HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Figures Only Full Text (PDF) Free Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Moncada, K. M. Articles by DeHaan, L. R. Search for Related Content PubMed Articles by Moncada, K. M. Articles by DeHaan, L. R. Agricola Articles by Moncada, K. M. Articles by DeHaan, L. R. Related Collections Ecosystem Restoration Plant Genetic Resources

PLANT GENETIC RESOURCES

Genetic Variation in Three Native Plant Species across the State of Minnesota

^a Dep. of Agronomy and Plant Genetics, Univ. of Minnesota, 411 Borlaug Hall, 1991 Upper Buford Cir., St. Paul, MN 55108
^b The Land Institute, 2440 E. Water Well Rd., Salina, KS 67401. This research was funded by the Minnesota Department of Transportation

^* Corresponding author (monc0003{at}umn.edu).

	ABSTRACT

TOP NOTES ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Restoration of native plant communities requires adapted germplasm; thus, information is needed to guide native seed collection and production. Analysis of genetic variation has potential to provide insight into diversity and relatedness of natural populations. Our objectives were to examine genetic diversity of native species in Minnesota to discover if variation is related to biomes or distance, and possibly to develop seed collection zones. Our study included prairie cordgrass (Spartina pectinata), purple prairie clover (Dalea purpurea), and spotted joe-pye weed (Eupatorium maculatum). Using amplified fragment length polymorphisms, we analyzed 100 samples from eight populations for prairie cordgrass, 152 samples from nine populations for purple prairie clover, and 127 samples from 10 populations for spotted joe-pye weed. We found moderate to high levels of genetic diversity within each species. Small populations were not necessarily lower in diversity than larger ones. Analysis of molecular variance results indicate clear population differentiation. However, rather than displaying geographic or ecological associations, the variation had discontinuous patterns. Therefore, we were not able to develop unambiguous recommendations for seed collection. We discuss the applicability of molecular markers in detecting adaptive potential.

Abbreviations: AFLP, amplified fragment length polymorphism • AMOVA, analysis of molecular variance • ECS, Ecological Classification System • FTA, fast technology for analysis • RAPD, random amplified polymorphic DNA • RFLP, restriction fragment length polymorphism • UPGMA, unweighted arithmetic average clustering algorithm

Genetic Variation in Three Native Plant Species across the State of Minnesota

^a Dep. of Agronomy and Plant Genetics, Univ. of Minnesota, 411 Borlaug Hall, 1991 Upper Buford Cir., St. Paul, MN 55108
^b The Land Institute, 2440 E. Water Well Rd., Salina, KS 67401. This research was funded by the Minnesota Department of Transportation

^* Corresponding author (monc0003{at}umn.edu).

Restoration of native plant communities requires adapted germplasm; thus, information is needed to guide native seed collection and production. Analysis of genetic variation has potential to provide insight into diversity and relatedness of natural populations. Our objectives were to examine genetic diversity of native species in Minnesota to discover if variation is related to biomes or distance, and possibly to develop seed collection zones. Our study included prairie cordgrass (Spartina pectinata), purple prairie clover (Dalea purpurea), and spotted joe-pye weed (Eupatorium maculatum). Using amplified fragment length polymorphisms, we analyzed 100 samples from eight populations for prairie cordgrass, 152 samples from nine populations for purple prairie clover, and 127 samples from 10 populations for spotted joe-pye weed. We found moderate to high levels of genetic diversity within each species. Small populations were not necessarily lower in diversity than larger ones. Analysis of molecular variance results indicate clear population differentiation. However, rather than displaying geographic or ecological associations, the variation had discontinuous patterns. Therefore, we were not able to develop unambiguous recommendations for seed collection. We discuss the applicability of molecular markers in detecting adaptive potential.

Abbreviations: AFLP, amplified fragment length polymorphism • AMOVA, analysis of molecular variance • ECS, Ecological Classification System • FTA, fast technology for analysis • RAPD, random amplified polymorphic DNA • RFLP, restriction fragment length polymorphism • UPGMA, unweighted arithmetic average clustering algorithm

	INTRODUCTION

TOP NOTES ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
THOUSANDS OF ACRES in Minnesota are being restored to native prairie, wetlands, and forests each year, all requiring the use of native plant seed. Seed supplies collected from remnant sites are sporadic in number and expensive to buy. As a result, most seed used by state agencies is bought from native seed producers and was grown in production fields that were originally collected from natural populations. The native seed industry in Minnesota consists of producers with growing sites in half the counties of the state (Dale, 2002). In Minnesota, more than 70 native plant species are in source-identified seed production (Dale, 2002). In recent years, demand for local native seed has surpassed the supply. Currently, state agencies have differing policies on how far away the source of seed can be for their restorations; however, all follow the policy to only use locally derived seed.

Suitably adapted native plants are important to the success of restorations. Consequently, many restorationists seek guidelines for seed collection zones for the establishment of production fields. The need for sound strategies has led to the new field of restoration genetics, which combines restoration ecology and population genetics. At this time, few guidelines exist for establishing self-sustaining, adapted populations (Hufford and Mazer, 2003). Genetic diversity studies, along with common garden and reciprocal transplantation studies, may provide insights concerning the ranges of adaptation of native plant populations. Knowledge of the patterns of genetic diversity and population genetics of the natural populations is a critical component in forming future restoration and conservation plans.

There are wide-ranging ideas on how to choose seed for restoration. The ultimate goal is to have adequate diversity to ensure future adaptability while still having currently adapted genotypes. Cultivars, composites of different genotypes, and local ecotypes are among the possible sources for seed in a restoration. However, there is more support for using local sources because of the numerous examples of plants exhibiting specific local adaptation (McKay et al., 2005). Several species used in restoration have shown strong evidence of being locally adapted (Hufford and Mazer, 2003). In a native annual legume, Chamaecristra fasciculata, from Kansas, Minnesota, and Oklahoma, seed production was significantly decreased when it was grown in nonnative environments (Etterson and Shaw, 2001). Big bluestem (Andropogon gerardii) populations from Kansas and Illinois appear to demonstrate local adaptation by growing larger and faster in their own soils than in other soils (Schultz et al., 2001). There is evidence of potential adverse effects of using nonlocal ecotypes, such as outbreeding depression and genetic swamping of local genotypes. See Hufford and Mazer (2003) for a review of these risks. The results suggest that seed should be collected locally and as close as possible to the site to be restored. Unfortunately, there is little consensus on what exactly constitutes "local" or adequately close (McKay et al., 2005). In addition, Knapp and Rice (1994) propose that simply collecting within a predetermined distance does not take into consideration environmental patchiness and habitat dissimilarities.

There likely are associations in Minnesota where native plants from one source are less able to thrive at an alternative site. Minnesota is an extremely ecologically diverse state with tallgrass prairies, wetlands, deciduous forests, and coniferous forests. This heterogeneous ecology may influence the distribution of adaptive variation of our native plants. The diverse environments of Minnesota can be categorized through the Ecological Classification System (ECS), a nationwide system that combines information on climate, geology, hydrology, topography, soil, and vegetation (Minnesota Department of Natural Resources, 2007). Biomes, based on major climate zones and native vegetation, are the highest level of classification in the system. Three biomes occur in Minnesota: Laurentian Mixed Forest, Eastern Broadleaf Forest, and Prairie Parkland (Fig. 1 ). Laurentian Mixed Forest is the most highly forested area, comprised of conifer, conifer-hardwood, or hardwood vegetation. The Eastern Broadleaf Forest is the transitional area between the prairie and the true forest. The Prairie Parkland once was primarily tallgrass prairie (Minnesota Department of Natural Resources, 2007). It is unique to have three biomes within one nonmountainous state, but much of the original natural habitat is gone. Less than 1% of the prairie ecosystem, less than 4% of the original old growth forest, and 47% of wetlands in Minnesota remain (Allmann, 1997).

View larger version (24K): [in this window] [in a new window]   Figure 1. Ecological Classification System (ECS) in Minnesota: site location for prairie cordgrass, purple prairie clover, and spotted joe-pye weed, and the two dendrogram clusters by site location for each species. The shading identifies the three biomes of the ECS in Minnesota. The sites for each species are shown on each individual map. The circles correspond to populations belonging to one cluster of the dendrogram (Fig. 2, 3, and 4). The squares correspond to populations belonging to the second cluster of the dendrogram (Fig. 2, 3, and 4). CP, Clinton Prairie; GFSP; Gooseberry Falls; HP, Holthe Prairie; LBSP, Lake Bemidji State Park; LLSP, Lake Louise State Park; MaP, Malmberg Prairie; MP, Mound Prairie; ND, Norway Dunes; OMSP, Old Mill State Park; OTP, Ottertail Prairie; PBS, Paul Bunyan Savanna; PC, Prairie Coteau; POP, Pin Oak Prairie; RLPL, Red Lake Peatland; RoP, Roscoe Prairie; ScP, Schaefer Prairie; SCS, Saint Croix Savanna; SCSP, Saint Croix State Park; TRAPP, Two Rivers Aspen Prairie; VMMP, Verlyn Marth Prairie; WIP, Wild Indigo Prairie.

View larger version (6K): [in this window] [in a new window]   Figure 2. Prairie cordgrass dendrogram. Unweighted arithmetic average clustering algorithm dendrogram (cophenetic correlation = 0.8571) of prairie cordgrass populations using Nei's genetic distance matrix. CP, Clinton Prairie; HP; Holthe Prairie; MaP, Malmberg Prairie; OTP, Ottertail Prairie; POP, Pin Oak Prairie; RoP, Roscoe Prairie; TRAPP, Two Rivers Aspen Prairie; WIP, Wild Indigo Prairie.

View larger version (6K): [in this window] [in a new window]   Figure 3. Purple prairie clover dendrogram. Unweighted arithmetic average clustering algorithm dendrogram (cophenetic correlation = 0.8578) of purple prairie clover populations using Nei's genetic distance matrix. MaP, Malmberg Prairie; MP, Mound Prairie; ND, Norway Dunes; PBS, Paul Bunyan Savanna; PC, Prairie Coteau; RoP, Roscoe Prairie; ScP, Schaefer Prairie; SCS, Saint Croix Savanna; VMMP, Verlyn Marth Prairie.

View larger version (6K): [in this window] [in a new window]   Figure 4. Spotted joe-pye weed dendrogram. Unweighted arithmetic average clustering algorithm dendrogram (cophenetic correlation = 0.8996) of spotted joe-pye weed populations using Nei's genetic distance matrix. GFSP; Gooseberry Falls; HP, Holthe Prairie; LBSP, Lake Bemidji State Park; LLSP, Lake Louise State Park; OMSP, Old Mill State Park; PC, Prairie Coteau; POP, Pin Oak Prairie; RLPL, Red Lake Peatland; SCSP, Saint Croix State Park; TRAPP, Two Rivers Aspen Prairie.

Prairie cordgrass is a long-lived perennial belonging to the Poaceae family (Gleason and Cronquist, 1991). It is stout-stemmed, growing up to 3 m tall. The scaly rhizomes of prairie cordgrass form dense stands of monotypic clones that are between 1 and 10 m across (Eggers and Reed, 1997). It blooms in the midsummer to early fall (Ladd, 1995). Members of this genus produce germinating seed only sparingly and can be a tetraploid, hexaploid, or octoploid (Gleason and Cronquist, 1991). Precise information for the breeding system of prairie cordgrass is not known, but another member of this genus (S. alterniflora) had significant levels of outcrossing (Travis et al., 2002). Prairie cordgrass occurs in wet prairies, prairie marshes, on the shores of lakes and rivers throughout the tallgrass region (Ladd, 1995), and also in mesic prairies, but usually only in the moist spots (Eggers and Reed, 1997). The geographic distribution of prairie cordgrass in Minnesota ranges throughout the Eastern Broadleaf Forest and Prairie Parkland biomes (University of Minnesota Herbarium, 2002; Fig. 1). Its overall distribution is throughout the United States, except for parts of the southwest and southeast, and the southeastern part of Canada (Gleason and Cronquist, 1991; USDA, NRCS, 2005).

Purple prairie clover is a perennial warm-season legume that belongs to the Fabaceae family (Gleason and Cronquist, 1991). Plants are slender, upright, and about 0.6 m tall (Ladd, 1995). Blooming in the late spring to summer, the flowers of the inflorescence have purple petals with bright orange protruding stamens (Ladd, 1995). Purple prairie clover is a diploid (Gleason and Cronquist, 1991), primarily cross-pollinated by insects (USDA, NRCS, 2005). Its habitat is dry to mesic prairies throughout the tallgrass region (Ladd, 1995) or in woodland openings, and along railroads (USDA, NRCS, 2005). The geographic distribution of purple prairie clover within Minnesota is primarily within the Eastern Broadleaf Forest and Prairie Parkland biomes (University of Minnesota Herbarium, 2002; Fig. 1). Its overall distribution is the north and south-central United States and extends to the prairies of southern Canada (Gleason and Cronquist, 1991; USDA, NRCS, 2005).

Spotted joe-pye weed is a perennial that belongs to the Asteraceae family (Gleason and Cronquist, 1991). It grows up to 1.5 m tall with unbranched stems that usually have purple spots. The flower heads occur in corymbiform inflorescences consisting of many purplish tubular flowers (Gleason and Cronquist, 1991), blooming from late spring to early fall (Ladd, 1995). Gleason and Cronquist (1991) list spotted joe-pye weed as an outcrossing tetraploid but note that diploid forms can also occur. Spotted joe-pye weed is common in permanently moist sites and occurs in sedge meadows, calcareous fens, and shallow marshes (Ladd, 1995). The geographic distribution of spotted joe-pye weed within Minnesota ranges throughout the entire state (University of Minnesota Herbarium, 2002; Fig. 1). Its overall distribution occurs in two-thirds of the United States and extends into southern Canada (Gleason and Cronquist, 1991; USDA, NRCS, 2005).

Our original goal was to investigate the geographical range of relatedness and, presumably, the extent of adaptation of these species' populations across Minnesota to potentially develop seed collection zones for seed producers for use in restoration projects. Using amplified fragment length polymorphisms (AFLPs) analysis (Vos et al., 1995), the genetic diversities of the three species were examined. Two hypotheses were considered: (i) the patterns of variation would relate to simple geographic proximity, suggesting populations that were nearer would be more closely related than those farther apart; and (ii) diversity would vary according to the three biomes in Minnesota, implying populations within a biome would be more closely related to each other. The overall purpose of this study was (i) to determine the level of genetic diversity of the populations of these species, (ii) to determine whether the populations are genetically different from one another, and (iii) to postulate the underlying causes for the patterns of variation.

	MATERIALS AND METHODS

TOP NOTES ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Sample Collection
Collection sites were chosen by consulting historical records and local natural resource authorities to ensure the likelihood of target species being present with minimal introgression. They were distributed over the 13,921 km² of Minnesota and across the three biomes to cover the distribution of the species as much as possible (Fig. 1). The collection site habitats were variable, including prairies, wetlands, and forested areas (Table 1 ). At each site, 20 random plant samples were collected (when possible) for each species to represent a population. The population size for each species was estimated visually. Ten meters were paced between each sample collected, except for prairie cordgrass, which was collected at 20 m apart due to clone size. Tissue from each plant was preserved onsite with Whatman fast technology for analysis (FTA) (Whatman Inc., Clifton, NJ) microcards using the direct leaf press method. Gloves were worn while collecting, and alcohol was swabbed over all equipment to prevent cross-contamination between samples. Leaf material was placed on the FTA card, covered with parafilm, and pressed into the card using needle-nose pliers until plant extract was drawn through the back of the card. Samples were stored in barrier pouches with silica gel packets in coolers with ice until arrival in the laboratory, where the samples were stored at room temperature (25°C). The number of samples collected for each site for each species and the overall estimated populations' sizes are shown in Table 2 .

Amplified products were separated by capillary electrophoresis on the ABI Prism 3100 Genetic Analyzer (Applied Biosystems, Foster City, CA) in the Advanced Genetics Analysis Center at the University of Minnesota. The ABI Prism Genescan Analysis Software detected fragments from the ABI Prism 3100. The prairie cordgrass samples were run singly, while purple prairie clover and spotted joe-pye weed samples were multiplexed. The peak detection threshold was set at 100 fluorescence units, and the internal size standard used was ROX-500. The data was converted into text format with ABI Prism's Genotyper software and used to generate electropherograms. The electropherograms were visually examined to check for proper fluorescence level and to eliminate false peaks using Genoprofiler (You and Luo, 2003). The text files were then imported into Microsoft Excel (Microsoft, Redmond, WA). Peakmatcher (DeHaan et al., 2002), an Excel macro, was used to assign marker categories and to determine the presence and absence of each marker, whether polymorphic or monomorphic, generating profiles for all individuals. All markers used for the three species were in the range of 50 to 500 base pairs. The default values of Peakmatcher were used, with the exception of repeatability, which was set to 90% so that each marker that did not have a repeatability of 90% across all the replicates was removed to eliminate ambiguous fragments. To eliminate possible bias, markers with a null frequency of less than 0.03 of the number of samples were removed.

Statistical Analysis
POPGENE V 1.31 software (Yeh et al., 1997) was used to estimate Nei's unbiased measure of genetic distance, Nei's total average gene diversity (H_t), and within-population gene diversity (H_s) (Nei, 1987). The genetic distances from POPGENE were analyzed using NTSYSpc Version 2.1 (Rohlf, 2000) to generate dendrograms using the unweighted arithmetic average clustering algorithm (UPGMA). A goodness-of-fit test of the clustering of the dendrograms was generated by calculating the cophenetic correlation coefficient to the original distance matrix using NTSYSpc. Principle coordinate analysis of the three species also was performed using NTSYSpc. For analysis of molecular variance (AMOVA) analysis and statistics (Excoffier et al., 1992), Arlequin software (Schneider et al., 2000) was used on the individual AFLP profiles. The number of permutations for significance set at 1000. Two- and three-level hierarchy AMOVAs were generated: among populations and among individuals within populations, and with the clusters that were apparent in the dendrograms. Analysis of molecular variance was also used to test the original hypothesis that the patterns of genetic variation would correspond to the ECS and to determine if the diversity patterns were related to ecological factors. The longitude and latitude of the sites were used to create a geographic distance matrix, which was compared to the Nei's unbiased measure of genetic distance matrix from Popgene using NTSYS with a Mantel test. Correlations between diversity and population size (Table 2) were compared using Microsoft Excel.

	RESULTS

TOP NOTES ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Polymorphisms and Diversity
To determine the extent of genetic variation of each species and individual populations, we generated diversity estimates using the marker profiles. Prairie cordgrass, spotted joe-pye weed and purple prairie clover had 460, 317, and 279 markers, respectively, corresponding to the size of their genomes. A marker was considered polymorphic when the most common allele was less than or equal to 0.95. We found polymorphism levels of 99, 97, and 97% for prairie cordgrass, purple prairie clover and spotted joe-pye weed, respectively. Prairie cordgrass had the highest H_t value at 0.2780 ± 0.0207. Spotted joe-pye weed was at 0.2100 ± 0.0160, and purple prairie clover had the lowest H_t at 0.1928 ± 0.0161. The within-population diversity (H_s) for prairie cordgrass was 0.2016 ± 0.0097. Purple prairie clover had the lowest 0.1402 ± 0.0060, and spotted joe-pye weed had a value of 0.1716 ± 0.0086. The total gene diversity (H_t) of each individual population for each species is shown in Table 2. No correlation was found between population size and population diversity.

Population Differentiation and Patterns of Variation
We wanted to determine how different the populations were from one another and how the genetic variation was distributed across the landscape by using AMOVA and Mantel tests. The population differentiation values derived from the statistics from the "All populations" portion of the AMOVA table (Table 3 ) were 0.23, 0.27, and 0.12 (p < 0.0001) for prairie cordgrass, purple prairie clover, and spotted joe-pye weed, respectively. On the basis of these values, prairie cordgrass and purple prairie clover would be considered greatly differentiated, while spotted joe-pye weed would be moderately differentiated.

	DISCUSSION

TOP NOTES ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Our study explored some of the population characteristics of three native plant species in Minnesota. At the species level, purple prairie clover and spotted joe-pye weed had somewhat lower diversity than would be predicted, but the diversity of prairie cordgrass was within the expected range when compared with the allozyme research overview of plant species by Hamrick and Godt (1990). At the level of individual population diversity, one corroboration with other similar research was that small populations were not lacking in diversity compared with large populations for any of the species. The results most pertinent to developing seed collection zones for restorations were the amount of population differentiation and patterns of variation. All species had populations that were fairly differentiated overall from one another, with prairie cordgrass and purple prairie clover populations much more differentiated than spotted joe-pye weed, which was only moderately differentiated by the qualitative guidelines for population differentiation given by Wright (1978). The most surprising result of our research was how discontinuous the relationships between the populations of each species were. The patterns of variation did not support either of our hypotheses that genetic diversity would be related to distance or to the ECS. This leads to the question of what is behind these associations and what it means for restorations.

Species and Individual Populations Are Diverse
The populations of all the species had markers ranging from 97 to 99% polymorphic. Other native plant populations have shown similarly high polymorphism levels. Fu et al. (2004) found 100% polymorphic bands in little bluestem (Schizachryrium scoparium), while Merchanda et al. (2004) found >99% rates of polymorphism in Echinacea spp. The observed level of polymorphism may not be surprising because of the considerable area from which the natural populations were sampled. It also may be due to the outcrossing nature of these species or because of characteristics of the primers used.

Genetic diversity can indicate the genetic "health" of populations; low diversity estimates can indicate genetically depauperate populations (Barrett and Kohn, 1991). Diversity estimates can be applied to individual populations or to whole species. All the species in this study are long-lived, wide-ranging, outcrossing perennials. Hamrick and Godt (1990) in their review of diversity studies using allozymes found that species with this type of life history would have high average gene diversities in the range of 0.300. Prairie cordgrass had a high diversity with an Ht of 0.2780, similar to results from big bluestem, another native grass, where the genetic diversity found was 0.28 in Ohio using random amplified polymorphic DNA (RAPD) markers (Selbo and Snow, 2005). Purple prairie clover and spotted joe-pye weed showed more moderate values of 0.2100 and 0.1928, respectively. These lower values may be related to fragmentation and low levels of outcrossing, differences between marker systems, or other details of their biology that have yet to be characterized. Purple prairie clover collected in Illinois prairie remnants exhibited a low level of diversity, especially when compared with other members of Fabaceae (Gustafson et al., 2002). The values in this research may be due to the marker system. Amplified fragment length polymorphisms have a tendency to underestimate diversity because they do not always detect the number of copies for a given marker (Milbourne et al., 1997). The ploidy level of the species in this study varies from diploid (purple prairie clover) to tetraploid (spotted joe-pye weed) and up to octoploid for prairie cordgrass, so there is the potential that some of the diversity present is being overlooked.

Our research included many relatively small populations. Population size and diversity of the individual population were not correlated with any of the species. In simulation models, Lesica and Allendorf (1991) found that small populations may not necessarily be of lower diversity, especially those small populations that are under environmental stress. Preservation of genetic variation should enable populations to adapt to changing biotic and abiotic stresses in the environment; thus, determining the levels of diversity could be a useful predictor of the sustainability of populations in conservation and restoration projects. Research in other native species, such as big bluestem, Indian grass (Sorghastrum nutans), and purple prairie clover, has also found that small populations do not necessarily correlate with lower diversity (Gustafson et al., 1999, 2002, 2004). Because small sites seem to have adequate diversity, Gustafson et al. (2005) recommends that management techniques in seed collection do not need to compensate for prevention of inbreeding or founder effects. Our research also supports the idea that small populations are worth preserving and could serve as diverse seed sources for restorations.

Populations Are Genetically Distinct
We found that the populations are genetically different from one another. Population differentiation may be expected because these species, due to intense habitat fragmentation, are not growing as a continuous population with high gene flow. Prairie sites in Minnesota are especially fragmented (Allmann, 1997). The abundance of prairie cordgrass and purple prairie clover tended to be more sporadic on the landscape compared with spotted joe-pye weed. The sites at which prairie cordgrass and purple prairie clover were found also tended to be smaller, with an average size less than 110 ha (250 acres), while spotted joe-pye weed sites averaged over 800 ha (2000 acres). Spotted joe-pye weed populations with a (st) value of 0.12 appear to be more genetically similar relative to the other two species. This could be because spotted joe-pye weed is not limited to pristine habitats and is found more often in disturbed places, which could mean greater gene flow because there are simply more populations or that the selective pressures are similar for these populations. Prairie cordgrass has greater population divergence with a (st) value of 0.23, which could be partly due to its breeding system, clonal growth, and low seed viability. Purple prairie clover populations also seem to be more differentiated from one another with a (st) value of 0.27, which may be due to the small population sizes and narrow range of habitat in which this species was found.

Using allozymes, species with similar life histories would have G(st) values [another type of differentiation statistic analogous to (st) and F(st) values] in the range of 0.21 (Hamrick and Godt, 1990). In a study of purple prairie clover populations in Illinois and Kansas, slight to moderate population differentiation was found with values of 0.09 and 0.23 using Shannon's diversity, which partitions variance similar to AMOVA, and a F(st) value of 0.042 (Gustafson et al., 2002). Big bluestem and Indian grass in Illinois using RAPDs had F(st) values of 0.125 and 0.121, respectively (Gustafson et al., 2004). In research that covered a smaller area than our study, Travis et al. (2002) found that smooth cordgrass (Spartina alternifolia) from four natural and recolonized sites in Louisiana were differentiated by a range of F(st) values of 0.0490 to 0.1101 using AFLPs, indicating little to moderate differentiation. The exact level of differentiation, however, may not be precise in our study because dominant markers like AFLPs can overestimate population differentiation with values up to twice as high as codominant markers (Gaudeul et al., 2004).

Patterns of Variation Are Discontinuous
We found the patterns of genetic variation to be discontinuous across the state of Minnesota. Our original hypotheses that the variation may correlate to the ECS biomes or to distance were not supported. While these relationships were not observed, other ecological factors may affect genetic variation. There could be numerous other reasons for the observed clustering of populations, including random genetic drift, divergent selection forces within a biome, asymmetrical habitat fragmentation, or even introgression from past restorations. However, a certain genotype may be ascribed to a site by chance genetic drift and not by adaptation (Rice and Knapp, 2000). This research found some evidence that the genetic variation in prairie cordgrass may be related to whether a site is mesic or wet. Patterns like this, which seem to fit a certain selection model, do not necessarily mean that selection was acting because genetic drift can often generate the same patterns (Epperson, 2003). Strong patterns are only produced when selection is very intense. Prairie cordgrass is a wetland species, so the moisture regime of a site may be considered a strong selection factor.

We did not find a relationship between genetic variation and geographic proximity for any of the species examined. The genetic relationships of some native species have varied with geography, while others have not. In little bluestem, Huff et al. (1998) found that populations varied more between sites with high or low fertility than varied with geography. In the same species, Fu et al. (2004), using AFLPs, found a relationship of variation with geographic origin. The genetic relatedness of big bluestem did not correlate well with geographic distance in Arkansas (Gustafson et al., 1999). With smooth cordgrass in Louisiana, Travis and Hester (2005) found a significant correlation between genetic differentiation and geographical distance over the range of 400 km. Gustafson et al. (2002) found that the genetic relationships correlated well with geographic proximity for purple prairie clover in Illinois. Distance may not be the best gauge of genetic similarity, especially in cases of environmental heterogeneity; additional factors like elevation, soil, climate, and the species' life history may have a greater impact (Hufford and Mazer, 2003).

Molecular Markers and Adaptive Potential
We wanted to characterize the population structure of these native plant species in the hope that this information could be applied to creating seed transfer zones in Minnesota for restoration projects. However, our research focused on differences detected with neutral molecular markers, which may or may not be indicative of adaptive differences. There are increasing concerns over the use of molecular diversity estimates in the field of ecology. Latta (2004) in his analysis of genetic variation across landscapes argued that molecular markers are unlikely to be predictive of adaptive traits. In a meta-analysis of 71 data sets, Reed and Frankham (2001) found that molecular markers did not detect differentiation due to natural selection and primarily determined effects due to genetic drift. They asserted that neutral molecular markers like AFLPs are not useful to conservation biologists as a way to reveal population differentiation due to local adaptation. In their review of issues in restoration genetics, McKay et al. (2005) stated that analyzing the patterns of molecular marker variation does not address the species' adaptive potential, the aspect that concerns those involved in restoration most. One suggestion may be to use molecular markers along with additional measures of diversity (Hufford and Mazer, 2003).

Without any indications of strong geographic correlations, and with the issues of using molecular markers to detect adaptation, it is difficult to provide clear recommendations for seed collection zones from our research. Determining whether the differences in our research are due to geographic isolation or drift, versus adaptive selection would be important for determining seed collection zones. Ideally, a useful supplement to this data would be to run common garden and reciprocal transplantation experiments on the same species on the same sites. The combination of all this research would provide more definitive answers for delineating collection zones for restorations. However, these types of experiments are very labor intensive and expensive, in addition to being impractical to perform on every species used in restorations. Our research does not necessarily support or contradict the current policies of native seed provenance of Minnesota state agencies, but there is increasing empirical support from other common garden studies on native species for using only seed in the vicinity for restorations (Gustafson et al., 2005). One recommendation, without other information, is to use seed from as close as feasible and match environmental conditions from collection site to the site to be restored (McKay et al., 2005). Another similar option for determining sources of seeds in restorations is to use Plant Adaptation Regions, a classification mapping system that combines information from the ECS and from plant hardiness zone maps (Vogel et al., 2005). The best alternative until further research is performed may be for state agencies to apply the commonsense approaches like the two described above in their guidelines for native seed collection and production fields.

	ACKNOWLEDGMENTS

 
The authors would like to thank the Minnesota Department of Transportation for funding this research and the Minnesota Department of Natural Resources, the Minnesota Chapter of the Nature Conservancy, and the Minnesota State Park System for allowing access to their sites.

	NOTES

TOP NOTES ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
All rights reserved. No part of this periodical may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without permission in writing from the publisher. Permission for printing and for reprinting the material contained herein has been obtained by the publisher.

Received for publication February 14, 2007.

	REFERENCES

TOP NOTES ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 

• Allmann, L. 1997. Natural areas: Protecting a vital community asset. Minnesota Dep. of Natural Resources, St. Paul.
• Barrett, S.C.H., and J.R. Kohn. 1991. Genetic and evolutionary consequences of small population size in plants: Implications for conservation. p. 3–30. In D.A. Falk and K.E. Holsinger (ed.) Genetics and conservation of rare plants. Oxford Univ. Press, New York and Oxford.
• Dale, C. 2002. Native seed sale in Minnesota. Available at http://www.fhwa.dot.gov/environment/greenerroadsides/win02p3.htm (verified 29 Aug. 2007). Greener Roadsides (Winter).
• DeHaan, L.R., R. Antoinodes, K. Belina, and N.J. Ehlke. 2002. Peakmatcher: Software for semi-automated fluorescence-based AFLP. Crop Sci. 42:1361–1365.[Abstract/Free Full Text]
• Eggers, S.D., and D.M. Reed. 1997. Wetland plants and plant communities of Minnesota and Wisconsin. U.S. Army Corps of Engineers, St. Paul, MN.
• Epperson, B.K. 2003. Geographical genetics. Princeton Univ. Press, Princeton, NJ.
• Etterson, J.R., and R.G. Shaw. 2001. Constraint to adaptive evolution in response to global warming. Science 294:151–154.[Abstract/Free Full Text]
• Excoffier, L., P.E. Smouse, and J.M. Quattro. 1992. Analysis of molecular variance inferred from metric distances among DNA haplotypes: Application to human mitochondrial DNA restriction data. Genetics 131:479–491.[Abstract]
• Fu, Y.B., A.T. Phan, B. Coulman, and K.W. Richards. 2004. Genetic diversity in natural populations and corresponding seed collections of little bluestem as revealed by AFLP markers. Crop Sci. 44:2254–2260.[Abstract/Free Full Text]
• Gaudeul, M., I. Till-Bottraud, F. Barjon, and S. Manel. 2004. Genetic diversity and differentiation in Erygium alpinum L. (Apiaceae): Comparison of AFLP and microsatellite markers. Heredity 92:508–518.[CrossRef][Web of Science][Medline]
• Gleason, H.A., and A. Cronquist. 1991. Manual of vascular plants of northeastern United States and adjacent Canada. 2nd ed. New York Botanical Garden, Bronx.
• Groh, S., A. Zacharias, S.F. Kianian, G.A. Penner, J. Chong, H.W. Rines, and R.L. Phillips. 2001. Comparative AFLP mapping in two hexaploid oat populations. Theor. Appl. Genet. 102:876–884.[CrossRef][Web of Science]
• Gustafson, D.J., D.J. Gibson, and D.L. Nickrent. 1999. Random amplified polymorphic DNA variation among remnant big bluestem (Andropogon gerardii Vitman) populations from Arkansas' Grand Prairie. Mol. Ecol. 8:1693–1701.[CrossRef][Medline]
• Gustafson, D.J., D.J. Gibson, and D.L. Nickrent. 2002. Genetic diversity and competitive abilities of Dalea purpurea (Fabaceae) from remnant and restored grasslands. Int. J. Plant Sci. 163:979–990.[CrossRef][Web of Science]
• Gustafson, D.J., D.J. Gibson, and D.L. Nickrent. 2004. Conservation genetics of two co-dominant grass species in an endangered grassland ecosystem. J. Appl. Ecol. 41:389–397.[CrossRef]
• Gustafson, D.J., D.J. Gibson, and D.L. Nickrent. 2005. Using local seeds in prairie restoration: Data support the paradigm. Native Plants J. 6:25–28.
• Hamrick, J.L., and M.J.W. Godt. 1990. Allozyme diversity in plant species. p. 43–63. In A.H.D. Brown, M.T. Clegg, A.L. Kalher, and B.S. Weir (ed.) Plant population genetics, breeding, and genetic resources. Sinauer Associates, Sunderland, MA.
• Huff, D.R., J.A. Quinn, B. Higgins, and A.J. Palazzo. 1998. Random amplified polymorphic DNA (RAPD) variation among native little bluestem [Schizachyrium scoparium (Michx.) Nash] populations from sites of high and low fertility in forest and grassland biomes. Mol. Ecol. 7:1591–1597.[CrossRef]
• Hufford, K.M., and S.J. Mazer. 2003. Plant ecotypes: Genetic differentiation in the age of ecological restoration. Trends Ecol. Evol. 18:147–155.[CrossRef]
• Knapp, E.E., and K.J. Rice. 1994. Starting from seed: Genetic issues in using native grasses for restoration. Restoration Manage. Notes 12:40–45.
• Ladd, D.M. 1995. Tallgrass prairie wildflowers. Falcon Press, Helena, MT.
• Latta, R.G. 2004. Relating processes to patterns of genetic variation across landscapes. For. Ecol. Manage. 197:91–102.[CrossRef]
• Lesica, P., and F.W. Allendorf. 1991. Are small populations of plants worth preserving? p. 56–60. In David Ehrenfeld (ed.) Readings from Conservation Biology: Genes, populations, and species. Society for Conservation Biology and Blackwell Science, Cambridge, MA.
• McKay, J.K., C.E. Christian, S. Harrison, and K.J. Rice. 2005. "How local is local?" A review of practical and conceptual issues in the genetics of restoration. Restor. Ecol. 13:432–440.[CrossRef]
• Merchanda, S.M., B.R. Baum, D.A. Johnson, and J.T. Arnason. 2004. Sequence assessment of comigrating bands in Echinacea: Implications for comparative biological studies. Genome 47:15–25.[Medline]
• Milbourne, D., R. Meyer, J.E. Bradshaw, E. Baird, N. Bonar, J. Provan, W. Powell, and R. Waugh. 1997. Comparison of PCR-based marker systems for the analysis of genetic relationships in cultivated potato. Mol. Breed. 3:127–136.[CrossRef]
• Minnesota Department of Natural Resources. 2000. Going native: A prairie restoration handbook for Minnesota landowners. Minnesota Department of Natural Resources, Section of Ecological Services, St. Paul.
• Minnesota Department of Natural Resources. 2007. Ecological classification system website. Available at http://www.dnr.state.mn.us/ecs/index.html (verified 29 Aug. 2007). Minnesota Department of Natural Resources, St. Paul.
• Nei, M. 1987. Molecular evolutionary genetics. Columbia Univ. Press, New York.
• Reed, D.H., and R. Frankham. 2001. How closely correlated are molecular and quantitative measures of genetic variation? A meta-analysis. Evolution Int. J. Org. Evolution 55:1095–1103.[CrossRef][Web of Science][Medline]
• Rice, K.J., and E.E. Knapp. 2000. Evolutionary factors affecting the probability of local adaptation or should we expect to see ecotypes behind every rock? p. 178–194. In J.E. Keeley, M. Baer-Keeley, and C.J. Fotheringham (ed.) 2nd interface between ecology and land development in California. Open File Report 00-62. USGS, Denver, CO.
• Rohlf, F.J. 2000. NTSYSpc numerical taxonomy and multivariate analysis system version 2.1. Exeter Software, Setauket, NY.
• Schneider, S., D. Roessli, and L. Excoffier. 2000. Arlequin: A software for population genetics data analysis, version 2.0000. Available at http://cmpg.unibe.ch/software/arlequin3/ (verified 29 Aug. 2007). Genetics and Biometry Lab, Department of Anthropology, Univ. of Geneva.
• Schultz, P.A., R.M. Miller, J.D. Jastrow, C.V. Rivetta, and J.D. Bever. 2001. Evidence of a mycorrhizal mechanism for the adaptation of Andropogon gerardii (Poaceae) to high and low-nutrient prairies. Am. J. Bot. 88:1650–1656.[Abstract/Free Full Text]
• Selbo, S.M., and A.A. Snow. 2005. Flowering phenology and genetic similarity among local and recently introduced populations of Andropogon gerardii in Ohio. Restor. Ecol. 13:441–447.[CrossRef]
• Travis, S.E., and M.W. Hester. 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. J. Ecol. 93:417–430.[CrossRef]
• Travis, S.E., C.E. Proffitt, R.C. Lowenfeld, and T.W. Mitchell. 2002. A comparative assessment of genetic diversity among differently-aged populations of Spartina alterniflora on restored versus natural wetlands. Restor. Ecol. 10:37–42.[CrossRef]
• University of Minnesota Herbarium. 2002. Minnesota vascular plant database. Available at http://wildflowers.umn.edu/ (verified 29 Aug. 2007). J.F. Bell Museum of Natural History Herbarium, Minneapolis, MN.
• USDA. NRCS. 2005. The PLANTS database, version 3.5. Available http://plants.usda.gov (verified 29 Aug. 2007). USDA, Natural Resources Conservation Service, Baton Rouge, LA.
• Vogel, K.P., M.R. Schmer, and R.B. Mitchell. 2005. Plant adaptation regions: Ecological and climatic classification of plant materials. Rangeland Ecol. Manage. 58:315–319.[CrossRef]
• Vos, P., R. Hogers, M. Bleeker, M. Reijans, M. Vandalee, M. Hornes, A. Frijters, J. Pot, J. Peleman, M. Kuiper, and M. Zabeau. 1995. AFLP: A new technique for DNA fingerprinting. Nucleic Acids Res. 23:4407–4414.[Abstract/Free Full Text]
• Wright, S. 1978. Evolution and the genetics of populations. Vol. 4. Variability within and among natural populations. Univ. of Chicago Press, Chicago, IL.
• Yeh, F.C., R.-C. Yang, B.J.T. Boyle, Z.-H. Ye, and J.X. Mao. 1997. POPGENE, the user-friendly shareware for population genetic analysis. Available at http://www.ualberta.ca/fyeh/ (verified 29 Aug. 2007). Molecular Biology and Biotechnology Centre, Univ. of Alberta, Canada.
• You, F.M., and M.C. Luo. 2003. Genoprofiler version 1.06. Available at http://wheat.pw.usda.gov/PhysicalMapping/tools/genoprofiler/genoprofiler.htm (verified 29 Aug. 2007). Univ. of California, Davis.

This Article Abstract Figures Only Full Text (PDF) Free Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Moncada, K. M. Articles by DeHaan, L. R. Search for Related Content PubMed Articles by Moncada, K. M. Articles by DeHaan, L. R. Agricola Articles by Moncada, K. M. Articles by DeHaan, L. R. Related Collections Ecosystem Restoration Plant Genetic Resources