Population genetic structure and demographic history of the fat greenling Hexagrammos otakii

Fat greenling (Hexagrammos otakii) is an important commercial fish in the Northwestern Pacific, being distributed along the coastal waters of the East/Japan Sea and the Yellow Sea. To investigate population genetic structure and demographic history of this species, one hundred and fifty five individuals were collected from five localities in the distribution range of the species and sequence variations in the mitochondrial genes COI, COIII-ND3-ND4L, and cytochrome b were examined. For all the genes in every sampling location, the nucleotide diversities were very low (0.001 ～ 0.005) although the haplotype diversities were relatively high, 0.55 ～ 0.81 for COI, 0.79 ～ 0.84 for COIII-ND3-ND4L, and 0.95 ～ 0.97 for cytochrome b. Hierarchical analysis of molecular variance (AMOVA), the conventional population statistic FST, and exact test of population differentiation revealed no significant genetic structuring among the samples, indicating that fat greening is panmictic throughout the range of its distribution. Neutrality tests such as Tajima’s D and Fu’s FS statistics and mismatch distribution analyses suggested that fat greening has undergone the demographic history of population expansion during the late Pleistocene period approximately 91,000 ～ 327,000 years ago. The star-burst patterns of haplotype networks and low nucleotide diversities also indicated recent population expansion. These results help establish the fisheries management strategy for fat greenling in the Northwestern Pacific.     

Isolation and characterization of polymorphic microsatellite loci from fat greenling (Hexagrammos otakii)

Fat greenling (Hexagrammos otakii) is an economocally important marine fish species. Thirtyfive microsatellite loci were isolated from two enriched genomic library of Hexagrammos otakii. Ten of these loci were polymorphic in a test population with alleles per locus ranging from two to six, and observed and expected heterozygosities per locus from 0.2581 to 1.0000 and from 0.2892 to 0.7726, respectively. Four loci significantly deviated from Hardy–Weinberg equilibrium after Bonferroni correction and no significant linkage disequilibrium was found between pairs of loci. These polymorphic microsatellite loci would be useful for investigating genetic diversity of Hexagrammos otakii.     

Genetic structure and historical demography of Collichthys lucidus inferred from mtDNA sequence analysis

Fluctuations of global climate changes during the Pleistocene had a huge impact on the marginal seas of the Northwestern Pacific. To examine historical demography and the population genetic structure of Collichthys lucidus, a total of 151 individuals from 7 locations were sequenced at the 5′ end of mitochondrial DNA control region and 380-bp fragments were obtained. The results of AMOVA showed that the genetic variation among the two groups was 78.39 % (P?=?0.78) and among populations within groups was 0.43 % (P?=?0.02). Both mismatch distribution analysis and neutrality tests showed C. lucidus has experienced a recent population expansion. Three distinct clades suggested that C. lucidus might be isolated in the three marginal seas during the Pleistocene periods. No genetic divergence within groups should be explained by insufficient time to attain migration-drift equilibrium. The changes in sea level during the Pleistocene had a major effect on the C. lucidus phylogeographical pattern because it limited its dispersal and at times created isolated populations.     

秦岭细鳞鲑群体遗传结构

秦岭细鳞鲑(Brachymystax lenok tsinlingensis)是秦岭地区特有鱼类,近年来由于环境恶化及人类活动的加剧,已对其造成严重影响,种群处于濒危状态,因此研究秦岭细鳞鲑的群体遗传结构、演化历史、分布动态等对其进行有效保护具有重要意义。本研究用线粒体D-loop区序列对秦岭地区6个群体(n=112)进行了遗传结构和群体演化分析。D-loop区扩增出的891bp序列在112个个体中,检测到42个变异位点,共26个单倍型;碱基序列总的单倍型多样度较高为0.883,核苷酸多样度为0.00799。AMOVA分析显示,60.05%的分子差异位于群体内,39.95%的分子差异位于群体间,Fst值统计检验表明,除那布大河群体与漳河群体和千河群体之间差异不显著之外,其余两两群体之间Fst值统计检验均为显著。系统树和单倍型网络图分析表明,6个地理群体的单倍型按照渭河上游和渭河中游两个河段形成两个大的类群,且5个群体共享一个单倍型H2,表明这些群体具有相同的演化历史,为同一个祖先群体演化而来。中性检验和歧点分布显示,秦岭细鳞鲑种群大小保持相对稳定,未经历明显的种群扩张。同时建议将渭河上游秦岭细鳞鲑群体作为一个整体进行重点保护。     

Genetic structure and diversity of green sea turtle (Chelonia mydas) from South China Sea inferred by mtDNA control region sequence

We analyzed 88 control region sequences of green sea turtle (Chelonia mydas) from around Hainan Island in the South China Sea. These sequences had a length of 489 bp and revealed 8 mtDNA haplotypes of which four haplotypes (CMC1, CMC4, CMC7, and CMC8) had not been discovered before. Haplotype diversity (h) and nucleotide diversity (π) were 0.45 ± 0.054 and 0.0035 ± 0.0014, respectively. Neighbor-Joining tree based on control region sequences revealed that genetic relationship between green sea turtles from the South China Sea and from Japan Sea were very close. Clustering relationship based on control region sequences indicated that the South China Sea is an important breeding site and feeding habitat for green sea turtles, which connects with the Middle East Pacific, the Southwest Pacific and the Indian Ocean.     

线粒体序列分析黑龙江流域哲罗鲑的种群遗传结构

哲罗鲑是我国土著的重要经济鱼类,近些年来,由于环境的恶化及人类活动的加剧,已对其资源量、栖息地、产卵场等造成了严重的破坏,种群处于濒危状态,被列入濒危物种红色名录中,因此研究哲罗鲑的群体遗传结构、演化历史、分布动态等对其进行有效保护具有重要意义.用线粒体Cox1和ND1基因序列对黑龙江流域内9个群体(n＝30)进行了遗传结构、群体演化分析.在30个个体中,Cox1扩增出1550bp的片段,群体间序列分歧距离在0～0.0028之间,共发现10个单倍型;ND1基因扩增出1000bp的片段,群体间序列分歧距离在0～0.0013之间,共发现7个单倍型.单倍型网络(TCS)和AMOVA分析结果均表明黑龙江流域哲罗鲑存在显著的群体分化,Cox1基因、ND1基因及组合数据的群体间Fst分别为0.0704、0.0491、0.0792,均达到显著性水平(P<0.05),根据配对群体间Fst(Pairwised Fst)可以将黑龙江流域哲罗鲑群体分为黑龙江上游群体、呼玛河群体、乌苏里江群体、内蒙古伊敏河上游群体4个地理群.9个群体共享同一个单倍型(BH11),表明这些群体具有相同的演化历史,为同一个祖先群体(黑龙江上游群体)演化而来.     

四川地区猕猴线粒体DNA 控制区遗传多样性及其种群遗传结构

猕猴是最理想的医学实验灵长类动物,并且是国家二级保护动物。四川地区的猕猴数量多、分布广,全面了解其遗传背景对于该地区猕猴资源的保护与合理利用具有重要的意义。本研究对来自四川8 个地理种群的231个不同猕猴个体的线粒体DNA 控制区部分序列进行了测定和群体分析,发现了110 个变异位点(22. 49% ),定义了56 种单倍型,其单倍型多样性(h)平均值为0. 686、核苷酸多样性(π)平均值为0. 01483,种群总体遗传多样性较高;进一步分析表明,8 个地理种群间存在着明显的遗传分化(F_st = 0. 70412,P < 0. 05),种群间基因交流较低(N_m < 1);系统发育树显示,四川猕猴8 个地理种群的单倍型基本上成簇分布在系统树上,与地理位置呈现一定的对应关系,说明四川猕猴具有明显的系统地理分布格局。地理隔离和人类活动可能是促使四川猕猴种群分化的主要因素。     

Minding the gap: frequency of indels in mtDNA control region sequence data and influence on population genetic analyses

Insertions and deletions (indels) result in sequences of various lengths when homologous gene regions are compared among individuals or species. Although indels are typically phylogenetically informative, occurrence and incorporation of these characters as gaps in intraspecific population genetic data sets are rarely discussed. Moreover, the impact of gaps on estimates of fixation indices, such as F(ST), has not been reviewed. Here, I summarize the occurrence and population genetic signal of indels among 60 published studies that involved alignments of multiple sequences from the mitochondrial DNA (mtDNA) control region of vertebrate taxa. Among 30 studies observing indels, an average of 12% of both variable and parsimony-informative sites were composed of these sites. There was no consistent trend between levels of population differentiation and the number of gap characters in a data block. Across all studies, the average influence on estimates of PhiST was small, explaining only an additional 1.8% of among population variance (range 0.0-8.0%). Studies most likely to observe an increase in PhiST with the inclusion of gap characters were those with < 20 variable sites, but a near equal number of studies with few variable sites did not show an increase. In contrast to studies at interspecific levels, the influence of indels for intraspecific population genetic analyses of control region DNA appears small, dependent upon total number of variable sites in the data block, and related to species-specific characteristics and the spatial distribution of mtDNA lineages that contain indels.     

Global population structure of the tope (Galeorhinus galeus) inferred by mitochondrial control region sequence data

In order to properly manage and conserve exploited shark species, detailed analyses of their population structure is needed. Global populations of Galeorhinus galeus are in decline due to the exploitation of the fishery over the past 80 years. Currently, the genetic structure of eastern Pacific populations of G. galeus is not known and recent observations in the northeastern Pacific suggest an increase in numbers. To evaluate gene flow among populations of G. galeus , 116 samples were collected and analysed from six geographically dispersed locations: Australia, North America, South Africa, South America (Argentina and Peru), and the UK. Analysis of 968 to 1006 bp of the 1068-bp mitochondrial control region revealed 38 unique haplotypes that were largely restricted to their collecting locality. Significant genetic structure was detected among populations (Φ_ST = 0.84; P  < 0.000001) and migration estimates were low ( Nm  = 0.05–0.97). Due to an apparent lack of migration, populations of G. galeus appear to be isolated from each other with little to no gene flow occurring among them. As a consequence of this isolation, increasing numbers of G. galeus in the northeastern Pacific can be best explained by local recruitment and not by input from geographically distant populations.     

Lack of genetic structure in endangered large yellow croaker Larimichthys crocea from China inferred from mitochondrial control region sequence data

The large yellow croaker Larimichthys crocea is an important commercial marine fish species in China. However, information about the population structure of this species is limited. In the present study, mitochondrial DNA (mtDNA) control region was sequenced from four populations of the yellow croaker in the southern Yellow Sea, East China Sea and South China Sea to investigate the genetic structure of this species. A total of 54 haplotypes were detected among 62 individuals of large yellow croaker. High levels of population genetic diversity were observed. Among the four populations, the haplotype diversity was between 0.9895 ± 0.0193 (Xiamen) and 1.0000 ± 0.0524 (Lvsi, Zhoushan). The nucleotide diversity ranged from 0.0208 ± 0.0108 (Xuwen) to 0.0246 ± 0.0138 (Lvsi). The results of AMOVA detected no significant differences among populations. The conventional F_ST statistics were negative and insignificant values. These indicated lack of population genetic structure throughout the Yellow Sea, East China Sea and South China Sea, and random mixing of individuals among the samples. Biological characters of large yellow croaker and lack of physical barrier in the studied area could be the reasons for lack of genetic structure in this species.     

基于线粒体控制区的序列变异分析中国东南部沿海凤鲚种群遗传结构

为了解中国东南部沿海凤鲚(Coilia mystus)的种群遗传多样性和遗传结构, 本文分析了长江口(CJ)、钱塘江口(QT)、闽江口(MJ)和九龙江口(JL)4个凤鲚地理群体的mtDNA控制区561 bp片段的序列变异。65尾样本共检测到28个单元型。4个群体总的单元型多样性和核苷酸多样性均较高(h = 0.9433 ± 0.0168, π = 0.0317 ± 0.0158), 但单个群体的核苷酸多样性水平却很低, 其中以CJ最高(π = 0.0080 ± 0.0046), MJ最低(π = 0.0015 ± 0.0013)。MJ与JL群体之间以及CJ与QT群体之间的平均K2P遗传距离很小, 分别为0.3%和0.8%; 而CJ、QT分别与MJ、JL群体之间的遗传距离均较大, 达到了6%。采用最大似然法(ML)、最大简约法(MP)和邻接法(NJ)分别构建的单元型间的系统发育树揭示, 4个凤鲚群体构成CJ-QT 和MJ-JL 2个支系, 且具有极高的支持率。单元型的网络分析也显示这两个支系间有高达28步的突变次数。AMOVA分析显示大部分的遗传变异来自这两支系群体间(90.77%), 表明凤鲚群体间存在着显著的地理分化。种群分化指数和基因流分析也表明, 支系间群体有着明显的遗传分化(F_ST > 0.9, N_m < 0.03)。所有分析结果支持所研究的凤鲚标本属于两个不同的地理种群, 且种群的分化至少已达到亚种水平。采用BEAST和TRACER软件得到凤鲚两个亚种的最近共同祖先约在0.34–0.46百万年前, 处于更新世晚期, 推测可能是第四纪晚期的气候旋回和海平面的升降导致了凤鲚的种群分化。     

Population structure of the yellow-footed rock-wallaby Petrogale xanthopus (Gray, 1854) inferred from mtDNA sequences and microsatellite loci

The yellow-footed rock-wallaby Petrogale xanthopus is considered to be potentially vulnerable to extinction. This wallaby inhabits naturally disjunct rocky outcrops which could restrict dispersal between populations, but the extent to which that occurs is unknown. Genetic differences between populations were assessed using mitochondrial DNA (control region) sequencing and analysis of variation at four microsatellite loci among three geographically close sites in south-west Queensland (P. x. celeris) and, for mtDNA only, samples from South Australia (P. x. xanthopus) as well. Populations from South Australia and Queensland had phylogenetically distinct mtDNA, supporting the present classification of these two groups as evolutionarily distinct entities. Within Queensland, populations separated by 70 km of unsuitable habitat differed significantly for mtDNA and at microsatellite loci. Populations separated by 10 km of apparently suitable habitat had statistically homogeneous mtDNA, but a significant difference in allele frequency at one microsatellite locus. Tests for Hardy-Weinberg equilibrium and micro-geographical variation at microsatellite loci did not detect any substructuring between two wallaby aggregations within a colony encircling a single rock outcrop. Although the present study was limited by small sample sizes at two of the three Queensland locations examined, the genetic results suggest that dispersal between colonies is limited, consistent with an ecological study of dispersal at one of the sites. Considering both the genetic and ecological data, we suggest that management of yellow-footed rock-wallabies should treat each colony as an independent unit and that conservation of the Queensland and South Australian populations as separate entities is warranted.     

Population genetic structure of Chilo suppressalis (Walker) (Lepidoptera: Crambidae): strong subdivision in China inferred from microsatellite markers and mtDNA gene sequences

Chilo suppressalis (Walker) displays significant geographical differences in ecological preference that may be congruent with patterns of molecular variation. To test this, we collected and analysed 381 individuals of this species from cultivated rice at 18 localities in China during the rice-growing season of 2005–2006. We used four microsatellite DNA markers and four mitochondrial DNA gene fragments. We found that this species is highly differentiated, coupled with an estimated population expansion date of at least 60 000  bp . Phylogenetic analyses, Bayesian clustering, and phylogeographical analyses of statistical parsimony haplotype network consistently divided the populations into three clades: a central China (CC) clade, a northern plus northeastern China (NN) clade and a southwestern China (SW) clade. Analysis of molecular variance indicated a high level of geographical differentiation at different hierarchical levels [ F _ST for microsatellite markers, COI, COII, 16S and ND1 is 0.06004 ( P  < 0.0001), 0.27607 ( P  < 0.0001), 0.22949 ( P  < 0.0001), 0.19485 ( P  < 0.0001) and 0.29285 ( P  < 0.0001), respectively]. Isolation by distance appeared among the samples from within China ( r  = 0.404, P  = 0.0002); N_em values estimated using a coalescent-based method were small (< 2 migrants per generation), suggesting that the observed levels of differentiation are a result of migration–drift equilibrium. Our results imply that the genetic differentiation of this borer, which is approximately in accordance with its observed number of generations per year in different Chinese geographical regions, is probably attributed to climatic and/or geological events (e.g. the last glacial maximum) and subsequently strengthened by the domestication of rice.     

Phylogeography and population structure of the Yunnan snub-nosed monkey (Rhinopithecus bieti) inferred from mitochondrial control region DNA sequence analysis

Rhinopithecus bieti, the Yunnan snub-nosed monkey, is the nonhuman primate with the highest altitudinal distribution and is also one of the 25 most globally endangered primate species. Currently, R. bieti is found in forests between 3000 and 4500 m above sea level, within a narrow area on the Tibetan Plateau between the Yangtze and Mekong rivers, where it is suffering from loss of habitat and shrinking population size (approximately 1500). To assess the genetic diversity within this species, its population structure and to infer its evolutionary history, we sequenced 401 bp of the hypervariable I (HVI) segment from the mitochondrial DNA control region (CR) for 157 individuals from 11 remnant patches throughout the fragmented distribution area. Fifty-two variable sites were observed and 30 haplotypes were defined. Compared with other primate species, R. bieti cannot be regarded as a taxon with low genetic diversity. Phylogenetic analysis partitioned haplotypes into two divergent haplogroups (A and B). Haplotypes from the two mitochondrial clades were found to be mixed in some patches although the distribution of haplotypes displayed local homogeneity, implying a strong population structure within R. bieti. Analysis of molecular variance detected significant differences among the different geographical regions, suggesting that R. bieti should be separated into three management units (MUs) for conservation. Based on our results, it can be hypothesized that the genetic history of R. bieti includes an initial, presumably allopatric divergence between clades A and B 1.0-0.7 million years ago (Ma), which might have been caused by the Late Cenozoic uplift of the Tibetan Plateau, secondary contact after this divergence as a result of a population expansion 0.16-0.05 Ma, and population reduction and habitat fragmentation in the very recent past.     

Native South American genetic structure and prehistory inferred from hierarchical modeling of mtDNA

Genetic diversity in Native South Americans forms a complexpattern at both the continental and local levels. In comparingthe West to the East, there is more variation within groupsand smaller genetic distances between groups. From this pattern,researchers have proposed that there is more variation in theWest and that a larger, more genetically diverse, founding populationentered the West than the East. Here, we question this characterizationof South American genetic variation and its interpretation.Our concern arises because others have inferred regional variationfrom the mean variation within local populations without takinginto account the variation among local populations within thesame region. This failure produces a biased view of the actualvariation in the East. In this study, we analyze the mitochondrial DNA sequence betweenpositions 16040 and 16322 of the Cambridge reference sequence.Our sample represents a total of 886 people from 27 indigenouspopulations from South (22), Central (3), and North America(2). The basic unit of our analyses is nucleotide identity bydescent, which is easily modeled and proportional to nucleotidediversity. We use a forward modeling strategy to fit a seriesof nested models to identity by descent within and between allpairs of local populations. This method provides estimates ofidentity by descent at different levels of population hierarchywithout assuming homogeneity within populations, regions, orcontinents. Our main discovery is that Eastern South America harbors moregenetic variation than has been recognized. We find no evidencethat there is increased identity by descent in the East relativeto the total for South America. By contrast, we discovered thatpopulations in the Western region, as a group, harbor more identityby descent than has been previously recognized, despite thefact that average identity by descent within groups is lower.In this light, there is no need to postulate separate foundingpopulations for the East and the West because the variabilityin the East could serve as a source for the Western gene pools.     

Population genetic structure and historical demography of grey mullet, Mugil cephalus, along the coast of China,inferred by analysis of the mitochondrial control region

Population genetic structure and historical demography of grey mullet, Mugil cephalus, along the coast of China were examined with a 389-bp segment of mtDNA control region. In total, 117 samples were collected from seven locations and 77 haplotypes were obtained. No haplotype was concurrently presented in all the samples, and samples in Southern and Northern regions had distinctly different haplotypes. Sorting of the data cladistically into phylogenetic trees indicated that these haplotypes can be divided into two main clades. Analysis of molecular variance (AMOVA) indicated significant genetic differentiation among subpopulations in different regions (φ_CT = 0.9505, p < 0.01). The average pairwise differences and φ_ST values (0.8943–0.9464) between regional populations were significant. The net Tumura and Nei genetic distance (0.5577 ± 0.1421) between the two groups was large. The deep and unique divergence between the two groups suggested that each group is likely to represent a distinct species. In contrast to the great genetic divergence on a broad scale, small but significant population subdivisions at a less spatial scale were also detected both in Southern and Northern regions, which support the conclusion that M. cephalus has limited dispersal potential. High haplotype diversity (h = 0.8716 ± 0.1462) and low nucleotide diversity (π = 0.0046 ± 0.0030) were found among the three Northern samples. Among the four Southern samples, however, extraordinarily high levels of both haplotype diversity (h = 0.9886 ± 0.0057) and nucleotide diversity (π = 0.0425 ± 0.0212) were detected. The different polymorphisms suggest an apparent distinguishing demographic history between regionally Southern and Northern populations, though mismatch distribution analyses and neutrality tests implied a late Pleistocene population expansion of both Northern and Southern populations.     

大石鸡兰州亚种的mtDNA种群遗传结构、单倍型分布和遗传多样性

大石鸡(Alectorismagna)是中国西北部的特有种。我们测定了大石鸡兰州亚种(A.m.lanzhouensis)8个地理种群106个样本的线粒体DNA控制区5′端458bp序列,研究其种群遗传结构和遗传多样性。27个变异位点共确定25种单倍型,其中单倍型M2广泛分布,而许多单倍型为一些地方种群特有。单倍型分布沿着南北方向变化,存在明显的地理结构。8个种群中核苷酸多样性最高的是定西种群,0.0069,最低的是海原种群,0.0028;单倍型多样性最高的是武山种群,0.86,最低的是北道种群,0.52。北方种群比南方种群具有更高的遗传多样性。系统发生树和单倍型分布表明,大石鸡兰州亚种存在两个明显的分支。溯祖理论、更新世冰期和花粉支持兰州亚种起源于兰州盆地,这个盆地是其遗传多样性的中心。     

华南沿海西部美丽小条鳅基于线粒体控制区的种群遗传变异及亲缘地理格局

为研究其种群遗传变异和亲缘地理格局,分析了107尾采自华南西部和海南岛的12条水系的美丽小条鳅(Micronoema-cheilus pulcher Nichols)控制区934-938 bp的序列,其中有79个核苷酸变异位点。分子变异分析(AMOVA)表明,种群间的遗传变异占46.88%,种群内的遗传变异占55.06%。基于36个单倍型的系统树显示,12条水系的种群聚成两支。其中,广西沿海诸独立水系(防城河、峒中河、北仑河、南流江)和西江水系与广东漠阳江和潭江水系关系密切,而海南岛万泉河和南渡江与广东鉴江水系关系密切。根据嵌套进化枝系地理分析(NCPA)推测,防城河周边地区可能是美丽小条鳅的扩散中心,该物种可由此区域通过两条途径扩散:(1)沿西江水系向广西沿海独立水系至广东漠阳江和潭江水系扩散;(2)向海南岛诸水系再至雷州半岛的鉴江水系扩散。在演化过程中,曾发生片断化事件、长距离建群和持续的分布区扩张。     

Biogeographical patterns of genetic differentiation in dung beetles of the genus Trypocopris (Coleoptera, Geotrupidae) inferred from mtDNA and AFLP analyses

Aim To examine the phylogeography and population structure of three dung beetle species of the genus Trypocopris (Coleoptera, Geotrupidae). We wanted to test whether genetic differences and genealogies among populations were in accordance with morphologically described subspecies and we aimed to establish times of divergence among subspecies to depict the appropriate temporal framework of their phylogeographical differentiation. We also wished to investigate the historical demographic events and the relative influences of gene flow and drift on the distribution of genetic variability of the different populations. Location Europe (mostly Italy). Methods We collected adult males from dung pats from 15 Italian localities over the period 2000–2002. For sequence analysis, some dried specimens from Albania, Croatia, Slovakia and Spain were also used. We applied cytochrome oxidase I mitochondrial DNA sequencing and the amplified fragment length polymorphism (AFLP) technique to determine whether phylogeographical patterns within the three species support the proposed hypotheses of subspecies designations, and to detect further structure among populations that might mediate diversification. Results and main conclusions The results show a high concordance between the distribution of mtDNA variation and the main morphological groups recognized as subspecies, which thus may represent independent evolutionary units. The degree of mitochondrial divergence suggests that speciation events occurred during the Pliocene, while diversification of the main subspecific lineages took place in the Pleistocene, from c. 0.3 to 1.5 Ma. Mitochondrial and nuclear data also reveal that there is phylogeographical structuring among populations within each of the main groups and that both contemporary and historical processes determined this pattern of genetic structure. Geographical populations form monophyletic clades in both phylogenetic and network reconstructions. Despite the high levels of intrapopulational diversity, F_ST values indicate moderate but significant genetic differentiation among populations, and a Bayesian clustering analysis of the AFLP data clearly separates the geographical populations. Nucleotide and gene diversity estimates reveal interspecific differences in the degree of diversification among populations that may be related to the different ecological requirements of the three species.