(ANDROPOGON GERARDII)</I>

Clemson University TigerPrints

All Dissertations Dissertations

5-2011 AN ECOLOGICAL, GENETIC, AND REPRODUCTIVE STUDY OF BIG BLUESTEM (ANDROPOGON GERARDII) POPULATIONS IN THE CAROLINAS Robert Tompkins Clemson University, [email protected]

Follow this and additional works at: https://tigerprints.clemson.edu/all_dissertations Part of the Plant Sciences Commons

Recommended Citation Tompkins, Robert, "AN ECOLOGICAL, GENETIC, AND REPRODUCTIVE STUDY OF BIG BLUESTEM (ANDROPOGON GERARDII) POPULATIONS IN THE CAROLINAS" (2011). All Dissertations. 724. https://tigerprints.clemson.edu/all_dissertations/724

This Dissertation is brought to you for free and open access by the Dissertations at TigerPrints. It has been accepted for inclusion in All Dissertations by an authorized administrator of TigerPrints. For more information, please contact [email protected].

AN ECOLOGICAL, GENETIC, AND REPRODUCTIVE STUDY OF BIG BLUESTEM (ANDROPOGON GERARDII) POPULATIONS IN THE CAROLINAS ______

A Dissertation Presented to The Graduate School of Clemson University ______

In Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy in Plant and Environmental Science ______

by Robert David Tompkins May 2011 ______

Accepted by Dr. William Stringer, Committee Chair Dr. Halina Knap Dr. Saara DeWalt Dr. Elena Mikhailova

ABSTRACT

AN ECOLOGICAL, GENETIC, AND REPRODUCTIVE STUDY OF BIG BLUESTEM (ANDROPOGON GERARDII) POPULATIONS IN THE CAROLINAS

Andropogon gerardii (Big Bluestem) is a dominant grass of the North American tallgrass prairie and is also found in remnant populations in the eastern U.S., including

North and South Carolina. This research included a systematic study of the ecology, genetics, and reproductive potential within and among A. gerardii populations in the

Carolinas. Floristic composition and edaphic (soil) features were characterized for eight

A. gerardii populations across various physiographic regions of North and South

Carolina. A total of 362 quadrats (1 m x 1 m) were sampled during the 2006-2008 growing seasons. A total of 306 vascular plant species was identified comprising 64 families, including 99 (32%) graminoids. Andropogon gerardii had the highest

Commonness Index (CI) value (5900), followed by Rubus spp. (1260). Community

Coefficient (CC) values were <0.5 for all pairings between sites, indicating high dissimilarity in species composition among sites. Soil pH values varied among the sites from 4.8 to 6.9. Calcium and Mg nutrient values were also highly variable. An out- crossing reciprocity study was conducted with five A. gerardii populations. Fifteen treatments were established at four garden sites. Seed germination was low for both out- crossed (4.8%) and selfed (2.4%) pairings. However, germination was significantly higher for both out-crossed and selfed seeds from ramets from the Suther Prairie population. There were no significant differences in seed set among plants from the five populations. Protein electrophoresis was used to assess genetic variation within and among nine local A. gerardii populations from various physiographic regions of both

North and South Carolina. Twenty-seven polymorphic loci and one monomorphic locus were resolved. Populations contained high levels of genotypic diversity. Genetic diversity was high at both the pooled species-wide level and at the mean within-population level.

For the species-wide sample, percent polymorphic loci (Ps) was 96.7%, the number of alleles per polymorphic locus (APsp) was = 4.07, the mean number of alleles per locus

(As) was = 3.97, and the genetic diversity (Hes) was 0.422. For within- populations, the mean Pp = 81.2%, APp, = 2.63, Ap = 2.38, and Hep = 0.340. Levels of within population

(Hep) genetic diversity ranged from 0.181 for the Sumter National Forest I population to

0.462 for the Bowman Barrier population. Genetic structure among populations (Gst) was

0.175, indicating that 83% of the genetic variation is found within-populations. The level of genetic diversity for the three larger populations (He = 0.387) and for the six smaller

populations (He = 0.330) was not significantly different (p = 0.5535). Nei‟s unbiased genetic identity between pairs of populations ranged from 0.6386 (populations

BlackJacks Heritage Preserve and Sumter National Forest I) to 0.9692 (populations Buck

Creek Serpentine Barrens and Suther Prairie). Of the population sites, the Blackjacks

Heritage Preserve population had the lowest measure of genetic identity with the other population sites, while the Suther Prairie population had the highest. The Mantel test showed no significant genetic isolation by geographic distance (r = 0.065; p =0.614).

Allelic richness values were high for the populations. Banding patterns were consistent with disomic inheritance. However for two loci (PGI3; UGPP1), tetrasomic patterns were indicated.

iii

ACKNOWLEDGEMENTS

There are many individuals and agencies to thank for their contributions to making this research project possible. First of all I would like to acknowledge the following landowners and agencies for allowing me access to do research on their properties; the

Suther family, Rick Harris, George Bowman, the South Carolina Department of Natural

Resources, and the U.S. Forest Service. Grant support for the research was made possible by the U.S. Forest Service and Dr. Elena Mikhailova. I will always be indebted as well to individuals who provided technical support for different aspects of the research; Dr.

William Bridges, Jr. for help with the statistical analysis, Dr. Patrick McMillan for help with the identification of difficult taxa, Dr. Jim Hamrick and Dr. Dorset Trapnell of the

University of Georgia for help with both the analysis and comments on the isozyme study; and Keith Richardson for help in the field and traveling with me to numerous sites.

In addition, I want to thank the members of my committee, Dr. Saara DeWalt, Dr.

Halina Knap, and Dr. Elena Mikhailova for their support throughout this project. I would like to especially thank my advisor, Dr. William Stringer, who helped me throughout this project with many different aspects of the research and analysis, for which I will be forever grateful. Lastly, I would like to thank my family for their patience and sacrifice over the five plus years of my program in completing this study.

TABLE OF CONTENTS Page

TITLE PAGE ...... i

ABSTRACT ...... ii

ACKNOWLEDGEMENTS ...... iv

TABLE OF CONTENTS ...... v

LIST OF TABLES ...... vii

LIST OF FIGURES ...... ix

INTRODUCTION ...... 1

CHAPTER 1 - BIG BLUESTEM (ANDROPOGON GERARDII VITMAN) COMMUNITIES IN THE CAROLINAS: COMPOSITION AND ECOLOGICAL FACTORS ...... 7 Abstract ...... 8 Introduction ...... 9 Study Sites ...... 10 Materials and Methods ...... 15 Results ...... 16 Discussion ...... 25 Literature Cited ...... 30 Appendix ...... 35

CHAPTER 2 - AN OUT-CROSSING RECIPROCITY STUDY BETWEEN REMNANT BIG BLUESTEM (ANDROPOGON GERARDII VITMAN) POPULATIONS IN THE CAROLINAS ...... 53 Abstract ...... 54 Introduction ...... 55 Materials and Methods ...... 57 Results ...... 62 Discussion ...... 70 Literature Cited ...... 77

CHAPTER 3 - GENETIC VARIATION WITHIN AND AMONG BIG BLUESTEM (ANDROPOGON GERARDII VITMAN) POPULATIONS IN THE CAROLINAS ...... 82 Abstract ...... 83 Introduction ...... 84 Materials and Methods ...... 86

Results ...... 91 Discussion ...... 98 Literature Cited ...... 103 Appendix ...... 109

CHAPTER 4 – SUTHER PRAIRIE: VASCULAR FLORA, SPECIES RICHNESS, AND EDAPHIC FACTORS – A UNIQUE EASTERN PRAIRIE ...... 117 Abstract ...... 118 Introduction ...... 119 Materials and Methods ...... 122 Results ...... 126 Discussion ...... 132 Literature Cited ...... 136 Appendix I ...... 139 Appendix II ...... 152

CHAPTER 5 - A NEWLY DOCUMENTED AND SIGNIFICANT PIEDMONT PRAIRIE SITE WITH A HELIANTHUS SCHWEINITZII TORREY & A. GRAY (SCHWEINITZ‟S SUNFLOWER) POPULATION ...... 155 Abstract ...... 156 Introduction ...... 157 Materials and Methods ...... 159 Results ...... 162 Discussion ...... 168 Literature Cited ...... 171 Appendix ...... 175

CONCLUSIONS...... 185

LIST OF TABLES

Table Page

1.1. Site features along with sampling dates and quadrat numbers sampled for Andropogon gerardii study sites...... 14

1.2. Associate species site and quadrat frequency with CI values of 150 or more. . 18 1.3. Sorenson‟s Community Coefficient values for pairings between sites...... 19 1.4. Rare and watch listed species and their status occurring in A. gerardii population sites in the Carolinas...... 20

1.5. Soil characteristics for Andropogon gerardii study sites...... 22 1.6. Soil parameter means for all eight study sites at 0-10-cm, 11-20-cm, and 21-30- cm depths...... 23

1.7. Soil parameter means from soil samples combined for all three depths from within and outside A. gerardii population sites...... 24

2.1. Treatments pairings of ramets from five Andropogon gerardii source populations at each experimental site...... 59

2.2.A. ANOVA results summarizing the experimental treatment effects for seed germination for source populations with replicate block (treatment) by year. . 63

2.2.B. Mean percent seed germination for A. gerardii by replicate by source population for years 2007 and 2008...... 64

2.3.A. ANOVA for percent germination of out-crossed treatments...... 66

2.3.B. Maternal and paternal contributions for out-crossed source populations for percent seed germination (years combined)...... 67

2.4.A. ANOVA for percent seed set for out-crossed treatments...... 68

2.4.B. Maternal and paternal contributions for out-crossed source populations for percent seed set (years combined)...... 68

2.5. Means for seed germination and seed set of selfed treatments for source populations (combined years)...... 69

3.1. Site descriptions from sampled populations of A. gerardii...... 88

3.2. Genetic variation for the Andropogon gerardii populations...... 93 vii

Table Page

3.3. Statistics of genetic diversity for loci from the sampled A. gerardii populations...... 94

3.4. Mean levels of genetic diversity (He), total number of alleles per locus (A), and number of alleles per polymorphic locus (APP) for small vs. large populations of A. gerardii...... 95

3.5. Nei‟s Genetic Identity values for sampled Andropogon gerardii populations...... 95

3.6. Nei‟s mean genetic identity value of each A. gerardii population...... 96

3.7. Summary of number of alleles and allelic richness per population...... 97

4.1. Species richness of vascular plants from Suther Prairie, Cabarrus County, North Carolina in relation to life form and wetland status...... 127

4.2. Species at Suther Prairie with highest frequency values in sampled quadrats during the 2006-07 growing seasons...... 130

4.3. Soil chemical properties of Suther Prairie...... 131

4.4. Climate data for Concord, North Carolina 2006-07...... 133

5.1. Taxonomic categories for vascular plant species for Troy Prairie...... 164

5.2. Species richness of vascular plants for Troy Prairie in relation to life form, habitat affinity, and wetland status...... 165

5.3. Frequencies for associate species for Troy Prairie from sampled quadrats (n=75) with the highest values...... 166

5.4. Comparison of upland vs. hydric soil chemical nutrient means at three sampled depths at Troy Prairie...... 167

viii

LIST OF FIGURES

Figure Page

1.1. Andropogon gerardii study sites in both North and South Carolina...... 13

2.1. Source population sites of Andropogon gerardii used in this study...... 58

2.2. X-ray images of Andropogon gerardii seeds showing filled vs. non-filled (empty) seeds...... 61

3.1. Map of sampled Andropogon gerardii populations for North and South Carolina...... 87

3.2. UPGMA phenogram displaying levels of genetic identity between populations of Andropogon gerardii...... 97

4.1. Location of Suther Prairie, Cabarrus County, North Carolina...... 123

4.2. View of Suther Prairie in the spring growing season...... 124

5.1. Location of Troy Prairie, Montgomery County, North Carolina ...... 160

5.2.A. Troy Prairie hydric sections...... 161

5.2.B. Troy Prairie Soil Map Units...... 161

INTRODUCTION

Andropogon gerardii Vitman (Big Bluestem) is a tall, native, C4,, warm-season perennial grass. Its range includes all states east of the Rocky Mountains and it is a major component of the North American tallgrass prairie (Weaver and Fitzpatrick 1934).

Although common in the Midwestern U.S., A. gerardii is relatively rare in much of the eastern U.S. (Radford et al. 1968). However, it is thought to have been more abundant throughout the Carolinas, particularly in the pre-settlement period (Barden 1997; Davis et al. 2002). At present, A. gerardii populations are scattered in the Piedmont, coastal plain and mountain regions of North Carolina, as well as in southern and coastal South

Carolina (Radford et al. 1968). Additionally, isolated populations have also been reported in upstate South Carolina (Chappell 2003). Where it does occur, it is often found in small, fragmented populations that most likely consist of clonal units (Chappell 2003). In a study of population structure of A. gerardii at Konza Prairie in Manhattan, Kansas,

Keeler (2002) found that the diameter of established clones of A. gerardii was relatively small with a mean area of 3.2m2.

Local A. gerardii populations have been described as being part of, but not restricted to, what are known as Piedmont prairie communities (Davis et al. 2002; Schafale and

Weakley 1990). Other prairie grass species often exist in association in these communities, including Sorghastrum nutans, Tridens flavus, Andropogon virginicus var. virginicus, Tripsacum dactyloides var. dactyloides, Schizachyrium scoparium and various

Panicum/Dichanthelium spp. (Davis et al. 2002). These sites also often contain rare endemic forbs such as Helianthus schweintzii, Lotus helleri, and Symphyotrichum georgianum (North Carolina Natural Heritage Program 2006; Schafale and Weakley

1990). The original range of A. gerardii and other native prairie-like grasses in the

Piedmont have probably been much reduced. This reduction has been primarily attributed to modern land use practices, and the fact that periodic fires are no longer common occurrences in the landscape (Barden 1997). Andropogon gerardii populations appear to be less common for the Carolinas than any of the other major prairie-relict grass species

(Radford et al. 1968). Beginning with European settlement, cultivation practices and fire suppression have led to the nearly complete disappearance of Piedmont prairies (Barden

1997).

Unlike A. gerardii of the Midwestern prairies, to date only generalized descriptions of

A. gerardii populations exist for the eastern United States (Schafale and Weakley 1990).

Previous ecological studies have been conducted at three sites in the Carolinas including

Suther Prairie, Cabarrus County, NC (J. Matthews unpublished data 2006); Buck Creek

Serpentine Barren, in Clay County, NC (Mansberg 1981; Marx 2007); and at the

Blackjacks Heritage Preserve, in York County, SC (Schmidt and Barnwell 2002).

Andropogon gerardii is an obligate out-crossing species (McKone et al. 1998;

Norrmann et al. 1997). It has been reported to be self-incompatible in Midwestern populations, and to possess a pre-zygotic incompatibility mechanism that results in the failure of the pollen tube to penetrate and grow into the style (McKone et al. 1998). This results in low reported seed set (0.2-6.0%) following self-pollination. In addition, post- zygotic mechanisms are suspected as well due to the low survivability of the offspring beyond the first growing season (Norrmann et al. 1997). Chappell (2003) reported reduced caryopsis production within four small A. gerardii populations in South

Carolina. Even in larger Midwestern populations of A. gerardii where genetically

compatible mates are more likely to occur, individuals have been reported to have highly variable seed set and seed mass (Norrmann et al. 1997; Law and Anderson 1940).

With the exception of Chappell (2003), which focused on the genetic diversity within and among two small local A. gerardii populations, and seed production within and among four local A. gerardii populations, no systematic study focusing on the ecology, genetics, and reproduction for A. gerardii populations in the eastern United States had been undertaken. Only general descriptions of local A. gerardii sites were known. Both the floristic composition and edaphic (soil) factors of A. gerardii populations in the

Carolinas had largely been unexplored. This study provides descriptive characterization of the floristic composition and edaphic (soil) features of eight A. gerardii populations from various physiographic regions of the Carolinas.

Two of the populations, Suther and Troy Prairies, represent among the best examples of what have been described as Piedmont Prairie community types, and until now their floristic composition and soil feature have not had complete descriptions. This project included extensive descriptions of each site‟s unique floral and edaphic features.

Although genetic studies are reported for Midwestern A. gerardii populations, with the exception of Chappell (2003), the genetics of A. gerardii populations in the Carolinas have been unexamined. In order to better assess the genetic structure of populations across the Carolinas, this project included an electrophoretic isozyme analysis of nine local A. gerardii populations.

Finally, in order to better understand the reproductive potential within and among local

A. gerardii populations, a two-year garden study was used to determine selfing and out- crossing potential between transplanted ramets from five local A. gerardii populations. It

is hoped that information gained from the out-crossing study will help in the development of seed accession lines for future restoration and establishment of local A. gerardii populations.

LITERATURE CITED

Barden, L. S. 1997. Historic prairies in the Piedmont of North and South Carolina, USA.

Natural Areas Journal 17:149-152.

Chappell, J. 2003. Lack of caryopsis production in remnant big bluestem (Andropogon

gerardii) populations located in upstate South Carolina. M.S. Thesis. Clemson

University, Clemson, South Carolina.

Davis, J. E. Jr., C. McRae, B. L. Estep, L. S. Barden, and J. F. Matthews. 2002. Vascular

flora of Piedmont prairies: Evidence from several prairie remnants. Castanea 67:1-

12.

Keeler, K. H., C. F. Williams, and L. S. W. Vescio. 2002. Clone size of Andropogon

gerardii Vitman (Big Bluestem at Konza Prairie, Kansas. American Midland

Naturalist 147:295-304.

Law, A. G. and K. L. Anderson. 1940. The effect of selection and inbreeding on the

growth of Big Bluestem (Andropogon Furcatus, Muhl.). Journal of the American

Society of Agronomists 32:931-943.

Mansberg, L. and T. R. Wentworth. 1984. Vegetation and soils of a serpentine barren in

western North Carolina. Bulletin of the Torrey Botanical Club 111:273-286.

Marx, E. 2007. Vegetation dynamics of the Buck Creek Serpentine Barrens, Clay

County, North Carolina. Honor‟s Thesis. University of North Carolina, Chapel

Hill, North Carolina.

McKone, M. J., C. P. Lund and J. M. O‟Brien. 1998. Reproductive biology of two

dominant prairie grasses (Andropogon gerardii and Sorghastrum nutans, Poaceae):

Male-biased sex allocation in wind-pollinated plants? American Journal of Botany

85:776-783.

Norrmann, G. A., C. L. Quarin, and K. H. Keeler. 1997. Evolutionary implications of

meiotic chromosome behavior, reproductive biology, and hybridization in 6x and

9x cytotypes of Andropogon gerardii (Poaceae). American Journal of Botany

84:201-207.

Radford, A. E., H. E. Ahles, and C. R. Bell. 1968. Manual of the Vascular Flora of the

Carolinas. University of North Carolina Press. Chapel Hill, North Carolina.

Schafale, M. P. and A. S. Weakley. 1990. Classification of the Natural Communities of

North Carolina. Third Approximation. North Carolina Natural Heritage Program.

Division of Parks and Recreation. North Carolina Department of Environment,

Health, and Natural Resources, Raleigh, North Carolina.

Schmidt, J. M. and J. A. Barnwell. 2002. A flora of the Rock Hill Heritage Preserve,

York County, South Carolina. Castanea 67:247-279.

Weaver, J. E. and T. J. Fitzpatrick. 1934. The prairie. Ecological Monographs 4:109-29.

Chapter 1

Big Bluestem (Andropogon gerardii; Poaceae) Communities in the

Carolinas: Composition and Ecological Factors

ABSTRACT

Andropogon gerardii (Big Bluestem) is a dominant grass of the North American tallgrass prairie. It is also found in remnant populations in the eastern U.S., including

North and South Carolina, often in association with other species with prairie affinities.

Eight A. gerardii population sites were sampled across various physiographic regions of

North and South Carolina. A total of 362 quadrats (1 m x 1 m) were sampled during the

2006-2008 growing seasons for species occurrence and site and quadrat frequency.

Associated species were assigned a commonness index (CI). A Sorenson‟s Community

Coefficient (CC) was used to determine floristic similarities among the sites. In addition, soil samples in three quadrats were sampled at each site at three depths (0-10 cm, 11-20 cm, and 21-30 cm) and analyzed for pH; organic C, and N contents; extractable P, K, Ca,

Mg, Zn, Mn; and CEC (cation exchange capacity). A total of 306 vascular plant species was identified comprising 64 families, including 99 (32%) graminoids. There were 61

(20%) Poaceae and 63 (20%) Asteraceae. Species per quadrat ranged from 1 to 13 with a mean of 5. Andropogon gerardii had the highest CI value (5900), followed by Rubus spp.

(1260). CC values were <0.5 for all pairings between sites, indicating high divergence in species composition among even nearby sites. There were 14 rare or watch-listed species identified, including the federally endangered Helianthus schweinitzii at Troy Prairie. A total of 153 (50%) of the species had been previously described as occurring in prairie- like associations. Soil pH values varied among the sites from 4.8 to 6.9. Calcium and Mg nutrient values were also highly variable.

INTRODUCTION

Andropogon gerardii (Big Bluestem) is a tall, native, warm-season perennial grass. Its range includes all states east of the Rocky Mountains and it is a major component of the

North American tallgrass prairie (Weaver and Fitzpatrick 1934). Although never as expansive as Midwestern prairies, eastern prairies are thought to have formed pockets interspersed throughout the pre-settlement landscape (Barden 1997; Davis et al. 2002).

Suppression of fire and the conversion of land for agricultural use probably played important roles in the decline of these communities, which are now predominantly relegated to roadside margins and utility rights-of-way (Davis et al. 2002).

The pre-settlement range and distribution of A. gerardii in the Carolinas is largely unknown. Radford et al. (1968) reported populations scattered throughout the Piedmont, coastal plain and mountain regions of North Carolina, as well as in southern and coastal

South Carolina. Additionally, isolated populations have also been reported in upstate

South Carolina (Chappell 2003). Local A. gerardii populations have been described as being part of, but not restricted to, what are known as Piedmont prairie communities

(Davis et al. 2002; Schafale and Weakley 1990). Other prairie grass species often exist in association in these communities, including Sorghastrum nutans, Tridens flavus,

Andropogon virginicus var. virginicus, Tripsacum dactyloides var. dactyloides,

Schizachyrium scoparium and various Panicum/Dichanthelium spp. (Davis et al. 2002).

These sites also often contain rare endemic forbs such as Helianthus schweintzii, Lotus helleri, and Symphyotrichum georgianum (North Carolina Natural Heritage Program

2006; Schafale and Weakley 1990).

Unlike A. gerardii of the Midwestern prairies, to date only generalized descriptions of these populations exist for the eastern United States (Schafale and Weakley 1990).

Previous studies have been conducted at three sites in the Carolinas: (1) Suther Prairie,

Cabarrus County, NC (J. Matthews unpublished data 2006); (2) Buck Creek Serpentine

Barren, Clay County, NC (Mansberg 1981; Marx 2007); and (3) Blackjacks Heritage

Preserve, York County, SC (Schmidt and Barnwell 2002). However, none of these had the ecology of A. gerardii as a primary focus. This study was conducted to provide a better understanding of the ecological factors associated with these plus five additional populations of A. gerardii in North and South Carolina. The objectives of this study were to document and compare floristic composition and edaphic characteristics for the eight sites. These data will provide a baseline for future study and can be used in ongoing restoration and maintenance of these and other A. gerardii sites.

STUDY SITES

A total of eight A. gerardii population sites were selected in North and South Carolina

(Figure 1.1; Table 1.1). These included the three largest populations (>0.5 ha; Suther

Prairie, Cabarrus County, NC; Buck Creek Serpentine Barren, Clay County, NC; Troy

Prairie, Montgomery County, NC) plus five additional sites from various physiographic regions of North and South Carolina. These sites were among the few known population sites for A. gerardii in both North and South Carolina with >100 culms. The following is a brief description of each site.

Suther Prairie (SP). This site is located on private property near SR 2408 (35.45˚ N,

80.46˚ W) about 8 km east of Concord in Cabarrus County, North Carolina. It lies along the floodplain of Dutch Buffalo Creek and a portion of the site remains wet for much of

the year. This site is the best known example of a remnant Piedmont prairie. It is unique in not having been regulary tilled and for its range of mesic to hydric soil conditions (L.

Barden pers. comm. 2006).

Troy Prairie (TP). This is a recently documented A. gerardii population site located along a utility right-of-way in Montgomery County, North Carolina (35.35˚ N, 79.87˚ W) approximately 3 km southeast of Troy. The northern portion of the site is located along a slope that grades to a hydric area fed by a nearby seepage spring. A second hydric area, also fed by a nearby spring, is located on the southern end of the property. As a result, hydric conditions are present over the growing season within small areas of the site.

Rock Hill Blackjacks Heritage Preserve (BJ). This site is located within the 117.3 ha

Rock Hill Blackjacks Heritage Preserve in York County, South Carolina (34.90˚ N,

81.01˚ W) just outside the city limits of Rock Hill. The underlying parent material consists of a gabbro pluton containing plagioclase feldspar with abundant outcrops and residual boulders. Large areas of the site contain Iredell series soils, which have high concentrations of Ca, Mg, and Fe (Schmidt and Barnwell 2002) and also contain montmorillonitic clay. Iredell soils often support unique floral communities (Batson

1952; Schmidt and Barnwell 2002). Schmidt and Barnwell (2002) documented 172 prairie-like species at the site, including many rare to uncommon species such as

Helianthus schweinitzii, Silphium terebinthinaceum, and Rosa carolina.

Buck Creek Serpentine Barren (BC). This site is located on a west-facing Appalachian mountain slope near NC Hwy. 64 in Clay County, North Carolina (35.08˚ N, 83.61˚ W) within Nantahala National Forest. Its geology is unique, comprising a peridotite outcrop that is approximately 90% olivine and serpentinized dunite (Mansberg and Wentworth

1984). According to Kauffman et al. (2004) this site is the most floristically distinctive serpentine area in the southern Appalachians. It has remained undisturbed since 1932 when it was acquired by the U.S. Forest Service, after being mined from 1875 to 1906 for corundum (Mansberg 1981; Mansberg and Wentworth 1984).

Orton Site (OS). This site is located in Brunswick County, North Carolina (34.07˚ N,

77.95˚ W) along the right shoulder of SR 133 N, approximately 3 km from Orton. It is restricted to the road backslope and is the easternmost site in the study. In addition, it is the only one located in the coastal plain region.

Bowman Barrier Rd. (BB). This site is located in Cabarrus County, North Carolina

(35.38˚ N, 80.42˚ W) near SR 2610 about 3 km southeast of Mt. Pleasant. It occurs within a floodplain along Dutch Buffalo Creek approximately 15 km down stream from the Suther Prairie site.

Sumter National Forest Site I (SNFI). This site is located on the shoulder of Battle

Creek Rd. in Oconee County, South Carolina in Sumter National Forest (34.76˚ N, 83.27˚

W) about 3.25 km west of Long Creek community. This is the smallest site in the study.

Sumter National Forest Site II (SNFII). Also located on Sumter National Forest land

(34.75˚ N, 83.27˚ W) along Damascus Church Rd., this site is approximately 3 km from the previous site located on the edge along the road.

Figure 1.1. Andropogon gerardii study sites in both North and South Carolina.

1 = Suther Prairie, 2 = Bowman Barrier Rd., 3 = Blackjack Heritage Preserve, 4 = Buck Creek Serpentine Barren, 5 = Troy Prairie, 6 = Orton Site, 7 = Sumter National Forest I, and 8 = Sumter National Forest II.

Table 1.1 Site features along with sampling dates and quadrat numbers sampled for

Andropogon gerardii study sites.

Site Area (ha) Hydric Conditions Land Management Dates Sampled # of Quadrats

SP 2.8 hydric to upland Mowing; Controlled burn 5/07; 8/07 90

TP 5.0 hydric to dry Clear cut 6/07; 9/07 75

BJ 9.0 x 10-3 dry Clear cut 5/07; 9/07 75

BC 3.0 dry Controlled burn 9/07; 6/08 37

OS 4.0 x 10-3 upland Mowing 6/07; 8/08 37

BB 0.5 floodplain Agriculture 9/06; 6/08 24

SNFI 1.2 x 10 -3 upland Controlled burn 7/07; 5/08 12

SNFII 2.4 x 10 -3 upland Controlled burn 7/07; 5/08 12

MATERIALS AND METHODS

Sites were visited during the 2006-2008 growing seasons to identify vascular species and take soil samples. Each site was sampled two times during the study period, and included early- and late-season sampling to ensure a more complete record of species presence at each site. The number of quadrats (1 m x 1 m) sampled for botanical data was determined by the relative size of each location. The sampling domain was determined by either the physical layout of the site (dimensions and/or physical surroundings), or by using the releve′ method (Mueller-Dombois and Ellenberg 1974) to select the area that contained the largest number of species at the site so that the floristic composition would be sufficiently represented in the sampling. In the three larger sites, quadrats were located every 10 m, randomly to the right or left of transects located across the length and width of the entire site. In the five smaller sites, quadrats were located contiguously along one or more transects within the areas of greatest density of A. gerardii. Additional transects were established and quadrats sampled adjacent to the initial transects, so that the sampling included a large representation of the species within each community. All vascular species at each site, both within and outside of the quadrats, were identified.

Voucher specimens were deposited in the herbaria of Clemson University and Belmont

Abbey College. Nomenclature follows Weakley (2007).

Frequency values were calculated for each species for both the number of sites in which it occurred, and for occurrence in quadrats among the sites. A commonness index (CI) was calculated by multiplying site frequency times quadrat frequency (% density). Site frequency was the percentage of sites (out of 8 total) in which a species occurred. Percent density was the number of quadrats in which a species occurred, divided by the total

number of quadrats sampled (362), multiplied by 100 (Anderson 1954 and Henderson

1995). Sorenson‟s Community Coefficient (CC) method was used to determine the floristic similarities among the sites based on the presence of all species found at each site, both within and outside of the sampled quadrats. It was calculated using the formula

CC = 2C/(A + B), where A = the total number of species for site A; B = the number of species for site B; and C = number of species common to both (Barbour et al. 1999;

Thompson and Poindexter 2006).

Soil characteristics including soil series type, texture, parent material, drainage and depth were recorded for all sites. In each population soil samples were collected in three quadrats at random at three depths (0-10 cm, 11-20 cm, and 21-30 cm; n = 72). In addition, three samples were collected at random at three depths (0-10 cm, 11-20 cm, and

21-30 cm; n = 72) along the periphery of each population site where A. gerardii was absent for comparison. Soil samples were analyzed for soil pH; organic C and N contents; extractable P, K, Ca, Mg, Zn, Mg; and cation exchange capacity (CEC)

(Donohue 1992; Isaac 1983). An ANOVA was performed to compare the mean values for soil characteristics across the sites, and across depths among sites. Post hoc differences in means were determined using Fisher‟s LSD. All statistical analyses were performed with SAS v. 9.1. (SAS Institute, Inc., Cary, NC 2002).

RESULTS

A total of 362 quadrats were sampled (Table 1.1) with 306 species identified from 64 families (Appendix). Of those, 111 were monocots (56 genera in 13 families), while 190 were dicots (129 genera in 46 families). In addition, one gymnosperm, three ferns and one Lycopsid were also identified. Ninety-nine graminoid species (Poaceae, Cyperaceae,

and Juncaceae) were collected, accounting for 32% of the total species. Of those, there were 61 (20%) Poaceae, 33 (11%) Cyperaceae, and 5 (<1%) Juncaceae. The Asteraceae accounted for the largest number of species with 63 (20%). Species per quadrat ranged from 1-13 with a mean of 5 species per quadrat.

Commonness Index (CI) values for plant species from sampled quadrats ranged from

13 to 5900. Species with CI values >150 are listed in Table 1.2. Andropogon gerardii had the highest calculated CI value (5900) followed by Rubus spp. (CI =1260) and Lespedeza cuneata (CI =1197). The high CI value for L. cuneata validates the high degree of invasiveness ascribed to this introduced species (Weakley 2007). Six grasses in addition to A. gerardii had CI values ≥150, with Tripsacum dactyloides var. dactyloides having the second highest (CI =684). Along with A. gerardii, this grass is an important species of moist and wetland areas in Midwestern prairies (Henderson 1995; Weakley 2007).

This was consistent with it being found at the two floodplain sites (SP and BB) and at

Troy Prairie which also had hydric sections. Two other grasses with prairie affinities,

Dichanthelium dichotomum var. dichotomum and D. sphaerocarpon, also had high CI values (304 and 315, respectively). The more widespread D. scoparium had a CI value of

350, and the introduced Holcus lanatus had a CI value of 150.

Only 40 of the species (13%) identified occurred in three or more sites, while only 14

(4%) occurred in four or more. Sorenson‟s Community Coefficient (CC) values for all eight sites when paired were all <0.5 (Table 1.3). According to Barbour et al. (1999), communities with a CC value of >0.5 are considered to be the same community type; thus, these values indicate a high degree of divergence among those sites in terms of species composition.

Table 1.2. Associate species site and quadrat frequency with CI values of 150 or more.

Taxon Frequency % Density % N CI

Andropogon gerardii Vitman 100 59 215 5900

Rubus spp. L. 63 20 73 1260

Lespedeza cuneata (Dumont-Cours.) G. Don 63 19 68 1197

Tripsacum dactyloides (L.) L. var. dactyloides 38 18 66 684

Campsis radicans (L.) Seeman ex Bureau 38 17 60 646

Pteridium aquilinum (L.) Kuhn var. pseudocaudatum 50 12 43 600

Diospyros virginiana L. 50 11 39 550

Symphyotrichum dumosum L. 63 7 25 441

Fraxinus spp. L. 38 11 38 418

Rhus copallinum L. var. copallinum 38 10 38 380

Dichanthelium scoparium (L.) Gould 25 14 51 350

Dichanthelium sphaerocarpon (Elliot) Gould 63 5 18 315

Festuca eliator L. 38 8 30 304

Dichanthelium dichotomum (L.) Gould 38 8 30 304

Ageratina altissima King & H.E. Robinson 38 7 25 266

Cephalanthus occidentalis L. 25 10 35 250

Vitis rotundifolia Michx. 25 9 32 225

Silphium compositum Michx. var. compositum 38 5 17 190

Prunella vulgaris L. 38 4 16 152

Holcus lanatus L. 25 6 23 150

Allium canadense L. var. canadense 25 6 20 150

Frequency is the % of occurrence in all 8 study sites. Density is the % occurrence in 362 sampled quadrats for all 8 sites. N is the total number of quadrats in which each species occurred over the 8 study sites and was used in calculating the density value. CI is the “commonness index” (Anderson 1954) derived by multiplying site frequency times density ( Henderson 1995).

Table 1.3. Sorenson‟s Community Coefficient values for pairings between sites.

SP TP 0.34 BJ 0.29 0.24 BC 0.13 0.19 0.08 OS 0.04 0.07 0.16 0.04 BB 0.17 0.13 0.12 0.06 0.07 SNFI 0.17 0.13 0.13 0.05 0.07 0.04 SNFII 0.06 0.13 0.06 0.07 0.08 0.05 0.05 SP TP BJ BC OS BB SNFI SNFII

Four sites contained rare species including Troy Prairie with the federally endangered

Helianthus schweinitzii. Thirteen other species documented were rare or ranked as watch listed for either North or South Carolina (Table 1.4; North Carolina Natural Heritage

Program 2006; South Carolina Department of Natural Resources 2008), and 18 others were considered uncommon (Weakley 2007). Many species were newly documented for the three previously surveyed sites, including 75 for Suther Prairie (Tompkins et al.

2010), 4 for BlackJacks Heritage Preserve, and 13 for Buck Creek.

Soils among the study sites varied widely in series, texture, parent material and drainage. They ranged from alluvial to upland and were derived from mafic to felsic parent material (Table 1.5). Although reported to be a potential Mollisol (Mansberg and

Wentworth 1984), Buck Creek has recently been classified as an Inceptisol in the Great

Group Eutrochrept, which reflects its recent history of disturbance as a mine site (Soil

Survey of Clay County, North Carolina 1998). The Chastain soil series for Suther Prairie is rare for the Piedmont and the first documented site with this series for Cabarrus

Table 1.4. Rare and watch listed species and their status occurring in A. gerardii

population sites in the Carolinas.

Taxon Study Site State Listing State Rank U.S. Rank

Carex bushii Mackenzie SP NC SR S1 __

Carex tenera Dewey var. tenera SP NC W7 S1 __

Eleocharis tricostata Torrey SP NC W1 S2,S3 __

Helianthus schweinitzii Torrey & A. Gray TP NC E S3 E

Liatris squarrulosa Michx. BJ NC SR S2 __

Muhlenbergia glomerata (Willdenow) Trinius BC NC SR S1 __

Oligoneuron album (Nuttall) Nesom BJ SC SR S1 __

Scirpus pendulus Muhlenberg SP NC SR S1 __

Solidago gracillima Torrey & A. Gray TP NC W1 S3 __

Solidago puberula Nuttall var. puberula BC NC W7 S2 __

Solidago radula Nuttall TP NC SR S1 __

Symphyotrichum rhiannon Weakley & Govus BC NC SR S1 FSC

Thalictrum macrostylum Small & Heller BC NC SR S2 FSC

Tridens strictus (Nuttall) Nash BJ SC SR S1 __

NC = North Carolina, SC = South Carolina, U.S. = Federal. SR = state rare, W = watch listed. U.S. Rank: E = endangered, FSC = Federal Species of Concern. State Rank: S1 = critically imperiled, S2 = imperiled, S3 = vulnerable. Watch Category: W1 = rare but relatively scarce, W7 = rare and poorly known.

County, NC. It is a more common soil series of the eastern coastal plain region (Natural

Resources Conservation Service 2009).

Soil pH for the sites ranged from very strongly acidic (4.8) to neutral (6.9). Four sites had pH values >6.0 (SP, BJ, BC and OS) that were significantly higher than the other sites (Table 1.6). Three of these high-pH sites (SP, BJ, and OS) had very high Ca levels, while a fourth (BC) had very high Mg. A fifth site (BB), the only one with a recent history of agricultural management, had high Ca, but the pH was acidic (5.3-5.6).

Magnesium levels were high at Suther Prairie and BlackJacks Heritage Preserve, and very high at Buck Creek, which is consistent with its serpentine parent material

(Mansberg and Wentworth 1984). The remaining sites (TP, SNFI, and SNFII) had moderately to strongly acidic pH, as well as only moderate levels of Ca and Mg.

Soil pH was significantly higher in sites where A. gerardii was present compared to the adjacent areas without A. gerardii (Table 1.7). Nutrient availability was also significantly higher for magnesium and manganese where A. gerardii occurred. However, nutrient availability was significantly higher for copper in adjacent areas without A. gerardii

(Table 1.7).

Table 1.5. Soil characteristics for Andropogon gerardii study sites.

Site Soil Series Texture Parent Material Drainage Depth

SP Chastain (s) loam alluvium poor 180+cm

Enon (ws) sandy loam mafic well -----#

Iredell (ws) sandy loam ferromagnesium moderately well -----#

Mecklenberg (ws) loam mafic well -----#

TP Badin (s) silt loam ferromagnesium well 150+cm

Tarrus (s) silt loam ferromagnesium well 100+cm

Herndon (s) loam metavolcanic well 150+cm

BJ Iredell (s) loam ferromagnesium moderately well 150+cm

BC Eutrochrept^ sandy loam felsic to mafic well 190+cm

OS Baymeade (s) sandy loam alluvium well 200+cm

BB Chewacla (s) loam alluvium poor 200+cm

SNFI Madison (s) sandy loam felsic to igneous well 170+cm

SNFII Madison (s) sandy loam felsic to igneous well 170+cm

(s) = indicates the soil series occupying the site; (ws) = indicates soil series occupying the watershed upstream from a floodplain sample site; -----# = depth of soils off the sample site deemed irrelevant; ^ = non-soil series (Great Group).

Table 1.6. Soil parameter means for all eight study sites at 0-10-cm, 11-20-cm, and 21-30-cm

depths.

Site cm pH N% C% P* K* Ca* Mg* Zn* Mn* CEC SP 10 6.1B 0.26B,A 2.9B 7.8C,B 275.3A 2339B,A 819B 3.1C 42.1D,C 12.5B,A TP 10 4.8D 0.21B,A 3.5B,A 9.7B 107.6B,A 348C 59D 7.3B,A 82.9B,A 2.5C BJ 10 6.2B,A 0.18B,C 2.6B,C 6.7C,B 66.1B,A 2381B,A 912B 2.9C 73.2B,A 12.3B,A BC 10 6.7A 0.24B,A 3.5B,A 1.5C 89.6B,A 574B,C 2392A 2.9C 85.8A 13.6A OS 10 6.5B,A 0.05D 0.9D 26.2A 42.1B 2953A 165D,C 7.8B,A 10.4E 1.0C BB 10 5.3C 0.26A 2.6B,C 13.1B 84.7B,A 1555B,A, 412C 10.1A 55.6B,C 9.3B,C SNFI 10 5.2D,C 0.11D,C 1.6D,C 7.1C,B 132.6B,A 422C 87D 5.6B,C 21.9D,E 5.0D SNFII 10 5.1D,C 0.22B,A 4.2A 9.3C,B 144.0B,A 683B,C 120D,C 4.5B,C 35.8D,C,E 7.3C,D Std D 0.72 0.07 1.08 6.45 64.1 925 704.3 2.4 25.1 3.1 SP 20 6.1B 0.13B 1.2B 3.4C,B 104.3A 1813A 775.7B 1.3C 29.5C 10.4A TP 20 4.8D 0.08C 1.6B 5.6C,B 52.7B,A 142B 29.9D 1.9B,C 61.7B 4.3B BJ 20 6.3B,A 0.11B,C 1.4B 4.9C,B 42.9B,A 2016A 806.3B 1.3C 84.4A 10.5A BC 20 6.7A 0.19A 3.1A 1.1C 67.6B,A 433B 1992.7A 1.8C 84.4A 11.1A OS 20 6.6A 0.02D 0.6C 22.8A 26.1B 1414A 90.8D 4.9B,A 48.6B 4.2B BB 20 5.4C 0.15B,A 1.6B 12.0B 58.9A 1650A 382.3C 6.8A 48.6B 9.3A SNFI 20 4.9D,C 0.08C 1.3B 4.1C,B 35.5B,A 191B 29.5D 1.9B,C 8.6D 4.6B SNFII 20 5.1D,C 0.07C 1.3B 2.6C,B 33.7B,A 280B 65.6D 1.2C 9.3D 5.0B Std D 0.76 0.05 0.69 6.37 21.5 716 584.4 1.8 25.3 3.1 SP 30 6.1B,C 0.07C 0.6E 2.5B 52.3A 1678A 806.4B 0.9B 38.1C,B 10.1A TP 30 4.9E 0.04D,E 0.6E 3.3B 41.1A 130B 29.4D 1.6B 26.6C 3.1B BJ 30 6.4B,A 0.10B 1.3C,B 3.3B 37.3A 2084A 859.9B 0.9B 70.3A 11.1A BC 30 6.9A 0.16A 1.9A 1.4B 50.7A 361B 1917.4A 1.2B 78.1A 10.8A OS 30 6.7A 0.02E 0.5E 18.2A 23.8A 1446A 83.7D 4.2A 5.3D 4.6B BB 30 5.6D,C 0.15A 1.4B 13.4A 43.3A 1734A 384.2C 6.2 B 46.7B 9.1A SNFI 30 5.2D,E 0.07C 1.1C,D 4.0B 29.4A 235B 38.9D 1.4B 5.6D 4.8B SNFII 30 5.2D,E 0.05D,C 0.8E,D 1.1B 25.0A 272B 77.3C 0.7B 5.2D 5.3B Std D 0.75 0.04 0.4 5.6 9.3 727 586.5 1.7 26.0 7.3

Superscripts denote significant different means within soil depths compared among sites (LSD Test, P<0.05). *kg/ha; Std D = standard deviation among sites.

Table 1.7. Soil parameter means from soil samples combined for all three depths from

within and outside A. gerardii population sites. Mean values with different

superscripts denote significant differences (p <0.05).

Soil Parameter Within A. gerardii pops. Outside A. gerardii pops. pH 5.8A±0.7 5.6B±0.5 Organic Carbon (%) 1.8A±1.05 1.5A±0.7 Organic Nitrogen (%) 0.1A±0.07 0.1A±0.08 Phosphorus (kg/ha) 7.7A±6.8 26.0A±44.0 Potassium (kg/ha) 70.0A±54.9 53.8A ±27.3 Calcium (kg/ha) 1130.3A± 874.5 1195.5A ±1119.3 Magnesium (kg/ha) 552.4A± 673.5 412.3B±402.9 Copper (kg/ha) 3.3A±2.6 6.1B±7.4 Zinc (kg/ha) 3.4A±2.6 4.1A±4.3 Manganese (kg/ha) 40.0A±28.4 32.6B±24.3 CEC (meq/100g) 7.6A±3.1 7.4A±4.1

DISCUSSION

Although species richness was high at several of the A. gerardii population sites, there was much variation in the species composition found within each. This was indicated by the low CC values (<0.5) for all site comparisons (Table 1.3). With the exception of A. gerardii, only five other species appeared in at least five of the sites, including Lespedeza cuneata, Rubus spp., Symphyotrichum dumosum, Dichanthelium sphaerocarpon and

Pycnanthemum tenuifolium. The higher CC values between the pairings, SP–TP, SP–BJ, and TP–BJ were expected, as all three sites are located in the central Piedmont region of

North and South Carolina. Although Bowman Barrier is also located within the Piedmont region very close to Suther Prairie, it was less similar to the other three Piedmont sites, most likely due to its history of agricultural use. Low CC values were also found when the Piedmont locations were paired with Buck Creek (mountain region), Orton Site

(coastal plain), along with Sumter National Forest I and Sumter National Forest II

(foothills). It should be noted that the 2007 growing season, when a large proportion of the sampling took place, was among the driest on record for much of the eastern U.S.

(State Climate Office of North Carolina 2007). As a result, the number of species observed may have been reduced. When compared with the other sites, Suther Praire may have been the most adversely affected by the drought as it was previously reported as having a larger number of mesic to hydric species. Many of those were not observed during the 2007 growing season (Tompkins et al. 2010).

Even the highest CC values reflect rather weak species associations, and make it difficult to assign a strict overall community classification for the study sites. This, however, does not negate the fact that there was a high percentage of species with prairie

affinities among the sites. A total of 153 (50%) species had been previously described as occurring in prairie-like associations; many had also been reported in association with A. gerardii (Barone and Hill 2007; Davis et al. 2002; Henderson 1995; Edgin and Ebinger

2000; Klips 2003; Laughlin and Uhl 2003; Mack and Boerner 2004; Schmidt and

Barnwell 2002; Thompson and Poindexter 2006). The presence of many prairie- associated species was consistent with earlier descriptions of Piedmont prairie locations

(Davis et al. 2002; Schmidt and Barnwell 2002).

Thus, these sites are characterized as having unique floral assemblages made up of prairie-like associates. The composition of the floral assemblage present at a particular site is most likely the result of factors such as having a nearby seed source and their arrival time at that site. As a result, many different species may be present at these sites, yet many of them, provided there is a nearby seed source, may be plants with prairie-like affinities.

Brady et al. (2005) have suggested that unique floristic associations and high species diversity reflect plant adaptation to unique soil conditions (e.g., the presence of certain heavy metals in serpentine sites, such as Buck Creek). As a result, for some sites, floristic composition may be more predictable based on the presence of unique geological and soil features. For other sites, however, the relationship between specific soil chemistry and floristic endemism is unclear. For example, the Suther Prairie and Troy Prairie sites had the highest floristic similarity among the pairings (Table 1.3), despite having significantly different means (P<0.05) for Ca, Mg, pH, and CEC at all three sampled depths. Thus, this similarity was more likely due to the presence of upland to hydric conditions at both, which is uncommon among prairie-like sites in the Piedmont (Davis et al. 2002; Schafale

and Weakley 1990). In addition, neither of these sites has been tilled for many years. It is thought that Suther Prairie has never been tilled (Tompkins et al. 2010). Likewise, there was not a correlation between soil features present and species composition among the pairing between Sumter National Forest I and Sumter National Forest II, which produced a very low measure of similarity (CC=0.05), despite having the closest proximity to one another (3 km), and having very similar soil conditions. Soil chemistry and general soil features were almost identical for those two sites (Tables 1.5 and 1.6), which suggest that soil conditions alone were not the determinate cause for floristic variation. These differences in soil characteristics were consistent with earlier reports of high soil variability among sites with prairie-associated species (Schafale and Weakley 1990).

High levels of Mg in the soils of both BlackJacks Heritage Preserve and Buck Creek were consistent with their underlying parent material (Table 1.5). In the case of Suther

Praire, the high levels of both Mg and Ca were probably due to alluvial deposition material from nearby mafic-origin soils (Table 1.5). The high Ca values for Orton Site and consequent high pH values were attributable to its origin from marine sediments.

Both Suther Prairie and Bowman Barrier contained alluvial soils from within the same watershed. However, Suther Prairie was less acidic for all three sampled depths than

Bowman Barrier (Table 1.6). Again, the higher pH at Suther Prairie may be due to its closer proximity to soils with mafic properties.

Although in some cases, such as Buck Creek, floristic endemism and species richness were most likely due to the presence of unique soil conditions, the data suggest that soil features alone may not be predictive of any particular floristic composition, even among sites with similar physiography. With the possible exception of the presence of mesic-

hydric soil conditions, other factors such as seed rain, species establishment, and disturbance history may play more important roles in plant species composition at those sites.

Although A. gerardii is found in a limited number of communities in the Carolinas, these data suggest that it often attains high frequency and dominates the sites where it occurs. This is consistent with its dominant role in prairie communities of the Midwest

(Rosburg 1997; Silletti and Knapp 2001; Silletti et al. 2004). Davis et al. (2002) reported that the lack of periodic disturbance within otherwise appropriate sites may be the main limiting factor of A. gerardii establishment and maintenance in the Carolinas.

Although there were significant differences in pH and certain nutrient levels within and outside the A. gerardii populations, the apparent broad tolerance of A. gerardii to various soil conditions would seem to again suggest that its establishment and maintenance, like many of its prairie-like associates, is not solely a function of soil features or physiography. Its presence may be more related to having a nearby seed source and periodic disturbance events. In fact, sites with poorer soil conditions may allow less competitive plant species to have a competitive advantage over other species that are normally more common in habitats with greater nutrient availability. Low resource availability has been shown to be favorable to A. gerardii when compared to other grassland species such as Sorghastrum nutans (Silletti and Knapp 2001).

Although not observed in 2007, 5 individuals of the Federally-endangered species,

Helianthus schweinitzii, were noted during the 2008 growing season at Troy Prairie.

During an initial survey in 2003, more than 100 stems were reported (Santec Consulting,

Charlotte, NC 2003). It is thought that recent human disturbance after the initial survey

may have reduced the population size (M. Bates pers. comm. 2005). Of the sites in this study, Troy Prairie is potentially most threatened in the near future. A portion of the site may be impacted negatively by a state highway bypass project (G. Jordan, U.S. Fish and

Wildlife Service, pers. comm. 2008).

While this study provides a better understanding of the floristic composition and ecology of both A. gerardii population sites and Piedmont prairies in North and South

Carolina, additional study should be done to provide a more in-depth characterization of their ecology. The results presented here provide a baseline for future study, and may be useful in the long term preservation and restoration of those sites.

LITERATURE CITED

Anderson, O. 1954. The phytosociology of dry lime prairies of Wisconsin. Ph. D.

Dissertation, University Wisconsin-Madison, Wisconsin.

Barbour, M. G., J. H. Burk, W. D. Pitts, F. S. Gulliam, and M. W. Schwartz. 1999.

Terrestrial Plant Ecology, 3rd. ed. Benjamin/Cummings., Inc. Menlo Park,

California.

Barden, L. S. 1997. Historic prairies in the Piedmont of North and South Carolina, USA.

Natural Areas Journal 17:149-152.

Barone, J. A. and J. G. Hill. 2007. Herbaceous flora of Blackland prairie remnants in

Mississippi and Western Alabama. Castanea 72:226-234.

Batson, W. T. 1952. The floristics of the Iredell Soil Series in the central Piedmont of

North Carolina. Ph.D. Dissertation, Duke University, Durham, North Carolina.

Brady, K. U., A. R. Kruckeberg, and H. D. Bradshaw, Jr., 2005. Evolutionary ecology of

plant adaptation to serpentine soils. Annual Review of Ecology, Evolution, and

Systematics 36:243-266.

Chappell, J. 2003. Lack of caryopsis production in remnant big bluestem (Andropogon

gerardii) populations located in upstate South Carolina. M.S. Thesis. Clemson

University, Clemson, South Carolina.

Davis, J. E. Jr., C. McRae, B. L. Estep, L. S. Barden, and J. F. Matthews. 2002. Vascular

flora of Piedmont prairies: Evidence from several prairie remnants. Castanea 67:1-

12.

Donohue, S. J. 1992. Reference Soil and Media Diagnostic Procedures for the Southern

Region of the United States. Southern Cooperative Bulletin No. 374.

Edgin, B. R. and J. E. Ebinger. 2000. Vegetation of a successional prairie at Prairie Ridge

State Natural Area, Jasper County, Illinois. Castanea 65:139-146.

Henderson, R. A. 1995. Plant species composition of Wisconsin prairies. An aid to

selecting species for planting and restorations based upon University of Wisconsin.

Madison Plant Ecology Laboratory data. Technical Bulletin No. 188. Dept. of

Natural Resources, Madison, Wisconsin.

Isaac, R. A. 1983. Reference Soil Test Methods for the Southern Region of the United

States. Southern Cooperative Series Bulletin No. 289.

Kauffman, G. L., A. L. Weakley, G. L. Nesom, and T. E. Govus. 2004. A new species of

Symphyotrichum (Asteraceae:Astereae) from a serpentine barren in western North

Carolina. Sida 21:827-839.

Klips, R. A. 2003. Vegetation of Claridon Railroad Prairie, a remnant of the Sandusky

Plains of Central Ohio. Castanea 68:135-142.

Laughlin, D. C. and C. F. Uhl. 2003. The xeric limestone prairies of Pennsylvania.

Castanea 68:300-316.

Mack J. J. and R. E. J. Boerner. 2004. At the tip of the prairie peninsula: Vegetation of

Daughter Savannah, Crawford County, Ohio. Castanea 69:309-323.

Mansberg, L. 1981. Vegetation, soils, and canopy age-structure of the Buck Creek

Serpentine Barren, Clay County, North Carolina. M.S. Thesis, North Carolina State

University, Raleigh, North Carolina.

Mansberg, L. and T. R. Wentworth. 1984. Vegetation and soils of a serpentine barren in

western North Carolina. Bulletin of the Torrey Botanical Club 111:273-286.

Marx, E. 2007. Vegetation dynamics of the Buck Creek Serpentine Barrens, Clay

County, North Carolina. Honor‟s Thesis. University of North Carolina, Chapel

Hill, North Carolina.

Mueller-Dombois, D. and H. Ellenberg. 1974. Aims and Methods of Vegetation Ecology.

Wiley & Sons, New York.

Natural Resources Conservation Service. 2009. Soil Series Descriptions. USDA.

soils.usda.gov/technical/classification/osd/index.html

North Carolina Natural Heritage Program. 2006. Natural Heritage Program List of Rare

Plant Species of North Carolina. Franklin, M.A. and J.T. Finnegan, eds. North

Carolina Natural Heritage Program Office of Conservation and Community

Affairs. North Carolina Department of Environment and Natural Resources.

Raleigh, North Carolina.

Radford, A. E., H. E. Ahles, and C. R. Bell. 1968. Manual of the Vascular Flora of the

Carolinas. University of North Carolina Press. Chapel Hill, North Carolina.

Rosburg, T. R. 1997. Distribution and abundance of prairie plant species in the Loess

Hills. Journal of the Iowa Academy of Science 104:8-20.

Santec Consulting 2003. Ayrsley Town Blvd. Suite 300. Charlotte, North Carolina

28273.

SAS Institute. 2002. SAS User‟s Guide: Statistics. Version 9.1. SAS Institute,

Cary, North Carolina.

Schafale, M. P. and A. S. Weakley. 1990. Classification of the Natural Communities of

North Carolina. Third Approximation. North Carolina Natural Heritage Program.

Division of Parks and Recreation. North Carolina Department of Environment,

Health, and Natural Resources, Raleigh, North Carolina.

Schmidt, J. M. and J. A. Barnwell. 2002. A flora of the Rock Hill Heritage Preserve,

York County, South Carolina. Castanea 67:247-279.

Silletti, A. M. and A. K. Knapp. 2001. Responses of the codominant grassland species

Andropogon gerardii and Sorghastrum nutans to long-term manipulations of

nitrogen and water. American Midland Naturalist 145:159-167.

Silletti, A. M., A. K. Knapp, and J. M. Blair. 2004. Competition and coexistence in

grassland codominants: Responses to neighbor removal and resource availability.

Canadian Journal of Botany 82:450-460.

Soil Survey of Clay County, North Carolina. 1998. United States Department of

Agriculture. Hayesville, North Carolina.

South Carolina Department of Natural Resources. 2008. Rare species list. Columbia,

South Carolina.

State Climate Office of North Carolina. 2007. http://www.NC-climate.ncsu.edu/

Thompson, R. L. and D. B. Poindexter. 2006. Vascular flora of the Elk and Bison Prairie,

Land Between the Lakes National Recreation Area, Trigg County, Kentucky.

Castanea 71:105-123.

Tompkins, R. T., C. M. Luckenbaugh, W. C. Stringer, K. H. Richardson, E. A.

Mikhailova, and W. C. Bridges, Jr. 2010. Suther Prairie: Vascular flora, species

richness, and edaphic factors. Castanea 75:232-244.

Weakley, A. S. 2007. Flora of the Carolinas, Virginia, Georgia, and Surrounding Areas,

working draft of January 2007. University of North Carolina, Chapel Hill, North

Carolina.

Weaver, J. E. and T. J. Fitzpatrick. 1934. The prairie. Ecological Monographs 4:109-29.

APPENDIX

Species collected at eight Andropogon gerardii study sites in North and South Carolina.

Abbreviations are as follows: SP = Suther Prairie; TP = Troy Prairie; BJ = Blackjacks

Nature Preserve; BC = Buck Creek Serpentine Barren; OS = Orton Site; BB = Bowman

Barrier Rd.; SNFI = Sumter National Forest Site I; SNFII = Sumter National Forest

Site II; P = Prairie Associate Species; NCR = NC Rare; SCR = SC Rare; NCWL = NC

Watch Listed; NCE = NC Endangered; USE = US Endangered and UC = Uncommon.

Collection numbers are those of R. Tompkins (RT). Voucher specimens are housed

in the Clemson University Herbarium (CLEMS) and Belmont Abbey College.

PTERIDOPHYTA

DENNSTAEDTIACEAE

Pteridium aquilinum (L.) Kuhn var. pseudocaudatum (TP – RT0167, BC - RT0621, OS – RT0731, SNFII - RT0712)

LYCOPODIACEAE

Lycopodiella alopecuroides (L.) Cranfill (TP - RT0573)

OSMUNDACEAE

Osmunda cinnamomea L. var. cinnamomea (TP - RT0638)

Osmunda regalis L. var. spectabilis (Willdenow) A. Gray (TP - RT0763)

GYMNOSPERMAE

PINACEAEAE

Pinus rigida P. Miller (BC - RT0613)

ANGIOSPERMAE

ACANTHACEAE

Ruellia caroliniensis (J.F. Gmelin) Steudel (SP - RT0586, BJ - RT0385) [P]

ALISMATACEAE

Sagittaria latifolia Willdenow var. latifolia (TP - RT0352)

ALLIACEAE

Allium canadense L. var. canadense (SP - RT0073, BC - RT0618) [P]

ALTINGIACEAE

Liquidambar styraciflua L. (SP - RT0733, TP - RT0662)

AMARYLLIDACEAE

Zephyranthes atamasca (L.) Herbert (SP - RT0022)

ANACARDIACEAE

Rhus copallinum L. var. copallinum (TP - RT0669, BJ - RT0680, OS - RT0729) [P]

Rhus glabra L. (SNFII - RT0711) [P]

Toxicodendron radicans (L.) Kuntze var. radicans (SP - RT0734, TP - RT0667, BJ – RT0688) [P]

APIACEAE

Angelica venenosa (Greenway) Fernald (TP- RT0754, BB - RT0239, SNFII - RT0250)

Cicuta maculata L. var. maculata (SP - RT0232) [P]

Daucus carota L. (SP - RT0231)

Eryngium yuccifolium Michx. var. yuccifolium (TP - RT0517) [P, UC]

Ptilimnium capillaceum (Michx.) Rafinesque (SP - RT0012)

APOCYNACEAE

Amsonia tabernaemontana Walter var. salicifolia (Pursh) Woodson (SP - RT0043)

Apocyonum cannabinum L. (SP - RT0199) [P]

Asclepias incarnata L. var. pulchra (SP - RT0276)

Asclepias rubra L. (TP - RT0520) [UC]

Asclepias tuberosa L. var. rolfsi (Britton ex Vail) Shinners (TP - RT0139) [P, UC]

ARISTOLOCHIACEAE

Hexastylis arifolia (Michx.) Small var. ruthii Small (BC - RT0397) [UC]

ASTERACEAE

Achillea millefolium L. (TP - RT0668) [P]

Ageratina altissima King & H.E. Robinson var. altissima (SP - RT0267, TP - RT0755, BB - RT0238) [P]

Ambrosia artemisiifolia L. (SP - RT0292, TP - 0663, SNFI - RT0699) [P]

Bidens frondosa L. (SP - RT0006, BJ - RT0683) [P]

Carphephorus bellidifolius (Michx.) Torrey & A. Gray (TP - RT0528)

Chrysogonum virginianum L. var. virginianum (TP - RT0447) [P]

Chrysopsis mariana (L.) Elliot (SP - RT0441, BC - RT0408) [P]

Cirsium vulgare (Savi) Tenore (SP - RT0258) [P]

Conyza canadensis (L.) Cronquist var. canadensis (SP - RT0294, SNFI - RT0687) [P]

Coreopsis lanceolata L. (SP - RT0083) [P]

Coreopsis major Walter var. rigida (TP - RT0158, BC - RT0620) [P]

Coreopsis pubescens var. robusta (BC - RT0421)

Coreopsis tinctoria Nuttall var. tinctoria (TP - RT0523, BJ - RT0188) [P, UC]

Erechites hieracifolius (L.) Rafinesque ex de Candolle (SP - RT0437, TP - RT0591) [P]

Erigeron strigosus Muhlenberg ex Willdenow var. strigosus (SP - RT0097, TP – RT0666) [P]

Eupatorium album L. var. album (TP - RT0360) [P]

Eupatorium capillifolium (Lamarck) Small (SP - RT0440, TP - RT0665) [P]

Eupatorium hyssopifolium L. (TP - RT0670, BJ - RT0682) [P]

Eupatorium rotundifolium L. (TP - RT0365) [UC]

Euthamia caroliniana (L.) Greene ex Porter & Britton (TP - RT0596)

Eutrochium fistulosum (Barratt) E.E. Lamont (TP - RT0569)

Helenium amarum (Rafinesque) H. Rock var. amarum (SP - RT0444) [P]

Helenium autumale L. (SP - RT0433, BC - RT0392) [P]

Helianthus angustifolius L. (TP - RT0530)

Helianthus atrorubens L. (TP - RT0332, BC - RT0407) [P]

Helianthus divaricatus L. (TP - RT0325, BC - RT0619) [P]

Helianthus schweinitzii Torry & A. Gray (TP - not collected) [P, SCR, NCE, USE]

Krigia dandelion (L.) Nuttall (TP - RT0449) [P]

Lactuca canadensis L. (SP - RT0580, SNFI - RT0251) [P]

Leucanthemum vulgare Lamarck (SP - RT0090, SNFI - RT0709)

Liatris pilosa (Aiton) Willdenow (TP - RT0664) [P]

Liatris squarrulosa Michx. (BJ - RT0772) [P, NCR]

Marshellia obovata (Walter) Beadle & F.W. Boynton var. obovata (SP - RT0092, TP – RT0468, BJ - RT0681)

Mikania scandens (L.) Willdenow (TP - RT0589)

Oligoneuron album (Nuttall) Nesom (BJ - RT0381) [P, SCR]

Packera aurea (L.) A & D Love (SP - RT0038, SNFI - RT0698) [P]

Parthenium integrifolium L. var. integrifolium (TP - RT0156, SNFII - RT0710) [P]

Pityopsis aspera (Shuttleworth ex Small) Small var. adenolepsis (SP - RT0761, TP – RT0311, SNFII - RT0254)

Pseudognaphalium obtusifolium (L.) Hilliard & Burtt (TP - RT0595) [P]

Rudbeckia hirta L. var. hirta (BC - RT0543) [P]

Rudbeckia hirta L. var. pulcherrima Farwell (BJ - RT0387) [P]

Seriocarpus asteroides (L.) Britton, Sterns, & Poggenburg (SNFII - RT0249)

Seriocarpus linifolius (L.) Britton, Sterns & Poggenburg (TP - RT0330, BJ - RT0186) [P]

Silphium compositum Michx. var. compositum (TP - RT0629, BC - RT0621, SNFII – RT0253) [P]

Silphium terebinthinaceum Jacq. (BJ - RT0378) [P, UC]

Solidago altissima L. var. altissima (BB - RT0235) [P]

Solidago curtisii Torrey & A. Gray (BC - RT0419)

Solidago gigantea Aiton (SP - RT0306, TP - RT0649) [P]

Solidago gracillima Torrey & A. Gray (TP - RT0322) [NCR]

Solidago nemoralis Aiton var. nemoralis (SP - RT0432, TP - RT0344, BJ - RT0382, BC - RT0416) [P]

Solidago odora Aiton var. odora (TP - RT0343) [P]

Solidago patula Muhlenberg ex Wildenow var. patula (TP - RT0768) [UC]

Solidago pinetorum Small (TP - RT0336) [P]

Solidago puberula Nuttall var. puberula (BC - RT0417) [NCWL]

Solidago radula Nuttall (TP - RT0352) [P, NCR]

Solidago rugosa P. Miller (BC - RT0414) [P]

Symphyotrichum concolor (L.) Nesom var. concolor (TP - RT0616)

Symphyotrichum dumosum L. (SP - RT0443, TP - RT0341, BJ - RT0762, BC - RT0420, OS -RT0730) [P]

Symphyotrichum grandiflorum (L.) Nesom (TP - RT0363)

Symphyotrichum lateriflorum L. (SP - RT0442) [P]

Symphyotrichum rhiannon Weakly & Govus (BC - RT0403) [NCR]

Verbesina occidentalis (L.) Walter (BB - RT0236)

Vernonia noveboracensis (L.) Michx. (TP - RT0567)

BETULACEAE

Alnus serrulata (Aiton) Wildenow (TP - RT0326)

BIGNONIACEAE

Campsis radicans (L.) Seeman ex Bureau (SP - RT0740, TP - RT0635, BJ - RT0774) [P]

CAMPANULACEAE

Campanula divaricata Michx. (BC - RT0389)

Lobelia glandulosa Walter (TP - RT0329)

Lobelia nuttallii J.A. Schultes (TP - RT0152)

Lobelia puberula Michx. var. simulans Fernald (BC - RT0423) [P]

Triodanis perfoliata (L.) Nieuwland (SP - RT0089, TP - RT0462, SNFI -RT0700) [P]

CAPRIFOLIACEAE

Lonicera japonica Thunberg (SP - RT0752, BJ -RT0678)

CARYOPHYLLACEAE

Dianthus armeria L. var. armeria (SP - RT0229)

COMMELINACEAE

Tradescantia ohiensis Rafinesque (SP - RT0023) [P]

CONVOLVULACEAE

Ipomoea pandurata (L.) G.F.W. Meyer (TP - RT0576) [P]

CUCURBITACEAE

Melothria pendula L. var. pendula (SP - RT0275)

CYPERACEAE

Carex atlantica Bailey (SP - RT0044, TP - RT0460) [UC]

Carex bushii Mackenzie (SP - RT0066) [P, NCR]

Carex caroliniana Schweinitz (BJ -RT0193, BB - RT0513)

Carex complanata Torrey & Hooker (TP - RT0154) [P]

Carex crinita Lamarck var. crinita (TP - RT0146)

Carex flaccosperma Dewey (SP - RT0085, TP - RT0651, BJ - RT0686)

Carex lurida Wahlenberg (SP - RT0222, TP - RT0170)

Carex intumescens Rudge var. intumescens (TP - RT0142)

Carex scoparia ScKuhr ex Willdenow var. scoparia (SP - RT0756, TP - RT0493, BC – RT0548, BB - RT0498) [P]

Carex squarrosa L. (SP - RT0070) [P]

Carex stricta Lamarck (SP - RT0047) [P]

Carex styloflexa Buckley (SP - RT0039)

Carex swanii (Fernald) Mackenzie (BC - RT0547)

Carex tenera Dewey var. tenera (SP - RT0086) [P, NCWL]

Carex virescens Muhlenberg ex Willdenow (BC - RT0612)

Carex vulpinoidea Michx. (SP - RT0067) [P]

Cyperus echinatus (L.) Wood (SP - RT0221) [P]

Cyperus erythrorhizos Muhlenberg (SP - RT0280, TP - RT0592)

Cyperus pseudovegetus Steudel (SP - RT0279)

Cyperus strigosus L. (SP - RT0011)

Eleocharis tricostata Torrey (SP - RT0076) [NCWL]

Eleocharis tuberculosa (Michx.) Roemer & J. A. Schultes (TP - RT0347)

Fimbristylis autumnalis (L.) Roemer & J.A. Schultes (TP - RT0353)

Fuirena squarrosa Michx. (TP - RT0320)

Rhynchospora caduca Elliott (SP - RT0009, TP - RT0648) [P, UC]

Rhynchospora glomerata (L.) Vahl var. glomerata (SP - RT0010, TP - RT0323)

Rhynchospora gracilenta A. Gray (TP - RT0350)

Rhynchospora recognita (Gale) Kral (TP - RT0317)

Scirpus atrovirens Willdenow (SP - RT0068, TP - RT0634) [P]

Scirpus cyperinus (L.) Kunth (SP - RT0585)

Scirpus pendulus Muhlenberg (SP - RT0062) [P, NCR]

Scleria muehlenbergii Steudel (TP - RT0307)

Scleria triglomerata Michx. (TP - RT0149) [P]

EBENACEAE

Diospyros virginiana L. (SP - RT0739, TP - RT0630, BJ - RT0679, OS - RT0769) [P]

ERICACEAE

Lyonia ligustrina (L.) Augustin de Candolle var. foliosiflora (Michx.) Fernald (TP – RT0153)

Lyonia mariana L. (TP - RT0461)

Rhododendron calendulaceum (Michx.) Torrey (BC - RT0559)

Rhododendron viscosum (L.) Torrey (BC - RT0535)

Vaccinium stamineum L. var. stamineum (BC - RT0422, SNFII - RT0473) [P]

Vaccinium tenellum Aiton (TP - RT0349) [P]

EUPHORBIACEAE

Euphorbia corollata L. (SNFII - RT0244) [P]

Tragia urticifolia Michx. (SP - RT0738, BJ - RT0696) [P]

FABACEAE

Apios americana Medikus (SP - RT0005) [P]

Baptisia tinctoria (L.) Ventenat (TP - RT0177)

Chamaecrista fasciculata (Michx.) Greene var. fasciculata (SP - RT0263, BJ - RT0676) [P]

Clitoria mariana L. var. mariana (TP - RT0521)

Crotalaria rotundifolia Walter ex J.F. Gmelin var. vulgaris (SP - RT0273) [P]

Desmodium canescens (L.) Augustin de Candolle (SP - RT0003, BC - RT0629) [P]

Desmodium ciliare (Muhlenberg ex Willdenow) Augustin de Candolle (SP - RT0296) [P]

Desmodium paniculatum (L.) Augustin de Candolle var. paniculatum (SP - RT0445, TP – RT0588) [P]

Desmodium perplexum Schubert (SP - RT0299)

Galactia volubilis (L.) Britton var. volubilis (SP - RT0274) [P]

Lespedeza bicolor Turczaninow (TP - RT0314)

Lespedeza cuneata (Dumont-Cours.) G. Don (SP - RT0262, TP - RT0652, BJ - RT0685, BC - RT0622, SNFI - RT0702)

Lespedeza procumbens Michx. (SP - RT0298, TP- RT0633, BB - RT0508, SNFII – RT0713) [P]

Lespedeza virginica (L.) Britton (TP - RT0653) [P]

Mimosa microphylla Dryander (TP - RT0646)

Orbexilum pedunculatum (P. Miller) Rydberg var. psoralioides (BJ - RT0182) [P, UC]

Rhynchosia tomentosa (L.) Hooker & Arnott var. tomentosa (TP - RT0164) [P]

Stylosanthes biflora (L.) Britton, Sterns, & Poggenburg (TP - RT0310) [P]

Tephrosia virginiana (L.) Persoon (TP - RT0467, SNFI - RT0701) [P]

Trifolium campestre Schreber (SP - RT0737)

Trifolium pratense L. (SP - RT0268)

FAGACEAE

Quercus marilandica Muenchhausen var. marilandica (TP - RT0631, SNFII –

RT0714) [P]

Quercus stellata Wagenheim (TP - RT0647, BJ - RT0753) [P]

GELSEMIACEAE

Gelsemium sempervirens St. Hilaire (TP - RT0451)

GENTIANACEAE

Gentianella quinquefolia (L.) Small var. quinquefolia (BC - RT0418) [P]

Sabatia angularis (L.) Pursh (SP - RT0265) [P]

HYPERICACEAE

Hypericum denticulatum Walter (BJ - RT0386)

Hypericum drummondii (Greville & Hooker) Torrey & A. Gray (TP - RT0315) [P, UC]

Hypericum gentianoides (L.) Britton, Sterns, & Poggenburg (TP - RT0309) [P]

Hypericum mutilum L. var. mutilum (SP - RT0285) [P]

Hypericum prolificum L. (SP - RT0260, BB - RT0726, SNFI - RT0256)

Hypericum stragulum P. Adams & Robson (BC - RT0557)

Hypericum virgatum Lamarck (TP - RT0566)

IRIDACEAE

Iris verna L. var. verna (TP - RT0452) [P]

Sisyrinchium mucronatum Michx. (SP - RT0024) [P]

JUNCACEAE

Juncus acuminatus Michx. (SP - RT0063, TP - RT0757)

Juncus effusus L. var. soltus (SP - RT0224, TP - RT0758) [P]

Juncus scirpoides Lamarck var. scirpoides (SP - RT0069, TP - RT0150)

Juncus tenuis Willdenow (SP - RT0735, TP - RT0632) [P]

Luzula echinata (Small) F.J. Hermann (SP - RT0057) [P]

LAMIACEAE

Physostegia virginiana (L.) Bentham var. praemorsa (BJ - RT0389) [P]

Prunella vulgaris L. (SP - RT0233, TP - RT0623, BJ - RT0684, BC - RT0773) [P]

Pycanthemum incanum (L.) Michx. var. incanum (BC - RT0624, SNFII - RT0255)

Pycanthemum muticum (Michx.) Persoon (BC - RT0425)

Pycanthemum tenuifolium Schrader (SP - RT0198, TP - RT0316, BJ - RT0677, BC – RT0628, BB - RT0725) [P]

Salvia lyrata L. (SP - RT0743, TP - RT0661, SNFI - RT0708) [P]

Scutellaria integrifolia L. (SP - RT0091) [P]

MAGNOLIACEAE

Liriodendron tulipfera L. var. tulipfera (TP - RT0656)

MELANTHIACEAE

Veratrum virginicum (L.) Aiton (SP - RT0213) [UC]

MELASTOMATACEAE

Rhexia virginica L. (SP - RT0200, TP - RT0178)

MYRICACEAE

Morella caroliniensis (P. Miller) Small (TP - RT0459)

NARTHECIACEAE

Aletris farinosa L. (TP - RT0464, BC - RT0561)

OLEACEAE

Fraxinus spp. L. (SP - RT0751,TP - RT0660, BJ - RT0674) [P]

ONAGRACEAE

Ludwigia alternifolia L. (SP - RT0002)

Ludwigia decurrens Walter (TP - RT0312)

Oenothera biennis L. (SP - RT0088, TP - RT0642, BJ - RT0191) [P]

Oenothera fruticosa L. var. fruticosa (BC -RT0536)

ORCHIDACEAE

Spiranthes vernalis Engelmann & A. Gray (TP - RT0519)

OROBANCHACEAE

Agalinis tenuifolia (Vahl) Rafinesque var. tenuifolia (BC - RT0406)

Castilleja coccinea (L.) Sprengel (SP - RT0096) [P, UC]

PHRYMACEAE

Mimulus alatus Aiton (SP - RT0300, TP - RT0327)

PLANTAGINACEAE

Mercardonia acuminata (Walter) Small var. acuminata (SP - RT0269, BJ - RT0376)

Nuttallanthus canadensis (L.) D.A. Sutton (SP - RT0742) [P]

Penstemon laevigatus Aiton (SP - RT0054)

Plantago virginica L. (SP - RT0093) [P]

POACEAE

Agrostis elliottiana J. A. Schultes (BB - RT0512)

Agrostis hyemalis (Walter) Britton (SP - RT0058) [P]

Aira caryophyllea L. (SP - RT0101, TP - RT0466) [UC]

Alopecurus carolinianus Walter (BC - RT0556, BB - RT0500)

Andropogon gerardii Vitman (SP - RT0750, TP - RT0671, BJ - RT0694, BC - RT0764, OS - RT0725, BB - RT0727, SNFI - RT0706, SNFII - RT0718) [P]

Andropogon glomeratus (Walter) Britton, Sterns & Poggenburg var. glomeratus (TP – RT0600) [P]

Andropogon virginicus L. var. virginicus (SP - RT0434) [P]

Anthoxanthum odoratum L. (TP - RT0659, SNFI - RT0703)

Aristida dichotoma Michx. (TP - RT0372)

Aristida oligantha Michx. (TP - RT0342) [P]

Arundinaria gigantea (Walter) Walter (TP - RT0371, SNFI - RT0245)

Bromus commutatus Schrader (SP - RT0099)

Bromus japonicus Thunberg ex Murray (SP – RT0059, BB - RT0511)

Bromus secalianus L. (SP - RT0100)

Chasmanthium laxum (L.) Yates (TP - RT0319)

Cinna arundinacea L. (TP -RT 0335)

Cynodon dactylon (L.) Persoon var. dactylon (TP - RT0594)

Danthonia compressa Austin ex Peck (BC - RT0550)

Danthonia sericea Nuttall (TP - RT0654, SNFI - RT0476) [P]

Danthonia spicata (L.) Palisot de Beauvois (SP - RT0606, TP - RT0471, BJ – RT0689) [P]

Dichanthelium acuminatum (Swartz) Gould & Clark var. fasciculatum (Torrey) (SP - RT0080, TP - RT0168, SNFI - RT0248) [P]

Dichanthelium clandestinum (L.) Gould (SP - RT0203, TP - RT0643, BJ - RT0614, BC – RT0625) [P]

Dichanthelium dichotomum (L.) Gould var. dichotomum (SP - RT0303, TP - RT0157, SNFII - RT0482) [P]

Dichanthelium scoparium (L.) Gould (SP - RT0007, TP - RT0394)

Dichanthelium sphaerocarpon (Elliott) Gould (SP- RT0216, TP - RT0469, BC – RT0409, SNFI - RT0241) [P]

Dichanthelium villosissimum (Nash) Freckmann var. villosissimum (TP - RT0522, BC – RT0551) [P]

Echinochloa crusgalli (L.) Palisot de Beauvois var. crusgalli (SP - RT0304, BB – RT0723)

Elymus virginicus L. var. virginicus (SP - RT0215, TP - RT0658, BJ - RT0690, BB – RT0504) [P]

Eragrostis spectabilis (Pursh) Steudel (SP - RT0436, TP - RT0308, BJ - RT0675) [P]

Festuca eliatior L. (TP - RT0644, BJ - RT0195, BC - RT0770)

Festuca paradoxa Desvaux (SP - RT0214)

Festuca rubra L. (SNFI - RT0767 )

Glyceria septentrionalis A.S. Hitchcock (SP - RT0055) [UC]

Glyceria striata (Lamarck) A.S. Hitchcock var. striata (SP - RT0077) [P]

Holcus lanatus L. (SP -RT 0102, TP - RT0636)

Leersia oryzoides (L.) Swartz (TP - RT0321)

Lolium perenne L. var. aristatum (SP - RT0061, OS - RT0732)

Muhlenbergia glomerata (Willdenow) Trinius (BC - RT0411) [NCR]

Panicum anceps Michx. var. anceps (TP -RT0313, BB - RT0721) [P]

Panicum dichotomiflorum Michx. var. dichotomiflorum (SP - RT0008) [P]

Panicum rigidulum Bosc ex Nees var. rigidulum (SP - RT0426)

Panicum virgatum L. var. cubense Grisebach (TP - RT0515)

Panicum virgatum L. var. virgatum (BC - RT0546 TP - RT0565, SNFI - RT0247) [P]

Paspalum dilatatum Poiret (SP - RT0266)

Paspalum setaceum Michx. (TP - RT0577)

Pennisetum glaucum L.R. Brown (SP - RT0013, TP - RT0637, BB - RT0722) [P]

Piptochaetium avenaceum (L.) Parodi (SNFII - RT0475) [P]

Poa languida Hitchcock (BC - RT0545)

Poa sylvestris A. Gray (SP - RT0036)

Saccharum giganteum (Walter) Persoon (SP - RT0435, TP - RT0331) [P]

Schizachyrium scoparium (Michx.) Nash var. scoparium (TP - RT0598)

Sorghastrum nutans (L.) Nash (SP - RT0305, TP - RT0645, BB - RT0728) [P]

Sorghum halapense (L.) Persoon (SP - RT0234)

Sphenopholis obtusata (Michx.) Scribner (SP - RT0060, SNFI - RT0707) [P]

Sphenopholis pensylvanica (L.) A.S. Hitchcock (TP - RT0465) [UC]

Tridens flavus (L.) A.S. Hitchcock (TP – RT0590, BB - RT0240) [P]

Tridens strictus (Nuttall) Nash (BJ - RT0388) [NCR]

Tripsacum dactyloides (L.) L. var. dactyloides (SP- RT0204, TP - RT0657, BB – RT0724) [P]

Vulpia elliotea (Rafinesque) Fernald (SP- RT0741)

Vulpia myuros (L.) K.C. Gmelin (SP - RT0056, TP - RT0525)

Vulpia sciurea (Nuttall) Henrard (SP - RT0098)

POLEMONIACEAE

Phlox carolina L. carolina (TP - RT0174, BC - RT0759)

Phlox nivalis Loddiges ex Sweet var. nivalis (TP- RT0771, SNFII - RT0715) [P]

POLYGALACEAE

Polygala incarnata L. (TP - RT0572)

POLYGONACEAE

Persicaria maculosa S.F. Gray (SP - RT0289)

PORTULACACEAE

Claytonia virginica L. var. virginica (SP - RT0486)

RANUNCULACEAE

Clematis virginiana L. (SNFI - RT0252)

Thalictrum macrostylum Small & Heller (BC - RT0398)

RHAMNACEAE

Ceanothus americanus L. var. americanus (TP - RT0145, BC - RT0412, SNFII – RT0716) [P]

ROSACEAE

Crataegus macrosperma Ashe (TP - RT0446, SNFII - RT0479)

Crataegus uniflora Muenchhausen (TP - RT0348, BJ - RT0672) [P]

Fragaria virginiana P. Miller (SP - RT0766) [P]

Gillenia trifoliata (L.) Moench (BC - RT0558)

Physocarpus opulifolius (L.) Maximowicz var. opulifolius (BC - RT0538)

Potentilla simplex Michx. (SP - RT0282, BB - RT0237) [P]

Prunus serotina Ehrhart var. serotina (TP - RT0640) [P]

Rosa carolina L. (BJ - RT0185) [P]

Rosa multiflora Thunberg ex Murray (SP - RT0746, BJ - RT0693)

Rosa wichuraiana Crepin (SNFII - RT0478)

Rubus spp. L. (SP- RT0259, TP - RT0760, BJ - RT0692, SNFI - RT0704, SNFII – RT0717) [P]

Sanguisorba canadensis L. (BC - RT0405) [UC]

RUBIACEAE

Cephalanthus occidentalis L. (SP - RT0230, BB - RT0719) [P]

Diodia teres Walter (TP - RT0339) [P]

Diodia virginiana L. (SP - RT0001, TP - RT0641, BB - RT0720)

Galium circaezans Michx. var. circaezans (SP - RT0290) [P]

Galium obtusum Bigelow var. filifolium (Wiegand) Fernald (SP - RT0048, BJ - RT0695)

Houstonia caerula L. (SP - RT0745, TP - RT0448, BC - RT0626) [P]

SAXIFRAGACEAE

Heuchera americana L. (SP - RT0095) [P]

SMILACACEAE

Smilax glauca Walter (BC - RT0540)

Smilax rotundifolia L. (SP - RT0747, BC - RT0627) [P]

SOLANACEAE

Physalis pubescens L. var. pubescens (SP - RT0272) [P]

Solanum carolinense L. var. carolinense (SP - RT0748, BJ - RT0673, SNFI - RT0705) [P]

TETRACHONDRACEAE

Polypremum procumbens L. (SP- RT0579)

ULMACEAE

Ulmus alata Michx. (SP - RT0749, BJ - RT0691)

VALERIANACEAE

Valerianella radiata (L.) Dufresne (SP - RT0207) [P]

VERBENACEAE

Verbena urticifolia L. var. urticifolia (SP - RT0261) [P]

VIOLACEAE

Viola primulifolia L. (TP - RT0457)

VITACEAE

Parthenocissus quinquefolia (L.) Planchon

Vitis aestivalis Michx. var. aestivalis (TP - RT0496) [P]

Vitis rotundifolia Michx. var. rotundifolia (TP - RT0639, OS - RT0724)

XYRIDACEAE

Xyris jupicai L.C. Richard (TP - RT0354)

Chapter 2

An Out-Crossing Reciprocity Study Between Remnant Big

Bluestem (Andropogon gerardii) Populations in the Carolinas

ABSTRACT

An out-crossing reciprocity study was conducted with five Andropogon gerardii

Vitman (Big Bluestem) populations across North and South Carolina for growing seasons

2007-08. A total of 15 treatments each were established at four garden sites. Overall seed germination was low from both out-crossed (4.8%) and selfed (2.4%) pairings. However, germination was significantly higher for both out-crossed and selfed seeds from ramets from the Suther Prairie population. There were no significant differences in seed set among plants from the five populations. In addition, there were no significant differences in parental effect in either germination or seed set among plants from the five populations. Plants from the BlackJacks Heritage Preserve population source (paternal) x

Suther Prairie population (maternal) and it reciprocal SP x BJ had the highest combined percent seed germination (21.3% and 5.8%, respectively). Plants from the Suther Prairie population (paternal) x Buck Creek Serpentine Barren (maternal) and its reciprocal BC x

SP had the highest combined percentage of seed set (22.7% and 9.6%, respectively).

Plants from the Sumter National Forest II population failed to produce germinable seeds.

Andropogon gerardii ramets or seeds from the Suther Prairie population are recommended for incorporation into restoration and establishment efforts for local A. gerardii populations.

INTRODUCTION

Habitat fragmentation is a significant threat to the maintenance of biodiversity in terms of both the numbers of individuals and the genetic variation within an ecosystem.

Fragmentation of ecosystems often leads to the loss of genetic diversity through (1) genetic drift, (2) elevated inbreeding, (3) reduced inter-population gene flow, and (4) increased probability of local extinction of demes (Ellstrand and Elam 1993; Young et al.

1996; Templeton et al. 1990).

As a result of the fragmentation of populations, restoration of rare species in plant communities may require the introduction of local genotypes to increase the opportunities for increased genetic diversity within small or genetically similar populations. Local or regional ecotypes are often recommended for the restoration of rare species (Knapp and

Rice 1994; Gustafson et al. 2004; Havens 1998; Jones 2003). The use of either seeds or vegetative sources from local gene pools has been reported to incorporate genotypes with increased competitive fitness within plant populations, as compared with the use of non- local and cultivar sources (Knapp and Rice 1994; Gustafson et al. 2004; Havens 1998;

Jones 2003). The introduction of non-local or cultivated genotypes may negatively affect community dynamics by reducing fecundity, germination rates, and survivorship

(Gustafson et al. 2004; Knapp and Rice 1994; Jones 2003). If nearby populations are lacking as potential sources for restoration material, it has been suggested that source material should at least originate from regional populations for the restoration of a given species (Knapp and Rice 1994; Jones 2003).

Although common in the Midwestern U.S., A. gerardii Vitman (Big Bluestem) is relatively rare in much of the eastern U.S. (Radford et al. 1968). Andropogon gerardii is

thought to have been once more abundant throughout the Carolinas (Barden 1997).

Where it does occur, it is often found in small, fragmented populations that most likely consist of clonal units (Chappell 2003). In a study of population structure of A. gerardii at

Konza Prairie in Manhattan, Kansas, Keeler et al. (2002) found that the diameter of established clones of A. gerardii was relatively small with a mean area of 3.2 m2. The size of A. gerardii clonal units in the Carolinas are thought to be potentially larger.

Chappell (2003) reported low caryopsis production within four small A. gerardii populations in South Carolina. Even in larger Midwestern populations of A. gerardii where genetically compatible mates are more likely to occur, individuals have been reported to have highly variable seed set and seed mass (Norrmann et al. 1997; Law and

Anderson 1940). As a result of potential self-incompatibility, individuals in small isolated populations, such as A. gerardii, may be unable to find genetically compatible mates

(Waburton et al. 2000). This may limit viable seed production within those smaller populations. In addition, the distance between local populations of A. gerardii most likely precludes inter-population gene flow.

Other studies of rare or isolated plant populations have revealed that as population size decreases, there is often a corresponding decrease of within population genetic diversity from reduced heterozygosity that can reduce plant fitness, including reproductive fitness

(Fisher and Matthies 1998; Oostermeijer et al. 1994; Van Treuren et al. 1990; Polans and

Allard 1989; Warburton et al. 2000; Kery et al. 2000; Menges 1991; Wolf et al. 2000;

Heschel and Paige 1995; Raijmann et al. 1994). This includes having reports of seed abortion due to inbreeding (Kerry et al. 2000). Law and Anderson (1940) reported that inbred individuals of A. gerardii had reduced overall plant vigor, including a reduction in

seed set, and in some individuals the elimination of spikelet production all together.

A two-year out-crossing reciprocity study using ramets from five A. gerardii plants from populations in the Carolinas was established with the following goals: (1) to assess the reproductive reciprocity and viability of seeds produced between out-crossings of selected A. gerardii populations. This includes assessing the reproductive viability of both pollen donors and pollen acceptors from those populations; (2) to provide information leading to the development of local or regional ecotype seed sources that can be used as accession lines for future restoration of A. gerardii populations; (3) to determine if both seed germination and seed set is higher from out-crossed and selfed pairings with plants from larger vs. smaller A. gerardii populations, and (4) to determine if self-incompatibility is maintained within intra-clonal crosses from A. gerardii populations. Similar reciprocity studies have been conducted for other plant species

(Reinartz and Les 1994; Demauro 1993; Heschel and Paige 1995).

MATERIALS AND METHODS

Field Experimental Set-up

In the Spring of 2007 five populations of A. gerardii were selected for the study.

Populations were selected from various physiographic regions of both North and South

Carolina (Fig. 2.1). Both Suther Prairie (SP; 2.8 ha), Cabarrus County, North Carolina

(35.45˚N, 80.46˚W) and Buck Creek Serpentine Barren (BC; 3 ha), Clay County, North

Carolina (35.08˚N, 83.61˚W) are two of the largest known for the Carolinas. The remaining three sites, Sumter National Forest II (SNFII; 2.4 x 10-3 ha), Oconee County,

South Carolina (34.75˚N, 83.27˚W), Blackjacks Heritage Preserve (BJ; 9.0 x 10-3 ha),

York County, South Carolina (34.90˚N, 81.01˚W), and Central (C; 0.15 ha), Pickens

County, South Carolina (34.42˚N, 82.47˚W) have much smaller populations (Fig. 2.1).

Figure 2.1. Source population sites of Andropogon gerardii used in this study.

1 = Suther Prairie, 2 = Sumter National Forest II, 3 = BlackJacks Heritage Preserve, 4 = Buck Creek Serpentine Barrens, and 5= Central.

For each of the five populations, a vegetative clump approximately 1 m2 was removed consisting of clonal ramets of one genet. Each clump was then subdivided into 8-10 cm sections and planted in pots prior to transplanting to the field. Pairings of ramets from the five A. gerardii source populations were used to establish 15 treatments and transferred to four garden sites (replicate blocks) owned by Clemson University (Table 2.1; n=60). The garden sites included the Simpson Experimental Farm, upper section (1) and lower section (2); the South Carolina Botanical Garden (3); and the Calhoun Experimental

Farm (4).

Table 2.1. Treatment pairings of ramets from five Andropogon gerardii source

populations at each experimental site.

SP BC BJ C SNF SP x BC x x BJ x x x C x x x x SNF x x x x x

SP = Suther Prairie, BC = Buck Creek Serpentine Barrens, BJ = BlackJacks Heritage Preserve, C = Central, and SNF = Sumter National Forest II.

Each treatment was established at a minimum of 75 m from other treatments to reduce the possibility of pollination from other nearby treatments. Ramets from the paired source populations in each treatment were planted approximately 25 cm apart. All treatment pairings were visited over the growing season on a regular basis to water plants and to remove any invading vegetation.

When both ramets within each treatment pairing had exserted anthers and stigmas, the inflorescences were marked to document sexual synchrony, and thus the potential for both maternal and paternal contributions from both sources. At maturity seeds were collected just prior to shattering and dry stored in paper bags. At the beginning of the second year‟s growing season, any plants that had died from the previous year were replaced with ramets from the same vegetative clonal source, and the experiment was repeated.

Germination and Seed Set

A random subsample of seeds collected from the treatments was cold-moist stratified at

5˚C for two weeks. Additional petri dishes were then filled with sand as a substrate and moistened with a 1% solution of KNO3. The dishes were then placed in a controlled growth chamber with a 20-30˚C (16-8 hrs.) alternating temperature/light regime (Rules for Testing Seeds 1998). The second year‟s germination procedure was similar, except that seeds were instead placed in a greenhouse under similar temperature and light conditions. Germination counts were made after a two week period.

In treatment pairings where a sufficient number of seeds remained after germination testing, a random subsample of remaining seeds was tested to determine percent seed set

(seeds with embryo and endosperm). Seeds were evenly distributed in petri dishes and

then x-rayed at the USDA Seed Testing Laboratory, Dry Branch, Georgia. X-ray images were evaluated for the presence of x-ray dense masses within individual seeds (Fig. 2.2).

Germination and seed set data were analyzed using ANOVA with SAS v. 9.1. (SAS

Institute 2002).

Figure 2.2. X-ray images of Andropogon gerardii seeds showing filled vs. non-filled

(empty) seeds.

RESULTS

Germination/Seed Set

The 2007 and 2008 growing seasons were among the driest in the last century (State

Climate Office of North Carolina 2010). As a result, despite regular watering, one or both plants of some pairings failed to survive for either of the two growing seasons prior to reaching reproductive maturity. In addition, several planted pairs were destroyed by human disturbance during both years of the study. Despite those losses, all 15 treatments had at least one surviving replicate pairing for each growing season. For 2007 58% (n =

35) of the pairings had both members to survive and for 2008, 60% (n = 36) had both members to survive. Both flower synchrony and the subsequent production of seeds for all surviving replicate pairings were observed for both growing seasons. Seeds from 28

(47%) of the remaining replicate pairings were x-rayed for 2007 along with 31 (52%) for

2008. The means for percent seed set and germination for each source population were positively correlated for both years combined (p =0.011).

The replicate block effect for seed germination was significant on the strength of the

Simpson Upper block mean (p = 0.014; Table 2.2.A). However, germination for the source populations was consistent across all four replicate blocks (p = 0.062; Table

2.2.A). With the exception of seeds from Suther Prairie ramets from two of the replicate blocks, germination means were not significantly different for source populations with each replicate block when compared for both growing seasons (p = 0.346; Tables 2.2.A and 2.2.B).

In 11 of the 15 treatments, at least one of the two ramets within a pairing produced germinable seeds. Seed germination was relatively low and not significantly different (p

= 0.221) for out-crossed treatments (4.8%) compared to selfed (2.4%).

Table 2.2.A. ANOVA results summarizing the experimental treatment effects for seed

germination for source populations with replicate block (treatment) by year.

Source DF MS F Value Pr > F Pop 4 761.189 15.10 <0.0001 Trt (pop) 20 81.447 1.62 0.0617 Site 3 186.619 3.70 0.0140 Year 1 45.003 0.89 0.3469 Site x pop 12 88.405 1.75 0.0654 Year x site 3 95.910 1.90 0.1335 Year x pop 4 23.039 0.46 0.7671

Table 2.2.B. Mean percent seed germination for A. gerardii by replicate by source

population for years 2007 and 2008.

Replicate* ______SU SB BG CA ______

Source 07 08 07 08 07 08 07 08 ______

SP** A25.3 C,B10.7 B,C3.7 B,C2.2 B,C5.5 B11.0 B,C8.0 A28.3

BC B,C3.2 B,C8.0 C0.0 B,C3.5 C0.0 C,B4.5 C0.0 C,B2.8

BJ B,C3.4 B,C6.0 C0.0 C0.0 B,C,D1.4 D0.0 C0.0 B,C3.0

C B,C3.3 B,C3.3 B,C0.7 B,C2.3 B,C6.8 B,C3.0 B,C3.0 B,C8.0

SNF C0.0 C0.0 C0.0 C0.0 C0.0 C0.0 C0.0 C 0.0 ______

Means with the same letter are not significantly different (p=0.05) using Fisher‟s LSD. *SU = Simpson Upper, SB = Simpson Bottoms, BG = Botanical Garden, and CA = Calhoun Bottoms.**SP = Suther Prairie, BC = Buck Creek Serpentine Barrens, BJ = BlackJacks Heritage Preserve, C = Central, and SNF = Sumter National Forest II.

for both growing seasons combined. Germination was, however, significantly higher for both out-crossed and selfed treatments with the Suther Prairie source population respectively (p =0.003, Tables 2.3.A and 2.3.B; p = <0.001, Table 2.5). In addition, seed set was not significantly different (p = 0.1426) among the out-crossed source populations (Tables 2.4.A and 2.4.B). There were also no significant differences among out-crossed source populations for paternal effect for both germination (p = 0.443;

Table 2.3.A and 2.3.B) and seed set (p = 0.232; Tables 2.4.A and 2.4.B). For the treatments, BJ (paternal) x SP (maternal) and it reciprocal SP x BJ had the highest combined percent seed germination (21.3%; 5.8%; Table 2.3.B). Of the out-crossed treatments, only BC x BJ failed to produce germinable seeds in both ramets (Table

2.3.B).

Three of the five selfed treatments (BJ x BJ, SNFII x SNFII, C x C; Table 2.5) failed to produce seeds that germinated, and two of the selfed treatments (BC x BC and SNFII x

SNFII; Table 2.5) failed to produce filled seeds. For the treatments, SP (paternal) x BC

(maternal) and its reciprocal BC x SP had the highest combined percentage of filled seed

(22.7% and 9.6%, respectively) (Table 2.4.B). The SNFII source population did not have seed set in the out-crossed treatments with SP, BJ, and C source populations (Table

2.4.B).

Plants from the SNFII source population failed to produce any germinable seeds in either year, while all five source populations contributed viable pollen in out-crossing treatments leading to germinable seed production (Table 2.3.B). Five of the out-crossed treatments produced seeds that germinated for both source populations (Table 2.3.B). In contrast, seven of the out-crossed treatments had seed set when either source population

served as a pollen donor or pollen acceptor (Table 2.4.B).

Neither an increase in germination nor seed set was significantly correlated with larger population size of the source populations (p = 0.266). However, when the Buck Creek source population was excluded there was a significant correlation (p = 0.039).

Table 2.3.A. ANOVA for percent germination of out-crossed treatments.

Source DF MS F Value Pr > F Maternal 4 783.362 18.08 0.0037 Paternal 4 105.904 2.44 0.4432 Trt(mat x pat) 11 104.819 2.42 0.0156 Site 3 245.022 5.65 0.0019 Year 1 0.350 0.01 0.9287 Mat x Site 12 171.392 3.96 0.0002 Pat x Site 12 68.574 1.56 0.1241 Mat x Year 4 27.321 0.63 0.6428 Pat x Year 4 62.732 1.45 0.2307

Table 2.3.B. Maternal and paternal contributions for out-crossed source populations for

percent seed germination (years combined).

Paternal source

______

SP BJ BC C SNF

______Maternal source Maternal source means

SP --- A21.3 C,D5.2 A,B16.0 B,C,D7.5 A12.3 BJ C,D5.8 --- D0.0 C,D3.4 D0.6 B2.6 BC B,C,D8.1 D0.0 --- C,D5.0 D0.5 B4.3 C C,D6.5 C,D2.3 C,D6.7 --- C ,D2.8 B4.6 SNF D0.0 D0.0 D0.0 D0.0 --- B0.0

Paternal source means A5.6 A5.9 A3.2 A7.2 A3.2

Means with the same letter are not significantly different (p=0.05) Fisher‟s LSD. SP = Suther Prairie, BC = Buck Creek Serpentine Barrens, BJ = BlackJacks Heritage Preserve, C = Central, and SNF = Sumter National Forest II.

Table 2.4.A. ANOVA for percent seed set for out-crossed treatments.

Source DF MS F Value Pr > F Maternal 4 374.382 8.07 0.1426 Paternal 4 286.654 6.18 0.2321 Trt (mat x pat) 11 174.264 3.75 0.0021 Site 3 230.471 4.97 0.0067 Year 1 0.351 0.01 0.9313 Mat x Site 11 245.726 5.29 0.0002 Pat x Site 12 74.975 1.62 0.1420 Mat x Year 4 66.450 1.43 0.2486 Pat x Year 4 10.424 0.22 0.922

Table 2.4.B. Maternal and paternal contributions for out-crossed source populations for

percent seed set (years combined).

Paternal source

SP BJ BC C SNF ______Maternal Maternal source source means

SP --- A,B21.6 C,D9.6 B,C,D11.0 C,D4.7 A12.4 BJ C,D9.0 --- D0.3 C,D2.8 C,D4.7 A,B4.1 BC A22.7 C,D1.3 --- C,D2.5 C,D0.8 A,B6.7 C C,D9.6 C,D1.8 B,C11.5 --- C,D3.5 A,B6.4 SNF D0.0 D0.0 D0.3 D0.0 --- B0.05

Paternal source means A9.0 A6.6 A 5.2 A4.7 A3.5

Table 2.5. Means for seed germination and seed set of selfed treatments for source

populations (combined years).

Selfed-Treatments

Germination Seed Set

Pop. Source

SP B,C11.0 C,D9.6

BC D0.9 D0.0

BJ D0.0 C,D5.4

C D0.0 C,D3.2

SNF D0.0 D0.0

Means with the same letter are not significantly different (p=0.05) using Fisher‟s LSD. SP = Suther Prairie, BC = Buck Creek Serpentine Barrens, BJ = BlackJacks Heritage Preserve, C = Central, and SNF = Sumter National Forest II.

DISCUSSION

The results of this study are like those found from Midwestern populations of A. gerardii in that populations in the Carolinas appear to be also primarily maintained by vegetative reproduction (Benson and Hartnett 2006; Chappell 2003). Both percent seed set and germination were low among the treatments for the study and consistent with reported rates of low viable seed production from A. gerardii populations elsewhere

(McKone et al. 1998; Law and Anderson 1940; Chappell 2003). This also appears to be consistent with annual regeneration and maintenance of A. gerardii populations on

Midwestern tallgrass prairie that has been shown to be almost exclusively by vegetative

(99%) rather than seedling recruitment (1%) (Benson and Hartnett 2006).

The significantly higher germination for seeds from the Simpson Experimental Farm replicate block was primarily attributable to higher germination from treatments with

Suther Prairie ramets. However, seed germination for all population sources within the

Simpson Experimental Farm replicate block was consistent between the replicate blocks.

Also, with the exception of Suther Prairie seeds from two of the replicate blocks, germination was not significantly different by source population for each replicate block for both growing seasons. In addition, seed set did not deviate significantly from germination for all source populations for both years combined.

Although there were no significant differences in paternal effect for both germination and seed set among the source populations, germination of seeds from Suther Prairie ramets in out-crossed treatments were significantly higher overall than the other four source populations. However, paternal effects were significantly greater from pollen of the BlackJacks Heritage Preserve and Central source populations in conjunction with the

Suther maternal source, compared with the other maternal sources for seed germination.

With the exception of lower than expected seed germination and seed set from the ramets of the Buck Creek source population, the results were consistent with reported lower reproductive success in smaller populations in some plant species. In similar studies, lower seed germination has been found to be correlated with population size in small, isolated populations of Hymenoxys acaulis var. glabra (Lakeside Daisy) (Demauro

1993), Ipomopsis aggregata (Scarlet Gilia) (Heschel and Paige 1995), Santalum lanceolatum (Sandlewood) (Warburton et al. 2000), Banksia goodi (Good‟s Banksia)

(Lamont et al. 1993), and at least partially correlated for small populations of the prairie perennial Silene regia (Royal Catch-Fly) (Menges 1991).

The complete absence of seed germination in the selfed treatments of the three smaller source populations (C, BJ, SNF), and very low seed germination for the Buck Creek source population, was consistent with reports of a functioning self-incompatibility system within populations of obligate out-crossing species elsewhere, including A. gerardii from Midwestern populations (Norrmann et al. 1997). Individuals of A. gerardii have been reported to possess a pre-zygotic incompatibility mechanism that results in the failure of the pollen tube to penetrate and grow into the style (McKone et al.1998;

Norrmann et al. 1997). This results in low reported seed set (0.2-6.0%) following self- pollination. In addition, post-zygotic mechanisms are suspected as well due to the low survivability of the offspring beyond the first growing season (Norrmann et al. 1997).

However, with the exception of seed germination (BJ x SP; SP x C) and seed set (BC x

SP; BJ x SP) from certain out-crossed treatments with Suther Prairie plants, the lack of significant variation in both seed germination and seed set for out-crossed treatments

would seem to indicate only a slight increased probability for successful out-crossing between some of the source populations used in the study. This lack of out-crossing success occurred despite the fact that the ramets were from populations that were separated by large geographical distances. Chappell (2003) also reported low within- population seed set (0-1%) from four small, isolated A. gerardii populations in South

Carolina. However, seed set increased from 0.8-11%, with a mean of 6%, when transplanted ramets from the four populations were allowed to out-cross with one another in a garden study.

Allelic self-incompatibility systems have been documented in several grass species

(Chapman 1996). If plant populations are highly inbred and have reduced levels of genetic diversity, then they may contain fewer self-incompatibility alleles which may reduce viable seed production between genetically similar individuals, particularly from those from very small populations (Chapman 1996; Demauro 1993). However, in an associated allozyme study (Tompkins et al. in prep.), although there was less genetic diversity within three smaller A. gerardii populations used in the study, all five populations sampled were found to have relatively high levels of genetic diversity compared to other reported plant species (Godt and Hamrick 1998). It is unclear though whether high levels of genotypic diversity from the isozyme data indicates maintenance of a corresponding large number of self-incompatibility alleles within those same populations.

The increased germination from out-crossed Suther Prairie ramets may thus be due to a greater number of self-incompatibility alleles within the population, as a result of its larger relative size, and thus a greater likelihood of out-crossing potential with

individuals of other populations (Chapman 1996).

Because of the larger A. gerardii population at the Buck Creek site, it is unclear why the overall seed germination and seed set from out-crossed treatments with Buck Creek ramets were not higher. In addition, seed germination for the selfed treatments of Buck

Creek was significantly lower than those of Suther Prairie. The A. gerardii population at

Buck Creek occurs on serpentine-based soil conditions. Plant species occurring on serpentine soils have been shown to be adapted to the unusual soil features associated with those sites. Other studies of species occurring on serpentine have indicated high levels of genetic differentiation between serpentine and non-serpentine populations

(Brady et al. 2005; Patterson and Givnish 2004; Mengoni et al. 2003). However, allozyme data from sampled plants from Buck Creek found high levels of genetic identity between the Buck Creek population and seven other local A. gerardii populations

(Tompkins et al. in prep). This suggests that the Buck Creek population is not genetically distinct from other local non-serpentine A. gerardii populations, and as a result, its occurrence on serpentine conditions does not help to explain why ramets from the Buck

Creek source population had reduced out-crossing success as compared with the Suther

Prairie ramets.

Another potential explanation for relatively lower viable seed production in the smaller populations in this study is ploidal incompatibility (Fowler and Levin 1984). Norrmann et al. (1997) reported that 6x (hexaploid) and 9x (enneaploid) A. gerardii cytotypes are inter-fertile in Midwestern studies. However, meiosis in enneaploids was irregular leading to defective gametes. In addition, data from the allozyme study which included the five A. gerardii populations used in this study indicated banding patterns consistent

with both diploid and polyploid cytotypes (Tompkins et al. in prep.). If so, the two ploidal levels may form triploid progeny which may be sterile or inviable (Fowler and

Levin 1984). As a result, reproduction by sexual means may be greatly reduced or eliminated entirely in some local A. gerardii populations due to either a functioning self- incompatibility system, or a high proportion of cytotypic disequilibrium among individuals, or both (Fowler and Levin 1984).

Clonal plant species, such as A. gerardii in small populations, may often exhibit high levels of fitness for a particular site (Oostermeijer et al. 1994), and may be sustained solely by vegetative growth. Within larger populations of A. gerardii in the Carolinas, there may be sufficient levels of out-crossing to ensure occasional cohort recruitment from seeds. This out-crossing, along with somatic mutations, may provide for some level of genetic variability within those larger populations (Tompkins et al. 2010).

This potential genetic variation, coupled with periodic disturbance to eliminate competitive mid-successional species, may be enough to ensure the long-term maintenance of larger populations of A. gerardii. However, although viable seed production is reported to be low in larger Midwestern populations of A. gerardii, the almost complete absence of viable seed production within some local A. gerardii populations, based on this study and that of Chappell (2003), may have severe consequences for the survival of smaller populations in the Carolinas.

Even where nearby suitable habitat conditions occur in the Carolinas, the ability of A. gerardii to either colonize or re-colonize potential sites may be negatively affected by low fecundity. Thus, even with suitable habitat conditions the recruitment of A. gerardii by sexual reproduction and subsequent seedling recruitment may be significantly

inhibited in some populations in the Carolinas.

Along with the regular monitoring of local A. gerardii populations, further analysis of the genetic diversity within and among local populations is needed to better assess their current status. The findings of this study suggest that there may be a reduction in reproductive fitness in local A. gerardii populations. Thus, it may be necessary to introduce other genotypes into those populations by the use of either vegetative ramets or seeds from other populations in order to prevent extinction and to increase genetic diversity. Furthermore, these findings indicated that local or regional seed source material should include the use of seeds or ramet material from the Suther Prairie A. gerardii population. In addition, the use of seed accession lines developed from other local out- crossed A. gerardii populations, also indicated by the findings in this study, may also be beneficial. Polans and Allard (1989) developed composite accession seed sources from several highly inbred Lolium multiflorum (Ryegrass) populations to introduce new genotypes into inbred populations. They found that there was recovery from a loss of genetic vigor after only one generation of random mating after the introduction of new genotypes into those populations. Similarly, when Heschel and Paige (1995) introduced pollen from distant Ipomopsis aggregata populations into two small I. aggregata populations, both the seed mass and percentage of germination increased within those populations. Thus, the development of composite accession seed lines of A. gerardii from other local sources may also provide a viable option to increase the number of different alleles in current A. gerardii populations in the Carolinas, or in the establishment of new populations.

Efforts should be made to introduce new genotypes into current A. gerardii

populations that are in peril of extinction. The findings of this study should provide baseline information for the development of future A. gerardii seed accession lines for the restoration and establishment of local A. gerardii populations.

LITERATURE CITED

Barden, L. S. 1997. Historic prairies in the Piedmont of North and South Carolina, USA.

Natural Areas Journal 17:149-152.

Benson, E. J. and D. C. Hartnett. 2006. The role of seed and vegetative reproduction in

plant recruitment and demography in tallgrass prairie. Plant Ecology 187:163-177.

Brady, K. U., A. R. Kruckeberg, and H. D. Bradshaw, Jr. 2005. Evolutionary ecology of

plant adaptation to serpentine soils. Annual Review of Ecology, Evolution, and

Systematics 36:243-66.

Chapman, G. P. 1996. The Biology of Grasses. pp. 71-88. CAB International. New

York, New York.

Chappell, J. 2003. Lack of caryopsis production in remnant big bluestem (Andropogon

gerardii) populations located in upstate South Carolina. M.S. Thesis. Clemson

University, Clemson, South Carolina.

Demauro, M. M. 1993. Relationship of breeding system to rarity in the Lakeside Daisy

(Hymenoxys auaulis var. glabra). Conservation Biology 7:542-550.

Ellstrand, N. and D. Elam. 1993. Population genetic consequences of small population

size: Implications for plant conservation. Annual Review of Ecology, Evolution,

and Systematics 24:217-242.

Fischer, M. and D. Matthies. 1998. Effects of population size on performance in the rare

plant Gentianella germanica. Journal of Ecology 86:195-204.

Fowler, N. L. and D. A. Levin. 1984. Ecological constraints on the establishment of a

novel polyploidy in competition with its diploid progenitor. The American

Naturalist 124:703-711.

Godt, M. J. W. and J. L. Hamrick. 1998. Allozyme diversity in grasses. pp. 11-29. In G.

P. Cheplick (ed.). Populations Biology of Grasses. Cambridge University Press.

Gustafson, D. J., D. J. Gibson, and D. L. Nickrent. 2004. Competitive relationships of

Andropogon gerardii (Big Bluestem) from remnant and restored native populations

and select cultivated varieties. Functional Ecology 18:451-457.

Havens, K. 1998. The genetics of plant restoration: An overview and a surprise.

Restoration and Management Notes 16:68-72.

Heschel, M. S. and K. N. Paige. 1995. Inbreeding depression, environmental stress, and

population size variation in Scarlet Gilia (Ipomopsis aggregata). Conservation

Biology 9:126-133.

Jones, T. A. 2003. The restoration gene pool concept: Beyond the native versus non-

native debate. Restoration Ecology 11:281-290.

Keeler, K. H., C. F. Williams, and L. S. W. Vescio. 2002. Clone size of Andropogon

gerardii Vitman (Big Bluestem) at Konza Prairie, Kansas. American Midland

Naturalist 147:295-304.

Kery, M., D. Matthies, and H. Spillmann. 2000. Reduced fecundity and offspring

performance in small populations of the declining grassland plants Primula veris

and Gentiana lutea. Journal of Ecology 88:17-30.

Knapp, E. E. and K. J. Rice. 1994. Starting from seed: Genetic issues in using native

grasses for restoration. Restoration & Management Notes 12:41-45.

Lamont, B. R., P. G. L. Klinkhamer and E. T .F. Witkowski. 1993. Population

fragmentation may reduce fertility to zero in Banksia goodi – a demonstration of

the Alee effect. Oecologia 94:446-450.

Law, A. G. and K. L. Anderson. 1940. The effect of selection and inbreeding on the

growth of Big Bluestem (Andropogon Furcatus, Muhl.). Journal of the American

Society of Agronomists 32:931-943.

McKone, M. J., C. P. Lund and J. M. O‟Brien. 1998. Reproductive biology of two

dominant prairie grasses (Andropogon gerardii and Sorghastrum nutans, Poaceae):

Male-biased sex allocation in wind-pollinated plants? American Journal of Botany

85:776-783.

Menges, E. S. 1991. Seed germination percentage increases with population size in a

fragmented prairie species. Conservation Biology 5:158-164.

Mengoni, A., C. Gonnelli, E. Brocchini, F. Galardi, S. Pucci, R. Gabbrielli, and M.

Bazzicalupo. 2003. Chloroplast genetic diversity and biogeography in the

serpentine endemic Ni-hyperaccumulator Alyssum bertolonii. New Phytologist

157:349-356.

Norrmann, G. A., C. L. Quarin, and K. H. Keeler. 1997. Evolutionary implications of

meiotic chromosome behavior, reproductive biology, and hybridization in 6x and

9x cytotypes of Andropogon gerardii (Poaceae). American Journal of Botany

84:201-207.

Oostermeijer, J. G. B., M W. van Eijck, and J. C. M. den Nijs. 1994. Offspring fitness in

relation to population size and genetic variation in the rare perennial plant species

Gentiana pneumonanthe (Gentianaceae). Oecologia 97:289-296.

Patterson, T. B. and T. J. Givnish. 2004. Geographic cohesion and parallel adaptive

radiations in Calochortus (Calochortaceae): Evidence from a cpDNA sequence

phylogeny. New Phytologist 161:253-264.

Polans, N. O. and R. W. Allard. 1989. An experimental evaluation of the recovery

potential of ryegrass populations from genetic stress resulting from restriction of

population size. Evolution 43:1320-1324.

Radford, A. E., H. E. Ahles, and C. R. Bell. 1968. Manual of the Vascular Flora of the

Carolinas. University of North Carolina Press. Chapel Hill, North Carolina.

Raijmann, L. E. L., N. C. Van Leeuwen, R. Kersten, J. G. B. Oostermeijer, H. C. M. D.

Nijs, and S. B. J. Menken. 1994. Genetic variation and outcrossing rate in relation

to population size in Gentiana pneumonanthe L. Conservation Biology 8:1014-

1026.

Reinartz, J. A. and D. H. Les. 1994. Bottleneck-induced dissolution of self-

incompatibility and breeding system consequences in Aster furcatus (Asteraceae).

American Journal of Botany 81:446-455.

Rules for Testing Seeds. 1998. Association of Official Seed Analysis.

SAS Institute. 2002. SAS User‟s Guide: Statistics. Version 9.1. SAS Inst., Cary, North

Carolina.

State Climate Office of North Carolina. 2010. (http://www.nc-climate.ncsu.edu).

Templeton, A. R., K. Shaw, E. Routman, and S. K. Davis. 1990. The genetic

consequences of habitat fragmentation. Annals of the Missouri Botanical Garden

77:13-27.

Tompkins, R. T., C. M. Luckenbaugh, W. C. Stringer, K. H. Richardson, E. A.

Mikhailova, and W. C. Bridges, Jr. 2010. Suther Prairie: Vascular flora, species

richness, and edaphic factors. Castanea 75:232-244.

Van Treuren, R., R. Bijlsma, W. Van. Delden, and N. J. Ouborg. 1990. The significance

of genetic erosion in the process of extinction. I. Genetic differentiation in Salvia

pratensis and Scabiosa columbaria in relation to population size. Heredity 66:181-

189.

Warburton, C. L., E. A. James, Y. J. Fripp, S. J. Trueman, and H. M. Wallace. 2000.

Clonality and sexual reproductive failure in remnant populations of Santalum

lanceolatum (Santalaceae). Biological Conservation 96:45-54.

Wolf, A. T., R. W. Howe, and J. L. Hamrick. 2000. Genetic diversity and population

structure of the serpentine endemic Calystegia collina (Convolvulaceae) in

northern California. American Journal of Botany 87:1138-1146.

Young, A., T. Boyle, and T. Brown. 1996. The population genetic consequences of

habitat fragmentation for plants. Tree 11:413-418.

Chapter 3

Genetic Variation Within and Among Big Bluestem

(Andropogon gerardii) Populations in the Carolinas

ABSTRACT

Protein electrophoresis was used to assess genetic variation within and among nine local Andropogon gerardii populations from various physiographic regions of both North and South Carolina. Twenty-seven polymorphic loci and one monomorphic locus were resolved. Genetic diversity was high at both the pooled species-wide level and at the mean population level. For the species-wide sample, percent polymorphic loci (Ps) was

96.7%. The number of alleles per polymorphic locus (APsp) was = 4.07, the mean number of alleles per locus (As) was = 3.97, and the genetic diversity (Hes) was 0.422. For within populations, the mean Pp = 81.2%, APp, = 2.63, Ap = 2.38, and Hep = 0.340. Levels of within population (Hep) genetic diversity ranged from 0.181 for the Sumter National

Forest I population to 0.462 for the Bowman Barrier population. Genetic structure among populations (Gst) was 0.175, indicating that 83% of the genetic variation is found within- populations. The level of genetic diversity for the three larger populations (He = 0.387),

and for the six smaller populations (He = 0.330), was not significantly different (p =

0.5535). Nei‟s unbiased genetic identity between pairs of populations ranged from 0.6386

(populations BJ and SNF1) to 0.9692 (populations BC and SP). Of the population sites, the Blackjacks Heritage Preserve population had the lowest measure of genetic identity with the other population sites, while the Suther Prairie population had the highest. The

Mantel test showed no significant genetic isolation by geographic distance (r = 0.065; p =

0.614). Allelic richness values were high for the populations. Banding patterns were consistent with disomic inheritance. However for two loci (PGI3; UGPP1), tetrasomic patterns were indicated.

INTRODUCTION

The amount and distribution of genetic variation within species is a subject of considerable interest because of its important evolutionary and conservation implications

(Huenneke 1991). Such studies are critical to understanding the fate of the increasing number of rare and endangered species. As a consequence of genetic drift, inbreeding, and restricted gene flow, small and isolated populations often show decreased levels of genetic variation compared to larger populations. If gene flow among populations is low, and genetic drift is the dominant evolutionary factor affecting populations, genetic differentiation among populations will be high (Huenneke 1991). This has been supported by the finding that small populations are more differentiated from each other compared to larger populations (Heschel and Paige 1995; Fischer and Matthies 1998;

Oostermeijer et al. 1994; Demauro 1993; Kery et al. 2000; Pleasants and Wendel 1989;

Van Treuren et al. 1990). Small populations are also more prone to extinction than larger populations due to reduced genetic variability and reduced resistance to environmental stochasticity (Gilpin and Soule 1986).

Although common in the Midwestern U.S., Andropogon gerardii Vitman (Big

Bluestem) is relatively rare in much of the eastern U.S. (Radford et al. 1968).

Andropogon gerardii was once thought to have been more abundant throughout the

Carolinas (Barden 1997). Where it does occur, it is often found in small, fragmented populations that are thought to consist of clonal units (Keeler 2002; Chappell 2003). In an isozyme study of A. gerardii population structure at Konza Prairie in Manhattan, Kansas

(Keeler 2002), the diameter of established clones were reported to be relatively small with a mean area of 3.2m2.

Midwestern populations of A. gerardii have been reported to be self-incompatible, and to possess a pre-zygotic incompatibility mechanism that results in the failure of the pollen tube to penetrate the style (McKone et al. 1998). This results in reported low seed set

(0.2-6.0%) following self-pollination. In addition, post-zygotic mechanisms are thought to contribute to the low survivability of the offspring beyond the first growing season

(Norrmann 1997). Also, an initial reproductive reciprocity study with five A. gerardii populations from the Carolinas suggested reduced reproductive reciprocity between smaller, local A. gerardii populations compared to a larger population, including lower percent seed set and germination (Tompkins et al. in review). Reductions in the size and increased spatial isolation of local A. gerardii populations in the post-settlement period may have resulted in decreased genetic variability and increased population divergence as a result of genetic drift and reduced gene flow (Young et al. 1996).

Gel electrophoresis of allozymes is a commonly used technique to investigate the genetic structure of plant populations. This includes species with small geographical ranges. Numerous studies have documented that restricted species are more likely to have lower levels of genetic diversity when compared to more widespread species (Loveless and Hamrick 1984; Karron 1987; Hamrick et al. 1979). Reduced genetic variation may also indicate reduced fecundity within a population (Kery et al. 2000).

Protein electrophoresis was used to help address several questions related to the levels and partitioning of genetic variation of A. gerardii populations in the Carolinas: 1.) What are the levels of genetic variation within and among populations in the Carolinas? 2.) Is the genetic distance among populations a function of the spatial distance separating them? 3.) Is there a relationship between A. gerardii population size and the population

level genetic diversity? The findings from this investigation have implications for conservation strategies and may be applicable as well to other rare clonal species that occur in small populations.

MATERIALS AND METHODS

Study Sites and Sampling

Nine A. gerardii populations were selected from various physiographic regions of both North and South Carolina (Fig. 3.1). Three of the largest known populations for the

Carolinas were selected including Suther Prairie, Cabarrus County, North Carolina; Troy

Prairie, Troy, North Carolina; and Buck Creek Serpentine Barren, Clay County, North

Carolina (Table 3.1). Six smaller population sites were also selected, including Rock Hill

Blackjacks Heritage Preserve, York County, South Carolina; Orton Site, Brunswick

County, North Carolina; Bowman Barrier Rd., Cabarrus County, North Carolina; Sumter

National Forest Sites I and II Oconee County, South Carolina; and Central, Pickens

County, South Carolina (Table 3.1).

Leaf tissue was collected from 24-48 individuals from each of nine populations in July of the 2010 (n = 288; Table 3.1). Samples were collected at random by starting at one end of each population moving back and forth laterally across the short axis of the population until the entire population was sampled evenly. This allowed for a uniform sampling of each population site. Samples were kept chilled during transport to the University of

Georgia.

Enzyme Extraction and Electrophoresis

Within 48 hours of collection, samples (40 mg of leaf tissue) were crushed in chilled

Figure 3.1. Map of sampled Andropogon gerardii populations for North and South

Carolina.

BB= Bowman Barrier Rd., BC = Buck Creek Serpentine Barren, BJ = BlackJacks Heritage Preserve, C = Central Site, OS = Orton Site, SP = Suther Prairie, SNFI and II = Sumter National Forest Site I and II and TP = Troy Prairie.

Table 3.1. Site descriptions from sampled populations of A. gerardii.

Site Pop. Size # of Coordinates Parent Material Soil Features

Area (ha) Samples

BB 0.5 24 35.38˚N, 80.42˚W alluvium Floodplain; hydric to mesic; pH 5.5 BC 3.0 48 35.08˚N, 83.61˚W felsic to mafic Serpentine conditions; high Mg:Ca ratio; presence of heavy metals; pH 6.8 BJ 9.0 x 10-3 24 34.90˚N, 81.01˚W ferromagnesium Vertic soil; very high levels of Ca and Mg ; pH 6.3 C 0.15 24 34.42˚N, 82.47˚W felsic, igneous, very deep; well metamorphic drained; strongly acidic OS 4.0 x 10-3 24 34.07˚N, 77.95˚W alluvium Roadside; highly disturbed; pH 6.6 SP 2.8 48 35.45˚N, 80.46˚W alluvium Floodplain; hydric to mesic; very high levels of Ca and Mg; pH 6.1 SNFI 1.2 x 10 -3 24 34.76˚N, 83.27˚W felsic to igneous Upland; moderate nutrient levels; pH 5.1 SNFII 2.4 x 10-3 24 34.75˚N, 83.27˚W felsic to igneous Upland; moderate nutrient levels; pH 5.1 TP 5.0 48 35.35˚N, 79.87˚W ferromagnesium Upland; moderate nutrients; pH 4.8

mortars with a pestle, a pinch of sea sand and an enzyme extraction buffer (Wendel and Parks 1982). Extracts were absorbed on 4 x 6 mm wicks of Whatman 3mm chromatography paper and wicks stored in microtest plates at -70º C until used for electrophoresis. Wicks were placed in horizontal 10% starch (Connaught hydrolyzed potato starch) gels and electrophoresis performed. Sixteen enzyme stains in four buffer systems resolved one monomorphic and 27 putative polymorphic loci. The enzymes stained and loci identified in parentheses for each of the buffer systems were as follows:

1) buffer system 4, aconitase (ACO1, ACO2), 6-phosphogluconate dehydrogenase (6-

PGD1, 6-PGD2), and shikimic dehydrogenase (SKDH1, SKDH2); 2) buffer system 8-, aspartate aminotransferase (AAT1, AAT2), fluorescent esterase (FE1, FE2, FE4), triosephosphate isomerase (TPI1, TPI2); 3) buffer system 11, isocitrate dehydrogenase

(IDH1), malate dehydrogenase (MDH1, MDH2), UTP-glucose-1-phosphate (UGPP1,

UGPP2); 4) buffer system 6, alcohol dehydrogenase (ADH2), diaphorase (DIA1, DIA2), menadione reductase (MNR5), peroxidase (PER2), phosphoglucoisomerase (PGI1, PGI2,

PGI3), and phosphoglucomutase (PGM1, PGM2). All buffer and stain recipes were adapted from Soltis et al. (1983) except DIA (Cheliak and Pitel 1984), MNR (Cheliak and Pitel 1984) and UGPP (Manchenko 1994). Buffer system 8- is a modification of buffer system 8 as described by Soltis et al. (1983). Banding patterns were consistent with Mendelian inheritance patterns expected for each enzyme system (Weeden and

Wendel 1989).

Data Analysis

Genetic diversity measures were estimated using a computer program, LYNSPROG, designed by M.D. Loveless and A. F. Schnabel. Measures of genetic diversity were:

percent polymorphic loci, P; mean number of alleles per locus, A; effective number of

2 alleles per locus, Ae = 1/∑pi ; mean number of alleles per polymorphic locus, APP ; genetic diversity, He (the proportion of loci heterozygous per individual under Hardy-

Weinberg expectations; Nei 1972). Species level values for these parameters were calculated by pooling data from all populations. Population level values were calculated for each population and then averaged across all populations. Observed heterozygosity

(Ho) was compared with Hardy-Weinberg expected heterozygosity (He ) for each polymorphic locus in each population by calculating Wright‟s fixation indices (i.e. inbreeding coefficient; Wright 1922; Wright 1951).

Nei‟s (1972) genetic distance statistics were calculated for each locus (monomorphic and polymorphic). Genetic identity values were calculated for each pair of populations.

An Unweighted Pair Group Method with Arithmetic Mean (UPGMA) phenogram of genetic identities was generated using NTSys-PC ver. 2.11j (Rohlf 2003).

Pairwise Gst values were obtained for all possible pairs of populations using only polymorphic loci with FSTAT (Goudet 2001). A Mantel test of correspondence between these pairwise Gst values and geographic distances for each pair of populations was performed (Smouse et al. 1986) using NTSys-PC ver. 2.11j (Rohlf 2003).

Because the observed number of alleles in a sample is highly dependent on the size of the sample (i.e. smaller samples should have fewer alleles), allelic richness was assessed using FSTAT (Goudet 2001). Allelic richness is a measure of the number of alleles independent of sample size. FSTAT employs Hurlbert‟s rarefaction index (1971) that was modified for population genetics (El Mousadik & Petit 1996; Petit et al. 1988). For the rarefaction analysis only polymorphic loci were used and A. gerardii populations were

treated as consisting of 24 genes, because in one population (Orton Site) only 12 individuals could be genotyped. Five loci (FE1, FE2, FE4, PER2 and TPI) were removed because for some populations these data were missing. An ANOVA was performed to determine if their were any significant differences in He, A, and Ap between the larger populations (n =3) and the smaller populations (n=6).

RESULTS

All sampled A. gerardii populations contained high levels of genotypic diversity and indicate that the populations have undergone less clonal spreading than expected. Only

Sumter National Forest I, Sumter National Forest II, and Troy Prairie showed any evidence of clonality. Sumter National Forest I (n=24) had 14 genotypes, two of which were each shared by two ramets and one of which was shared by nine ramets. Sumter

National Forest II (n=24) had 22 genotypes, one of which was shared by two ramets.

Troy Prairie (n = 48) had 46 genotypes, two of which were each shared by two ramets.

Twenty-eight allozyme loci were resolved from the nine A. gerardii populations, twenty-seven of which were polymorphic (Table 3.3). Banding patterns were consistent with disomic inheritance. However for two loci (PGI3; UGPP1), tetrasomic patterns were indicated.

Genetic diversity was high at both the pooled species-wide level and at the mean population level. For the species-wide sample, percent polymorphic loci (Ps) was 96.7%.

The number of alleles per polymorphic locus (APsp) was = 4.07, the mean number of alleles per locus (As) was = 3.97, and the genetic diversity (Hes) was 0.422 (Table 3.2).

For within-populations, the mean Pp = 81.2%, APp, = 2.63, Ap = 2.38, and Hep = 0.340

(Table 3.2).

The BlackJacks Heritage Preserve population had the most private alleles. The highest number of alleles (88) occurred for the Bowman Barrier population and the fewest (48) occurred in the Sumter National Forest II population site (Appendix). However, when the sample size effect was removed through rarefaction, the number of alleles/population ranged from 38-76 with Sumter National Forest I and Sumter National Forest II having the fewest and Bowman Barrier and BlackJacks Heritage Preserve having the most.

Levels of within population (Hep) genetic diversity ranged from 0.181 for the Sumter

National Forest I population to 0.462 for the Bowman Barrier population (Table 3.2).

Genetic structure among populations (Gst) was 0.175, indicating that 83% of the genetic variation is found within-populations (Table 3.3). Out of the 207 χ2 tests for departure from Hardy-Weinberg expectations, 34% were significant. The Fis value indicated slight overall excess heterozygosity across the loci for the populations (Table 3.3).

For the three large populations (SP, TP, BC), the mean number of alleles (A) = 2.64 and polymorphic alleles (APp) = 2.72. For the six small populations (BB, BJ, C, OS,

SNFI and II) (A) = 2.2 and (APp) = 2.58. The level of He (p = 0.553), and the means for

A (p = 0.203) , and Ap (p = 0.554) were not significantly different for the larger populations compared to the smaller populations (Table 3.4).

Nei‟s unbiased genetic identity between pairs of populations ranged from 0.6386

(populations BJ and SNF1) to 0.9692 (populations BC and SP; Table 3.5). Of the population sites, the Blackjacks Heritage Preserve population had the lowest mean genetic identity with the other population sites, while the Suther Prairie population had the highest (Table 3.6).

Allelic richness values among the populations were relatively consistent with their genetic diversity values, with the larger population sites having higher values compared to the smaller populations (Table 3.7).

The Mantel test showed no significant genetic isolation by geographic distance (r =

0.065; p =0.614). With the exception of some clustering of populations, the UPGMA phenogram did not indicate a strong relationship between physiographic location and genetic identity among the populations (Fig. 3.2).

Table 3.2. Genetic variation for the sampled Andropogon gerardii populations.

Pop N Pp APp Ap Ae Hop (SD) Hep (SD) BB 24 96.67 3.00 2.93 2.02 0.375 (0.090) 0.462 (0.028) BC 48 96.67 2.72 2.67 1.75 0.420 (0.063) 0.384 (0.031) BJ 24 88.46 3.43 3.15 2.06 0.340 (0.081) 0.450 (0.043) C 24 90.0 2.48 2.33 1.71 0.404 (0.083) 0.364 (0.035) OS 24 60.0 2.28 1.77 1.61 0.353 (0.080) 0.294 (0.044) SNFI 24 50.00 2.20 1.60 1.35 0.299 (0.039) 0.181 (0.043) SNFII 24 58.62 2.12 1.66 1.44 0.366 (0.058) 0.229 (0.042) SP 48 93.33 2.75 2.63 1.66 0.363 (0.060) 0.355 (0.033) TP 48 96.67 2.69 2.63 1.65 0.393 (0.058) 0.343 (0.035) Mean 32 81.16 2.63 2.38 1.69 0.368 0.340 SD 2.13 0.42 0.57 0.23 0.023 0.013 Ps APsp As Aes Hes Pooled 96.7 4.07 3.97 1.86 0.422

For within-populations, the mean percent polymorphic loci = Pp, the number of alleles per polymorphic locus = APp, the mean number of alleles per locus = Ap, the effective number of alleles per locus = Ae, the observed genetic diversity = Hop, and the expected genetic diversity = Hep. For the pooled species-wide sample, percent polymorphic loci = Ps, the number of alleles per polymorphic locus = APsp, the mean number of alleles per locus = As, the effective number of alleles per locus = Aes, and the genetic diversity = Hes.

Table 3.3. Statistics of genetic diversity for loci from the sampled A. gerardii populations

Ht = total genetic diversity in all populations, Hs = within-population genetic diversity, Gst= total among population genetic diversity, Fis = deviation from Hardy-Weinberg

Locus Ht Hs Gst Fis ADH2 0.2440 0.2178 0.1071 0.0652 PGI1 0.4952 0.4528 0.0858 -0.1774 PGI2 0.4035 0.3304 0.1812 -0.1937 PGI3A 0.3526 0.2346 0.3346 -0.1314 PGI3B 0.5607 0.3874 0.3091 0.4913 MNR5 0.2481 0.2142 0.1366 -0.1204 6P1 0.4878 0.4420 0.0937 -0.0641 6P2 0.2917 0.2582 0.1147 -0.0120 ACO1 0.5432 0.4672 0.1399 -0.1932 ACO2 0.4799 0.3292 0.3140 -0.0161 SKDH1 0.0000 0.0000 0.0000 0.0000 SKDH2 0.7170 0.6274 0.1250 0.1336 DIA1 0.5546 0.4455 0.1968 -0.3451 DIA2 0.5577 0.3332 0.4024 0.0986 PER2 0.3671 0.3409 0.0711 -0.3237 PGM1 0.1622 0.0977 0.3976 -0.1407 PGM2 0.4738 0.4046 0.1460 0.4760 AAT1 0.3170 0.2301 0.2741 0.2256 AAT2 0.4034 0.3312 0.1789 0.0667 FE1 0.5193 0.4113 0.2080 -0.1648 FE2 0.4772 0.4210 0.1177 -0.0880 FE4 0.5966 0.4248 0.2879 -0.0128 TPI1 0.6251 0.5656 0.0951 -0.4132 IDH1 0.3113 0.2871 0.0777 -0.2287 MDH1 0.4063 0.3538 0.1292 -0.4285 MDH2 0.1653 0.1509 0.0868 0.1809 UGPP1A 0.2857 0.2502 0.1243 0.0990 UGPP1B 0.5130 0.4765 0.0712 -0.1294 UGPP2 0.5752 0.4380 0.2384 0.0375 Avg. (Poly) 0.4363 0.3596 0.1748 -0.0700

Table 3.4. Mean levels of genetic diversity (He), total number of alleles per locus (A), and

number of alleles per polymorphic locus (APp) for small vs. large populations of

A. gerardii.

n He A APp

Small Populations 6 0.330 2.2 2.58

Large Populations 3 0.387 2.64 2.72

Table 3.5. Nei‟s Genetic Identity Values for sampled Andropogon gerardii

populations.

BC BJ C OS SNFI SNFII SP TP BB 0.9401 0.7562 0.9010 0.8349 0.8340 0.8617 0.9269 0.9065 BC 0.7310 0.9266 0.8874 0.8618 0.8974 0.9692 0.9233 BJ 0.7027 0.7165 0.6386 0.6707 0.7085 0.7167 C 0.8825 0.8760 0.8953 0.9147 0.8855 OS 0.8241 0.8334 0.9080 0.8791 SNFI 0.8725 0.8860 0.8709 SNFII 0.9093 0.8961 SP 0.9475

Table 3.6. Nei‟s mean genetic identity value of each A. gerardii population.

Site Mean Identity Value

BB 0.8701

BC 0.8921

BJ 0.7051

C 0.8730

OS 0.8457

SNFI 0.8329

SNFII 0.8545

SP 0.8962

TP 0.8782

Table 3.7. Summary of number of alleles and allelic richness per population. Allelic

richness is based on a rarefaction analysis where N = 12 individuals.

No. of Allelic Population alleles richness BB 74 71.28 BC 64 56.06 BJ 76 69.92 C 56 52.40 OS 42 41.49

SNF1 38 36.59 SNF2 41 40.18

SP 63 54.83 TP 64 54.65 Mean 57 60.66

Figure 3.2. UPGMA phenogram displaying levels of genetic identity between populations of Andropogon gerardii. The coefficient of similarity is indicated at the

bottom of the phenogram.

DISCUSSION

Results from this study indicated that local A. gerardii populations possess levels of genotypic diversity consistent with the findings of Keeler et al. (2002) in their study of A. gerardii genetic diversity within sampled plots at Konza Prairie, KS. They found 31.8 genetic individuals in an area of 100 m2. Results from this study suggest that local A. gerardii clonal size may be smaller (<1 m2) than the mean of 3.2m2 from sampled plots reported from Konza Prairie. Other studies have also indicated significant genetic variation within populations of other clonal species (Ellstrand and Roose 1987; Hamrick and Godt 1990).

Of the A. gerardii populations, the Blackjacks Heritage Preserve site was the most genetically distinct. However, the populations had high levels of genetic identity with each other and did not show significant isolation by distance. This suggests that these populations most likely had some degree of gene flow between them at one time.

Although there was no significant difference in the levels of genetic diversity within the small and large A. gerardii population sites, the mean values for A, APp, and He were all lower for the small versus larger populations. This indicates that smaller populations of

A. gerardii in the Carolinas may have lower genetic diversity when compared to larger populations. However, the overall genetic diversity for the A. gerardii populations in this study was higher than that reported for other plant species. Godt and Hamrick (1998) reported an overall mean genetic diversity (He) of 0.149 for plant species, and (He) of

0.191 for 21 surveyed grass species. However, the lower levels of genetic diversity for the six smaller A. gerardii populations in this study, when compared to larger

populations, was consistent with reported lower genetic diversity levels found for comparisons between large and small populations for other plant species (Godt and

Hamrick 1998).

The high relative within-population genetic diversity found for the populations in this study was consistent with reports of higher levels of genetic diversity within A. gerardii populations elsewhere (Keeler 1992; Gustafson et al. 1999; Selbo and Snow 2005), and for other reported out-crossing grasses (Godt and Hamrick 1998). Other out-crossing perennial grasses also have been reported to have similar levels of genetic variation (5-

15%) among populations (Huff et al. 1993; 1998; Huff 1997). Both Gustafson et al.

(1999) and Selbo and Snow (2005) reported higher levels of within-population genetic variation compared to among (89%; 84% respectively) from A. gerardii populations in

Arkansas and Ohio. However, Chappell (2003) found lower levels of genetic diversity within, compared to between, the Sumter National Forest I population and another small, local A. gerardii population. Chappell‟s study, however, involved a much smaller sample size (n=7), and used Principle Components Analysis (PCA) to compare the genetic diversity within and among the two populations. As a result, this may explain the apparent difference in the results of the Chappell study compared with those from this study (J. L. Hamrick pers. comm. 2011).

The higher than expected levels of genetic diversity for the A. gerardii populations in this study is most likely due to the clonal habit of this species. Even though these populations are most likely fragmented, their clonal habit has most likely allowed them to maintain higher within-population genetic diversity due to the long generational time associated with their clonal lifespan. It has been postulated that long-lived species may be

less likely to lose genetic variation within populations because individuals have longer periods to pass on alleles (Pleasants and Wendel 1989). It is possible that some of the genotypes found in ramets that occur within the local A. gerardii populations today are from early founding genotypes for those populations (Pleasants and Wendel 1989). As a result, high levels of within-population genetic structure have been maintained in these eastern populations, as has been suggested for Midwestern A. gerardii populations from similar studies (Keeler 2002).

Similar high levels of genetic diversity have been reported for the clonal species

Populus tremuloides (Quaking Aspen), which is believed to have reproduced primarily by vegetative means since having been initially established by seed some 8-10,000 years

BP (Cheliak and Dancik 1982). The high levels of genetic diversity for P. tremuloides indicates that there has been little genetic erosion in the long time period since its establishment, despite having relatively small populations (Pleasants and Wendel 1989).

The ability of species like A. gerardii to reproduce primarily via vegetative reproduction has likely allowed individuals in smaller populations to persist longer than they would by sexual reproduction alone. Thus, despite what has most likely been drastic reductions in population size of A. gerardii in the post-settlement period, these populations have been able to persist as fragmented populations while still maintaining relatively high levels of genetic diversity.

Other asexual or primarily asexual species, in addition to P. tremuloides, have been reported to have relatively high levels of genetic diversity as well (Pleasants and Wendel

1989). This includes clonal grass species such as Spartina patens (Salt Meadow Cord

Grass) (Pleasants and Wendel 1989), and in some lines of Panicum maximum (Guinea

100

Grass) (Usberti and Jain 1978). Coates (1988) also found high levels of genetic diversity in small isolated populations of the clonal Acacia anomala (Chittering Grass Wattle).

Results from this study suggest that there was once gene flow between these remnant local populations that would have maintained genetic diversity among populations. As a result, high levels of genetic diversity are still evident in these populations, although likely reduced, in some smaller populations.

Thus, although A. gerardii populations have most likely been severely fragmented, they still maintain high levels of genetic diversity. Even within smaller local populations of A. gerardii from this study, the data indicated comparatively higher levels of diversity than that reported for other non-clonal selfing species that have been subjected to severe reductions in overall population size (Raijmann et al. 1994; Demauro 1993).

Polyploidy is reported for populations of A. gerardii from Midwestern studies (Keeler et al 1992; Norrman et al. 1997). Most loci from populations in this study indicated banding patterns consistent with disomic inheritance (J.L. Hamrick pers. comm. 2011).

However, two loci displayed tetrasomic patterns of inheritance. The predominance of the pattern of disomic inheritance pattern may have been due to gene silencing at some loci.

Flow cytometric examination is needed to determine the ploidy levels in the eastern populations for comparison to those populations of the Midwestern U.S.

Although Keeler et al. (2002) suggested that the small scale genetic variation within the

A. gerardii population at Midwestern Konza Prairie indicated a high frequency of sexual reproduction, Carolina and Midwestern populations of A. gerardii have low rates of reported fertile seed production. In addition, due to the overall size of the Konza Prairie

A. gerardii population, there may be a greater likelihood of increased out-crossing

101

success, and thus potential viable seed production, when compared to the smaller populations of A. gerardii found in the Carolinas. Thus, despite their relatively high levels of genetic diversity, the smaller A. gerardii populations in the Carolinas may now be less likely to re-colonize by sexual reproduction in current sites or to colonize new habitats by seeds. Some populations may now very well be below the necessary threshold size for natural recovery on their own without the introduction of additional genotypes.

Further study is necessary to better understand the genetic structure of local A. gerardii populations and factors that may affect their conservation.

102

LITERATURE CITED

Chappell, J. 2003. Lack of caryopsis production in remnant big bluestem

(Andropogon gerardii) populations located in upstate South Carolina. M.S.

Thesis. Clemson University, Clemson, South Carolina.

Cheliak, W. M. and B. P. Dancik. 1982. Genetic diversity of natural populations of a

clone-forming tree Populus tremuloides. Canadian Journal of Genetics and

Cytology 24:611-616.

Cheliak, W. M. and J. A. Pitel. 1984. Techniques for starch gel electrophoresis of

enzymes from forest tree species. Petawawa National Forestry Institute,

Information Report P1-X-42, pp. 1-49. Canadian Forestry Service. Agriculture,

Chalk River, Ontario, Canada.

Coats, D. J. 1988. Genetic diversity and population genetic structure in the rare Chittering

Grass Wattle, Acacia anomala Court. Austaralian Journal of Botany 36:273-286.

Demauro, M. M. 1993. Relationship of breeding system to rarity in the Lakeside Daisy

(Hymenoxys auaulis var. glabra). Conservation Biology 7:542-550.

Ellstrand, N. C. and M. L. Roose. 1987. Patterns of genotypic diversity in clonal plant

species. American Journal of Botany. 74:123-131.

El Mousadik, A. and R. J. Petit. 1996. High level of genetic differentiation for allelic

richness among populations of the argan tree [Argania spinosa (L.) Skeels]

endemic to Morocco. Theoretical Applied Genetics 92:832-839.

Fischer, M. and D. Matthies. 1998. Effects of population size on performance in the rare

plant Gentianella germanica. Journal of Ecology 86:195-204.

103

Gilpin, M. E. and M. E. Soule. 1986. Minimum viable populations: Processes of species

extinction. In: Conservation Biology: the Science of Scarcity and Diversity (ed.

Soule, M.E.) pp. 19-34. Sinauer Associates, Sunderland.

Godt, M. J. W. and J. L. Hamrick. 1998. Allozyme diversity in grasses. pp. 11-29. In G.

P. Cheplick (ed.). Populations Biology of Grasses. Cambridge University Press.

Goudet, J. 2001. FSTAT, a program to estimate and test gene diversities and fixation

indices (version 2.9.3). Available from

http://www.unil.ch/izea/softwares/fstat.html.

Gustafson, D. J., D. J. Gibson, and D. L. Nickrent. 1999. Random amplified polymorphic

DNA (RAPD) variation among remnant big bluestem (Andropogon gerardii

Vitman) populations from Arkansas‟ Grand Prairie. Molecular Ecology 8:1693-

1701.

Hamrick, J. L. and M. J. W. Godt. 1990. Allozyme diversity in plant species, pp. 43-63.

In: A. D. H. Brown, M.T. Clegg, A. L. Kahler, and B.S. Weir (eds.) Plant

Population Genetics, Breeding and Genetic Resources.

Hamrick, J. L, Y. B. Linhart, and J. B. Mitton. 1979. Relationships between life history

characteristics and electrophoretically detectable genetic variation in plants.

Annual Review of Ecology and Systematics 10:173-200.

Heschel, M. S. and K. N. Paige. 1995. Inbreeding depression, environmental stress, and

population size variation in Scarlet Gilia (Ipomopsis aggregata). Conservation

Biology 9:126-133.

104

Huenneke, L. F. 1991. Ecological implications of genetic variation in plant populations.

In Genetics and Conservation of Rare Plants. eds. D. A. Falk and K. E. Holsinger.

Oxford University Press.

Huff, D. R. 1997. RAPD characterization of heterogeneous perennial rye grass cultivars.

Crop Science 37:557-564.

Huff, D. R., R. Peakall, and P. E. Smouse. 1993. RAPD variation within and among

natural population of outcrosssing buffalograss [Buchloe dactyloides (Nutt.)

Engelm.] Theoretical and Applied Genetics 86:927-934.

Huff, D. R., J. A. Quinn, B. Higgins, and A. J. Palazoo. 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. Molecular Ecology 7:1591-1597.

Hulbert, S. H. 1971. The nonconcept of species diversity; a critique and alternative

parameters. Ecology 52:577-586.

Karron, J. D. 1987. A comparison of levels of genetic polymorphism and self-

compatibility in geographically restricted and widespread plant congeners.

Evolutionary Ecology 1:47-58.

Keeler, K. H., C. F. Williams, and L. S. W. Vescio. 2002. Clone size of Andropogon

gerardii Vitman (Big Bluestem) at Konza Prairie, Kansas. American Midland

Naturalist 147:295-304.

Keeler, K. H. 1992. Local polyploidy variation in the native prairie grass Andropogon

gerardii. American Journal of Botany 79:1229-1232.

105

Kery, M., D. Matthies, and H. Spillmann. 2000. Reduced fecundity and offspring

performance in small populations of the declining grassland plants Primula veris

and Gentiana lutea. Journal of Ecology 88:17-30.

Loveless, M. D. and J. L. Hamrick. 1984. Ecological determinants of genetic structure in

plant populations. Annual Review of Ecology and Systematics. 15:65-95.

Manchenko, G. P. 1994. Handbook of Detection of Enzymes on Electrphorectic Gels.

CRC Press, Ann Arbor, Michigan.

McKone, M. J., C. P. Lund and J. M. O‟Brien. 1998. Reproductive biology of two

dominant prairie grasses (Andropogon gerardii and Sorghastrum nutans, Poaceae):

Male-biased sex allocation in wind-pollinated plants? American Journal of Botany

85:776-783.

Nei, M. 1972. Genetic distance between populations. American Naturalist 106:283-292.

Norrmann, G. A., C. L. Quarin, and K. H. Keeler. 1997. Evolutionary implications of

meiotic chromosome behavior, reproductive biology, and hybridization in 6x and

9x cytotypes of Andropogon gerardii (Poaceae). American Journal of Botany

84:201-207.

Oostermeijer, J. G. B., M. W. van Eijck, and J. C. M. den Nijs. 1994. Offspring fitness in

relation to population size and genetic variation in the rare perennial plant species

Gentiana pneumonanthe (Gentianaceae). Oecologia 97:289-296.

Petit, R. J., A. El Mousadik, and O. Pons 1988. Identifying populations for conservation

on the basis of genetic markers. Conservation Biology 12:844-855.

106

Pleasants, J. M. and J. F. Wendel. 1989. Genetic diversity in a clonal narrow endemic,

Erthronium propullans, and in its widespread progenitor, Erythronium albidum.

American Journal of Botany 76:1136-1151.

Radford, A. E., H. E. Ahles, and C. R. Bell. 1968. Manual of the Vascular Flora of the

Carolinas. University of North Carolina Press. Chapel Hill, North Carolina.

Raijmann, L. E. L., N. C. Van Leeuwen, R. Kersten, J. G. B. Oostermeijer, H. C. M. D.

Nijs, and S. B. J. Menken. 1994. Genetic variation and outcrossing rate in relation

to population size in Gentiana pneumonanthe L. Conservation Biology 8:1014-

1026.

Rohlf, F. J. 2003. Numerical taxonomy and multivariate analysis system version 2.11j.

Exeter Software, Setauket, New York.

Selbo, S. M. and A. A. Snow. 2005. Flowering phenology and genetic similarity among

local and recently introduced populations of Andropogon gerardii in Ohio.

Restoration Ecology 13:441-447.

Smouse, P. E, J. C. Long and R. R. Sokal. 1986. Multiple regression and correlation

extensions of the Mantel test of matrix correspondence. Systematic Zoology

35:627-632.

Soltis, D. E., C. H. Haufler, D. C. Darrow, and G. J. Gastony. 1983. Starch gel

electrophoresis of ferns: A compilation of grinding buffers, gel and electrode

buffers, and staining schedules. American Fern Journal 73:9-27.

Usberti, Jr. J. A. and S. K. Jain. 1978. Variation in Panicum maximum: A comparison of

sexual and asexual populations. Botanical Gazette 139:112-116.

107

Van Treuren, R., R. Bijlsma, W. Van. Delden, and N. J. Ouborg. 1990. The

significance of genetic erosion in the process of extinction. I. Genetic

differentiation in Salvia pratensis and Scabiosa columbaria in relation to

population size. Heredity 66:181-189.

Weeden, N. F. and J. F. Wendel. 1989. Genetics of plant isozymes. pp. 46-72, In

Isozymes in Plant Biology. D. E. and P. S. Sotis (eds.).

Wendel, J. F. and C. R. Parks. 1982. Genetic control of isozyme variation Camellia

japonica L. Journal of Heredity 73:197-204.

Wright, S. 1922. Coefficents of inbreeding and relationship. American Naturalist 56:330-

338.

Wright, S. 1951. The genetical structure of populations. Annals of Eugenics 15:323-354.

Young, A. T. Boyle and T. Brown. 1996. The population genetic consequences of habitat

fragmentation for plants. Trends in Ecology and Evolution 11:413-418.

108

APPENDIX. LOCI, GENE FREQUENCIES AND SAMPLE

SIZES.

GENE FREQUENCIES AND SAMPLE SIZES ADH2 PGI1 PGI2 PGI3 PGI3B MNR5 6P1 6P2 ACO1 BB 1 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 2 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.1250 0.0000 0.0000 3 0.0833 0.7917 0.6667 0.2917 0.1667 0.0417 0.1458 0.1250 0.4792 4 0.8542 0.2083 0.3333 0.7083 0.6667 0.8542 0.7292 0.8125 0.0833 5 0.0625 0.0000 0.0000 0.0000 0.1667 0.1042 0.0000 0.0625 0.4375 6 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 7 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 8 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 N 24 24 24 24 24 24 24 24 24 #A 3 2 2 2 3 3 3 3 3 BC 1 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 2 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0729 0.0000 0.0000 3 0.1979 0.6458 0.6042 0.4375 0.0000 0.0625 0.1042 0.2813 0.6146 4 0.7917 0.3542 0.3958 0.5625 0.4688 0.8021 0.7604 0.7188 0.0729 5 0.0000 0.0000 0.0000 0.0000 0.5313 0.1354 0.0625 0.0000 0.3125 6 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 7 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 8 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 N 48 48 48 48 48 48 48 48 48 #A 3 2 2 2 2 3 4 2 3 BJ 1 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 2 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.2708 0.0000 0.0000 3 0.1667 0.5000 0.3958 0.7500 0.6522 0.4167 0.3542 0.1458 0.0208 4 0.5625 0.5000 0.6042 0.2292 0.0000 0.5208 0.3750 0.7708 0.0000 5 0.2708 0.0000 0.0000 0.0208 0.3478 0.0208 0.0000 0.0833 0.9792 6 0.0000 0.0000 0.0000 0.0000 0.0000 0.0417 0.0000 0.0000 0.0000 7 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 8 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 N 24 24 24 24 23 24 24 24 24 #A 3 2 2 3 2 4 3 3 2 C 1 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 2 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 3 0.0417 0.6042 1.0000 0.3333 0.0000 0.0000 0.1875 0.0833 0.7500 4 0.9583 0.3958 0.0000 0.6667 0.5208 0.8125 0.7917 0.9167 0.0208 5 0.0000 0.0000 0.0000 0.0000 0.4792 0.1875 0.0208 0.0000 0.2292 6 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 7 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 8 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 N 24 24 24 24 24 24 24 24 24 #A 2 2 1 2 2 2 3 2 3 OS 1 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 2 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.1250 0.0000 0.0000 3 0.0000 0.2083 0.6190 0.0000 0.0000 0.1667 0.2917 0.0000 0.2500 4 1.0000 0.7917 0.3810 1.0000 1.0000 0.8333 0.3958 1.0000 0.2500 5 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.1875 0.0000 0.5000 6 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 7 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 8 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 N 12 24 21 12 12 12 24 24 24

109

#A 1 2 2 1 1 2 4 1 3 SNF1 1 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 2 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 3 0.0000 0.5208 1.0000 0.0000 0.0000 0.0000 0.0625 0.0000 0.5000 4 1.0000 0.4792 0.0000 1.0000 1.0000 1.0000 0.9375 1.0000 0.0000 5 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.5000 6 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 7 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 8 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 N 24 24 24 24 24 24 24 24 24 #A 1 2 1 1 1 1 2 1 2 SNF2 1 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 2 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 3 0.2174 0.3261 0.5870 0.0000 0.0000 0.0000 0.1957 0.1087 0.5000 4 0.7826 0.6739 0.4130 1.0000 0.0000 1.0000 0.8043 0.8913 0.0000 5 0.0000 0.0000 0.0000 0.0000 1.0000 0.0000 0.0000 0.0000 0.5000 6 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 7 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 8 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 N 23 23 23 23 23 23 23 23 23 #A 2 2 2 1 1 1 2 2 2 SP 1 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 2 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 3 0.0729 0.6042 0.6354 0.0000 0.0000 0.0000 0.3125 0.3750 0.5000 4 0.8958 0.3958 0.3438 0.9063 0.6042 0.9688 0.6667 0.6250 0.0104 5 0.0313 0.0000 0.0000 0.0938 0.3750 0.0313 0.0208 0.0000 0.4896 6 0.0000 0.0000 0.0000 0.0000 0.0208 0.0000 0.0000 0.0000 0.0000 7 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 8 0.0000 0.0000 0.0208 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 N 48 48 48 48 48 48 48 48 48 #A 3 2 3 2 3 2 3 2 3 TP 1 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 2 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0313 0.0000 0.0000 3 0.0000 0.5625 0.9479 0.0000 0.0000 0.0104 0.3750 0.0833 0.3021 4 0.9479 0.4375 0.0521 0.9688 0.5833 0.8854 0.5938 0.9167 0.0313 5 0.0521 0.0000 0.0000 0.0104 0.3958 0.1042 0.0000 0.0000 0.6667 6 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 7 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 8 0.0000 0.0000 0.0000 0.0208 0.0208 0.0000 0.0000 0.0000 0.0000 N 48 48 48 48 48 48 48 48 48 #A 2 2 2 3 3 3 3 2 3

GENE FREQUENCIES AND SAMPLE SIZES ACO2 SKDH1 SKDH2 DIA1 DIA2 PER2 PGM1 PGM2 AAT1 BB 1 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 2 0.0000 0.0000 0.1042 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 3 0.1667 0.0000 0.3333 0.2708 0.1087 0.0000 0.1667 0.0208 0.0833 4 0.3542 1.0000 0.1458 0.5417 0.4130 0.7609 0.7917 0.4583 0.7708 5 0.3333 0.0000 0.4167 0.1875 0.4783 0.2391 0.0000 0.4792 0.1458 6 0.1458 0.0000 0.0000 0.0000 0.0000 0.0000 0.0417 0.0417 0.0000 7 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 8 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 N 24 24 24 24 23 23 24 24 24 #A 4 1 4 3 3 2 3 4 3

110

BC 1 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 2 0.0000 0.0000 0.0208 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 3 0.3229 0.0000 0.2917 0.0000 0.0938 0.0000 0.0000 0.0000 0.0729 4 0.6771 1.0000 0.1771 0.5729 0.8854 0.8021 0.9896 0.6087 0.9271 5 0.0000 0.0000 0.5000 0.4271 0.0208 0.1979 0.0000 0.3370 0.0000 6 0.0000 0.0000 0.0104 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 7 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 8 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0104 0.0543 0.0000 N 48 48 48 48 48 48 48 46 48 #A 2 1 5 2 3 3 2 3 2 BJ 1 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0417 2 0.0000 0.0000 0.1250 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 3 0.0000 0.0000 0.5000 0.0000 0.1957 0.0000 0.5000 0.0455 0.1667 4 0.1458 1.0000 0.2500 0.0000 0.4783 0.0000 0.2708 0.3864 0.2083 5 0.4167 0.0000 0.1042 1.0000 0.1957 0.0000 0.0000 0.2500 0.5833 6 0.3958 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 7 0.0417 0.0000 0.0208 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 8 0.0000 0.0000 0.0000 0.0000 0.1304 0.0000 0.2292 0.3182 0.0000 N 24 24 24 16 23 0 24 22 24 #A 4 1 5 1 4 0 3 4 4 C 1 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 2 0.0000 0.0000 0.0208 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 3 0.5000 0.0000 0.5417 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 4 0.5000 1.0000 0.0833 0.0625 0.2917 0.6458 1.0000 0.7083 0.9375 5 0.0000 0.0000 0.3542 0.5625 0.7083 0.3542 0.0000 0.2500 0.0625 6 0.0000 0.0000 0.0000 0.3750 0.0000 0.0000 0.0000 0.0000 0.0000 7 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 8 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0417 0.0000 N 24 24 24 24 24 24 24 24 24 #A 2 1 4 3 2 2 1 3 2 OS 1 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 2 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 3 0.6250 0.0000 0.6250 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 4 0.3750 1.0000 0.1875 0.0000 1.0000 0.9583 1.0000 0.6667 1.0000 5 0.0000 0.0000 0.1875 1.0000 0.0000 0.0417 0.0000 0.0000 0.0000 6 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.3333 0.0000 7 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 8 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 N 24 24 24 12 12 12 24 24 12 #A 2 1 3 1 1 2 1 2 1 SNF1 1 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 2 0.0000 0.0000 0.0417 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 3 0.0000 0.0000 0.2292 0.0000 0.0000 0.0000 0.0000 0.0000 0.0208 4 1.0000 1.0000 0.0833 0.5000 0.0000 0.5417 1.0000 1.0000 0.9792 5 0.0000 0.0000 0.6458 0.5000 1.0000 0.4583 0.0000 0.0000 0.0000 6 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 7 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 8 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 N 24 24 24 24 24 24 24 24 24 #A 1 1 4 2 1 2 1 1 2

111

SNF2 1 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 2 0.0000 0.0000 0.0435 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 3 0.0000 0.0000 0.2391 0.0000 0.0000 0.0000 0.0000 0.0000 0.0652 4 1.0000 1.0000 0.3261 0.5000 0.0000 0.9348 1.0000 1.0000 0.9348 5 0.0000 0.0000 0.3913 0.5000 1.0000 0.0652 0.0000 0.0000 0.0000 6 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 7 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 8 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 N 23 23 23 23 23 23 23 23 23 #A 1 1 4 2 1 2 1 1 2 SP 1 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 2 0.0000 0.0000 0.1042 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 3 0.1354 0.0000 0.4063 0.0000 0.1354 0.0000 0.0000 0.0000 0.0417 4 0.8333 1.0000 0.0729 0.6667 0.7604 0.7083 1.0000 0.6250 0.9583 5 0.0313 0.0000 0.4167 0.3333 0.1042 0.2813 0.0000 0.3646 0.0000 6 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0104 0.0000 7 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 8 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 N 48 48 48 48 48 48 48 48 48 #A 3 1 4 2 3 3 1 3 2 TP 1 0.0000 0.0000 0.0417 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 2 0.0000 0.0000 0.4792 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 3 0.0000 0.0000 0.2083 0.0000 0.0000 0.0000 0.0417 0.0000 0.3125 4 0.9583 1.0000 0.2604 0.5833 0.5938 0.8333 0.9583 0.7396 0.6458 5 0.0417 0.0000 0.0104 0.4167 0.4063 0.1667 0.0000 0.2604 0.0417 6 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 7 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 8 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 N 48 48 48 48 48 24 48 48 48 #A 2 1 5 2 2 2 2 2 3

GENE FREQUENCIES AND SAMPLE SIZES AAT2 FE1 FE2 FE4 TPI1 TPI2 IDH1 MDH1 MDH2 BB 1 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 2 0.0000 0.1667 0.0000 0.0417 0.0000 0.0000 0.0208 0.0000 0.0000 3 0.1667 0.0625 0.5217 0.1875 0.2083 0.0000 0.3542 0.0000 0.0000 4 0.7083 0.7708 0.4783 0.6458 0.3750 0.5417 0.6250 0.6250 0.8333 5 0.1250 0.0000 0.0000 0.1250 0.3542 0.4583 0.0000 0.1667 0.1667 6 0.0000 0.0000 0.0000 0.0000 0.0625 0.0000 0.0000 0.2083 0.0000 7 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 8 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 N 24 24 23 24 24 24 24 24 24 #A 3 3 2 4 4 2 3 3 2 BC 1 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 2 0.0000 0.0000 0.0000 0.0104 0.0000 0.0000 0.0000 0.0000 0.0000 3 0.1146 0.4063 0.5938 0.2083 0.1354 0.0000 0.2396 0.0000 0.0000 4 0.8854 0.5938 0.3750 0.4896 0.4271 0.5625 0.7500 0.9167 0.8750 5 0.0000 0.0000 0.0313 0.2917 0.4375 0.4375 0.0104 0.0729 0.1250 6 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0104 0.0000 7 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 8 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 N 48 48 48 48 48 24 48 48 48 #A 2 2 3 4 3 2 3 3 2

112

BJ 1 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 2 0.0417 0.0000 0.0000 0.0000 0.0208 0.0417 0.0208 0.0000 0.0000 3 0.1458 0.0000 0.0000 0.0000 0.6458 0.1250 0.2292 0.0000 0.0000 4 0.3333 0.0000 0.0000 0.0000 0.1042 0.6250 0.6667 0.7083 1.0000 5 0.0000 0.0000 0.0000 0.0000 0.1250 0.1667 0.0208 0.2708 0.0000 6 0.4792 0.0000 0.0000 0.0000 0.1042 0.0417 0.0625 0.0208 0.0000 7 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 8 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 N 24 0 0 0 24 24 24 24 24 #A 4 0 0 0 5 5 5 3 1 C 1 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 2 0.0000 0.0000 0.0000 0.0625 0.0000 0.0000 0.0000 0.0000 0.0000 3 0.2500 0.7708 0.4583 0.1250 0.1875 0.0000 0.2083 0.0000 0.0000 4 0.7500 0.2292 0.1458 0.5417 0.3750 0.5000 0.7917 0.9167 0.8958 5 0.0000 0.0000 0.3958 0.2708 0.4375 0.5000 0.0000 0.0208 0.1042 6 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0625 0.0000 7 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 8 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 N 24 24 24 24 24 24 24 24 24 #A 2 2 3 4 3 2 2 3 2 OS 1 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 2 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 3 0.2917 0.6458 0.6875 0.5417 0.0000 0.0000 0.1250 0.0000 0.0000 4 0.7083 0.3542 0.0000 0.4583 0.7500 0.5000 0.8750 0.7500 1.0000 5 0.0000 0.0000 0.3125 0.0000 0.2500 0.5000 0.0000 0.2500 0.0000 6 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 7 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 8 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 N 12 24 24 12 12 12 12 12 12 #A 2 2 2 2 2 2 2 2 1 SNF1 1 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 2 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 3 0.0000 0.9583 0.9583 0.0000 0.0417 0.0000 0.0000 0.0000 0.0000 4 0.5417 0.0417 0.0417 0.0000 0.5000 0.5000 1.0000 0.5625 1.0000 5 0.4583 0.0000 0.0000 1.0000 0.4583 0.5000 0.0000 0.4375 0.0000 6 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 7 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 8 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 N 24 24 24 24 24 24 24 24 24 #A 2 2 2 1 3 2 1 2 1 SNF2 1 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 2 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 3 0.0000 0.8261 0.8478 0.0000 0.0000 0.0000 0.3043 0.0000 0.0000 4 1.0000 0.1739 0.1522 1.0000 0.5682 0.0000 0.6957 0.5000 1.0000 5 0.0000 0.0000 0.0000 0.0000 0.4318 0.0000 0.0000 0.0000 0.0000 6 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.5000 0.0000 7 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 8 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 N 23 23 23 23 22 0 23 23 23 #A 1 2 2 1 2 0 2 2 1

113

SP 1 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 2 0.0104 0.0000 0.0104 0.0104 0.0000 0.0000 0.0000 0.0000 0.0000 3 0.1146 0.6146 0.6458 0.1042 0.0208 0.0000 0.1354 0.0000 0.0000 4 0.8750 0.3854 0.3229 0.7604 0.5938 0.5104 0.8646 0.6875 0.9688 5 0.0000 0.0000 0.0208 0.1250 0.3646 0.4896 0.0000 0.2188 0.0313 6 0.0000 0.0000 0.0000 0.0000 0.0208 0.0000 0.0000 0.0938 0.0000 7 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 8 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 N 48 48 48 48 48 48 48 48 48 #A 3 2 4 4 4 2 2 3 2 TP 1 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 2 0.0729 0.0313 0.0000 0.3854 0.0000 0.0000 0.0000 0.0000 0.0000 3 0.1667 0.3438 0.7396 0.0208 0.0729 0.0000 0.0625 0.0000 0.0000 4 0.7604 0.5938 0.2500 0.5938 0.5000 0.5417 0.9375 0.8438 0.7708 5 0.0000 0.0313 0.0104 0.0000 0.3958 0.4583 0.0000 0.1563 0.2292 6 0.0000 0.0000 0.0000 0.0000 0.0313 0.0000 0.0000 0.0000 0.0000 7 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 8 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 N 48 48 48 48 48 48 48 48 48 #A 3 4 3 3 4 2 2 2 2

GENE FREQUENCIES AND SAMPLE SIZES UGPP1A UGPP1B UGPP2 BB 1 0.0000 0.0000 0.0000 2 0.1458 0.0000 0.0000 3 0.2500 0.0833 0.0000 4 0.5625 0.4375 0.1667 5 0.0417 0.3750 0.6458 6 0.0000 0.1042 0.1875 7 0.0000 0.0000 0.0000 8 0.0000 0.0000 0.0000 N 24 24 24 #A 4 4 3 BC 1 0.0000 0.0000 0.0000 2 0.0104 0.0000 0.0000 3 0.1250 0.0000 0.0208 4 0.8542 0.6250 0.4583 5 0.0104 0.3750 0.5000 6 0.0000 0.0000 0.0208 7 0.0000 0.0000 0.0000 8 0.0000 0.0000 0.0000 N 48 48 48 #A 4 2 4 BJ 1 0.0000 0.0000 0.0000 2 0.0000 0.0000 0.0000 3 0.0833 0.0000 0.0000 4 0.7917 0.7917 0.1458 5 0.1250 0.1667 0.0417 6 0.0000 0.0417 0.8125 7 0.0000 0.0000 0.0000 8 0.0000 0.0000 0.0000 N 24 24 24 #A 3 3 3

114

C 1 0.0000 0.0000 0.0000 2 0.0208 0.0000 0.0000 3 0.3958 0.0000 0.0000 4 0.5833 0.6250 0.3125 5 0.0000 0.3125 0.6875 6 0.0000 0.0625 0.0000 7 0.0000 0.0000 0.0000 8 0.0000 0.0000 0.0000 N 24 24 24 #A 3 3 2 OS 1 0.0000 0.0000 0.0000 2 0.0000 0.0000 0.0000 3 0.0000 0.0000 0.0000 4 1.0000 1.0000 0.5833 5 0.0000 0.0000 0.3958 6 0.0000 0.0000 0.0208 7 0.0000 0.0000 0.0000 8 0.0000 0.0000 0.0000 N 12 12 24 #A 1 1 3 SNF1 1 0.0000 0.0000 0.0000 2 0.0000 0.0000 0.0000 3 0.0000 0.0000 0.0000 4 1.0000 0.5000 0.9583 5 0.0000 0.5000 0.0417 6 0.0000 0.0000 0.0000 7 0.0000 0.0000 0.0000 8 0.0000 0.0000 0.0000 N 24 24 24 #A 1 2 2 SNF2 1 0.0000 0.0000 0.0000 2 0.0000 0.0000 0.0000 3 0.0000 0.0000 0.0000 4 1.0000 0.5000 0.5000 5 0.0000 0.5000 0.5000 6 0.0000 0.0000 0.0000 7 0.0000 0.0000 0.0000 8 0.0000 0.0000 0.0000 N 23 23 23 #A 1 2 2 SP 1 0.0000 0.0000 0.0000 2 0.0208 0.0000 0.0000 3 0.1563 0.0000 0.0000 4 0.8229 0.6563 0.6250 5 0.0000 0.1771 0.3750 6 0.0000 0.1667 0.0000 7 0.0000 0.0000 0.0000 8 0.0000 0.0000 0.0000 N 48 48 48 #A 3 3 2

115

TP 1 0.0000 0.0000 0.0000 2 0.0625 0.0208 0.0000 3 0.0208 0.0104 0.0000 4 0.9167 0.5833 0.6875 5 0.0000 0.3854 0.2813 6 0.0000 0.0000 0.0313 7 0.0000 0.0000 0.0000 8 0.0000 0.0000 0.0000 N 48 48 48 #A 3 4 3

116

Chapter 4

Suther Prairie: Vascular Flora, Species Richness, and

Edaphic Factors – A Unique Eastern Prairie

117

ABSTRACT

Piedmont prairie communities of the southeastern United States were common prior to

European settlement. Suther Prairie in Cabarrus County, North Carolina is among the best-known examples of such an ecosystem. This study provides a complete floristic list of species observed at the site from 1997-2007. Sampling over the ten-year period included fixed transects and general floristic inventories. During the 2006-07 growing seasons, additional transects and 90 randomly placed 1m x 1m quadrats were established for sampling species frequency. Soil in six of these random quadrats was sampled at three depths (0-10 cm, 11-20 cm, 21-30 cm) for pH, organic C, total N, and extractable P, K,

Ca, Mg, Zn, Mn, Cu, B, and Na.

There were 208 species documented during the 10 year sampling period. In 2006-07

142 new species were identified, but 49 previously documented species were not re- located. Of the 208 species, 66 were graminoids (36 grasses and 30 sedges/rushes).

Ninety of these had previously been reported from prairie habitats. Obligate or facultative wetland species comprised 32% of the list. Thirteen species were rare, watch-listed, or uncommon for North Carolina. Big Bluestem (Andropogon gerardii) and Eastern

Gamagrass (Tripsacum dactyloides var. dactyloides) had the highest frequency of occurrence during the 2006-07 sampling. Species richness was 6.8 species per m2. Levels of Ca and Mg were much higher than normal for Piedmont soils, and there was significant variation in soil C, N, P and Zn levels among the sampled depths.

118

INTRODUCTION

Early European explorers to the Piedmont reported openings in the forests, which they referred to as prairies or savannas (Barden 1997; Logan 1859). These open sites were primarily maintained through the use of fire by Native Americans to clear land for agriculture or to improve habitat for game. Although anthropogenic fires were likely the predominant factor, lightning-caused fires also played a role in maintaining prairie-like conditions (Barden 1997). Beginning with European settlement, cultivation practices and fire suppression have led to the nearly complete disappearance of Piedmont prairies

(Barden 1997).

Piedmont prairies may have little in common with prairie habitat in other geographic areas. Indeed, the term prairie is a loosely-defined concept that encompasses a broad range of habitat characteristics under the umbrella of disturbance and the lack of a well- developed tree canopy or shrub layer. Soils, in particular, exhibit a broad array of characteristics among prairies in different geographic locations. While Midwestern prairies often have deep, fertile soils, those of Piedmont prairies often have shallow, infertile soils with high shrink/swell capacities and wide seasonal variation in available water. Additionally, Piedmont prairies are limited in size by a number of factors including the expanse of the underlying soil and the frequency of wooded riparian areas, which serve to break up the open canopy. In contrast, Midwestern prairies can be quite expansive. Piedmont prairies fall under the Xeric Hardpan Forest community classification and also the Piedmont Basic Hardpan Woodland – Prairie Barren Subtype

(NatureServe 2009; Schafale and Weakley 1990).

119

Presently in the eastern U.S., plant species adapted to prairie conditions are primarily relegated to roadside margins and utility rights-of-way. These relict habitats often support unique floral assemblages and rare species (Davis et al. 2002). These include the occurrence of dominant grassland species such as Indiangrass (Sorghastrum nutans),

Purpletop (Tridens flavus), Broomsedge (Andropogon virginicus var. virginicus), Eastern

Gamagrass (Tripsacum dactyloides var. dactyloides), and various Panic grasses

(Panicum/Dichanthelium spp.) They also often contain large numbers of both Asteraceae and Fabaceae species (Davis et al. 2002).

Suther Prairie, located in Cabarrus County, North Carolina, is the best-known surviving example of a Piedmont prairie site (L. Barden pers. comm. 2008). It has been owned by the Suther family since the 1700s. This prairie is an unusual wet prairie that exhibits hydric soil conditions over most of its area (Davis et al. 2002). Unlike most remnant

Piedmont prairies such as Mineral Springs Preserve in Union County, N.C. and the

Blackjacks Heritage Preserve in York County, S.C. (Schmidt and Barnwell 2002), which are adjacent to roads, railroads, or utility rights-of-way, Suther Prairie is located in the interior of a large Piedmont farm and is protected by a wide surrounding buffer of undeveloped farmland. Additionally, Suther Prairie is located in an alluvial area with soils that remain consistently wet throughout the year. In contrast, Mineral Springs

Preserve and the Blackjacks Heritage Preserve each exhibit greater seasonal fluctuations in available soil moisture, likely a result of their non-alluvial locations. While the open canopy at Suther Prairie has been maintained by annual mowing, Mineral Springs

Preserve and the Blackjacks Heritage Preserve have most likely remained open as a result of stress to tree roots related to the seasonal fluctuations in soil moisture and the high

120

shrink-swell properties of the soils at those locations (Schmidt and Barnwell 2002;

McRae 1999).

Currently, Suther Prairie is comprised of a 2 ha section that has been managed for more than 200 years by mowing, and a 0.8 ha section that was cultivated until the mid 1970s and then converted to fescue pasture. These two sections were previously separated by a hedgerow, which was removed in 2004 to connect the formerly cultivated portion with the adjacent prairie remnant (L. Barden pers. comm. 2008), thereby providing additional suitable habitat for the prairie species found in the prairie remnant (Estep and McRae

1997). The larger remnant portion has probably never been plowed because of the extremely hydric condition of the soil during the growing season; rather, it has been historically mown for hay once or twice a year. Recently, in addition to regular mowing for hay, Suther Prairie has been managed with prescribed burns approximately every 2 to

3 years (Louis Suther pers. comm. 2008). The remnant portion includes numerous regionally significant species and has been used as a seed source by a number of conservation organizations and one commercial seed company.

This study provides a complete floristic inventory obtained from the 2006-07 growing seasons, combined with unpublished data from voucher specimens collected over a 10 year period from the site. This data will help assess species richness at the site, allowing for a comparison of richness to similar Piedmont sites. Data collected will also provide a baseline for identifying potential species compositional changes over time in response to current management practices. Frequency values obtained from the 2006-07 portion of the inventory, in combination with data collected by Lee Lehman for graduate research at

UNC Charlotte and by the Habitat Assessment and Restoration Program (HARP), can be

121

used to determine predominant and less common species and to detect future changes in community species structure. Moreover, this study provides an edaphic characterization of soils which may be useful for additional studies of edaphic association with species‟ occurrence. Both floristic and soil data may also be helpful in the on-going restoration and/or maintenance of Suther Prairie and other potential eastern prairie sites.

MATERIALS AND METHODS

Study Area

Suther Prairie is located on private property (35˚27‟026”N, 80˚28‟057”W) about 8 km northeast of Concord, North Carolina (Fig. 4.1). It lies along the floodplain of Dutch

Buffalo Creek and remains wet for much of the year, with the notable exception of the

2007 growing season, which was exceptionally dry. The soil mapping unit for the site is the Chastain series (Cabarrus Soil and Water Conservation pers. comm. 2008). The area is approximately 2.8 ha with a maximum length of 175 m and a maximum width of 145 m. The elevation along the margins is slightly higher than in the center, creating a poorly drained central depression. It is encircled by early successional woodlands on the mesic border area (Fig. 4.2).

Botanical Sampling

A floristic inventory of the site has been ongoing since 1997. Sampling has included periodic visits to update a maintained species list, and has also included establishing transects for annual sampling events. During both early- and late-seasons of 2006-07, additional temporary transects were established at points along both the greatest width and length of the site. At 10 m intervals along each transect, a 1x1 m quadrat was randomly placed on the left or right side of the transect line, with a total of 90 quadrats.

122

Figure. 4.1. Location of Suther Prairie, Cabarrus County, North Carolina.

123

Figure 4.2. View of Suther Prairie in the spring growing season.

(photo courtesy of the Suther family)

124

All vascular species within each quadrat were identified. Species frequency was determined as the percentage of quadrats in which a species was noted. In addition, the site was visited every 3 to 4 weeks during the growing season to find species not noted within the sampled quadrats. Voucher specimens were stored in the herbaria of Clemson

University, Belmont Abbey College, and Mecklenburg County (formerly the UNC

Charlotte Herbarium). Botanical nomenclature follows Weakley (2007).

Soil Analysis

Soil samples were collected from six quadrats located at random on the site. Within each of the sampled quadrats, soil samples were taken at three depths (0-10 cm, 11-20 cm, and 21-30 cm). The samples were analyzed at the Clemson University Agricultural

Service Laboratory (http://www.clemson.edu/agsrvlb/) for pH, organic carbon, total nitrogen, extractable phosphorus, potassium, calcium, magnesium, zinc, manganese, copper, boron, and sodium. Mineral analyses were performed using Mehlich No. 1 extraction solution and element quantification by inductively coupled plasma atomic emission spectroscopy (ICP-AES) (Jones 2001). Total carbon (C) and nitrogen (N), equivalent to organic C and N for these particular soil samples, were measured on 200 mg samples using a combustion analyzer (Elementar Vario Macro, Mt. Laurel, NJ). To test the null hypothesis of no nutrient difference among depths, a one-way ANOVA of the soil chemical data was performed using PROC GLM with six observations per sample depth. Post hoc differences in means were determined using Tukey‟s Test. All statistical analyses were performed with SAS v. 9.1 (SAS Institute 2002).

125

RESULTS

Flora

A total of 208 vascular species (Table 4.1; Appendix I) were documented over 10 years of sampling. Of those, 77 were monocots, comprising 41 genera in 11 families, and 130 were dicots in 102 genera, representing 48 families. There was also one gymnosperm.

Twenty-three species (11%) were obligate wetland (OBL); 43 (21%) were facultative wetland (FACW, FACW+ or FACW-), 61 (29%) were facultative (FAC, FAC+ or FAC-

), 41 (19%) were facultative upland (FACU, FACU+ or FACU-), 12 (5%) were upland

(UPL), and 28 (14%) were undesignated (Kartesz 1996; Table 4.1). Sixty-six graminoid species were found, accounting for 31% of the total species. Among these were 36

Poaceae (17%), 24 Cyperaceae (11%), and 6 Juncaceae (3%) (Table 4.1). In addition, the

Asteraceae accounted for 29 species (13%) and the Fabaceae for 17 species (8%).

One-hundred and fifty-nine species were collected during the 2006-07 growing seasons, of which 142 were newly documented for the site. Forty-nine of the previously documented species were not re-located during the 2006-07 surveys. In addition, 36 undocumented species (lacking voucher specimens) (Appendix II) were also not located during the 2006-07 surveys.

Of the species collected, 13 were rare, watch-listed, or uncommon for either North

Carolina or the Piedmont region (Weakley 2007; North Carolina Natural Heritage

Program 2006). These include Slender Sedge (Carex tenera var. tenera) (N.C. Watch

List), Bush‟s Sedge (Carex bushii) (N.C. Rare), Rufous Bulrush (Scirpus pendulus) (N.C.

Rare), Three-angled Spikerush (Eleocharis tricostata) (N.C. Watch List), Redroot

Flatsedge (Cyperus erythrorhizos), Angle-stem Beaksedge (Rhynchospora caduca), Bog

126

Table 4.1. Species richness of vascular plants from Suther Prairie, Cabarrus County,

North Carolina in relation to life form and wetland status.*

Groupings Type Number of Species Percent of Species

Life form Forbs 119 57%

Grass 36 17%

Sedge/Rush 30 14%

Tree/Shrub 15 07%

Vine 8 04%

Wetland Status OBL 23 11%

FACW 43 21%

FAC 61 29%

FACU 41 19%

UPL 12 05%

UD 28 14%

*OBL = Obligate Wetland, FACW = Facultative Wetland, FAC = Facultative, FACU = Facultative Upland, UPL = Upland, UD = Undesignated

127

Bunchflower (Veratrum virginicum), Eastern Mannagrass (Glyceria septentrionalis),

Indian Paintbrush (Castilleja coccinea), Sweet Betsy (Trillium cuneatum), Little

Sundrops (Oenothera perennis), Virginia Yellow Flax (Linum virginianum), and Red

Canada Lily (Lilium canadense var. editorum). The population of Red Canada Lily is disjunct from its typical mountain habitat and was only concentrated in a small section of the site (Weakley 2007; North Carolina Natural Heritage Program 2006).

Ninety of the species had previously been reported as occurring in prairie-like associations (Edgin et al. 2005; Klips 2003; Laughlin and Uhl 2003; Davis et al. 2002;

Edgin and Ebinger 2000; Hill 1997, revised 2000; Leidolf and McDaniel 1998; Webb et al. 1997). Of these, 18 are reported from wet-mesic prairie conditions in Midwestern, mid-atlantic, and southern mesic prairies. These species include Wild Onion (Allium canadense var. canadense), Hemp Dogbane (Apocynum cannabinum), Swamp Milkweed

(Asclepias incarnata var. pulchra), Water Hemlock (Cicuta maculata var. maculata),

Old-field Five-fingers (Potentilla simplex), Golden Ragwort (Packera aurea), Upright

Sedge (Carex stricta), Broom Sedge (Carex scoparia var. scoparia), Big Bluestem

(Andropogon gerardii), Indian Paintbrush (Castilleja coccinea), Smooth Spiderwort

(Tradescantia ohiensis), Common Rush (Juncus effusus var. soltus), Button Bush

(Cephalanthus occidentalis), Eastern Gray Goldenrod (Solidago nemoralis var. nemoralis), Indian Grass (Sorghastrum nutans), Little Sundrops (Oenothera perennis),

Nodding Ladies‟-tresses (Spiranthes cernua), and White Turtlehead (Chelone glabra)

(Edgin et al. 2005; Mack and Boerner 2004; Klips 2003; Laughlin and Uhl 2003; Edgin and Ebinger 2000; Leidolf and McDaniel 1998; Webb et al. 1997; Henderson 1995; Taft and Solecki 1990).

128

Quadrat sampling conducted during the 2006-07 growing seasons detected 87 species, an average of 6.8 species per 1m2 quadrat. Andropogon gerardii had the highest frequency within the quadrats (60%) followed by Tripsacum dactyloides var. dactyloides

(53%) (Table 4.2).

Soils

The predominant soil at Suther Prairie has recently been mapped as a Chastain series

(fine, mixed semi-active, acid, thermic Fluvaquentic Endoaquepts) (Cabarrus Soil and

Water Conservation pers. comm. 2008). This soil mapping unit is quite uncommon on

Piedmont landscapes, being most commonly found on frequently flooded and poorly drained eastern Coastal Plain sites. Calcium and magnesium levels were very high throughout the surface of the soil, while phosphorous levels were low. In addition, C, N,

P, and Zn decreased with soil depth (Table 4.3). Mean soil pH for all three depths was 5.8

(Table 4.3).

129

Table 4.2. Species at Suther Prairie with highest frequency values in sampled

quadrats during the 2006-07 growing seasons.

Species Frequency Andropogon gerardii 60%

Tripsacum dactyloides var. dactyloides 53%

Apocynum cannabinum 44%

Cephalanthus occidentalis 39%

Dichanthelium scoparium 36%

Carex vulpinoidea 31%

Holcus lanatus 24%

Allium canadense var. canadense 22%

Apios americana 18%

Scirpus atrovirens 15%

Agrostis hyemalis 14%

Sphenopholis obtusata 13%

130

Table 4.3. Soil chemical properties of Suther Prairie (mean + std) of 6 samples per

each soil sampling depth. Superscripts denote significant different means

between soil depths (Tukey‟s Test, P < 0.05).

Parameter Measured Soil depth, 0-10 cm Soil depth, 11-20 cm Soil depth, 21-30 cm pH 5.7 ± 0.4 5.9 ± 0.2 6.0 ± 0.2 N (%) 0.2 ± 0.08a 0.1 ± 0.05a,b 0.1 ± 0.04b C (%) 2.7 ± 0.5a 1.5 ± 0.5b 0.9 ± 0.3b C/N 10.1 ± 1.3 9.2 ± 1.3 8.5 ± 1.0 P (kg/ha) 6.7 ± 2.5a (low)* 2.98 ± 1.1b (low) 2.6 ± 0.9b (low) K (kg/ha) 160.8 ± 257.8 (medium) 73.2 ± 80.9 (low) 39.2 ± 29.6 (low) Ca (kg/ha) 2041.4 ± 445.7 (high) 1829.4 ± 208.7 (high) 1818.2 ± 247.5 (high) Mg (kg/ha) 717.1 ± 177.7 (high) 744 ± 170.8 (high) 787 ± 223.9 (high) Zn (kg/ha) 2.98 ± 0.9a 1.4 ± 0.3b 1.1 ± 0.3b Mn (kg/ha) 48.9 ± 17.2 42.0 ± 20.6 45.6 ± 14.5 Cu (kg/ha) 4.9 ± 2.0 4.4 ± 1.1 4.1 ± 1.1 B (kg/ha) 0.8 ± 0.5 0.4 ± 0.2 0.3 ± 0.1 Na (kg/ha) 44.0 ± 20.5 45.2 ± 21.8 47.8 ± 19.6

* low = low level amounts for that nutrient; medium = medium level amounts for thay nutrient; high = high level amounts for that nutrient

131

DISCUSSION

Plant species richness at Suther Prairie is similar to that found in prairies in western states with similar edaphic conditions (Henderson 1995; Taft and Solecki 1990). The large proportion (32%) of obligate and facultative wetland species is indicative of the mesic-hydric conditions. Many species documented from the site prior to 2006-07 were not re-located, likely as a result of drought. The 2007 growing season, in particular, was among the driest on record for the region (Table 4.4). Many of the species not observed during the later sampling events were associated with the wetter portions of the site, and it is likely that the drought conditions more adversely affected hydrophytic species. Even several facultative and facultative upland species previously documented from the site were not re-located in the

2007 survey. In addition, it is not clear to what extent either management or environmental changes not related to available moisture over the last 10 years, may have also influenced changes in the floristic composition.

The historically wetter soil conditions at Suther Prairie make it difficult to compare with other Piedmont prairies that have greater seasonal soil moisture variability. Several species that are generally associated with Piedmont prairies, that have drier summer soil conditions, were not observed at Suther Prairie including Kidneyleaf Rosinweed (Silphium compositum var. compositum), Prairie Dock (Silphium terebinthinaceum), Blazing Star (Liatris virgata),

Scaly Blazing Star (Liatris squarrosa var. squarrosa), Smooth Coneflower (Echinacea laevigata), Common Wild Quinine (Parthenium integrifolium var. integrifolium), Georgia

Aster (Symphyotrichum georgianum), Northern Rattlesnake-master (Eryngium yuccifolium var. yuccifolium) and Winged Sumac (Rhus copallinum var. copallinum). Conversely, Suther

132

Table 4.4. Climate data for Concord, North Carolina 2006-07 (Data from State

Climate Office of North Carolina; normal values from 1971-2000).

2006 2007 Normal

Temperature (˚C) 16.3 16.8 15.3

Precipitation (cm) 110.7 72.4 120.1

Prairie includes a number of unusual species such as Indian Paintbrush (Castilleja coccinea) and Red Canada Lily (Lilium canadense var editorum) that are not commonly associated with other Piedmont prairie sites.

The current species list includes 31 (15%) non-native species. This represents an increase from the percentage of non-native species documented at the site in 2002 (6%). The increase is likely a result of two factors: logging of the adjacent forest and associated disturbance, and inclusion of the previously plowed fescue field, which likely had a higher percentage of non- native species than the original prairie site. However, species data from the former fescue field section is from a more limited duration compared with that of the original prairie portion. As a result, it will be necessary to monitor species compositional changes in the future to better assess whether this increase in non-native species is potentially a threat to the native species composition of the prairie.

The introduction of prescribed burns, beginning in 1997, may explain why several oak species previously documented were not observed during the 2006-07 sampling periods.

These species may have been documented from the margins of the hedgerow that previously divided the site, or small oak seedlings that were able to persist prior to the prescribed burns may have been present within the prairie. Additionally, woodland species such as Sweet

133

Betsy (Trillium cuneatum) and May-apple (Podophyllum peltatum) were previously documented from the margin of the site, adjacent to the hardwood forest. The adjacent margin of the prairie is now comprised of early successional woodland habitat.

The uniqueness of the site as a remnant prairie is well demonstrated by the large number of prairie species present. Forty-three % of the documented species from Suther Prairie are known to form prairie-like associations. Thus, the species composition, as well as the high frequency of Andropogon gerardii, is consistent with Midwestern, mid-atlantic, and southern mesic prairies (Edgin et al. 2005; Klips 2003; Laughlin and Uhl 2003; Edgin and Ebinger

2000; Leidolf and McDaniel 1998; Webb et al. 1997).

Soil chemical analysis data indicated above-normal levels of both calcium and magnesium, and below-normal phosphorus levels (Clemson University Agriculture Service Laboratory

2008). Extractable Ca and Mg fell within the top 1-5% of soil analyses performed on

Piedmont soil samples from forested or tree planting sites by the Clemson soil testing lab

(Soil Test Yearly Summaries - www.clemson.edu/agsrvlb/2007SoilSummary.htm). More than 25 years ago, the remnant portion of the prairie had received one treatment of lime, but since that time no soil amendments have been added (Louis Suther pers. comm. 2008).

The high level of both calcium and magnesium may be related to the presence of Enon and

Mecklenburg soil series on surrounding upland areas of the watershed. These series are derived from basic igneous and metamorphic rocks that are high in ferro-magnesian minerals

(Soil Survey of Cabarrus County, North Carolina 1988). The colluvial deposition from these soil series may have contributed to the buildup of extractable Ca and Mg in the soil. Soil pH from all three sampled depths was found to be moderately acidic. However, recent additional soil data from Suther Prairie suggest that soil pH may be lower in some areas of the prairie

134

than samples taken for this study indicated, particularly in portions of the site that are more hydric (Lee Lehman pers. comm. 2008).

Much of the uniqueness of the Suther Prairie vegetation may be attributed to the long history of annual mowing for hay from the unplowed soil. Mowing suppressed invasion of mid-successional species, while the absence of plowing permitted perennial herbaceous rootstocks to survive. In addition, this helped to retain nutrients, such as Ca and Mg, in the surface layers of the soil. These factors and soil conditions ranging from mesic to hydric, create ideal conditions for high herbaceous plant species richness.

Periodic floristic surveys should be conducted at this site to determine how species richness and composition may be changing, particularly with recent climatic changes that include higher annual temperatures. The results of this study, along with future findings, should contribute to improved management protocols for not only this prairie, with its unique hydrology and soils, but also for protection and restoration of other Piedmont prairies in the eastern United States.

135

LITERATURE CITED

Barden, L.S. 1997. Historic prairies in the Piedmont of North and South Carolina, USA.

Natural Areas Journal 17:149-152.

Clemson University Agricultural Service Laboratory. 2008. Clemson University.

Clemson, South Carolina. (http://www.clemson.edu/agsrvlb/).

Davis, Jr., J. E., C. McRae, B. L. Estep, L. S. Barden and J. F. Matthews. 2002. Vascular

flora of Piedmont prairies: evidence from several prairie remnants. Castanea

67:112.

Edgin, B. R. and J. E. Ebinger. 2000. Vegetation of a successional prairie at Prairie Ridge

State Natural Area, Jasper County, Illinois. Castanea 65:139-146.

Edgin, B. R., R. Beadles, and J. E. Ebinger. 2005. Vascular flora of Beadles Barrens

Nature Preserve, Edwards County, Illinois. Castanea 70:47-58.

Estep, B. L., and C. A. McRae. 1997. Prairie restoration and management plan, Suther

Farm Prairie, Cabarrus County, North Carolina. Unpublished.

Henderson, R. A. 1995. Plant species composition of Wisconsin prairies. An aid to

selecting species for planting and restorations based upon University of Wisconsin-

Madison Plant Ecology Laboratory data. Technical Bulletin No. 188. Dept. of

Natural Resources, Madison, Wisconsin.

Hill, S. R. 1997, revised 2000. The prairie element in the flora of Oconee County,

South Carolina. Unpublished list.

Jones, J. B. 2001. Laboratory guide for conducting soil tests and plant analysis. CRC,

London, p. 363.

136

Kartesz, J. T. 1996. National List of Vascular Plant Species that Occur in Wetlands: Biota

of North America Program (BONAP) Chapel Hill, North Carolina.

Klips, R. A. 2003. Vegetation of Claridon Railroad Prairie, a remnant of the Sandusky

Plains of Central Ohio. Castanea 68:135-142.

Laughlin, D. C. and C. F. Uhl. 2003. The xeric limestone prairies of Pennsylvania.

Castanea 68:300-316.

Leidolf, A., and S. McDaniel. 1998. A floristic study of Black Prairie plant communities

at Sixteen Section Prairie, Oktibbeha County, Mississippi. Castanea 63:51-62.

Logan, J. H. 1859. A history of the upper country of South Carolina from the earliest

periods to the close of the War of Independence. S.G. Courtenay and Co.,

Charleston, South Carolina. 551 pp.

Mack J. J. and R. E. J. Boerner. 2004. At the tip of the prairie peninsula: vegetation of

Daughter Savannah, Crawford County, Ohio. Castanea 69:309-323.

McRae C. A. 1999. Experiments in Prairie Restoration. M.S. Thesis, University of North

Carolina at Charlotte.

NatureServe. 2009. NatureServe Explorer: An online encyclopedia of life [web

application]. Version 7.1. NatureServe, Arlington, Virginia. Available

http://www.natureserve.org/explorer. (Accessed: October 7, 2009).

North Carolina Natural Heritage Program. 2006. Natural Heritage Program List of

Rare Plant Species of North Carolina. Franklin, M.A. and J.T. Finnegan (eds.)

North Carolina Natural Heritage Program Office of Conservation and Community

Affairs. North Carolina Department of Environment and Natural Resources.

Raleigh, North Carolina.

137

SAS Institute. 2002. SAS User‟s Guide: Statistics. Version 9.1. SAS Institute, Cary,

North Carolina. SAS Institute.

Schafale, M. P. and A. S. Weakley. 1990. Classification of the Natural Communities of

North Carolina. Third Approximation. North Carolina Natural Heritage Program,

Division of Parks and Recreation, North Carolina Department of Environment,

Health, and Natural Resources, Raleigh, North Carolina.

Schmidt, J. M. and J. A. Barnwell. 2002. A flora of the Rock Hill Heritage Preserve,

York County, South Carolina. Castanea 67:247-279.

Soil Survey of Cabarrus County, North Carolina. 1988. United States Department

of Agriculture. Soil Conservation Service.

Taft , J. B. and M. K. Solecki. 1990. Vascular flora of the wetland and prairie communities

of Gavin Bog and Prairie Nature Preserve, Lake County, Illinois. Rhodora 92:142-

165.

Weakley, A. S. 2007. Flora of the Carolinas, Virginia, Georgia, and Surrounding Areas,

working draft of January 2007. University of North Carolina, Chapel Hill, North

Carolina.

Webb, D. H., H. R. Deselm and W. M. Dennis. 1997. Studies of prairie barrens of

northwestern Alabama. Castanea 62:173-184.

138

APPENDIX I

Plant species identified at Suther Prairie, Cabarrus County, N.C. from 1997 through 2007.

Species collected during the 2006-2007 growing seasons are noted with ^. OBL = Obligate

Wetland; FACW (+ or -) = Facultative Wetland; FAC (+ or -) = Facultative; FACU (+ or -) =

Facultative Upland; UPL = Upland. P = Prairie-like Species; R = North Carolina Rare

Species; UC = Uncommon Species; WL = North Carolina Watch List Species. Collectors followed by accession numbers: RT = Robert Tompkins; JM = Jim Matthews; BE = Bret

Estep; CM = Catherine McRae. CLEMS = Clemson University Herbarium; UNCC =

University of North Carolina at Charlotte Herbarium (now housed at Reedy Creek Nature

Preserve in Charlotte, N.C.). ______

ACANTHACEAE

Ruellia caroliniensis (J.F Gmelin) Steudel [P] (JM31752) UNCC

ALLIACEAE

Allium canadense L. var. canadense [FAC, P] ^(RT0073) CLEMS

ALTINGIACEAE

Liquidambar styraciflua L. [FAC+] ^(RT0794) CLEMS

AMARYLLIDACEAE

Zephyranthes atamasca (L.) Herbert [FAC] ^(RT0022) CLEMS

ANACARDIACEAE

Toxicodendron radicans (L.) Kuntze var. radicans [FAC] ^(RT0667) CLEMS

APIACEAE

Cicuta maculata L. var. maculata [OBL, P] ^(RT0232) CLEMS

Daucus carota L. [UPL] ^(RT0231) CLEMS

139

Ptilimnium capillaceum (Michx.) Rafinesque [OBL] ^(RT0012) CLEMS

APOCYNACEAE

Amsonia tabernaemontana Walter var. salicifolia (Pursh) Woodson [FACW, P] ^(RT0043) CLEMS

Apocyonum cannabinum L. [FAC-, P] ^(RT0199) CLEMS

Asclepias incarnata L. var. pulchra [OBL, P] ^(RT0276) CLEMS

Asclepias tuberosa L. var. tuberosa ^(RT0903) CLEMS

ARISTOLOCHIACEAE

Endodeca serpentaria (L.) Rafinesque [FACU] (JM31964) UNCC

ASTERACEAE

Ageratina altissima King & H.E. Robinson var. altissima [FAC-, P] ^(RT0267) CLEMS

Ambrosia artemisiifolia L. [FACU, P] ^(RT0699) CLEMS

Bidens frondosa L. [FACW, P] ^(RT0683) CLEMS

Chrysopsis mariana (L.) Elliott [UPL, P] ^(RT0408) CLEMS

Cirsium horridulum Michx. var. horridulum [FAC+] (BE34505) UNCC

Cirsium vulgare (Savi) Tenore [FAC] ^(RT0258) CLEMS

Conyza canadensis (L.) Cronquist var. canadensis [FACU, P] ^(RT0687) CLEMS

Coreopsis lanceolata L. [UPL, P] ^(RT0083) CLEMS

Elephantopus carolinianus Raeuschel [FAC] (JM32427) UNCC

Erechtites hieracifolius (L.) Rafinesque ex de Candolle [FACU, P] ^(RT0437) CLEMS

Erigeron strigosus Muhlenberg ex Willdenow var. strigosus [FAC, P] ^(RT0097) CLEMS

Eupatorium capillifolium (Lamarck) Small [FACU, P] ^(RT0440) CLEMS

Eupatorium hyssopifolium L. [P] (JM35085) UNCC

140

Helenium amarum (Rafinesque) H. Rock var. amarum [FAC-] ^(RT0444) CLEMS

Helenium autumnale L. [FACW, P] ^(RT0433) CLEMS

Krigia dandelion (L.) Nuttall [FACU] (CM34508) UNCC

Leucanthemum vulgare Lamarck [UPL] ^(RT0090) CLEMS

Marshallia obovata (Walter) Beadle & F.W. Boynton var. obovata ^(RT0092) CLEMS

Packera aurea (L.) A. D. Love [FACW, P] ^(RT0038) CLEMS

Pityopsis aspera (Shuttleworth ex Small) Small var. adenolepsis ^(RT0761) CLEMS

Pseudognaphalium obtusifolium (L.) Hilliard & Burtt [P] ^(RT0781) CLEMS

Pyrrhopappus carolinianus (Walter) A.P. de Candolle [P] ^(RT0904) CLEMS

Solidago altissima L. var. altissima [FACU+] ^(RT0905) CLEMS

Solidago gigantea Aiton (FACW) ^[RT0306] CLEMS

Solidago nemoralis Aiton var. nemoralis [P] ^(RT0432) CLEMS

Symphyotrichum dumosum L. [FAC, P] ^(RT0443) CLEMS

Symphyotrichum lateriflorum L. [FAC, P] ^(RT0442) CLEMS

Symphyotrichum puniceum (L.) A. D. Love var. puniceum [OBL] (JM31960) UNCC

Vernonia noveboracensis (L.) Michx. [FAC+] (JM35084) UNCC

BERBERIDACEAE

Podophyllum peltatum L. [FACU] (BE34470) UNCC

BETULACEAE

Betula nigra L. [FACW] ^(RT0789) CLEMS

Carpinus caroliniana Walter [FAC] ^(RT0780) CLEMS

BIGNONIACEAE

Campsis radicans (L.) Seeman ex Bureau [FAC, P] ^(RT0740) CLEMS

141

BORAGINACEAE

Myosotis verna Nuttall [FAC-] (BE32616) UNCC

CAMPANULACEAE

Lobelia cardinalis L. [FACW+] (JM31974) UNCC

Triodanis perfoliata (L.) Nieuwland [FACU+, P] ^(RT0089) CLEMS

CAPRIFOLIACEAE

Lonicera japonica Thunberg [FAC-] ^(RT0752) CLEMS

CARYOPHYLLACEAE

Cerastium fontanum Baumgartner var. vulgare (Hartman) Greuter & Burdet [FAC-] (JM35105) UNCC

Dianthus armeria L. var. armeria [UPL] ^(RT0229) CLEMS

Stellaria pubera Michx. (CM35234) UNCC

COMMELINACEAE

Tradescantia ohiensis Rafinesque [FAC-, P] ^(RT0023) CLEMS

CORNACEAE

Cornus amomum P. Miller [FACW+] (BE34478) UNCC

Cornus florida L. [FACU] ^(RT0791) CLEMS

CUCURBITACEAE

Melothria pendula L. var. pendula [FACW-] ^(RT0275) CLEMS

CUPRESSACEAE

Juniperus virginiana L. var. virginiana [FACU-, P] ^(RT0793) CLEMS

CYPERACEAE

Carex annectens (Bicknell) Bicknell [FACW] (BE34515) UNCC

Carex atlantica Bailey [FACW-] ^(RT0044) CLEMS

142

Carex blanda Dewey [FAC-] (CM32642) UNCC

Carex bushii Mackenzie [FACW, P, UC] ^(RT0066) CLEMS

Carex flaccosperma Dewey [FAC+] ^(RT0085) CLEMS

Carex lupulina Muhlenburg ex Willdenow [OBL] (JM35072) UNCC

Carex lurida Wahlenberg [OBL] ^(RT0222) CLEMS

Carex scoparia ScKuhr ex Willdenow var. scoparia [FACW, P] ^(RT0756) CLEMS

Carex squarrosa L. [FACW] ^(RT0070) CLEMS

Carex stricta Lamarck [OBL, P] ^(RT0047) CLEMS

Carex styloflexa Buckley [FACW] ^(RT0039) CLEMS

Carex tenera Dewey var. tenera [FACW, P, WL] ^(RT0086) CLEMS

Carex vulpinoidea Michx. [OBL, P] ^(RT0067) CLEMS

Cyperus croceus Vahl [FAC] ^(RT0906) CLEMS

Cyperus echinatus (L.) Wood [FAC] ^(RT0221) CLEMS

Cyperus erythrorhizos Muhlenberg [OBL, UC] ^(RT0280) CLEMS

Cyperus pseudovegetus Steudel [FACW] ^(RT0279) CLEMS

Cyperus strigosus L. [FACW+] ^(RT0011) CLEMS

Eleocharis tricostata Torrey [FACW+, WL] ^(RT0076) CLEMS

Rhynchospora caduca Elliott [FACW+, UC] ^(RT0009) CLEMS

Rhynchospora glomerata (L.) Vahl var. glomerata [OBL] ^(RT0010) CLEMS

Scirpus atrovirens Willdenow [OBL, P] ^(RT0068) CLEMS

Scirpus cyperinus (L.) Kunth [OBL] ^(RT0585) CLEMS

Scirpus pendulus Muhlenberg [OBL, P, R] ^(RT0307) CLEMS

143

EBENACEAE

Diospyros virginiana L. [FAC] ^(RT0739) CLEMS

EUPHORBIACEAE

Tragia urticifolia Michx. [P] ^(RT0738) CLEMS

FABACEAE

Apios americana Medikus [FACW, P] ^(RT0005) CLEMS

Chamaecrista fasciculata (Michx.) Greene var. fasciculata [FACU, P] ^(RT0263) CLEMS

Crotalaria rotundifolia Walter ex J.F. Gmelin var. vulgaris [FACU, P ] ^(RT0273) CLEMS

Desmodium canescens (L.) Augustin de Candolle [FACU, P] ^(RT0003) CLEMS

Desmodium ciliare (Muhlenberg ex Willdenow) Augustin de Candolle [P] ^(RT0296) CLEMS

Desmodium paniculatum (L.) Augustin de Candolle var. paniculatum [FACU, P] ^(RT0445) CLEMS

Desmodium perplexum Schubert ^(RT0299) CLEMS

Desmodium viridiflorum (L.) A.P. de Candolle [P] (JM35086) UNCC

Galactia volubilis (L.) Britton var. volubilis [FACU, P] ^(RT0274) CLEMS

Lathyrus hirsutus L. (JM34514) UNCC

Lespedeza cuneata (Dumont-Cours.) G. Don [UPL] ^(RT0262) CLEMS

Lespedeza procumbens Michx. [P] ^(RT0298) CLEMS

Trifolium campestre Schreber [UPL] (JM35106) UNCC

Trifolium dubium Sibthorp [FACU-] (JM34625) UNCC

Trifolium pratense L. [FACU-] ^(RT0268) CLEMS

Vicia sativa L. var. nigra (L.) Ehrhart [FACU] ^(RT0782) CLEMS

Vicia tetrasperma (L.) Schreber (JM35101) UNCC

144

FAGACEAE

Quercus alba L. [FACU] ^(RT0792) CLEMS

Quercus phellos L. [FACW-] ^(RT0784) CLEMS

Quercus stellata Wangenheim [FACU, P] ^(RT0786) CLEMS

GENTIANACEAE

Sabatia angularis (L.) Pursh [FAC] ^(RT0265) CLEMS

GERANIACEAE

Geranium carolinianum L. var. carolinianum [P] (JM35107) UNCC

HYPERICACEAE

Hypericum mutilum L. var. mutilum [FACW, P] ^(RT0285) CLEMS

Hypericum prolificum L. [FACU] ^(RT0260)

IRIDACEAE

Sisyrinchium mucronatum Michx. [FACW, P] ^(RT0024) CLEMS

JUNCACEAE

Juncus acuminatus Michx. [OBL] ^(RT0063) CLEMS

Juncus biflorus Elliott [FACW] (JM31770) UNCC

Juncus effusus L. var. solutus Fernald & Wiegand [FACW+, P] ^(RT0224) CLEMS

Juncus scirpoides Lamarck var. scirpoides [FACW+] ^(RT0069) CLEMS

Juncus tenuis Willdenow [FAC, P] ^(RT0735) CLEMS

Luzula echinata (Small) F.J. Hermann [FAC, P] ^(RT0057) CLEMS

LAMIACEAE

Mentha x. piperita L. (pro sp.) var. Piperita [Mentha aquatica x spicata] [FACW] (JM31728) UNCC

Prunella vulgaris L. [FAC-, P] ^(RT0233) CLEMS

145

Pycnanthemum tenuifolium Schrader [FAC-, P] ^(RT0198) CLEMS

Salvia lyrata L. (FAC-) ^(RT0743) CLEMS

Scutellaria integrifolia L. [FAC, P] ^(RT0091) CLEMS

LILIACEAE

Lilium canadense L. var. editorum Fernald [FAC, R] (JM31764) UNCC

LINACEAE

Linum virginianum L. [FAC+, UC] (JM31761) UNCC

MAGNOLIACEAE

Liriodendron tulipifera L. var. tulipifera [FAC] ^(RT0790) CLEMS

MELANTHIACEAE

Veratrum virginicum (L.) Aiton [FACU-, UC] ^(RT0213) CLEMS

MELASTOMATACEAE

Rhexia mariana L. var. mariana [FACW+] (JM31762) UNCC

OLEACEAE

Fraxinus pennsylvanica Marshall [ FACW] ^(RT0751) CLEMS

ONAGRACEAE

Ludwigia alternifolia L. [OBL] ^(RT0002) CLEMS

Oenothera biennis L. [FACU, P] ^(RT0088) CLEMS

Oenothera fruticosa L. var. fruticosa [FACU, P] (CM32634) UNCC

Oenothera perennis L. [FAC-, P, UC) (JM34624) UNCC

ORCHIDACEAE

Spiranthes cernua (L.) L.C. Richard [FACW, P] (JM31947) UNCC

Spiranthes vernalis Engelmann & A. Gray [FACW-] (JM31759) UNCC

146

OROBANCHACEAE

Castilleja coccinea (L.) Sprengel [FAC, P, UC] ^(RT0096) CLEMS

OXALIDACEAE

Oxalis dillenii Jacquin (BE34458) UNCC

Oxalis violacea L. [P] (JM35102) UNCC

PHRYMACEAE

Mimulus alatus Aiton [OBL] ^(RT0300) CLEMS

PLANTAGINACEAE

Chelone glabra L. [OBL, P] (JM32425) UNCC

Gratiola neglecta Torrey [OBL] (JM31767) UNCC

Gratiola viscidula Pennell [OBL] (JM31741) UNCC

Mecardonia acuminata (Walter) Small var. acuminata [FACW] ^(RT0269) CLEMS

Penstemon australis Small [P] (BE34510) UNCC

Penstemon laevigatus Aiton [FAC] ^(RT0054) CLEMS

Plantago virginica L. [FACU-, P] ^(RT0093) CLEMS

POACEAE

Agrostis hyemalis (Walter) Britton [FAC, P] ^(RT0058) CLEMS

Aira caryophyllea L. ^(RT0466) CLEMS

Andropogon gerardii Vitman [FAC, P] ^(RT0750) CLEMS

Andropogon virginicus L. var. virginicus [FAC-, P] ^(RT0434) CLEMS

Bromus commutatus Schrader [UPL] ^(RT0099) CLEMS

Bromus japonicus Thunberg ex Murray [FACU] ^(RT0059) CLEMS

Bromus secalinus L. ^(RT0100) CLEMS

Dactylis glomerata L. [FACU] ^(RT0901) CLEMS 147

Danthonia spicata (L.) Palisot de Beauvois [P] ^(RT0606) CLEMS

Dichanthelium acuminatum (Swartz) Gould & Clark var. fasciculatum (Torrey) Freckmann [FAC, P] ^(RT0080) CLEMS

Dichanthelium clandestinum (L.) Gould [FACW, P] ^(RT0203) CLEMS

Dichanthelium dichotomum (L.) Gould var. dichotomum [FAC, P] ^(RT0303) CLEMS

Dichanthelium scoparium (Lamarck) Gould [FACW] ^(RT0007) CLEMS

Dichanthelium sphaerocarpon (Elliott) Gould [FACU, P] ^(RT0216) CLEMS

Echinochloa crusgalli (L.) Palisot de Beauvois var. crusgalli [FACW-] ^(RT0304) CLEMS

Elymus hystrix L. var. hystrix [UPL] ^(RT0907) CLEMS

Elymus virginicus L. var. virginicus [FAC, P] ^(RT0215) CLEMS

Eragrostis spectabilis (Pursh) Steudel [FACU, P] ^(RT0436) CLEMS

Festuca paradoxa Desvaux [FAC-] ^(RT0214) CLEMS

Glyceria septentrionalis A.S. Hitchcock [OBL, UC] ^(RT0055) CLEMS

Glyceria striata (Lamarck) A.S. Hitchcock var. striata [OBL, P] ^(RT0077) CLEMS

Holcus lanatus L. [FACU-] ^(RT0636) CLEMS

Lolium perenne L. var. aristatum Willdenow [FACU] ^(RT0061) CLEMS

Panicum dichotomiflorum Michx. var. dichotomiflorum [OBL, P] ^(RT0008) CLEMS

Panicum rigidulum Bosc ex Nees var. elongatum (Pursh) Lelong [FACW] (JM32439) UNCC

Paspalum dilatatum Poiret [FAC+] ^(RT0266) CLEMS

Pennisetum glaucum L.R. Brown [FAC] ^(RT0013) CLEMS

Poa sylvestris A. Gray [FAC+] ^(RT0036) CLEMS

Saccharum giganteum (Walter) Persoon [FACW, P] ^(RT0435) CLEMS

Setaria parviflora (Poiret) Kerguelen [FAC] (CM35240) UNCC

148

Sorghastrum nutans (L.) Nash [FACU, P] ^(RT0305) CLEMS

Sorghum halepense (L.) Persoon [FACU] ^(RT0234) CLEMS

Sphenopholis obtusata (Michx.) Scribner [FAC+, P] ^(RT0060) CLEMS

Tripsacum dactyloides (L.) L. var. dactyloides [FAC+] ^(RT0204) CLEMS

Vulpia myuros (L.) K.C. Gmelin [FACU] ^(RT0056) CLEMS

Vulpia sciurea (Nuttall) Henrard [UPL] ^(RT0098) CLEMS

POLYGONACEAE

Persicaria maculosa S.F. Gray [FACW] ^(RT0289) CLEMS

Persicaria setacea (Baldwin) Small [FACW] (JM31742) UNCC

PORTULACACEAE

Claytonia virginica L. var. virginica ^(RT0486) CLEMS

RANUNCULACEAE

Ranunculus pusillus Poiret [FACW+] (BE34468) UNCC

Ranunculus sardous Crantz [FAC+] (BE34465) UNCC

ROSACEAE

Fragaria virginiana P. Miller [FAC-, P] ^(RT0766) CLEMS

Potentilla simplex Michx. [FACU, P] ^(RT0237) CLEMS

Prunus serotina Erhart var. serotina [FACU] ^(RT0785) CLEMS

Rosa multiflora Thunberg ex Murray [UPL] ^(RT0693) CLEMS

Rubus argutus Link [FAC, P] ^(RT0259) CLEMS

RUBIACEAE

Cephalanthus occidentalis L. [OBL, P] ^(RT0230) CLEMS

Diodia virginiana L. [FACW] ^(RT0720) CLEMS

Galium circaezans Michx. var. circaezans [FAC-, P] ^(RT0290) CLEMS 149

Galium obtusum Bigelow var. filifolium (Wiegand) Fernald [FACW-] ^(RT0048) CLEMS

Houstonia caerulea L. [FAC, P] (CM32636) UNCC

SAURURACEAE

Saururus cernuus L. [OBL] (JM31769) UNCC

SAXIFRAGACEAE

Heuchera americana L. [FACU, P] ^(RT0095) CLEMS

SMILACACEAE

Smilax glauca Walter [FAC] (JM35074) UNCC

Smilax rotundifolia L. [FAC, P] ^(RT0747) CLEMS

SOLANACEAE

Physalis heterophylla Nees [P] (JM34689) UNCC

Physalis pubescens L. var. pubescens [UPL, P] ^(RT0272) CLEMS

Solanum carolinense L. var. carolinense [FACU, P] ^(RT0748) CLEMS

TETRACHONDRACEAE

Polypremum procumbens L. [FACU-] ^(RT0579) CLEMS

TRILLIACEAE

Trillium cuneatum Rafinesque [UC] (JM35098) UNCC

ULMACEAE

Ulmus alata Michx. [FACU+] ^(RT0749) CLEMS

VALERIANACEAE

Valerianella radiata (L.) Dufresne [FAC-, P] ^(RT0207) CLEMS

VERBENACEAE

Verbena urticifolia L. var. urticifolia [FAC+, P] ^(RT0261) CLEMS

150

VIOLACEAE

Viola sororia Willdenow [FAC-] (JM34535) UNCC

VITACEAE

Parthenocissus quinquefolia (L.) Planchon [FAC] ^(RT0783) CLEMS

______

151

APPENDIX II

Plant species historically identified, but lacking voucher specimens, from Suther Prairie.

APIACEAE

Conium maculatum L.

Thaspium barbinode (Michx.) Nuttall

APOCYNACEAE

Trachelospermum difforme (Walter) A. Gray

QUIFOLIACEAE

Ilex opaca Aiton var. opaca

ARACEAE

Peltandra virginica (L.) Schott

ASTERACEAE

Eutrochium fistulosum (Barratt) E.E. Lamont

Lactuca canadensis L.

Pluchea camphorata (L.) A.P. de Candolle

Symphyotrichum pilosum (Willdenow) Nesom var. pilosum

CAMPANULACEAE

Triodanis biflora (Ruiz & Pavon) Greene

CYPERACEAE

Scleria pauciflora Muhlenburg ex Willdenow var. pauciflora

FABACEAE

Lespedeza repens (L.) W. Barton

Medicago lupulina (L.)

152

Wisteria frutescens (L.) Poiret

FAGACEAE

Quercus falcata Michx.

Quercus marilandica Muenchhausen var. marilandica

HALORAGACEAE

Proserpinaca pectinata Lamarck

HYPERICACEAE

Hypericum punctatum Lamarck

JUGLANDACEAE

Juglans nigra L.

LAURACEAE

Sassafras albidum (Nuttall) Nees

MALVACEAE

Hibiscus moscheutos L. var. moscheutos

MORACEAE

Morus rubra L.

MYRSINACEAE

Anagallis arvensis L. var. arvensis

PASSIFLORACEAE

Passiflora incarnata L.

PINACEAE

Pinus taeda L.

153

PLATANACEAE

Platanus occidentalis L. var. occidentalis

POACEAE

Andropogon glomeratus (Walter) Britton, Sterns, & Poggenburg var. glomeratus

Arthraxon hispidus (Thunberg) Makino var. hispidus

Chasmanthium latifolium (Michx.) Yates

Microstegium vimineum (Trinius) A. Camus

Panicum anceps Michx. var. anceps

Saccharum brevibarbe (Michx.) Persoon

Schizachyrium scoparium (Michx.) Nash var. scoparium

Tridens flavus (L.) A.S. Hitchcock

ROSACEAE

Rosa carolina L.

ULMACEAE

Ulmus americana L. var. americana

154

Chapter 5

A Newly Documented and Significant Piedmont

Prairie Site with a Helianthus schweinitzii Torrey & A.

Gray (Schweinitz‟s Sunflower) Population

155

ABSTRACT

This study documented and described a recently located (2003) and significant Piedmont prairie site (Troy Prairie) from Montgomery County, N.C. From 2006-08 a complete floristic inventory was conducted. In addition, 75 quadrats (1m x 1m) were sampled for species frequency, and 12 quadrats were sampled at 3 depths (0-10 cm, 11-20 cm, 21-30 cm) for edaphic (soil) chemistry to determine pH; organic C and N contents; extractable P, K, Ca,

Mg, Zn, and Mn. A total of 163 vascular plant species were identified, including 62 monocots, 97 dicots, 3 ferns and one Lycopsid. There were 57 (35%) graminoids and 39

(24%) Asteraceae. Of the Asteraceae, Helianthus schweinitzii (Schweinitz‟s Sunflower) is

Federally endangered, while Solidago gracillima and S. radula are both rare-listed for N.C.

Only five individuals of H. schweinitzii were found in this study of an estimated 100 in 2003.

Eighty-seven (53%) species had previously been reported as occurring in prairie-like associations. Andropogon gerardii and Rubus sp. had the highest frequency at 48%. Troy

Prairie is one of only three known A. gerardii populations in North or South Carolina >0.5 ha. The site contains upland to hydric conditions, and 42 (26%) of the species listed were obligate to facultative wetland. The mean soil pH was very strongly acidic (4.9). Phosphorus,

K and Ca levels were low for all three sampled depths for both upland and hydric sections, while there were medium levels of Mg. A portion of the site is under current consideration for a highway improvement project.

156

INTRODUCTION

The ongoing assessment of biodiversity plays an important role in both species and habitat protection. This is not only important in large scale settings, but also in smaller, peripheral habitats. It has been suggested that these peripheral populations have more genetic variation than those located in core areas, as a result of the variable conditions which induce differential selection, and thus help to maintain high genetic diversity (Safriel et al. 1999). In addition, due to the marginal conditions often found among these sites, populations that are small and isolated have higher between-population genetic diversity due to genetic drift. It is also more likely that these populations also evolve tolerances to extreme environmental conditions (Crawford 2008).

Several small, remnant Piedmont prairie sites have been documented for North and South

Carolina (Schafale and Weakley 1990; Davis et al. 2002; Schmidt and Barnwell 2002). These relict communities are thought to have been common during the pre-European settlement period (Barden 1997). Historically these sites were maintained by pre-settlement anthropogenic activity or periodic lightning fires. Although fire frequency is uncertain, it is thought to have been more common in the past (Barden 1997). Most of these remnant sites are now found along roadsides and utility rights-of-way that are regularly maintained to reduce or eliminate invading mid-successional species (Davis et al. 2002).

Although highly variable in species composition, these communities typically consist of prairie-like woodlands with scattered trees dominated by species such as Quercus stellata and Q. marilandica var. marilandica (Schafale and Weakley 1990). Grass species reported for these sites include Sorghastrum nutans, Tridens flavus, Andropogon virginicus var. virginicus, Tripsacum dactyloides var. dactyloides, Schizachyrium scoparium, Andropogon

157

gerardii, and various Panicum/Dichanthelium species (Davis et al. 2002). Compared with these other grass species, A. gerardii occurs less widely in the Piedmont of the Carolinas

(Radford et al. 1968).

Examples of rare forbs reported from these sites include Liatris squarrulosa, Silphium terebinthinaceum, Oligonueron album and the Federally endangered Helianthus schweinitzii

(Schweinitz‟s Sunflower) (Schafale and Weakley 1990; Davis et al. 2002; Schmidt and

Barnwell 2002). Even though H. schweinitzii is endemic to both North and South Carolina, it is only found within a radius of 100 km of Charlotte, N.C. in south-central N.C. and north- central S.C. To date, its ecology is still not well understood (Matthews et al. 1997).

The edaphic features that have been described for these prairie communities include unique soil characteristics that often restrict root growth, and thus impede or slow the establishment and subsequent successional development of forest communities. These would include the presence of shallow impermeable clay subsoils, that slow internal drainage and are subject to xeric-like conditions in the summer (Schafale and Weakley 1990; Davis et al. 2002). These extremely dry conditions also probably contributed to the pre-settlement maintenance of these sites by increasing the likelihood of periodic fires, reducing competition and leaf litter, and thus enhancing grass and forb recruitment.

These edaphic conditions may support xeric hardpan forests (Schafale and Weakley 1990).

Logan (1859) and Brown (1953) described these prairie communities growing back to stands of Quercus marilandica var. marilandica and Q. stellata in their observations of the upper

Piedmont of South Carolina. In addition, these sites may possess vertic soils (shrink-swell clays) that may cause poor drainage as well and produce subsequent tree-root disease

(Schafale and Weakley 1990; NatureServe 2009).

158

This study describes a recently located Piedmont prairie site (Troy Prairie) in Montgomery

County, N.C. The objectives of the study included: a complete floristic inventory of all vascular plants; determination of species frequency; characterization of the plant community structure; identification of any rare and threatened species; and the characterization of edaphic (soil) features. Although surveyed in 2003 and 2004 for rare species, this study was the first to provide a complete floristic list for the site. This study should also provide baseline information on Piedmont prairie sites, and the factors that contribute to their establishment and maintenance in the landscape.

MATERIALS AND METHODS

Study Area

Troy Prairie is located along a utility right-of-way in Montgomery County, in south-central

North Carolina (35.3514˚ N by 79.8798˚ W) approximately 3 km southeast of Troy, N.C. off of SR 1554 (Fig. 5.1). The site is approximately 5 ha with a north-south orientation. Of which, the northern portion has the highest elevation, and facing south it then slopes left to right along the length of the site. The northern end of the site grades into a small (5 m x 5 m) depression that has hydric conditions. This is one of two hydric sections that are fed by nearby seepage springs that run across the long axis of the site. The second hydric section is also located in a small depression approximately of the same dimensions as the first section, and is located at the southern end of the property (Fig. 5.2A).

159

Figure 5.1. Location of Troy Prairie, Montgomery County, North Carolina.

160

Figures 5.2.A. Troy Prairie hydric sections* and 5.2.B. Troy Prairie Soil Map Units.**

* = hydric areas noted by arrows. ** = Soil series map unit 475C = Herndon Series, map units 130 and 131 = Badin-Tarrus Complex.

The site is periodically maintained to remove woody vegetation along the utility right-of- way. Up until recently herbicidal application had been used for much of this maintenance.

However, due to the recent recognition of the site‟s unique flora, utility maintenance personnel have now eliminated the use of widespread herbicide application and instead have returned to the mechanical removal of woody species (L. Fogo USFW pers. comm. 2006).

The site is privately owned and surrounded by a larger pine plantation that has recently been logged, and a portion of the property is also located within an area that is under current consideration for a highway improvement project to N.C. Hwy. 24/27 (N.C. DOT TIP

Project No. R-623; Santec 2003).

Botanical Survey and Sampling

An initial survey of the site was performed in October 2003, and a number of rare and uncommon plant species were documented, including a large population of A. gerardii

(Santec 2003). Also, during the initial 2003 survey, an estimated population of 100 individuals of Helianthus schweinitzii was documented at the site (Santec 2003). A follow-up survey in the fall of 2004 found a significant reduction of these individuals due to what

161

appeared to be recent human disturbance (North Carolina Natural Heritage Program pers. comm. 2005).

The site was visited during regular intervals over the 2006-08 growing seasons to identify and collect all vascular plant species. Transect data were also taken during the 2007 summer and fall growing seasons. Seventy-five quadrats (1m x 1m) were established along two parallel transect lines along the long axis of the property. Quadrats were located at intervals of 10 m, alternating to the right and left along transects. All species were identified and voucher specimens collected (Appendix). Duplicate specimens were stored in the herbaria of

Clemson University and Belmont Abbey College. Botanical nomenclature follows Weakley

(2007). Presence/absence data for the sampled quadrats were used to determine species frequency values.

Soil Analysis

Six soil samples each were taken at random from both hydric and upland sections and analyzed for soil chemistry at three depths (0-10 cm, 11-20 cm, 21-30 cm; n=36). Laboratory soil analyses were conducted to determine soil pH; extractable phosphorus, potassium, calcium, magnesium, zinc, and manganese (Isaac 1983; Donohue 1992). An ANOVA was performed to compare the mean values for soil characteristics between the three depths for hydric and upland samples. Post hoc differences in means were determined using Tukey‟s

Test. All statistical analyses were performed with SAS v. 9.1. (SAS Institute 2002).

RESULTS

Botanical Survey and Sampling

A total of 163 vascular species were identified including 62 (38%) monocots, 97

162

(60%) dicots, 3 ferns and one Lycopsid (2%) (Table 6.1). There were fifty-seven

(35%) graminoids and 82 forbs (50%), with 39 (24%) Asteraceae and 11 (7%) Fabaceae species (Tables 6.1 and 6.2). The Asteraceae included Helianthus schweinitzii, Solidago gracillima, S. radula and S. patula var. patula. Both S. gracillima and S. radula are rare- listed species for North Carolina (North Carolina Natural Heritage Program 2006) and S. patula var. patula is uncommon for the Piedmont region of North Carolina (Weakley 2007).

In addition, 8 other species identified from the site were uncommon for North Carolina, and

11 others were uncommon for the Piedmont region of North Carolina (Weakley 2007;

Appendix). Of the graminoids there were 36 Poaceae, 17 Cyperaceae, and 4 Juncaceae.

Twenty-four (67%) Poaceae and 5 (24%) of the Cyperaceae/Juncaceae species were prairie species (Appendix). Uncommon graminoids for North Carolina for the site included A. gerardii, Aira caryophyllea, Carex atlantica and Sphenopholis pennsylvanica (Weakley

2007; Appendix).

The mean number of species per quadrat was 7 (± 2). Andropogon gerardii and

Rubus sp. had the highest frequency values at 48% each, followed by Dichantheliun dichotomum var. dichotomum with 38% (Table 5.3). Ten of 14 (71%) species with the highest frequency values were species with prairie affinities (Table 5.3), with a total of eighty-seven (53%) of the species reported as occurring in prairies (Table 5.2).

Forty-two (26%) species were either obligate (n=18) or facultative wetland (n=24), including the rare obligate wetland species Solidago gracillima (Kartesz 1996; Table 5.2).

Sixty-seven (41%) species had either facultative, facultative upland, or upland wetland statuses, and 54 (33%) were undesignated (Kartesz 1996; Table 5.2). Only five plants from the estimated original 100 H. schweinitzii individuals were found during the 2008 growing

163

season.

Table 5.1. Taxonomic categories for vascular plant species for Troy Prairie.

Category Species % Species Genera Families Fern/Lycopsid 4 02 3 3

Dicot 97 60 70 29

Asteraceae 39 24 24 n/a

Fabaceae 11 07 8 n/a

Monocot 62 38 39 8

Graminoid 57 35 34 3

164

Table 5.2. Species richness of vascular plants for Troy Prairie in relation to life form,

habitat affinity, and wetland status.

Grouping Type Number of Species % of Species Lifeform Forb 82 50 Grass 36 22 Sedge/Rush 21 13 Shrub 9 6 Tree 7 4 Vine 4 2 Fern/Lycopsid 4 2

Habitat Affinity Prairie 87 53 Wetland 42 24 Non-native 8 5 Other 28 17

Wetland Status UPL 1 .06 FACU 25 15 FAC 41 25 FACW 24 15 OBL 18 11 UD 54 33

(UPL) = Upland, (FACU+, FACU, FACU-) = Facultative Upland, (FAC+, FAC, FAC-) = Facultative, (FACW) = Facultative Wetland, (OBL) = Obligate Wetland, (UD) = Undesignated. “Wetland” species have FACW or OBL indicator statuses according to Kartesz 1996 (note: prairie and wetland designations are not necessarily mutually exclusive to one another).

165

Table 5.3. Frequencies for associate species for Troy Prairie from sampled

quadrats (n=75) with the highest values (* notes prairie species).

Species Frequency Value Andropogon gerardii * 48% Rubus sp.* 48% Dichanthelium dichotomum var. dichotomum* 32% Tephrosia virginiana* 31% Rhus copallinum* 31% Pteridium aquilinum 31% Coreopis major var. rigida* 27% Symphyotrichum dumosum* 25% Dichanthelium scoparium 24% Silphium compositum var. compositum* 21% Quercus stellata* 17% Liquidambar styraciflua 15% Toxicodendron radicans* 15% Vitis rotundifolia var. rotundifolia 13%

Soil Analysis

There were no significant differences in soil nutrient levels for hydric vs. upland samples for all three sampled depths, except for Ca at the 20 cm depth. Overall soil nutrient levels were of moderate to very low values. Phosphorous levels had very low nutrient values, and nutrient levels were low for both K and Ca, while Mg had medium nutrient values (Clemson

University Agricultural Service Laboratory 2008; Table 5.4). The soils were very acidic with a mean pH of 4.9 (Tisdale et al. 1993; Table 5.4). The soils belong to the Badin-Tarrus series complex and Herndon Soil series (Fig. 5.2.B). Both are well drained, moderately permeable, and have low to moderate water capacity. They were derived from fine-grained, metavolcanic rock parent material (Soil Survey of Montgomery County, North Carolina 2006).

166

Table 5.4. Comparison of upland vs. hydric soil chemical nutrient means at three

sampled depths at Troy Prairie.

Soil Depth Parameter 0-10 cm 11-20 cm 21-30 cm Mean Measured pH upland 4.9A 5.0A 4.9A 4.93 pH hydric 4.9A 4.8A 4.9A 4.86 P upland 6.3A (VL) 4.3A (VL) 2.3A (VL) 4.3 (VL) P hydric 4.9A (VL) 2.8A (VL) 2.3A (VL) 3.33 (VL) K upland 72.3A (M) 41.1A (L) 42.0A (L) 51.8 (L) K hydric 78.1A (M) 32.8A (L) 34.1A (L) 48.3 (L) Ca upland 329.3A (L) 193.0B (L) 204.6A (L) 242.2 (L) Ca hydric 468.8A (M) 268.1A (L) 264.0A (L) 333.6 (L) Mg upland 51.0A (M) 32.0A (L) 47.5A (M) 43.5 (M) Mg hydric 73.1A (M) 45.5A (M) 48.5A (M) 55.7 (M) Zn upland 6.7A 2.4A 2.0A 3.7 Zn hydric 7.1A 4.1A 3.7A 4.96 Mn upland 48.5A 21.1A 16.8A 28.8 Mn hydric 72.1A 25.0A 19.8A 38.9

Superscripts denote significant different means between each soil property comparison of upland vs. hydric at each depth within a column (Tukey‟s Test, P = 0.05). (L = low nutrient levels, VL = very low nutrient levels, M = medium nutrient levels)

167

DISCUSSION

Troy Prairie is an important Piedmont prairie remnant site with high species richness. In addition, the presence of rare species, such as Helianthus schweinitzii, Solidago gracillima, and S. radula, also further contributes to the site‟s uniqueness and is consistent with earlier reported descriptions of rare endemics occurring within Piedmont prairies (Schafale and

Weakley 1990; Davis et al. 2002; Schmidt and Barnwell 2002). In addition to being rare- listed for North Carolina, both S. gracillima and S. radula were new county records for

Montgomery County, N.C. (North Carolina Natural Heritage Program 2006). The occurrence of both S. radula and S. patula var. patula was particularly unexpected when they are reported as occurring over mafic soils (Weakley 2007).

The large number of prairie species identified at the site was consistent with previous reports of species with prairie affinities within those communities (Schafale and Weakley

1990; Davis et al. 2002; Schmidt and Barnwell 2002). In addition, a large proportion of those species are commonly found in prairie associations in Midwestern, mid-atlantic, and southern mesic prairies of the U.S. (Henderson 1995; Webb et al. 1997; Leidolf and McDaniel 1998;

Edgin and Ebinger 2000; Klips 2003; Laughlin and Uhl 2003; Edgin et al. 2005).

The presence of a large graminoid component is also indicative of prairie communities.

Furthermore, five of the grasses at the site are dominant in Midwestern prairies, including A. gerardii, Schizachyrium scoparium, Tripsacum dactyloides var. dactyloides, Tridens flavus, and Sorghastrum nutans (Henderson 1995; Edgin and Ebinger 2000).

The A. gerardii population at Troy Prairie is one of only three known populations >0.5 ha for both North and South Carolina. The presence of A. gerardii as a dominant species at the site is consistent with its dominant role in prairie communities of the Midwest (Rosburg

168

1997; Silletti and Knapp 2001; Silletti et al. 2004). In addition to A. gerardii, the occurrence of the uncommon graminoids Aira caryophyllea, Carex atlantica and Sphenopholis pennsylvanica at the site is of significance for North Carolina (Weakley 2007).

Even though the soils on the site did not belong to mapping units with typical hydric soils and redoxomorphic features, soil moisture at the surface was, however, sufficient enough in those sections to support a large percent of obligate and facultative wetland species. The large number of wetland species was unexpected because the wet sections constituted <5% of the total site. The presence of Solidago gracillima warrants special concern because of its rarity in N.C. and its small population size (<20), restricted to the two wet areas of the site.

Other uncommon North Carolina wetland species identified from the site were Asclepias rubra and Carex atlantica. Also, this is only the second documented Piedmont prairie community in both North and South Carolina to possess hydric conditions with a large proportion of hydrophytic species.

In addition to the disturbance noted at the site in 2004, the more recent clear-cut logging of the surrounding area of the site may have further contributed to the significant reduction of

Helianthus schweinitzii individuals found during the 2008 growing season. As a result, the author suggests that the population be monitored and a management strategy implemented to both stabilize and potentially increase the current number of individuals of the population.

The soils on the site did not possess vertic soils, and are more permeable than those reported for xeric hardpan forest communities. However, the occurrence of species such as

Quercus stellata, Q. marilandica var. marilandica, Helianthus schweinitzii, Eryngium yuccifolium var. yuccifolium, Sorghastrum nutans, Hypericum virgatum, Rhynchosia tomentosa var. tomentosa, Silphium compositum var. compositum and Carphephorus

169

bellidifolius is consistent with reported descriptions of woodland-prairie communities with xeric conditions (Brown 1953; Logan 1859; Schafale and Weakley 1990; Weakley 2007;

Natureserve 2009). In particular, C. bellidifolius is reported as occurring in the driest of habitats in the Carolinas (Weakley 2007). In addition, the presence of Q. marilandica var. marilandica is also indicative of very poor soil conditions often associated with these xeric communities (Weakley 2007).

The floristic composition at the site does exhibit adaptation to edaphic extremes in soil moisture regimes, based on the presence of both wetland and xeric-like species occurrence. Thus, the presence of both hydric to upland habitat conditions at the site has contributed to its high overall species richness. This, coupled with periodic thinning of woody vegetation that has eliminated or reduced mid-successional species, has created optimal conditions for prairie species establishment and maintenance.

With its high species richness, presence of Helianthus schweinitzii and other rare species, as well as a large population of A. gerardii, Troy Prairie warrants long term maintenance and study. I recommend that Troy Prairie be monitored on both a regular and long term basis to determine any species compositional changes that may be occurring at the site, in particular those that might negatively impact species richness. In addition, it is suggested that that the site‟s ecological uniqueness be considered in any future plans involving highway construction on or near the site that might alter the site‟s flora in any form.

170

LITERATURE CITED

Barden, L. S. 1997. Historic prairies in the Piedmont of North and South Carolina, USA.

Natural Areas Journal 17:149-152.

Brown, D. S. 1953. A city without cobwebs: Rock Hill, South Carolina. University of

South Carolina Press, Columbia, South Carolina. 334 pp.

Clemson University Agricultural Service Laboratory. 2008. Clemson University, Clemson,

South Carolina (http://www.clemson.edu/agsrvlb/).

Crawford, R. M. M. 2008. Plants on the Margin: Ecological Limits and Climatic Change. Ch.

2 pp. 31-59. Cambridge University Press. Cambridge, UK.

Davis, Jr., J. E., C. McRae, B. L. Estep, L. S. Barden and J. F. Matthews. 2002. Vascular

flora of Piedmont prairies: evidence from several prairie remnants. Castanea 67:1-12.

Donohue, S. J. 1992. Reference Soil and Media Diagnostic Procedures for the Southern

Region of the United States. Southern Cooperative Bulletin No. 374.

Edgin, B. R. and J. E. Ebinger. 2000. Vegetation of a successional prairie at Prairie Ridge

State Natural Area, Jasper County, Illinois. Castanea 65:139-146.

Edgin, B. R., R. Beadles and J. E. Ebinger. 2005. Vascular flora of Beadles Barrens Nature

Preserve, Edwards County, Illinois. Castanea 70:47-58.

Henderson, R. A. 1995. Plant species composition of Wisconsin prairies. An aid to selecting

species for planting and restorations based upon University of Wisconsin-Madison

Plant Ecology Laboratory data. Technical Bulletin No. 188. Department of Natural

Resources. Madison, Wisconsin.

Isaac, R. A. 1983. Reference Soil Test Methods for the Southern Region of the United States.

Southern Cooperative Series Bulletin No. 289.

171

Kartesz, J. T. 1996. National List of Vascular Plant Species that Occur in Wetlands:Biota of

North America Program (BONAP) Chapel Hill, North Carolina.

Klips, R. A. 2003. Vegetation of Claridon Railroad Prairie, a remnant of the Sandusky Plains

of Central Ohio. Castanea 68:135-142.

Laughlin, D. C. and C. F. UHL. 2003. The xeric limestone prairies of Pennsylvania. Castanea

68:300-316.

Leidolf, A. and S. McDaniel. 1998. A floristic study of Black Prairie plant communities at

Sixteen Section Prairie, Oktibbeha County, Mississippi. Castanea 63:51-62.

Logan, J. H. 1859. A History of the Upper Country of South Carolina from the Earliest

Periods to the Close of the War of Independence. S.G. Courtenay and Co., Charleston,

South Carolina. 521 pp.

Matthews, J. F., L. S. Barden and C. R. Matthews. 1997. Corrections of the chromosome

number, distribution and misidentifications of the Federally endangered sunflower,

Helianthus schweinitzii T & G. Journal of the Torrey Botanical Society 124:198-209.

Natureserve. 2009. NatureServe Explorer: An online encyclopedia of life [web

application]. Version 7.1. NatureServe. Arlington, Virginia. Available

http://www.natureserve.org/explorer. (Accessed: October 7, 2009).

North Carolina Natural Heritage Program. 2006. Natural Heritage Program List of Rare Plant

Species of North Carolina. Franklin, M.A. and J.T. Finnegan (eds.) North Carolina

Natural Heritage Program Office of Conservation and Community Affairs. North

Carolina Department of Environment and Natural Resources. Raleigh, North Carolina.

Radford, A. E., H. E. Ahles, and C. R. Bell. 1968. Manual of the Vascular Flora of the

Carolinas. University of North Carolina Press. Chapel Hill, North Carolina.

172

Safriel, U. N., Volis, S. and S. Kark. 1994. Core and peripheral populations and global

climate change. Israel Journal of Plant Sciences 42:331-345.

Santec Consulting 2003. Ayrsley Town Blvd. Suite 300. Charlotte, North Carolina 28273.

SAS Institute. 2002. SAS User‟s Guide: Statistics. Version 9.1. SAS Institute, Cary, North

Carolina.

Schafale, M. P. and A. S. Weakley. 1990. Classification of the Natural Communities of

North Carolina. Third Approximation. North Carolina Natural Heritage Program.

Division of Parks and Recreation. North Carolina Department of Environment, Health,

and Natural Resources. Raleigh, North Carolina.

Schmidt, J. M. and J. A. Barnwell. 2002. A flora of the Rock Hill Heritage Preserve, York

County, South Carolina. Castanea 67:247-279.

Soil Survey OF Montgomery County, North Carolina. 2006. United States Department of

Agriculture. Soil Conservation Service.

Silletti, A. M. and A. K. Knapp. 2001. Responses of the codominant grassland species

Andropogon gerardii and Sorghastrum nutans to long-term manipulation of nitrogen

and water. American Midland Naturalist 145:159-167.

Silletti, A. M., A. K. Knapp, and J. M. Blair. 2004. Competition and coexistence in grassland

codominants: Responses to neighbor removal and resource availability. Canadian

Journal of Botany 82:450-460.

Tisdale, S. L., W. L. Nelson, J. D. Beaton and J. L. Havlin 1992. Soil acidity and basicity. pp.

364-404. In Soil Fertility and Fertilizers, 5th ed. Macmillan Publ. New York.

173

Weakley, A. S. 2007. Flora of the Carolinas, Virginia, Georgia, and Surrounding Areas,

working draft of January 2007. University of North Carolina. Chapel Hill, North

Carolina.

Webb, D. H., H. R. Deselm and W. M. Dennis. 1997. Studies of prairie barrens of

Northwestern Alabama. Castanea 62:173-184.

174

APPENDIX

Species list for Troy Prairie, Montgomery County, North Carolina.

P = Prairie Species; UCP = Uncommon for Piedmont region of North

Carolina; UCNC = Uncommon for North Carolina; NCR = North Carolina Rare-

Listed Species; USE = United States Endangered; (UPL) = Upland; (FACU+,

FACU, FACU-) = Facultative Upland; (FAC+, FAC, FAC-) = Facultative; (FACW)

= Facultative Wetland; (OBL) = Obligate Wetland. RT (Robert Tompkins) =

Collector‟s initials followed by accession number. All specimens are housed at CLEMS=

Clemson University Herbarium and Belmont Abbey College.

ALISMATACEAE

Sagittaria latifolia Willdenow var. latifolia [OBL] (RT0352)

ALTINGIACEAE

Liquidambar styraciflua L. [FAC+] (RT0662)

ANACARDIACEAE

Rhus copallinum L. var. copallinum [P, FAC-] (RT0669)

Toxicodendron radicans (L.) Kuntze var. radicans [P, FAC] (RT0667)

APIACEAE

Angelica venenosa (Greenway) Fernald (RT0754)

Eryngium yuccifolium Michx. var. yuccifolium [P, UCNC, FAC] (RT0517)

APOCYNACEAE

Asclepias rubra L. [UCNC, OBL] (RT0520)

Asclepias tuberosa L. var. rolfsi (Britton ex Vail) Shinners [P, UCNC] (RT0139)

175

ASTERACEAE

Achillea millefolium L. [P, FACU] (RT0668)

Ageratina altissima King & H.E. Robinson var. altissima [P] (RT0755)

Ambrosia artemisiifolia L. [P, FACU] (RT0663)

Carphephorus bellidifolius (Michx.) Torrey & A. Gray (RT0528)

Chrysogonum virginianum L. var. virginianum [P] (RT0447)

Coreopsis major Walter var. rigida [P] (RT0158)

Coreopsis tinctoria Nuttall var. tinctoria [P, UCP, FAC] (RT0523)

Erechites hieracifolius (L.) Rafinesque ex de Candolle [P, FAC-] (RT0591)

Erigeron strigosus var. strigosus Muhlenberg ex Willdenow [P, FAC] (RT0666)

Eupatorium album L. var. album [P, UCP] (RT0360)

Eupatorium capillifolium (Lamarck) Small [P, FACU] (RT0665)

Eupatorium hyssopifolium L. [P] (RT0682)

Eupatorium rotundifolium L. [P, UCP, FAC] (RT0365)

Euthamia caroliniana (L.) Greene ex Porter & Britton (RT0596)

Eutrochium fistulosum (Barratt) E.E. Lamont (RT0569)

Helianthus angustifolius L. [UCP, FAC+] (RT0530)

Helianthus atrorubens L. [P] (RT0332)

Helianthus divaricatus L. [P] (RT0325)

Helianthus schweinitzii Torrey & A. Gray [P, NCR, USE] not collected

Krigia dandelion (L.) Nuttall [P, FACU] (RT0449)

Liatris pilosa (Aiton) Willdenow [P] (RT0664)

Marshellia obovata (Walter) Beadle & F.W. Boynton var. obovata (RT0468)

176

Mikania scandens (L.) Willdenow [FACW] (RT0589)

Parthenium integrifolium L. var. integrifolium [P] (RT0156)

Pityopsis asperis (Shuttleworth ex Small) Small var. adenolepsis (RT0311)

Pseudognaphalium obtusifolium (L.) Hilliard & Burtt [P] (RT0595)

Seriocarpus linifolius (L.) Britton, Sterns & Poggenburg [P] (RT0330)

Silphium compositum Michx. var. compositum [P] (RT0629)

Solidago gigantea Aiton [P, FACW] (RT0649)

Solidago gracillima Torrey & A. Gray [OBL, NCR] (RT0322)

Solidago nemoralis Aiton var. nemoralis [P] (RT0344)

Solidago odora Aiton var. odora [P] (RT0343)

Solidago patula Muhlenberg ex Wildenow var. patula [UCP, FACU] (RT0768)

Solidago pinetorum Small [P] (RT0336)

Solidago radula Nuttall [P, NCR] (RT0352)

Symphyotrichum concolor (L.) Nesom var. concolor (RT0616)

Symphyotrichum dumosum L. [P, FAC] (RT0341)

Symphyotrichum grandiflorum (L.) Nesom (RT0363)

Vernonia noveboracensis (L.) Michx. [FAC+] (RT0567)

BETULACEAE

Alnus serrulata (Aiton) Wildenow [FACW] (RT0326)

BIGNONIACEAE

Campsis radicans (L.) Seeman ex Bureau [P, FAC] (RT0635)

CAMPANULACEAE

Lobelia glandulosa Walter [FAC] (RT0329)

177

Lobelia nuttallii J.A. Schultes [FACW] (RT0152)

Triodanis perfoliata (L.) Nieuwland [P, FACU+] (RT0462)

CONVOLVULACEAE

Ipomoea pandurata (L.) G.F.W. Meyer [P, FACU] (RT0567)

CYPERACEAE

Carex atlantica Bailey [UCNC, FACW] (RT0460)

Carex complanata Torrey & Hooker [P, FAC+] (RT0154)

Carex crinita Lamarck var. crinita [FACW] (RT0146)

Carex flaccosperma Dewey [FAC+] (RT0651)

Carex intumescens Rudge var. intumescens [FACW] (RT0142)

Carex lurida Wahlenberg [OBL] (RT0170)

Carex scoparia ScKuhr ex Willdenow var. scoparia [FACW] (RT0756)

Cyperus erythrorhizos Muhlenberg [OBL] (RT0592)

Eleocharis tuberculosa (Michx.) Roemer & J.A. Schultes [UCP, FACW] (RT0347)

Fimbristylis autumnalis (L.) Roemer & J.A. Schultes [OBL] (RT0353)

Fuirena squarrosa Michx. [OBL] (RT0320)

Rhynchospora glomerata (L.) Vahl var. glomerata [OBL] (RT0323)

Rhynchospora gracilenta A. Gray [UCP, OBL) (RT0350)

Rhynchospora recognita (Gale) Kral (RT0317)

Scirpus atrovirens Willdenow [P, OBL] (RT0634)

Scleria muehlenbergii Steudel (RT0307)

Scleria triglomerata Michx. [P, FAC] (RT0149)

178

DENNSTAEDTIACEAE

Pteridium aquilinum (L.) Kuhn var. pseudocaudatum [FAC] (RT0167)

EBENACEAE

Diospyros virginiana L. [P, FAC] (RT0630)

ERICACEAE

Lyonia ligustrina (L.) Augustin de Candolle var. foliosiflora (Michx.) Fernald [FACW] (RT0153)

Lyonia mariana L. [UCP, FAC] (RT0461)

Vaccinium tenellum Aiton [P, UCP, FACU] (RT0349)

FABACEAE

Baptisia tinctoria (L.) Ventenat (RT0177)

Clitoria mariana L. var. mariana (RT0521)

Desmodium paniculatum (L.) Augustin de Candolle var. paniculatum [P, UCP, FACU] (RT0588)

Lespedeza bicolor Turczaninow (RT0314)

Lespedeza cuneata (Dumont-Cours.) G. Don [UPL] (RT0652)

Lespedeza procumbens Michx. [P] (RT0633)

Lespedeza virginica (L.) Britton [P] (RT0653)

Mimosa microphylla Dryander (RT0646)

Rhynchosia tomentosa (L.) Hooker & Arnott var. tomentosa [P] (RT0164)

Stylosanthes biflora L. Britton, Sterns, & Poggenburg [P] (RT0310)

Tephrosia virginiana (L.) Persoon [P] (RT0701)

FAGACEAE

Quercus marilandica Muenchhausen var. marilandica [P] (RT0631)

179

Quercus stellata Wagenheim [P, FACU] (RT0753)

GELSEMIACEAE

Gelsemium sempervirens St. Hilaire [FAC] (RT0451)

HYPERICACEAE

Hypericum drummondii (Greville & Hooker) Torrey & A. Gray [P, UCNC, FACU] (RT0315)

Hypericum gentianoides (L.) Britton, Sterns, & Poggenburg [P, FACU] (RT0309)

Hypericum virgatum Lamarck (RT0566)

IRIDACEAE

Iris verna L. var. verna [P] (RT0452)

JUNCACEAE

Juncus acuminatus Michx. [OBL] (RT0757)

Juncus effusus L. var. soltus [P, FACW] (RT0758)

Juncus scirpoides Lamarck var. scirpoides [FACW] (RT0150)

Juncus tenuis Willdenow [P, FAC] (RT0632)

LAMIACEAE

Prunella vulgaris L. [P, FAC-] (RT0623)

Pycanthemum tenuifolium Schrader [P, FAC-] (RT0316)

Salvia lyrata L. [P, FAC-] (RT0661)

LYCOPODIACEAE

Lycopodiella alopecuroides (L.) Cranfill [OBL] (RT0573)

MAGNOLIACEAE

Liriodendron tulipfera L. var. tulipfera [FACU] (RT0656)

180

MELASTOMATACEAE

Rhexia virginica L. [UCP, FACW] (RT0178)

MYRICACEAE

Morella caroliniensis (P. Miller) Small (RT0459)

NARTHECIACEAE

Aletris farinosa L. [FAC+](RT0464)

OLEACEAE

Fraxinus sp. L. [P] (RT0660)

ONAGRACEAE

Ludwigia decurrens Walter [OBL] (RT0312)

Oenothera biennis L. [P, FACU] (RT0642)

ORCHIDACEAE

Spiranthes vernalis Engelmann & A. Gray [FACW, UCP] (RT0519)

OSMUNDACEAE

Osmunda cinnamonea L. var. cinnamomea [FACW] (RT0638)

Osmunda regalis L. var. spectabilis (Willdenow) A. Gray [OBL] (RT0763)

PHRYMACEAE

Mimulus alatus Aiton [OBL] (RT0327)

POACEAE

Aira caryophyllea L. [UCNC] (RT0466)

Andropogon gerardii Vitman [P, UCNC, FAC] (RT0671)

Andropogon glomeratus (Walter) Britton, Sterns & Poggenburg var. glomeratus [P, FACW] (RT0600)

Andropogon virginicus L. var. virginicus [P, FAC-] (RT0434)

181

Anthoxanthum odoratum L. (RT0659)

Aristida dichotoma Michx. [FACU] (RT0372)

Aristida oligantha Michx. [P] (RT0342)

Arundinaria gigantea (Walter) Walter [FACW] (RT0371)

Chasmanthium laxum (L.) Yates [FACW] (RT0319)

Cinna arundinacea L. [P, FACW] (RT0335)

Cynodon dactylon (L.) Persoon var. dactylon [FACU] (RT0594)

Danthonia sericea Nuttall [P, FAC-] (RT0654)

Danthonia spicata (L.) Palisot de Beauvois [P] (RT0471)

Dichanthelium acuminatum (Swartz) Gould & Clark var. fasciculatum (Torrey) [P, FAC] (RT0168)

Dichanthelium clandestinum (L.) Gould [P, FACW] (RT0643)

Dichanthelium dichotomum (L.) Gould var. dichotomum [P, FAC] (RT0157)

Dichanthelium scoparium (L.) Gould [FACW] (RT0394)

Dichanthelium sphaerocarpon (Elliott) Gould [P, FAC] (RT0469)

Dichanthelium villosissimum (Nash) Freckmann var. villosissimum [P] (RT0522)

Elymus virginicus L. var. virginicus [P, FAC] (RT0658)

Eragrostis spectabilis (Pursh) Steudel [P, FAC-] (RT0308)

Festuca eliatior L. (RT0644)

Holcus lanatus L. [FAC-] (RT0636)

Leersia oryzoides (L.) Swartz [OBL] (RT0321)

Panicum anceps Michx. var. anceps [P, FAC-] (RT0313)

Panicum virgatum L. var. cubense Grisebach [P, FAC+] (RT0515)

182

Panicum virgatum L. var. virgatum [P, FAC+] (RT0565)

Paspalum setaceum Michx. [P, FAC] (RT0577)

Pennisetum glaucum L.R. Brown [P] (RT0637)

Saccharum giganteum (Walter) Persoon [P, FACW] (RT0331)

Schizachyrium scoparium (Michx.) Nash var. scoparium [P, FACU] (RT0598)

Sorghastrum nutans (L.) Nash [P, FACU] (RT0645)

Sphenopholis pennsylvanica (L.) A.S. Hitchcock [UCNC, OBL] (RT0465)

Tridens flavus (L.) A.S. Hitchcock [P, FACU] (RT0590)

Tripsacum dactyloides (L.) L. var. dactyloides [P, FAC+] (RT0657)

Vulpia myuros (L.) K.C. Gmelin [FACU] (RT0525)

POLEMONIACEAE

Phlox carolina L. carolina [FACU] (RT0174)

Phlox nivalis var. nivalis [P] (RT0771)

POLYGALACEAE

Polygala incarnata L. [FAC-] (RT0572)

RHAMNACEAE

Ceanothus americanus L. var. americanus [P] (RT0145)

ROSACEAE

Crataegus macrosperma Ashe (RT0446)

Crataegus uniflora Muenchhausen [P] (RT0348)

Prunus serotina Ehrhart var. serotina [P, FACU] (RT0640)

Rubus sp. L. [P] (RT0760)

183

RUBIACEAE

Diodia teres Walter [P, FACU-] (RT0339)

Diodia virginiana L. [FACW] (RT0641)

Houstonia caerula L. [P, FAC] (RT0448)

VIOLACEAE

Viola primulifolia L. [FACW] (RT0457)

VITACEAE

Vitis aestivalis Michx. var. aestivalis [P, FAC-] (RT0496)

Vitis rotundifolia Michx. var. rotundifolia [FAC] (RT0639)

XYRIDACEAE

Xyris jupicai L.C. Richard [OBL] (RT0354)

184

CONCLUSIONS

Plant communities that contain A. gerardii populations in the Carolinas are highly diverse in terms of species composition, edaphic (soil) factors, physiography, and land use history.

As a result, it is problematic to assign a strict community type to these ecosystems. The majority of the sites, however, have high species richness. Many are characterized by having an assemblage of prairie-associated species, often with a large proportion of composite and graminoid species. Individuals of A. gerardii appear well adapted to various soil conditions.

Many of the A. gerardii populations in the Carolinas have most likely been fragmented and may be potentially threatened with extinction. Those populations appear to have relatively high levels of genetic diversity and to reproduce primarily via vegetative reproduction. The data from this study suggest that viable seed production may be inhibited, particularly in smaller populations. However, despite population fragmentation, individuals within current populations are so-called „proven survivors‟ and have been selected from cohorts. Those individuals most likely have higher fitness and have been competitively successful compared to other individuals that are now extinct. Despite relatively low rates of viable seed production and reduced subsequent germination, these smaller eastern A. gerardii populations most likely persist due to periodic disturbance events that eliminates mid- successional competition from other species. However, once these populations are reduced to a certain point, they may be unable to re-colonize sexually. As a result, human intervention may be necessary to re-introduce new genotypes into those populations to increase the likelihood of their continued existence.

185

(ANDROPOGON GERARDII)&lt;/I&gt;

(ANDROPOGON GERARDII)</I>