Effective number of breeding adults in Bufo bufo estimated from age-specific variation at minisatellite loci

Estimates of the effective number of breeding adults were derived for three semi-isolated populations of the common toad Bufo bufo based on temporal (i.e. adult-progeny) variance in allele frequency for three highly polymorphic minisatellite loci. Estimates of spatial variance in allele frequency among populations and of age-specific measures of genetic variability are also described. Each population was characterized by a low effective adult breeding number ( N _b) based on a large age-specific variance in mini-satellite allele frequency. Estimates of N _b (range 21–46 for population means across three loci) were ≊ 55–230-fold lower than estimates of total adult census size. The implications of low effective breeding numbers for long-term maintenance of genetic variability and population viability are discussed relative to the species' reproductive ecology, current land-use practices, and present and historical habitat modification and loss. The utility of indirect measures of population parameters such as N _b and N _e based on time-series data of minisatellite allele frequencies is discussed relative to similar measures estimated from commonly used genetic markers such as protein allozymes.     

Stability of genetic structure and effective population size inferred from temporal changes of microsatellite DNA polymorphisms in the land snail Helix aspersa (Gastropoda: Helicidae)

Temporal evolution of genetic variability may have far-reaching consequences for a diverse array of evolutionary processes. Within the polders of the Bay of Mont-Saint-Michel (France), populations of the land snail Helix aspersa are characterized by a metapopulation structure with occasional extinction processes resulting from farming practices. A temporal survey of genetic structure in H . aspersa was carried out using variability at four microsatellite loci, in ten populations sampled two years apart. Levels of within-population genetic variation, as measured by allelic richness, H _e or F _is , did not change over time and similar levels of population differentiation were demonstrated for both sampling years. The extent of genetic differentiation between temporal samples of the same population established (i) a stable structure for six populations, and (ii) substantial genetic changes for four populations. Using classical F -statistics and a maximum likelihood method, estimates of the effective population size ( N _e) illustrated a mixture of stable populations with high N _e, and unstable populations characterized by very small N _e estimates (of 5–11 individuals). Owing to human disturbances, intermittent gene flow and genetic drift are likely to be the predominant evolutionary processes shaping the observed genetic structure. However, the practice of multiple matings and sperm storage is likely to provide a reservoir of variability, minimizing the eroding genetic effects of population size reduction and increasing the effective population size.  © 2004 The Linnean Society of London, Biological Journal of the Linnean Society , 2004, 82 , 89–102.     

Loss of genetic variation and effective population size of Kirikuchi charr: implications for the management of small, isolated salmonid populations

Small, isolated populations may face extinction due to a combination of inbreeding depression and other threats. Effective population size ( N _e) is one comprehensive measure that allows us to evaluate the genetic status of a population, and to make management decisions regarding genetic viability. We simulated loss of genetic variation and estimated N _e for two small, isolated populations of Kirikuchi charr Salvelinus leucomaenis japonicus , the endangered, southernmost local populations of the genus Salvelinus in the world, using VORTEX, an individual-based stochastic PVA model. Approximately half of the genetic variation was lost over 200 years regardless of census population size and demographic parameters, and N _e estimates were roughly 50 in each of the two populations, suggesting the possibility of inbreeding depression. The target population size of N _e>500, by securing long-term viability, is several times that of the present size of each of the populations studied, and no local habitats maintaining such a target number are considered to exist. The results strongly indicate a need for recovering natural connections and potential gene flow among local populations. However, the impending threat to these populations from non-native charr widely distributed throughout the drainage has prevented the recovery of the connections. Given the small N _e of the two populations, it would be necessary to retain gene flow artificially within or across local populations. This will be true of many other salmonid populations that have been isolated or fragmented recently.     

Relative contribution of contemporary pollen and seed dispersal to the effective parental size of seedling population of California valley oak (Quercus lobata, Née)

For plant populations, gene movement through pollen and seed dispersal governs the size of local genetic neighbourhoods and shapes the opportunities for natural selection and genetic drift. A critical question is how together these two processes influence the evolutionary dynamics of local populations. To assess the respective contributions of pollen and seed flow, we propose a novel indirect assessment of the separate male and female gametic contributions to total effective parental size ( N _e), based on parental correlations estimated via kinship coefficients, that can be applied to data sets that include unambiguous genotypes for male and female gametic contributions. Using the endemic Californian valley oak ( Quercus lobata ) as our study species, we apply this method to a set of microsatellite genotypes for two distinct ecological sets of naturally recruiting seedlings with acorns attached. We found that the effective numbers of contributing male parents ( N _ep) exceed effective numbers of female parents ( N _em) for seedlings established beneath adult trees ( N _ep = 8.1 and N _em = 1.1), as well as for seedlings established away from adult trees ( N _ep = 15.4 and N _em = 2.7), illustrating that seed dispersal enhances pollen dispersal and increases the effective number of seed sources in open seedling patches. The resulting effective parental size of seedling populations translates into smaller effective numbers of parents for undispersed vs. dispersed seedlings ( N _e = 3.6 and N _e = 6.7, respectively). This study introduces a novel statistic method and provides important new evidence that, on a short-term temporal scale, seed dispersal shapes the local neighbourhood size of new recruits.     

Genetic estimates of contemporary effective population size: what can they tell us about the importance of genetic stochasticity for wild population persistence?

Genetic stochasticity due to small population size contributes to population extinction, especially when population fragmentation disrupts gene flow. Estimates of effective population size ( N _e) can therefore be informative about population persistence, but there is a need for an assessment of their consistency and informative relevance. Here we review the body of empirical estimates of N _e for wild populations obtained with the temporal genetic method and published since Frankham's (1995 ) review. Theoretical considerations have identified important sources of bias for this analytical approach, and we use empirical data to investigate the extent of these biases. We find that particularly model selection and sampling require more attention in future studies.
We report a median unbiased N _e estimate of 260 (among 83 studies) and find that this median estimate tends to be smaller for populations of conservation concern, which may therefore be more sensitive to genetic stochasticity. Furthermore, we report a median N _e/ N ratio of 0.14, and find that this ratio may actually be higher for small populations, suggesting changes in biological interactions at low population abundances. We confirm the role of gene flow in countering genetic stochasticity by finding that N _e correlates strongest with neutral genetic metrics when populations can be considered isolated. This underlines the importance of gene flow for the estimation of N _e, and of population connectivity for conservation in general. Reductions in contemporary gene flow due to ongoing habitat fragmentation will likely increase the prevalence of genetic stochasticity, which should therefore remain a focal point in the conservation of biodiversity.     

Genetic diversity in the endangered dysploid larkspur Delphinium bolosii and its close diploid relatives in the series Fissa of the Western Mediterranean area

Allozyme diversity was evaluated in four closely-related taxa of the Delphinium series Fissa distributed throughout the Western Mediterranean area. All are considered threatened plants. Delphinium bolosii and Delphinium mansanetianum are narrowly endemic to the Eastern Iberian Peninsula, whereas Delphinium fissum ssp. sordidum is found in a few populations across the Peninsula. Delphinium fissum ssp. fissum is more widely distributed but often in small and isolated populations. In this group, Delphinium bolosii is dysploid (2 n  = 18) whereas the other taxa are diploid (2 n  = 16). A total of 12 populations were surveyed, including all known locations for D. bolosii , D. mansanetianum , and D. fissum ssp. sordidum . Eleven enzyme systems were assayed and 15 loci were resolved. Markedly depauperate values for genetic diversity were obtained for D. mansanetianum ( H _e = 0.013) and D. fissum ssp. sordidum ( H _e = 0.044). The estimates for D. fissum ssp. fissum ( H _e = 0.071) were below the values expected for widespread species. Small population size and marginal distribution have probably contributed to the low variability observed in this group. By contrast, D. bolosii exhibited comparatively larger populations and greater genetic diversity ( H _e = 0.138). We suggest that, apart from population size and local adaptation, genetic diversity during speciation may have been promoted by dysploidy through genomic recombination.  © 2007 The Linnean Society of London, Biological Journal of the Linnean Society , 2007, 92 , 773–784.     

Population genetic structure and effective population size of ayu (Plecoglossus altivelis), an amphidromous fish

The temporal and spatial population genetic structure of ayu Plecoglossus altivelis (Salmoniformes: Plecoglossidae), an amphidromous fish, was examined using analysis of variation at six microsatellite DNA loci. Intracohort genetic diversities, as measured by the number of alleles and heterozygosity, were similar among six cohorts (2001–2006) within a population (Nezugaseki River), with the mean number of alleles per cohort ranging from 11·0 to 12·5 and the expected heterozygosity ranging from 0·74 to 0·77. Intrapopulational genetic diversities were also similar across the three studied populations along the 50 km coast, with the mean number of alleles and the expected heterozygosity ranging from 11·33 to 11·67 and from 0·75 to 0·76, respectively. The authors observed only one significant difference in pair-wise population differentiation ( F _ST-value) between the cohorts within a population and among three populations. Estimates of the effective population size ( N _e) based on maximum-likelihood method yielded small values (ranging from 94·8 to 135·5), whereas census population size ranged from c. 4800 to 24 000. As a result, the ratio of annual effective population sizes to census population size ( N _e/ N ) ranged from 0·004 to 0·023. These estimates of N _e/ N agree more closely with estimates for marine fishes than that of the larger estimates for freshwater fishes. The present study suggests that ayu which is highly fecund and shows low survival during the early life stages is also characterized by having low value of N _e/ N , similar to marine species with a pelagic life cycle.     

Contrasting population structures in two sympatric anurans: implications for species conservation

A general prediction of the neutral theory of evolution is that genetic diversity should correlate positively with effective population size. We show here that diversity across eight microsatellite loci was consistently and substantially lower in one common amphibian (Bufo bufo) than in another with similar life history traits (Rana temporaria) despite B. bufo having the larger breeding assemblage sizes. However, B. bufo breeding assemblages were much more highly differentiated than those of R. temporaria according to both Fst and Rst estimators. These differences occurred in shared habitats across identical geographical distances. The patterns of genetic diversity and differentiation detected in these two species were probably a consequence of high gene flow in R. temporaria but much lower gene flow among the larger but more dispersed B. bufo assemblages. These observations highlight the difficulty of defining the boundaries of wild populations, and show how two broadly similar species can exhibit very different population dynamics.     

Contrasting levels of genetic diversity between the common, self-compatible Liparis kumokiri and rare, self-incompatible Liparis makinoana (Orchidaceae) in South Korea

Levels of allozyme variation and intrapopulation spatial genetic structure of the two terrestrial clonal orchids Liparis kumokiri , a self-compatible relatively common species, and L. makinoana , a self-incompatible rare species, were examined for 17 ( N  = 1875) and four ( N  = 425) populations, respectively, in South Korea. Populations of L. makinoana harboured high levels of genetic variation ( H _e = 0.319) across 15 loci. In contrast, L. kumokiri exhibited a complete lack of allozyme variation ( H _e = 0.000). Considering the lack of genetic variability, it is suggested that current populations of L. kumokiri in South Korea originated from a genetically depauperate ancestral population. For L. makinoana , a significant deficit of heterozygosity (mean F _IS = 0.198) was found in population samples excluding clonal ramets, suggesting that pollen dispersal is localized, generating biparental inbreeding. The significant fine-scale genetic structuring (≤ 2 m) found in a previous study, in addition to the moderate levels of population differentiation ( F _ST = 0.107) and the significant relationship between genetic and geographical distances ( r  = 0.680) found here, suggests a leptokurtic distribution of seed dispersal for L. makinoana . Although populations of L. makinoana harbour high levels of genetic variation, they are affected by a recent genetic bottleneck. This information suggests that genetic drift and limited gene flow could be the main evolutionary forces for speciation of a species-rich genus such as Liparis .  © 2007 The Linnean Society of London, Botanical Journal of the Linnean Society , 2007, 153 , 41–48.     

Genetic variability in the European bison (Bison bonasus) population from Białowieża forest over 50 years

The aim of this study was to assess potential post-bottleneck temporal genetic differentiation following the reintroduction of the species into the Białowieża Forest. Variability of 12 polymorphic microsatellite markers was analysed for 178 individuals born between 1955 and 2005, divided by birth year into five temporal groups. Low overall allelic richness (AR) per locus (AR = 2.0) and a low overall expected heterozygosity (H_E = 0.28) were observed. The overall F _IS was not significantly different from zero. The mean F _IS values were, however, significantly different from zero for individuals born between 1955 and 1965 ( F _IS = 0.19). A Bayesian computation was used to estimate effective population size (N_e) for each temporal group. We observed relatively small values of N_e ranging from 7.9 to 28.4. The low N_e values confirm that, despite a rapid growth of the bison population following the founder event, N_e increased only slowly.  © 2009 The Linnean Society of London, Biological Journal of the Linnean Society , 2009, 97 , 801–809.     

Genetics of the peripheral populations of the alpine bullhead, Cottus poecilopus (Scorpaeniformes, Cottidae) in Poland

The alpine bullhead ( Cottus poecilopus ) in Poland is highly polymorphic ( P for examined populations in the range of 0.0–0.45, mean P  = 0.11, P for species 0.64) in seven out of 11 loci studied ( Pgd-1, Mdh-1, Ck-1, Gpd-1, Gpd-2, Gpi-1, Gpi-2 ). However, they reveal very low heterozygosity (mean H _o = 0.007, mean H _e = 0.019), which may result from population subdivisions, small genetically effective size ( N _e) of some populations and frequent inbreeding, the latter being probably mediated through polygyny. The index of genetic differentiation ( G _ST) for the Polish metapopulation equals 0.48, which is relatively high. By far, the strongest are genetic differences between the Sudeten-Carpathian group and the Hańcza population. Gene flow between the Sudeten-Carpathian group and the Hańcza population is very limited and their divergence may have started as early as the last (Eemian) interglacial period.     

Comparing direct vs. indirect estimates of gene flow within a population of a scattered tree species

The comparison between historical estimates of gene flow, using variance in allelic frequencies, and contemporary estimates of gene flow, using parentage assignment, is expected to provide insights into ecological and evolutionary processes at work within and among populations. Genetic variation at six microsatellite loci was used to quantify genetic structure in the insect-pollinated, animal-dispersed, low-density tree Sorbus torminalis L. Crantz, and to derive historical estimates of gene flow. The neighbourhood size and root-mean-squared dispersal distance inferred from seedling genotypes ( N _b  = 70 individuals, σ _e  = 417 m) were similar to those inferred from adult genotypes ( N _b  = 114 individuals, σ _e  = 472 m). We also used parentage analyses and a neighbourhood model approach after an evaluation of the statistical properties of this method on simulated data. From our data, we estimated even contributions of seed- and pollen-mediated dispersal to the genetic composition of established seedlings, with both fat-tailed pollen and seed dispersal kernels, and slightly higher mean distance of pollen dispersal (248 m) as compared to seed dispersal (135 m). The resulting contemporary estimate of gene dispersal distance (σ _c  = 211 m) was ∼twofold smaller than the historical estimates. Besides different assumptions and statistical nuances of both approaches, this discrepancy is likely to reflect a recent restriction in the scale of gene flow which requires manager's attention in a context of increasing forest fragmentation.     

Loss of genetic variation in hatchery-reared Indian major carp, Catla catla

The hypothesis that effective population sizes are low in hatchery-reared catla ( Catla catla ) from Bangladesh, possibly leading to inbreeding and loss of variation, was tested. The study was based on analysis of seven microsatellite loci in three samples of hatchery-reared catla and four samples representing wild populations. Pair-wise estimates of genetic differentiation between samples were low between wild samples (θ ranging from 0·012 to 0·034), but high between hatchery samples (θ ranging from 0·153 to 0·185), suggesting strong genetic drift in hatcheries. Genetic variation, both in terms of expected heterozygosity and allelic richness, was significantly lower in hatchery samples than in samples of wild catla. Application of a method for reconstructing families among offspring without parental genetic data showed that the hatchery samples consisted of very few half- and full-sib families, whereas the wild samples consisted of a high number of families, suggesting that most individuals were unrelated. Finally, estimation of the effective number of parents ( N _b) in the largest sample of hatchery fish confirmed that effective population size was low ( N _b= 14·9 for multiallelic loci and N _b= 10·6 if alleles were pooled into two composite alleles). The results show that low effective population sizes leading to loss of variation and possibly inbreeding depression should be a matter of serious concern in aquaculture production of catla.     

High genetic variability in Neotropical myophilous orchids

The patterns of genetic variability of seven Bulbophyllum Thouars species were investigated using 14 enzymatic loci, and compared with those of other Neotropical myophilous orchid species. The genetic variability estimated was very high ( H _e = 0.39–0.61, P  = 86–100%, A  = 2.6–3.8), despite the small population sizes, which may be explained by the vegetative reproduction. Of the three species with multiple populations, only B. epiphytum showed moderate values of genetic structuring. Environmental characteristics and seed dispersal mechanisms are important in understanding the differences in the population structure observed. Almost all pairs of species showed low genetic similarity, indicating a long period of divergence. Our results suggest that B. adiamantinum and B. insectiferum , species currently classified in different sections of the genus, are more closely related than previously thought.  © 2007 The Linnean Society of London, Botanical Journal of the Linnean Society , 2007, 153 , 33–40.     

Genetic diversity and effective size of the Atlantic salmon Salmo salar L. inhabiting the River Eo (Spain) following a stock collapse

Historic angling records suggest the occurrence of a drastic decline in the River Eo (Spain) Atlantic salmon population size during the past two decades, as a result of overexploitation and habitat deterioration. In recent years, the population has been apparently recovering, and the present study is aimed to report information on the level of genetic diversity and the effective size of the current population as these may have immediate consequences for its conservation. Eighty-six salmon from two temporal groups (1998–1999 and 2004–2006), representing three generations, were genotyped using a panel of eight microsatellites. Inspite of the recent decline in census numbers and the detection of the signs of a population bottleneck, the population exhibits a high level of genetic diversity, similar to that from other populations, and almost unchanged during the period of study [average allelic richness ( A ) = 14·0 and 13·9, and average heterozygosity ( H _e) = 0·843 and 0·851 in 1998–1999 and 2004–2006, respectively]. The effective population size ( N _e) estimated by two different temporal methods showed a consistent value around 80 salmon, whereas the estimates from the linkage disequilibrium (LD) method provided a value around 165 individuals for either sample. The recent growing number of salmon, as indicated by fisheries records, the relatively large estimates of the ratio N _e/ N (with range 0·23–0·44 for the temporal estimates and 0·31–0·59 for the LD estimates) and the high levels of diversity found suggest that the population has not been greatly affected by the historical census declines and can be expected to recover in the future.     

Alkane composition diversity among populations of dwarf forms of Juniperus communis L.: comparison between western Europe and northern American populations

The cuticular wax composition of leaves has been analysed in three western European populations (Corsica, central Pyrenees, northern Alps) of Juniperus communis var. saxatilis Pall. (=  J. nana Willd., nom illeg.) and in one population of J. communis L. var. depressa Pursh. from North America (Sierra Nevada). Gas chromatography shows the presence of 13 alkanes in all samples ranging from C₂₃ to C₃₅ with important intraspecific polymorphism in alkane content. The dominant alkanes range from C₃₃ to C₃₅. Alkanes C₂₁ and C₂₂ were found only in Corsica and Sierra Nevada populations. Canonical discriminant analysis separated the J. communis L. var. depressa Pursh. of the population of Sierra Nevada from other populations of J. communis var. saxatilis Pall. on the basis of their higher C₃₁ content and the constant presence of C₂₁ and C₂₂ alkanes. J. communis var. saxatilis Pall. populations from the Pyrenees are close to northern Alps populations characterized by high concentrations of C₃₃, C₃₄ and C₃₅ alkanes. This paper confirms the existence of Juniperus var. saxatilis Pall. in the Pyrenees (France).  © The Linnean Society of London, Botanical Journal of the Linnean Society , 2002, 140 , 165–168.     

Spatio-temporal population genetics of the Danish pine marten (Martes martes)

A spatio-temporal study of genetic variation in the Danish pine marten ( Martes martes ) populations from the Jutland peninsula and from the island of Sealand was performed using 11 microsatellite markers. Samples obtained from 1892 to 2003 were subdivided into historical (prior to 1970) and recent (from 1970) groups. As compared with the historical samples, there was a significant loss of genetic variation in the recent Jutland population, but not in Sealand. Effective population sizes were estimated using Bayesian-based software (TMVP). Historical effective population sizes were 5897 (90% highest probability density, HPD, limits: 1502–6849) in Jutland and 1300 (90% HPD limits: 224–5929) in Sealand, whereas recent effective population sizes were 14.7 (90% HPD limits: 10.9–23.5) in Jutland and 802 (90% HPD limits: 51.8–5510) in Sealand. Significant genetic differentiation ( F _ST) was found between the two historical samples, between the two recent samples, and between the historical and the recent sample in Jutland; whereas the F _ST value between the historical and the recent sample in Sealand was not significant. The significant genetic differentiation between the historical and the recent samples indicates changes in the genetic compositions over time, and the higher F _ST values between the two recent samples, as compared with the two historical samples, indicates that the populations in Sealand and Jutland have drifted apart within a short time span. No deviation from Hardy–Weinberg equilibrium was found within populations, indicating no further substructuring.  © 2008 The Linnean Society of London, Biological Journal of the Linnean Society , 2008, 93 , 457–464.     

Population divergence in the amphicarpic species Amphicarpaea edgeworthii Benth. (Fabaceae): microsatellite markers and leaf morphology

Comparative analyses of the genetic differentiation in microsatellite markers ( F _ST) and leaf morphology characters ( Q _ST) of Amphicarpaea edgeworthii Benth. were conducted to gain insight into the roles of random processes and natural selection in the population divergence. Simple sequence repeat analyses on 498 individuals of 19 natural populations demonstrate that a significant genetic differentiation occurs among populations (mean F _ST = 0.578), and A. edgeworthii is a highly self-fertilized species (mean selfing rate s  = 0.989). The distribution pattern of genetic diversity in this species shows that central populations possess high genetic diversity (e.g. population WL with H _E = 0.673 and population JG with H _E = 0.663), whereas peripheral ones have a low H _E as in population JD (0.011). The morphological divergence of leaf shape was estimated by the elliptical Fourier analysis on the data from 11 natural and four common garden populations. Leaf morphology analyses indicate the morphological divergence does not show strong correlation with the genetic differentiation ( R  = 0.260, P  = 0.069). By comparing the 95% confidence interval of Q _ST with that of F _ST, Q _ST values for five out of 12 quantitative traits are significantly higher than the average F _ST value over eight microsatellite loci. The comparison of F _ST and Q _ST suggests that two kinds of traits can be driven by different evolutionary forces, and the population divergence in leaf morphology is shaped by local selections.  © 2009 The Linnean Society of London, Biological Journal of the Linnean Society , 2009, 96 , 505–516.     

Allozyme variation and population structure of the very narrow endemic Seseli farrenyi (Apiaceae)

Analyses have been made of allozyme variation of the narrow endemic species Seseli farrenyi Molero & J. Pujadas (Apiaceae), which has only three known populations in Catalonia with a total of around 2000 individuals. All three populations were sampled and subjected to starch gel electrophoresis. Nine enzymes were resolved and 14 loci were interpreted. We detected high values of polymorphism ( P = 83.3%, A = 3.0, H _e = 0.297), far exceeding those expected for endemic species ( P = 26.3%, A = 1.39, H _e = 0.063). Genetic diversity was greater within populations than among populations, and the value of gene flow was very high ( Nm = 5.85). Most loci showed deviations from Hardy– Weinberg equilibrium, possibly due to the presence of subpopulations. The main threats to this species are human activities (tourism, fires), while natural threats are minimal due to its high genetic diversity. Finally, we propose some conservation measures which include both in situ and ex situ strategies. © 2000 The Linnean Society of London, Botanical Journal of the Linnean Society , 2002, 138 , 305–314.     

Extremely low levels of genetic diversity in the terrestrial orchid Epipactis thunbergii (Orchidaceae) in South Korea: implications for conservation

Levels of allozyme variation, population genetic structure, and fine-scale genetic structure (FSGS) of the rare, both sexually and clonally reproducing terrestrial orchid Epipactis thunbergii were examined for eight ( N  = 734) populations in a 20 × 20-km area in South Korea. Twenty-three putative allozyme loci resolved from 15 enzyme systems were used. Extremely low levels of allozyme variation were found within populations: the mean frequency of polymorphic loci was 3.8% [isocitrate dehydrogenase ( Idh-2 ) with two alleles was polymorphic across populations], the mean number of alleles per locus was 1.04, and the mean expected heterozygosity was 0.013. The overall fixation index was not significantly different from zero ( F _IS = 0.069), although the species is self-compatible. However, a significantly high degree of population differentiation was found between populations at Idh-2 ( F _ST = 0.388) in the studied area. Furthermore, spatial autocorrelation analyses revealed a significant FSGS (up to 3 m) within populations. These observations suggest that the main explanatory factors for the extremely low levels of genetic diversity and the shaping of the population genetic structure of E. thunbergii are genetic drift as a result of a small effective population size, a restricted gene flow, and the isolation of populations. Considering the current genetic structure of E. thunbergii , three guidelines are suggested for the development of conservation strategies for the species in South Korea: (1) protection of habitats of standing populations; (2) prohibition by law of any collection of E. thunbergii ; and (3) protection of nearby pollinator populations, given the fact that fruit set in natural habitats is very low.  © 2007 The Linnean Society of London, Botanical Journal of the Linnean Society , 2007, 155 , 161–169.