Genetic evidence of human mediated, historical seed transfer from the Tyrolean Alps to the Romanian Carpathians in Larix decidua (Mill.) forests

Abstract

Key message

Historic transfer of larch from Alpine sources to Southern and Eastern Carpathians has been verified by means of nuclear genetic markers. Tyrolean populations can be differentiated into a north-western and south-eastern group, while Romanian populations are separated according to the Southern and Eastern Carpathians. Low-level introgression from Alpine sources is found in autochthonous Carpathian populations.

Context

Large scale human mediated transfer of forest reproductive material may have strongly modified the gene pool of European forests. Particularly in European larch, large quantities of seeds from Central Europe were used for plantations in Southern and Eastern Europe starting in the mid nineteenth century.

Aims

Our main objective was to provide DNA marker based evidence for the anthropogenic transfer of Alpine larch reproductive material to native Carpathian populations.

Methods

We studied and compared 12 populations (N = 771) of Larix decidua in the Alps (Austria, Italy) and in the Southern and Eastern Carpathians (Romania) using 13 microsatellites.

Results

High genetic diversity (He = 0.752; RS = 9.4) and a moderate genetic differentiation (FST = 0.13; GST = 0.28) among populations were found; Alpine and Carpathian populations were clearly separated by clustering methods. A Tyrolean origin of plant material was evident for one out of four adult Romanian populations. In the transferred population, a genetic influence from Carpathian sources was found neither in adults nor in juveniles, while the natural regeneration of two Romanian populations was genetically affected by Alpine sources to a minor degree (2.2 and 2.9% allochthonous individuals according to GeneClass and Structure, respectively).

Conclusion

Tracing back of plant transfer by means of genetic tools is straightforward, and we propose further studies to investigate gene flow between natural and transferred populations.

Introduction

European larch (Larix decidua Mill.) is a monoecious, anemophilous pioneer tree (Pinaceae) with a disjunct natural distribution in Europe. This deciduous conifer naturally occurs in the montane to the subalpine altitudinal belt, but can also be found in the Polish lowlands; in addition to pure stands, it occurs also in mixed stands associated with stone pine (Pinus cembra), Norway spruce (Picea abies), Swiss pine (Pinus mugo), Silver fir (Abies alba), and European beech (Fagus sylvatica) (Mayer 1992). From the forelands of the Alps and Carpathians, which represented the main refuge areas of the species during the last glacial period, it has re-colonised Central Europe (Wagner et al. 2015b). The natural distribution is split into five regions: (1) Alps, (2) Eastern Sudetes, (3) Polish lowlands, (4) Western Carpathians with scattered populations in the High and Low Tatra Mountains as well as Beskids, Fatra, and Ore Mountains, and finally (5) scattered occurrences in the Eastern and Southern Carpathians and in the Apuseni Mountains of Romania (Wagner et al. 2015a). Differences in seed weight (Simak 1967), stomata rows (Maier 1992), and female flower colour (Geburek et al. 2007) exist among populations and provide an indication about their geographical origin.

European larch is an economically important forest tree species mainly used for its timber, while its turpentine as well as larch shingles are locally also of importance (Øyen 2006). In the second half of the nineteenth century, the rapid establishment of the railway system in Central and Northern Europe enhanced the trade with plant material and triggered large-scale cultivation outside and inside of the native range (Pardé 1957; Jansen and Geburek 2016). In consequence, the current distribution area of L. decidua has more than doubled to 1.5 million ha in Europe (Pâques et al. 2013); the species has been extensively cultivated in temperate European forests especially in Germany, Poland, France, Denmark, Great Britain, Sweden, and Norway.

In this study, we focused on L. decidua in the Southern and Eastern Carpathians, where it was affected by an intense seed transfer from other areas of the native range, especially from the Tyrolean Alps. In particular, forests in Transylvania and along the Prahova Valley were affected by these transfers starting in the middle of the nineteenth century (Rubţov 1965). Cultivations of unknown origin were reported in Transylvania (Sighişoara), the Southern (Bucegi Mountains, region of Azuga) and Eastern Carpathians (Dofteana), as well as south of the Carpathian arc (Câmpina, Piteşti) (Gava 1963; Rubţov and Mocanu 1958). Although the origin of this material was not recorded, the use of Alpine material is likely. During that time, Transylvania was part of the Austro-Hungarian Empire and cultivations with Alpine seed sources were promoted (Rubţov and Mocanu 1958). Also, the Austrian seed trading companies “Jenewein” (Innsbruck) and “Steiner” (Wiener Neustadt) advertised the use of larch material especially from Tyrol and the Vienna Woods in the Southern Carpathians (Rubţov 1965). Since 1890, the trade of Alpine plant material has been documented and Austria was probably the main supplier (Gava 1963). However, which Alpine sources were used and where exactly the material was planted in Romania remains unknown. Historic records indicate, that the transferred material originated mainly not only from the Mieminger Plateau (North Tyrol, Austria) and the Etsch and Eisack valleys (South Tyrol, Italy) (Gothe 1961) but also from South (e.g. St. Johann, the Fiemme valley) and North Tyrol (e.g. Götzens, Seefeld) (Jansen and Geburek 2016).

Larch plantings with transferred Alpine provenances, especially when seeds from the Western Alps were used outside of the Alpine range, were reported to perform poorly (Weisgerber and Šindelár 1992). Through such a plant transfer the gene pool of recipient autochthonous larch populations may have been adversely affected. Outbreeding depression may be one of the effects, i.e. the reduction in progeny fitness after populations have been mixed compared to the fitness of progenies from crosses between trees that are more closely related (Frankham et al. 2017). Theoretically such effects resulting from chromosomal incompatibilities causing (partly) sterility, deleterious epistatic interactions between diverged genes, and/or disruption of co-adapted gene complexes cannot be ruled out in transferred larch populations, but these effects are not very likely. Instead, local adaptation losses are more probable due to the introduction of maladapted genes from Alpine larch sources. Theory predicts that the genetic impact on the native population depends on the degree of maladaptation of the transferred plant material and on the size of the recipient population (Kopp and Matuszewski 2014; Kremer et al. 2012). In any case, a transfer can cause genetic swamping, i.e. loss of the integrity of the local gene pool as a result of the introgression of genes of the transferred population through extensive hybridization with the local source (Hufford and Mazer 2003).

Romanian larch populations are scattered and hence are often genetically isolated. Located at the southern fringe of the native range they represent rear-edge populations and are therefore important from a gene conservation point of view (Fady et al. 2016). Therefore, we addressed the following research questions based on data from nuclear microsatellite markers: Can a potential Alpine origin be verified and—if so—can the Alpine sources be narrowed down to certain regions? Are there detectable effects on the gene pool of the Carpathian populations from Alpine seed sources in adults and/or juveniles? And if this is true, how large is the genetic impact on a local scale? Moreover, our data are of relevance for natural population differentiation based on in situ samples in the studied regions.

Materials and methods

Population sampling

Needle or cambium samples were collected from 12 populations (in situ) from the Alps and Carpathians. In the Alps material was collected in North Tyrol (Austria) and South Tyrol (Italy) (hereafter, the denotation “Tyrol” stands for both South and North Tyrol); Carpathian material was sampled in the Vâlcea, Prahova, and Braşov counties of Romania representing the native population groups, except for R3, which presumably originated from a Tyrolean source (Fig. 1, Table 1). The sample sites were selected based on documented evidence (Gothe 1961; Rubţov and Mocanu 1958) and the help of local experts. Additionally, we sampled individual trees in three locations (spot checks, SC; overall N = 10; Table 1) in the vicinity of the sampled native stands due to their Tyrolean-like habitus (Geburek et al. 2007; Rubner and Svoboda 1944). Plant tissue was collected from 50 adult dominant trees in forest stands, considering at least 30 m of inter-tree distance to avoid sampling of closely related individuals. To determine the potential recent influence of the Tyrolean variety on the Carpathian gene pool, we furthermore sampled 50 juveniles (diameter at breast height < 10 cm or height < 1.3 m) from the natural regeneration in each of these stands. Geographical coordinates were recorded with a 60CSx GPS device (Garmin International, Inc.). Approximately 3-cm2 sized cambium samples at the stem base were collected by using a hollow punch, while mainly needle samples were taken from juveniles. Collected materials were dried and stored in zip lock plastic bags containing silica gel.

Fig. 1
figure1

Sampled populations of Larix decidua within the natural distribution range (dashed line). Colours indicate geographical regions (blue—Austria (North Tyrol), green—Italy (South Tyrol), red—Romania). Map from Wagner (2013) was modified based on data from Stănescu et al. (1997)

Table 1 Geographical origin of 12 Larix decidua populations used in this study

DNA extraction and microsatellite genotyping

Total DNA was extracted from approximately 100 mg of cambium (dried in silica gel) or needle tissue, after freezing in liquid nitrogen and homogenisation using a Qiagen Tissue Lyser device (Qiagen Inc.). Genomic DNA was extracted by using the DNeasy 96 Plant Kit (Qiagen Inc.) following the manufacturer’s protocol. Quality and concentration of obtained DNA were measured using an ND-1000 spectrophotometer (NanoDrop Inc.). DNA was stored at 4 °C. All samples were genotyped with 13 highly polymorphic nuclear microsatellite loci (Wagner et al. 2012). Multiplex PCR amplification was optimised to be performed in a 10-μl reaction volume containing 2–10 ng of genomic DNA, 5-μl HotStarTaq Master Mix (Qiagen Inc.), double distilled water, and 0.3 μM of forward and reverse primers each. We used the following cycling protocol on a TC-412 Programmable Thermal Controller (Techne): 35 cycles with 94 °C for 30 s, 56 °C for 90 s, and 72 °C for 60 s. Before the first cycle, a prolonged denaturation step (95 °C for 15 min) was included and the last cycle was followed by a 30-min extension at 72 °C. Genotyping was outsourced to a commercial provider (ecogenics, Balgach, Switzerland) using Applied Biosystems (Foster City, CA, USA) chemistry and allele calling tools with manual checking of scores.

Data analyses

Genotypes with missing data for more than two loci were excluded from the analysis. Individuals from SC1, SC2, and SC3 were only used for Structure and GeneClass analyses (see below) (Table 1).

Genetic diversity within populations

Microsatellite data were analysed for allele dropout, null alleles, and slight changes in allele sizes during PCR amplification using Micro-Checker 2.2.3 (van Oosterhout et al. 2004). Furthermore, the Bayesian approach of individual inbreeding coefficient Fi implemented in INEst 2.2 (Chybicki and Burczyk 2009) was used to check raw data for null alleles, to estimate unbiased inbreeding coefficients Fi, and unbiased observed and expected heterozygosity within populations (settings: 500,000 Markov chain Monte Carlo iterations, burn-in of 50,000, thinning parameter of 50). Deviations from Hardy-Weinberg expectation (HWE) expressed as heterozygote excess or deficiency and the genotypic disequilibrium among pairs of loci were assessed using Genepop 4.7.0 implementing Monte Carlo Markov chain simulations of Fischer’s test with 10,000 dememorisations, 100 batches, and 10,000 iterations (Raymond and Rousset 1995; Rousset 2008).

Genetic diversity indices, including deviations from HWE, number of alleles, effective number of alleles (Ne), observed (Ho), expected heterozygosity (He), and number of private alleles (Ap) were calculated per locus and as means over all loci with corresponding standard errors (SE) using GenAlEx 6.502 (Peakall and Smouse 2006, 2012). The number of alleles corrected for equal sample size (allelic richness RS) with a rarefaction to 39 individuals (minimum sample size; Table 1) was calculated using Fstat 2.9.3.2 (Goudet 1995). Descriptive summarised statistics (e.g. mean, SE) were performed using R 3.3.2 (R Development Core Team 2016).

Additionally, we characterised the spatial genetic structure (SGS) within the (sub)populations using the Nason estimator of multilocus kinship coefficient Fij (Loiselle et al. 1995) implemented in SPAGeDi 1.5 (Hardy and Vekemans 2002). Ten distance intervals were defined by the program so that the number of pairwise comparisons within each distance interval was approximately constant. Individual sampling locations and gene copies were each permuted 10,000 times. SE of mean Fij values per distance class was generated by jack-knifing over loci. We assessed the significance of estimated values by comparing them with the confidence interval of the coefficients under the assumption of no SGS. Natural and non-natural (e.g. afforestation) populations should be different as in planted populations SGS assessed with selectively neutral markers should be random. Evidence of recent bottlenecks within populations was assessed using the graphical test of Luikart et al. (1998) implemented in Bottleneck 1.0.02 (Piry et al. 1999), which in most cases is not sensitive to deviations from Hardy-Weinberg proportions (Luikart et al. 1998).

Genetic diversity among populations

To determine population differentiation and the relationships among populations, we first performed a principle co-ordinate analysis (PCoA) for visualising population clusters due to relationships among inter-individual genetic distances and to identify a set of reduced dimension traits (eigenvectors) according to the number of populations using GenAlEx (Peakall and Smouse 2006, 2012). In order to take into account the unequal sample sizes, we used Nei’s unbiased genetic distance (Nei 1978) as algorithm for the PCoA. Secondly, we computed an unweighted pair group method arithmetic average dendrogram (UPGMA) based on Nei’s standard genetic distance (Nei 1972) using 1000 bootstrapped matrices created by a microsatellite analyser (Dieringer and Schlötterer 2003). For consensus tree construction, we used the programs Neighbour and Consense implemented in the Phylip 3.63 package (Felsenstein 1989).

Furthermore, we used Structure 2.3.4 (Pritchard et al. 2000) to infer the most likely number of population clusters and attempt to assign individuals to populations by reference to their genotypes. The software uses a Bayesian clustering algorithm to pool individuals into a predefined number of clusters (K) by minimising deviations from HWE and gametic-phase disequilibrium within the clusters. We used the admixture-model, where each individual does not have any information about population affiliation, with K values ranging from 2 to 10 and a run length of 800,000 iterations with a burn-in period of 200,000. Four runs per K were performed for reasons of iteration. In these computations, we included all available samples, also groups (spot checks) with a sample size < 10 (cf. Table 1). To predict the appropriate number of clusters for the whole data set, we used ΔK, an ad hoc quantity related to the rate of change of the log probability of data with respect to the number of clusters (Evanno et al. 2005), implemented in Structure Harvester 0.6.94 (Earl and von Holdt 2012). We defined individuals with more than 30% probability to belong to a different genetic cluster as introgressed. In addition, following the analysis of the whole data set, also putative autochthonous Romanian and Tyrolean adult populations were run separately with the same settings to analyse unbiased population structure on the regional level.

Information about population differentiation at various levels of population aggregations was obtained using an analysis of molecular variance (AMOVA) implemented in Arlequin 3.5 (Excoffier and Lischer 2010), with which global FST (Weir and Cockerham 1984) and RST (Slatkin 1995) were calculated separately with 10,000 permutations. Thereby, we defined the population aggregations according to the two major groups (Tyrolean vs. Romanian populations), as well as to their demographic grouping (adults vs. juveniles). To take allele size and stepwise mutations into account, we compared the observed differentiation FST with RST. If stepwise mutations have contributed to differentiation, RST is expected to be significantly higher than FST. In addition, we calculated GST (Hedrick 2005), which unlike the GST value (Nei 1972), also regards the restricted amount of genetic variation given by overlapping sets of alleles among subpopulations (Hedrick 1999); this was done using GenAlEx (Peakall and Smouse 2006, 2012) with 10,000 permutations and bootstraps, respectively. To assess the effect of geographical proximity (and perhaps population history) on genetic diversity of the two regions, we compared diversity indices (RS, Ho, He, Fis; using Fstat [Goudet 1995]) and differentiation (FST, GST; using a specific R script [R Development Core Team 2016], C. Dobeš, unpublished) among groups. Populations were grouped according to the Structure analysis, and significance was assessed based on 10,000 permutations.

Finally, we analysed the existence of first-generation migrants in the Romanian juvenile populations using GeneClass 2.0 (Piry et al. 2004). Thereby, we used the Bayesian criterion of Baudouin and Lebrun (2000) for computing the likelihood of occurrence of the individual genotype within the population where the individual has been sampled. L_home as the test statistic was used in order to prevent a potential bias in likelihood ratios caused by missing source populations (ghost populations) (Paetkau et al. 2004). For probability computation and better performance of calculation in case of ghost populations (Leblois 2011), we chose the Monte-Carlo re-sampling algorithm of Cornuet et al. (1999) with 10,000 simulated individuals and a type I error (probability of detecting a resident as immigrant) of < 0.05.

Results

Genetic diversity within populations

Average null allele frequency across all loci per population ranged between 1.9 and 8.5% (mean of 0.045; SE = 0.004; data not shown). These values fall well within the range between 5 and 8%, which according to a simulation study (Chapuis and Estoup 2007) have only minor effects on classical estimates of population differentiation. Neither large allele dropouts nor scoring of stutter peaks were identified in any of the loci. We found no evidence for genotypic disequilibrium between loci within populations (P < 0.05). However, all populations showed a significant deficit of heterozygotes, except population T9 and T12 in North Tyrol, thus in 11 populations highly significant (P < 0.001) deviations from HWE were detected (Table 2). In total, 113 alleles, ranging from six (locus Ld101_565) to a maximum of 38 (locus bcLK263_550) with an average of 18.46 alleles per locus were detected. The lowest mean number of observed alleles (Na) was calculated in population R2 (6.77), whereas the highest value was found in ST5 (11.23). Expected heterozygosity (He) values were mainly below values of observed heterozygosity (Ho) also indicated by positive fixation indices. Generally, lower diversity values were found in Romanian compared to Tyrolean samples (Table 3). We detected a total of 32 private alleles at low frequency in 13 populations with a minimum of one private allele both in R3 and T10 and a maximum of five private alleles in ST7. In most cases, SGS did not deviate from random expectation and significant Fij values were only obtained in first distance classes, allowing no estimation of gene dispersal distances. The mean pairwise kinship coefficient Fij (Loiselle et al. 1995) over all loci was significant with 0.015 in the first distance class in Romanian and with 0.010 in Tyrolean populations in each first distance class. The highest, significant Fij detected was 0.025 in the first distance class (up to 100 m) in R1 (juveniles). Moreover, R1 (adults, 81 m), R2 (juveniles, 72 m), R3 (adults, 58 m; the planted stand), R4 (juveniles, 11 m), as well as ST7 (70 m), ST8 (133 m), and T9 (67 m) showed significant positive deviations from the permuted mean values in the first distance class. Recent bottlenecks in each population were not detected.

Table 2 Genetic diversity of all Larix decidua populations and age cohorts based on 13 SSR markers studied
Table 3 Differences in genetic diversity between Tyrolean and Romanian populations

Genetic diversity among populations

The PCoA based on Nei’s unbiased genetic distance (Nei 1978) revealed clusters for Romanian and Tyrolean populations. However, a strong affiliation of the Romanian population R3 to Tyrolean populations was evident (Fig. 2). The first two axes of the PCoA explained 75.7 and 11.2% of the observed variation, respectively. Similar results of relationships among populations according to geographic proximity were provided by the UPGMA dendrogram, where each region, i.e. Romania, North and South Tyrol formed its own clade (Fig. 3).

Fig. 2
figure2

Principal co-ordinate analysis showing the multivariate relationships of 12 Larix decidua populations. The first and second axes explain 75.7 and 11.2% (86.9%) of the observed variation, respectively. Circles indicate possible clusters of related populations

Fig. 3
figure3

UPGMA consensus tree showing relationships between 12 Larix decidua populations. Numbers beside the branches indicate bootstrap support for nodes. Circles indicate possible grouping of related populations

Results of Structure consistently showed high assignment coefficients for Tyrolean and Romanian clusters, resulting from strong differentiation between these population groups (Fig. 4a). Results interpreted using the method of Evanno et al. (2005) revealed two clusters (K = 2) as the most likely group supporting the results from PCoA as well as from the UPGMA dendrogram. The Tyrolean cluster can be further differentiated into two groups, i.e. a north-western (T9, T10, T11, T12, ST8) and a south-eastern group (ST5, ST6, ST7) (Fig. 4b). Also, in the pure Romanian populations (excluding R3), a strong clustering was present, especially when these were analysed separately (Fig. 4c), corroborating the findings from PCoA and UPGMA. Population R3 and spot check SC3 (in close vicinity to R2) growing in Romania were assigned to the north-western Tyrolean cluster, while SC1 and SC2 (in close vicinity to R1) were assigned to the south-eastern cluster. According to our threshold, we found three juveniles in R1 and one individual both in the juveniles and adults, respectively, in R4 that were assigned to the Tyrolean clusters (Fig. 4a).

Fig. 4
figure4

a Bar plots of individual population assignment performed with Structure for all sampled Larix decidua individuals for two assumed ancestral clusters (K = 2). Immigrants were identified both by Structure and GeneClass are marked by a black arrow; immigrants identified by Structure alone by a blue arrow. b Bar plots of individual population assignment performed with Structure for the Tyrolean populations with K = 2. c Bar plots of individual population assignment performed with Structure for the native Romanian populations with K value of 3

The AMOVA revealed an FST value of 13.3% and an RST value of 8.01% within the adults when groups were defined according to the Structure analysis (Tyrolean vs. Romanian populations; Fig. 4). Within the juveniles, an FST value of 13.6% and an RST value of 9.22% were calculated. GST was estimated within adults as 27.7 and 32.6% for juveniles, respectively (Table 4).

Table 4 Results of analysis of molecular variance separated for adults and juveniles

For the GeneClass assignments, we checked whether first-generation migrants of Tyrolean origin could be identified in the Romanian putatively autochthonous juveniles (natural regeneration). Here, we found two individuals in R1 (juveniles) to be linked to T11 and R3, respectively, and one individual in R4 (juveniles) linked to ST7. Of particular interest is that in R3 (which is obviously from north-eastern Tyrolian origin), no first-generation migrants from autochthonous Romanian populations were detected.

Discussion

Genetic diversity within populations

Our results of nuclear microsatellite diversity are similar to other studies on L. decidua (King et al. 2013; Pluess 2011; Wagner et al. 2012) and to other conifer species with similar habitat demands in boreal or temperate forest associations such as P. abies (King et al. 2013; Tollefsrud et al. 2009), P. cembra (Dzialuk et al. 2014; Lendvay et al. 2014), or Pinus sylvestris (Pazouki et al. 2016). In general, high levels of diversity within populations in long-lived, outcrossing, and late successional coniferous taxa were obtained. However, such comparisons are always limited as samples sizes and markers differ among studies. Nevertheless, comparisons with the results of Wagner (2013) obtained by the same marker-set showed similar results of FST values, but slightly higher inbreeding coefficients than ours, which might be the effect of the presence of null alleles. Similar to Wagner (2013), in our study marker bcLK263 and bcLK211 had the highest and Ld42 had the lowest number of alleles. Comparisons with species within the genus Larix reveal higher microsatellite diversity for L. decidua (He = 0.75) than for L. lyallii and L. occidentalis (0.42 and 0.58; Khasa et al. 2006) as well as for L. gmelinii (0.41–0.60; Zhang et al. 2015), L. sibirica, and L. cajanderi (0.63 and 0.56; Oreshkova et al. 2013). He estimates in L. kaempferi (0.72–0.76; Nishimura and Setoguchi 2011) are similar to our results. In principle, differences can be due to historical bottlenecks affecting populations besides other evolutionary processes.

Almost all populations show a significant homozygote excess compared to HWE. Similar results are also reported in other studies of Larix spp. (Larionova et al. 2004; Nishimura and Setoguchi 2011; Oreshkova et al. 2013). However, when null alleles in our study were taken in account as proposed by Chybicki and Burczyk (2009), only a very small homozygote excess was observed. Unexpectedly, we did not find a general trend that in juveniles an excess of homozygotes at neutral gene loci was more pronounced than in adults, as to be expected because inbred individuals are predominantly selected against. Juvenile cohorts often differ genetically from adults, as diversity (e.g. loss of rare alleles; Kettle et al. 2007) or homozygote excess (Fujio and von Brand 1991; Stoeckel et al. 2006) might decrease during growth depending on the different selection rate against homozygotes (deleterious homozygote recessive alleles). However, it remains open whether early postzygotic selection due to embryonic lethals (recessive lethal genes that eliminate most selfed embryos during seed development; Savolainen et al. 1992) or during later ontogenetic stages other forms of viability selection occurred as in our study we did not analyse seed material but rather established young trees.

Genetic diversity was lower in the Carpathians compared to the Tyrolean populations, especially in the south-eastern cluster. Larch forests were present in the Carpathians approximately from 11,500 years ago (Magyari et al. 2012). At the same time also in Tyrol, larch forests became established as macrofossil-pollen records indicate (Heiss et al. 2005; Oeggl and Wahlmüller 1994). Lower genetic diversity of the Romanian autochthonous populations (R1, R2, R4) might be explained by the origin of these populations from small, isolated, and genetically depauperate refugia in the western part of the Southern Carpathians (Wagner et al. 2015b). On the other hand, the gene pool of Tyrolean larch might have become enriched by the merging of populations derived from several large Alpine refugia, as the data of Wagner et al. (2015b) suggest, but this would need an in-depth analysis based on of a larger set of Alpine populations. Small differences in genetic diversity in Tyrol shown by the decrease of the number of private alleles from South (12) to North Tyrol (7) could be due to genetic bottlenecks affecting diversity during migration into the Alps. We did not find any evidence of recent bottlenecks in the studied populations; however, this does not mean that there were no former bottlenecks, as our sample sizes might have been too small to detect them (Luikart et al. 1998).

Generally, SGS did not deviate significantly from random expectation. Only in very few cases, statistically significant—but very weak—clumping was observed. In the planted population R3, no deviation from random SGS was expected. However, we found a statistically significant Loiselle’s coefficient in the first distance class in R3 (adults). We interpret this significant deviation as well as those in other populations as biologically not meaningful as Loiselle’s coefficients never exceeded 0.025. Our sampling was not ideal to compare SGS among populations as distance classes could not directly be compared, as adults and juveniles were partly found in patches rather than evenly distributed across space. Disregarding these methodical limitations, it is noteworthy that larch seeds are already released in spring and may be blown over long distances on smooth snow surfaces. Also low, but significant Loiselle’s kinship coefficients (< 0.05) in distance classes up to 40 m have been already reported in L. decidua (Pluess 2011). Seed dispersal in tamarack (L. laricina) (Brown et al. 1988) is probably more limited than in L. decidua, which may explain a more pronounced non-random spatial structure in naturally regenerated tamarack stands (Knowles et al. 1992).

Genetic diversity among populations

A high level of genetic diversity and moderate differentiation among populations and groups were found in the AMOVA analysis. This pattern is in accordance with other outcrossing, wind pollinating conifers (Porth and El-Kassaby 2014). L. decidua is similarly or only slightly higher differentiated than P. abies (Androsiuk et al. 2013) or P. sylvestris (Belletti et al. 2012). Our GST of approximately 30% is similar to that found in Larix lyallii (GST = 0.26; 19 populations studied), but much higher than in L. occidentalis (GST = 0.13; 9 populations studied) (see Supplemental material in Heller and Siegismund 2009, based on data from Khasa et al. 2006). Considering the allele size, RST could not explain more of the observed variance than FST, suggesting the absence of a phylogeographic pattern. Our results also indicate (corroborated by UPGMA, PCoA and Structure analyses) that the genetic differentiation is higher in the Southern and Eastern Carpathians than in the Tyrolean Alps. While in Tyrol a continuous distribution is found, the distribution of larch in the Southern and Eastern Carpathians is restricted to small and isolated populations probably limiting gene flow and thus enhancing population differentiation.

Our analysis confirms the strong range wide structure reported by Wagner et al. (2015a) by analysing material collected in a provenance trial covering a gradient over the native distribution range on the basis of mitochondrial and nuclear markers. Our study reveals a more detailed genetic differentiation in the Tyrolean Alps and in the Carpathians. We have provided evidence that in putative autochthonous Tyrolean populations two slightly contrasting genetic clusters exist. These two genetic clusters (northern vs. southern populations) reflect the post-glacial immigration of larch (Wagner et al. 2015b). Today’s Tyrolean populations survived probably close to Verona and migrated northwards through the Etsch valley and then split into different valleys. A major split into a western (following the river Etsch) and eastern route following the river Eisack probably occurred where today the city of Bozen is located. This hypothetical migration pattern is supported by our data.

Translocation evidence

It is obvious from our results that population R3 is of north-western Tyrolean origin. The Eastern and Southern Carpathians were in particular affected by seed transfers originating from Alpine (Tyrolean) sources (Jansen and Geburek 2016). This transfer started already in the middle of the nineteenth century (Gava 1963; Rubţov 1965; Rubţov and Mocanu 1958). Therefore, we expected that the natural regeneration of Romanian putatively autochthonous stands may have been affected by the Tyrolean gene pool. This was the case to a minor extent for R1 and R4 as shown by Structure (2.9% allochthonous individuals) and GeneClass indicated 2.2% migrants from Tyrolean origin into the Southern and Eastern Carpathians. However, a general impact is impossible to be assessed from our study as any genetic introgression depends on the geographical and ecological distance as well as on the size of introduced populations. However, for R1, at least seven trees of Tyrolean origin (SC1 and SC2) were in close or relative close neighbourhood to R1 and at least three allochthonous trees (SC3) were in close proximity to R2; hence, we expected a higher gene flow from these sources. Unfortunately, no detailed information on the extent of allochthonous larch plants used in this area was available. Between 5 and 10% of intraspecific hybrids were detected in the offspring in P. abies stands in Norway, although only 4% of allochthonous trees was initially present (Dietrichson 1991). Introgression was also studied both for Spanish autochthonous Pinus pinaster and P. sylvestris populations surrounded by 15-fold and 3-fold, respectively, area of allochthonous plantations in close proximity. For both species, a male gametic gene flow of 6–8% was estimated (Unger et al. 2014). In our study, neither Structure nor GeneClass indicated an influence of autochthonous Romanian populations on the juvenile population in R3. According to the forest records, in the neighbourhood of R3, other larch stands originating from unknown (very probably Alpine) sources are located but no autochthonous populations. But if R3 was surrounded by autochthonous larch populations, a much higher introgression rate would be expected at least in the R3 juveniles unless flowering between Romanian and Tyrolian populations is badly synchronised.

Conclusions

In this study, we show that populations of European larch from the Alpine and Carpathian regions differ both in genetic diversity and in population genetic structure among and within regions. The anthropogenic transfer of larch from the Alpine region (Tyrol) to the Carpathians was clearly demonstrated. We show that translocated populations as well as introgressed individuals can be identified with relatively little effort. Our data indicate a minimal influence of this introduced material on the native gene pool. Hence, genetic swamping has been probably insignificant and, therefore, the intrinsic value of the gene pool of autochthonous, scattered larch populations as rear-edge populations in the Southern and Eastern Carpathians remains high. However, a general assessment of the intraspecific introgression in this region must remain unanswered, as our samples size was too small, providing only a snapshot of the situation in the region. To fully assess, the impact of translocations on the natural gene pool in a first step maternally inherited markers should be used to help identifying allochthonous populations and in a second step a gene flow study between native and introduced stands should be done. Ideally, this would be supplemented by an assessment of the flowering phenology and other mating system parameters of translocated Tyrolean compared to Romanian larch trees.

Data availability

The data sets generated during the current study are available in the Open Science Framework repository. https://doi.org/10.17605/OSF.IO/37RYM).

References

  1. Androsiuk P, Shimono A, Westin J, Lindgren D, Fries A, Wang XR (2013) Genetic status of Norway spruce (Picea abies) breeding populations for northern Sweden. Silvae Genet 62:127–136

    Article  Google Scholar 

  2. Baudouin L, Lebrun P (2000) An operational Bayesian approach for the identification of sexually reproduced cross-fertilized populations using molecular markers. Acta Hortic 546:81–93

    Google Scholar 

  3. Belletti P, Ferrazzini D, Piotti A, Monteleone I, Ducci F (2012) Genetic variation and divergence in Scots pine (Pinus sylvestris L.) within its natural range in Italy. Eur J For Res 131:1127–1138

    Article  Google Scholar 

  4. Brown KR, Zobel DB, Zasada JC (1988) Seed dispersal, seedling emergence, and early survival of Larix laricina (Du Roi) K. Koch in the Tanana Valley, Alaska. Can J For Res 18:306–314

    Article  Google Scholar 

  5. Chapuis M-P, Estoup A (2007) Microsatellite null alleles and estimation of population differentiation. Mol Biol Evol 24:621–631

    CAS  Article  Google Scholar 

  6. Chybicki IJ, Burczyk J (2009) Simultaneous estimation of null alleles and inbreeding coefficients. J Hered 100:106–113

    CAS  Article  Google Scholar 

  7. Cornuet JM, Piry S, Luikart G, Estoup A, Solignac M (1999) New methods employing multilocus genotypes to select or exclude populations as origins of individuals. Genetics 153:1989–2000

    CAS  PubMed  PubMed Central  Google Scholar 

  8. Development Core Team R (2016) R: a language and environment for statistical computing. R Foundation for Statistical Computing, Vienna

    Google Scholar 

  9. Dieringer D, Schlötterer C (2003) Microsatellite analyser (MSA): a platform independent analysis tool for large microsatellite data sets. Mol Ecol Notes 3:167–169

    CAS  Article  Google Scholar 

  10. Dietrichson J (1991) Genspredning fra plantet mellomeuropeisk gran (Picea abies [L.] Karst.) på Syd-Østlandet. Rapport fra Skogforsk 11:1–11

    Google Scholar 

  11. Dzialuk A, Chybicki I, Gout R, Mączka T, Fleischer P, Konrad H, Curtu AL, Sofletea N, Valadon A (2014) No reduction in genetic diversity of Swiss stone pine (Pinus cembra L.) in Tatra Mountains despite high fragmentation and small population size. Conserv Genet 15:1433–1445

    CAS  Article  Google Scholar 

  12. Earl DA, von Holdt BM (2012) STRUCTURE HARVESTER: a website and program for visualizing STRUCTURE output and implementing the Evanno method. Conserv Genet Resour 4:359–361

    Article  Google Scholar 

  13. Evanno G, Regnaut S, Goudet J (2005) Detecting the number of clusters of individuals using the software STRUCTURE: a simulation study. Mol Ecol 14:2611–2620

    CAS  Article  Google Scholar 

  14. Excoffier L, Lischer HEL (2010) Arlequin suite ver 3.5: a new series of programs to perform population genetics analyses under Linux and Windows. Mol Ecol Res 10:564–567

    Article  Google Scholar 

  15. Fady B, Aravanopoulos FA, Alizoti P, Mátyás C, von Wühlisch G, Westergren M, Belletti P, Cvjetkovic B, Ducci F, Huber G, Kelleher CT, Khaldi A, Kharrat MBD, Kraigher H, Kramer K, Mühlethaler U, Peric S, Perry A, Rousi M, Sbay H, Stojnic S, Tijardovic M, Varela MC, Vendramin GG, Zlatanov T (2016) Evolution-based approach needed for the conservation and silviculture of peripheral forest tree populations. For Ecol Manag 375:66–75

    Article  Google Scholar 

  16. Felsenstein J (1989) Phylip—phylogeny inference package vers. 32 Cladistics 5:164–166

    Google Scholar 

  17. Frankham R, Ballou JD, Ralls K, MDB E, Dudash MR, Fenster CB, Lacy RC, Sunnucks P (2017) Genetic management of fragmented animal and plant populations. Oxford University Press, United Kingdom 400 p

    Google Scholar 

  18. Fujio Y, von Brand E (1991) Differences in degree of homozygosity between seed and sown populations of the Japanese scallop Patinopecten yessoensis. Nippon Suisan Gakk 57:45–50

    Article  Google Scholar 

  19. Gava M (1963) Le meleze Roumanie. Rev For Fr 6:541–545

    Article  Google Scholar 

  20. Geburek T, Robitscheck K, Milasowski N, Schadauer C (2007) Different cone colour pay off: lessons learnt from European larch (Larix decidua Mill.) and Norway spruce (Picea abies [Karst.] L.). Can J Bot 85:132–140

    Article  Google Scholar 

  21. Gothe H (1961) Samenkundliche und pflanzenzüchterische Untersuchungen an der Schlitzer Lärche. PhD Thesis University of Munich, Munich, p 136

    Google Scholar 

  22. Goudet J (1995) Fstat vers. 1.2: a computer programme to calculate F-statistics. J Hered 86:485–486

    Article  Google Scholar 

  23. Hardy OJ, Vekemans X (2002) SPAGeDi: a versatile computer program to analyse spatial genetic structure at the individual or population levels. Mol Ecol Notes 2:618–620

    Article  Google Scholar 

  24. Hedrick PW (1999) Perspective: highly variable loci and their interpretation in evolution and conservation. Evolution 53:313–318

    Article  Google Scholar 

  25. Hedrick PW (2005) A standardized genetic differentiation measure. Evolution 59:1633–1638

    CAS  Article  Google Scholar 

  26. Heiss A, Kofler W, Oeggl K (2005) The Ulten Valley in South Tyrol, Italy: vegetation and settlement history of the area, and macrofossil record from the Iron Age cult site of St. Walburg. Palyno-Bull. Inst Bot Univ Innsbruck 1:63–73

    Google Scholar 

  27. Heller R, Siegismund HR (2009) Relationship between three measures of genetic differentiation G ST, D EST and GST: how wrong have we been? Mol Ecol 18:2080–2083

    CAS  Article  Google Scholar 

  28. Hufford KM, Mazer SJ (2003) Plant ecotypes: genetic differentiation in the age of ecological restoration. Tr Ecol Evol 18:147–155

    Article  Google Scholar 

  29. Jansen S, Geburek T (2016) Historic translocations of European larch (Larix decidua Mill.) genetic resources across Europe—a review from the 17th until the mid-20th century. For Ecol Manag 379:114–123

    Article  Google Scholar 

  30. Kettle CJ, Hollingsworth PM, Jaffre T, Moran B, Ennos RA (2007) Identifying the early genetic consequences of habitat degradation in a highly threatened tropical conifer, Araucaria nemorosa Laubenfels. Mol Ecol 16:3581–3591

    CAS  Article  Google Scholar 

  31. Khasa DP, Jaramillo-Correa JP, Jaquish B, Bousquet J (2006) Contrasting microsatellite variation between subalpine and western larch, two closely related species with different distribution patterns. Mol Ecol 15:3907–3918

    CAS  Article  Google Scholar 

  32. King GM, Gugerli F, Fonti P, Frank DC (2013) Tree growth response along an elevational gradient: climate or genetics? Oecologia 173:1587–1600

    Article  Google Scholar 

  33. Knowles P, Perry DJ, Foster HA (1992) Spatial genetic structure in two tamarack [Larix laricina (Du Roi) K. Koch] populations with differing establishment histories. Evolution 46:572–576

    Article  Google Scholar 

  34. Konrad H (2018) RawDataRaffletal_Larix. V 22 Aug. 2018. OSF. [Dataset]. https://doi.org/10.17605/OSF.IO/37RYM.

  35. Kopp M, Matuszewski S (2014) Rapid evolution of quantitative traits: theoretical perspectives. Evol Appl 7:169–191

    Article  Google Scholar 

  36. Kremer A, Ronce O, Robledo-Arnuncio JJ, Guillaume F, Bohrer G, Nathan R, Bridle JR, Gomulkiewicz R, Klein EK, Ritland K, Kuparinen A, Gerber S, Schueler S (2012) Long-distance gene flow and adaptation of forest trees to rapid climate change. Ecol Lett 15:378–392

    Article  Google Scholar 

  37. Larionova AY, Yakhneva NV, Abaimov AP (2004) Genetic diversity and differentiation of Gmelin larch Larix gmelinii populations from Evenkia (Central Siberia). Russ J Genet 40:1127–1134

    CAS  Article  Google Scholar 

  38. Leblois R (2011) Assignment and Clustering algorithms for individual multilocus genotypes. Centre de Biologie et de Gestion des Populations. Erasmus Mundus Master Programme in Evolutionary Biology, University Montpellier II: Genetic Data Analysis, 2012-2013. Montpellier. France. Available from http://raphael.leblois.free.fr/ressources/cours/MEME_30-03-2011_Clustering_LEBLOIS_small.pdf. Accessed 15 Dec 2017

  39. Lendvay B, Höhn M, Brodbeck S, Mîndrescu M, Gugerli F (2014) Genetic structure in Pinus cembra from the Carpathian Mountains inferred from nuclear and chloroplast microsatellites confirms post-glacial range contraction and identifies introduced individuals. Tree Genet Genom 10:1419–1433

    Article  Google Scholar 

  40. Loiselle BA, Sork VL, Nason J, Graham C (1995) Spatial genetic structure of a tropical understory shrub, Psychotria officinalis (Rubiaceae). Am J Bot 82:1420–1425

    Article  Google Scholar 

  41. Luikart G, Allendorf FW, Cornuet J-M, Sherwin WB (1998) Distortion of allele frequency distributions provides a test for recent population bottlenecks. J Hered 89:238–247

    CAS  Article  Google Scholar 

  42. Magyari EK, Jakab G, Bálint M, Kern Z, Buczkó K, Braun M (2012) Rapid vegetation response to Lateglacial and early Holocene climatic fluctuation in the South Carpathian Mountains (Romania). Quat Sci Rev 35:116–130

    Article  Google Scholar 

  43. Maier J (1992) Herkunftsunterschiede in der Länge der Stomatareihen bei Larix decidua Mill. Flora 186:169–176

    Article  Google Scholar 

  44. Mayer H (1992) Waldbau auf soziologisch-ökologischer Grundlage. Gustav Fischer, Stuttgart

    Google Scholar 

  45. Nei M (1972) Genetic distance between populations. Am Nat 106:283–292

    Article  Google Scholar 

  46. Nei M (1978) Estimation of average heterozygosity and genetic distance from a small number of individuals. Genetics 89:583–590

    CAS  PubMed  PubMed Central  Google Scholar 

  47. Nishimura M, Setoguchi H (2011) Homogeneous genetic structure and variation in tree architecture of Larix kaempferi along altitudinal gradients on Mt. Fuji. J Plant Res 124:253–263

    Article  Google Scholar 

  48. Oeggl K, Wahlmüller N (1994) Holozäne Vegetationsentwicklung an der Waldgrenze der Ostalpen: die Plancklacke (2140m)/Sankt Jakob im Defreggen, Osttirol. Diss Bot 234:389–411

    Google Scholar 

  49. Oreshkova NV, Belokon MM, Jamiyansuren S (2013) Genetic diversity, population structure, and differentiation of Siberian Larch, Gmelin Larch, and Cajander Larch on SSR marker data. Russ J Genet 49:178–186

    CAS  Article  Google Scholar 

  50. Øyen B (2006) Lerk (Larix) i Norge – del. Dyrkningshistorien Aktuelt fra skogforskningen 2:1–16

    Google Scholar 

  51. Paetkau D, Slade R, Burden M, Estoup A (2004) Genetic assignment methods for the direct, real-time estimation of migration rate: a simulation based exploration of accuracy and power. Mol Ecol 13:55–65

    CAS  Article  Google Scholar 

  52. Pâques LE, Foffová E, Heinze B, Lelu-Walter M-A, Liesebach M, Philippe G (2013) Larches (Larix sp.). In: Pâques LE (ed) Forest tree breeding in Europe; current state-of-the-art and perspectives, managing forest ecosystems. Springer Science & Business Media, p 13–122

  53. Pardé J (1957) Plaidoyer pour Mélèze. Rev For Franç 9:634–650

    Article  Google Scholar 

  54. Pazouki L, Shanjani PS, Fields PD, Martins K, Suhhorutšenko M, Viinalass H, Niinemets Ü (2016) Large within-population genetic diversity of the widespread conifer Pinus sylvestris at its soil fertility limit characterized by nuclear and chloroplast microsatellite markers. Eur J For Res 135:161–177

    CAS  Article  Google Scholar 

  55. Peakall ROD, Smouse PE (2006) GENEALEX 6: genetic analysis in Excel. Population genetic software for teaching and research. Mol Ecol Notes 6:288–295

    Article  Google Scholar 

  56. Peakall ROD, Smouse PE (2012) GENEALEX 6.5: genetic analysis in Excel. Population genetic software for teaching and research—an update. Bioinformatics 28:2537–2539

    CAS  Article  Google Scholar 

  57. Piry S, Luikart G, Cornuet JM (1999) BOTTLENECK: a computer program for detecting recent reductions in the effective population size using allele frequency data. J Hered 90:502–503

    Article  Google Scholar 

  58. Piry S, Alapetite A, Cornuet JM, Paetkau D, Baudouin L, Estoup A (2004) GeneClass2: a software for genetic assignment and first-generation migrant detection. J Hered 95:536–539

    CAS  Article  Google Scholar 

  59. Pluess AR (2011) Pursuing glacier retreat: genetic structure of a rapidly expanding Larix decidua population. Mol Ecol 20:473–485

    Article  Google Scholar 

  60. Porth I, El-Kassaby YA (2014) Assessment of the genetic diversity in forest tree populations using molecular markers. Diversity 6:283–295

    Article  Google Scholar 

  61. Pritchard JK, Stephens P, Donnelly P (2000) Inference of population structure using multilocus genotype data. Genetics 155:945–959

    CAS  PubMed  PubMed Central  Google Scholar 

  62. Raymond M, Rousset F (1995) GENEPOP (version 1.2): population genetics software for exact tests and ecumenicism. J Hered 86:248–249

    Article  Google Scholar 

  63. Rousset F (2008) Genepop'007: a complete reimplementation of the Genepop software for Windows and Linux. Mol Ecol Res 8:103–106

    Article  Google Scholar 

  64. Rubner K, Svoboda P (1944) Untersuchungen an Lärchenzapfen verschiedener Herkunft. Intersylva 4:121–146

    Google Scholar 

  65. Rubţov S (1965) Arealui si ecologia Laricelui - Centrele de raspindire naturala a laricelui in Romania. In: Rubţov S (ed) Laricele - Ecologia si Cultura. Editura Agro-Silvica, p 65–74

  66. Rubţov S, Mocanu V (1958) Raspindirea laricelui spontan si cultivat, in R.P.R. St cerc biol 1:5–53

    Google Scholar 

  67. Savolainen O, Kärkkäinen K, Kuittinen H (1992) Estimating numbers of embryonic lethals in conifers. Heredity 69:308–314

    Article  Google Scholar 

  68. Simak M (1967) Seed weight of larch from different provenances (Larix decidua Mill.). Stud For Suec 57:1–31

    Google Scholar 

  69. Slatkin M (1995) A measure of population subdivision based on microsatellite allele frequencies. Genetics 139:1463–1463

    PubMed Central  Google Scholar 

  70. Stănescu V, Șofletea N, Popescu O (1997) Flora forestieră lemnoasă a României. Editura Ceres, București

    Google Scholar 

  71. Stoeckel S, Grange J, Fernandez-Manjarres J, Biger I, Frascaria-Lacoste N, Mariette S (2006) Heterozygote excess in a self-incompatible and partially clonal forest tree species—Prunus avium L. Mol Ecol 15:2109–2118

    CAS  Article  Google Scholar 

  72. Tollefsrud MM, Sønstebø JH, Brochmann C, Johnsen Ø, Skrøppa T, Vendramin GG (2009) Combined analysis of nuclear and mitochondrial markers provide new insight into the genetic structure of North European Picea abies. Heredity 102:549–562

    CAS  Article  Google Scholar 

  73. Unger GM, Vendramin GG, Robledo-Arnuncio JJ (2014) Estimating exotic gene flow into native pine stands: zygotic vs. gametic components. Mol Ecol 23:5435–5447

    CAS  Article  Google Scholar 

  74. van Oosterhout C, Hutchinson WF, Wills DPM, Shipley P (2004) MICROCHECKER: software for identifying and correcting genotyping errors in microsatellite data. Mol Ecol Notes 4:535–538

    Article  Google Scholar 

  75. Wagner S (2013) History of the European larch (Larix decidua Mill.). Doctoral thesis. Rheinische Friedrich-Wilhelms-Universität Bonn and the Université Bordeaux I, Bonn and Bordeaux, 163 pp

  76. Wagner S, Gerber S, Petit RJ (2012) Two highly informative dinucleotide SSR multiplexes for the conifer Larix decidua (European larch). Mol Ecol Res 12:717–725

    CAS  Article  Google Scholar 

  77. Wagner S, Liepelt S, Gerber S, Petit RJ (2015a) Within-range translocations and their consequences in European larch. PLoS One 10:e0127516. https://doi.org/10.1371/journal.pone.0127516

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  78. Wagner S, Litt T, Sànchez-Goni MF, Petit R (2015b) History of Larix decidua Mill. (European larch) since 130 ka. Quat Sci Rev 124:224–247

    Article  Google Scholar 

  79. Weir BS, Cockerham CC (1984) Estimating F statistics for the analysis of population structure. Evolution 38:1358–1370

    CAS  PubMed  Google Scholar 

  80. Weisgerber H, Šindelár J (1992) IUFRO’s role in coniferous tree improvement. History, results, and future trends of research and international cooperation with European larch (Larix decidua Mill.). Silvae Genet 41:150–161

  81. Zhang G, Sun Z, Zhou D, Xiong M, Wang X, Yang J, Wei Z (2015) Development and characterization of novel EST-SSRs from Larix gmelinii and their cross-species transferability. Molecules 20:12469–12480

    CAS  Article  Google Scholar 

Download references

Acknowledgements

We thank the forestry authority of the Autonomous Province of Bolzano and North Tyrol for providing information about the spatial position of larch stands and Manuel Fauner for his support in the field. Thomas Thalmayr helped to create figures. Christoph Dobeš provided an R script to calculate FST and GST among population groups. We also gratefully acknowledge the input of the handling editor Bruno Fady and three anonymous reviewers.

Author information

Affiliations

Authors

Corresponding author

Correspondence to Thomas Geburek.

Ethics declarations

Conflict of interest

The authors declare that they have no conflict of interest.

Additional information

Contribution of co-authors: Hannes Raffl did the sampling, lab work, computations, and statistics and leads the writing of the manuscript. Heino Konrad supported lab work, assisted in data analysis, took part in the discussion of the results, and provided input to the manuscript. Lucian Curtu was the local support in sampling in Romania, contributed to discussion and manuscript writing. Thomas Geburek conceived the study, supported the interpretation of the results, and provided input to the manuscript.

Handling editor: Bruno Fady

Electronic supplementary material

ESM 1

(DOCX 130 kb)

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Raffl, H., Konrad, H., Curtu, L.A. et al. Genetic evidence of human mediated, historical seed transfer from the Tyrolean Alps to the Romanian Carpathians in Larix decidua (Mill.) forests. Annals of Forest Science 75, 98 (2018). https://doi.org/10.1007/s13595-018-0776-9

Download citation

Keywords

  • Genetic pollution
  • Genetic swamping
  • Intraspecific introgression
  • Microsatellites
  • Spatial genetic structure