Biodiversity in Swedish Cyprinid Fish: Insights Into Processes of Divergence

Full text

(1)

(2)

(3)

(4)

(5)

(6)

(7)

(8) .

(9) 3WEDISH#YPRINID )NSIGHTS)NTO0ROCESSES OF$IVERGENCE !"#$# "%.

(10)

(11) . "&'&(&) "*+*&'')(&( , -

(12) ,, .&/+**.

(13) .

(14)

(15)

(16)

(17)

(18) !"!"" #"$""%

(19) &

(20) %

(21)

(22)

(23) % '&

(24)

(25) &()&

(26) *

(27) + &( ,-(!""(.

(28) /*&0 &$1 & 1

(29) '

(30)

(31) % (2 . (

(32)

(33) .

(34)

(35)

(36)

(37)

(38) 344(56( ( 47#66843!#!(

(39) &

(40) & &

(41) &

(42)

(43) *

(44)

(45)

(46) %%

(47)

(48) (/

(49)

(50)

(51) %

(52)

(53) 9

(54) *

(55) &

(56) & %

(57)

(58) & .

(59) :

(60) % (1 & &1

(61) %

(62)

(63)

(64) &

(65) /*& %&

(66) ( 2 & &

(67) & &

(68)

(69) %

(70)

(71)

(72) &

(73) 1 &%%

(74) %

(75). 9

(76)

(77)

(78)

(79) %

(80) &(2 1%

(81) &

(82)

(83) ; . & *

(84) %

(85)

(86) &

(87)

(88) &

(89) ()&

(90)

(91) % & ;

(92) %

(93)

(94) &

(95) * ;

(96) &

(97)

(98)

(99) 1

(100) &

(101) * &

(102)

(103) *& & % & &

(104) %% &

(105)

(106) %

(107) *

(108) (<

(109) %

(110) *

(111) ; &%%

(112) % %

(113)

(114)

(115) =>

(116) %

(117) *? &

(118) % % (@

(119)

(120)

(121)

(122)

(123) %

(124) & & &

(125) *

(126) %

(127) %%

(128)

(129) ;( -

(130) * &

(131)

(132) &

(133)

(134) %

(135) % &

(136) * & &%%

(137) % % (2

(138) % &

(139) &

(140) % &#7

(141) %/*&

(142) *

(143)

(144) % &

(145)

(146) *

(147) %% 9 ( )&

(148)

(149)

(150) % & &

(151) &

(152) &

(153) %

(154)

(155) *&

(156)

(157) % (A & %& %

(158)

(159) *

(160) * * & . . & 1%

(161)

(162)

(163) & %& &

(164) * &

(165) ( <

(166) & % & & 1 & ; & &

(167) %

(168)

(169)

(170)

(171) & 9 %

(172)

(173)

(174) %

(175)

(176) (

(177)

(178) & ;

(179)

(180). 9 % &

(181)

(182)

(183)

(184) *&

(185) &/*& %&0 ! " #

(186) $

(187)

(188) $

(189)

(190) #% $

(191)

(192) #&

(193) '()# #$*+,-. # B, - !"" 1// #36#3!#8 1/. 47#66843!#! $ $$$ #"73=&. $:: (9(:

(194) C > $ $$$ #"73?.

(195) List of Papers. This thesis is based on the following papers, which are referred to in the text by their Roman numerals. I. II. III. IV. V. Demandt, M. H., Björklund, M. (2007) Loss of genetic variability in reintroduced roach populations. Journal of Fish Biology (Supplement B), 70:255-261. Demandt, M. H., Bergek, S. (2009) Identification of cyprinid hybrids by using geometric morphometrics and microsatellites. Journal of Applied Ichthyology (In Press). Demandt, M. H. (2009) Stable levels of gene diversity despite low effective population size in isolated perch and roach populations. Accepted manuscript (Conservation Genetics). Demandt, M. H. (2009) Phylogenetic relationship in Swedish cyprinid fish: evidence from mitochondrial and nuclear data. Submitted manuscript. Demandt, M.H., Björklund, M. (2009) Rates of diversification in fishes. Manuscript.. Paper I and II are published with kind permission of the publisher..

(196)

(197) Contents. Introduction ..................................................................................................... 7 Biodiversity and Speciation ....................................................................... 7 Study species .............................................................................................. 9 Common bream, white bream and roach - members of the cyprinid family ..................................................................................................... 9 Species identification in common bream and white bream ................... 9 Sampling of fish................................................................................... 11 Geometric morphometrics ................................................................... 11 Molecular tools .................................................................................... 12 Phylogenetic methods .......................................................................... 13 Aims of the thesis ..................................................................................... 14 Results and discussion .................................................................................. 15 Comparison of genetic variability in natural and reintroduced populations of roach (paper I) ...................................................................................... 15 Interspecific hybridization between common bream and silver bream (paper II)................................................................................................... 16 Gene diversity in isolated perch and roach populations (paper III) ......... 18 Phylogenetic relationship in Swedish cyprinid fish (paper IV)................ 19 Rates of diversification in fishes (paper V) .............................................. 21 Concluding remarks ...................................................................................... 23 Sammanfattning på svenska .......................................................................... 25 Zusammenfassung......................................................................................... 27 Acknowledgements ....................................................................................... 30 References ..................................................................................................... 32.

(198) Species names. Scientific name. English/ Swedish/ German. Abramis brama Blicca bjoerkna Rutilus rutilus Leuciscus leuciscus Leuciscus idus Squalius cephalus Phoxinus phoxinus Scardinius erythrophthalmus Aspius aspius Leucaspius delineatus Tinca tinca Gobio gobio Alburnus alburnus Ballerus ballerus Vimba vimba Pelecus cultratus Carassius carassius Cyprinus carpio. Common bream/ braxen/ Brachsen White bream/ björkna/ Güster Roach/ mört/ Rotauge Dace/ stäm/ Hasel Ide/ id/ Aland Chub/ färna/ Döbel Minnow/ elritsa/ Elritze Rudd/ sarv/ Rotfeder Asp/ asp/ Asp Sunbleak/ groplöja/ Moderlieschen Tench/ sutare/ Schleie Gudgeon/ sandkrypare/ Gründling Bleak/ benlöja/ Ukelei Zope/ faren/ Zope Vimba bream/ vimma/ Zährte Sichel/ skärkniv/ Sichelfisch Crucian carp/ ruda/ Karausche Common carp/ karp/ Karpfen. Perca fluviatilis Salvelinus alpinus. European perch/ abborre/ Flussbarsch Arctic charr/ röding/ Seesaibling. Species names follow the common names for Europe according to FishBase (Froese & Pauly, 2007).

(199) Introduction. Biodiversity and Speciation When Darwin published his book ‘The origin of species’ (Darwin 1859) he challenged the traditional way biologists had viewed the origin of biological diversity. Together with Alfred Russell Wallace, Charles Darwin proposed that biological diversity is the result of natural selection acting on heritable variation in populations (Darwin & Wallace 1858, Stern & Orgogozo 2009). However, how variation was inherited from one individual to its offspring was unknown to both of them. At the very end of the 19th century the studies of Mendel were rediscovered and laid the foundation of the discipline of population genetics. Today, one of the principal aims of population genetics is to quantify the amount of heritable variation present in nature. By doing so, biologists attempt to answer what determines the genetic differences between species, populations and individuals. These differences between species are created by forces such as natural and sexual selection, mutations, gene flow and random genetic drift. The advent of molecular tools (such as DNA sequencing or microsatellites) made it possible to gather fine scale information on species differences and also led to the rise of molecular systematics, revealing information about the speciation process. One of the central problems of speciation is to understand what isolating barriers (preventing gene flow between populations) are important for a given speciation event (Coyne & Orr 2004). Traditionally, allopatric speciation, where populations become physically isolated by a barrier (e.g. mountain range), has been considered the default mode of how species arise. But more and more evidence accumulates that speciation can also occur in sympatry (Barluenga et al. 2006, Mavárez et al. 2006), i.e. populations diverge within the same geographic area. Irrespective of the mode of speciation, most of the concepts describing the process of speciation assume that new species are the result of splitting of old ones (Figure 1a). However, the possibility that new species arise by hybridization of older species (Figure 1b) is widely accepted among botanists, but zoologists perceived natural hybridization to be of little long-term evolutionary importance (Arnold 1997).. 7.

(200) Figure 1. Speciation process as the result of lineage splitting (a) or hybridization (b). Hybridization between divergent evolutionary lineages may result in progeny with lower levels of fertility and /or viability, but may also create hybrid genotypes with equivalent or higher levels of fitness in certain environments. In the former case, the parental species are not affected as F1 hybrids are removed by natural selection (Arnold & Hodges 1995). However, in the latter case, fertile F1 hybrids can either cross with the initial hybrid generation or backcross with their parental species leading to introgression. During the last two decades, a number of zoologists have concluded that natural hybridization is frequent and evolutionary important in taxa as divergent as birds and fish (Grant & Grant 1992, DeMarais et al. 1992). Despite the fact that the subject of speciation has attracted considerable attention, our understanding of the process of speciation still remains fragmentary (Coyne & Orr 2004). Ever since Darwin presented his concept of evolution by natural selection it has been debated whether phenotypic evolution takes place gradually or suddenly upon speciation (e.g. the theory of ‘punctuated equilibrium’, Eldredge & Gould 1972). From the fossil record inferences upon the features of evolutionary radiations can be made and reveal the tendency for diversification rates to decline through time. Recent accumulation of phylogenetic information has become increasingly important in understanding the modes of diversification leading to the present diversity. Increasing availability of gene sequence data and the development of phylogenetic reconstruction from these data provide tools for analyzing the history of diversification in cyprinid fishes which allow estimates of rates of speciation and extinction (Nee et al. 1992, Rabosky & Lovette 2008). By using cyprinid species that hybridize with each other and thus have not reached the final stage of speciation, important insights into the evolution of isolation mechanisms can be gained, allowing us to understand the steps leading to speciation.. 8.

(201) Study species Common bream, white bream and roach - members of the cyprinid family In this thesis focus was laid on the common bream (Abramis brama L.), white bream (Blicca bjoerkna L.), and roach (Rutilus rutilus L.). These three species belong to the family of Cyprinidae. The Cyprinidae represent one of the largest fish families of the world (Nelson 1994) and form the largest family of freshwater fish with distribution in Eurasia, Africa and Northern America (Billard 1999). A large proportion of the freshwater ichthyofauna of Sweden belongs to the family of Cyprinidae. Out of the 46 fish species reproducing naturally in Sweden (Lehtonen et al. 2008) 18 cyprinid species are found in Swedish freshwaters, though most of them are restricted to the southern parts of the country (Pethon & Svedberg 2000). Traditionally, the Cyprinidae are split into the two lineages of the Cyprininae and the Leuciscinae (Zardoya & Doadrio 1999). This division is supported by morphological studies (Howes 1991). Within these lineages, the relationship between sub-families has not been totally resolved yet (Hänfling & Brandl 2000) and numerous approaches have been made to shed light onto the systematics using morphological (Chen et al. 1984), allozyme (Hänfling & Brandl 2000), and mitochondrial data (Zardoya & Doadrio 1999). Interspecific and intergeneric hybridization has been reported for the family of Cyprinidae. Their widespread distribution in Europe, the temporal overlap in spawning activities (Wheeler 1969) as well as the uniform karyotype of this group of fishes, weakens the barrier to hybridization (Bianco et al. 2004). About 62 different hybrids between the species of European Cyprinidae have been described, mostly between different Leuciscinae genera (Kottelat & Freyhof 2007). In particular, the hybrids of common bream and roach are recorded from almost every water body in which the parental species occur (Kottelat & Freyhof 2007) and represent the most common hybrid fish in waters of the British Isles (Pitts et al. 1997). Although hybridization within the cyprinid family has been studied intensely, little scientific attention has been paid to the sibling species common bream and white bream (but see Swinney & Coles 1982). Because sibling species share a common ancestor, white bream and common bream can be used as a model system for the study of mechanisms leading to speciation.. Species identification in common bream and white bream Species were identified based on phenotypic differences. In white bream the eye diameter is larger than the distance from the tip of the snout to the anterior edge of the eye. In contrast, the snout-eye distance exceeds the eye di9.

(202) ameter in common bream. Additionally a red base is visible on the pectoral and dorsal fins of white bream, whereas common bream lack this colouring (Gebhardt & Ness 1998, Pethon & Svedberg 2000, Wheeler 1969). Based on these differences I clustered the fishes into three groups. Group 1: A. brama, Group 2: B. bjoerkna and Group 3: individuals with an intermediate phenotype hereafter called hybrid (Figure 2).. Figure 2. Common bream (Abramis brama L.), hybrid, and white bream (Blicca bjoerkna L.). 10.

(203) Sampling of fish Many fish contributed to the findings presented here (Table 1). The majority of fish were sampled by the Swedish Board of Fisheries in the context of various national monitoring projects. Sampling of common bream and white bream has been conducted by me and colleagues by means of gill net fishing. Samples used for the phylogeny were collected by several different people. All contributions have been acknowledged in the respective articles. During all experimental procedures, the authors followed Swedish Guidelines concerning the care and welfare of experimental fish. Table 1. Species sampled for this thesis and their sampling location. SBF = Swedish board of Fisheries, MD = Marnie Demandt & colleagues Species. sampling location. Year. paper. collector. R. rutilus R. rutilus R. rutilus R. rutilus R. rutilus R. rutilus R. rutilus R. rutilus A. brama, B. bjoerkna A. brama, B. bjoerkna A. brama, B. bjoerkna A. brama, B. bjoerkna A. brama, B. bjoerkna A. brama A. brama A. brama A.brama A. brama R. rutilus, P. fluviatilis R. rutilus, P. fluviatilis R. rutilus, P. fluviatilis R. rutilus, P. fluviatilis R. rutilus, P. fluviatilis. Lake Fiolen Lake Allgjuttern Lake Stora Envättern Lake Bysjön Lake Nässjön Lake Surtesjön Lake Husevattnet Lake Tinnsjön Lake Funbosjön Lake Gisslaren Lake Långsjön Lake Lötsjön Lake Södersjön Lake Bredsjön Lake Gimo Damm Lake Fibysjön Lake Velången Lake Siggeforasjön Biotest basin Biotest basin Biotest basin Biotest basin Biotest basin. 2002 2002 2002 2002 2002 2002 2002 2002 2003 2003 2003 2003+2004 2003 2004 2003 2003 2004 2003+2004 1977 1982 1988 1994 2000. I I I I I I I I II II II II II II II II II II III III III III III. SBF SBF SBF SBF SBF SBF SBF SBF MD MD MD MD MD MD MD MD MD MD SBF SBF SBF SBF SBF. Geometric morphometrics To quantify the morphology (shape) of common bream and white bream I used landmark-based morphometrics (Bookstein 1991). 13 corresponding landmarks (Figure 3) were digitalized and recorded from photos of each specimen. The computer program TpsRelw (Rohlf 2003) was applied to process the x and y coordinates of each specimen using Procrustes Superimposition, which translates, rotates and scales to unit size. Through superim11.

(204) position the non-shape variation in the dataset is eliminated (Adams et al. 2004) and the variables become shape variables, which in turn can be used to statistically compare samples and depict changes in shape. Changes in shape are described by both partial warps (shape changes due to local deformations) and the uniform scores (shape changes due to bending or compression). Multivariate analysis of variance (MANOVA) and canonical discriminant function analysis were applied to the dataset to determine the degree of morphometric differentiation between common bream, white bream and hybrids. Geometric morphometrics allow graphical representations in terms of the configuration of landmark points (Adams et al. 2004) and the differences in shape among groups of individuals can be depicted in deformation grids. Thin-plate spline deformation grids were generated, where the shape differences of one specimen are represented as a set of bent grid lines relative to an undeformed grid representing the consensus form of the entire sample.. Figure 3. Position of the landmarks used to define body shape.. Molecular tools The molecular tools applied in this thesis were microsatellite DNA and sequencing of mitochondrial and nuclear DNA. Microsatellites Microsatellites are short segments of DNA in which a specific motif of one to six basepairs is repeated, usually up to a maximum of approximately 60 times. They are assumed to be neutral markers (but see Selkoe & Toonen 2006) that follow Mendelian inheritance. Their high mutation rate generates high levels of allelic diversity necessary for genetic studies. A combination of several microsatellite markers decreases sampling errors and provides a more precise picture of the population under study. Furthermore, the use of multiple loci increases statistical power (Björklund & Bergek 2009). Here, microsatellite markers were used to detect whether populations reveal signs of former bottlenecks, to estimate population size, and the level of relationship among species. 12.

(205) Mitochondrial and nuclear DNA Mitochondria are distinct cellular organelles with their own genome. They harbour a small, circular genome (mtDNA) encoding mostly unique sequences: 13 protein-coding genes, two rRNA genes, 22 tRNA genes, and a control region that contains sites for replication and transcription initiation (Li 1997). Due to its predominantly maternal inheritance, the absence of recombination, and a relatively high mutation rate, mtDNA has become a powerful tool for tracking the ancestry of many species back for hundreds of generations. The majority of phylogenetic studies involve analyses of the control region and the cytochrome b gene. However, other regions of the mitochondrion (e.g. ATP synthase subunit 6 and 8) have been shown to be as good or even better in terms of phylogenetic resolution (Ödeen 2001). Nuclear DNA is located in the nucleus of eukaryotic cells and organized in chromosomes. The bi-parental mode of inheritance makes recombination possible. Since nuclear genes can be selected from different chromosomes, gene trees reflect a different evolutionary history than mitochondrial genes. Phylogenetic studies should use a combination of both types of DNA in order to infer the relationship among species.. Phylogenetic methods To explain diversity one needs to have good knowledge on the evolutionary history of the group under study. Data for phylogenetic inference can stem from sources such as morphology, behaviour, and geographic distribution. With the molecular revolution in the 20th century, data on nucleotide sequences have been used predominantly. In this study, nucleotide sequence data were obtained for the mitochondrial ATP synthase subunit 6 and 8 and for the nuclear gene ITS1. The following two approaches are to date widely used in phylogenetic practice and were applied to the molecular data. Maximum parsimony Maximum parsimony is a character-based (cladistic) method that infers a phylogenetic tree by minimizing the total number of evolutionary steps required to explain a given set of data, or in other words by minimizing the total tree length. The evolutionary steps are base- or amino acid substitutions from sequence data. Bayesian Inference In model-based methods such as Bayesian analysis, inferences of phylogeny are based upon the posterior probabilities of phylogenetic trees (Huelsenbeck & Ronquist, 2001). Bayesian methods incorporate prior information (also called ‘prior probability’) and update estimates of the evidence in favour of the different hypotheses by combining the prior probabilities and the probabilities of obtaining the data under each of the hypothesis (McCarthy, 2007). 13.

(206) Both Maximum parsimony and Bayesian inference were used to study the phylogenetic relationship of the Swedish Cyprinidae. However, these methods produced incongruent phylogenies and were affected by problematic taxa.. Aims of the thesis The main aim of this thesis was to investigate the processes such as hybridization, gene flow, and random genetic drift involved in creating the diversity in Swedish cyprinid fish. Main questions addressed in this thesis are: • Is the demographic history of a population mirrored in the genome? What are the genetic consequences of population bottlenecks? (Paper I) • Are traditional methods of species identification sufficient to identify specimens of common bream, white bream and their hybrids to avoid misclassification? How does hybridization between common bream (Abramis brama) and white bream (Blicca bjoerkna) affect the phenotype? (Paper II) • How large is genetic drift in isolated populations? Can genetic diversity be maintained with decreasing effective population size? (Paper III) • What are the phylogenetic relationships among the 18 cyprinid species occurring in Sweden? (Paper IV) • What do molecular phylogenies of fishes tell us about the speciation rate through time? (Paper V). 14.

(207) Results and discussion. Comparison of genetic variability in natural and reintroduced populations of roach (paper I) The aim of paper I was to determine the level of genetic variability in natural and reintroduced populations of roach (Rutilus rutilus) and to assess whether the sizes of the introductions were large enough to prevent loss of genetic variability with all its consequences. Acidification of lakes has been a major problem in Western Europe and has led to the extinction of fish populations over large areas. The pH in many lakes has been restored by liming (Appelberg et al. 1992), and most lakes are now healthy enough to sustain fish populations. Colonization by fishes has either happened naturally or by human transfer and release. However, the genetic impact of human mediated reintroductions is largely unknown and evidence suggests that reintroductions can lead to several problems (Ryman & Laikre 1991, Lynch & O’Healy 2001, Theodorou & Couvet 2004). I analysed five microsatellite markers and sequenced a 550 bp long segment of mtDNA, including the ATP synthase subunit 6 and 8 in order to determine the level of genetic variability in natural (control) and reintroduced populations of roach (Figure 4). For the reintroduced populations a record of number of introduced specimen was available. The microsatellite analysis revealed larger allelic richness in the control group for three of the five loci. Additionally, I found allelic richness to be positively correlated with abiotic factors such as lake size, lake depth and species diversity. The range of allele size was larger in the control group than in the introduced group for all loci. Sequencing of mtDNA revealed that the number of variable sites was almost twice as large in the control group as in the introduced group, not significantly though. Small variance in haplotype frequencies was found for all populations except one (Stora Envättern) indicating the effects of a bottleneck in population size. To summarize, the results of the microsatellite analysis, and to a lesser extent the mtDNA sequence analysis, revealed that the reintroduced populations have less genetic variability than natural populations. The findings of this study allow the conclusion that the number of introduced individuals has to be maximised in order to avoid a reduction in ge15.

(208) netic variability a) already before collection of the founding stock (due to incomplete sampling) and b) after the introduction event (due to survival and skewed reproductive success). High quality in habitat contributes to successful reintroductions.. Figure 4. Map over southern Sweden with locations of lakes with bottlenecked (*) and non-affected populations of roach.. Interspecific hybridization between common bream and silver bream (paper II) The aim of paper II was to assess whether species identification using phenotypic/morphological characters is sufficient to identify the closely related species pair Abramis brama and Blicca bjoerkna and their hybrids correctly or if identification should be complemented with genetical analyses. Interspecific hybridization is a widespread phenomenon (Grant & Grant 1992, Dowling & DeMarais 1993) and fish species hybridize probably more than any other animal taxon (Turner 1999). Although hybridization within the cyprinid family has been studied intensely, little scientific attention has been paid to the sibling species common bream and white bream (but see Swinney & Coles 1982). Sibling species share a common ancestor and can be used as an important model system for the study of mechanisms leading to speciation. Populations are defined as species when gene flow between them is sufficiently small that each remains distinct (modified version of Mayr’s biological species concept [1995], as suggested by Coyne & Orr [2004]). Ongoing gene flow between hybridizing populations ultimately leads to the fusion 16.

(209) into a single species if not the evolution of pre- or post-zygotic isolation barriers reduce the frequency of hybridization and lead to completion of the speciation process. For hybridization studies, the use of accurate taxonomic information is essential. However, distinguishing between pure breeds of common bream (Abramis brama) and white bream (Blicca bjoerkna) and their hybrids by assessing their external morphology can be extremely difficult. Nuclear markers have proven to be an effective tool for the differentiation of cyprinid species and the detection of interspecific hybrids (Hänfling et al. 2005, Pitts et al. 1997, Freyhof et al. 2005, Hamilton & Tyler 2008). The analysis of mitochondrial DNA in hybrids additionally allows the identification of the maternal species and thus gives information about the direction of hybridization. I used a combination of landmark-based techniques together with microsatellite markers and sequence data to investigate natural hybridization between the cyprinids common bream and white bream and their hybrids. Geometric morphometrics revealed significant differences in body shape among common bream, white bream and hybrids and thus confirmed the classification based on phenotypic markers. Hybrids were, as expected, of intermediate body shape but tended to cluster more with specimen of common bream (Figure 5).. Figure 5. Scatterplot depicting clusters of common bream, white bream and hybrids.. Reports from other studies concerning the resemblance of common bream and white bream hybrids are contradictory. Wheeler (1969) describes hybrids as resembling white bream, whereas Swinney & Coles (1982) point out that hybrids are more similar to common bream. The observed phenotypic similarity of hybrids with common bream is concordant with the results from the sequence data of the ATP synthase subunit 6 & 8 revealing that most hybrids (89%) actually were mothered by common bream. The microsatellite data revealed significant departures from HardyWeinberg equilibrium (HWE) in two loci in the common bream group and in 17.

(210) three loci in the hybrid group. These departures from HWE can either be due to null alleles, heterozygote deficiencies or to a subpopulation structure (Wahlund effect). Since I pooled the specimen from different lakes, nonrandom sampling of members from a limited numbers of families is the most probable explanation for the observed heterozygote deficiency. Genic differentiation was significant for each pair of groups in all comparisons. Although 94% of the genetic variability was found within groups, a highly significant amount of the variation was accounted for by differences among groups (global FST = 0.055, Analysis of molecular variance [AMOVA], p<0.01). This finding supports the results of the geometric morphometrics and sequence data. To summarize, for common bream and white bream the effect of ongoing hybridization has not threatened the genetic integrity of these two species. The observed unidirectional hybridization, however, and possible introgression through fertile F1 hybrids cannot be overlooked and needs further attention.. Gene diversity in isolated perch and roach populations (paper III) The aim of paper III was to investigate the effect of genetic drift on genetic variation in enclosed populations of perch and roach. Anthropogenic environmental disturbances such as habitat/population fragmentation and its effects on population structures are no longer deniable and we can observe that affected populations often show decreased genetic variation (Borell et al. 2008, Yamamoto et al. 2004). Since genetic diversity is the basis for evolutionary change, temporal changes in gene frequencies are a simple yet effective tool to assess a population´s evolutionary potential. Changes in gene frequencies over time allow the direct study of the causes of evolutionary change (Lessios et al. 1994) such as genetic drift, selection or gene flow. Genetic drift, together with natural selection, is the major source of changes in allele frequencies in small and isolated populations (Begon et al. 1996). Under laboratory conditions it has been shown that genetic drift leads to temporal changes in gene frequencies (Shikano et al. 2001). However, assessing the long-term impact of genetic drift in natural populations is far more difficult since such populations are rarely isolated from conspecific populations and ongoing gene flow antagonizes the effects of drift. Information on how much of the genetic variation is lost due to random drift can be gained by estimating the effective population size (Ne) of a population. Most estimates of Ne to date are based on measures of changes in allele frequencies between temporal samples from the same population. 18.

(211) In this study I analysed allele frequencies at five polymorphic microsatellite loci of an enclosed population of perch (Perca fluviatilis) and roach (Rutilus rutilus) covering a period of 23 years. The observations of temporal shifts in genetic variability were used to estimate Ne for both species applying four different methods. I expected to find a decrease in genetic diversity over time. In contrast to my expectation the analysis of microsatellite data revealed stable levels of gene diversity over time for both species. Estimates of genetic differentiation (FST) showed a slight but highly significant divergence between 1977 and 2000 for both perch and roach. This differentiation is consistent with isolation by time. The observed FST values were lower than FST values obtained by simulations only for the first time period (1977-1982), indicating that genetic drift is lower than expected, but fell within the range of the simulated values for the remaining time period. Estimates of effective population size varied largely among methods but follow the same trend with decreasing Ne until the end of the 80ies and increased afterwards (Figure 6). For roach, the population consisted of one interbreeding population. However, for perch four subpopulations were found which aggregate independent of kin. This finding is consistent with other studies on small scale genetic differentiation in perch (Bergek & Björklund 2007). 260. 700. Roach. Perch 600. 240. harmonic mean Ne TempoFs NeEstimator LDNe. 220 200. 500. harmonic mean Ne TempoFs NeEstimator LDNe. 180 160. 400. 140 120. 300 100 80. 200. 60 40. 100. 20 0. 0 1977-1982. 1982-1988. 1988-1994. 1994-2000. 1977-1982. 1982-1988. 1988-1994. 1994-2000. Figure 6. Different estimates of effective population size Ne for populations of perch and roach in the Biotest basin from 1977 to 2000.. Phylogenetic relationship in Swedish cyprinid fish (paper IV) The aim of paper IV was to determine the phylogenetic relationship among the 18 species of cyprinids occurring in Sweden using mitochondrial and nuclear markers. The phylogenetic relationship among the European cyprinids remains unresolved and even the monophyly of the whole family is sometimes in doubt 19.

(212) (Howes 1991). Despite approaches to resolve the relationships including morphological (Chen et al. 1984), allozyme (Hänfling & Brandl 2000) and mitochondrial (Zardoya & Doadrio 1999) studies, the relationships are far from being solved and different classifications are proposed by different authors (Kottelat & Freyhof 2007). The cyprinids are widely distributed in North America, Eurasia, and Africa, but are naturally absent from South America, Madagascar and Australia. The greatest diversity of cyprinid species is found in China and south-east Asia (Billard 1999). From Asia, cyprinids colonized Europe in the Oligocene (30-35 million years ago) (Banarescu 1960). However, how their subsequent dispersal throughout Europe took place is unclear but colonization from Russia across northern Europe via river connections and through freshwater lakes is the most likely scenario (Levy et al. 2009). 40 cyprinid species are extant in the area of central Europe (Banarescu & Coad 1991). To date, the existing phylogenies on European cyprinids are restricted in taxon sampling and do not include all taxa found in Sweden. Aligned sequences of the 18 species occurring in Sweden were analysed by maximum parsimony (MP) and Bayesian inference (BI). Analyses were based on one mitochondrial gene (ATP synthase subunit 6 and 8) and one nuclear gene (internal transcribed spacer region 1 (ITS1)). As outgroup, I chose a member of the family of Salmonidae (Salvelinus alpinus L.). Phylogenetic reconstructions both with maximum parsimony and Bayesian inference support the traditional subdivision into the subfamilies Cyprininae and Leuciscinae (Figure 7). The relationship between Leuciscus delineatus and Alburnus alburnus was consistently recovered in all trees. The position of the tench (Tinca tinca) was not consistent among the different trees, and thus its position relative to the other species could not be resolved. However, based on my data, the tench should be included into the subfamily of Leuciscinae. Surprisingly, I could not confirm the status of common bream (A. brama) being sister species to white bream (B. bjoerkna), as postulated by other studies. Since these two breams clustered together with Vimba their relationship needs further investigation. The bushlike topology of the Leuciscinae suggests speciation events in an evolutionary very short time span.. 20.

(213) Salvelinus alpinus. Vimba vimba. Blicca bjoerkna. Abramis brama 1.00 Scardinius erythrophthalmus. Squalius cephalus 0.96 Ballerus ballerus 0.77 0.91 Rutilus rutilus. 0.79. 0.99 Tinca tinca. Aspius aspius 0.96 1.00 Gobio gobio 1.00. 0.96. Leuciscus idus 1.00 0.88 Leuciscus leuciscus Pelecus cultratus 0.94 Phoxinus phoxinus. Alburnus alburnus 1.00 Leuciscus delineatust. Carassius carassius 0.79 Cyprinus carpio 0.1. Figure 7. Tree resulting from Bayesian analysis for the concatenated dataset.. Rates of diversification in fishes (paper V) The aim of paper V was to investigate divergence times in different taxa of fish on the basis of molecular phylogenies. The branching pattern of a phylogenetic tree can be used to detect changes in speciation rate through time. The growing number of molecular phylogenies provides us with an increasing amount of data that can be used for studying diversification rates. Analyses of molecular phylogenies often show a tendency of bursts of diversification rates in the early stages of species radiations, followed by a decrease in speciation rates through time (Rüber & Zardoya 2005, Phillimore & Price 2008). This pattern of evolution is already found in the fossil record where little evolutionary change over long time is followed by rapid cladogenesis and gave rise to the ‘theory of punctuated equilibrium’ (Eldredge & Gould 1972). Slowdowns of speciation rates can be explained by density dependent cladogenesis, a pattern describing a decrease in speciation rates due to decreased availability of free niche space. The tools for inferring rates of diversification have improved (Pybus & Har21.

(214) vey 2000, Rabosky 2006) but have in common that they rely on the generalized birth-death process. This process models the situation in which a species has equal probabilities of either speciating or going extinct (Nee et al. 1994). A literature search was conducted for species-level molecular phylogenies of fishes. 15 phylogenies were chosen, covering a variety of taxa. Divergence times were estimated using the R-package LASER and tested for fit of different models of diversification. Additionally, changes in speciation rates were estimated by the application of -statistics. In all phylogenies diversification rates were best described by ratevariable models, and changes in diversification rates followed always a change in speciation rate model. The majority of phylogenies revealed a negative value of , indicating a decline in diversification through time. This study reveals that incomplete taxon sampling leads to a violation of the assumptions of the applied -statistics, leading to erroneous conclusions about the process of divergence. A negative correlation between clade size and has been found in other studies, but could not be confirmed by our data. The findings of this study indicate that density-dependent cladogenesis is the most probable evolutionary scenario for fishes.. 22.

(215) Concluding remarks. In this thesis, molecular methods were utilized to obtain data on several cyprinid species, such as common bream, white bream, and the roach as well as on two non-cyprinids, the perch and the charr. Since genetic variability is the basis for evolutionary change it is important to maintain the variability found in populations especially when it comes to endangered species. By studying common species such as the roach, information can be gathered and transferred to rare species, e.g. on how to prevent the loss of genetic diversity when it comes to reintroduction events. Reintroductions can result in a bottleneck with consequent loss of genetic variation. I found that genetic variation was lost in populations used for reintroduction compared to persistent populations of roach (paper I). However, loss of genetic variation must not necessarily occur, since the effect of bottlenecks can be highly stochastic and strongly related to duration of the bottleneck and population size. This study confirmed that large propagules are necessary to maintain genetic variability. I found that common bream and white bream are genetically differentiated but that identification of the pure species and the hybrids based on phenotypic markers is not easy. First generation hybrids backcross presumably with the parental species as well as other hybrids resulting in a broad continuum of morphotypes. Geometric morphometric revealed differences in body shape among common bream, white bream and their hybrids (paper II). In order to secure the correct identification of these species and their hybrids I suggest a combination of molecular and traditional identification methods. The development of genetic markers, sensitive to discriminating hybrids from their parental species, would simplify the study of hybridization and its prevalence. By obtaining estimates of effective population size (Ne) for enclosed populations of perch and roach I found large differences in estimates of Ne depending on which method was chosen (paper III). Investigations on temporal changes in genetic diversity revealed stable levels of genetic diversity over time despite low Ne. Being affected by random genetic drift, genetic divergence was created over time within the population of perch and roach, respectively. Phylogenetic inference, using Maximum Parsimony and Bayesian analysis for one nuclear and one mitochondrial gene, confirmed the difficulties in resolving the relationship among European Cyprinidae. In contrast to earlier 23.

(216) studies, my data did not support the status of common bream and white bream as sister species. Unresolved polytomies for the subfamily of Leuciscinae suggest rapid species divergence through time. Speciation is the ultimate source of new species. By analyzing published phylogenies of fish I found that rates of divergence follow density dependent cladogenesis, i.e. that a burst of speciation is followed by a slow down with niches becoming filled (paper V). These findings are consistent with divergence rates estimated in other taxa (birds) and seem to mirror a general pattern in species diversification through time. The work presented here in this thesis contributes to insights into the processes of divergence creating the biodiversity observed in general and into (Swedish) cyprinid fish in particular.. 24.

(217) Sammanfattning på svenska. Den diversitet av organismer, som vi ser på vår jord, är ett resultat av ett flertal olika faktorer och processer. En av processerna kallas för mikroevolution och beskriver bland annat evolutionen som sker till följd av ändrade genfrekvenser i en population. Små förändringar ackumuleras och leder så småningom till genetiska skillnader mellan populationer av samma art. Makroevolution däremot är en process som beskriver evolutionen av tydligt skilda ordningar/familjer och släkten. När Darwin för 150 år sedan i sin bok ”Arternas uppkomst” ifrågasatte biologernas traditionella syn på arternas uppkomst, menade han makroevolution. Darwin postulerade i sin teori att biodiversitet är ett resultat av naturligt urval på ärftlig variation hos en population. Hur variation kunde överföras från en generation till nästa hade han ingen kunskap om eftersom Gregor Mendels upptäckter dröjde ända till 1869 för att sen återupptäckas först år 1900. Även om artbildningsprocessen fortfarande idag fascinerar forskarna, så finns det luckor i kunskapen om hur nya arter bildas. Generellt skiljer man mellan två olika mekanismer av artbildning. Vid allopatrisk artbildning delas en population, A, i två underenheter (A1 och A2) genom externa faktorer (t. ex. formation av en bergskedja). Varje underenhet, A1 och A2, utsätts för olika selektionstryck och utvecklas längs olika evolutionära banor. Artbildningsprocessen är avslutad när A1 och A2 inte längre kan reproducera sig med varandra och utbytet av genetiskt material uteblir. Den här processen är allmänt accepterad. I motsats till den allopatriska artbildningen sker processen sympatrisk artbildning inom samma geografiska område och fysiska barriärer, som förhindrar genflöde, existerar inte. Den evolutionära betydelsen av sympatrisk artbildning betraktas dock som marginell. Ända sedan Darwin introducerade teorin om evolution genom naturligt urval, har det diskuterats om evolutionen sker linjärt eller stegvis. Fossiler tillåter bara en ofullständig rekonstruktion av den förflutna evolutionen. Ackumulering av fylogenetisk information och vidareutveckling av fylogenetiska rekonstruktionsmetoder, baserad på genetiska data, ger ytterligare kunskaper över artbildningens historia. De genetiska skillnaderna mellan arter, populationer och individer är en följd av naturlig och sexuell selektion, mutationer, genflöde och genetisk drift. Dessa kan kvantifieras genom användning av molekylära metoder som mikrosatelliter och/eller sekvensering av DNA. Användning av arter, där artbildningsprocessen inte är avslutat än, är följaktligen av stor betydelse för 25.

(218) forskning som rör processer som leder till artbildning. Karpfiskarna braxen (Abramis brama) och björkna (Blicca bjoerkna) är nära släkt och hybridiserar med varandra. Mellan de här två arterna kan det alltså fortfarande ske ett utbyte av gener och de har ännu inte nått artbildningens slutfas. Därför kan vi få viktig information om de evolutionära processerna bakom artbildning med hjälp av de här modellorganismerna. Det övergripande syftet med denna avhandling är att undersöka processer som genflöde, genetisk drift och hybridisering hos naturliga fiskpopulationer, för att kunna förstå deras bidrag till uppkomsten av den diversitet vi ser hos karpfiskar i Sverige. Med hjälp av morfometriska och genetiska analyser hos braxen och björkna och deras hybrider kunde jag visa att det finns fenotypiska och genetiska skillnader mellan dem (artikel II). Genom sekvensering av en mitokondriell gen visade jag att hybridisering mellan braxen och björkna huvudsakligen sker mellan braxen-honor och björkna-hanar. Den genetiska differentieringen mellan båda arterna var dock så stor att en fusion anses som osannolikt i nuläge. Jag undersökte även en population av mörtar (Rutilus rutilus), som reducerades i antal genom en flaskhals (artikel I). Genetiska analyser visade att populationer, som decimerats genom en flaskhals, har mindre genetisk variation, jämfört med en kontrollgrupp av samma art, vars populationsstorlek var konstant över tiden. Effekten av genetisk drift på den genetiska variationen undersökte jag hos populationer av mört och abborre (Perca fluviatilis) som under lång tid varit isolerade (artikel III). Min förväntning var att den genetiska variationen skulle minska med tiden. Detta bekräftades inte, utan den genetiska variationen var konstant. Den effektiva populationsstorleken minskade dock hos båda arterna, ett resultat som överensstämmer med påverkan av genetisk drift. Släktförhållandena mellan de europeiska karpfiskarna i allmänhet och i synnerhet hos de svenska karpfiskarna, är delvis oklara. Genom sekvensering av en mitokondriell och en nukleär gen sammanställde jag de fylogenetiska förhållanden för de 18 karpfiskar som finns i Sverige (artikel IV). Det mitokondriella och det nukleära släktträdet visade inte samma resultat, vilket kan förklaras med olika mutationshastigheter och selektionstryck för de båda generna. Förhållandet mellan braxen och björkna som systerarter kunde inte bekräftas. Deras släktskap måste undersökas närmare i samband med ytterligare en art, vimman. Artbildningsprocesser är den ultimata källan till nya arter. Genom analyser av redan publicerade fylogenier av olika fiskarter kunde jag visa att förgreningshastigheten är beroende av den befintliga arttätheten i habitatet (s. k. density dependent cladogenesis). Det betyder att artbildningen sker explosionsartad i början men mattas av på grund av en ökad konkurrens mellan arterna (artikel V). Mina resultat bekräftar liknande estimat av divergenshastigheter hos andra taxa (t. ex. fåglar) och återspegla ett allmänt mönster inom artbildningsprocessen. 26.

(219) Zusammenfassung. Die Diversität von Organismen, die wir auf unserer Erde sehen, ist das Resultat einer Vielzahl von Faktoren und Prozessen. Ein als Mikroevolution benannter Prozess beschreibt u.a. die Evolution als Folge veränderter Genfrequenzen in einer Population. Kleine Veränderungen akkumulieren sich und können so langfristig zu genetischen Unterschieden zwischen Populationen der gleichen Art führen. Diesem Prozess gegenüber steht die Makroevolution, mit der die Evolution von deutlich zu unterscheidenden Gattungen/Familien und höheren Taxa gemeint ist. Als Darwin vor 150 Jahren mit seinem Buch „The origin of species“ die traditionelle Sichtweise der Biologen zur Entstehung der Arten herausforderte, meinte er genau diesen Prozess der Makroevolution. Darwin postulierte in seiner Theorie, dass Biodiversität das Resultat natürlicher Selektion auf vererbbare Variation in Populationen sei. Wie aber Variation von einer Generation zur nächsten weitervererbt wird, war ihm nicht bekannt, denn die Erkenntnisse Gregor Mendels lagen noch nicht vor und sollten erst im Jahr 1869 entdeckt, bzw. im Jahr 1900 wiederentdeckt werden. Obwohl Forscher auch noch heutzutage noch von der Artbildung fasziniert sind, ist das Wissen über die Prozesse in einigen Bereichen immer noch lückenhaft. Generell unterscheidet man zwischen verschiedenen Mechanismen der Artbildung. Beim Vorgang der allopatrischen Artbildung wird durch externe Faktoren (z. B. die Auffaltung einer Bergkette) eine Population A in zwei Untereinheiten (A1 und A2) aufgespalten. Jede Untereinheit A1 und A2 ist unterschiedlichen Selektionsdrücken ausgesetzt und entwickelt sich entlang unterschiedlicher evolutionärer Pfade. Der Prozess der Artbildung ist abgeschlossen, wenn sich A1 und A2 nicht mehr miteinander reproduzieren können und somit der Austausch von genetischem Material unterbleibt. Dieser Prozess ist allgemein akzeptiert. Im Gegensatz zur allopatrischen Artbildung findet der Prozess der sympatrischen Artbildung innerhalb des gleichen geographischen Gebiets statt und ohne physische Barrieren, die den Genfluss unterbinden. Die evolutionäre Bedeutung der sympatrischen Artbildung wird jedoch als marginal betrachtet. Seit Darwin seine Theorie der Evolution durch natürliche Selektion vorstellte, wird diskutiert, ob Evolution geradlinig oder sprunghaft stattfindet. Funde von Fossilien erlauben nur eine sehr lückenhafte Rekonstruktion der vergangenen Evolution. Doch die Akkumulation phylogenetischer Information (Phylogenie: Abstammung) und auch die Weiterentwicklung phylogene27.

(220) tischer Rekonstruktionsmethoden, basierend auf genetischen Daten, geben zusätzliche Erkenntnisse über die Geschichte der Artbildung. Die genetischen Unterschiede zwischen Arten, Populationen und Individuen sind Folge von natürlicher und sexueller Selektion, Mutation, Genfluss und genetischer Drift und können durch den Gebrauch von molekularen Methoden wie Mikrosatelliten und/oder der Sequenzierung der DNA quantifiziert werden. Der Gebrauch von Arten, bei denen der Artbildungsprozess noch nicht ganz abgeschlossen ist, d.h. zwischen denen weiterhin Gene ausgetauscht werden, ist daher für das Studium der Artbildungsprozesse von großer Bedeutung. Die Karpfenartigen Brachsen (Abramis brama) und Güster (Blicca bjoerkna) sind nah miteinander verwandt und hybridisieren miteinander. Zwischen diesen Arten kann also weiterhin ein Austausch von Genen stattfinden und sie haben somit noch nicht das Endstadium der Artbildung erreicht. Aus diesem Grund können anhand dieser Modellorganismen wichtige Einblicke in die evolutionären Prozesse der Artbildung gewonnen werden. Ziel dieser Doktorarbeit ist, die Prozesse wie Genfluss, genetische Drift und Hybridisierung bei natürlichen Fischpopulationen zu untersuchen, um Aussagen über deren Beitrag zur Bildung der Diversität bei (schwedischen) Karpfenartigen machen zu können. Anhand von morphometrischen und genetischen Analysen bei Brachsen und Güstern und deren Hybriden konnte ich zeigen, dass phänotypische und genetische Unterschiede zwischen diesen drei Typen bestehen (Artikel II). Mit Hilfe der Sequenzierung eines mitochondriellen Gens ließ sich nachweisen, dass die Hybridisierung zwischen Brachsen und Güster hauptsächlich zwischen Brachsen Weibchen und Güster Männchen stattfand. Beide Arten waren jedoch so weit genetisch differenziert, dass zum jetzigen Zeitpunkt nicht davon ausgegangen werden kann, dass eine Fusion stattfindet. Bei Rotaugen (Rutilus rutilus), deren Population durch einen Flaschenhals dezimiert wurde (Artikel I), zeige ich anhand genetischer Analysen, dass Populationen, die durch einen Flaschenhals stark in Anzahl verringert worden waren, eine verringerte genetische Variation besaßen, verglichen mit Populationen von Rotaugen, deren Populationsgröße konstant war. Den Effekt von genetischer Drift auf genetische Variation untersuchte ich an Langzeit-isolierten Populationen von Rotaugen und Flussbarschen (Perca fluviatilis) (Artikel III). Meine Erwartung, dass sich die genetische Variation mit der Zeit verringert, wurde nicht bestätigt. Die genetische Variation zeigte sich konstant während des Zeitraums der Untersuchung. Die effektive Populationsgrösse nahm jedoch für beide Arten ab, dieses Resultat ist konsistent mit dem Einfluss genetischer Drift. Die Verwandtschaftsverhältnisse der europäischen Karpfenartigen im Allgemeinen und im Besonderen der schwedischen, sind teilweise ungeklärt. Mit Hilfe der Sequenzierung eines mitochondriellen und eines nuklearen Gens habe ich die Abstammungsverhältnisse der 18 in Schweden zu findenden Karpfenarten erstellt (Artikel IV). 28.

(221) Zwischen dem mitochondriellen und nuklearem Stammbaum bestand eine Inkongruenz, die auf die unterschiedliche Mutationsgeschwindigkeit und Selektionsgeschichte der beiden Gene zurückzuführen ist. Den Status von Brachsen und Güster als Schwesterarten konnte ich nicht bestätigen. Deren Abstammungsverhältnisse müssen in Zusammenhang mit der Zährte näher geklärt werden. Der Prozess der Artbildung ist die ultimative Quelle neuer Arten. Durch die Analyse von bereits veröffentlichten Phylogenien von diversen Fischarten konnte ich zeigen, dass das Tempo der Aufspaltung von der bereits befindlichen Artendichte im Habitat abhängig ist (density dependent cladogenesis). Das bedeutet, dass die Artbildung in der Anfangsphase explosionsartig erfolgt, um dann mit zunehmender Konkurrenz zwischen den Arten um Habitat, Nahrung etc. abzunehmen, bis ein Gleichgewicht hergestellt ist (Artikel V). Meine Ergebnisse bestätigen ähnliche Schätzungen von Divergenzraten in anderen Taxa (z. B. Vögel) und spiegeln ein generelles Muster des Artbildungsprozesses wider.. 29.

(222) Acknowledgements. Many, many thanks to my supervisors Mats Björklund and Erik Petersson. Mats, for giving me the freedom to develop my own thoughts and for always answering my questions whenever I knocked at your office door. Thanks for your optimism and confidence. Erik, your enthusiasm and compassion for fish deeply impressed me and made me change my view of handling fishes as a study organism. Thank you for being an enormous help when collecting, keeping, injecting, spawning, fertilizing and rearing fish. Emma, thank you for helping me during field work. It would have been impossible without you. I will never forget our numerous fishing tours in tiny boats and with way too many kanelbullar. Thank you for our nonscientific morning chats about children, men, and life in general. Sara, you started out as my exam student. You became a great colleague with who I could exchange fishy ideas. Thank you so much for your help in the field, stirring eggs in the lab, and being a perfect room-mate during conference stays. Johanna, you also wanted to escape your lab work with the beetles and gave me a helping hand during field work. Your way of unsentimentally knocking out a fish led to a re-validation of my first impression of you. Thank you for being a great colleague and friend, for your happy state of mind and for always having a laugh on your lips. Olivia, thank you for being such a good friend, for dragging me out of the lab or office to eat pancakes and for all the nice evenings we had playing games or watched films. Norbert, thank you for being a great colleague, your patience with the students at Klubban deeply impressed me. Thanks to Anders Ö for good laboratory cooperation. Thanks to Ingela for all help with office-related stuff. Thanks to Reija for analyzing samples. Thanks to Maria Traki for trying to analyze the Biotest basin data. Thanks to Laura Winter for analyzing the stomach contents. Thanks to all the people at the Animal Ecology department for creating such a pleasant atmosphere. Special thanks to Ingrid and Anders for encouraging my teaching approaches, and to Niclas K, for looking at hundreds of fish photographs. Thanks to Amber and Sandra for improving my manuscripts. Niclas V, thanks for patiently bearing the morning chats I had with Emma. Last but not least, I would like to thank my friends and family. Ablahad, Gunilla and Mofid, many thanks for all the lunches and dinners we had together, giving the girls the impression, that meeting you equals food consumption! Nina, best sister on earth, you are now only 30 minutes away! 30.

(223) You taught me how to be Queen of Science! Philipp, master of computers, thanks for walking the dog and entertaining the children from while to while. Thanks to my parents for their never ending encouragement and support. I know I can always count on you! Christian, you are the reason why I stayed in Uppsala. Thank you for your ambition to understand what I am doing. Thanks to my daughters Viveca and Linnea, for your never ending ambition to decorate my office with drawings. You are my mitochondrial link into the next generation of the human genealogy.. 31.

(224) References. Adams, D.C., Rohlf, J.F., Slice, D.E. (2004) Geometric morphometrics: Ten years of progress following the 'revolution'. Italian Journal of Zoology 71, 5-16. Appelberg, M., Degerman, E., Norrgren, L. (1992) Effects of acidification and liming on fish in Sweden - a review. Finnish Fisheries Research 13, 13-23. Arnold, M.L. (1997) Natural hybridization and evolution. Oxford University Press. Arnold, M.L., Hodges, S.A. Are natural hybrids fit or unfit relative to their parents? Trends in Ecology and Evolution 10: 67-71. Banarescu, P. (1960) Einige Fragen zur Herkunft und Verbreitung der Süsswasserfischfauna der europäisch-mediterranen Unterregion. Archiv für Hydrobiologie 57, 16-134. Banarescu, P., Coad, B.W. (1991) Cyprinids of Eurasia. In: Cyprinid fishes, systematics, biology and exploitation. I.J. Winfield & J.S. Nelson (ed) p 127-155 Chapman & Hall, London Barluenga, M., Stölting, K.M., Salzburger, W., Muschick, M., Meyer, A. (2006) Sympatric speciation in Nicaraguan crater lake cichlid fish. Nature 439, 719-723. Begon, M., Harper, J.L., Townsend, C.R. (1996), Ecology: individuals, populations and communities, Blackwell Science, Oxford. Bergek, S., Björklund, M. (2007) Cryptic barriers to dispersal within a lake allow genetic differentiation of Eurasian perch. Evolution 61, 2035-2041. Bianco, P.G., Aprea, G., Balletto, E., Capriglione, T., Fulgione, D., Odierna, G. (2004) The karyology of the cyprinid genera Scardinius and Rutilus in southern Europe. Ichthyology Research 51, 274-278. Billard, R. (1999), Carp biology and Culture, Springer. Björklund, M., Bergek, S. (2009) On the relationship between population differentiation and sampling effect: is more always better? Oikos 118, 1127-1129. Bookstein, F.L. (1991), Morphometric tools for landmark data, Cambridge University Press, New York. Borell, Y.J., Bernardo, D., Blanco, G., Vásquez, E., Sánchez, J.A. (2008) Spatial and temporal variation of genetic diversity and estimation of effective population sizes in Atlantic salmon (Salmo salar, L.). 32.

(225) populations from Asturias (Northern Spain) using microsatellites. Conservation Genetics 9, 807-819. Chen, X.-L., Yue, P.Q., Lin, R.D. (1984) Major groups within the family Cyprinidae and their phylogenetic relationships. Acta Zootaxonomica Sinia 9, 424-440. Coyne, J.A., Orr, H.A. (2004) Speciation, Sinauer Associates, Sunderland. Darwin, C., Wallace, A.R. (1858), On the tendency of species to form varieties; and on the perpetuation of varieties and species by natural means of slection. Journal of the Linnean Society of London (Zoology) 3, 45-62. Darwin, C. (1859) On the origin of species by means of natural selection or the preservation of favored races in the struggle for life, J. Murray, London. DeMarais, B..D., Dowling, T.E., Douglas, M.E., Minckley, W.L., Marsh, P.C. (1992) Origin of Gila seminude (Teleostei: Cyprinidae) through introgressive hybridization: implications for evolution and conservation. Proceedings of the National Academy of Sciences of the United States of America 89: 2747-2751. Dowling, T.E., DeMarais, B.D. (1993) Evolutionary significance of introgressive hybridization in cyprinid fishes. Nature 362, 444-446. Eldredge, N., Gould, S.J. (1972) Punctuated equilibria: an alternative to phyletic gradualism. In: Models in paleobiology. T.J.M. Schopf (ed) p. 82115 Freeman, Cooper and Company, San Francisco, CA. Freyhof, J., Lieckfeldt, D., Pitra, C., Ludwig, A. (2005) Molecules and morphology: Evidence for introgression of mitochondrial DNA in Dalmatian cyprinids. Molecular Phylogenetics and Evolution 37, 347-354. Froese, R., Pauly D. (2007) FishBase. http://www.fishbase.org Gebhardt, H., Ness, A. (1998), Fische: die heimischen Süsswasserfische sowie Arten der Nord- und Ostsee, BLV Naturführer. Grant, P.R., Grant, B.R. (1992) Hybridization of bird species. Science 256, 193-197. Hamilton, P.B., Tyler, C.R. (2008) Identification of microsatellite loci for parentage analysis in roach Rutilus rutilus and eight other cyprinid fish by cross-species amplification, and a novel test for detecting hybrids between roach and other cyprinids. Molecular Ecology Resources 8, 462-465. Hänfling, B., Brandl, R. (2000) Phylogenetics of European cyprinids: insights from allozymes. Journal of Fish Biology 57, 265-276. Hänfling, B., Bolton, P., Harley, M., Carvalho, G.R. (2005) A molecular approach to detect hybridisation between crucian carp (Carassius carassius) and non-indidenous carp species (Carassius spp. and Cyprinus carpio). Freshwater Biology 50, 403-417. Howes, G.J. (1991), Systematics and biogeography: an overview. In: Cyprinid Fishes Systematics, biology and exploitation (ed. by I.J. Winfield. & J.S. Nelson (eds) pp 1-33. Chapman & Hall.. 33.

No results found