2011 SOUTHEASTERN NATURALIST 10(4):751–760 No Evidence of Local Adaptation in Uniola paniculata L. (Poaceae), a Coastal Dune Grass Cara L. Gormally1,* and Lisa A. Donovan1 Abstract - Studies of local adaptation generally investigate plants growing in relatively stable habitats. We asked whether populations of the long-lived clonal grass Uniola paniculata (Sea Oats) are locally adapted to microhabitats in the southeastern US coastal dunes, a habitat characterized by dynamic environmental gradients spanning relatively small distances. Although vegetative zonation is well characterized across these gradients, little is known about intraspecific evolutionary responses of species spanning the gradients. Plants from the foredune and backdune areas of the gradient (less than 10 m and 40–60 m from the shoreline, respectively) were reciprocally transplanted into experimental plots in both habitats. Although foredune plots were washed away by storms before harvest, the foredune plants demonstrated no early advantage in stem diameter or height growth, and thus there was no support for local adaptation in foredune plants. In the backdune plots, the backdune plants demonstrated no early growth advantage, and additionally demonstrated no advantage in survival, nor in growth or total biomass of surviving plants at harvest. Thus, there was again no support for local adaptation. In frequently disturbed environments such as the coastal dunes, plants may be more likely to respond with phenotypic plasticity than through local adaptation. Introduction Natural selection can drive local adaptation when plant species encounter heterogeneous environmental conditions (Clausen et al. 1940, Hereford 2009, Turesson 1922), both at the landscape level across the span of their ranges, as well as at finer, microhabitat scales (Antonovics and Bradshaw 1970). Local adaptation is the association of certain genotypes with particular habitats, with the locally adapted genotypes outperforming “foreign” genotypes (Kawecki and Ebert 2004, Linhart and Grant 1996). Most field-based reciprocal transplant experiments of local adaptation target plant species growing in relatively stable environments characterized by differences in abiotic factors such as heavy metals, fertility, herbicides, elevation, light, temperature, and soil-water availability (Leimu and Fischer 2008, Linhart and Grant 1996). However, much less is known about local adaptation of plant species located in frequently disturbed habitats, particularly for long-lived clonal plants (Galloway and Fenster 2000, Gregor 1930, Leimu and Fischer 2008, Miller and Fowler 1994). Coastal sand dunes are spatially and temporally heterogeneous, in part due to frequent disturbances that can result in the removal of low-lying foredunes, partial destruction of larger dunes, the erosion of dune profiles and even complete habitat destruction (e.g., submerging of an entire barrier island) (Claudino-Sales et al. 1Department of Plant Biology, University of Georgia, Athens, GA 30602. 2Current address - School of Biology, Georgia Institute of Technology, 310 Ferst Drive, Atlanta, GA 30322. *Corresponding author - cara.gormally@biology.gatech.edu. 752 Southeastern Naturalist Vol. 10, No. 4 2007). In addition to physical disturbance, environmental conditions—including salt spray, sand accretion, and soil nutrients—may differ dramatically across distances as small as 0.5 m, driving changes in plant community composition and species’ distributions (Barbour et al. 1999, Boyce 1957, Cowles 1899, Doing 1985, Maun and Perumal 1999, Oosting and Billings 1942, Smith and Smith 2001, Wagner 1964). Plants in these habitats demonstrate intraspecific variation in morphological, physiological, and phenological traits (Gormally and Donovan 2010, Jerling 1985, Wagner 1964). Among the grasses studied, as the distance from the shoreline increases, plant size generally decreases, and proportions of flowering, seeding, and establishment decline (Gormally and Donovan 2010, Jerling 1985, Wagner 1964). While extensive ecological research has been conducted in order to understand the role of abiotic conditions in species zonation and primary succession on coastal dunes (Barbour et al. 1999, Boyce 1957, Cowles 1899, van der Valk 1974, Wilson and Sykes 1999), we know relatively little about the evolutionary responses of plant populations to the environmental heterogeneity of coastal dune systems. Tests of local adaptation in coastal dunes are affected not only by the dynamic coastal dune environment, which is frequently altered by sand erosion and accretion, but also by the growth habit of dune plants, most of which are long-lived and clonal (Cheplick and White 2002, Gray 1985, Knight and Miller 2004). Vegetative reproduction may reinforce local adaptation, through the level of clonal diversity as well as through clonal growth patterns, e.g., the establishment of independent daughter ramets in the same habitat as the maternal plant. Alternatively, guerrilla clonal growth patterns—spreading rather than clumping placement of ramets— may prevent the association of particular genotypes with certain habitats, decreasing the potential for population differentiation and local adaptation. Phenotypic plasticity is another response to heterogeneous environmental conditions (Schlichting and Pigliucci 1998, Via and Lande 1985), which may evolve when selective pressures fluctuate frequently (Schlichting and Pigliucci 1998). The ability to respond plastically is a means to deal with environmental variability, as adjusting its phenotype in response to particular environmental cues may allow an individual to maximize its fitness potential (Schlichting and Pigliucci 1998). Disturbances to the coastal dune habitat are frequent relative to its plant inhabitants’ long life spans. Erosion can result in the removal of foredune habitat, exposing plant populations previously situated on the backdunes to the unprotected foredune conditions. Alternately, sand accretion can result in the building of new dunes along the shoreline, so that established plant populations experience changed conditions when new dunes act as buffers from salt spray and sand movement. If the time scale of disturbance is shorter than the lifespan of an individual genet, one individual may produce multiple phenotypes—through clonal reproduction of daughter ramets—in response to fluctuating environmental conditions (van Kleunen and Fischer 2001). Thus, processes such as phenotypic plasticity may facilitate an individual’s success in environments which experience frequent disturbances. We used Uniola paniculata L. (Poaceae) (Sea Oats) to investigate populationlevel evolutionary responses, since its natural history, as a rhizomatous perennial grass which reproduces both clonally and sexually, is representative of many dune plant species (Wagner 1964). Uniola paniculata is a federally protected plant due to its role in stabilizing dune habitats, and is frequently used in dune restoration 2011 C.L. Gormally and L.A. Donovan 753 projects. Uniola paniculata is the dominant plant species on the primary dunes where it occurs, from Virginia southward along the Atlantic coast to the Bahamas and along the Gulf coast to Veracruz, Mexico (Wagner 1964). On the dunes, U. paniculata’s habitat spans a localized dune gradient from the dynamic foredunes (embryonic dunes) to the older, stabilized backdunes situated farther inland. We investigated whether foredune and backdune populations of U. paniculata were locally adapted to microhabitats across its shoreline-to-landward range on the coastal sand dunes. In a previous study, we documented intraspecific trait variation in populations of Uniola paniculata located in the dynamic foredunes and on the stabilized backdunes that mirrored the underlying habitat variation (Gormally and Donovan 2010). However, that study could not identify the underlying process driving this variation. A previous allozyme study demonstrated that there is no strong regional pattern of genetic structure among populations of U. paniculata, but fine-scale clonal structure and diversity vary widely, perhaps resulting from localized microhabitat differences (Franks et al. 2004). In this manipulative study, we asked whether populations of U. paniculata were locally adapted to foredune and backdune microhabitats as evidenced by ramets transplanted from their source habitats outperforming ramets from the other habitat. Field Site and Species Description All fieldwork was conducted on the coastal sand dunes at the National Estuarine Research Reserve System (NERRS) on Sapelo Island, located on the Atlantic coast of Georgia. Sapelo Island beach dunes are protected through the NERRS, as well as through the Georgia Coastal Ecosystems Long Term Ecological Research Network (GCE-LTER). Thus, these dunes are ideal settings to document the evolutionary responses of natural coastal dune plant populations, as the dune habitats are not currently impacted by anthropogenic development or foot traffic. Vegetation on the Sapelo Island dunes is typical of southeastern Atlantic coastal dunes, consisting primarily of clonal perennial grass and vine species. Uniola paniculata reproduces both vegetatively and sexually, although vegetative reproduction may contribute more to fitness than seed production, as a high percentage of ovules are aborted (Hester and Mendelssohn 1987). On Sapelo, U. paniculata’s growing season begins in late March, when plants develop new leaves to replace the previous year’s leaves, and ends in October, and plants are quiescent during the winter (Wagner 1964). On average, annual precipitation is 115 cm, with about 80% occurring during the growing season (2000–2005 Georgia Coastal Ecosystems Long-Term Ecological Research network data). From July 2006 until November 2008, conditions in coastal Georgia ranged from a moderate to severe drought (National Oceanic Atmospheric Administration 2009). Methods Reciprocal transplant plots We conducted two reciprocal transplant experiments, designated as Summer 2006 and Fall 2007 (referring to the date plants were transplanted into experimental plots). The foredune habitat was defined as the first 10 m closest to the shoreline, 754 Southeastern Naturalist Vol. 10, No. 4 and the backdune habitat was defined as the farthest inland edge of the species’ range, based on previous characterization of the environment at this study site (Gormally and Donovan 2010). The foredune habitat is characterized by greater levels of soil boron, potassium, magnesium, manganese, as well as higher soil salinity, pH, and sand accretion, than the backdune habitat, which experiences less sand movement in terms of accretion and erosion and is consequently a more stable habitat (Gormally and Donovan 2010). For both the Summer 2006 and Fall 2007 experiments, we set up three 3-m2 reciprocal transplant plots in each habitat (foredune and backdune), with the plots located at least 300 m apart within each of the habitats (corresponding to 3 of the transects in Gormally and Donovan 2010). Plot locations in the backdune habitat varied from 40 to 60 m inland from the foredune, according to the presence of the species, which varied with island geomorphology (Stallins and Parker 2003). Existing vegetation was cleared from each plot at least four months prior to transplanting. Plots were weeded to remove other vegetative growth throughout the course of both experiments. The U. paniculata ramets were collected from the foredune and backdune habitats and are referred to as the foredune (FD) and backdune (BD) populations, respectively. To ensure that ramets represented distinct genets, they were always dug up from locations separated by at least 3 m (Franks et al. 2004). When collected, the ramet consisted of a tiller—an aboveground stem formed from an axillary bud on a maternal ramet—and a variable amount of roots. For the Summer 2006 experiment, we collected 412 total ramets in June 2006 (n = 207 FD ramets and n = 205 BD ramets). Ramets were treated with Hormex to induce rooting, fertilized once with Jack’s Peat-Lite fertilizer (NPK 20-10-20; 200 ppm N), and planted in 25.4-cm3 pots containing sand. Plants were maintained in an open-air shaded greenhouse on Sapelo to facilitate root growth and minimize environmental or maternal effects (Roach and Wulff 1987). In August 2006, the surviving ramets were transplanted into experimental plots, 30 ramets per plot (n = 89 FD ramets and n = 91 BD ramets), for a total of n = 180 ramets in the Summer 2006 experiment. Plants were watered for two weeks following transplant to promote establishment. We measured the stem basal diameter and stem height (from the base of the plant at the soil surface, to the uppermost node) of the Summer 2006 ramets at transplant (August 2006) to account for initial size, and then at six months (February 2007) to assess growth. The Summer 2006 foredune plots were completely washed away during storms in April 2007. Backdune plots remained, and surviving ramets were measured for height and stem diameter at 1 year (August 2007), and aboveground biomass was harvested at 2 years (September 2008). At harvest, ramets were assessed for stem diameter, height, and number of tillers, i.e., number of new stems. Biomass was weighed, after drying at 60 °C. The Fall 2007 experiment was set up after the Summer 2006 foredune plots were washed away. For this experiment, we adjusted the experimental design to include more ramets and collected two ramets per genet in order to place the same genotype into plots in each habitat (360 ramet pairs for a total of n = 720 individual ramets). Ramets were treated with Hormex to induce rooting and fertilized once with Jack’s Peat-Lite fertilizer (NPK 20-10-20; 200 ppm N). The ramets were initially planted in 25.4-cm3 pots of sand and maintained in the University of Georgia Plant Biology greenhouses in Athens, GA, for approximately 4.5 2011 C.L. Gormally and L.A. Donovan 755 months to facilitate root growth and minimize environmental or maternal effects (Roach and Wulff 1987). Ramets were watered regularly. In November 2007, the 326 surviving ramets (161 FD and 165 BD ramets) were transplanted into experimental plots, and were watered for two weeks to promote establishment. We measured the wet weight of the Fall 2007 ramets at transplant to account for initial size. The Fall 2007 foredune plots washed away during storms in December 2007, so no growth measurements were possible for these plots. The Fall 2007 backdune plots remained and surviving ramets were harvested after 1 year (September 2008). At harvest, ramets were assessed for stem diameter, height, number of tillers, and dry biomass (dried at 60° C). Statistical analysis For the Summer 2006 experiment, we compared survival and growth of FD and BD ramets in foredune and backdune plots at 6 months, and in backdune plots at 1 year and at harvest. Survival was analyzed with a chi square (χ2) test. Growth traits (stem diameter, height, and biomass) of surviving ramets met assumptions of normality and heterogeneity and were analyzed with analysis of covariance (ANCOVA), with source population as the main predicting variable. For the Summer 2006 experiment, stem diameter or height at transplant was included as the covariate. For the Fall 2007 experiment, wet weight at transplant was included as the covariate. To determine whether the amount of vegetative reproduction, as indicated by number of tillers, differed between populations, we used analysis of variance (ANOVA). Results Summer 2006 experiment At six months after transplantation into reciprocal transplant plots, there was no evidence that either FD or BD ramets were locally adapted to their respective home habitats, using non-destructive growth measurements of height and stem diameter (Tables 1 and 2). Although there was a significant population effect for stem diameter in the backdune plots (Table 2), with FD plant stem diameter having increased more than that of BD ramets during this interval, this is opposite of the pattern that would support local adaptation in the backdune habitat. Nor was there evidence of differential survival in either habitat (foredune: χ2 = 0.37, df = 3, n = 89; backdune: χ2 = 0, df = 3, n = 90). After a year of growth in the backdune plots, there were no differences between FD and BD ramets for survival (χ2 = 0.79, df = 3, n = 90), height or stem diameter (Table 2). At harvest, only 15 ramets were alive (7 FD and 8 BD ramets), and there was no differential survival of FD and BD ramets (χ2 = 0.00891, df = 3, n = 90). There were also no differences in size of surviving FD and BD ramets at harvest: stem diameter, height, total dry biomass, and vegetative reproduction (estimated as number of tillers produced). Fall 2007 experiment At harvest, only 13 ramets (7 FD and 6 BD ramets) remained alive in the backdune plots, with no differential survival by population (χ2 = 0.05829, df = 3, 756 Southeastern Naturalist Vol. 10, No. 4 n = 156). At harvest, there were no differences between populations in stem diameter or height, total dry biomass or number of tillers produced (Table 3). Discussion Our study contributes to the small but growing body of evidence that clonal plants are more likely to respond through the strategy of phenotypic plasticity than local adaptation in frequently disturbed habitats. Using measurements of growth in the field and, most significantly, survival in the field during an extreme drought, we found no evidence that populations of U. paniculata were locally adapted to specific microhabitats across a shoreline-to-landward environmental gradient. This finding helps to explain the results from a previous study that showed that FD ramets were morphologically and physiologically different (larger in height, stem diameter, number of nodes, with higher aboveground tissue concentrations of N and K) and more likely to flower than ramets located farther inland (Gormally and Donovan 2010). The phenotypic variation detected in the previous study mirrored the underlying environmental variation, with increased soil salinity, sand accretion, soil pH, and concentrations of soil nutrients (B, K, Mg, and Na) in the 10-m interval closest to the shoreline. In order to demonstrate that populations were locally adapted, transplant populations would have to outperform populations from other sites in the reciprocal transplant plots located in Table 1. Uniola paniculata growth for Summer 2006 experiment foredune plots measured at 6 months after transplant. Least square means (LSmeans) ± 1 SE are presented. (*** signifies P ≤ 0.0001). Foredune plot ANCOVA effects Covariate BD ramets FD ramets (initial size) Population Height (cm) 14.173 ± 0.952 13.422 ± 0.967 F(1,62) = 45.44*** F(1,62) = 0.29 Stem diameter (cm) 2.450 ± 0.165 2.898 ± 0.187 F(1,68) = 2.05 F(1,68) = 3.17 Table 2. Uniola paniculata growth (measured in cm) for Summer 2006 experiment backdune plots measured at 6 months, 1 year, and harvest (2 years). LSmeans ± 1 SE are presented. (* signifies P < 0.05, ** signifies P < 0.01, *** signifies P ≤ 0.0001). Backdune plot ANCOVA effects Covariate BD ramets FD ramets (initial size) Population Measurements at 6 months Height 11.821 ± 1.086 13.058 ± 1.073 F(1,82) = 39.26*** F(1,82) = 0.61 Stem diameter 2.327 ± 0.187 3.051 ± 0.182 F(1,81) = 0.69 F(1,81) = 7.34** Measurements at 1 year Height 53.872 ± 2.691 54.228 ± 2.812 F(1,72) = 9.08* F(1,72) = 0.01 Stem diameter 3.808 ± 0.264 3.774 ± 0.272 F(1,67) = 0.03 F(1,67) = 0.01 Measurements at harvest Biomass 9.899 ± 1.752 10.431 ± 1.874 F(1,12) = 1.01 F(1,12) = 0.04 Height 50.333 ± 3.817 57.705 ± 4.083 F(1,12) = 1.79 F(1,12) = 1.72 Stem diameter 3.370 ± 0.323 3.838 ± 0.426 F(1,10) = 0.02 F(1,10) = 0.65 Tillers 2.375 ± 0.676 2.429 ± 0.723 n/a F(1,13) = 0 2011 C.L. Gormally and L.A. Donovan 757 their source habitat (Linhart and Grant 1996, Kawecki and Ebert 2004). However, there was no evidence from our reciprocal transplant experiments to support the hypothesis that the trait variation previously documented in naturally occurring populations was due to local adaptation or genetic differentiation of any kind. Early growth measurements from both the foredune and backdune plots provided no evidence of population differentiation. At harvest, there was no differential survival by population in the backdune plots. Though the plots in the backdune habitat were relatively unaffected by the nor’easters, mortality was substantial, with <10% of ramets from both the Summer 2006 and Fall 2007 experiments surviving to the time of harvest. Growth and survival following transplantation were likely negatively impacted by the moderate to extreme drought conditions which lasted from May 2006 to August 2008. Nearly 30% of the ramets in the Summer 2006 experiment died during the six-month period of May–December 2007, characterized as a severe to extreme drought, according to the NOAA Palmer Drought Index (National Oceanic Atmospheric Administration 2009) Our results are consistent with the few studies that explicitly test for local adaptation of a coastal perennial. Populations of Triplasis purpurea (Walter) Chapm. (Purple Sandgrass) located 15 m and 80 m from the shoreline, were not locally adapted (Cheplick and White 2002). Foredune and mature dune populations of Ammophila arenaria (L.) Link (European Beachgrass) were not locally adapted (Gray 1985). Although a study of the coastal perennial vine Hydrocotyle bonariensis Comm. Ex Lam (Largeleaf Pennywort) did find evidence of local adaptation, these populations were adapted to microhabitats located at high and low dune heights (Knight and Miller 2004), different from the microhabitats tested in our study. Additionally, local adaptation of H. bonariensis appeared to be associated with interactions of the surrounding vegetation at each microhabitat (Knight and Miller 2004). Most dune plant species are clonal perennials, propagating both by sexual and vegetative reproduction. Though clonal reproduction does not include recombination, clonal plants are not less genetically diverse (Hamrick and Godt 1989). One expectation is that clonality would increase the likelihood of local adaptation through the reduction of gene flow and increased placement of a genet’s ramets into preferentially locally adaptive sites. However, clonality may reduce local adaptation if genets are adapted to historical environmental conditions, so that populations might be genetically differentiated, but not locally adapted to their current habitat, but instead to past conditions (Callaghan et al. 1996). Local adaptation Table 3. Uniola paniculata measurements at harvest (1 year) for the Fall 2007 experiment. LSmeans ± 1 SE are presented. (* signifies P < 0.05). Backdune plot ANCOVA effects Covariate BD ramets FD ramets (initial size) Population Biomass (g) 4.756 ± 0.80 3.22 ± 0.738 F(1,10) = 7.66* F(1,10) = 1.91 Height (cm) 33.259 ± 7.743 32.564 ± 7.144 F(1,10) = 1.33 F(1,10) = 0 Stem diameter (cm) 2.686 ± 0.395 2.259 ± 0.365 F(1,10) = 3.22 F(1,10) = 0.60 Tillers (#) 1.333 ± 0.182 1.143 ± 0.169 F(1,10) = 0.05 F(1,10) = 0.41 758 Southeastern Naturalist Vol. 10, No. 4 may also be constrained when strong gene flow prevents differentiation, when selection is constrained by low amounts of genetic variation, or when natural selection fluctuates due to strong spatial variability. Given the changeable nature of the coastal sand dune environment, one genet might experience multiple environments over the course of its life. Under fluctuating environmental conditions, plasticity may be more likely to have evolved than local adaptation (Mitchell-Olds 1992, Richards et al. 2010). In a greenhouse study of the clonal salt marsh perennial Borrichia frutescens L. (Bushy Seaoxeye), Richards et al. (2010) determined that trait variation along a salinity gradient was due to phenotypic plasticity. In this common garden study, plants from microhabitats along the salinity gradient did not respond differentially to salinity treatments; instead, plants responded plastically for traits measured. It seems likely that the trait variation we previously documented (Gormally and Donovan 2010) was a result of phenotypic plasticity rather than local adaptation (Schlichting and Pigliucci 1998, Via and Lande 1985). Our study highlights some of the challenges implicit in addressing the question of how long-lived plants respond to selection in a variable environment that may change through the course of an individual’s lifetime. Since U. paniculata is a long-lived perennial grass, identifying the environments to which these populations may be adapted is difficult. On the coastal dunes, selective pressures may fluctuate frequently, particularly due to periods of sand accretion and erosion, sometimes resulting in the removal of foredunes, exposing populations of plants previously situated on the backdunes to increased disturbance from sand accretion and erosion. Further, as a clonal plant capable of sexual and vegetative reproduction, U. paniculata may respond to its environment at both the level of the genet—through the processes of local adaptation or phenotypic plasticity— and the ramet—through physiological integration, developmental plasticity, and selective placement of daughter ramets (van Kleunen and Fischer 2001). We used ramets rather than seedlings due to concerns about seedling survival following transplantation, but it is possible that seedlings might respond differently than ramets to transplantation and might differ in average lifespan. Additional knowledge about the average lifespan of an individual ramet, as well as the average lifetime of a genet, would enhance the study of evolutionary responses of plant populations to the environmental heterogeneity of coastal dune systems. Our finding of no evidence of local adaptation in U. paniculata may have positive implications for conservation and restoration efforts. Despite dramatic environmental differences across the coastal dunes, we found no evidence that populations are locally adapted across dune microhabitats. This means that plant material for this species can be sourced without concern for the visible intraspecifi c variation that exists in these microhabitats, since genotypes from each microhabitat should be equally likely to flourish in any habitat. However, studies of populations across the species’ entire range are needed in order to understand larger geographic patterns. Current conservation and restoration considerations underscore the necessity for acquiring a better understanding of the evolutionary responses of natural plant populations on the coastal dunes, despite the challenges implicit in addressing these questions. 2011 C.L. Gormally and L.A. Donovan 759 Acknowledgments The authors wish to thank the following people for help with fieldwork: Beau Brouillette, Jason Bonner, Kate Seader, Allison Hennigan, Anna Johnson, Nicole Umberger, Meredith Barton, Victor Thompson, Scott Gevaert, Katherine Hale, Anna Harvey, Mike Boyd, Samantha Carvalho, Patrick Gormally, Maggie Kilgo, and Haley Zapal. Research at the Sapelo Island National Estuarine Research Reserve (SINERR) was facilitated by the research reserve coordinator, Dorset Hurley, and by Jon Garbisch at the University of Georgia Marine Institute. Tom Patrick at the Georgia Department of Natural Resources Wildlife Resources Center provided invaluable help in the permitting process for ramet collection. We thank Jim Hamrick and Eleanor Pardini for insightful comments throughout the experimental design process. We thank SINERR (NOAA) (NA07NO54200039), Sea Grant (NA04OAR4170033), and the Georgia Botanical Society for financial support. This is contribution number 988 from the University of Georgia Marine Institute. Literature Cited Antonovics, J., and A.D. Bradshaw. 1970. Evolution in closely adjacent plant populations. VIII. Clinal patterns at a mine boundary. Heredity 25:349–362. Barbour, M.G., J.H. Burk, W.D. Pitts, F.S. Gilliam, and M.W. Schwartz (Eds.). 1999. Terrestrial Plant Ecology. Benjamin Cummings, Menlo Park, CA. Boyce, S.G. 1957. The salt spray community. Ecological monographs 24:29–67. Callaghan, T.V., A. Carlsson, and B. M. Svensson. 1996. Some apparently paradoxical aspects of the life cycle, demography, and population dynamics of plants from the subarctic Abisco area. Ecological Bulletin 45:133–143. Cheplick, G.P., and T.P. White. 2002. Saltwater spray as an agent of natural selection: No evidence of local adaptation within a coastal population of Triplasis purpurea (Poaceae) American Journal of Botany 89:623–631. Claudino-Sales, V., P. Wang, and M.H. Horwitz. 2007. Factors controlling the survival of coastal dunes during multiple hurricane impacts in 2004 and 2005: Santa Rosa barrier island, Florida. Geomorphology 95:295–315. Clausen, J., D.D. Keck, and H.M. Hiesey. 1940. Experimental studies on the nature of species. I. Effects of varied environments on western North American plants. Carnegie Institute of Washington, Washington, DC. Cowles, H.C. 1899. The ecological relations of the vegetation of the sand dunes of Lake Michigan. Botanical Gazette 27:95–391. Doing, H. 1985. Coastal fore-dune zonation and succession in various parts of the world. Vegetatio 61:65–75. Franks, S.J., C.L. Richards, E. Gonzales, J.E. Cousins, and J.L. Hamrick. 2004. Multiscale genetic analysis of Uniola paniculata (Poaceae): A coastal species with a linear, fragmented distribution. American Journal of Botany 91:1345–1351. Galloway, L.F., and C.B. Fenster. 2000. Population differentiation in an annual legume: Local adaptation. Evolution 54:1173–1181. Gormally, C., and L.A. Donovan. 2010. Responses of Uniola paniculata L.(Poaceae), an essential dune-building plant, to complex changing environmental gradients on the coastal dunes. Estuaries and Coasts 33:1237–1246. Gray, A.J. 1985. Adaptation in perennial coastal plants, with particular reference to heritable variation in Puccinellia maritima and Ammophila arenaria. Vegetatio 61:179–188. Gregor, J.W. 1930. Experiments on the genetics of wild populations part I. Plantago maritima. L. Journal of Genetics 22:15–25. Hamrick, J.L., and M.J. Godt. 1989. Allozyme diversity in plant species. Pp. 43–63, In A.H.D. Brown, M.T. Clegg, A.L. Kahler, and B.S. Weir (Eds.). Plant Poulation Genetics, Breeding, and Germplasm Resources. Sinauer, Sunderland, MA. 760 Southeastern Naturalist Vol. 10, No. 4 Hereford, J. 2009. A quantitative survey of local adaptation and fitness trade-offs. The American Naturalist 173:579–588. Hester, M.W., and I.A. Mendelssohn. 1987. Seed production and germination response of four Louisiana populations of Uniola paniculata (Gramineae). American Journal of Botany 74:1093–1101. Jerling, L. 1985. Population dynamics of Plantago maritima along a distributional gradient on a Baltic seashore meadow. Vegetatio 61:155–161. Kawecki, T.J., and D. Ebert. 2004. Conceptual issues in local adaptation. Ecology Letters 7:1225–1241. Knight, T.M., and T.E. Miller. 2004. Local adaptation within a population of Hydrocotyle bonariensis. Evolutionary Ecology Research 6:103–114. Leimu, R., and M. Fischer. 2008. A meta-analysis of local adaptation in plants. PLoS ONE 3. Linhart, Y.B., and M.C. Grant. 1996. Evolutionary significance of local genetic differentiation in plants. Annual Review of Ecology and Systematics 27:237–277. Maun, M.A., and J. Perumal. 1999. Zonation of vegetation on lacustrine coastal dunes: Effects of burial by sand. Ecology Letters 2:14–18. Miller, R.E., and N.L. Fowler. 1994. Life-history variation and local adaptation within two populations of Bouteloua rigidiseta (Texas Grama). Journal of Ecology 82:855– 864. Mitchell-Olds, T. 1992. Does environmental variation maintain genetic variation? A question of scale. TRENDS in Ecology and Evolution 7:397–398. National Oceanic Atmospheric Administration. 2009. National Climatic Data Center. Available online at http://www.noaa.gov. Accessed 1 September 2009. Oosting, H.J., and W.D. Billings. 1942. Factors affecting vegetational zonation on coastal dunes. Ecology 23:131–142. Richards, C.L., S.N. White, M.A. McGuire, S.J. Franks, L.A. Donovan, and R. Mauricio. 2010. Plasticity, not adaptation to salt level, explains variation along a salinity gradient in a salt marsh perennial. Estuaries and Coasts 33(4):840–852. Roach, D.A., and R.A. Wulff. 1987. Maternal effects in plants. Annual Review of Ecology and Systematics 18:209–235. Schlichting, C.D., and M. Pigliucci. 1998. Phenotypic evolution: A reaction norm perspective. Sinauer Associates, Inc., Sunderland, MA. Smith, R.L., and T.M. Smith (Eds.). 2001. Ecology and Field Biology. Benjamin Cummings, New York, NY. Stallins, J.A., and A.J. Parker. 2003. The influence of complex systems interactions on barrier island dune vegetation pattern and process. Annals of the Association of American Geographers 93:13–29. Turesson, G. 1922. The genotypical response of the plant species to habitat. Hereditas 3:211–350. van der Valk, A.G. 1974. Environmental factors controlling the distribution of forbs on coastal foredunes in Cape Hatteras National Seashore. Canadian Journal of Botany 52:1057–1073. van Kleunen, M., and M. Fischer. 2001. Adaptive evolution of plastic foraging responses in a clonal plant. Ecology 82:3309–3319. Via, S., and R. Lande. 1985. Genotype-environment interaction and the evolution of phenotypic plasticity. Evolution 39 505–522. Wagner, R.H. 1964. The ecology of Uniola paniculata L. in the dune-strand habitat of North Carolina. Ecological Monographs 34:79–96. Wilson, J.B., and M.T. Sykes. 1999. Is zonation on coastal sand dunes determined primarily by sand burial or by salt spray? A test in New Zealand dunes. Ecology Letters 2:233–236.