker number and length, show high heritability in many studies, and are associated with foraging ability implying that differentiation of these four morphs are likely to have been driven by natural selection.

Conclusion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Material and methods
  5. Study area and sampling
  6. Morphology
  7. Life-history
  8. Genetic markers
  9. Data analysis
  10. Results
  11. Morphological divergence of populations in Lake Femund
  12. Life-history divergence of populations in Lake Femund
  13. Genetic differentiation of populations in Lake Femund
  14. Comparison of morphological and genetic divergence in Lake Femund
  15. Morphological relationships of Lake Femund vs. nearby lakes
  16. Genetic relationships of Lake Femund vs. nearby lakes
  17. Discussion
  18. Trait differentiation in Lake Femund compared with other coregonid systems
  19. Are phenotypic traits in Lake Femund morphs associated with use of niches?
  20. Sympatric origin of the divergent morphs in Lake Femund?
  21. Could traits separating Lake Femund morphs be altered by natural selection?
  22. Conclusion
  23. Acknowledgments
  24. References
  25. Appendices

The European whitefish in Lake Femund feature four morphs, with distinct biology, being most divergent in traits related to foraging. Two or more populations exist within each of the four morphs being phenotypically similar despite the micro-geographical separation. Genetic divergence between morphs is moderate to large, with values being generally small between populations within morphs. A higher proportion of molecular variance is also found between than within morphs. The evolutionary origin of the four morphs seems to be intra-lacustrine, where divergent selection has been sufficiently strong to maintain phenotypic differentiation in spite of gene flow. Apparently, the adaptive gene combinations that have produced the four Lake Femund morphs have originated within this lake, although the genes themselves could have arrived from more than one source population outside the lake.

Acknowledgments

  1. Top of page
  2. Abstract
  3. Introduction
  4. Material and methods
  5. Study area and sampling
  6. Morphology
  7. Life-history
  8. Genetic markers
  9. Data analysis
  10. Results
  11. Morphological divergence of populations in Lake Femund
  12. Life-history divergence of populations in Lake Femund
  13. Genetic differentiation of populations in Lake Femund
  14. Comparison of morphological and genetic divergence in Lake Femund
  15. Morphological relationships of Lake Femund vs. nearby lakes
  16. Genetic relationships of Lake Femund vs. nearby lakes
  17. Discussion
  18. Trait differentiation in Lake Femund compared with other coregonid systems
  19. Are phenotypic traits in Lake Femund morphs associated with use of niches?
  20. Sympatric origin of the divergent morphs in Lake Femund?
  21. Could traits separating Lake Femund morphs be altered by natural selection?
  22. Conclusion
  23. Acknowledgments
  24. References
  25. Appendices

This work was funded by the Research Council of Norway, the Norwegian Directorate for Nature Management, and by our host institutions. We are grateful to O. Elgåen from Femund Fiskerlag for collecting the samples. The following persons were invaluable during laboratory work; T. Balstad, K. Kvaløy, C. Landry, L. Papillon, M. Valcourt, R. Saint-Laurent and C. Potvin. Fruitful discussions with J. Blais, G. Lu, O. Ugedal, D. Garant, R. Saint-Laurent, D. Fraser, M. Schulz, V. Castric, J. Turgeon, E. Østbye, and constructive comments from O. Seehausen, A. Vøllestad, and an anonymous referee greatly improved the manuscript.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Material and methods
  5. Study area and sampling
  6. Morphology
  7. Life-history
  8. Genetic markers
  9. Data analysis
  10. Results
  11. Morphological divergence of populations in Lake Femund
  12. Life-history divergence of populations in Lake Femund
  13. Genetic differentiation of populations in Lake Femund
  14. Comparison of morphological and genetic divergence in Lake Femund
  15. Morphological relationships of Lake Femund vs. nearby lakes
  16. Genetic relationships of Lake Femund vs. nearby lakes
  17. Discussion
  18. Trait differentiation in Lake Femund compared with other coregonid systems
  19. Are phenotypic traits in Lake Femund morphs associated with use of niches?
  20. Sympatric origin of the divergent morphs in Lake Femund?
  21. Could traits separating Lake Femund morphs be altered by natural selection?
  22. Conclusion
  23. Acknowledgments
  24. References
  25. Appendices
  • Adams, C.E. & Huntingford, F.A. 2004. Incipient speciation driven by phenotypic plasticity? Evidence from sympatric populations of Arctic charr. Biol. J. Linn. Soc. 81: 611618.
  • Albertson, R.C., Streelman, J.T. & Kocher, T.D. 2003. Genetic basis of adaptive shape differences in the cichlid head. J. Heredity 94: 291301.
  • Alexander, G.D. & Adams, C.E. 2004. Exposure to a common environment erodes inherited between-population trophic morphology differences in Arctic charr. J. Fish Biol. 64: 253257.
  • Amundsen, P.-A., Knudsen, R., Klemetsen, A. & Kristoffersen, R. 2004. Resource competition and interactive segregation between sympatric whitefish morphs. Ann. Zol. Fenn. 41: 301307.
  • AppliedBiosystems. 1996a. GENESCAN, Software, Version 2.1. User's Manual. ABI, Foster city.
  • AppliedBiosystems. 1996b. GENOTYPER, Software, Version 2.0. User's Manual. ABI, Foster city.
  • Bagenal, T.B. & Tesch, F.W. 1978. Age and growth. In: Methods for assessment of fish production in fresh waters, IBP handbook No. 3 (T. B.Bagenal, ed.), pp. 101130. Blackwell Scientific Publishers, Oxford.
  • Bentzen, P. & McPhail, J.D. 1984. Ecology and evolution of sympatric sticklebacks (Gasterosteus): specialization for alternative trophic niches in the Enos Lake species pair. Can. J. Zool. 62: 22802286.
  • Berg, L.S. 1962. Freshwater Fishes of the USSR and Adjacent Countries, 4th edn. Israel Program for Scientific Translations, Jerusalem.
  • Bergstrand, E. 1982. The diet of four sympatric whitefish species in Lake Parkijaure. Rep. Inst. Freshw. Res., Drottningholm 60: 514.
  • Bernatchez, L. 1995. A role for molecular systematics in defining evolutionary significant units in fishes. Am. Fish. Soc. Symp. 17: 114132.
  • Bernatchez, L. 2004. Ecological theory of adaptive radiation: an empirical assessment from Coregonine fishes (Salmoniformes). In: Evolution Illuminated: Salmon and their Relatives (A. P.Hendry & S. C.Stearns, eds), pp. 175207. Oxford University Press, Oxford.
  • Bernatchez, L. & Dodson, J.J. 1994. Phylogenetic relationships among palearctic and nearctic whitefish (Coregonus sp.) populations as revealed by mitochondrial DNA variation. Can. J. Fish. Aquat. Sci. 51: 240251.
  • Bodaly, R.A. 1979. Morphological and ecological divergence within the lake whitefish (Coregonus clupeaformis) species complex in Yukon Territory. J. Fish. Res. Bd. Canada 36: 12141222.
  • Bodaly, R.A., Vuorinen, J., Wards, R.D., Luczynski, M. & Reist, J.D. 1991. Genetic comparisons of New and Old World coregonid fishes. J. Fish Biol. 38: 3751.
  • Bodaly, R.A., Clayton, J.W., Lindsay, C.G. & Vuorinen, J. 1992. Evolution of lake whitefish (Coregonus clupeaformis) in North America during the Pleistocene: genetic differentiation between sympatric populations. Can. J. Fish. Aquat. Sci. 49: 769779.
  • Bronte, C.R., Fleischer, G.W., Maistrenko, S.G. & Pronin, N.M. 1999. Stock structure of Lake Baikal omul as determined by whole-body morphology. J. Fish Biol. 54: 787798.
  • Campbell, D. & Bernatchez, L. 2004. Generic scan using AFLP markers as a means to assess the role of directional selection in the divergence of sympatric whitefish ecotypes. Mol. Ecol. Evol. 21: 945956.
  • Cornuet, J.-M., Piry, S., Luikart, G., Estoup, A. & Solignac, M. 1999. New methods employing multilocus genotypes to select or exclude populations as origins of individuals. Genetics 153: 19892000.
  • Day, T. & McPhail, J.D. 1996. The effect of behavioural and morphological plasticity on foraging efficiency in the threespined stickleback. Oecologia 108: 380388.
  • Day, T., Pritchard, J. & Schluter, D. 1994. Ecology and genetics of phenotypic plasticity: a comparison of two sticklebacks. Evolution 48: 17231734.
  • Dill, L.M. 1983. Adaptive flexibility in the foraging behaviour of fishes. Can. J. Fish. Aquat. Sci. 40: 398408.
  • Douglas, M.R., Brunner, P.C. & Bernatchez, L. 1999. Do assemblages of Coregonus (Teleostei: Salmoniformes) in the central Alpine region of Europe represent species flocks? Mol. Ecol. 8: 589603.
  • El Mousadik, A. & Petit, R.J. 1996. High level of genetic differentiation for allele richness among populations of the argan tree [Argania spinosa (L.) Skeels] endemic to Morocco. Theor. Appl. Genet. 92: 823839.
  • Excoffier, L., Smouse, P.E. & Quattro, J.M. 1992. Analysis of molecular variance inferred from metric distances among DNA haplotypes: application to human mitochondrial DNA restriction data. Genetics 131: 479491.
  • Felsenstein, J. & Sober, E. 1986. Parsimony and likelihood: an exchange. Syst. Zool. 35: 617626.
  • Foote, C.J., Moore, K., Stenberg, K., Craig, K.J., Wenburg, J.K. & Wood, C.C. 1999. Genetic differentiation in gill raker number and length in sympatric anadromous and nonanadromous morphs of sockeye salmon, Oncorhynchus nerka. Environ. Biol. Fish. 54: 263274.
  • Goldstein, D.B. & Schlötterer, C. 1999. Microsatellites: Evolution and Applications, 1st edn. Oxford University Press, Oxford.
  • Goudet, J. 1995. FSTAT (vers. 1.2): a computer program to calculate F-statistics. J. Hered. 86: 485486.
  • Goudet, J. 1999. PCA-GEN for Windows, vers. 1.2 [WWW document]. URL http://www2.unil.ch/popgen/softwares/pcagen.html.
  • Guo, S.W. & Thompson, E.A. 1992. Performing the exact test of Hardy–Weinberg proportion for multiple alleles. Biometrics 48: 361372.
  • Hagen, D.W. 1973. Inheritance of numbers of lateral plates and gill rakers in Gasterosteus aculeatus. Heredity 30: 303312.
  • Haldane, J.B.S. 1948. The theory of a cline. J. Genet. 48: 277283.
  • Hatfield, T. 1997. Genetic divergence in adaptive characters between sympatric species of stickleback. Am. Nat. 149: 10091029.
  • Hendry, A.P. & Stearns, S.C. 2004. Evolution Illuminated: Salmon and their Relatives. Oxford University Press, Oxford.
  • Hendry, A.P., Taylor, E.B. & McPhail, J.D. 2002. Adaptive divergence and the balance between selection and gene flow: lake and stream stickleback in the Misty system. Evolution 56: 11991216.
  • Hermida, M., Fernández, C., Amaro, R. & San Miguel, E. 2002. Heritability and ‘evolvability’ of meristic characters in a natural population of Gasterosteus aculeatus. Can. J. Zool. 80: 532541.
  • Himberg, M. & Lehtonen, H. 1995. Systematic and nomenclature of coregonid fishes. Particularly in Northwest Europe. Arch. Hydrobiol. Spec. Issues. Advanc. Limnol. 46: 3947.
  • Hindar, K. 1994. Alternative life histories and genetic conservation. In: Conservation Genetics (V.Loeschcke, J.Tomiuk & S. K.Jain, eds), pp. 323336. Birkhäuser Verlag Basel, Switzerland.
  • Huitfeldt-Kaas, H. 1918. Ferskvandsfiskenes utbredelse og innvandring i Norge. Centraltrykkeriet, Kristiania [in Norwegian].
  • Humphries, C.J. & Parenti, L.R. 1999. Cladistic Biogeography: Interpreting Patterns of Plant and Animal Distributions, 2nd edn. Oxford University Press, Oxford.
  • Jensen, A.L. 1985. Relations among net reproductive rate and life history parameters for lake whitefish (Coregonus clupeaformis). Can. J. Fish. Aquat. Sci. 42: 164168.
  • Jonsson, B. & Jonsson, N. 2001. Polymorphism and speciation in Arctic charr. J. Fish Biol. 58: 605638.
  • Kahilainen, K. & Lehtonen, H. 2002. Habitat use and growth of three sympatric forms of European whitefish, Coregonus lavaretus (L.), in subarctic Lake Muddusjärvi. Arch. Hydrobiol. Spec. Issues. Advanc. Limnol. 57: 277290.
  • Kahilainen, K., Malinen, T., Tuomaala, A. & Lehtonen, H. 2004. Diel and seasonal habitat and food segregation of three sympatric Coregonus lavaretus forms in a subarctic lake. J. Fish Biol. 64: 117.
  • Kirchhofer, A. 1990. Limnologische und ichtyologische untersuchungen im Brienzersee unter besonderer berücsichtigung der differenzierung der sympatrischen felchenpopulationen, Dissertation. Phil. – nat. Fakultät der Universität Bern.
  • Kirchhofer, A. & Tschumi, P.-A. 1986. Age structure and growth of coregonid fish populations in Lake Thun. Arch. Hydrobiol. Beih. Ergebn. Limnol. 22: 303318.
  • Kirkpatrick, M. & Selander, R.K. 1979. Genetics of speciation in lake whitefishes in the Allegash basin. Evolution 33: 478485.
  • Klemetsen, A., Elliot, J.M., Knudsen, R. & Sørensen, P. 2002. Evidence for genetic differences in the offspring of two sympatric morphs of Arctic charr Salvelinus alpinus (L.). J. Fish Biol. 60: 933950.
  • Kottelat, M. 1997. European freshwater fishes. An heuristic checklist of the freshwater fishes of Europe (exclusive of former USSR), with an introduction for non-systematists and comments on nomenclature and conservation. Biol. Sec. Zool. 52: 1271.
  • Løvik, J.E. & Kjellberg, G. 1982. Glåma i Hedmark. Delrapport om dyreplankton. Undersøkelser i tidsrommet 1978–80. Norwegian Institute of Water Research, Report No. 0-78045-III, pp. 158 [in Norwegian].
  • Larson, G.L. 1976. Social behavior and feeding ability of two phenotypes of Gasterosteus aculeatus in relation to their spatial and trophic segregation in a temperate lake. Can. J. Zool. 54: 107121.
  • Lavin, P.A & McPhail, J.D. 1987. Morphological divergence and the organization of trophic characters among lacustrine populations of the threespine stickleback (Gasterosteus aculeatus). Can. J. Fish. Aquat. Sci. 44: 18201829.
  • Leary, R.F., Allendorf, F.W. & Knudsen, K.L. 1992. Genetic, environmental, and developmental causes of meristic variation in rainbow trout. Act. Zool. Fenn. 191: 7995.
  • Lehtonen, H. & Niemelä, E. 1998. Growth and population structure of whitefish (Coregonus lavaretus (L.)) in mountain lakes of Finland. Arch. Hydrobiol. Spec. Issues. Advanc. Limnol. 49: 8195.
  • Li, W.-H. 1997. Molecular Evolution, 2nd edn. Sinauer Associates, Incorporate Publishers, Sunderland, MA, USA.
  • Lindroth, A. 1957. A study of the whitefish (Coregonus) of the Sundsvall Bay district. Rep. Inst. Freshw. Res., Drottningholm 38: 70108.
  • Linnaeus, C. 1758. Systema Naturae, 10th edn. Impensis direct. Laurentii salvii, Stockholm.
  • Liou, L.W. & Price, T.D. 1994. Speciation by reinforcement of premating isolation. Evolution 48: 14511459.
  • Lópes-Fanjul, C., Fernández, A. & Toro, M.A. 2003. The effect of neutral nonadditive gene action on the quantitative index of population divergence. Genetics 164: 16271633.
  • Losos, J.B., Jackman, T.R., Larson, A., de Queiroz, K. & Rodriguez-Schettino, L. 1998. Contingency and determinism in replicated adaptive radiations of island lizards. Science 279: 21152118.
  • Lu, G. & Bernatchez, L. 1999. Correlated trophic speciation and genetic divergence in sympatric lake whitefish ecotypes (Coregonus clupeaformis): support for the ecological speciation hypothesis. Evolution 53: 14911505.
  • Lu, G., Basley, D.J. & Bernatchez, L. 2001. Contrasting patterns of mitochondrial DNA and microsatellite introgressive hybridization between lineages of lake whitefish (Coregonus clupeaformis): relevance for speciation. Mol. Ecol. 10: 965985.
  • Malmquist, H.J. 1992. Phenotype-specific feeding behaviour of two arctic charr Salvelinus alpinus morphs. Oecologia 92: 354361.
  • Mayr, E. 2001. What Evolution is, 1st edn. Basic books, New York.
  • Merilä, J. & Crnokrak, P. 2001. Comparison of genetic differentiation at marker loci and quantitative traits. J. Evol. Biol. 14: 892903.
  • Merilä, J., Björklund, M. & Baker, A.J. 1997. Historical demography and present day population structure of the greenfinch, Carduelis chloris. Evolution 51: 946956.
  • Meyer, A. 1987. Phenotypic plasticity and heterochrony in Cichlasoma managuense (Pisces, Cichlidae) and their implications for speciation in cichlid fishes. Evolution 41: 13571369.
  • Mousseau, T.A. & Roff, D.A. 1987. Natural selection and the heritability of fitness components. Heredity 59: 181197.
  • Næsje, T.F., Sandlund, O.T. & Saksgård, R. 1992. Siken i Femund: effekter og anbefalinger etter ti års næringsfiske. NINA-oppdragsmelding 145: 24 (ISBN-82-426-0258-1) [In Norwegian with English summary].
  • Næsje, T.F., Sandlund, O.T. & Saksgård, R. 1998. Selective predation of piscivorous brown trout (Salmo trutta L.) on polymorphic whitefish (Coregonus lavaretus L.). Arch. Hydrobiol. Spec. Issues. Advanc. Limnol. 50: 283294.
  • Nei, M. 1987. Molecular Evolutionary Genetics. Columbia University Press, New York, USA.
  • Nei, M., Tajima, F. & Tateno, Y. 1983. Accuracy of estimated phylogenetic trees from molecular data. II. Gene frequency data. J. Mol. Evol. 19: 153170.
  • Nielsen, J.L. & Powers, D. (eds.) 1995. Evolution and the aquatic ecosystem: defining unique units in population conservation. Am. Fish. Soc. Symp. 17: 1435.
  • Nikolskii, G.V. 1963. The Ecology of Fishes, 1st edn. Academic press, New York.
  • Patton, J.C., Gallaway, B.J., Fechhelm, R.G. & Cronin, M.A. 1997. Genetic variation of microsatellite and mitochondrial DNA markers in broad whitefish (Coregonus nasus) in the Colville and Sagavanirktok rivers in northern Alaska. Can. J. Fish. Aquat. Sci. 54: 15481556.
  • Pigeon, D., Chouinard, A. & Bernatchez, L. 1997. Multiple modes of speciation involved in the parallel evolution of sympatric morphotypes of lake whitefish (Coregonus clupeaformis, Salmonidae). Evolution 51: 196205.
  • Price, T.D., Qvarnström, A. & Irwin, D.E. 2003. The role of phenotypic plasticity in driving genetic evolution. Proc. R. Soc. Lond. B 270: 14331440.
  • Raymond, M. & Rousset, F. 1995. GENEPOP version 1.2: population genetics software for exact test and ecumenism. J. Heredity 86: 248249.
  • Reist, J.D. 1985. An empirical evaluation of several univariate methods that adjust for size variation in morphometric data. Can. J. Zool. 63: 14291439.
  • Resethnikov, Y.S. 1968. Coregonid fishes in recent conditions. Finn. Fish. Res. 9: 1116.
  • Rice, W.R. & Hostert, E.E. 1993. Laboratory experiments on speciation: what have we learned in 40 years? Evolution 47: 16371653.
  • Robinson, B.W. & Pearson, K.J. 2002. Changing times, spaces, and faces: tests and implications of adaptive morphological plasticity in the fishes of northern postglacial lakes. Can. J. Fish. Aquat. Sci. 59: 18191833.
  • Robinson, B.W. & Wilson, D.S. 1994. Character release and displacement in fishes: a neglected literature. Am. Nat. 144: 596627.
  • Rogers, S.M., Gagnon, V. & Bernatchez, L. 2002. Genetically based phenotype-environment association for swimming behavior in lake whitefish ecotypes (Coregonus clupeaformis Mitchill). Evolution 56: 23222329.
  • Rogers, M.R., Marchand, M. & Bernatchez, L. 2004. Isolation, characterization, and cross-salmonid amplification of 31 microsatellite loci in the lake whitefish (Coregonus clupeaformis, Mitchill). Mol. Ecol. Notes 4: 8992.
  • Rufli, H.V. 1978. Die heutigen sympatrischen felchenpopulationen (Coregonus spp.) des Thuner- und Bielersees und ihre morphologie. Schweiz. Z. Hydrol. 40: 731.
  • Saint-Laurent, R., Legault, M. & Bernatchez, L. 2003. Divergent selection maintains adaptive differentiation despite high gene flow between sympatric rainbow smelt ecotypes (Osmerus mordax Mitchill). Mol. Ecol. 12: 315330.
  • Saksgård, R., Næsje, T.F., Sandlund, O.T. & Ugedal, O. 2002. The effect of fish predators on European whitefish (Coregonus lavaretus L.) habitat use in Lake Femund, a deep Norwegian lake. Arch. Hydrobiol. Spec. Issues Advanc. Limnol. 57: 537552.
  • Sambrook, J., Fritch, E.F. & Maniatis, T. 1989. Molecular Cloning: A Laboratory Manual, 2nd edn. Cold Spring Harbour Laboratory Press, New York.
  • Sandlund, O.T. & Næsje, T.F. 1989. Impact of a pelagic gill-net fishery on the polymorphic whitefish (Coregonus lavaretus L. s.l.) population in Lake Femund, Norway. Fish Res. 7: 8597.
  • Sandlund, O.T., Næsje, T.F. & Saksgård, R. 1995. Ecological diversity in whitefish Coregonus lavaretus: ontogenetic niche shifts and polymorphism. Arch. Hydrobiol. Spec. Issues Advanc. Limnol. 46: 4959.
  • SAS. 2002. JMP Software, JMP®, Version 5.0. Copyright © 2002. SAS Institute Inc., Cary, NC, USA.
  • Schliewen, U.K, Tautz, D. & Paabo, S. 1994. Sympatric speciation suggested by monophyly of crater lake cichlids. Nature 368: 629632.
  • Schliewen, U.K., Rassman, K., Markmann, M., Market, J., Kocher, T. & Tautz, D. 2001. Genetic and ecological divergence of a monophyletic cichlid species pair under fully sympatric conditions in Lake Ejegham, Cameroon. Mol. Ecol. 10: 14711488.
  • Schluter, D. 2000. The Ecology of Adaptive Radiation, 1st edn. Oxford University Press, New York.
  • Schluter, D. 2004. Frequency dependent natural selection during character displacement in sticklebacks. Evolution 57: 11421150.
  • Schluter, D., Clifford, E.A., Nemethy, M. & McKinnon, J.S. 2004. Parallel evolution of inheritance of quantitative traits. Am. Nat. 163: 809822.
  • Schneider, S., Roessli, D. & Excoffier, L. 2000. ARLEQUIN v. 2.000; A Software for Population Genetic Data Analysis. Genetics and Biometry Laboratory, University of Geneva, Switzerland.
  • Schulz, M. & Freyhof, J. 2003. Coregonus fontanae, a new spring-spawning cisco from Lake Stechlin, northern Germany (Salmoniformes: Coregonidae). Ichtyol. Explor. Freshwaters 14: 209216.
  • Skúlason, S. & Smith, T.B. 1995. Resource polymorphisms in vertebrates. TREE 10: 366370.
  • Skúlason, S., Snorrason, S.S. & Jónsson, B. 1999. Sympatric morphs, populations and speciation in freshwater fish with emphasis on Arctic charr. In: Evolution of Biological Diversity (A. E.Magurran & R.May, eds), pp. 7073. Oxford University Press, Oxford.
  • Skurdal, J., Vøllestad, L.A. & Quenild, T. 1985. Comparison of scales and otoliths for age determination of whitefish Coregonus lavaretus. Fish. Res. 3: 237243.
  • Slatkin, M. 1973. Gene flow and selection in a cline. Genetics 75: 733756.
  • Smirnov, V.V. 1992. Intraspecific structure of Baikal Omul, Coregonus autumnalis migratorius (Georgi). Pol. Arch. Hydrobiol. 39: 325333.
  • Smith, G.R. & Todd, T.N. 1984. Evolution of species flocks of fishes in north temperate lakes. In: Evolution of Fish Species Flocks (A. A.Echelle & I.Kornfield, eds), pp. 4568. University of Maine at Orono Press, Orono.
  • Spitze, K. 1993. Population structure in Daphnia obtusa: quantitative genetic and allozymic variation. Genetics 135: 367374.
  • Steinmann, P. 1950a. Monographie der Schweizerischen Koregonen. Beitrag zum problem der entstehung neur arten. Einleitung. Schweiz. Z. Hydrol. 12: 109189.
  • Steinmann, P. 1950b. Monographie der Schweizerischen Koregonen. Beitrag zum problem der entstehung neur arten. Spezieller teil. Schweiz. Z. Hydrol. 12: 340491.
  • Steinmann, P. 1951. Monographie der Schweizerischen Koregonen. Beitrag zum problem der entstehung neur arten. Spezieller teil (2nd part). Schweiz. Z. Hydrol. 13: 54191.
  • Svärdson, G. 1950. The coregonid problem: II. Morphology of two coregonid species in different environments. Rep. Inst. Freshw. Res., Drottningholm 31: 151162.
  • Svärdson, G. 1957. The coregonid problem. VI. The Palearctic species and their intergrades. Rep. Inst. Freshw. Res., Drottningholm 38: 267356.
  • Svärdson, G. 1979. Speciation of Scandinavian Coregonus. Rep. Inst. Freshw. Res., Drottningholm 57: 195.
  • Svärdson, G. 1998. Postglacial dispersal and reticulate evolution of nordic coregonids. Nordic J. Freshw. Res. 74: 332.
  • Svarvar, P.-O. & Müller, R. 1982. Die felchen des Alpnachersees. Schweiz. Z. Hydrol. 44: 295314.
  • Tåning, V. 1958. Experimental study of meristic characters in fishes. Biol. Rev. 27: 169193.
  • Taylor, E.B. 1999. Species pairs of north temperate freshwater fishes: taxonomy, evolution and conservation. Rev. Fish Biol. Fish. 9: 299324.
  • Taylor, E. & McPhail, J. 2000. Historical contingency and ecological determinism interact to prime speciation in sticklebacks, Gasterosteus aculeatus. Proc. Roy. Soc. Lond. B 267: 23752384.
  • Todd, T.N. 1998. Environmental modification of gillraker number in coregonine fishes. Arch. Hydrobiol. Spec. Issues. Advanc. Limnol. 50: 305315.
  • Todd, T.N. & Smith, G.R. 1992. A review of differentiation in Great Lakes ciscoes. Pol. Arch. Hydrobiol. 39: 261267.
  • Turelli, M., Barton, N. & Coyne, J.A. 2001. Theory and speciation. TREE 16: 330343.
  • Turgeon, J., Estoup, A. & Bernatchez, L. 1999. Species flock in the North American Great Lakes: Molecular ecology of Lake Nipigon ciscoes (Teleostei: Coregonidae: Coregonus). Evolution 53: 18571871.
  • Via, S. & Lande, R. 1985. Genotype–environment interaction and the evolution of phenotypic plasticity. Evolution 39: 505522.
  • Voloshenko, B.B. 1973. A comparative analysis of the feeding of underyearlings of the Pelyad [Coregonus peled (Gmelin)], the Broad whitefish [Coregonus nasus (Pallas)] and their hybrids when reared together. J. Ichtyol. 13: 569576.
  • Wimberger, P.H. 1992. Plasticity of fish body shape. The effects of diet, development, family and age in two species of Geophagus (Pisces; Cichlidae). Biol. J. Linn. Soc. 45: 197218.
  • Wright, S. 1951. The gentical structure of populations. Ann. Eugenics 15: 323354.

Appendices

  1. Top of page
  2. Abstract
  3. Introduction
  4. Material and methods
  5. Study area and sampling
  6. Morphology
  7. Life-history
  8. Genetic markers
  9. Data analysis
  10. Results
  11. Morphological divergence of populations in Lake Femund
  12. Life-history divergence of populations in Lake Femund
  13. Genetic differentiation of populations in Lake Femund
  14. Comparison of morphological and genetic divergence in Lake Femund
  15. Morphological relationships of Lake Femund vs. nearby lakes
  16. Genetic relationships of Lake Femund vs. nearby lakes
  17. Discussion
  18. Trait differentiation in Lake Femund compared with other coregonid systems
  19. Are phenotypic traits in Lake Femund morphs associated with use of niches?
  20. Sympatric origin of the divergent morphs in Lake Femund?
  21. Could traits separating Lake Femund morphs be altered by natural selection?
  22. Conclusion
  23. Acknowledgments
  24. References
  25. Appendices
Table Appendix1.  The mean (SD) values of log10 transformed and allometrically scaled traits in all the 18 surveyed populations.
Code1.2.3.4.5.6.7.8.9.10.11.12.13.14.15.16.17.18.R2P-value
  1. The population and trait codes are found in Tables 1–3 and Figs 1,2. In last two columns test statistics for univariate anova is given.

HEL1.795 (0.024)1.784 (0.011)1.789 (0.019)1.789 (0.015)1.781 (0.013)1.789 (0.014)1.786 (0.011)1.785 (0.012)1.790 (0.011)1.780 (0.012)1.786 (0.011)1.828 (0.016)1.775 (0.013)1.794 (0.022)1.799 (0.023)1.805 (0.014)1.804 (0.013)1.774 (0.013)0.33<0.0001
HEH1.418 (0.016)1.418 (0.014)1.424 (0.014)1.415 (0.016)1.433 (0.024)1.427 (0.017)1.432 (0.013)1.435 (0.012)1.432 (0.011)1.419 (0.017)1.430 (0.012)1.329 (0.020)1.415 (0.010)1.416 (0.017)1.405 (0.017)1.395 (0.022)1.429 (0.012)1.390 (0.016)0.65<0.0001
PRL1.206 (0.019)1.196 (0.020)1.189 (0.018)1.204 (0.023)1.220 (0.022)1.201 (0.023)1.215 (0.018)1.219 (0.021)1.205 (0.027)1.224 (0.026)1.234 (0.020)1.041 (0.039)1.181 (0.016)1.174 (0.022)1.183 (0.030)1.175 (0.024)1.195 (0.016)1.129 (0.026)0.74<0.0001
ORL1.234 (0.017)1.229 (0.016)1.236 (0.016)1.227 (0.019)1.241 (0.019)1.237 (0.017)1.231 (0.017)1.230 (0.014)1.239 (0.015)1.230 (0.021)1.233 (0.015)1.197 (0.025)1.249 (0.017)1.244 (0.015)1.229 (0.020)1.236 (0.029)1.259 (0.014)1.219 (0.021)0.26<0.0001
ORH1.157 (0.018)1.156 (0.026)1.166 (0.021)1.147 (0.020)1.171 (0.024)1.169 (0.016)1.158 (0.015)1.174 (0.019)1.175 (0.019)1.149 (0.015)1.156 (0.016)1.106 (0.023)1.184 (0.017)1.167 (0.016)1.157 (0.018)1.167 (0.029)1.170 (0.025)1.120 (0.018)0.41<0.0001
POL1.483 (0.003)1.486 (0.002)1.494 (0.001)1.490 (0.003)1.502 (0.002)1.501 (0.002)1.506 (0.002)1.498 (0.001)1.504 (0.002)1.514 (0.001)1.509 (0.001)1.503 (0.002)1.416 (0.003)1.484 (0.002)1.473 (0.003)1.462 (0.002)1.493 (0.002)1.468 (0.002)0.71<0.0001
UPL1.227 (0.019)1.223 (0.024)1.216 (0.022)1.220 (0.015)1.238 (0.025)1.237 (0.015)1.243 (0.015)1.237 (0.016)1.235 (0.019)1.227 (0.025)1.262 (0.011)1.098 (0.023)1.211 (0.018)1.223 (0.019)1.207 (0.028)1.203 (0.025)1.227 (0.011)1.160 (0.018)0.71<0.0001
LOL1.365 (0.014)1.360 (0.014)1.362 (0.018)1.352 (0.012)1.371 (0.020)1.365 (0.014)1.386 (0.011)1.388 (0.014)1.384 (0.013)1.379 (0.016)1.405 (0.016)1.310 (0.015)1.372 (0.015)1.367 (0.022)1.366 (0.019)1.371 (0.024)1.374 (0.017)1.340 (0.016)0.56<0.0001
INB1.240 (0.021)1.244 (0.021)1.254 (0.016)1.254 (0.019)1.262 (0.022)1.266 (0.013)1.268 (0.012)1.267 (0.020)1.274 (0.014)1.261 (0.023)1.270 (0.014)1.149 (0.034)1.248 (0.018)1.245 (0.022)1.244 (0.026)1.223 (0.029)1.281 (0.015)1.208 (0.022)0.63<0.0001
SNW1.044 (0.022)1.055 (0.022)1.048 (0.022)1.042 (0.023)1.076 (0.031)1.052 (0.021)1.063 (0.022)1.063 (0.022)1.066 (0.019)1.083 (0.027)1.066 (0.017)0.858 (0.030)1.060 (0.022)1.024 (0.023)0.997 (0.030)0.995 (0.030)1.066 (0.021)0.981 (0.023)0.79<0.0001
SNH0.861 (0.022)0.863 (0.016)0.861 (0.020)0.861 (0.020)0.892 (0.032)0.872 (0.032)0.857 (0.021)0.863 (0.025)0.854 (0.025)0.879 (0.030)0.883 (0.024)0.631 (0.036)0.836 (0.022)0.841 (0.036)0.814 (0.030)0.786 (0.043)0.871 (0.025)0.772 (0.029)0.78<0.0001
UPW1.191 (0.027)1.192 (0.024)1.175 (0.024)1.182 (0.022)1.205 (0.026)1.191 (0.016)1.212 (0.024)1.189 (0.017)1.180 (0.022)1.207 (0.023)1.215 (0.017)1.015 (0.040)1.188 (0.026)1.173 (0.022)1.146 (0.032)1.126 (0.035)1.169 (0.024)1.104 (0.022)0.74<0.0001
UGL1.150 (0.014)1.135 (0.021)1.138 (0.020)1.148 (0.020)1.173 (0.027)1.159 (0.024)1.170 (0.020)1.146 (0.019)1.156 (0.025)1.160 (0.021)1.150 (0.015)1.019 (0.037)1.186 (0.016)1.128 (0.022)1.107 (0.025)1.108 (0.029)1.162 (0.021)1.104 (0.016)0.68<0.0001
LGL1.366 (0.014)1.366 (0.016)1.371 (0.015)1.371 (0.014)1.388 (0.018)1.380 (0.016)1.374 (0.014)1.374 (0.015)1.372 (0.019)1.367 (0.019)1.360 (0.012)1.251 (0.028)1.379 (0.015)1.357 (0.016)1.334 (0.022)1.346 (0.020)1.369 (0.016)1.332 (0.017)0.71<0.0001
GIL0.758 (0.060)0.723 (0.067)0.729 (0.055)0.732 (0.071)0.768 (0.047)0.773 (0.058)0.917 (0.031)0.897 (0.050)0.891 (0.047)0.970 (0.037)0.962 (0.043)0.396 (0.063)0.666 (0.046)0.796 (0.048)0.754 (0.056)0.717 (0.076)0.878 (0.045)0.621 (0.050)0.84<0.0001
INL1.475 (0.029)1.462 (0.025)1.474 (0.021)1.465 (0.031)1.488 (0.031)1.485 (0.030)1.496 (0.027)1.484 (0.020)1.506 (0.018)1.493 (0.030)1.480 (0.015)1.359 (0.027)1.510 (0.023)1.457 (0.038)1.441 (0.037)1.433 (0.035)1.492 (0.017)1.443 (0.028)0.56<0.0001
ULO0.529 (0.062)0.388 (0.111)0.503 (0.104)0.540 (0.087)0.456 (0.194)0.445 (0.141)0.428 (0.112)0.365 (0.094)0.369 (0.101)0.509 (0.116)0.342 (0.116)0.253 (0.101)0.291 (0.06)0.305 (0.026)0.282 (0.113)0.299 (0.001)0.382 (0.095)0.364 (0.086)0.42<0.0001
UPG1.020 (0.031)1.026 (0.029)1.021 (0.029)1.031 (0.034)1.049 (0.031)1.050 (0.035)1.126 (0.027)1.127 (0.025)1.121 (0.030)1.204 (0.031)1.187 (0.024)1.028 (0.033)0.981 (0.037)1.086 (0.026)1.109 (0.040)1.083 (0.036)1.078 (0.036)1.062 (0.031)0.78<0.0001
LOG1.262 (0.031)1.257 (0.034)1.261 (0.029)1.260 (0.032)1.282 (0.033)1.271 (0.036)1.356 (0.021)1.362 (0.023)1.353 (0.026)1.458 (0.029)1.434 (0.024)1.247 (0.031)1.199 (0.037)1.330 (0.033)1.333 (0.045)1.290 (0.048)1.315 (0.020)1.258 (0.032)0.81<0.0001
LAS1.935 (0.019)1.930 (0.018)1.931 (0.017)1.929 (0.011)1.939 (0.017)1.924 (0.019)1.936 (0.015)1.947 (0.017)1.947 (0.016)1.957 (0.020)1.954 (0.016)1.952 (0.016)1.965 (0.017)1.956 (0.017)1.979 (0.021)1.993 (0.016)1.945 (0.019)1.958 (0.016)0.52<0.0001
Table Appendix2.  Allelic variability at six microsatellite loci in the 18 studied populations estimated as the number of alleles at each locus (A), allele richness (Ar), the range of allele size in base pairs (R), the observed heterozygosity (Ho: proportion of heterozygous individuals per sample), and genetic diversity (He; Nei, 1987). All the locus deviating from the Hardy–Weinberg equilibrium is marked as * after a tests of sequential Bonferroni adjustments. The overall column gives the populations measure.
Location-populationN tot SsBgIIIM.26Cocl-23Bwf2C1-gC2-157Bwf1Overall
1. Storvika45A1095175551
Ar8.27.44.213.74.54.47.1
R155–219268–294151–161198–254121–153209–233121–294
Ho0.8220.4440.4670.6440.5560.2670.533
He0.7150.5290.4340.8640.5240.2990.561
2. Hullet36A995107646
Ar8.27.94.99.36.65.37.0
R155–221266–288149–161206–248121–155209–231121–288
Ho0.8060.4720.5000.543*0.6110.3610.549*
He0.7840.5060.5630.8490.5560.3640.603
3. Vestfjorden56A8852011456
Ar7.16.54.215.98.23.57.6
R155–219268–288149–161200–248117–161207–229117–288
Ho0.5850.3960.5740.642*0.8150.2590.545*
He0.6710.4880.5220.9060.6420.3220.591
4. Joneset45A775168447
Ar6.46.04.614.16.43.66.9
R155–219268–288149–161200–248121–155209–231121–288
Ho0.6220.5330.5330.591*0.5780.2440.517*
He0.7270.5620.5110.8740.6490.3590.613
5. Hallsteinvika49A8771710655
Ar6.76.05.614.38.54.57.6
R155–219268–288151–163200–248121–161207–229121–288
Ho0.7020.208*0.5000.625*0.5630.3130.484*
He0.7130.4410.5630.9150.6200.4460.616
6. Femundsenden42A8831510448
Ar7.26.63.012.77.93.86.9
R155–219268–286155–161200–252115–153207–229115–286
Ho0.8050.279*0.6050.7440.3950.4190.539*
He0.7630.3530.5210.8930.5760.4350.589
7. Tufsinga45A895158550
Ar6.78.14.113.37.43.17.1
R155–221268–286155–165206–254121–155207–229121–286
Ho0.8090.532*0.4260.553*0.7610.4040.580*
He0.7640.5970.4410.8650.8250.4440.655
8. Sorkelva56A775158547
Ar6.36.23.912.36.73.46.5
R155–219268–286155–161206–252117–161207–229117–286
Ho0.7450.5820.3570.519*0.6070.2860.512*
He0.7070.6110.3380.870.712.0.3510.595
9. Tjønnan45A1085178452
Ar8.77.94.214.17.13.27.5
R155–223268–286155–175198–250121–155207–233121–286
Ho0.614*0.533*0.3560.578*0.7780.4440.550*
He0.7760.7460.3440.8800.7820.3830.651
10. Kvernvika45A910479544
Ar7.48.14.06.37.64.26.3
R155–223258–290155–161206–242121–159209–233121–290
Ho0.400*0.533*0.6890.289*0.6220.356*0.481*
He0.7020.6170.7130.5550.6770.5120.629
11. Gløtfossen45A863108742
Ar7.24.82.88.86.45.76.0
R155–223266–286155–159206–250121–161209–233121–286
Ho0.8570.5330.5560.556*0.6000.3330.569*
He0.7140.5900.5210.8090.5690.3650.593
12. Isteren-dwarf30A744107436
Ar6.76.83.916.38.63.87.7
R155–219274–286155–163204–242121–151207–229121–286
Ho0.6550.6330.5000.7330.7000.2330.575
He0.5780.7070.5720.8660.7160.2160.609
13. Isteren-normal30A674179447
Ar6.03.93.89.76.73.95.7
R155–219248–286151–161204–254121–155209–233121–286
Ho0.6300.3330.4330.9330.6670.7330.621
He0.6270.3950.4300.8090.7240.6100.599
14. Engeren45A77399439
Ar6.26.23.08.47.73.45.8
R155–223268–288155–161202–248121–155209–229121–288
Ho0.863*0.5330.5780.523*0.6440.2000.556*
He0.7190.6870.5110.7690.7410.2750.616
15. Drevsjø45A171182613984
Ar14.19.57.121.410.88.311.8
R155–219260–314147–173198–264121–159209–229121–314
Ho0.7560.6670.6220.659*0.8670.7330.717*
He0.8870.8240.6340.9410.8450.8030.822
16. Vurrusjø45A1316826121085
Ar12.114.07.422.210.99.412.7
R177–219258–336147–173184–260121–161209–233121–336
Ho0.7950.8440.7330.738*0.7780.8670.793*
He0.8410.8590.7710.9520.8760.8470.856
17. Storsjøen30A565105839
Ar5.05.95.09.74.87.86.4
R155–199270–308153–163200–248121–157207–231121–308
Ho0.6430.4140.5330.310*0.633*0.6330.528*
He0.6260.4170.6280.5650.5510.7900.597
18. Lossnen29A1096127852
Ar10.08.95.911.86.97.88.5
R155–219272–306147–161204–250121–153209–229121–306
Ho0.741*0.8620.7240.5000.5170.7590.684*
He0.8560.8500.6470.7570.5090.7650.729