| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Population Biology |
Department of Biology, University of South Dakota, Vermillion, South Dakota 57069 USA
Received for publication November 30, 2005. Accepted for publication March 30, 2006.
ABSTRACT
Knowledge of genetic structure at different scales is necessary for evaluating the importance of interactions between the genome and environment and for inferring underlying processes that bring about evolutionary diversification. Here, genetic and morphological variation was assessed for 154 individuals of Platanthera aquilonis and P. dilatata in Maine, using RAPD and PCR-RFLP markers and measurements of five morphological traits to determine the prevalence of interspecific hybrids and underlying spatial genetic structure of the population. Both species co-mingled in the population, but clumping was observed. Platanthera aquilonis was more abundant, but genetic variation was lower (polymorphic loci [40%], expected heterozygosity [0.137]) than that of P. dilatata (Pp = 72%; He = 0.245). Interspecific hybrids were rare (eight individuals), and morphology was not consistently reliable for determining hybrid status even though the species differed significantly in several traits. Spatial autocorrelation analyses showed significant genetic structure at small distances for both species, most likely due to restricted seed dispersal. Platanthera aquilonis did not exhibit a pattern of spatial genetic structure consistent with other selfing species. This suggests that the species is facultative autogamous, which allows for limited production of hybrid individuals and reduces the strength of spatial genetic structure relative to P. dilatata.
Key Words: hybridization • Orchidaceae • PCR-RFLP • Platanthera • RAPD • spatial autocorrelation • spatial genetic structure • sympatry
Studies of genetic structure within populations provide insight into microevolutionary patterns by elucidating the movement of genes at a range of spatial scales (Heywood, 1991
; Epperson, 1993
; Vekemans and Hardy, 2004
). Such studies help to elucidate the link between genetics and ecology and offer theoretical and empirical applications to the fields of conservation biology (Escudero et al., 2003
), ecology (Berry, 1989
), and genetics (Epperson and Li, 1997
) by contributing to an evolutionary explanation of adaptation and organismal diversification. For example, mating systems may evolve in response to biparental inbreeding if related genotypes become established as neighbors due to pollen and/or seed dispersal (Lande and Schemske, 1985
; Uyenoyama, 1986
; Charlesworth and Charlesworth, 1987
). Genetic substructuring resulting from limited seed and pollen dispersal and/or genetic drift can enhance the potential for adaptive evolution (Wright, 1977
). Character displacement may evolve in response to selection against hybrid genotypes between closely related species occurring in sympatry (Brown and Wilson, 1956
; Howard, 1993
). In each of these examples, an understanding of genetic relatedness among individuals at a very fine scale is integral to evaluating the evolutionary significance of nonrandom mating, genetic drift, and hybridization.
Spatial genetic structure develops as a result of numerous genetic and environmental factors that can vary in intensity across populations (e.g., Torres et al., 2003
; Chung et al., 2005
; Tero et al., 2005
), generations (Hamrick, et al., 1993
; Tonsor et al., 1993
; Epperson and Alvarez Buylla, 1997
; Kalisz et al., 2001
; Kittelson and Maron, 2001
; Cornman et al., 2004
), years (Travis and Hester, 2005
), and microhabitats (Kittelson and Maron, 2001
). For many plant species, seed dispersal more strongly influences local patterns of gene flow and genetic structure than pollen dispersal (Hamrick and Nason, 1996
) because of the increased genetic content inherent in sexually produced seeds. Limited seed dispersal causes genetically distinct seed shadows around maternal plants. If pollen dispersal is also limited, more intense genetic structure should develop within populations over cycles of regeneration (Epperson, 1993
). Thus, the patterns of fine-scale genetic structure that develop within populations are typically consistent with an isolation-by-distance effect, whereby geographically close genotypes are more closely related to one another than a randomly chosen pair of individuals and geographically distant individuals exhibit a nearly random pattern of genotypic similarity (Hardy, 2003
; Vekemans and Hardy, 2004
).
Another factor influencing fine-scale genetic structure of many plant populations is the potential for hybridization between closely related species occurring in sympatry (Chung and Park, 2000
; Cornman et al., 2004
; Wu and Campbell, 2005
). Depending on dispersal abilities, pollinator behavior, and the strength of selection against hybrids, clinal patterns of genetic structure can become established (Rieseberg and Carney, 1998
). Additionally, differences in the direction of gene flow between hybridizing species can lead to asymmetric patterns of cytoplasmic and nuclear introgression (Asmussen et al., 1989
; Oddou-Muratorio et al., 2001
). Thus, studies of spatial genetic structure in closely related species that occur in sympatry are important for identifying the presence of a hybrid zone and for elucidating how species boundaries are maintained or eroded.
In this study, fine-scale genetic variation was examined in two North American temperate orchid species, Platanthera aquilonis Sheviak and Platanthera dilatata var. dilatata (Pursh) Lindley ex Beck, that purportedly hybridize when they come into contact. To better understand local patterns of genetic variation, including seed dispersal and the importance of interspecific hybridization, genetic and morphological characters were evaluated within a large sympatric population in northern Maine. Platanthera aquilonis and P. dilatata are members of section Limnorchis, a taxonomically complex group in which species boundaries are blurred by intraspecific morphological variation and putative interspecific hybridization (Kraenzlin, 1893
; Rydberg, 1901
; Ames, 1910
; Luer, 1975
; Schrenk, 1978
; Sheviak, 1999
). Platanthera aquilonis and P. dilatata are two of the most widely distributed species of the section. Because they frequently occur in sympatry, there is great potential for interspecific hybridization to occur. No one, though, has quantified the extent to which these species hybridize in areas of overlap. By examining morphological and genetic variation across congeneric and conspecific individuals within a mixed population, this study will aid in understanding the taxonomic distinctiveness of these species and factors important for the maintenance of genetic variation at the population level. The specific objectives of this study are (1) to evaluate the spatial distribution of individuals of the two species in the sample population, (2) to compare levels of genetic and morphological variation between the species, (3) to evaluate fine-scale patterns of genetic variation in each of the species, and (4) to determine the prevalence of interspecific hybridization.
MATERIALS AND METHODS
Study species
Platanthera aquilonis and P. dilatata are herbaceous, perennial orchid species commonly found in wet habitats such as roadside ditches, fens, wet mountain meadows, and stream banks throughout much of North America. The species exhibit considerable variation in height and the number of flowers per stem, but they are distinguishable by flower color (white in P. dilatata, green in P. aquilonis), lip shape (dilated in P. dilatata, tapered in P. aquilonis), and column structure (pointed in P. dilatata, rounded in P. aquilonis; (Sheviak, 1999
; Wallace, 2002
). The flowers of both species are small (1–2 cm wide), and inflorescences can be sparsely or densely populated with ca. 10–70 flowers. The two species differ in their mode of pollination. Platanthera dilatata has flowers with features typical of pollination by small skippers and noctuid moths, characterized by white flowers, a spicy scent, and the production of nectar in a floral spur (Boland, 1993
). By contrast, P. aquilonis is facultatively autogamous and has small scentless green flowers and loosely organized pollinia (Gray, 1862
; Catling, 1983
; Catling and Catling, 1991
; Sheviak, 1999
, 2001
).
Study site
A total of 152 plants, 109 individuals of P. aquilonis and 43 individuals of P. dilatata, were studied in a sympatric population in Crystal Bog, Aroostook County, Maine, USA, in July 2000. Crystal Bog is a large peatland complex, ca. 1470 ha in size, containing numerous fens, bogs, and wooded areas. The Crystal Bog complex is considered one of the most pristine peatlands in Maine because it contains numerous vegetation types and is floristically diverse (Davis and Anderson, 1999
). Several other orchid species were also present in Crystal Bog, including Liparis loeselii, Arethusa bulbosa, Calypso bulbosa, several Spiranthes species and Platanthera clavellata. None of these species is known to hybridize with the focal species of this study, and there was no evidence in floral or vegetative characters to suggest hybridization of P. aquilonis or P. dilatata with any of these other orchid species at the time of sampling. The species of interest were identified in the field based on published species descriptions (Luer, 1975
; Sheviak, 1999
). Both vegetative and flowering individuals were present at the time of sampling, but distinguishing between the two focal species when flowers are not present is extremely difficult. Therefore, only flowering plants were included in this study. Individuals were mapped by their compass direction and distance from a central point. These variables were translated into x–y coordinates, which were used to calculate Euclidean distances between all pairs of individuals. The mean distance between nearest neighbors (d) was determined from the matrix of Euclidean distances for each of the species. The density of individuals of each species was determined as D = n/
ri2, where n is the sample size and ri is the distance from individual (I) to its nearest neighbor (Byth and Ripley, 1980
).
Collection and analysis of morphological data
For every flowering individual, 2–3 flowers and a leaf sample were collected. Flowers were chosen near the top of the inflorescence. Because the flowers open consecutively from the bottom to the top of the inflorescence, this sampling scheme ensured collection of the freshest flowers. The flowers were stored in an ethanol/formalin preservative until measurements were made on individual floral characters within 6 months of the field visit. Floral measurements were taken after dissection under a microscope using a metric miniscale (BioQuip, Rancho Dominguez, California, USA). Additionally, voucher specimens (LEW 225–228) were collected and deposited at OS. Measurements were made on five traits (lip length, lip basal width, lip apex width, spur length, and column length) to document intraspecific morphological variation and to identify additional species-delimiting characters. Variation in each of these traits was compared between species using t tests carried out with the SPSS statistical software (SPSS, Inc., 1999
).
Collection and analysis of RAPD data
Leaf samples were kept on ice in the field and stored at –80°C until DNA was extracted. Total genomic DNA was extracted using CTAB and the method of Doyle and Doyle (Doyle and Doyle, 1987
). A sample of five individuals of each of the two species was used to screen for variation across 62 RAPD primers. Four 10-mer primers (OPA-20, OPC-02, OPX-04, OPX-06; Operon Biotechnologies, Huntsville, Alabama, USA; primer sequences are available in Appendix S1 in Supplemental Data accompanying online version of this article) having high levels of polymorphism and consistency across independent amplifications were selected for this study. In total, 25 polymorphic RAPD fragments were amplified by the four primers. Each 25-µL reaction contained 1x PCR buffer (20 mM Tris-HCl and 50 mM KCl; Invitrogen, Carlsbad, California, USA), 200 µM of each dNTP (Invitrogen), 2 mM MgCl2, 5 pmoles of primer, 0.5 units of Taq DNA polymerase (Invitrogen), and 0.4 µL template DNA. Reactions were subjected to the following thermal cycler program: 94°C for 2 min; 35 cycles of 94°C for 1 min, 36°C for 1 min, 72°C for 2 min, and a final extension of 72°C for 7 min. A negative control, including all ingredients except template DNA, was included with each set of reactions to detect contamination. Duplicate reactions and gels were run for all primers and individuals. The total product for each individual and reaction was separated on 1.5% Tris-acetate-EDTA (TAE) agarose gels, stained with ethidium bromide, and visualized with UV light. A 1 kb-plus DNA ladder (Invitrogen) was run on each gel as a size standard. Assessment of band homology across individuals was based on the similarity of molecular masses. Then, suspected homologous bands were affirmed by running individuals containing those bands side-by-side on a gel. Bands were scored as present or absent in all individuals, and bands that were not replicated in duplicate amplifications were eliminated from the data set.
Genetic variation in the RAPD data set across the two species, excluding hybrid individuals, was quantified by calculating percentage polymorphic loci (Pp) and expected heterozygosity (He) in Tools for Population Genetic Analysis (TFPGA) software v. 1.3 (Miller, 1997
). The Taylor expansion (Lynch and Milligan, 1994
) was employed in the estimation of allele frequencies used to calculate He. Patterns of genetic structure between the two species and across patches of individuals within each species were explored through an analysis of molecular variance (AMOVA) based on pairwise distances between individuals (Excoffier et al., 1992
) performed in ARLEQUIN, v. 2.0 (Schneider et al., 2000
). Significance of variance components was assessed using 10 000 permutations of the data. Finally, a multivariate principal components analysis was performed on the matrix of RAPD bands in NTSYS-pc, v. 2.0 (Rohlf, 1997
), using the distance coefficient of Nei and Li (1979)
.
Collection of chloroplast PCR-RFLP data
A restriction fragment length polymorphism in the rpL16 intron of the chloroplast genome was also used to distinguish the species. Platanthera aquilonis contains a restriction site in the rpL16 intron recognized by EcoRV, whereas P. dilatata lacks this restriction site (Wallace, 2002
). The rpL16 intron was amplified by PCR using primers F71 (Jordan et al., 1996
) and R622 (Les et al., 2002
) in all individuals. Each 25 µL reaction contained 1x PCR buffer (20 mM Tris-HCl and 50 mM KCl; Invitrogen), 200 µM of each dNTP, 3 mM MgCl2, 0.24 µM of each primer, 0.5 units of Taq DNA polymerase (Invitrogen), and 1.0 µL template DNA. The thermal cycler program included an initial denaturation step of 94°C for 5 min, followed by 35 cycles of 94°C for 1 min, 53°C for 1 min, 72°C for 2 min, and a final extension at 72°C for 5 min. Successful amplification was verified by the presence of a band of the correct size on a 1.0% TAE agarose gel. PCR products were then cleaned by precipitation with an equal volume of PEG : NaCl (20% : 2.5 M). Four microliters of the cleaned product were digested with two units of EcoRV (Invitrogen) for 24 h according to the manufacturer's instructions, followed by separation on 1.2% agarose TBE gels. A 1 kb-plus DNA ladder was run on each gel as a size standard. Gels were stained with ethidium bromide and visualized under UV light. Bands of similar mobility on a gel were assumed to be homologous, and therefore, individuals with similar bands were assumed to have the restriction site (or lack thereof) in common. Individuals were scored by the presence or absence of the restriction site because no other variation was found with EcoRV in any of the sampled individuals or in a much wider survey of the species (Wallace, 2002
, 2003
).
Identification of hybrid individuals
Individuals were considered to be of hybrid ancestry if they exhibited intermediacy in floral characters and/or molecular genetic patterns between P. aquilonis and P. dilatata. First, the RAPD data were subjected to Bayesian analyses of hybrid ancestry using the methods implemented in the programs STRUCTURE (Pritchard et al., 2000
) and NEWHYBRIDS (Anderson and Thompson, 2002
). STRUCTURE uses a model-based clustering method to assign individuals to groups in which Hardy–Weinberg equilibrium is realized. Individuals assigned to two sources with non-trivial probabilities are potential hybrids. NEWHYBRIDS uses an inheritance model defined in terms of genotype frequencies to compute, for each individual, the posterior probability of inclusion in each of six classes: pure P. aquilonis, pure P. dilatata, F1, F2, backcross with P. aquilonis, and backcross with P. dilatata. The RAPD data were coded as recommended by the authors for dominant markers. Individuals were not distinguished by species before conducting the analyses, and the analysis with each of the programs was repeated 10 times to ensure consistency of results among runs. STRUCTURE was run using K = 2 with an admixture model of ancestry, correlated allele frequencies, inferred alpha based on an initial value of 1.0, a burn-in period of 150 000 steps, and 1 million MCMC iterations following the burn-in phase. NEWHYBRIDS was run using the default parameters for the six genotype class frequencies, uniform priors (although using uninformative Jeffreys priors produced similar results), a burn-in phase of 100 000 steps, and 500 000 MCMC sweeps. Presented results are average values of the 10 replicate runs from each program. A q-value of 0.10 was used to infer mixed ancestry in STRUCTURE or inclusion in each of the hybrid classes in NEWHYBRIDS.
Flower color was also compared to the chloroplast RFLP pattern to infer hybridization events. If a green-flowered individual lacked the rpL16 restriction site, as in P. dilatata, it was considered to be of hybrid ancestry. Similarly, if a white-flowered individual contained the rpL16 restriction site, as in P. aquilonis, it was also considered to be a hybrid. Additional floral characters were examined in greater detail in hybrid individuals to assess the effectiveness of morphological characters alone for detecting hybrids between these species.
Analysis of spatial genetic structure
Spatial genetic structure at RAPD loci was detected using autocorrelation analyses. Analyses were performed independently on the two species, and hybrid individuals were not included in these analyses. The spatial autocorrelation analyses were conducted using Hardy's kinship coefficient (Fij) between individuals vs. distance in logarithmic scale using SPAGeDi, version 1.1b (Hardy and Vekemans, 2002
; Hardy, 2003
). When estimating kinship coefficients, the inbreeding coefficient (FIS) was set to 0.10 for P. dilatata and 0.70 for P. aquilonis. Although there is no direct evidence of the extent of inbreeding in P. dilatata, it is expected to be low since P. dilatata is an outcrossing species (Boland, 1993
). By contrast, because P. aquilonis is capable of autogamy (Gray, 1862
; Catling, 1983
; Catling and Catling, 1991
; Sheviak, 1999
, 2001
), the inbreeding coefficient is assumed to be higher. Additional analyses with variable levels of assumed inbreeding (FIS = 0.35, 0.50, 0.95 for P. aquilonis; FIS = 0.05, 0.50 for P. dilatata) resulted in nearly identical kinship coefficients (data not presented), supporting Hardy's (2003) suggestion that parameter estimation is robust to errors made on the assumed level of inbreeding.
The number of distance classes was set to 25 for P. aquilonis and 8 for P. dilatata such that at least 100 pairwise comparisons were considered for each distance interval as suggested by the authors (Hardy and Vekemans, 2002
). A jackknife procedure (over loci) was used to estimate standard errors for each distance class and 10 000 randomizations of spatial locations were conducted to test for overall spatial structure (Hardy and Vekemans, 2002
). Bonferroni corrections (P < 0.05/k, where k is the number of comparisons conducted) were applied in the assessment of significant P-values. Spatial genetic structure of each of the species in Crystal Bog was quantified by calculating the Sp statistic from the spatial autocorrelation analyses. The Sp statistic was determined as b/(1 – F1), where b is the regression slope of F on distance classes and F1 is the kinship coefficient estimate for adjacent individuals in the first distance interval (Hardy and Vekemans, 2002
; Hardy, 2003
; Vekemans and Hardy, 2004
). The Sp statistic accounts for differences in spatial genetic structure due to variation in sampling schemes by considering average kinship across individuals relative to the extent of the decrease in F across distance intervals. Thus, it can readily be used to compare the extent of spatial genetic structure across species and studies (Vekemans and Hardy, 2004
).
RESULTS
Distribution of individuals
Green-flowered individuals were far more abundant than white-flowered plants in Crystal Bog in the year of sampling. In fact, P. aquilonis individuals, excluding hybrids (N = 102), outnumbered P. dilatata individuals (N = 42), by nearly three times. The mean distance between neighboring P. aquilonis individuals was 1.9 m, whereas neighboring P. dilatata individuals were separated by a mean distance of 3.9 m. The density of individuals for both species in the given area was low (0.01–0.04 individuals/m2; Table 1), but individuals exhibited a clumped distribution as indicated by aggregation indices, R, (Clark and Evans, 1954
) of less than 1 (P. aquilonis R = 0.403; P. dilatata R = 0.551). Despite their clumped occurrence, the two species, however, did not occur in separate areas of the bog. Instead, in many instances, heterospecific plants were found growing alongside one another (Fig. 1). The nearest neighbor for P. aquilonis was a conspecific individual in 75% of the cases, whereas the nearest neighbor for P. dilatata was a conspecific individual in only 31% of the cases, reflecting the greater abundance of individuals of the former species and the mosaic distribution of the two species in this population.
|
|
|
|
Three individuals (1–3; Fig. 1) were considered to be hybrids when discordance between morphological traits and rpL16 patterns was used as a criterion of hybrid ancestry. However, both Bayesian analyses suggested an extremely low probability (<1%) of inclusion within their morphologically defined group for hybrid individuals 1–3. That is, there was concordance between the RAPD and rpL16 data sets in the assumed ancestry of individuals 1–3, a finding that is counter to their morphologically defined group. The Bayesian analyses of the RAPD data set suggested the presence of 3–5 additional cryptic hybrids with green flowers and the rpL16 RFLP pattern of P. aquilonis. The five green-flowered hybrids (individuals 4–8) identified by STRUCTURE showed varying patterns of inheritance in the NEWHYBRIDS analysis. For example, individual 4 had a 62% probability of belonging to the pure P. aquilonis group, and a 17% probability of inclusion in the F2 and backcross with P. aquilonis classes. Individual 5 had the highest probability, 60%, of belonging to the F2 class, indicating that it is likely a later generation hybrid or a backcrossed individual. Individual 6 was placed in the pure P. aquilonis class with a probability of 69% and in the F2 class with a 21% probability. Lastly, individuals 7–8, identified in the STRUCTURE analysis as potential hybrids, had a high posterior probability (97%) of belonging to the pure P. aquilonis group.
Hybrid ancestry was not easily deduced on the basis of the morphological traits alone. This is due to the high degree of intraspecific morphological variation exhibited by these species and the fact there is overlap in the sizes of floral features between the species (Table 2). Individuals 1 and 5 were morphologically indistinguishable from nonhybrid P. aquilonis, and individual 3 was indistinguishable from nonhybrid P. dilatata. The five other hybrids exhibited some floral features that suggest hybrid ancestry. For example, individuals 2, 4, 6, and 7 had slightly dilated lips, and the flowers of individual 8 contained a particularly long spur and large column (Table 2). There was no apparent separation of hybrids within the population because they were found growing within centimeters of nonhybrid plants of both species (Fig. 1).
Spatial genetic structure
Spatial autocorrelation analysis indicated highly significant spatial genetic structure (P < 0.001) for both species in Crystal Bog. Mean values of the regression slopes (b) were –0.069 and –0.048 for P. aquilonis and P. dilatata, respectively (Table 1). For both species, positive values of Fij were found at short distances, indicating that neighboring individuals had a higher genetic relatedness than random pairs of individuals. The kinship coefficient for the first distance interval was slightly higher for P. aquilonis (0.123 ± 0.075) than for P. dilatata (0.109 ± 0.039; Table 1). Negative values of Fij occurred at larger distances within both species. Kinship coefficients deviated significantly (P < 0.05) from the mean kinship coefficient only for the first distance interval in P. dilatata (Fig. 3). In P. aquilonis, kinship coefficients were positive at the smallest distance classes and negative at larger distance classes (Fig. 3). The Sp statistic was higher in P. aquilonis (0.079) than in P. dilatata (0.054; Table 1), indicating somewhat stronger genetic structure for the former species in Crystal Bog.
|
Spatial distribution and genetic variation of P. aquilonis and P. dilatata
The spatial distribution of plants within a population is determined by the direction and distance of dispersal and microhabitat suitability. Although individuals of P. aquilonis and P. dilatata appeared to be clumped in areas throughout Crystal Bog, the species were not segregated. Instead, individuals were often found growing within 1 m of one another. There was a notable difference in the abundance of each species, which may be related to differences in reproductive output or an artifact of the sampling method. Because P. dilatata is biotically pollinated, seed production in any given year may be small compared to P. aquilonis, which is capable of self-fertilization. Alternatively, P. dilatata may be relatively infrequent in this population as a result of a recent colonization event from few founders or decrease in seedling recruitment rates in the recent past. Although these species have persisted for a long time in Crystal Bog (K. Stockwell, The Nature Conservancy, personal communication), the relative abundance of the species as well as effective population sizes may have changed over time. Finally, the sampling scheme used in this study did not include vegetative or dormant individuals, and it is difficult to evaluate the effective population sizes of either species based on census data from a single year.
Differing levels of genetic variation between the species in Crystal Bog were not associated with differences in abundance. The greater level of genetic variation observed in P. dilatata compared to P. aquilonis in Crystal Bog is similar to previous findings in which a larger sample of populations was considered and ISSR markers were used to quantify genetic variation (Wallace, 2004
). Different mating systems exhibited by the two species may also be important determinants of the overall levels of genetic variation maintained by each of these species. Because P. dilatata is pollinated by Lepidopteran insects (Boland, 1993
) and probably incapable of self-pollination (Sheviak, 1999
), there is a high probability that most matings will occur between unrelated individuals. In contrast, the autogamous mating system of P. aquilonis permits the production of inbred offspring, which could act to reduce genetic variation in the population if there is selection against deleterious mutants (Charlesworth and Charlesworth, 1987
). Alternatively, differences in current levels of genetic variation in Crystal Bog between P. aquilonis and P. dilatata could reflect different rates of gene flow due to pollen dispersal between neighboring populations and/or differences in the gene pools of the founding individuals. Additional studies in other northeastern populations are clearly needed to fully evaluate the role that different methods of reproduction have in determining the amount of genetic variation between these two species.
Prevalence of interspecific hybridization
Convincing genetic evidence of ongoing hybridization between P. aquilonis and P. dilatata has never been presented. The morphological and molecular genetic markers used in this study indicate that P. aquilonis and P. dilatata produce mature hybrid offspring with flowers, indicating that these species are genetically compatible. The fertility of hybrid offspring remains to be empirically established, but the identification of individuals 4, 5, and 6 as F2 or backcrossed hybrids with at least a 17% posterior probability suggests that hybrids retain a high degree of fertility.
Complete incongruence between floral morphology and genetic markers in the identification of individuals 1–3 is quite curious because there is no readily apparent explanation for this finding. These individuals were not simply misidentified because the shapes and sizes of their floral organs match those of conspecific individuals with flowers of the same color. If the RAPD markers used to identify hybrids reside in the chloroplast genome, they do not represent an independent assessment of hybrid status relative to the rpL16 intron and would exhibit concordance with the cp-RFLP patterns. However, Bayesian analyses of other hybrid individuals based on the RAPD data indicated with high posterior probabilities inclusion in two groups. If the majority of RAPD markers resided in the chloroplast genome, these analyses would have placed these individuals in either the P. aquilonis or the P. dilatata group. Alternatively, individuals1–3 may represent very old instances of introgression resulting in the decoupling of RAPD markers from genes controlling floral development, or they may be floral mutants. The discordance between morphological and genetic markers is an interesting phenomenon, and the latter hypotheses warrant further testing with genetic markers of known origin in the nuclear genome.
The infrequent occurrence of hybrid individuals in Crystal Bog is, on one hand, not surprising given the differences in breeding systems, and on the other hand, very surprising given that these species frequently co-occur and have produced a stable allotetraploid species, Platanthera huronensis (Nuttall) Lindl. (Wallace, 2003
). Divergence in floral characters, including flower color, scent, and size (Table 2) support the idea that these species utilize different pollination systems. However, there is also evidence that the flowers of P. aquilonis are not obligately autogamous. First, at the time of sampling, the flowers were open and available to pollinators. Additionally, reciprocal patterns of hybridization were observed. Hybrid individuals 1 and 2 had green flowers but were mothered by P. dilatata (i.e., they contain the rpL16 profile of P. dilatata), hybrid individual 3 had white flowers but was mothered by P. aquilonis (i.e., it contained the rpL16 restriction site in common with P. aquilonis), and hybrid individuals 4–8 are likely later generation hybrids that have green flowers and the chloroplast profile of P. aquilonis. Thus, P. aquilonis has been a donor and recipient of pollen with P. dilatata. The combination of a leaky autogamous system in P. aquilonis and low abundance of pollinators for P. dilatata, which is characteristic of many orchid populations (Tremblay et al., 2005
), may reproductively isolate these species as well as allow for occasional hybridization events.
In contrast to the pattern observed in Crystal Bog, large hybrid swarms between P. aquilonis and P. dilatata, recognized as P. x media (Rydberg) Luer, have been noted around the Great Lakes (Luer, 1975
; Case, 1987
). Case (1987)
has noticed that on lakeshore bogs, habitat segregation breaks down and the species intergrade completely such that the "flip of a coin probably becomes as reliable as any structural character for separating the two species." Schrenk (1978)
has even suggested that P. x media is the most common and widely distributed orchid within section Limnorchis. However, many accounts of P. x media may actually be P. huronensis, which frequently occurs in large populations in the absence of P. aquilonis or P. dilatata and is widely distributed from New England to the Pacific Northwest to southern Utah (Sheviak, 1999
; Wallace, 2003
). Additionally, morphological variants have likely been confused with true interspecific hybrids. The low incidence of hybridization observed in Crystal Bog certainly casts doubt on the regularity of diploid hybridization between P. aquilonis and P. dilatata. Occasional hybridization events have the potential to produce new species such as P. huronensis, but on-going production of diploid hybrids does not seem to occur, at least in Crystal Bog. Nevertheless, the prevalence of interspecific hybridization needs to be examined in other populations across the species' distributions and where hybridization is suspected such that factors involved in the maintenance or breakdown of species boundaries between P. aquilonis and P. dilatata can be fully evaluated. Given the difficulty of distinguishing hybrids in Crystal Bog on the basis of morphological characters alone, it is important that these studies use comparisons between morphological variation and genetic variation to document interspecific hybridization. Additionally, further studies are needed to assess the molecular genetic and cytological differences between P. huronensis and P. x media.
Spatial genetic structure
Substantial genetic structure at fine scales has been found in most plant species studied (Vekemans and Hardy, 2004
), as a result of limited seed and/or pollen dispersal. In species that are capable of widespread dispersal of seeds via wind currents, strong genetic structure within or between populations is not expected (Ackerman and Ward, 1999
; Tremblay et al., 2005
). However, empirical and theoretical studies of orchids have demonstrated that seed dispersal and seedling establishment most often happen over distances less than 10 m (Peakall and Beattie, 1996
; Chung et al., 1998
, 2004
, 2005
; Murren and Ellison, 1998
; Trapnell et al., 2004
). All six studies of spatial genetic structure in orchids have found significant genetic clustering at less than 5 m, suggesting that most seeds fall and become established in the immediate vicinity of the maternal plant. Murren and Ellison (1998)
used wind tunnel experiments to show that seeds of an epiphytic orchid, Brassavola nodosa L. are dispersed less than 6 m from the maternal plant. The pattern of genetic structure among individuals of P. aquilonis and P. dilatata in Crystal Bog is consistent with isolation-by-distance because there was a significant and negative correlation between genetic similarity and physical distance (Fig. 3). The significant genetic clustering at short inter-plant distances in both species likely reflects restricted seed dispersal leading to limited seedling recruitment around the maternal plant (Campbell and Dooley, 1992
; Loiselle et al., 1995
; Kalisz et al., 2001
).
The Sp statistic allows patterns of spatial genetic structure across species and studies to be directly compared even though different sampling schemes may have been used. Although a synthetic quantification of spatial genetic structure, the Sp statistic facilitates the identification of trends in fine-scale patterns of genetic structure across a wide survey of plant species with varying life histories and population densities. Higher values of the Sp statistic indicate stronger genetic structure at small spatial scales. In a review of previous studies of fine-scale patterns of genetic variation in plants, Vekemans and Hardy (2004)
found significant differences in the Sp statistic corresponding to breeding system and life form but not for pollen or seed dispersal mechanisms. Selfing and herbaceous species generally have high Sp statistics. Additionally, highly dense populations tend to exhibit smaller values of the Sp statistic than low density populations. In comparison to mean values for other species, the Sp statistics for P. aquilonis (0.079) and P. dilatata (0.054) are smaller and larger, respectively, than other self-pollinating (0.1431) and outcrossing (0.0126) species (Vekemans and Hardy, 2004
). Additionally, the values for both species are higher than values reported for other species with wind-dispersed seeds (Sp = 0.0120).
Given the differences in breeding system and density exhibited by species in Crystal Bog, they were also expected to show differences in the strength of spatial genetic structure. Specifically, the effect of inbreeding should increase the Sp statistic in highly selfing species by a factor of two compared to outcrossing species (Vekemans and Hardy, 2004
). The slightly stronger degree of spatial structure observed in P. aquilonis may reflect its ability to self-pollinate, which can substantially increase the rate of genetic drift and reduce effective population size. Additionally, because gene dispersal is largely controlled by seed dispersal in selfing species, seeds falling close to the mother plant will act to increase the overall degree of relatedness among near neighbors. That a two-fold difference in the Sp statistic of P. aquilonis and P. dilatata was not found for plants in Crystal Bog suggests other forces have influenced the patterns of genetic structure. As noted before, P. aquilonis flowers are open and can effectively donate and receive pollen from pollinators, and pollen dispersal, therefore, may have a stronger role in determining genetic structure than first thought. Additional studies of pollinator visitation rates and paternity analyses would help to clarify modes of reproduction in P. aquilonis and to evaluate this as a determinant of spatial genetic structure.
Density can also be a major determinant of spatial genetic structure by affecting the strength of genetic drift. Plant species with high adult densities typically exhibit weaker fine-scale spatial genetic structure than species with lower densities (Hamrick and Nason, 1996
) due in part to an increase in pollinator flight distances in low-density populations (Schmitt, 1983
; Fenster, 1991
) of animal-pollinated species. In Crystal Bog, the four-fold difference in the density of individuals of the two species was also expected to result in a stronger difference in spatial genetic structure between the species. Differences in gene dispersal mechanisms between the two species may act to reduce the impact of density on spatial genetic structure as suggested by Vekemans and Hardy (2004)
. Additionally, gene flow from nearby populations via pollen or seeds could act to reduce spatial structure at small spatial scales in both species.
Natural selection after seed dispersal may also play an important role in reducing spatial genetic structure in orchid species as a response to environmental characteristics. Most Platanthera species are not clonal, but individual plants are often observed growing within a few centimeters of one another (L. Wallace, personal observation; Sieg and King, 1995
). Orchid seeds require a symbiotic relationship with endomycorrhizal fungi for proper germination and seedling establishment (Dressler, 1981
; Rasmussen, 1995
). Although no studies have demonstrated co-evolved and highly specific relationships between fungal species and the orchids with which they are symbiotic, some studies have reported limitations to seedling recruitment due to microsite suitability (Calvo, 1993
; Kull, 1998
; Taylor and Bruns, 1999
). Thus, if sites close to other orchids are more hospitable because they already contain mycorrhizal fungi, environmental selection may result in the establishment of orchid seedlings only in areas close to the maternal plant, thereby creating a pattern of isolation-by-distance.
Conclusions
The characterization of genetic structure at the smallest of scales provides essential information about the mechanisms of gene dispersal. For orchids, studies of population level phenomena are important for understanding the complex nature of their life cycles. Similar to other orchid species, conspecific individuals in Crystal Bog were more genetically similar at short distances. Isolation by distance in this population is most likely the result of limited seed dispersal, rather than pollen dispersal, given that P. aquilonis and P. dilatata displayed similar spatial genetic patterns despite their different breeding systems and differing levels of genetic variation.
Despite anecdotal accounts of widespread hybridization between P. aquilonis and P. dilatata, this study was the first to document genetic intermediacy of hybrid individuals. Surprisingly few hybrid individuals were found in Crystal Bog, but even infrequent hybridization events can have dramatic and long-lasting evolutionary effects by providing a bridge to allow alleles to pass between species. Thus, the importance of hybridization as a creative evolutionary mechanism warrants further study, particularly in Orchidaceae, given its size, diversity, and the high degree of genetic compatibility between species. This study also demonstrated the challenges of identifying hybrids on the basis of morphological characters alone as well as the importance of using comparisons between morphological variation and genetic variation to document the frequency and prevalence of interspecific hybridization.
FOOTNOTES
1 The author thanks M. Wallace for help in the field, K. Stockwell of The Nature Conservancy for permission to conduct this study in Crystal Bog, S. Datwyler and C. Randle for help with lab procedures and thoughtful discussions, C. Sheviak for sharing his knowledge of the study species, and two anonymous reviewers for helpful comments that improved the manuscript. This study was supported by funds from the American Orchid Society. L.E.W. was supported by a fellowship from the American Orchid Society and NSF award DEB-0445410. 
LITERATURE CITED
Ackerman J. D. Ward S.. 1999. Genetic variation in a widespread epiphytic orchid: where is the evolutionary potential?. Systematic Botany 24: 282-291.[CrossRef][ISI]
Ames O.. 1910. Orchidaceae: illustrations and studies of the family Orchidaceae, vol. IV, the genus Habenaria in North America Merrymount Press, Boston, Massachusetts, USA.
Anderson E. C. Thompson E. A.. 2002. A model-based method for identifying species hybrids using multilocus genetic data. Genetics 160: 1217-1229.
Asmussen M. A. Arnold M. L. Avise J. C.. 1989. The effects of assortative mating and migration on cytonuclear associations in hybrid zones. Genetics 122: 923-934.
Berry R. J.. 1989. Ecology: where genes and geography meet. Journal of Animal Ecology 58: 733-759.[CrossRef][ISI]
Boland J. T.. 1993. The floral biology of Platanthera dilatata (Pursh.) Lindl. (Orchidaceae). M.S. thesis, Memorial University of Newfoundland, St Johns, Newfoundland, Canada.
Brown W. L. J. Wilson E. O.. 1956. Character displacement. Systematic Zoology 5: 49-64.[Medline]
Byth K. Ripley B. D.. 1980. On sampling patterns by distance methods. Biometrics 36: 279-284.[CrossRef][ISI]
Calvo R. N.. 1993. Evolutionary demography of orchids: intensity and frequency of pollination and the cost of fruiting. Ecology 74: 1033-1042.[CrossRef][ISI]
Campbell D. R. Dooley J. L.. 1992. The spatial scale of genetic differentiation in a hummingbird-pollinated plant: comparison with models of isolation by distance. American Naturalist 139: 706-734.[CrossRef][ISI]
Case F. W. Jr.. 1987. Orchids of the western Great Lakes region Cranbrook Institute of Science, Bloomfield Hills, Michigan, USA.
Catling P. M.. 1983. Autogamy in eastern Canadian Orchidaceae: a review of current knowledge and some new observations. Le Naturaliste Canadien 110: 37-53.
Catling P. M. Catling V. R.. 1991. A synopsis of breeding systems and pollination in North American orchids. Lindleyana 6: 187-210.
Charlesworth D. Charlesworth B.. 1987. Inbreeding depression and its evolutionary consequences. Annual Review of Ecology and Systematics 18: 237-268.[CrossRef][ISI]
Chung M. Y. Chung G. M. Chung M. G. Epperson B. K.. 1998. Spatial genetic structure in populations of Cymbidium goeringii (Orchidaceae). Genes and Genetic Systems 73: 281-285.[CrossRef]
Chung M. Y. Nason J. D. Chung M. G.. 2004. Spatial genetic structure in populations of the terrestrial orchid Cephalanthera longibracteata (Orchidaceae). American Journal of Botany 91: 52-57.
Chung M. Y. Nason J. D. Chung M. G.. 2005. Spatial genetic structure in populations of the terrestrial orchid Orchis cyclochila (Orchidaceae). Plant Systematics and Evolution 254: 209-219.[CrossRef][ISI]
Chung M. G. Park C.-W.. 2000. Notes on spatial genetic structure in a hybrid population between Aconitum japonicum subsp. napiforme and A. jaluense (Ranunculaceae). Annales Botanica Fennici 37: 243-247.
Clark P. J. Evans F. C.. 1954. Distance to nearest neighbor as a measure of spatial relationships in populations. Ecology 35: 445-453.[Medline]
Cornman R. S. Burke J. M. Wesselingh R. A. Arnold M. L.. 2004. Contrasting genetic structure of adults and progeny in a Louisiana iris hybrid population. Evolution 58: 2669-2681.[ISI][Medline]
Davis R. B. Anderson D. S.. 1999. A numerical method and supporting database for evaluation of Maine peatlands as candidate natural areas Technical Bulletin 175. Maine Agricultural and Forest Experiment Station, University of Maine, Orono, Maine, USA.
Doyle J. J. Doyle J. L.. 1987. A rapid DNA isolation procedure for small amounts of fresh leaf tissue. Phytochemical Bulletin 19: 11-15.
Dressler R. L.. 1981. The orchids: natural history and classification Harvard University Press, Cambridge, Massachusetts, USA.
Epperson B. K.. 1993. Recent advances in correlation analysis of spatial patterns of genetic variation. Evolutionary Biology 27: 95-155.
Epperson B. K. Alvarez-Buylla E. R.. 1997. Limited seed dispersal and genetic structure in life stages of Cecropia obtusifolia. Evolution 51: 275-282.[CrossRef][ISI]
Epperson B. K. Li T.-Q.. 1997. Gene dispersal and spatial genetic structure. Evolution 51: 672-681.[CrossRef][ISI]
Escudero A. Iriondo J. M. Torres E.. 2003. Spatial analysis of genetic diversity as a tool for plant conservation. Biological Conservation 113: 351-365.[CrossRef][ISI]
Excoffier L. Smouse P. E. Quattro J. M.. 1992. Analysis of molecular variance inferred from metric distances among DNA haplotypes: application to human mitochondrial DNA restriction data. Genetics 131: 479-491.[Abstract]
Fenster C. B.. 1991. Gene flow in Chamaecrista fasciculata (Leguminosae). I. Gene dispersal. Evolution 45: 398-409.[CrossRef][ISI]
Gray A.. 1862. Fertilization of orchids through the agency of insects. American Journal of Science and Arts 2: 420-429.
Hamrick J. L. Murawski D. A. Nason J. D.. 1993. The influence of seed dispersal mechanisms on the genetic structure of tropical tree populations. Vegetatio 107: /108 281-297.
Hamrick J. L. Nason J. D.. 1996. Consequences of dispersal in plants. In E. Rhodes Jr., R. K. Chesser, M. H. Smith [eds.], Population dynamics in ecological space and time 203-236 University of Chicago Press, Chicago, Illinois, USA.
Hardy O. J.. 2003. Estimation of pairwise relatedness between individuals and characterization of isolation-by-distance processes using dominant genetic markers. Molecular Ecology 12: 1577-1588.[CrossRef][Medline]
Hardy O. J. Vekemans X.. 2002. SPAGeDi: a versatile computer program to analyse spatial genetic structure at the individual or population levels. Molecular Ecology Notes 2: 618-620.[CrossRef][ISI]
Heywood J. S.. 1991. Spatial analysis of genetic variation in plant populations. Annual Review of Ecology and Systematics 22: 335-355.[CrossRef][ISI]
Howard D. J.. 1993. Reinforcement: origin, dynamics, and fate of an evolutionary hypothesis. In R. G. Harrison [ed.], Hybrid zones and the evolutionary process 46-49 Oxford University Press, New York, New York, USA.
Jordan W. C. Courtney M. W. Neigel J. E.. 1996. Low levels of intraspecific genetic variation at a rapidly evolving chloroplast DNA locus in North American duckweeds (Lemnaceae). American Journal of Botany 83: 430-439.[CrossRef][ISI]
Kalisz S. Nason J. D. Hanzawa F. M. Tonsor S. J.. 2001. Spatial population genetic structure in Trillium grandiflorum: the roles of dispersal, mating, history, and selection. Evolution 55: 1560-1568.[CrossRef][ISI][Medline]
Kittelson P. M. Maron J. L.. 2001. Fine-scale genetically based differentiation of life history traits in the perennial shrub Lupinus arboreus. Evolution 55: 2429-2438.[CrossRef][ISI][Medline]
Kraenzlin F.. 1893. Beitraege zur einer monographie der gattung Habenaria. Botanische Jahrbucher fur Systematic, Pflanzengeshichte und Pflanzengeographie 16: 52-223.
Kull T.. 1998. Fruit-set and recruitment in populations of Cypripedium calceolus L. in Estonia. Botanical Journal of the Linnean Society 126: 27-38.[CrossRef]
Lande R. Schemske D. W.. 1985. The evolution of self-fertilization and inbreeding depression in plants. I. Genetic models. Evolution 39: 24-40.[CrossRef][ISI]
Les D. H. Crawford D. J. Landolt E. Gabel J. D. Kimball R. T.. 2002. Phylogeny and systematics of Lemnaceae, the duckweed family. Systematic Botany 27: 221-240.[ISI]
Loiselle B. A. Sork V. L. Nason J. D. Graham C.. 1995. Spatial genetic structure of a tropical understory shrub, Psychotria officinalis (Rubiaceae). American Journal of Botany 82: 1420-1425.[CrossRef][ISI]
Luer C. A.. 1975. The native orchids of the United States and Canada, excluding Florida New York Botanical Garden, Bronx, New York, USA.
Lynch M. Milligan B. G.. 1994. Analysis of population genetic structure with RAPD markers. Molecular Ecology 3: 91-99.[Medline]
Miller M. P.. 1997. TFPGA (Tools for Population Genetic Analysis): a Windows program for the analysis of allozyme and molecular population genetic data Utah State University, Logan, Utah, USA.
Murren C. J. Ellison A. M.. 1998. Seed dispersal characteristics of Brassavola nodosa (Orchidaceae). American Journal of Botany 85: 675-680.[Abstract]
Nei M. Li W. H.. 1979. Mathematical model for studying genetic variation in terms of restriction endonucleases. Proceedings of the National Academy of Sciences, USA 76: 5269-5273.
Oddou-Muratorio S. Petit R. J. Le Guerroue B. Guesnet D. Demesure B.. 2001. Pollen versus seed-mediated gene flow in a scattered forest tree species. Evolution 55: 1123-1135.[CrossRef][ISI][Medline]
Peakall R. Beattie A. J.. 1996. Ecological and genetic consequences of pollination by sexual deception in the orchid Caladenia tentactulata. Evolution 50: 2207-2220.[CrossRef][ISI]
Pritchard J. K. Stephens M. Donnelly P.. 2000. Inference of population structure using multilocus genotype data. Genetics 155: 945-959.
Rasmussen H. N.. 1995. Terrestrial orchids from seed to mycotrophic plant Cambridge University Press, Cambridge, UK.
Rieseberg L. H. Carney S. E.. 1998. Tansley review no. 102: plant hybridization. New Phytologist 140: 599-624.[CrossRef][ISI]
Rohlf F. J.. 1997. NTSYS-pc: numerical taxonomy and multivariate analysis system Exeter Software, Setauket, New York, USA.
Rydberg P. A.. 1901. The American species of Limnorchis and Piperia, north of Mexico. Bulletin of the Torrey Botanical Club 28: 605-643.[CrossRef]
Schmitt J.. 1983. Density-dependent pollinator foraging, flowering phenology, and temporal pollen dispersal patterns in Linanthus bicolor. Evolution 37: 1247-1257.[CrossRef][ISI]
Schneider S. Roessli D. Excoffier L.. 2000. ARLEQUIN: a software for population genetics data analysis. University of Geneva, Geneva, Switzerland Website http://anthro.unige.ch/arlequin/.
Schrenk W. J.. 1978. North American platantheras: evolution in the making. American Orchid Society Bulletin 47: 429-437.
Sheviak C. J.. 1999. The identities of Platanthera hyperborea and P. huronensis, with the description of a new species from North America. Lindleyana 14: 193-203.
Sheviak C. J.. 2001. A role for water droplets in the pollination of Platanthera aquilonis (Orchidaceae). Rhodora 103: 380-386.[ISI]
Sieg C. H. King R. M.. 1995. Influence of environmental factors and preliminary demographic analyses of a threatened orchid, Platanthera praeclara. American Midland Naturalist 134: 307-323.[CrossRef][ISI]
SPSS, Inc.. 1999. SPSS Base 10.0 for Windows User's Guide. SPSS Inc., Chicago, Illinois, USA.
Taylor D. L. Bruns T. D.. 1999. Population, habitat and genetic correlates of mycorrhizal specialization in the ‘cheating' orchids Corallorhiza maculata and C. mertensiana. Molecular Ecology 8: 1719-1732.[CrossRef][Medline]
Tero N. Aspi J. Siikamaki P. Jakalaniemi A.. 2005. Local genetic population structure in an endangered plant species, Silene tatarica (Caryophyllaceae). Heredity 94: 478-487.[CrossRef][ISI][Medline]
Tonsor S. J. Kalisz S. Fisher J.. 1993. A life-history based study of population structure: seed bank to adults in Plantago lanceolata. Evolution 47: 833-843.[CrossRef][ISI]
Torres E. Iriondo J. M. Escudero A. Perez C.. 2003. Analysis of within-population spatial genetic structure in Antirrhinum microphyllum (Scrophulariaceae). American Journal of Botany 90: 1688-1695.
Trapnell D. W. Hamrick J. L. Nason J. D.. 2004. Three-dimensional fine-scale genetic structure of the neotropical epiphytic orchid, Laelia rubescens. Molecular Ecology 13: 1111-1118.[CrossRef][Medline]
Travis S. E. Hester M. W.. 2005. A space-for-time substitution reveals the long-term decline in genotypic diversity of a widespread salt marsh plant, Spartina alterniflora, over a span of 1500 years. Journal of Ecology 93: 417-430.[CrossRef]
Tremblay R. L. Ackerman J. D. Zimmerman J. K. Calvo R. N.. 2005. Variation in sexual reproduction in orchids and its evolutionary consequences: a spasmodic journey to diversification. Biological Journal of the Linnean Society 84: 1-54.[CrossRef]
Uyenoyama M. K.. 1986. Inbreeding and the cost of meiosis: the evolution of selfing in populations practicing biparental inbreeding. Evolution 40: 388-404.[CrossRef][ISI]
Vekemans X. Hardy O. J.. 2004. New insights from fine-scale spatial genetic structure analyses in plant populations. Molecular Ecology 13: 921-935.[CrossRef][Medline]
Wallace L. E.. 2002. Biological investigations in the genus Platanthera (Orchidaceae): conservation issues in Platanthera leucophaea and evolutionary diversification in section Limnorchis Ph.D. dissertation, Ohio State University, Columbus, Ohio, USA.
Wallace L. E.. 2003. Molecular evidence for allopolyploid speciation and recurrent origins in Platanthera huronensis (Orchidaceae). International Journal of Plant Sciences 164: 907-916.[CrossRef][ISI]
Wallace L. E.. 2004. A comparison of genetic variation and structure in the allopolyploid Platanthera huronensis and its diploid progenitors, Platanthera aquilonis and Platanthera dilatata (Orchidaceae). Canadian Journal of Botany 82: 244-252.[CrossRef]
Wright S.. 1977. Evolution and the genetics of populations University of Chicago Press, Chicago, Illinois, USA.
Wu C. A. Campbell D. R.. 2005. Cytoplasmic and nuclear markers reveal contrasting patterns of spatial genetic structure in a natural Ipomopsis hybrid zone. Molecular Ecology 14: 781-792.[CrossRef][Medline]
| ||||||||||||||||
