Barrier islands include some of the most endangered and fragmented ecosystems on the Atlantic coast, providing critical habitat for many species, including some that are threatened and endangered. As the vast majority of these islands have been developed for human usage study and protection of the few remaining undeveloped and undisturbed islands is critical. This study was undertaken in order to characterize the vascular plant communities on Bird Island, an uninhabited, undeveloped barrier island on the border of North and South Carolina, with the objectives of a thorough survey of flora, vegetation, and environment, classification of plant communities, and multivariate analysis of vegetation and environmental data. A floristic inventory of the island and its associated marshes was conducted during the growing season (May-November) of 2002 and 2003. One hundred four 100m² plots were inventoried for vegetation and environment using protocols developed by the Carolina Vegetation Survey. Plant communities were identified according to the National Vegetation Classification, the Classification of the Natural Communities of North Carolina, and the Carolina Vegetation Survey. Interpretation of vegetation patterns was based on multivariate analysis of vegetation and environmental data. Ninety-one vascular plant species in 35 families, including 4 exotic species, were distributed across 12 communities. Communities on Bird Island appear to be distinctive when compared to those described for other barrier islands in the region. Additionally, the vegetation survey on Bird Island revealed suitable habitat for the federally listed Seabeach amaranth (<I>Amaranthus pumilus</I>); an important dune-building annual of the North American Atlantic coast. Surveys of the late 1980s and early 1990s documented small populations of Seabeach amaranth on Bird Island, but our work found no indication of a population in either 2002 or 2003. Seabeach amaranth?s existence range-wide is threatened by habitat loss and fragmentation, herbivory, and anthropogenic disturbances such as beach driving, hardening, and raking. Published work on this species has been limited, but observations of natural, experimental, and restored populations have indicated relationships between distance from the ocean and both survivorship and reproduction. To quantify this relationship, 314 plants were transplanted at varying distances from the ocean. Plants were monitored monthly from June until December 2003. We found that distance from the ocean had a significant effect on survivorship, size, and reproduction; however, presence of webworms and ghost crabs did not have significant effects on size or reproduction. Distance from the ocean combined with complex factors such as over wash and soil salinity appear to be important indicators of success for both natural and restored populations of this species, and landscape position should be incorporated into future conservation and restoration efforts for A. pumilus in the Carolinas and elsewhere. Overall, we found that the vegetation of Bird Island is mostly intact, with few exotic, invasive species present. Bird Island?s protected status and limited presence of invasive species make it suitable habitat for continued protection and further restoration of rare, threatened, or endangered species, such as Seabeach amaranth. Atlantic barrier islands in general provide distinct community types combined with a small species pool, a combination that may provide a model for examination of larger ecological questions.

The potential for plantations to catalyze forest regeneration on highly degraded land and the need to research management methods to increase biodiversity within plantations have been widely recognized. Our study investigated the effects of creating artificial canopy gaps by girdling exotic timber trees in plantations on the germination, growth, and survival of native tree species that may be important food plants for the Puerto Rican parrot. We found that seedling growth increased significantly in gaps; there were no differences in survival or germination between gap and closed plots. Percentage cover of grasses, shrubs, and vines increased but did not prevent tree seedlings from growing significantly more in gaps. Leaf litter removal had no effect on germination, growth, or survival. Both local and landscape level diversity is predicted to increase in gaps if large saplings present in the understory replace the girdled timber trees, but plantations will still be dominated by exotics and timber species. Our results suggest that restoration of native forest diversity in plantations will require continued management to remove exotic species and promote growth of tree species with high wildlife habitat value.

Carolina Bays are unique elliptical depressional wetlands that are abundant in the Atlantic Coastal Plain, particularly in North Carolina. A large percentage of these bays have been ditched, drained, logged, or otherwise disturbed by humans in the past 200 years. Currently, Carolina Bay restoration provides a means for mitigating wetlands lost to development. The purpose of this study was to characterize compositional variation of vegetation relative to gradients of environmental conditions in three Carolina Bays in order to provide reference data for restoring impacted bays. Within three unaltered Carolina Bays, soil sampling, water table monitoring, and vegetation assessments were conducted. Four community types were identified: pond pine woodland, non-riverine swamp forest, high pocosin, and bay forest. Distributions of these community types were strongly correlated to depth of organic material and associated soil properties and hydrologic regimes. Pond pine woodland dominated mineral soils; both pond pine woodland and nonriverine swamp forest were found on histic and shallow organic soils; both high pocosin and bay forest were found on deep organic soils. Examination of the data included simple averaging, single factor analyses of variance, Fisher?s protected least significant differences, chi-square tests of independence, cluster analyses, and ordination procedures. The application of this material was intended for a North Carolina Department of Transportation Carolina Bay mitigation project and future Carolina Bay restoration projects in this region.

Hybrid zones have long been touted by plant evolutionary biologists as an unrivaled phenomenon through which the mechanisms of evolution may be studied as they are occurring. Speciation, introgression, and adaptation may all occur as a result of their formation and thus their study may provide new insight into how these processes occur. The purpose of this study is to determine the manner in which a broad hybrid zone of <i>Aesculus</i> was formed, through the use of chloroplast DNA analysis. This hybrid zone encompasses parts of central and northern Georgia and includes hybrids among three species of <i>Aesculus</i> sect. <i>Pavia</i>: <i>Aesculus flava</i>, <i>A. pavia</i>, and <i>A. sylvatica</i>. These species currently have distinct geographic distributions, with both <i>A. pavia</i> and <i>A. flava</i> absent from the hybrid zone. Previous hypotheses have purported that the <i>Aesculus</i> hybrid zone was formed through either 1) secondary contact of previously isolated species or 2) recurrent long-distance pollen dispersal via the ruby-throated hummingbird, <i>Archilochus colubris</i>. It is also possible that the zone may have originated through a combination of both of these forces. This study consists of two parts, the first involving the verification of maternal inheritance of chloroplasts in <i>Aesculus</i>, and the second involving the assessment of the hypothesis of historical secondary contact, based on patterns of cpDNA variation in hybrid and parental populations. Verification of the inheritance of chloroplasts was accomplished through the sequencing of the <i>matK</i> gene from the parents and progeny of 17 crosses among various <i>Aesculus</i> species. The relative contribution of historical secondary contact to the formation of the hybrid zone was accomplished through PCR-RFLP analysis of three loci in the chloroplast genome: <i>matK</i>, <i>trnD</i>-<i>trnT</i>, and <i>trnH</i>-<i>trnK</i>, from 29 natural populations of <i>Aesculus</i> located within, adjacent to, and broadly separate from the hybrid zone. Haplotypes identified from RFLP analysis of the three loci were sequenced and subjected to phylogenetic and population genetic analyses. Results from the sequencing of the <i>matK</i> gene from controlled crosses verified that chloroplasts are inherited maternally in <i>Aesculus</i>, as in most angiosperms. Twenty-one unique haplotypes were identified via analysis of RFLPs, indicating that the chloroplast genome of <i>Aesculus</i> is highly polymorphic. Phylogenetic and population genetic analyses of sequence and restriction site data revealed that cpDNA haplotypes do not correlate with either spatial or taxonomic boundaries. Haplotypes of <i>A. pavia</i>, a species that is presently physically absent from the hybrid zone, were detected in hybrid populations. Additionally, most populations were fixed for a single haplotype that was unique to that population. These results suggest that while secondary contact has played a historical role in the formation of the hybrid zone, current gene flow via seeds is highly restricted among populations. Furthermore, multiple types of cpDNA, originating from the last common ancestor of the group, are maintained in hybridizing species, suggesting that the common ancestor of species of sect. <i>Pavia</i> was polymorphic. Comparison of these results to those of previous analysis of allozyme markers suggests that intermittent long-distance pollen dispersal has helped to maintain the hybrid zone while localized gene flow due to secondary contact of divergent species was responsible for the initial formation of the hybrid zone. Future studies of the <i>Aesculus</i> hybrid zone might focus on the fitness of hybrid individuals as a determining factor in the continued maintenance of the hybrid zone, as well as genotype-by-environment interactions that may be influencing the physical shape and geographic location of the <i>Aesculus</i> hybrid zone. This study could lead to the development of a new hybrid zone model, which would be particularly well-suited for plants capable of long-distance dispersal.

Across plant taxa there are large-scale correlations among ecophysiological leaf traits: Specific leaf area (SLA), maximum photosynthetic rate (Amax), Leaf nitrogen content (NL), and leaf lifespan (LL). Suite syndromes of leaf traits are associated among r and K-selected species and Arabidopsis thaliana ecotypes exhibit groth syndromes similar to the r and K-selection theory and range from a wide variety of latitudes of origin. The purpose of this study was to determine if the ecoyphysiological leaf traits: SLA, Amax, Nl, LL and other growth-rleated traits were correlated across A. thaliana ecotypes from a wide range of latitudes of origin. Fifteen A. thaliana ecotypes were grown in an experimental growth chamber to determine if these ecophysiological and other growth-related leaf traits such as leaf initiation rate and bolting time were correlated across the ecytpes. Results indicate that SLA, Amax and NL were correlated across the ecotypes as consistent with global trends while LL was not significantly correlated. The life history trait, bolting time, was correlated with leaf tratis. There was a wide range of leaf age and change in longevity among the leaves, making leaf lifespan difficult to determine. The results from this study indicate that A. thaliana will be a good model in order to understand the developmental and genetic links with these correlations.

Echinacea laevigata is a federally endangered species and a close relative of the medicinally important E. purpurea. The species has 24 recognized populations restricted to four states (VA, NC, SC, GA). To determine the population structure and outcrossing rate across the range of the species, we conducted AFLP analysis using four primer combinations for 22 populations. The genetic diversity of this species was found to be high based on the level of polymorphic loci (200 of 210 loci; 95.24%) and Nei?s gene diversity (ranging from 0.1398 to 0.2606; overall 0.2611). There was significant population genetic differentiation (GST of 0.2941), suggesting possible adaptation to local environments. Results from the AMOVA analysis suggest that a majority of the genetic variance is attributed to variation within populations (70.26%). These results are congruent to a previous allozyme study that examined the genetic makeup of 11 of the 24 populations, excluding any Georgia populations. An isolation by distance (IBD) analysis indicated that genetic differentiation among populations is a function of geographic distance, although long-distance gene dispersal (LDGF) between some populations was suggested based on the NJ tree. An estimate of the outcrossing rate based on genotypes of progenies from six of the 22 populations using a multilocus estimate was 0.833-1.2, where 1.2 is considered complete outcrossing, suggesting that the species is predominantly outcrossing. The remaining populations of E. laevigata have significant levels of population diversity, which is encouraging because the long-term survival of this species will depend on moderate to high levels of genetic diversity and management efforts can focus on other issues instead of increasing the genetic diversity.

Cool Springs Environmental Education Center (CSEEC), owned by Weyerhaeuser Company, includes a 591 ha State Significant Natural Area. It is located in Craven County, North Carolina, in the floristically rich Atlantic Coastal Plain. A comprehensive inventory of the vascular flora at CSEEC has documented the occurrence of approximately 567 specific and sub-specific taxa and 303 genera in 118 plant families as well as populations of the Atlantic Coastal Plain endemics Litsea aestivalis (L.) Fern. (Pondspice) and Solidago villosicarpa LeBlond (Coastal Goldenrod). Twenty distinct plant community types were identified including the uncommon Longleaf Pine (Pinus palustris P. Mill) Woodland, Taxodium distichum (L.) L.C. Richard (Bald Cypress?Tupelo Gum) ? Nyssa aquatica L. (Swamp Tupelo) Swamp, a number of botanically heterogeneous small depression wetlands, and the novel Quercus hemisphaerica Bartr. ex Willd (Sand Laurel Oak). ? Pinus taeda L. (Loblolly Pine) Woodland. The high diversity of plant communities reflects the high diversity of soil types ranging from excessively drained deep sands to very poorly drained organic soils.

Wounding stress is a continuous threat to the survival of all organisms, and, in crop plants it leads to a dramatic reduction in crop yield. The signaling pathways that allow plants to respond to wounding stress are known to be complex. Many plant hormones such as jasmonic acid (JA), salicylic acid (SA), and ethylene have been identified as critical factors for the induction of wound-responsive genes in a number of plant species. However, the specific roles of each hormone in vivo, and how different hormones interact in the wound-stress response are largely unknown. Elucidation of the specific interactions of these plant hormones in response to wound stress is a major focus of wound research. The major goal of this research is to utilize Arabidopsis JA, SA, and ethylene biosynthetic and signaling mutants to analyze the role of each hormone separately and together in wound-responsive gene induction. Arabidopsis wild type, ein2, npr1, jar1, npr1/ein2, and npr1/jar1 plants were wounded and sampled at different times, after which RNA was extracted, subjected to electophoresis, transferred to filters and probed for several putative wound-regulated transcripts, which included PDF1.2, PR1, WAK1, and, LOX2. The constitutive levels of transcript expression as well as patterns of accumulation of these transcripts in response to the wounding stimulus varied in the different mutants.

<p>A method of rapid field assessment and predictive modeling was developed to characterize vegetation communities of the Gorges State Park and Gamelands and to create predictive community maps for the area. This method placed an emphasis on locating rare communities using expert information, existing databases, aerial photography, and random encounters in efforts to provide information to researchers and park personnel about community locations. Approaches for classifying these communities were examined to identify which would provide suitable units for modeling community types. I sampled 102 field locations and assigned their vegetation to 16 different community types. Predictive community maps were generated using discriminant functions incorporating digital terrain data, including elevation, slope, relative slope position, terrain shape index, and weighted landform index. Three sets of discriminant functions were created to meet the different needs of persons interested in using these maps. Photo-interpreted cover classes were also including in the modeling process as filters. Map accuracies ranged from 65% to 75%, with those using only discriminant functions (without filtering) yielding higher accuracies.<P>

<p>Dinoflagellates represent a diverse group of both auxotrophic and heterotrophic protists. Most heterotrophic dinoflagellates are raptorial feeders that encounter prey using ?temporal-gradient sensing? chemotaxis wherein cells move along a chemical gradient in a directed manner toward the highest concentration. Using short-term ?memory? to determine the orientation of the gradient, dinoflagellates swim in a ?run-and-tumble? pattern, alternating directed swimming with rapid changes in orientation. As the extracellular concentration of the attractant increases, a corresponding increase in the ratio of net-to-gross displacement results in overall movement toward the stimulus.The dinoflagellates Pfiesteria piscicida and P. shumwayae are heterotrophic estuarine species with complex life cycles that include amoeboid, flagellated, and cyst stages, that have been implicated as causative agents in numerous major fish kills in the southeastern United States These organisms show documented ?ambush-predator? behavior toward live fish in culture, including rapid transformations among stages and directed swimming toward fish prey in a manner that suggests the presence of a strong signalling relationship between live fish and cells of Pfiesteria spp.Zoospores of the two species of Pfiesteria can be divided into three functional types: TOX-A designates actively toxic isolates fed on fish prey; TOX-B refers to temporarily non-toxic cultures that have recently (1 week to 6 months) been removed from fish prey (and fed alternative algal prey); and NON-IND refers to isolates without apparent ichthyotoxic ability (tested as unable to kill fish in the standardized fish bioassay process; or without access to fish for ca. 1.5 years). Several Pfiesteria-like dinoflagellates have been isolated from samples in which P. piscicida and P. shumwayae are also present, including several cryptoperidiniopsoid species that have repeatedly been tested as lacking ichthyotoxic capability under ecologically relevant conditions (cell densities that occur in estuaries).Microcapillary assay techniques were employed to determine the attraction of P. piscicida and P. shumwayae zoospores to sterile-filtered fish mucus and excreta. Differences in attraction were measured among functional types, and between these two species and several isolates of cryptoperidiniopsoids, in ten-minute trials in which zoospores entering tubes filled with test substances were observed and counted. TOX-A zoospores of both P. piscicida and P. shumwayae were strongly and comparably attracted to fish secreta/excreta, relative to their behavior toward microcapillary tubes that were filled with filtered seawater. TOX-B zoospores of both Pfiesteria species showed intermediate attraction toward fish materials that appeared to be inversely related to time isolated from fish prey. NON-IND zoospores exhibited low attraction to fish materials. Cryptoperidiniopsoid zoospores showed moderate attraction with no apparent influence of previous exposure to fish.In an additional experiment that examined the signal activity in fish materials over time after collection from fresh fish, unfiltered fish materials ceased to attract P. piscicida zoospores after approximately 48 hours and ultrafiltered materials maintained attractive ability over the duration of the experiment (72 hours). These data show that filtration of fish materials may be used to extend the useful life of the chemical signal, possibly by removing bacteria that consume or degrade it.In recent years, several researchers have identified fish kairomones (pheromones that benefit the recipient) present in fish mucus that induce life history and behavioral changes in a range of zooplankters. It is likely that one or more of these kairomones, or similar compounds, are responsible for the behavioral and developmental changes observed in Pfiesteria spp. in the presence of live fish. Data from these experiments support the current understanding that significant behavioral differences exist between functional types of Pfiesteria spp., and between these toxic dinoflagellate species and known lookalike dinoflagellates without ichthyotoxic activity under ecologically relevant conditions.<P>