<%BANNER%>

Seed Banking and Vesicular-Arbuscular Mycorrhizae in Pasture Restoration in Central Florida


PAGE 1

SEED BANKING AND VESICULAR-ARBUSCULAR MYCORRHIZAE IN PASTURE RESTORATION IN CENTRAL FLORIDA By AMY MILLER JENKINS A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2003

PAGE 2

Copyright 2003 by Amy Miller Jenkins

PAGE 3

To my husband, Michael, who is always an inspiration!

PAGE 4

ACKNOWLEDGMENTS This study would not have been possible without the help of many friends and advisors. I would like to acknowledge my two committee members, Dr. Doria Gordon and Dr. Kaoru Kitajima for their guidance, wisdom, and inspiration. I would especially like to thank Dr. Doria Gordon for being a wonderful mentor over the past five years. I also thank Dr. Kaoru Kitajima for her valuable insight and use of her laboratory space and equipment for all of my lab work. I thank the graduate students and staff of the School of Natural Resources and Environment (SNRE) and the Botany Department as well as Dr. Humphrey, the Dean of SNRE for his insightfulness and advice. SNRE provided a research assistantship which made my graduate program possible. The Nature Conservancy (TNC) and the staff at the Disney Wilderness Preserve and Gainesville Office have made this research possible through generous use of their land and facilities for my research purposes. Dr. David Sylvia and Dr. James Graham provided invaluable help with the mycorrhizae chapter in experimental design and results interpretation. Abid Alagely and Sarah Bray provided technical assistance in mycorrhizae structural identification and methods. Nancy Bissett (The Natives Inc.) and Michael Byrne (TNC) provided tremendous help with plant identification. I thank Richard Abbott and Kent Perkins for help with seed identification and use of the UF herbarium facilities. Dr. Doug Levey gave advice on the direct count seed bank methods. The University of Florida School of Forest Resources provided shadehouse space that made the seed bank germination trials possible. Sarah Bray, iv

PAGE 5

Catherine Cardels, and Michael Jenkins provided helpful editorial comments on this thesis. I thank Clea Paz and Sarah Bray for their support and friendship throughout this graduate experience. I would particularly like to thank my parents for their love, support, and encouragement to pursue graduate school. Mostly I thank Michael, my husband, for his endless love, support, and assistance both in my coursework and with my research. He happily provided countless hours of fieldwork, laboratory assistance, editing, brainstorming, shadehouse construction, coffee making, and much more! v

PAGE 6

TABLE OF CONTENTS Page ACKNOWLEDGMENTS.................................................................................................iv LIST OF TABLES...........................................................................................................viii LIST OF FIGURES...........................................................................................................ix ABSTRACT.......................................................................................................................xi CHAPTER 1 GENERAL INTRODUCTION AND SITE DESCRIPTION......................................1 2 SEED BANKING IN A PASTURE RESTORATION SITE AND MESIC PINE FLATWOODS IN CENTRAL FLORIDA...................................................................9 Introduction...................................................................................................................9 Methods......................................................................................................................13 Germination Assay..............................................................................................14 Direct Count Assay..............................................................................................17 Vegetation Survey...............................................................................................18 Data Analysis.......................................................................................................18 Results.........................................................................................................................20 Germination Assay..............................................................................................20 Direct Count Assay..............................................................................................22 Vegetation Survey...............................................................................................24 Aristida stricta Recruitment................................................................................25 Methods Comparison...........................................................................................25 Discussion...................................................................................................................26 Methods Comparison...........................................................................................30 Management Implications...................................................................................32 3 VESICULAR-ARBUSCULAR MYCORRHIZAE IN UPLAND RESTORATION SITES, MESIC PINE FLATWOODS, AND PASPALUM NOTATUM PASTURES IN CENTRAL FLORIDA..........................................................................................40 Introduction.................................................................................................................40 Methods......................................................................................................................45 Experiment 1: First-year and Pre-restoration Complexes...................................46 Experiment 2: Five-year Complex......................................................................49 vi

PAGE 7

Data Analysis.......................................................................................................51 Experiment 1................................................................................................51 Experiment 2................................................................................................52 Results.........................................................................................................................53 Experiment 1.......................................................................................................53 Experiment 2.......................................................................................................54 Discussion...................................................................................................................55 Experiment 1.......................................................................................................55 Experiment 2.......................................................................................................58 Conclusions/Applications....................................................................................61 APPENDIX SPECIES LISTS...........................................................................................72 LIST OF REFERENCES...................................................................................................82 BIOGRAPHICAL SKETCH.............................................................................................91 vii

PAGE 8

LIST OF TABLES Table page 2-1 Species in common between the restoration and flatwoods soils and unique species in the flatwoods and restoration soils in the germination assay..................34 3-1 Composition of the modified Hoaglands fertilizer solution.....................................63 3-2 Spearman correlation matrix for MIP and soil chemistry in the pre-restoration complex....................................................................................................................63 3-3 Spearman correlation matrix for MIP and soil chemistry variables in the one-year complex....................................................................................................................64 3-4 Results of two-way nested ANOVA for each soil chemistry variable on restoration treatment, elevation and their interaction within the one-year complex....................................................................................................................64 3-5 Results of Tukey mean comparisons of the three treatments in Experiment 1 for each significant variable...........................................................................................65 3-6 Results of the Tukey mean comparisons of the five elevation strata in Experiment 1 for significant variables.....................................................................65 3-7 P-values from the Spearmans correlation matrix of eight soil variables in Experiment 2............................................................................................................65 A-1 Species list for the germination assay with life form and category..........................73 A-2 Species list for the direct count assay with life form and category..........................75 A-3 Species list for the field vegetation sampling by elevation strata (n=5 strata) and frequency by plot size..............................................................................................76 A-4 Species observed in fifteen healthy pine flatwoods surveyed at DWP in 1997.......80 viii

PAGE 9

LIST OF FIGURES Figure page 1-1 Location of the The Nature Conservancys Disney Wilderness Preserve within the central Florida region...........................................................................................8 2-1 The mean number of seedlings per sample (+1 SE) in the germination assay.........35 2-2 Mean species accumulation curves of observed species richness for the restoration and flatwoods soils.................................................................................35 2-3 Relative abundance of species categories at the 5 elevation strata..........................36 2-4 Mean (+1 SE) density of Aristida stricta (m-2, n=5) for each of the 5 elevation strata six months post-seeding in the first-year restoration site...............................37 2-5 Comparison of the mean species accumulation curves of observed species richness for the restoration soils...............................................................................38 2-6 Number of seeds/seedlings per m2...........................................................................39 3-1 Correlation between MIP (n = 5) and pH (n = 3) for soils from flatwoods and one-year unit restoration soils in June 2002.............................................................66 3-2 Mean mycorrhizal inoculation potential (+1 SE, n = 25) of the one-year unit, and adjacent flatwoods and pasture..........................................................................66 3-3 Mean inoculation potential (+1 SE, n = 15) of the 5 elevation strata in June 2002..........................................................................................................................67 3-4 Mean gravimetric soil moisture content (+1 SE, n = 5) for each treatment by elevation strata in the one-year complex..................................................................67 3-5 Mean pH (+1 SE, n = 3) for each treatment by elevation in the one-year complex....................................................................................................................68 3-6 Mean soil P content (mg/kg) (+1 SE, n=3) for each treatment in the one-year complex....................................................................................................................68 3-7 Mean soil total Kjeldahl nitrogen (g/kg) (+1 SE, n=3) for each treatment in the one-year complex.....................................................................................................69 ix

PAGE 10

3-8 Percent colonization (n = 1) of Aristida stricta roots by VAM fungi in the six restoration sites and adjacent flatwoods in the Experiment 2..................................69 3-9 Mean inoculation potential (+1 SE, n = 30) of the soils in the five treatments of the five-year complex...............................................................................................70 3-10 Mean soil total Kjeldahl nitrogen (+1 SE, n = 18) of the restoration and adjacent flatwoods and pasture soils in Experiment 2............................................................70 3-11 Mean soil gravimetric moisture content (+1 SE, n = 30) of the restoration and adjacent flatwoods and pasture soils in Experiment 2.............................................71 3-12 Mean soil extractable P content (+1 SE, n = 18) of the restoration and adjacent flatwoods and pasture soils in Experiment 2............................................................71 x

PAGE 11

Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science SEED BANKING AND VESICULAR-ARBUSCULAR MYCORRHIZAE IN PASTURE RESTORATION IN CENTRAL FLORIDA By Amy Miller Jenkins December 2003 Chair: Doria R. Gordon Major Department: Natural Resources and Environment Increasing efforts to restore longleaf pine systems that have been converted to non-native pasture grasses are underway in Florida. Native vegetation does not recover quickly in abandoned pastures unless restoration efforts are introduced. The goal of this study was to examine the contribution of relict seed banks and vesicular-arbuscular mycorrhizae (VAM) fungi in pasture restoration sites in central Florida. I compared the seed bank in a restoration site and an adjacent flatwoods and evaluated the mycorrhizal inoculation potential (MIP) of three restoration areas in different stages of progress. In Chapter 2, I investigated the species composition of the soil seed bank in a pasture restoration area and adjacent pine flatwoods. In this study, I examined the seed bank with germination and direct count assays and evaluated the aboveground vegetative species composition of the restoration area 6 months post-sowing. The restoration and flatwoods seed banks differed in species composition. The restoration soils had more non-native and weedy native species, while the flatwoods had more native species xi

PAGE 12

characteristic to pine flatwoods. The flatwoods soils had more unique species than the restoration soils. Although several perennial grasses were recorded, none of the perennial grasses most characteristic of flatwoods (e.g., Aristida stricta) were found in the seed bank with my methods. The restoration area seed bank is rich with native species. However, the seedbank will not regenerate a natural flatwoods community because most of the characteristic flatwoods species are absent. Seed of A. stricta and other perennial grasses must be reintroduced and weedy species controlled in restoration projects. In Chapter 3, I examined the relationship of VAM fungi to soil chemistry, restoration efforts, and temporal variation. Mycorrhizal inoculation potential was measured in pre-restoration, first-year restoration and five-year post restoration sites coupled with their adjacent pasture and natural flatwoods. Mycorrhizal inoculation potential of the restoration area before restoration seeding with a native seed mix was low (~3%) and similar to that found in pasture. There was a significant increase in MIP in the restoration area just 6 months post seeding not seen in the pasture or flatwoods. The temporal variation in MIP followed the phenology of the aboveground vegetation. The MIP in natural flatwoods was low in all tests (~2%). I also compared the VAM fungi colonization of A. stricta roots in 5-year post restoration sites with plants in adjacent flatwoods and found no significant difference. The MIP data coupled with the A. stricta results, suggest that inoculum is not lacking in restoration areas in central Florida; and that VAM fungi appear to infect native plants in restoration areas at a rate similar to that found in pine flatwoods. Overall, these results suggest that seed, but not mycorrhizae, will need to be augmented in pasture restoration efforts. xii

PAGE 13

CHAPTER 1 GENERAL INTRODUCTION AND SITE DESCRIPTION The natural environment of Florida has been dramatically changed through development pressures, habitat fragmentation, and agricultural practices. Pastures planted with Paspalum notatum Flgg (bahia grass), a non-native species, are widespread across Florida and the Southeast for cattle forage and sod production. Restoration of degraded upland landscapes such as abandoned pastures and agricultural fields may be an important mode for reintroducing many diminishing community types with increasing development pressures and habitat fragmentation in Florida. Native vegetation does not recover quickly in abandoned pastures dominated by P. notatum unless restoration efforts reintroduce native species (Violi 1999). Many land managers in Florida are restoring pastures to native upland habitat for various other reasons, including upland mitigation and phosphate-mine reclamation (The Nature Conservancy 2000). Additionally, with the continuing programs in the State of Florida to purchase conservation lands through bond issues (Florida Forever 2003), more former agricultural lands and semi-degraded properties that may have associated restorable pastures are being purchased. Many upland restoration projects are underway in Florida today. State agencies such as St. Johns River Water Management District are restoring old agricultural fields; phosphate companies such as CF Industries Inc. (Plant City, FL) and IMC Agrico Inc. (Fort Green, FL) are incorporating restoration of pine flatwoods into their mine-reclamation process; and The Nature Conservancy is restoring both sandhill in old windrowed pine plantations and pine flatwoods in pastures (The Nature Conservancy 2000). 1

PAGE 14

2 Are there certain ecological features of these pastures and restoration sites that, if better understood, would aid in restoration success? In this project, I investigated the contribution of seed banks and mycorrhizae to restoration of improved pastures at The Nature Conservancys Disney Wilderness Preserve (DWP) in Kissimmee, Florida. The Disney Wilderness Preserve is a 4,797-ha preserve owned and managed by The Nature Conservancy. The preserve is located just south of Orlando, Florida on the north side of Lake Hatchineha, at the base of the Reedy Creek/Lake Marion Creek watershed in Osceola and Polk Counties in central Florida (Figure 1-1). Many native terrestrial plant communities exist at the preserve. Mesic pine flatwoods is the most common among a mosaic of community types such as scrub, upland hardwood forest, dry prairie, wet prairie, baygall, cypress swamps, and depressional marsh, (FNAI and FDNR 1990). The average for the maximum daily temperature for this region is 31 C in summer and 18 C in winter. Freezing temperatures do occur but are infrequent and of short duration (Jordan 1984). This region of Florida has pronounced wet and dry seasons and long-term periods of flooding and drought with an average annual rainfall of 1270 mm (Jordan 1984). Soil types at our study site were Myakka and Smyrna series belonging to the order spodosols, which is common in pine flatwoods and characterized by a spodic horizon and sandy poorly drained conditions (Soil Conservation Service 1990, Brady and Weil 2002). These soils in Central Florida have low nutrients and organic matter and variable pH (Abrahamson and Hartnett 1990). The Disney Wilderness Preserve, previously a cattle ranch, has approximately 486 ha of improved and unimproved pastures scattered throughout the property in a matrix of natural flatwoods and wetlands. These pastures can alter ecosystem function by impeding

PAGE 15

3 movement of natural processes such as naturally occurring fire (The Nature Conservancy 1996). Additionally, the pastures are composed of non-native pasture grasses such as Paspalum notatum, Cynodon dactylon (Bermuda grass) Panicum repens (torpedo grass) and Digitaria pentzii (pangolagrass), and also non-native invasive species such as Solanum viarum (tropical soda apple) (Langeland and Burks 1998) and Imperata cylindrica (cogongrass). At DWP, old aerial photographs show that pine flatwoods occupied the sites before they were converted to pasture (The Nature Conservancy 1996). For these reasons, The Nature Conservancy sought to restore the pastures on DWP to a plant community assemblage that more closely resembled the pine flatwoods community. The goal of the restoration is to restore the structure and function of the ecosystem. The pine flatwoods community is the most widespread community type in Florida and historically extended over much of the southeastern coastal plain (Abrahamson and Hartnett 1990). Unfortunately pine flatwoods are one of the communities most impacted by human activities (Abrahamson and Hartnett 1990). Pine flatwoods are characterized by scattered longleaf pine (Pinus palustris) trees with a dense, species-rich ground layer of perennial herbaceous plants, which historically were maintained by naturally occurring high-frequency (5-10 years, Platt 1988), low-intensity forest fires (Abrahamson and Hartnett 1990, Pyne et al. 1996). Such fires were carried by the fuels of the herbaceous understory, especially Aristida stricta Trin. & Rupr. (wiregrass (Kesler et al. 2003)). Presently, many landowners and land managers manage the pinelands with prescribed fire to maintain community species composition and structure. Upland restoration in Florida, in the pine flatwoods and sandhill communities, is often modeled after restoration that has occurred in the prairie states for many years

PAGE 16

4 (Cottam 1987, Packard and Mutel 1997). Groundcover in the pine flatwoods strongly resembles the tallgrass prairie of the Midwest both in structure and function. This restoration has focused primarily on the native groundcover under the assumption that the groundcover (dominated by A. stricta in Florida) is what fuels the growing-season fires that maintain pine flatwoods and sandhills (Pyne et al. 1996). Restoring A. stricta cover is a primary goal of pasture restoration. Restorers also assume that longleaf pine trees can always be added after groundcover is established (N. Bissett, The Natives 2002, pers. comm.). Restoration begins with site preparation to remove the non-native cover and reduce the weedy seed bank. Many practitioners recommend that preparation occur over a period of about one year (N. Bissett, The Natives, Inc. 2002 pers. comm.; B. Wertschnig, CF Industries, Inc. 2000 pers. comm.). Removal of non-native pasture grasses is through herbicide application and soil disking (Bissett 1996). Soil disking is a type of plowing that shallowly turns the soil to about 15 cm depth. Disking breaks up dense root mats of pasture grasses and tends to scarify and raise seeds in the seed bank to the surface. Soil disking is done for various reasons and is usually repeated often throughout site preparation. Disking repeatedly surfaces new seeds so they can germinate and then turns the seedlings back under to kill them (The Nature Conservancy 1996). This technique effectively kills P. notatum and exhausts the seed banks weedy component that tends to dominate restoration efforts in pastures (N. Bissett, 2002, pers. comm.; The Nature Conservancy 1996). These sites are then sown with native understory/herbaceous seed collected from as close a source as possible.

PAGE 17

5 Seed-collection sites are burned in the growing season and the seed is sown in the winter (Bissett 1996). Seeds are collected in the fall in central Florida (using various methods, but most commonly with a silage cutter mounted on the front of a tractor). The tractor is driven through a flatwoods and it cuts and collects everything above 31-45 cm above the ground (Bissett 1996). Alternatively, a Woodward Flail-vac (used at DWP; Ag Renewal Inc., Manhattan, KS) is driven through the flatwoods and strips the seeds off the stalks of mostly grasses and forbs with a large rotating brush (approximately 45 cm above the ground). The seed then accumulates in a holding chamber. Once collected, the seeds are stored and sown onto the site in the winter. The seed is sown with a hay blower onto the prepared soil and then pressed into the soil with a roller. Rolling improves seed contact with soil and increases seed germination in A. stricta (Gordon et al. 2000). Invasive non-native species control is required for many years in these sites after seeding. Upland restoration at DWP began in 1995 with the creation of an experimental pilot study (The Nature Conservancy 1996). This study aimed to identify the best method of site preparation for restoration. Six randomly placed sites (one within each of six P. notatum pastures) were chosen; and 5 methods of P. notatum removal (single disk, single herbicide, disk and herbicide, multiple disk, and multiple herbicide) were administered to each site. Then the sites were sown with native flatwoods seed mix collected on-site (The Nature Conservancy 1996). Vegetation was monitored for 3 years after seeding to determine if significant differences existed among the site-preparation methods in the percent cover of native and non-native species. After 2 years of monitoring, the multiple herbicide method had a significantly higher percent cover of native species and significantly lower percent cover of P. notatum, while the other three methods were not

PAGE 18

6 significantly different from one another in percent cover of P. notatum (The Nature Conservancy 1999). There was tremendous native species richness that became established in the sites in all treatments (The Nature Conservancy 1999). Total species richness recorded was 219 of which 141 were native species characteristic to pine flatwoods, 38 were native weedy species, 37 were non-native weedy species, and 3 were non-native invasive species (The Nature Conservancy 1999). Species richness ranged from 34 to 47 species per 450 m2 in 1997, 35 to 61 per 450 m2 species in 1998 (Gordon et al. 2000), and 30 to 45 per 450 m2 in 1999 (Gordon et al. 2001). Average species richness per site was 37 species in 1997, 46 species in 1998, and 39 species in 1999 (Gordon et al. 2001). However, no further site management occurred and many of the characteristic native species were no longer present by 2001 (A. Jenkins 2001, pers. obs.). Management methods for maintaining native species in such restoration sites for more than a few years, therefore, requires further investigation. Results of this experiment led to the implementation of larger restoration units on DWP. A 10.7 ha restoration effort at DWP began with site preparation (disking and herbicide application) in 2001 and seeding occurred in early February 2002. Another adjacent unit of approximately 26.39 ha was sown in January 2003 while a 5.26 ha unit was site prepared during 2002-2003 and will be sown in 2004. The results of the pilot experiment and other restoration projects around the state raised questions that led to the work in this thesis. Those original experimental pilot sites and the new 10.7 ha and 5.26 ha restoration units at DWP provided suitable locations to investigate whether the tremendous species richness recorded in the DWP study could be attributed to a persistent seed bank of native flatwoods species, a seed bank that remained

PAGE 19

7 in the soil since pasture conversion. Gordon et al. (2000) concluded that existing seed banks or dispersal from adjacent sites clearly contributed significantly to the high species richness that resulted. They did not attribute the high species richness to the applied seed mix alone (Gordon et al. 2000). In Chapter 2, I compare the composition of the soil seed bank in the 18-ha pasture restoration unit and an adjacent flatwoods at DWP. The other puzzling result from the experimental pilot study was the lack of persistence by many native species that germinated. This result led me to question whether some soil factor was responsible. Factors such as the lack of beneficial mycorrhizal fungi may reduce the competitive advantage of the newly established native species. In Chapter 3, I examine the effects of restoration on soil mycorrhizal inoculum potential by comparing restoration soils with soils in flatwoods and P. notatum pastures. I also compared mycorrhizal field colonization of A. stricta plants growing in the previously mentioned experimental sites and adjacent plants in natural flatwoods. Addressing these questions will lead us to better understand the ecology of pasture restoration and to more successfully establish the native pine flatwoods community in restored pastures in Florida.

PAGE 20

8 Figure 1-1. Location of the The Nature Conservancys Disney Wilderness Preserve within the central Florida region.

PAGE 21

CHAPTER 2 SEED BANKING IN A PASTURE RESTORATION SITE AND MESIC PINE FLATWOODS IN CENTRAL FLORIDA Introduction Restoration efforts may be simplified when a seed bank of native species from the historical native community remains in the sites involved. A seed bank is a pool of seeds in the soil profile that can germinate at some time and replace adult plants of the aboveground floristic community (Baker 1989). Restoration of native communities by using a relict native seed bank would be ideal since collecting native seed is extremely time consuming and expensive, and seed is often unavailable. It is unknown, however, how long native seeds remain viable in the seed bank in soils that have been converted to agriculture (van der Valk and Pederson 1989). Seed banks have been successfully used in certain areas such as heathlands in Britain where native propagule banks (seeds and buds) were used as donors to reestablish heathlands in restoration sites (Putwain and Gillham 1990). However, persistent seed banks are not always present. For example, research in limestone prairie revealed that the dominant species were absent in the seed bank and needed to be actively reintroduced (Laughlin 2003). This research investigates the composition of the native pine flatwoods relict seed bank in the soils of pastures in central Florida in an applied context for future restoration. By comparing the seed bank composition of native pine flatwoods soils to pasture restoration areas, seeding needs of the site can be better evaluated. 9

PAGE 22

10 Pastures in Florida are usually vegetatively dominated by non-native grass species such as Paspalum notatum, Cynodon dactylon, Panicum repens, and Digitaria pentzii, and harbor other non-native invasive species as well as weedy and aggressive native species such as Eupatorium capillifolium (dogfennel). Efforts to restore such degraded landscapes to the natural assemblage of native pine flatwoods species is often hindered by competition from these non-native and weedy native species (Harper-Lore 1998, Brown and Bugg 2001) that are present in the seed bank (D'Antonio and Meyerson 2002). As a result, extensive site preparation is essential to restoration (Cottam 1987), especially in pastures because it removes competing non-native seeds from the seed bank through continued herbicide application and disking of the soil which promotes flushes of seedling growth before native seeds are introduced to the site (Harper-Lore 1998). The pine flatwoods in Florida are dominated by a species-rich herbaceous ground cover (Hardin and White 1989), which is the target suite of species for The Nature Conservancys pasture restoration efforts. Pine flatwoods can also have a non-continuous shrub layer consisting of Serenoa repens (saw palmetto), Ilex glabra (gallberry), various species of Ericaceous plants, and a pine canopy which is primarily Pinus palustris (longleaf pine) with occasional Pinus elliottii (slash pine) along wetland margins. Aristida stricta (wiregrass) (Kesler et al. 2003) is the dominant perennial grass species, a keystone in the pine flatwoods community because it produces fine fuel which carries low-intensity community-maintaining fires (Platt 1988, Clewell 1989, Noss 1989, Abrahamson and Hartnett 1990). Many pine flatwoods species rely on these low-intensity fires for their existence (Clewell 1989, Noss 1989). For example, Pinus palustris is a poor competitor and needs the low-intensity fires to control hardwoods while many

PAGE 23

11 herbaceous species reproduce only after such low-intensity fires (Clewell 1989). As a result, A. stricta is primarily targeted as a target species to reestablish in upland restoration of both pine flatwoods and in the similar community type, sandhill, because it aids in reestablishing community structure and function. Extensive research has been dedicated to seed germination and establishment of A. stricta (Seamon and Myers 1992, Mulligan et al. 2002). Sandhill restoration studies found that A. stricta only becomes established when directly sown in revegetation efforts (Hattenbach et al. 1998, Cox et al. 2003). For this reason, A. stricta seed is collected and directly sown into restoration sites across the state. Other species are sometimes hand-collected and added to the seed mix but the quantities are far lower than that for A. stricta. Earlier work has demonstrated that A. stricta is often the only species introduced from the seed mix restoration efforts (Gordon et al. 2000). Seed bank composition can influence restoration approach and objectives (van der Valk and Pederson 1989). The composition of the relict native seed bank in pastures that were formerly flatwoods has been little studied (Violi 1999), particularly comparing the composition of the pasture seed bank to that of intact flatwoods soils. Knowing the differences and similarities in the seed bank between pastures and flatwoods will be useful to upland restoration efforts in pastures because practitioners will better be able to evaluate seeding and future non-native species management requirements. Differences among agricultural and disturbed land uses could be an important influence on the relict seed bank composition for example, abandoned pastures in Lake Placid, Florida, were found to have a species rich native seed bank (Violi 1999), while very few obligate sandhill species were found in the seed bank of abandoned citrus groves in central Florida

PAGE 24

12 (Buchanan 1999). Hattenbach et al. (1998) and Cox et al. (2003) both found species rich native seed banks in disturbed sandhill sites in north Florida. Many different methods for quantifying the presence and abundance of seeds have been used in studies of soil seed banks. Most studies have quantified the seed bank by spreading soil samples into germination flats, placing them in a greenhouse or shadehouse, and identifying and counting any germinated seedlings (Rabinowitz 1981, Kitajima and Tilman 1996, Carrington 1997, Schott and Hamburg 1997, Butler and Chazdon 1998, Leckie et al. 2000). However, only viable seeds whose specific germination cues are met in the environmental conditions of the test are identified by this method. As a result, some seeds and species are likely underestimated or missed. Other studies have used direct separation of seeds from soil with a salt solution method (Malone 1967, Brown 1992, Buhler and Maxwell 1993), sieving (Carrington 1997), or elutriation (Gross 1989). These direct separation methods may overestimate the viable portion of the seed bank unless seed viability is also measured. At least two studies have compared more than one method of estimating the soil seed bank (Gross 1990, Brown 1992), with inconsistent results. Germination trials with cold-stratified soil resulted in more species than direct germination or elutriation in a ploughed field in Michigan (Gross 1990). In a woodlot in Ontario, 102 species were detected by elutriation and only 60 by the germination technique (Brown 1992). A combination of methods might be most suitable to accurately quantify the composition of a soil seed bank. One objective of this study was to investigate whether there were significant differences in species composition of the soil seed bank between a pasture restoration area (prior to sowing) and the adjacent, relatively undisturbed flatwoods. Another

PAGE 25

13 objective was to compare the seed bank composition in the restoration soils with that of the aboveground vegetation in the restoration area 6 months post restoration seeding to determine similarity. I also wanted to determine which species germinated in the restoration field but were not in the seed bank assays to evaluate which species could be attributed to the seed mix. In this study, I examined the composition of the seed bank with two methods. The seed bank species composition was evaluated through the germination method and a direct counting method (sieving) and frequency of plant species in the restoration area was examined. I hypothesized that the seed bank in the restoration and flatwoods soils would be similar in species composition, that both would be dominated by weedy native species, and that the restoration soils would have more non-native species than the flatwoods soils. I also hypothesized that the vegetation emerging in the restoration area would be composed of similar species as in the two seed bank assays, with the addition of the sown A. stricta. I predicted that the two methods of evaluating seed bank species compositions would yield differing results based on the findings of Gross (1990) and Brown (1992). Methods The Nature Conservancy initiated a 10.7 ha upland restoration project at the Disney Wilderness Preserve (DWP; see site description in Chapter 1) in June 2001. The site was treated with herbicide and disked six times over the course of the year to deplete the weedy seed bank and control non-native species. Seed collection of A. stricta and other native herbaceous species (as described in Chapter 1) occurred in November 2001 on-site in pine flatwoods stands (121.5 ha) with dense A. stricta plants that were burned in March and July 2001. Several species were also hand-collected and added to the seed mix. These species include: Buchnera americana, Panicum rigidulum, Rhexia nashii,

PAGE 26

14 Sorghastrum secundum, Pterocaulon virgatum, Dichanthelium spp., Pityopsis graminifolia, Panicum anceps, Solidago fistulosa, Axonopus spp., Lacnanthes caroliniana, Dichanthelium portoricense, Saccharum giganteum, and Eragrostis spp. Total seed mix collected was approximately 4.536 m3 (40, 30 gallon bags), which was stored in a cool, dry place until seeding occurred (K. Kosel 2001, pers. comm.). The seed was sown with a modified hay blower on 15 February 2002 onto the prepared 10.7 ha restoration site. Just before and after seeding, the soil was compacted with a cultipacker. Nomenclature throughout follows Wunderlin (Wunderlin 1998) except A. stricta which follows Kesler et al. (2003). Species were also separated into categories to better distinguish those that are characteristic of native pine flatwoods and represent the long-term restoration goal (hereafter native characteristic) from more weedy or early successional species (hereafter native weedy) (The Nature Conservancy, unpublished data). We also separated non-native weedy species and non-native invasive species, the latter identified through a listing of species invasive in natural areas in Florida (Florida Exotic Pest Plant Council 2003). Germination Assay To compare the seed banks in the restoration area and the adjacent pine flatwoods, a seed bank germination experiment was initiated. The restoration area was classified into five elevation strata to capture variability across the site. Using a topographical map and aerial photographs; one 50 m X 50 m quadrat was permanently marked within each stratum. Five strata at the same elevations were established in the adjacent native flatwoods. These flatwoods were chosen based on several criteria: proximity, species composition, and the same soil type and elevation range as those in the restoration area.

PAGE 27

15 On 9 February 2002, in each of the 5 strata in each site (restoration and flatwoods), 10 randomly placed soil samples were collected. Each sample was a composite of 2 soil cores taken 0.5 m apart and well mixed. Each soil core was 12 cm deep x 8 cm diameter (surface 0.0050 m2; 602 cm3). Composited samples were therefore 0.01 m2 (1204 cm3). We chose to sample at this depth because the soil in the restoration area had been repeatedly disked to roughly 12 cm (K. Kosel 2001, pers. comm.). Similar depths have been used in other seed bank studies (Rabinowitz 1981, Butler and Chazdon 1998). All samples were immediately placed in a cold room (10 C) and stored there until the shadehouse experiment was established on 20 April 2002. Soil was collected immediately prior to restoration seed sowing in the field (15 February 2002). The experiment was established in an open air shadehouse on the University of Florida campus in Gainesville (Alachua Co., FL). The shadehouse has benches on which the flats for the experiment were placed and a wooden frame, which supported 50% shade cloth. Plastic sheeting (4 mil polyethelene film) was also installed just under the shade cloth to prevent direct hard rain contact with the soil in the flats. This was done to avoid unintentional mixing of soil and seeds from flat to flat in heavy rain storms. A misting-style sprinkler was installed and the flats were irrigated daily for one hour. Flats were 25 cm X 50 cm in dimension and divided by a barrier into two 25 cm x 25 cm sections. Newspaper was placed in the bottom of each flat section to prevent soil loss and to aid in moisture retention. The soil used for this experiment was 50% sand and 50% vermiculite. The sand was autoclaved for 45 minutes at 121 C to kill any contaminating seeds. Each flat section was filled with 3 cm of the soil mix and watered heavily. Field soils were

PAGE 28

16 randomly assigned to the flat sections (50 flatwoods, 50 restoration area, and 20 control). The control flat sections were included to account for germinated seeds that were dispersed from the local area instead of the field site. The control soil was 50% store bought potting soil and 50% sand (both autoclaved for 45 minutes at 121 C to kill any contaminating seeds). In each flat section, 236.5 cm3 of the field soil (restoration or flatwoods) or control soil was spread to a depth of 0.38 cm over the moistened sand/vermiculite mixture. This amount equates to 39.4 % of each composited soil sample and therefore represents 0.002 m2 of soil at 0-12 cm depth. After the experiment was established, the unused field soils were air-dried and stored for use in the direct seed count assay (below). Germination was checked bi-weekly for the first 10 weeks and then weekly until germination ceased, around 15 August 2002. The arrangement of the flats on the benches was randomly repositioned weekly to minimize any irrigation or sunlight variability caused by bench position. Data on germinated seedlings were recorded each monitoring day for date of emergence and morphological characteristics. Most seedlings were also marked with color coded toothpicks for easier relocation. Most species of grasses and sedges were not identified until they produced reproductive structures which did not occur until approximately August/September 2002 or later. In many flat sections (primarily restoration soils) numerous Cyperus spp. seedlings emerged simultaneously. Because tracking that many seedlings was too difficult and competition among seedlings would reduce survival, I randomly removed a known number of the Cyperus spp. seedlings. As a result, all species in the genus Cyperus are listed throughout as Cyperus spp. However, at least 5 species were represented in each of the restoration and flatwoods

PAGE 29

17 soils in the germination assay: C. compressus, C. globulosus, C. polystachyos, C. retrorsus, and C. surinamensis were identified among the remaining Cyperus spp. seedlings left in the flats. On 15 September 2002, a heavy rain severely damaged the plastic sheeting, allowing all of the flats to be subjected to a heavy rain event. Several of the flats (8 flatwoods and 3 restoration) were completely destroyed by the water. All data from these flats were eliminated from the experimental analysis because many seedlings had not yet been identified. The plastic sheeting was replaced. No germination was observed after this date. We identified most of the species that germinated in the germination assay with the help of field keys (Wunderlin 1998) (Godfrey and Wooten 1979, Godfrey and Wooten 1981), field comparison, and experts (N. Bissett, The Natives Inc. pers comm.2002). However, individuals that produced no reproductive structures by the end of the experiment were not identified beyond monocots and dicots. They were included in estimates of seed density. Direct Count Assay I initiated a direct count seed bank investigation on 28 January 2003. The same soil samples that were used for the shadehouse experiment had been air-dried and stored in plastic bags for use in this investigation. Four samples from each elevation strata were randomly chosen and three 20 cm3 subsamples were taken from each. The three subsamples together represent 0.0005 m2 (60 cm3) of soil (at 12 cm depth). Each sub-sample was poured through a series of three sieves. The sieves were stacked together in size order with the largest on top (3.5 mm, 0.9 mm, 0.5 mm). The soil caught by each sieve was transferred to a single layer in a petri dish and examined under a dissecting

PAGE 30

18 microscope. The soil that fell through the 0.5 mm sieve was discarded following examination of 10 subsamples that yielded no seeds. Seeds were extracted from the soil under microscope examination, labeled, and set aside for identification. The seeds were identified when possible by comparison to herbarium specimens at the University of Florida, Florida Museum of Natural History herbarium, published line drawings (Godfrey and Wooten 1979, Godfrey and Wooten 1981), photographs (Martin and Barkley 1961, Landers and Johnson 1976), and expert consultation (R. Abbott 2003, pers. comm.). Identification was to species where possible and to morphospecies where not possible. Vegetation Survey To accompany the seed bank data, I also studied species assemblages in the restoration site which incorporated any plants that emerged in the restoration area from the seed bank and sown seed mix. I monitored the vegetation in the one-year restoration field in August 2002 (6 months post restoration seeding) in the same 5 strata that were used for the seed bank sampling. Each of the 50 X 50 m quadrats in the restoration area was divided into 100, 5 X 5 m subquadrats. Ten of these squares were randomly selected in each quadrat for vegetation sampling. In the northeast corner of each randomly-chosen subquadrat, I placed a nested 20 cm X 20 cm and 1 X 1 m frame (nested frequency sampling). Presence of any species within each 20 x 20 cm, 1 x 1 m, and 5 x 5 m quadrat was recorded. Additionally, density of A. stricta plants was recorded in the 1 x 1 m frame to examine A. stricta recruitment. All plants were keyed to species in the field when possible and unknown plants were collected and subsequently identified. Data Analysis Species accumulation curves were generated to compare the species richness between treatments (restoration and flatwoods) in the germination and direct count

PAGE 31

19 assays, and among the three types of data collected in the restoration soils (germination, count, and field emergence) (Colwell 1997, Chazdon et al. 1998). Species accumulation curves were generated with the computer software, EstimateS (Colwell 1997), from 50 randomizations of the samples to compare observed species richness among treatments. A non-parametric estimator of species richness was used to estimate true species richness based on the observed species (Colwell 1997, Chazdon et al. 1998). The Incidence-based Coverage Estimator (ICE) was calculated with EstimateS (Colwell 1997), and chosen based on the results of Chazdon et al. (1998) that compared multiple non-parametric estimators. Coleman curves were also generated with EstimateS (Colwell 1997), which evaluate patchiness in the distribution of species in the dataset by sub-sampling from the 50 randomizations. If the Coleman Curve overlaps the species accumulation curve, the distribution of the species is not considered patchy. Unique species are defined as species that only occur in one sample. The Jaccard (CJ) incidence-based similarity index (Magurran 1988) was calculated to compare seed bank species composition between the flatwoods soils and restoration soils in the germination and direct count assay. The Bray-Curtis quantitative dissimilarity index (Underwood and Chapman 1998), that accounts for abundance, was calculated to determine dissimilarity of seed bank species composition and abundance between the flatwoods and restoration soils in both laboratory assays. Differences in species richness among the five elevation strata in the vegetation survey for each of seven vegetation categories (native characteristic graminoid, native weedy graminoid, non-native invasive graminoid, non-native weedy graminoid, native characteristic forb, native weedy forb, and non-native weedy forb) were tested using

PAGE 32

20 ANOVA (SAS 1999). ANOVA was also used to determine if the density of A. stricta plants in the restoration field was dependent on the elevation strata. Tukey tests (p < 0.05) were used to determine post-hoc differences among means. Results Germination Assay Most seedlings germinated within 70 days after sowing with a total of 40 species recorded in the shadehouse experiment (see Table A-1). The flatwoods and restoration soils each had 31 species, 20 of them shared among the two soil types. Of the 40 species, 21 were native species characteristic to flatwoods, 10 were native weedy species, and 5 were non-native weedy species (no category determined for four unknowns). Several individuals were only determined to genus, one species was only determined to family (Asteraceae), and unknown dicot and sedge groups were used. Two of the flatwoods samples showed no germination. Species composition was 35% similar (CJ = 0.359) between the flatwoods and restoration soils. Of the 40 total species found, 55% were forb and 45% were graminoid species (Table A-1). Nine species were unique to each treatment and three of the species unique to the restoration soils were non-native species (Table 2-1). Several perennial grasses were recorded in the restoration soils (i.e., Andropogon sp. and Dichanthelium portoricense), however none of the perennial bunchgrasses most characteristic of mature flatwoods such as A. stricta, Sorghastrum secundum, and Schizacharium scoparium were found (Table A-4). The relative abundances of species in the two treatments were significantly different. The Bray-Curtis quantitative dissimilarity index revealed that the two treatments were 92.3% dissimilar in species composition when abundance was accounted

PAGE 33

21 for. Only two species, Cyperus spp. and Hedyotis uniflora, were among the top most abundant in both soil types (Figure 2-1). The five most common species in the restoration soils were Cyperus sp., Hedyotis uniflora, Scoparia dulcis, Ludwigia octavalvis, and Kyllinga brevifolia (Figure 2-1a). Cyperus sp. dominated all other species in these soils and consisted of at least 5 species (C. compressus, C. globulosus, C. polystachyos, C. retrorsus, C. surinamensis). These species together averaged 71.5 seedlings per sample and accounted for 83% of all seedlings in the restoration soils (Figure 2-1a). Cyperus spp. averaged only 2.05 seedlings per sample in the flatwoods soils (Figure 2-1b) and only accounted for 6% of the total seedlings germinated. The five most abundant species in the flatwoods soils were Eleocharis sp. (filiform type), Hedyotis uniflora, Dichanthelium sp., Cyperus spp., and Polypremum procumbens (Figure 2-1b). The most common species in the flatwoods soils was Eleocharis sp., which averaged 16.9 seedlings per flat section and accounted for 51% of all seedlings in the flatwoods soils, followed by Hedyotis uniflora which averaged 6.52 seedlings per plot (20% of all flatwoods seedlings) (Figure 2-1b). The mean species accumulation curve approached its asymptote more rapidly for the restoration soils than the flatwoods soils (Figure 2-2a). The ICE estimator for incidence-based species richness predicted higher species richness in the flatwoods (40.7) than in the restoration soils (36.4). The curves indicated the sufficiency of the sampling, since the observed species richness is close to reaching an asymptote (particularly for the restoration soils) and the ICE predicted values. The Coleman Curves (not shown here) for each soil type overlapped the species accumulation curves and therefore did not predict a patchy spatial distribution for either soil type. In flatwoods samples, 9 of the 31 species occurred only once, whereas in the restoration site samples 7 of the 31 species only

PAGE 34

22 occurred once. The seed bank in the restoration soils seems to be more homogeneous because of the steeper slope on the species accumulation curve (Figure 2-2) and fewer unique species. This pattern likely resulted from the soil mixing that occurred during pasture conversion and restoration site preparation. Greater patchiness would be expected in the undisturbed flatwoods soils. Estimated seed density in the restoration and flatwoods soils differed by almost 40% when numbers are extrapolated to seedlings per m2. Three times more seedlings emerged from the restoration soils (4045 seedlings per 0.094 m2 of soil = 43,032 seeds m-2) than the flatwoods soils (1402 seedlings per 0.084 m2 of soil = 16,690 seeds m-2). Direct Count Assay Direct seed count resulted in a total of 22 species. Thirteen species were found in the flatwoods soils and 15 in restoration soils with seven overlapping species (see Table A-2). Species were labeled as unknown when genus and family could not be determined. Many more seeds could not be identified in this experiment compared to the germination assay. In addition, the condition of many soil-extracted seeds was poor (due to herbivory, soil scarification, broken parts, etc.) and vital morphological features were often missing. Both the flatwoods and restoration soils were dominated by native graminoid seeds. However, the flatwoods graminoid species were native and characteristic of pine flatwoods whereas those in the restoration soils were dominated by native weedy species. Non-native species were only found in the restoration soils: Desmodium triflorum, Fimbristylis schoenoides, and Kummerowia striata. The two soil types were 29% similar in species composition according to the Jaccard index (CJ = 0.29). Again, as in the germination assay, none of the perennial bunchgrasses most characteristic of native flatwoods were found in either soil type.

PAGE 35

23 Although observed species number was similar in both the flatwoods and restoration soils, ICE estimated species richness was much higher for the flatwoods soils (45 species) than for the restoration soils (16 species). This large difference in species richness reflected their difference in accumulation curves (Figure 2-2b). The cumulative number of species in the restoration soils almost reached an asymptote within the 20 samples while those in the flatwoods did not (Figure 2-2b). The large difference in the ICE estimated species richness for the restoration and flatwoods for this direct seed counting reflects differences between the soil types in incidence of rare and common species. For the total area of 0.01 m2, 15 species were found in the restoration soils, two of which occurred only once, whereas 13 species were found in the flatwoods soils, and nine of them occurred only once. Although the Coleman Curves generated (not shown because they overlapped the species accumulation curve) did not predict patchy spatial distribution, the species accumulation curves and the number of unique species found demonstrate greater homogeneity in restoration site than flatwoods seed banks. Again, this difference is likely due to loss of species and the soil mixing that occurred in the restoration soils during pasture management and site preparation. Consistent with the germination study, only a few species accounted for the majority of the seeds extracted from the soil, and these dominant species were different for the two soil types. The Bray-Curtis dissimilarity index revealed the two treatments to be 91.3% dissimilar. The dominant species in the flatwoods soils were Fimbristylis autumnalis (present in the germination assay but not among the top five species) and Dichanthelium portoricense. Those two species accounted for 87% of all the seeds extracted from the flatwoods soil but only 4% of the seeds extracted from the restoration

PAGE 36

24 soils. Cyperus compressus, Cyperus globulosus, Cyperus polystachyos, Kyllinga brevifolia, and Fimbristylis dichotoma (all Cyperaceae) dominated the restoration soils and accounted for 91% (15%, 16%, 17%, 25%, and 18% respectively) of the total seeds extracted. None of these five species were found in the flatwoods soils. Again, many more seeds were extracted from the restoration (1258 in 0.01 m2 = 2,516,000 seeds m-2), than from the flatwoods soils (202 in 0.01 m2 of soil = 404,000 seeds m-2). There were four species found in the direct count that did not germinate in the germination assay but none of them were dominant components of the directly counted seeds and none were identified to species. Vegetation Survey The restoration area at the time of the vegetation survey was extremely dense with vegetation approximately 60 cm tall with scattered taller plants such as Phytolaca Americana (pokeweed) and Eupatorium capillifolium (dogfennel). Most species were annuals and producing reproductive structures at the time of sampling with the exception of several perennial species such as A. stricta plants which generally had 1-2 leaves about 15 cm tall. I found 84 species in the vegetation sampling (Table A-3), 31 of which were also found in the two seed bank assays. Of those 84 species, 52 were native species characteristic to flatwoods, 16 were native weedy species, 13 were non-native weedy species, and 3 were non-native invasive species (Table A-3). While the species varied across the elevation strata, the species richness observed in the vegetation survey were consistently higher than in either of the seed bank assays (Figure 2-3a). Of the 49 species that were observed in the vegetation survey but never in either seed bank assay, 67% were native characteristic species, 10% were native weedy species, 14% were non-native

PAGE 37

25 weedy species, and 6% were non-native invasive species (which were absent from all seed bank assays). The elevation strata differed in species richness (Figure 2-3b). Stratum 1 was at the lowest elevation, close to a cypress dome, and had the highest species richness. This richness was probably due to the additional wetland species that were specific to that stratum (i.e., Amphicarpum muhlenbergianum, Rhynchospora microcephala, Sabatia grandiflora, Rhexia nashii). The remaining strata had variable levels of species only occurring in one stratum (Figure 2-3b). All strata shared 21 species in and many species were sporadic (shared by at least two strata) (Figure 2-3b, see Table A-3). The species richness varied among the categories (F = 73.87, p< 0.0001) and elevation strata (F = 7.04, p < 0.0001) (Table A-3). While not consistent along the elevation gradient, I found significantly more native characteristic graminoids in Strata 1 and 4 than in Strata 5 and 3 (F = 7.18, p < 0.0009), non-native weedy graminoids in Stratum 1 than in the other strata (F = 25.58, p < 0.0001), and native characteristic forbs in Strata 1 and 2 higher than in Stratum 3 (F = 5397, p < 0.0025). Aristida stricta Recruitment Aristida stricta was found in all but four of the 25, 1 X 1 m plots monitored. The average density was 2.56 m2 per plot and ranged from 0 to 15 m2 per plot with no significant differences among the elevation strata (Figure 2-4). Differences may be attributable to uneven seed density. Methods Comparison Species richness per area of soil sampled for the restoration soils could be compared through the species accumulation curves for all trials conducted, scaled for the area sampled (Figure 2-5). As discussed earlier, the germination assay yielded many

PAGE 38

26 more species than did the direct count assay for the restoration soils (Figure 2-5a). For the restoration soils, the species richness was greatest in the vegetation survey which sampled far more soil area (625 m2 total) and incorporated both seed bank and sown species (Fig. 2-5b). Another comparison of the different methods for seed bank quantification is to examine the number of seeds/seedlings per m2. The species richness per sample in the restoration and flatwoods soils is fairly consistent in the germination assays when calculated for equivalent area (m2), but differs in the direct counting assay (Figure 2-6). Fewer seeds were identified in the direct count than in the germination assay from the flatwoods soils (Figure 2-6). This discrepancy is because Eleocharis sp. and Hedyotis uniflora, which were the most abundant species in the germination assay in the flatwoods soils, were never extracted from the soils in the direct count assay. Discussion I found clear differences in the seed bank composition of the flatwoods and restoration sites in both germination and direct counting assays. The two soil types were only 35% and 29% similar in species composition in the germination and direct count assays, respectively. The higher abundance of seeds in the restoration soils was also a noticeable difference (92.3 % and 91.3% dissimilar to flatwoods soils when the abundance of seeds are accounted for in the germination and direct count assays respectively). The flatwoods soils had higher incidence of unique species than the restoration soils. Additionally, the species that dominated the seed bank differed by soil type. The restoration soils had more seeds of non-native and weedy native species, while the flatwoods had more native characteristic flatwoods species (Tables 1, A-1, A-2). It is probable that disking during site preparation, which is meant to eliminate the weedy seed

PAGE 39

27 bank in the former pastures, also eliminated relict native characteristic species from the seed bank. Another distinction that can be made between the seed bank composition of the flatwoods and restoration soils is that of permanent and transient seed bank. Transient seed banks exist when certain seeds bank in the soil for a short period of time such as just until the next favorable season (Thompson 2000). Permanent seed banks exist in the soils for longer periods of time and have many reasons and mechanisms for their persistence (Thompson 2000). The seed bank composition in the restoration soils is primarily permanent seed bank since there were no plants reproducing in the field for about one year before the samples were taken. However, the flatwoods soils likely contained both seed bank types. Germination assays with soil taken at different times of year in the flatwoods could reveal which species might be only transiently banking in the soil. Elimination of any transient species from the flatwoods seed bank species composition would yield a more valuable comparison between the two soil types. The seed bank species composition results confirm what many restoration practitioners already suspected: native sedges, mostly in the genera Cyperus and Fimbristylis, and annual dicots, such as Hedyotis uniflora, Polypremum procumbens and Ludwigia spp., dominate the seed bank in restoration areas in central Florida. These results are similar to seed bank species composition found in pastures in Lake Placid, FL (Violi 1999), despite the greater estimated seed density in my study. The Cyperaceae family is frequently dominant in the seed banks of grasslands (Roberts 1981). The restoration area seed bank at DWP is rich and diverse with native species. Some are early successional weedy and some are characteristic of healthy, mature pine flatwoods

PAGE 40

28 systems. Restoration areas that were formerly sandhill in north Florida have also been found to have rich, diverse native seed banks (Cox et al. 2003). Many dominant perennial bunchgrasses were absent from all seed bank assays, most importantly the keystone species, A. stricta. Similar findings were reported from three sandhill studies in North Florida (Hattenbach et al. 1998, Cox et al. 2003), (G. Parks, unpublished data), a sandhill study in central Florida (Buchanan 1999), and a flatwoods study in south central Florida (Maliakal et al. 2000). Aristida stricta has only been found in sandhill restoration plots when directly sown (Cox et al. 2003). Apparently A.stricta must be sown in order to become established in restoration sites in central Florida. Because of the importance of this species for system function (Platt 1988, Clewell 1989, Noss 1989, Abrahamson and Hartnett 1990), where lost from the system through habitat degradation or destruction, A. stricta must be actively reintroduced. Others have observed that supplementation of the seed bank is necessary in restoration efforts when the dominant plant species is unlikely to be found in the seed bank (Laughlin 2003). Many guidelines have been presented for the best methods of reestablishing A. stricta both in pine flatwoods (Bissett 1996, Conservancy 2000) and in sandhill (Seamon and Myers 1992, Hattenbach et al. 1998, Cox et al. 2003). Species composition of the two seed bank assays and the restoration vegetation survey were different. (Figures 2-3a and 2-5). The vegetation survey contained many more native species characteristic to flatwoods and the only account of non-native invasive species in the study. Therefore, the majority of the species that the vegetation survey and the seed bank assays had in common were native weedy species. This difference likely results from several factors. Firstly, when the vegetation field site was

PAGE 41

29 sown, other species were also added through the seed mix (e.g., A. stricta, Andropogon virginicus var. glaucus, and Pityopsis graminifolia). Secondly, the presence of non-native species also contributes to this difference. Species such as Panicum repens, Cynodon dactylon, and Solanum viarum likely were not sown in with the seed mix and came from the seed bank or vegetative propagules in the restoration area. Non-native and weedy species were probably introduced during pasture creation and cattle or agricultural management, as the adjacent non-converted flatwoods did not contain these species. Similar findings have been reported from an abandoned agricultural area on former sandhills in central Florida in which a weedy seed bank had no native sandhill species (Buchanan 1999). Thus, historical land use significantly influences seed bank composition, with important implications for restoration (Roberts 1981, Bekker et al. 1997). While some species found in the vegetation survey plots may differ from species in soil collected prior to sowing because they were introduced in the seed mix, this explanation is not likely the primary source of species difference. Many of the additional species found in the field probably originated from the seed bank itself or as vegetative propagules (i.e., Paspalum notatum) in soil because of their absence from healthy flatwoods stands, their reproductive phenology (unlikely to have been in seed when seed was collected), and/or height (i.e., seed stalk is lower than the seed harvester could access). Dominant species in the vegetation survey such as Eupatorium capillifolium, were absent in the two seed bank assays. Post-seeding dispersal could be playing a large role in this difference as well.

PAGE 42

30 Finally, there were large differences in total soil area sampled among the three sampling methods: vegetation sampling plots were 625 m2; the direct count assay, 0.01 m2; and the germination assay, 0.094 m2 (Figure 2-5). Species area relationships (Gotelli 2001) suggest that area is likely also responsible for the difference in the species richness observed in the three methods. For all the methods, a larger sample area would have yielded a greater sampled species richness. Differences in richness and composition across the elevation strata found in the vegetation survey suggest that species distribution is patchy and likely to vary within and among restoration sites. However, aside from higher richness in the wettest stratum, no clear patterns with elevation were observed. While I am unable to extrapolate to other flatwoods restoration areas, these results suggest that prediction of species composition of a restored site, even among relatively homogeneous areas, may not be possible (Kltzli and Grootjans 2001). Methods Comparison I found detectable differences in the seed bank composition between the germination and direct count methods. Brown (1992) found similar differences in the estimates of the soil seed bank from the solution separation method and the seedling germination method. One of the most common species in Browns germination trial, Verbascum thapsus L., only represented 1 % of the seeds directly extracted from the soil (Brown 1992). Contrasting the two seed bank assays in the flatwoods soils of my study, the most common species found in the germination assay, Eleocharis sp., was not found in the direct count assay (Figure 2-6). Clearly this is due to an extraction error since the soils used for both assays originated from the same soil samples. Perhaps the 0.5 2.5

PAGE 43

31 mm (Godfrey and Wooten 1979) Eleocharis seeds were too small to directly extract from the soil. I found far more seeds per m2 through the direct count method (404,000 2,516,000 m-2) than the germination method (16,690 43,032 m-2) but fewer species, 22 and 41 respectively. Here, the seedling germination assay underestimated the seed bank density, whereas the direct count method may have overestimated it. Seeds extracted from the soils in the direct count assay may not all be viable. Thus, the extrapolated numbers likely overestimate the real seed bank, particularly as the estimated seed density is considerably higher than other studies in Florida pasture (Violi 1999) and sand pine scrub (20 seeds m-2) (Carrington 1997), old-field pine in North Carolina (1470 seeds m-2) (Oosting and Humphreys 1940), tall-grass prairie (6368 seeds m-2) (Rabinowitz 1981), pastures in Great Britain (400 70,000 seeds m-2) (Roberts 1981) and an annual grassland in California (5156 54,687 seeds m-2) (Heady 1958) all of which used the germination method. Results from this and other studies suggest that the use of only one method inadequately estimates species composition and seed abundance present in the soil seed bank. However, the time investment for the direct counting assay was great and I only found a few species in the direct count that were not represented in the germination assay. Therefore, based on my results for species richness, direct count assays may not be worthwhile. Composition of the aboveground vegetation in the flatwoods was not examined in this study. However, results from my two seed bank assays may be qualitatively compared with the aboveground species composition measured by The Nature Conservancy (The Nature Conservancy 1997) in 15 pine flatwoods in 1997. Most of the

PAGE 44

32 species from my seed bank assays are also in the species list observed by The Nature Conservancy in the flatwoods vegetation (Table A-4). However, the species that characterize the flatwoods community are not represented in the seed bank of either the flatwoods or restoration soils. These include: A. stricta, Serenoa repens, and Ilex glabra, more grass species than forbs (Abrahamson and Hartnett 1990), and many species in the Ericaceae (FNAI and FDNR 1990) (Tables A-1, A-2). Most of the perennial bunchgrasses, absent in my seed bank samples, were well represented in flatwoods vegetation. Maliakal et al. (2000) found that 54% of the species in A. stricta flatwoods in south-central Florida rely only on vegetative regeneration rather than on seed following fire. Thus, the species composition found in the two seed bank assays are not representative of the aboveground vegetation in a mature pine flatwoods. Similar findings were reported in sand pine scrub (Carrington 1997), flatwoods in south-central Florida (Maliakal et al. 2000), and limestone prairies of Pennsylvania (Laughlin 2003). Management Implications Understanding the composition of the seed bank in restoration sites can help restoration practitioners better gauge the direct seeding needs of their site. Most obvious is the finding that the pine flatwoods dominant perennial bunchgrasses are absent from the seed bank both in the flatwoods and restoration soils; thus direct seeding of these species into restoration projects is necessary, especially if there are no adjacent seed sources from which they could naturally disperse. These results also suggest that recruitment of these species may depend on specific factors influencing seed production or subsequent seedling establishment. Another important finding based on the vegetation survey was that many non-native species emerged from the restoration soils and likely came from the seed bank or

PAGE 45

33 vegetative propagules. These species were not thoroughly eliminated by the site preparation techniques (herbicide and disking) and will continue to require active management. This research is vital for the restoration of pine flatwoods communities because conversion of this land to pasture eliminates native bunchgrasses from the vegetation. While restoration of a diverse groundcover may not require direct seeding, many of the species present are early successional or weedy natives that persist in the seed bank under abandoned pastures. Conversely, seed of the dominant pine flatwoods species must be reintroduced into restoration projects as they will not naturally regenerate through relict seed banks.

PAGE 46

34 Table 2-1. Species in common between the restoration and flatwoods soils and unique species in the flatwoods and restoration soils in the germination assay. Non-native species are indicated with an asterisk. Common Species Flatwoods only Restoration only Andropogon sp. Asterceae basal rosette Chamaecrista nictitans Cyperus surinamensis Crotolaria rotundifolia Cyperus globulosus Cyperus compressus Drosera brevifolia Erechtites hieracifolia Cyperus polystachyos Eleocharis viviparous Fimbristylis schoenoides* Cyperus retrorsus Ludwigia maritima Hedyotis corymbosa* Cyperus sp. Rhynchospora fasicularis Kummerowia striata* Dichanthelium sp. Scleria reticularis Lindernia sp. Dichanthelium portericense Scleria sp. Ludwigia repens Desmodium triflorum Toxicodendron radicans Ludwigia sp. Eleocharis filiform type Fimbristylis autumnalis Fimbristylis dichotoma Gnaphalium sp. Hedyotis uniflora Kyllinga brevifolia Ludwigia round leaf Murdannia nudiflora Oxalis corniculata Ludwigia octavalvis Polypremum procumbens Scoparia dulcis Xyris-like

PAGE 47

35 Five most common species in the restoration soils CYSPHEUNSCDULUOCKYBRNumber of seedlings per sample 0510707580 Restoration Flatwoods Five most common species in the flatwoods soils ELSPHEUNDISPCYSPPOPR 0510152025707580 ab Figure 2-1. The mean number of seedlings per sample (+1 SE) in the germination assay for the five most common species in the a) restoration soils (n = 47) compared with the same species in the flatwoods soils and b) flatwoods soils (n = 42) compared with the same species in the restoration soils. Note different scales and species on axes. Species codes as follows: CYSP = Cyperus sp., HEUN = Hedyotis uniflora, SCDU = Scoparia dulcis, LUOC = Ludwigia octavalvis, KYBR = Kyllinga brevifolia, ELSP = Eleocharis sp., DISP = Dichanthelium portoricense, POPR = Polypremum procumbens. Cumulative number of samples 0510152025 01020304050 Cumulative number of samples 01020304050Cumulative number of species 01020304050 Restoration Flatwoods ab Figure 2-2. Mean species accumulation curves of observed species richness for the restoration and flatwoods soils in the a) germination assay and b) direct count assay. Isolated points set to the right are the ICE incidence-based estimators of species richness.

PAGE 48

36 Elevation strata 12345Total number of species 010203040506070 Common species Sporadic species Unique species 12345Total number of species 010203040506070 Species found in seedbank assays Species only found in field trial ab Figure 2-3. Relative abundance of species categories at the 5 elevation strata: a) species whether found in both the seed bank assays and vegetation survey, or species found only in vegetation survey and b) species whether common to all strata, found in between 2 and 4 strata (sporadic species), or species unique to each stratum.

PAGE 49

37 Elevation Strata 012345Plants m-2 02468 Lowest <-----------------------------------------------------------> HighestMean = 2.56 m-2 Figure 2-4. Mean (+1 SE) density of Aristida stricta (m-2, n=5) for each of the 5 elevation strata six months post-seeding in the first-year restoration site.

PAGE 50

38 0200400600800Cumulative number of species 20406080100 Field sampling Cumulative area sampled (m2) 0.000.020.040.060.080.10Cumulative number of species 0510152025303540 Germination Assay Direct Count Assay ab Figure 2-5. Comparison of the mean species accumulation curves of observed species richness for the restoration soils in the (a) direct count and germination trials, (b) field sampling scaled to the area of soil sampled. Isolated points set to the right are the ICE predicted species richness. Note the difference in scale of both axes.

PAGE 51

39 Number of seeds per sample 020406080100120140160180 abFlatwoods RestorationFlatwoods Restoration Figure 2-6. Number of seeds/seedlings per m2 for a) direct count assay (n = 40), b) germination assay (n = 89) in the flatwoods and restoration soils. Each point represents one sample.

PAGE 52

CHAPTER 3 VESICULAR-ARBUSCULAR MYCORRHIZAE IN UPLAND RESTORATION SITES, MESIC PINE FLATWOODS, AND PASPALUM NOTATUM PASTURES IN CENTRAL FLORIDA Introduction Increasing efforts to restore longleaf pine systems that have been converted to non-native pasture grasses (improved pasture) are underway in Florida. The goal of this research is to understand whether restoration of vesicular arbuscular-mycorrhizae (VAM) fungi needs to accompany that of native vegetation to ensure successful establishment of native plant species in upland restoration sites in central Florida. Understanding how VAM fungi respond to restoration efforts and the mycorrhizal role in the establishment of native pine flatwoods plant species such as Aristida stricta (wiregrass), can improve the success of pasture restoration efforts and facilitate future native plant restoration projects in the state. Vesicular-arbuscular mycorrhizae are soil-borne fungi that form symbiotic relationships with plants. VAM fungi associated with plant roots receive carbon from the plant while they aid plants in acquiring nutrients and water (Smith and Read 1997). This symbiosis is especially important for immobile nutrients like phosphorus because these fungi can bridge the phosphorus-depletion zone around plant roots and transport P to the plant (Jakobsen 1992, Sylvia 1999). Vesicular-arbuscular mycorrhizae fungi are vital ecological elements of ecosystems, can influence and be influenced by their aboveground plant community (Eom et al. 2000, Hart et al. 2001), and play a crucial role in plant community structure (van der Heidjen et al. 1998). Hart et al. (2001) suggest that VAM 40

PAGE 53

41 fungi are phenotypically variable and that host plant identity and environmental heterogeneity may be playing a major role in their behavior. Vesicular-arbuscular mycorrhizae fungi also aid in soil aggregation (Smith and Read 1997). Symbiotic relationships between plants and VAM fungi are widespread (Smith and Read 1997). Many plant species in the pine flatwoods including Serenoa repens (saw palmetto) (Fisher and Jayachandran 1999), A. stricta (Mullahey and Speed 1991), Andropogon virginicus var. glauca (chalky bluestem, listed as A. capillipes in Mullahey and Speed 1991), Andropogon virginicus (broomsedge bluestem) (Mullahey and Speed 1991), Liatris tenuifolia var. laevigata (Anderson and Menges 1997), Pityopsis graminifolia (Anderson and Menges 1997), and Balduina angustifolia (Anderson and Menges 1997) have symbiotic associations with VAM fungi. Restoration practitioners in Florida hope to introduce many of these native grasses and forbs to their upland restoration sites. Such herbaceous species, which produce ground fuels for the fast moving surface fires that are natural and essential in southeastern pine flatwoods (Pyne et al. 1996), aid in the restoration of the structure and function of the community. Vesicular-arbuscular mycorrhizae fungi presence in soils is closely related to soil chemistry. Vesicular-arbuscular mycorrhizae fungi tend to be more prevalent in low nutrient (Miller 1987) and acidic soils (Read et al. 1976), and less important in soils with high P availability (Read et al. 1976). The soils of the pine flatwoods are both acidic and low in nutrients (Abrahamson and Hartnett 1990). Land use and agricultural practices can also affect VAM fungi. Soil management such as tillage, soil disturbance, and fallow treatments decrease soil inoculum in some cases (Kabir 1999). Additions of phosphorus fertilizers can also decrease the efficacy of

PAGE 54

42 VAM fungi (Peng et al. 1993). Therefore in pastures where severe soil management has occurred and fertilizers are abundant, the presence of VAM fungi may be reduced. Vesicular-arbuscular mycorrhizae fungi have played an important role in land restoration projects such as coastal dune restoration (Sylvia and Burks 1988), restoration of mine spoil sites (Corbett 1996, Moynahan and Zabinski 2002), restoration of Sporobolus wrightii (big sacaton) grass in the desert southwest (Richter and Stutz 2002), and restoration of tallgrass prairie (Smith et al. 1998). Mycorrhizal fungi have been well studied in the tallgrass prairie community, which is analogous to the pine flatwoods understory in ecology and herbaceous species diversity and structure. The majority of tallgrass prairie species have symbiotic relationships with mycorrhizae, especially the warm-season grasses like Andropogon gerardii (big bluestem) and Sorghastrum nutans (Indian grass) (Miller 1997). At a tallgrass prairie restoration site, Smith et al. (1998) found increased percent cover and percent colonization of native species when VAM fungi inoculum was placed below the sown seeds. Miller (1997) suggests that in mesic sites with rich soils, the below-ground mycorrhizal changes that take place when an agricultural field is restored back to prairie mirror the changes in the above-ground plant community. Weedy mycorrhizal species are succeeded by a more diverse suite of non-weedy species that already exist in the soil and do not need to be reintroduced (Miller 1997). These generalizations may not be true for more xeric, nutrient poor, or disturbed sites (Miller 1997). Studies in the greenhouse (Kabir 1999), prairie restoration (Smith et al. 1998), arid regions (Allen 1989), and ultisols of Indonesia (Boddington and Dodd 2000) have shown that mycorrhizal fungi populations are significantly reduced by soil disturbance and/or

PAGE 55

43 fallow treatments. Reductions in soil inoculum could lead to early colonization, persistence, and competition by nonmycotropic plants. This response could have serious implications for pasture restoration in Florida since site preparation consisting of repeated cycles of soil disturbance and fallow periods is often required to reduce populations of non-native and weedy species (Bissett 1996, Harper-Lore 1998). Reductions in inoculum could also impact the ability of native plant species to persist in the restoration site. Mycorrhizae also show temporal variation in activity such as sporulation (Siguenza et al. 1996, Eom et al. 2000). Several researchers have shown that spore production tends to increase at the end of the growing season for the host plant (Ebbers et al. 1987, Siguenza et al. 1996). Phenology of VAM fungi have been shown to vary with the phenology of the plants with which they associate (Siguenza et al. 1996, Hartnett and Wilson 1999). During the course of pasture restoration, the floral species composition changes from a managed perennial monoculture, to a transient community dominated by weedy and annual species, to the target pine flatwoods natural community, characterized by diverse, perennial, herbaceous species. Because mycorrhizae can differ in their life history strategies (Bever et al. 2001), which can be influenced by their host plants, the mycorrhizal community is likely to reflect changes in the aboveground vegetative community. Therefore, in long-lived perennial dominated community types, the turnover and sporulation of VAM fungi would be much reduced compared to that of an annually dominated plant community (J. Graham 2002, pers. comm.). Additionally, diverse plant communities generally support diverse VAM fungi communities (Bever et al. 2001).

PAGE 56

44 It is not currently known whether the soil disturbance that is necessary for pasture restoration in Florida alters the soil mycorrhizal community and inoculation potential as has been shown in other studies. We also have no information on whether the suite of mycorrhizae species present in the soils of pastures and restoration areas are beneficial to native flatwoods plant species. If additions of mycorrhizae during restoration can increase percent cover and give the native species in Florida a competitive advantage over re-colonizing non-native species, the success of upland restoration efforts could be greatly improved. Understanding whether this important component of the floral community balance is negatively affected by restoration efforts seems vital to restoration success and persistence of the reintroduced floral community. In this study I examined VAM fungi in relation to soil chemistry, restoration activities, and temporal variation in pastures, pine flatwoods, and restoration areas in central Florida. By comparing mycorrhizae in each of three upland restoration areas with different starting conditions, I could evaluate immediate and short-term responses to restoration site preparation and native species planting efforts. The upland restoration complexes were: pre-restoration improved pasture, first-year restoration (hereafter one-year unit) and five-years following restoration (hereafter five-year unit) each coupled with an adjacent pasture and pine flatwoods. I hypothesized that the one-year unit would initially have significantly reduced MIP than the adjacent pasture and flatwoods, but would recover with time after seeding takes place. Therefore, the five-year unit should have higher MIP than the one-year unit, but lower MIP than the adjacent pasture and flatwoods. Further, I hypothesized that the VAM fungal root colonization of A. stricta plants in the five-year unit would be significantly less than that of A. stricta plants in

PAGE 57

45 natural flatwoods. Corollary predictions based on soil chemistry and temporal variation are as follows: Pastures and restoration sites, having been fertilized and limed in the past (The Nature Conservancy 1996), would have elevated concentrations of soil nutrients and pH relative to the flatwoods soils. Variations in soil chemistry, especially phosphorus and pH, would alter soil mycorrhizae inoculation potential (MIP) and A. stricta root colonization rates especially in the restoration and pasture soils. I predicted that there would be temporal variation among the three sampling dates for the MIP in the one-year complex, with higher soil inoculum in the fall and winter months because this is the time of greatest sporulation (Siguenza et al. 1996). Methods I evaluated the potential of the soil to colonize plants with VAM fungi through spores, root fragments or hyphae fragments using soil mycorrhizae inoculation potential (MIP) tests. MIP assays are a widely used method to evaluate soil inoculum (Sylvia 1994, Corbett 1996, Anderson and Menges 1997, Bray et al. 2003). The MIP method was chosen because I was interested in relative differences in different soils rather than the absolute number of propagules. I compared the MIP of soils in three restoration areas, each to their adjacent flatwoods, and pasture. In these sites field soils were used as inoculum in laboratory assays with host plants to estimate inoculation potential of the soils. In Experiment 1, I evaluated the MIP of a pre-restoration area and the one-year unit, each compared to adjacent pastures and flatwoods (Pre-restoration and First-year restoration complexes respectively). In Experiment 2, the five-year unit allows me to examine the longer-term effects of restoration on the MIP of the soils in restoration plots that were sown in 1998 compared to adjacent flatwoods and pastures. I also examined the

PAGE 58

46 VAM fungal colonization of A. stricta in the five-year unit and adjacent flatwoods. Soil chemistry, including nutrient content, pH, organic matter content, and soil moisture content, were also measured at all of the sites. The soils of the preserve have been mapped and all experiments for this study took place in Symrna or Myakka fine sands soil series. Experiment 1: First-year and Pre-restoration Complexes In the one-year unit, I established an experiment to determine whether soil disturbance during site preparation had a negative impact on the mycorrhizae inoculation potential (MIP) of the soil. Site preparation was initiated in the 10.7 ha (26.5 acre) one-year unit in June 2001. Glyphosate (3% concentrate) herbicide was broadcast on the pasture grasses, followed by soil disking (6 times over 6 months) to deplete the weedy seed bank. Seed collection (see chapter 1) occurred in November 2001. Immediately prior to seeding (February 2002), the site was prepared with a cultipacker to compact the disked soil. The seed was sown with a hay blower onto the prepared site. Non-native, invasive plant species were treated with herbicides, the only management performed on the site after seeding. The one-year unit was divided by elevation differences into five strata (1 to 5, lowest to highest elevation respectively). The overall elevation gradient in the restoration plot was only 0.45 m (1.5 ft.) The stratification was done by approximation using a topographical map and aerial photographs; 50 m X 50 m plot was marked at each elevation strata. Five plots were established in the adjacent flatwoods and pasture at corresponding elevations. The soils in these elevation strata were sampled several times throughout 2002. At each sampling period, in each stratum (15 total: 5 pasture, 5 restoration, 5 flatwoods) two soil samples (12 cm deep x 8 cm diameter) were

PAGE 59

47 composited and kept cold for one day during transportation back to the laboratory. I sampled the flatwoods soils on 28 June and 5 December 2002. I sampled the restoration area and the pasture soils on 2 March, 28 June, and 5 December 2002. The same sampling methods were used in a second restoration site at DWP, the pre-restoration complex. This area was scheduled for site preparation in 2002. As a result, I could compare the pre-restoration mycorrhizal community of this site and the experimental pasture (~9 km distant). In the pre-restoration complex, soils were collected on 30 May 2002 to evaluate the pre-restoration mycorrhizal conditions. The methods for the laboratory assays were from Sylvia (1994), International Culture Collection of Vesicular Arbuscular Mycorrhizal Fungi (2002), and personal communication with A. Alagely. For the laboratory assays, planting tubes were filled full with a sterilized soil mix composed of equal parts of sand, peat moss, and vermiculite. The sand was treated with 25% Muriatic acid for 24 hours and then rinsed for 24 hours with water to remove any phosphorus. A 20 ml sample of field soil from each elevation stratum was placed on the surface of the soil mix in each of 5 tubes (5 replicates). A 10 ml layer of the soil mix was then placed on top of the field soil and three Silver Queen corn seeds were placed on the soil. The corn seeds were purchased from a local feed store and were washed with soapy water for 30 minutes to remove the fungicide present. The seeds were covered with 15 ml of the soil mix. The sown tubes were placed in a growth chamber with a day/night cycle of 23 C night cycle for 8 h and a 28 C day cycle for 16 h. The light intensity inside the chamber for the day cycle was measured with a LI-250 (Li-Cor, Lincoln, Nebraska) and was 181.5 mol/m2/s. The plants were grown for 28 days (to detect only primary colonization, Sylvia 1994). They

PAGE 60

48 were fertilized on the day 10, 17, and 24 post-sowing with a modified Hoaglands solution (Table 3-1) (Sylvia et al. 2001). Occasionally, due to the vertical space limitations of the growth chamber the tips of the corn leaves had to be trimmed off to avoid growth into the lights. After 28 days the corn plants were removed from the growth chamber and root samples were taken from each plant. The root samples were placed into labeled tissue cassettes (OmniSette tissue cassettes, Fisher Scientific, Pittsburgh, PA). Root samples were cleared for 30 minutes in 1.8 M KOH to remove root pigments and then stained with Trypan Blue stain to enhance mycorrhizal structures according to methods in Sylvia (1994). Samples were stored in the refrigerator until they were examined microscopically. The Gridline-Intersect method was used to estimate the proportion of colonized root in the corn plants (Sylvia 1994). Stained root samples were spread out in a petri dish with scribed gridlines 1.27 cm (0.5 in) apart. The gridlines were scanned under a dissecting microscope and the total number of intersections of the gridlines with roots was counted as well as the number of roots intersected that were colonized by VAM. Colonization was determined when arbuscules, vesicles, or spores were seen. Hyphae that could not be traced to one of these structures were not counted. Any structures that could not be clearly identified through the dissecting scope were examined closely with a compound microscope. The root segment involved was mounted on a microscope slide in Polyvinyl-Lacto-Glycerol (PVLG) for future reference (International Culture Collection of Vesicular Arbuscular Mycorrhizal Fungi 2002), and examined with a compound

PAGE 61

49 microscope. Two hundred intersections were examined and the percentage of colonized to total intersections was calculated. Soil chemistry was evaluated at the same locations of the mycorrhizae tests for the one-year and pre-restoration complexes. Three randomly placed soil samples (12 cm depth x 4 cm diameter, from each stratum in each of the three locations (restoration, flatwoods, pasture)) were collected on 12 October 2002, air dried and sent to the Analytical Research Lab at the University of Florida for analysis of K, extractable P, Ca, Mg, total Kjeldahl nitrogen (TKN), and soil organic matter content (OM) (Analytical Research Laboratory 2003). pH of the same sample locations was measured using the methods for pH determination in water (Thomas 1996) on air dry soil. Soil samples were collected for soil moisture content analysis on 1 February 2003. Five randomly placed replicates within each elevation stratum were collected. Fresh mass was taken from each sample and recorded. All samples were dried to a constant mass at 60C. Dry mass was recorded and percent water content was calculated (% water content = mass of water/dry mass). Experiment 2: Five-year Complex Six restoration sites in the five-year complex (A-F) were established in 1997 throughout the preserve in six different pastures (with differing conversion and fertilization histories as pastures) (see Chapter 1) (The Nature Conservancy 1996). Paspalum notatum had been reduced by multiple herbicide (3% glyphosate concentrate (51% glyphosate, 49% inert ingredients)) applications prior to sowing with native flatwoods seed (The Nature Conservancy 1999). At the time of my sampling, these six restoration sites were five-years post restoration. All of the sites contained at least one A.

PAGE 62

50 stricta plant and high cover of Paspalum notatum and other non-native species (see Chapter 1). Roots from one A. stricta plant were harvested in each of the six restoration sites and their adjacent flatwoods on 30 July 2002. Three root sub-samples from each plant sample were separated, washed, cleared for 30 minutes in 1.8 M KOH, and stained (Sylvia 1994). A modified magnified intersections method described by McGonigle et al. (1990) was used to estimate percent colonization of A. stricta roots by VAM fungi. In this method, each sub-sample was spread thinly on a petri dish that was scored with a 1 cm grid. The roots were cut into one-cm segments and 30 one-cm segments were randomly chosen from the petri dish. These segments were mounted on microscope slides in PVLG. On each slide, two sets of five root samples were lined up parallel to the long side of the slide and covered with cover slips, resulting in three slides for each sub-sample. Data from the three slides were added together for each sub-sample. The slides were examined with a compound microscope at 40x. The microscope was fitted with a crosshair ocular that visually produced two thin crosshair lines in the field of view. The field of view was moved perpendicular to the root segments; colonization by VAM fungi was recorded and at each intersection of root with the center of the crosshairs (McGonigle et al.1990). Positive colonization was only recorded if arbuscles, vesicles, or coils were visible. On 12 October 2002, two composited soil cores were collected in five contrasting vegetation locations for each five-year unit for use in mycorrhizal inoculation procedure assays. The five vegetative locations including the following: random within restoration site, directly next to a random A. stricta plant within restoration site, random in improved

PAGE 63

51 pasture, random in flatwoods, and directly next to a random A. stricta plant in flatwoods. These soils were used to evaluate the mycorrhizal inoculation potential of the soil at those sites and vegetation locations to complement the A. stricta colonization data. From each soil sample (n=30) five 20 ml samples were assayed for MIP as described for Experiment 1 (See above, Sylvia 1994)). Soil chemistry was also evaluated as described for Experiment 1. Data Analysis Multiple regressions were performed on the one-year unit (June data only), pre-restoration, five-year unit data to see if the MIP and wiregrass colonization (for five-year unit) data were significantly related to the soil chemistry variables (SAS 1999). Spearmans correlations were included in the regression to determine whether any of the soil chemistry variables were correlated with each other (SAS 1999). For these analyses, means for each elevation stratum (for pre-restoration area and one-year unit) and each site (five-year unit) within restoration and flatwoods were used for the soil chemistry variables and MIP data. Raw data were used for A. stricta colonization. Pasture data was not included for the analysis of the one-year unit or the five-year unit because I expected pastures would be more affected by past fertilization activities but was included in the pre-restoration site because the measurements represent pre-restoration conditions (i.e., both the restoration and the pasture were bahia grass pastures). Experiment 1 Due to the skewness of the MIP dataset, normality was not achieved even with various transformations. A Kruskal-Wallace non-parametric test was used on the MIP data to determine if there were statistical differences among the elevation strata and treatments for each sampling date ( SAS 1999). Each soil chemistry variable was either

PAGE 64

52 normally distributed or transformed to approach normality. A two-way nested ANOVA was performed on each environmental variable to determine if there were statistical differences among the elevation strata and treatments for each of the sampling dates (SAS 1999). For these Kruskal-Wallace tests and ANOVAs, the five replicates were nested within elevation strata and Tukeys means comparisons (p < 0.05) were used to determine differences among all pairs of elevations and sites. A repeated measure ANOVA was also performed on the MIP data separately for the pasture and restoration treatments to see if there were significant differences among the three sampling dates (March, June, and December 2002) (SAS 1999). Since sampling only occurred in the flatwoods in June and December 2002, I performed an one-way ANOVA to see if date was a significant factor and used Tukey means comparisons (p < 0.05) to determine which date had higher MIP (SAS 1999). Experiment 2 A t-test was performed on the A. stricta colonization data to see if there were significant differences between those plants growing in natural flatwoods conditions and the restored five-year unit. As mentioned for Experiment 1, normality was not achieved in the MIP data and Kruskal-Wallace tests were used to test for differences in vegetation location (SAS 1999). ANOVAs were performed on each soil chemistry variable (P, K, Ca, Mg, TKN, pH, organic matter, and percent water content) to determine if there were statistical differences among the sites and vegetation locations (SAS 1999). Phosphorus and pH were log transformed to improve normality. For these soil chemistry ANOVAs, vegetation location was nested within site. Because I found no significant difference among the sites, all sites were lumped in order to test differences among vegetation

PAGE 65

53 locations. One-way ANOVAs were performed on each soil chemistry variable and a Kruskal-Wallace test was performed on the MIP data and means comparisons were made between all pairs of means using Tukey tests (p < 0.05) (SAS 1999). Results Experiment 1 I found no significant relationship between the MIP data and the soil environmental variables in the pre-restoration complex (restoration and flatwoods only) (Table 3-2). There were significant correlations among some of the soil chemistry variables, however (Table 3-2). In the one-year unit MIP was only positively correlated with soil water content and pH and many of the soil environmental variables were correlated with each other (Table 3-3). There was a significant positive relationship between MIP and pH in the one-year complex (restoration and flatwoods only) when restoration and flatwoods soils were combined (Regression: F = 7.04, p < 0.0291, r2 = 0.47) (Figure 3-1). Restoration site soil MIP was consistently higher than that of flatwoods site soil at equivalent pH (Figure 3-1). In the March 200 sampling of the one-year complex, I found no significant differences in MIP between pasture and restoration site or among elevation strata. In June, the restoration area soils had significantly higher MIP than the pasture and flatwoods soils (Kruskal-Wallace: F = 22.88, p < 0.0001, Figure 3-2). Additionally, elevation Stratum 2 (19.6 m) had a significantly higher MIP than did Stratum 4 (19.98 m) (Kruskal Wallace: F = 3.37, p < 0.03, Figure 3-3) In December there was no significant difference among the elevation strata but the pasture had significantly higher MIP levels than the flatwoods (Kruskal-Wallace: F = 6.35, p < 0.03, Figure 3-2).

PAGE 66

54 Results of the repeated measures analysis of variance indicate a significant difference among the three sampling dates for the restoration soils (over time: F = 13.56, p < 0.0002) but not for the pasture soils. In June, MIP of the restoration soils was significantly higher than on the other two sampling dates (Figure 3-2). In the flatwoods soils, there was no significant difference between the two sampling dates (Figure 3-2). All soil environmental variables except for TKN and organic matter varied significantly by restoration treatment (Tables 3-4, 3-5). Flatwoods soils had significantly higher soil water content than did pasture and restoration soils (Two-way nested ANOVA: p < 0.05, Figure 3-4). Pasture soils had significantly higher pH than restoration and flatwoods soils (Figure 3-5). The pasture soils had the highest means among the three treatments for all variables except for water content and organic matter; but differences were not always significant (Figure 3-4, 3-5, 3-6, Table 3-5). Similarly, variables except for K, pH, and organic matter varied significantly by elevation (Table 3-4). There was no consistent relationship between elevation stratum and concentration of soil nutrients, however, the lower elevations (strata 1 and 2) tended to have high means for soil chemistry compared to the mid elevations (strata 3 and 4) (Table 3-6). The interaction between treatment and elevation strata was significant for P, Ca, and TKN (Table 3-4) with no consistent pattern (Figures 3-4, 3-6, 3-7). Experiment 2 Aristida stricta root colonization by VAM fungi was 25.8% ( 6.1 SE) in the flatwoods and 25.5% ( 4.4 SE) in the five-year unit. However these data are extremely variable on a per plant and per site basis (Figure 3-8). The colonization level was dependent on none of the combinations of variables examined. The correlation matrix among the 10 soil variables (A. stricta colonization,

PAGE 67

55 MIP, P, K, Ca, Mg, TKN, pH, organic matter, and water content) resulted in several significant correlations among the variables. As seen in the one-year site, many of the nutrients were correlated with each other. Phosphorus (p < 0.02) and Ca (p < 0.01) were both positively correlated with organic matter. Potassium was positively correlated with TKN (p < 0.05). Water content was positively correlated with K (p < 0.04) and TKN (p < 0.01). I similarly found no dependence of MIP on the soil chemistry variables (Table 3-7). Because I found no significant difference among the 6 sites when vegetation locations were nested, location was directly tested across sites. Soil chemistry was not significantly different across vegetation locations for Ca, pH, K, Mg, and organic matter. Flatwoods soils had significantly higher TKN content than restoration soils (One-way ANOVA: F = 3.22, p < 0.05, Figure 3-10) and significantly higher water content than pasture and restoration soils (One-way ANOVA: F = 8.65, p < 0.01, Figure 3-11). Pasture soils had significantly higher P content than the restoration area soils (One-way ANOVA: F = 4.97, p < 0.01, Figure 3-12). The MIP differed significantly by vegetation location, (Kruskal-Wallace: F = 3.02, p < 0.0001; Figure 3-9) but no significant differences in MIP were found among the sites in the five-year unit. When sites are lumped and vegetation locations are tested, the soils from restoration wiregrass and pasture had significantly higher soil MIP than the flatwoods wiregrass soils (Kruskal-Wallace: F = 4.56, p = 0.0017) (Figure 3-9). Discussion Experiment 1 Results of the MIP data coupled with the A. stricta colonization rates, suggest that inoculum is not lacking in the restoration areas at DWP and that site-preparation

PAGE 68

56 activities are not resulting in long term decreases in soil inoculum. There were no important relationships found between the MIP and soil chemistry variables tested and variation in the soil inoculum at the dates sampled in these study sites seem to follow the temporal variation seen in the aboveground vegetation. As hypothesized, I found elevated levels of soil nutrients in the pasture soils. Although not always significant, there was a general trend of the pastures having the highest concentration of soil chemistry variables and the highest pH of the three treatments followed by the restoration soils and then the flatwoods soils (Tables 3-4 and 3-5). There was a significant interaction between treatment and elevation strata for several soil chemistry variables (water content, TKN, and Ca) but no consistent pattern for the interaction. pH was one of the only factors that was correlated with MIP in the one-year unit. While fungi are generally more dominant in acidic soils (Hartel 1999), my MIP levels were higher in the higher pH soils (Figure 3-1). However, the small pH range (4.1 4 .5) involved could minimize the importance of this result. The MIP of all the sites, including flatwoods dominated by perennial native plants, was low (Figures 3-2, 3-9), and similar to that of other studies in Florida sandhill (Anderson and Menges 1997) and warm-season grass areas of a prairie in Illinois (Corbett 1996). Results of the MIP data at the one-year unit suggest that mycorrhizae recovered from the soil disturbance caused by the restoration site preparation. The VAM fungi were not depleted by the soil disturbance and fallow treatment (Figure 3-2) as in other studies (Smith et al. 1998, Kabir 1999). The amount and severity of soil disturbance in our restoration methods may not be sufficient to result in long-term decreases in soil

PAGE 69

57 inoculum. These results indicate that mycorrhizal inoculum may not be needed in upland restoration efforts in central Florida. The significant increase in the MIP of the soil at the one-year unit six months post-sowing, however, provides no information about the species and diversity of mycorrhizae present in that soil. Studies in many areas have recently found that VAM fungi can be host specific (McGonigle and Fitter 1990, Hartnett and Wilson 1999, Eom et al. 2000, Bever et al. 2001). Therefore, just having a significant increase in VAM fungi post-restoration does not imply that reintroduced native species can utilize these beneficial fungi. Sylvia (1986) found that a beach dune restoration site also showed an increase in VAM fungi just after restoration. After further investigation, however, the increase was partially due to abundance of a non-symbiotic fungus (Sylvia 1986, Sylvia and Burks 1988). In an Illinois prairie restoration, Miller (1997) found that weedy mycorrhizal species are succeeded by a more diverse suite of species that already exist in the soil and do not need to be introduced. However, restoration of disturbed lands to diverse plant communities relies on the presence and diversity (both functional and taxonomic) of VAM fungi (Bever et al. 2001). High VAM fungi can lead to high plant diversity (van der Heidjen et al. 1998). Further work should investigate whether the VAM fungi present in our disturbed restoration site soil are beneficial to the native plant species. Root colonization of native desirable plant species growing in the restoration area throughout the first years of the restoration process (compared to those growing in undisturbed natural areas) would also be beneficial information to complement my results.

PAGE 70

58 Zea mays is the standard host plant for use in MIP assays because it is very mycotropic and the roots are cleared easily and fungal structures stain well ( International Culture Collection of Vesicular Arbuscular Mycorrhizal Fungi 2003). However, it is possible that the fungi colonizing the Zea mays plants in the soils I tested was only a subset of the total fungi in the soil or that Zea mays is not the correct assay plant for these flatwoods soils. Future studies could possibly use native plants as hosts in laboratory assays to investigate these possibilities. The temporal variation in the soil inoculum supports the observations that sporulation increases in the late growing season and the fungal phenology is related to plant species phenology (Siguenza et al. 1996, Eom et al. 2000). Although not significant over time in the repeated measures ANOVA, I saw a trend of increasing soil MIP in the pasture and flatwoods in the December sampling date which corresponds to the end of the growing season for P. notatum and the perennial flatwoods species (Figure 3-2). In the restoration soils, temporal variation was significant and MIP was highest in June, which corresponds to the phenology of the multitude of annual species at the end of their life cycle in the restoration area at that time (Figure 3-2). Experiment 2 I found no significant relationship between the A. stricta colonization and soil environmental variables in the five-year unit. Mullahey and Speed (1991) similarly found no correlation between root colonization of four Florida native grasses including A. stricta and soil nutrients. In another study, Medina et al. (1988) also found no relationship between root colonization of tropical forage legumes and soil nutrients in South Florida. The MIP results for the five-year complex were consistent with the one

PAGE 71

59 year complex. The flatwoods soils had low mycorrhizal inoculum while the restoration and pastures had higher levels. Since there was a significant difference among the vegetation locations in P content, I expected to see that difference influence the MIP data. However, P may have been at such low concentrations or over such a small range of values in both experiments that the correlation could not be determined (Schwandes et al. 2001) (D. Sylvia 2003, pers. comm.). In this study the flatwoods soils averaged 2.55 mg/kg, the restoration averaged 2.3 mg/kg, and the pasture soils averaged 3.15 mg/kg, which is substantially lower than the 4.7 mg/kg and 9.9 mg/kg reported for north and south Florida flatwoods soils respectively (Schwandes et al. 2001). The TKN content in the soils of the flatwoods was significantly higher than in the five-year unit restoration soils where considerable soil disturbance and fallow period had taken place. Perhaps the nitrogen in the restoration soils was leached out of the system during site preparation because of the low nitrogen demand during the period with no or few plants growing in the system. This result was not consistent across experiments. Aristida stricta plants growing in the restoration areas for five years reach the same colonization rates by mycorrhizae as those growing in natural flatwoods systems. This result suggests that the mycorrhizae present in the soils of the restoration areas can colonize A. stricta. Larger sample sizes were not possible in this experiment due to the lack of A. stricta plants in some restoration sites but would potentially clarify the variation seen on a per plant basis. However, while colonization rates are similar we have no information about species identity or richness and other native plants species were not tested in this study. Further evaluation of other native plant species and VAM fungi

PAGE 72

60 identity and diversity could increase our knowledge of how mycorrhizae are responding after restoration and how well other native plants utilize the mycorrhizal species present. This information could help explain the persistence problem in the five-year restoration unit. I found higher colonization rates in A. stricta than has been found in other studies of this species (Mullahey and Speed 1991, Anderson and Menges 1997). Anderson and Menges (1997) found no colonization of A. stricta by VAM fungi in a sandhill in Highlands County, FL Mullahey and Speed (1991) collected twelve A. stricta plants in native range in South Florida (three each from four locations) and concluded an average root colonization percentage by VAM fungi of 8%. Our average of 25% colonization is considerably higher. Therefore, A. stricta colonization rates may vary by soil or plant community type. The results of earlier studies at DWP indicated that diversity and abundance of native plant species in the five-year unit decreased over the three years of post-restoration vegetation monitoring (see Chapter 1) (The Nature Conservancy 1999). Many native species that emerged in the restoration did not persist over time while there was an increase in the cover of non-native species (The Nature Conservancy 1999). One possible explanation for this lack of persistence could be the lack of the correct suite of mycorrhizal species in the restoration soils. VAM fungi host specificity can be related to plant species specificity (Dhillion and Zak 1993). Significant levels of host specificity have been noted in the tallgrass prairie ecosystem (Hartnett and Wilson 1999, Eom 2000). It is therefore important to know whether the suite of VAM fungi species in pre-restoration sites are specialists or generalists and whether the restored vegetation can

PAGE 73

61 utilize the suite of VAM species present in the restoration site. Johnson et al (1991) found that forests and old fields were dominated by different species of VAM fungi. A. stricta can utilize the species of mycorrhizae present in the restoration soils (see Experiment 2) at DWP and persisted in the five-year restoration units but it is possible that other native plants cannot, which could affect persistence. Testing other native plant species colonization rates by VAM fungi at the early stages of restoration would clarify host specificity questions further. Conclusions/Applications Important distinctions that can be drawn from the data in both experiments are the differences between the MIP of the managed and unmanaged soils and of perennially and annually dominated plant assemblages. In all experiments the native pine flatwoods soil had lower inoculum than soils in the pastures or restoration areas, both areas under management activities. Lower spore yields were reported in soils of native woodlands compared to that of six agronomic crops that were planted in adjacent cleared woodlands in Northwest Florida (Schenck and Kinloch 1980). Above ground vegetation has been noted to influence mycorrhizal community (Bever et al. 2001) and seems to be the case for my study as well. The native pine flatwoods is also a system dominated by long-lived perennial plant species and therefore may not have high turnover of plants and VAM fungi. Low incidence of sporulation by native perennial plants in England (Read et al. 1976) and in a north Florida woodland (Schenck and Kinloch 1980) have been reported. The consistently higher MIP in the restoration soils than the perennially dominated flatwoods in this study may, therefore, have reflected the dominance of annual plants in the aboveground vegetation, high plant turnover, and relatively high soil inoculum.

PAGE 74

62 Clearly VAM fungi inoculum is more dynamic in the restoration area than in the native flatwoods in this central Florida site. This mirrors the substantial changes that are occurring with the aboveground vegetation in the site. This study did not clarify whether the suite of VAM fungal species in the restoration is appropriate for use by the native flatwoods species being introduced in the restoration. However, there is convincing evidence from the MIP data and the A. stricta root colonization data that mycorrhizae inoculum is not limiting in this restoration. This should come as good news to practitioners of restoration in Florida.

PAGE 75

63 Table 3-1. Composition of the modified Hoaglands fertilizer solution (Sylvia et al 2001). Stock Chemical ml/L 0.01M KH2PO4 0.3 1M KNO3 1.5 NaFeEDTA 0.3 0.1M NaCl 0.45 1M Ca(NO3)2 1.5 1M MgSO4 0.6 Micronutrient Stock 0.3 H3BO3.86 g/L MnCl2.81 g/L ZnSO4.22 g/L CuSO4.08 g/L NaMoO4.02 g/L Table 3-2. Spearman correlation matrix for MIP and soil chemistry flatwoods and pre-restoration treatments only; only p < 0.05 shown. Non-correlated variables are indicated with dashes (--). MIP P K Ca Mg TKN OM water MIP --------P --0.0066 0.0191 0.0042 ---K -0.0066 --0.0001 0.0313 --Ca -0.0191 ------Mg -0.0042 0.0001 --0.0300 --TKN --0.0313 -0.0300 -0.0220 -OM -----0.0220 --water --------

PAGE 76

Table 3-3. Spearman correlation matrix for MIP and soil chemistry variables in the one-year complex (restoration and flatwoods only; significant p-values shown) in June. Non-correlated variables are indicated with dashes (--). 64 MIP P K Ca Mg TKN pH OM water MIP ------0.0089 -0.0245 P --0.0061 --0.0111 -0.0072 -K -0.0061 -------Ca ----0.0008 ---0.0005 Mg ---0.0008 ----0.0022 TKN -0.0111 -----0.0392 -pH 0.0089 --------organic matter -0.0072 ---0.0392 ---water content 0.0245 --0.0005 0.0022 ----Table 3-4. Results of two-way nested ANOVA for each soil chemistry variable on restoration treatment (pasture, restored pasture, native flatwoods), elevation (5 strata) and their interaction (Treatment x Elevation) within the one-year complex. Non-significant tests are indicated with dashes (--). Treatment Elevation Treatment x Elevation Variable df F-value P-value df F-value P-value df F-value P-value Water content 2 21.99 0.0001 4 5.88 0.0027 8 4.44 0.0032 P 2 5.12 0.0160 4 5.97 0.0101 8 --K 2 7.07 0.0048 4 --8 --Ca 2 38.99 0.0001 4 10.36 0.0014 8 4.49 0.0151 Mg 2 11.25 0.0005 4 4.64 0.0223 8 --TKN 2 --4 3.99 0.0345 8 3.52 0.0333 pH 2 50.31 0.0001 4 --8 --Organic matter 2 --4 --8 --

PAGE 77

65 Table 3-5. Results of Tukey mean comparisons of the three treatments in Experiment 1 for each significant variable. Treatments for each variable are listed from highest to lowest mean (left to right respectively). Different letters indicate significant differences (p < 0.05). Variable Treatments water content Flatwoodsa Pastureb Restorationb P Pasturea Flatwoodsab Restorationb K Pasturea Restorationb Flatwoodsb Ca Pasturea Restorationa Flatwoodsb Mg Pasturea Restorationab Flatwoodsb pH Pasturea Restorationb Flatwoodsb Table 3-6. Results of the Tukey mean comparisons of the five elevation strata in Experiment 1 for significant variables. Elevation strata (numerically from lowest to highest elevation) for each variable are listed from highest to lowest mean (left to right respectively). Different letters indicate significant differences (p < 0.05). Variable Elevation Strata water content 1a 3ab 5ab 2b 4b P 2a 5a 1ab 4ab 3b TKN 2a 1ab 5ab 4ab 3b Ca 1a 2a 5ab 3b 4b Mg 1a 2ab 5ab 3ab 4b Table 3-7. P-values from the Spearmans correlation matrix of eight soil variables in Experiment 2 (significant p-values shown). Non-correlated variables are indicated with dashes (--). MIP P K Ca Mg TKN pH OM water content MIP ---------P -------0.0202 -K -----0.0548 --0.0415 Ca -------0.0102 -Mg ---------TKN --0.0548 -----0.0139 pH -0.0202 -------OM ---0.0102 -----Water content --0.0415 --0.0139 ---

PAGE 78

66 pH 4.04.14.24.34.44MIP (%) .5 0246810121416 p < 0.0291R2 = 0.47 Flatwoods Restoration Figure 3-1. Correlation between MIP (n = 5) and pH (n = 3) for soils from flatwoods and one-year unit restoration soils in June 2002. Treatments RestorationPastureFlatwoodsMean Inoculation Potential (%) 02468101214 March June December abbAABB Figure 3-2. Mean mycorrhizal inoculation potential (+1 SE, n = 25) of the one-year unit, and adjacent flatwoods and pasture. Different letters indicate significant differences and capital and lower case letters indicate separate univariate tests.

PAGE 79

67 Elevation strata 12345MIP (%) 024681012 abaabbab Figure 3-3. Mean inoculation potential (+1 SE, n = 15) of the 5 elevation strata in June 2002 (1 = lowest, 5 = highest) for restoration, flatwoods and pasture together in the Experiment 1. Different letters indicate significant differences. Elevation Strata 12345Gravimetric moisture content (%) 0.150.200.250.300.350.400.45 Flatwoods Pasture Restoration LOWEST <----------------------------------------------------------------------> HIGHEST Figure 3-4. Mean gravimetric soil moisture content (+1 SE, n = 5) for each treatment by elevation strata in the one-year complex.

PAGE 80

68 Evelation strata 0123456pH 34567 Flatwoods Pasture Restoration Figure 3-5. Mean pH (+1 SE, n = 3) for each treatment by elevation in the one-year complex. Blocks by elevation 12345Soil P Content (mg/kg) 1.01.52.02.53.03.54.04.55.0 Flatwoods Pasture Restoration Figure 3-6. Mean soil P content (mg/kg) (+1 SE, n=3) for each treatment in the one-year complex.

PAGE 81

69 Blocks by elevation 12345Soil TKN (g/kg) 0123456 Flatwoods Pasture Restoration Figure 3-7. Mean soil total Kjeldahl nitrogen (g/kg) (+1 SE, n=3) for each treatment in the one-year complex. Sites ABCDEFPercent colonization 0102030405060 Restoration area Flatwoods Figure 3-8. Percent colonization (n = 1) of Aristida stricta roots by VAM fungi in the six restoration sites and adjacent flatwoods in the Experiment 2.

PAGE 82

70 Treatments FlatwoodsFlatwiregrassPastureRestorationRestwiregrassMIP (%) 01234567 abbaaba Figure 3-9. Mean inoculation potential (+1 SE, n = 30) of the soils in the five treatments of the five-year complex. Different letters indicate significant differences among the five treatments. Treatments FlatwoodsPastureRestorationSoil TKN (g/kg) 0.00.51.01.52.02.5 aabb Figure 3-10. Mean soil total Kjeldahl nitrogen (+1 SE, n = 18) of the restoration and adjacent flatwoods and pasture soils in Experiment 2. Different letters indicate significant differences.

PAGE 83

71 Treatments FlatwoodsPastureRestorationGravimetric soil moisture (%) 051015202530 abb Figure 3-11. Mean soil gravimetric moisture content (+1 SE, n = 30) of the restoration and adjacent flatwoods and pasture soils in Experiment 2. Different letters indicate significant differences. Treatments FlatwoodsPastureRestorationSoil Extractable P (mg/kg) 01234 abab Figure 3-12. Mean soil extractable P content (+1 SE, n = 18) of the restoration and adjacent flatwoods and pasture soils in Experiment 2. Different letters indicate significant differences.

PAGE 84

APPENDIX SPECIES LISTS

PAGE 85

Table A-1. Species list for the germination assay with life form and category. Flatwoods Elevation Strata Restoration Elevation Strata species Life form category 1 2 3 4 5 mean 1 2 3 4 5 mean Andropogon sp. grass nc 0 0 0.17 0.1 0.22 0.10 0 0 0.2 0 0 0.04 Chamaecrista nictatans forb nc 0 0 0 0 0 0.00 0 0.11 0 0 0 0.02 Cyperus surinamensis forb nc 0 0.13 0 0.2 0 0.07 0 0 0 0.1 0 0.02 Dichanthelium sp. grass nc 0.67 0.75 0.67 0.7 0.56 0.67 0 0.11 0.1 0 0.13 0.07 Eleocharis viviparous sedge nc 0 0 0 0.1 0.11 0.04 0 0 0 0 0 0.00 Hedyotis uniflora forb nc 0.78 0.63 0.5 0.6 0.44 0.59 0.3 0.56 0.8 0.2 0 0.37 Lindernia sp. forb nc 0 0 0 0 0 0.00 0 0.11 0.2 0.2 0.38 0.18 Ludwigia maritima forb nc 0 0 0 0 0.11 0.02 0 0 0 0 0 0.00 Ludwigia repens forb nc 0 0 0 0 0 0.00 0.2 0 0.1 0.1 0.13 0.11 Polypremum procumbens forb nc 0 0.25 0 0 0 0.05 0.2 0.44 0.7 0.2 0.25 0.36 Toxicodendron radicans forb nc 0 0 0.17 0 0 0.03 0 0 0 0 0 0.00 Dichanthelium portericense grass nc 0.22 0 0.33 0.1 0.33 0.20 0 0 0 0.1 0 0.02 Crotolaria rotundifolia forb nc 0 0 0 0 0.11 0.02 0 0 0 0 0 0.00 Cyperus compressus sedge nc 0.11 0.13 0 0.1 0.22 0.11 0.5 0.89 0.6 0.9 0.88 0.75 Cyperus polystachyos sedge nc 0 0.13 0.17 0 0.22 0.10 0.6 1 0.8 0.8 1 0.84 Drosera brevifolia forb nc 0 0 0.33 0 0 0.07 0 0 0 0 0 0.00 Eleocharis filiform sedge nc 0.89 0.25 1 0.2 0.56 0.58 0 0.11 0.4 0.3 0 0.16 Rhynchospora fascicularis sedge nc 0.11 0 0 0 0 0.02 0 0 0 0 0 0.00 Scleria reticularis sedge nc 0.11 0 0 0 0 0.02 0 0 0 0 0 0.00 Scleria sp. sedge nc 0 0 0.17 0.1 0 0.05 0 0 0 0 0 0.00 Xyris like sedge nc 0.11 0 0.5 0 0 0.12 0.1 0 0 0 0 0.02 Gnaphalium sp. forb nw 0.22 0.38 0.17 0.2 0.33 0.26 0 0.11 0.2 0 0.38 0.14 Ludwigia octavalvis forb nw 0.11 0.13 0 0 0 0.05 0.1 0 0 0.1 0.13 0.07 Oxalis corniculata forb nw 0.44 0 0.5 0.2 0.11 0.25 0.3 0.33 0.2 0.3 0.5 0.33 Scoparia dulcis forb nw 0 0.13 0 0 0.11 0.05 0 0.33 0.6 0.2 0.38 0.30 73

PAGE 86

74 Table A-1. Continued Flatwoods Elevation Strata Restoration Elevation Strata species Life form category 1 2 3 4 5 mean 1 2 3 4 5 mean Cyperus glogulosus sedge nw 0 0 0 0 0 0.00 0 0 0 0.1 0 0.02 Cyperus retrorsus sedge nw 0 0 0.17 0.1 0 0.05 0.2 0.22 0.4 0.4 0.25 0.29 Erechtites hieracifolia forb nw 0 0 0 0 0 0.00 0 0 0 0 0.13 0.03 Fimbristylis autumnalis sedge nw 0.11 0.13 0 0 0.11 0.07 0.2 0.11 0.1 0.4 0.63 0.29 Fimbristylis dichotoma sedge nw 0 0 0.17 0 0 0.03 0.3 0.11 0.2 0.4 0.13 0.23 Kyllinga brevifolia sedge nw 0.11 0 0.17 0 0 0.06 0.4 0 0.2 0.4 0.13 0.23 Hedyotis corymbosa forb nnw 0 0 0 0 0 0.00 0.1 0 0 0 0.25 0.07 Kummerowia striata forb nnw 0 0 0 0 0 0.00 0.1 0 0 0 0 0.02 Murdannia nudiflora forb nnw 0 0.13 0 0 0 0.03 0 0.11 0.5 0.1 0.25 0.19 Desmodium triflorum forb nnw 0 0 0.17 0 0 0.03 0.2 0.22 0.1 0.2 0.13 0.17 Fimbristylis schoenoides sedge nnw 0 0 0 0 0 0.00 0.3 0 0 0.1 0 0.08 Asteraceae basal rosette forb 0 0 0 0.1 0 0.02 0 0 0 0 0 0.00 Ludwigia round leaf forb 0.11 0.13 0.33 0.1 0 0.13 0.1 0 0.1 0 0 0.04 Ludwigia sp. forb 0 0 0 0 0 0.00 0.1 0.22 0.2 0.3 0.13 0.19 Cyperus sp. sedge 0.33 0.13 0.17 0 0 0.13 0.1 0 0 0 0 0.02 # flats destroyed 1 2 4 0 1 1.60 0 1 0 0 2 0.60 Categories include: native characteristic = nc, native weedy = nw, non-native weedy = nnw, non-native invasive = nni (D. Gordon, unpublished data). Species without a category were not identified clearly enough to determine which category they belong to. Native species are list first.

PAGE 87

75 Table A-2. Species list for the direct count assay with life form and category Flatwoods Restoration species life form category species life form category Caryophylaceae/Portulaca/Chenopod forb Cyperus compressus sedge nc Dichanthelium portoricense grass nc Cyperus globulosus sedge nw Fimbristylis autumnalis sedge nw Cyperus polystachyos sedge nc legume small tan forb Desmodium triflorum forb nnw Paspalum/panicum round grass Dichanthelium portoricense grass nc Poaceae huge seed grass Fimbristylis autumnalis sedge nw Poaceae long awn grass Fimbristylis dichotoma sedge nw Rhynchospora microcephala? sedge nc Fimbristylis schoenoides? sedge Rhynchospora sp. sedge nc Kummerowia striata forb nnw Rubiaceae forb Kyllinga brevifolia sedge nw Scleria reticularis sedge nc legume small tan Scleria sp. sedge nc Rhynchospora sp. sedge nc unknown Rhynchospora tracyii? sedge nc Scleria reticularis sedge nc unknown Categories include: native characteristic = nc, native weedy = nw, non-native weedy = nnw, non-native invasive = nni (D. Gordon, unpublished data). Species without a category were not identified clearly enough to determine which category they belong to.

PAGE 88

Table A-3. Species list for the field vegetation sampling by elevation strata (n=5 strata) and frequency by plot size. Elevation Strata 1 1 1 2 2 2 3 3 3 4 4 4 5 5 5 mean mean mean Species life form category 0.2x0.2 1x1 5x5 0.2x0.2 1x1 5x5 0.2x0.2 1x1 5x5 0.2x0.2 1x1 5x5 0.2x0.2 1x1 5x5 .2x.2 1x1 5x5 Aster fistifolius Forb nc 0 0 0 0 0 0.2 0 0 0 0 0 0 0 0 0 0 0 0.04 Centella asiatica Forb nc 0.6 0.6 0.6 0.4 0.6 1 0 0.2 0.6 0 0.2 0.8 0 0 0 0.2 0.32 0.6 Chamaecrista nictatans Forb nc 0 0 0.6 0.4 0.6 1 0 0.8 1 0.4 0.8 1 0.2 1 1 0.2 0.64 0.92 Crotolaria rotundifolia Forb nc 0 0 0 0 0 0.2 0 0.2 0.4 0 0 0.2 0 0 0 0 0.04 0.16 Diodia virginiana Forb nc 0 0.2 1 0.4 0.4 0.8 0 0 0 0 0.2 0.4 0.2 0.2 0.4 0.1 0.2 0.52 Euthamia minor Forb nc 0 0 0.2 0 0 0 0 0 0.4 0 0 0 0 0 0 0 0 0.12 Hedyotis uniflora Forb nc 0.6 1 1 0.2 0.6 0.8 0 0.2 0.4 0.2 1 1 0.2 0.6 0.8 0.2 0.68 0.8 Hydrocotyle umbellata Forb nc 0 0.4 0.8 0.2 0.2 0.4 0 0.2 0.2 0 0 0.2 0.4 0.4 0.6 0.1 0.24 0.44 Hypericum cistifolium Forb nc 0 0 0.2 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.04 Hypericum tetrapetalum Forb nc 0 0 0.2 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.04 Lachnanthes caroliniana Forb nc 0 0 0.2 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.04 Lindernia group? Forb nc 0 0 0 0 0 0 0 0 0 0 0.2 0.2 0 0 0 0 0.04 0.04 Ludwigia arcuata Forb nc 0 0 0.2 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.04 Ludwigia maritima Forb nc 0 0 0.4 0 0.2 1 0 0.4 0.6 0 0.2 0.2 0 0.2 1 0 0.2 0.64 Ludwigia repens Forb nc 0 0.6 1 0 0.6 0.6 0 0 0 0 0.4 0.8 0 0.6 0.8 0 0.44 0.64 Phyla nodiflora Forb nc 0 0 0 0 0 0.2 0 0 0 0 0 0 0 0.2 0.2 0 0.04 0.08 Phytolaca americana Forb nc 0 0 0 0 0.2 0.4 0 0 0.2 0 0 0 0 0 0.4 0 0.04 0.2 Pityopsis graminifolia Forb nc 0 0 0 0 0 0 0 0 0 0 0 0.2 0 0 0 0 0 0.04 Pluchea odorata Forb nc 0 0 0.2 0 0 0 0 0 0 0 0 0.2 0 0 0 0 0 0.08 Pluchea rosea Forb nc 0 0 0.4 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.08 Polygonum punctatum Forb nc 0 0 0 0 0 0.2 0 0 0 0 0 0 0 0 0 0 0 0.04 76

PAGE 89

77 Table A-3. Continued Elevation Strata 1 1 1 2 2 2 3 3 3 4 4 4 5 5 5 mean mean mean Species life form category 0.2x0.2 0.2x0.2 1x1 5x5 1x1 5x5 0.2x0.2 1x1 5x5 0.2x0.2 1x1 5x5 0.2x0.2 1x1 5x5 .2x.2 1x1 5x5 Polypremum procumbens Forb nc 0.2 1 1 0.6 0.8 1 0.6 1 1 0.6 0.8 1 0.2 1 1 0.4 0.92 1 Rhexia nashii Forb nc 0 0 0.2 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.04 Sabatia grandiflora Forb nc 0 0 0.2 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.04 Amphicarpum muhlenbergianum grass nc 0 0.2 0.4 0 0 0 0 0 0 0 0 0.2 0 0 0 0 0.04 0.12 Andropogon glomeratus glaucopsis grass nc 0 0.2 0.4 0 0 0 0 0 0 0 0 0 0 0 0 0 0.04 0.08 Andropogon glomeratus hirsuitior grass nc 0.2 0.2 0.4 0 0 0 0 0 0 0 0.2 0.4 0 0 0 0 0.08 0.16 Andropogon virginicus grass nc 0 0 0 0 0 0 0 0 0 0 0 0.4 0 0 0 0 0 0.08 Andropogon virginicus glauca grass nc 0 0 0 0 0 0 0 0 0 0 0 0.4 0 0 0 0 0 0.08 Aristida beyrichiana grass nc 0.4 0.8 0.8 0 0.8 1 0.4 0.6 1 0 1 1 0.4 1 1 0.2 0.84 0.96 Axonopus furcatus grass nc 0 0.2 0.4 0 0 0.2 0 0 0 0 0 0.2 0 0 0 0 0.04 0.16 Dichanthelium ensifolium grass nc 0 0 0 0 0 0 0 0 0 0 0 0.4 0 0 0 0 0.04 0.08 Dichanthelium portoricense grass nc 0 0 0 0 0 0 0 0.2 0.2 0.2 0.4 0.4 0 0 0.4 0 0.12 0.2 Leersia hexandra grass nc 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.2 0 0 0.04 Paspalum laeve grass nc 0 0 0 0 0 0 0 0 0.2 0 0 0 0 0 0 0 0 0.04 Paspalum setaceum grass nc 0 0 0.2 0 0 0.2 0 0 0 0 0 0 0 0 0 0 0 0.08 Cyperus compressus sedge nc 0.2 0.2 0.4 0.4 1 1 0.6 0.8 1 0.4 0.8 1 0.6 1 1 0.4 0.76 0.88 Cyperus haspan sedge nc 0 0 0.4 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.08 Cyperus surinamensis sedge nc 0 0 0.6 0 0 0.4 0 0 0 0.2 0.6 0.6 0 0 0 0 0.12 0.32 Eleocharis filiform sedge nc 0 0.6 1 0.2 0.2 0.6 0 0 0 0.2 0.4 0.6 0 0 0 0.1 0.24 0.44

PAGE 90

78 Table A-3. Continued Elevation Strata 1 1 1 2 2 2 3 3 3 4 4 4 5 5 5 mean mean mean Species life form category 0.2x0.2 1x1 5x5 0.2x0.2 1x1 5x5 0.2x0.2 1x1 5x5 0.2x0.2 1x1 5x5 0.2x0.2 1x1 5x5 .2x.2 1x1 5x5 Fimbristylis caroliniana sedge nc 0 0 0 0 0 0 0 0 0 0 0.2 0.2 0 0 0 0 0.04 0.04 Fimbristylis puberula sedge nc 0 0 0 0 0 0 0 0 0 0 0.4 0.8 0 0.2 0.2 0 0.12 0.2 Fimbristylis tomentosa? sedge nc 0 0 0 0 0.6 0.6 0 0 0 0 0 0.6 0 0 0 0 0.12 0.24 Rhynchospora fascicularis sedge nc 0 0 0.4 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.08 Rhynchospora microcephala sedge nc 0.2 0.4 0.6 0 0 0 0 0 0 0 0 0 0 0 0 0 0.08 0.12 Rhynchospora rariflora sedge nc 0 0.2 0.2 0 0 0 0 0 0 0 0 0 0 0 0 0 0.04 0.04 Scleria retucularis sedge nc 0.2 0.6 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0.12 0.2 Scleria sp. sedge nc 0 0.2 0.2 0 0 0 0 0 0 0 0 0 0 0.2 0.2 0 0.08 0.08 Scleria triglomerata sedge nc 0 0.2 0.2 0 0 0 0 0 0 0 0 0 0 0 0 0 0.04 0.04 Xyris caroliniana sedge nc 0 0 0.4 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.08 Xyris sp. sedge nc 0 0 0.2 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.04 Circium nutallii Forb nw 0 0.2 0.4 0 0.2 0.4 0 0 0 0 0 0 0 0 0 0 0.08 0.16 Commelina diffusa Forb nw 0 0 0 0 0 0.2 0 0 0.4 0 0 0 0 0 0.2 0 0 0.16 Erechtites hieracifolia forb nw 0 0 0.6 0 0 0.2 0 0.2 0.4 0 0 0 0 0 0 0 0.04 0.24 Eupatorium capillifolium forb nw 0.2 0.4 1 0 0.4 0.8 0.2 0.8 1 0 0.2 1 0.2 0.8 1 0.1 0.52 0.96 Ludwigia octavalis Forb nw 0.4 0.8 1 0.2 0.4 0.8 0 0 1 0.2 0.6 0.8 0 0.2 0.4 0.2 0.4 0.8 Oxalis corniculata Forb nw 0 0.2 0.2 0 0 0 0 0 0.2 0 0.2 0.4 0 0.4 0.8 0 0.16 0.32 Portulaca amilis Forb nw 0 0 0 0 0 0 0 0 0 0 0 0.2 0 0 0.2 0 0 0.08 Scoparia dulcis Forb nw 0.4 0.8 0.8 1 1 1 0.6 0.8 1 0.2 0.8 1 0.4 1 1 0.5 0.88 0.96 Sesbannia herbacea Forb nw 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Digitaria serotina grass nw 0.4 0.6 0.8 0 0.4 0.8 0.2 0.6 0.8 0 0 0.6 0.4 0.6 1 0.2 0.44 0.8 Cyperus globulosus sedge nw 0 0 0.4 0 0.2 0.4 0.2 0.4 0.8 0 0.4 0.4 0 0.2 0.6 0 0.24 0.52

PAGE 91

79 Table A-3. Continued Elevation Strata 1 1 1 2 2 2 3 3 3 4 4 4 5 5 5 mean mean mean Species life form category 0.2x0.2 1x1 5x5 0.2x0.2 1x1 5x5 0.2x0.2 1x1 5x5 0.2x0.2 1x1 5x5 0.2x0.2 1x1 5x5 .2x.2 1x1 5x5 Cyperus polystachos sedge nw 0.2 0.6 0.8 0.8 1 1 0.6 1 1 1 1 1 0.8 1 1 0.7 0.92 0.96 Cyperus retrorsus sedge nw 0 0.2 0.6 0 0.4 0.8 0 0.6 0.8 0.2 0.6 0.8 0.2 0.8 0.8 0.1 0.52 0.76 Fimbristylis autumnalis sedge nw 0 0 0.2 0 0 0 0 0 0 0 0 0 0 0 0.2 0 0 0.08 Fimbristylis dichotoma sedge nw 0.2 0.6 1 0.8 1 1 0.4 0.8 0.8 0.2 0.8 0.8 0.4 0.8 1 0.4 0.8 0.92 Kyllinga brevifolia sedge nw 0 0.2 0.2 0 0.4 0.4 0.2 0.4 0.4 0.2 0.8 0.8 0 0 0.2 0.1 0.36 0.4 Cuphea carthogenesis Forb nnw 0 0 0 0 0 0 0 0 0 0 0 0.4 0 0 0 0 0 0.08 Desmodium triflorum Forb nnw 0.2 0.6 0.8 0.2 0.8 1 0.6 0.8 1 0 0 0 0.4 0.6 0.8 0.3 0.56 0.72 Hedyotis corymbosa Forb nnw 0 0 0 0.2 0.2 0.2 0 0.2 0.2 0 0.2 0.6 0 0 0.4 0 0.12 0.28 Indigofera hirsuta Forb nnw 0 0 0 0 0 0 0 0 0 0 0 0 0 0.2 0.2 0 0.04 0.04 Kummerowia striata Forb nnw 0.4 0.8 1 0.2 1 1 0.2 0.8 1 0.4 0.8 1 0 0.4 0.6 0.2 0.76 0.92 Macroptilium lathroides Forb nnw 0 0 0 0 0 0.2 0 0 0.4 0 0 0.2 0 0 0 0 0 0.16 Murdannia nudiflora Forb nnw 0 0.2 0.6 0.2 0.4 0.8 0.4 1 1 0 0.4 0.6 0.4 0.8 0.8 0.2 0.56 0.76 Solanum viarum Forb nnw 0 0 0.2 0 0 0.2 0 0 0.2 0 0 0.4 0 0 0.2 0 0 0.24 Eleusine indica grass nnw 0 0.2 0.2 0 0 0.2 0 0 0.2 0 0 0.2 0 0.2 0.2 0 0.08 0.2 Paspalum urvellii grass nnw 0.2 0.2 0.6 0 0 0 0 0 0.2 0 0 0 0 0 0 0 0.04 0.16 Sacciolepis indica grass nnw 0 0.6 1 0 0 0.2 0 0 0.2 0 0 0 0 0 0 0 0.12 0.28 Fimbristylis shoenoides sedge nnw 0.8 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0.2 0.2 0.2 Cynodon dactylon grass nni 0 0 0 0 0 0 0 0 0.2 0 0 0 0 0 0 0 0 0.04 Panicum repens grass nni 0 0 0 0 0 0 0 0 0 0 0 0.2 0 0 0.2 0 0 0.08 Paspalum notatum grass nni 0 0.2 0.8 0 0.2 0.6 0.2 0.2 0.8 0 0.2 0.8 0.2 0.4 1 0.1 0.24 0.8 Bulbostylis sp. sedge 0 0 0.2 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.04 Life form and category included for each species where possible (native characteristic = nc, native weedy = nw, non-native weedy = nnw, non-native invasive = nni (D. Gordon, unpublished data)). Native species are listed first.

PAGE 92

80 Table A-4. Species observed in fifteen healthy pine flatwoods surveyed at DWP in 1997 Acalypha graciliens Elephantopus elatus Pinus palustris Agalinis sp. Elephantopus sp. Pityopsis graminifolia Amphicarpum muhlenbergianum Erigeron vernus Pluchea rosea Andropogon brachystachys Eupatorium morhii Pluchea sp. Andropogon glomeratus glaucopsis Eupatorium rotundifolium Polygala rugelii Andropogon glomeratus glomeratus Euphorbia polyphylla Polygala setacea Andropogon glomeratus hirsutior Euthamia minor Pterocaulon virgatum Andropogon gyrans gyrans Fimbristylis puberula Quercus chapmanii Andropogon ternarius Galactia elliottii Quercus minima Andropogon virginicus Galactia sp. Quercus myrtifolia Andropogon virginicus brachystachys Gaylussacia dumosa Quercus virginianum Andropogon virginicus glaucus Gaylussacia frondosa nana Rhexia mariana Andropogon virginicus virginicus Gratiola hispida Rhexia nutallii Aristida spiciformis Gymnopogon chapmanianus Rhus copallina Aristida stricta Hedyotis procumbens Rhynchospora fernaldii Asclepias pedicellata Hedyotis uniflora Rhynchospora cephalantha Asimina reticulata Hypericum cistifolium Rhynchospora fascicularis Aster tortifolius Hypericum fasciculatum Rhynchospora fernaldii Axonopus furcatus Hypericum hypericoides Rhynchospora filiform Balduina angustifolia Hypericum myrtifolium/tetrapetalum Rhynchospora intermedia Befaria racemosa Hypericum reductum Rhynchospora microcephala Bigelowia nudata/Balduina uniflora Hypericum tetrapetalum Rhynchospora sp. Bulbostylis stenophylla Hypoxis juncea Rubus trivialis Carphephorus carnosus Ilex glabra Sabal palmetto Carphephorus corymbosus Juncus dichotomus Schizacharium scoparium Carphephorus paniculatus Lachnanthes caroliniana Scleria pauciflora Chamaecrista nictitans Lachnocaulon beyrichiana Scleria sp. Chamaecrista sp. Lachnocaulon anceps Scleria triglomerata Chapmannia floridana Lechea torreyi Sesbania vesicaria Chaptalia tomentosa Liatris laevigata Setaria geniculata Circium horridulum Liatris tenuifolia Solidago fistulosa Cnidoscolus stimulosus Licania michauxii Solidago stricta Composite unknown Ludwigia maritima Sorghastrum secundum Crotolaria rotundifolia Ludwigia not peruviana Styllingia sylvatica Ctenium aromaticum Ludwigia sp Syngonanthus flavidulus

PAGE 93

81 Table A-4. Continued Cuthbertia striped leaves Lygodesmia aphylla Tephrosia sp. Cyperus globulosus Lyonia fruiticosa Utricularia sp. Cyperus nashii Lyonia lucida Vaccinium myrsinites Cyperus polystachyos Myrica cerifera Vitus rotundifolia Cyperus retrorsus Opuntia sp. Xyris brevifolia Desmodium incanum Panicum anceps Xyris caroliniana Dichanthelium ensifolium Panicum longifolium Xyris difformis curtissii Dichanthelium ensifolium var. ensifolium Paspalum laeve Xyris elliottii Dichanthelium portoricense Paspalum setaceum Dichanthelium strigosum Phoebanthus grandiflora Eleocharis filiform Physalis sp. Eleocharis viviparous Pinus elliottii Reprinted with permission from The Nature Conservancy 1997. Fourth Annual Monitoring Report. The Nature Conservancy, Kissimmee, Florida. Table 4, Page 31.

PAGE 94

LIST OF REFERENCES Abrahamson, W. G., and D. C. Hartnett. 1990. Pine flatwoods and dry prairies. Pages 103-149 in R. L. Myers and J. J. Ewel, editors. Ecosystems of Florida. University of Central Florida Press, Orlando, Florida. Allen, E. B. 1989. The restoration of disturbed arid landscapes with special reference to mycorrhizal fungi. Journal of Arid Environment 17:279-286. Analytical Research Laboratory. 2003. Analytical Research Laboratory, University of Florida, Institute of Food and Agriculture Sciences URL: http://arl.ifas.ufl.edu/, January 2003. Anderson, R. C., and E.S. Menges. 1997. Effects of fire on sandhill herbs: nutrients, mycorrhizae, and biomass allocation. American Journal of Botany 84:938-948. Baker, H. G. 1989. Some aspects of the natural history of seed banks. In M. A. Leck, V. T. Parker, and R. L. Simpson, editors. Ecology of soil seed banks. Academic Press, Inc., San Diego, California, USA. Bekker, R. M., G. L. Verweij, R. E. N. Smith, R. Reine, J. P. Bakker, and S. Schneider. 1997. Soil seed banks in European grasslands: does land use affect regeneration perspectives? Journal of Applied Ecology 34:1293-1310. Bever, J. D., A. P. Schultz, A. Pringle, and J. P. Morton. 2001. Arbuscular mycorrhizal fungi: More diverse than meets the eye, and the ecological tale of why. BioScience 51:923-931. Bissett, N. J. 1996. Upland restoration challenge: Direct seeding of wiregrass and associated species. The Palmetto Summer:8-11. Boddington, C. L., and J. C. Dodd. 2000. The effect of agricultural practices on the development of indigenous arbuscular mycorrhizal fungi. I. Field studies in an Indonesian ultisol. Plant and Soil 218:137-144. Brady, N. C., and R. R. Weil. 2002. The Nature and Properties of Soils, Thirteenth edition. Prentice Hall, Upper Saddle River, New Jersey. Bray, S. R., K. Kitajima, and D. M. Sylvia. 2003. Mycorrhizae differentially alter growth, physiology, and competitive ability of an invasive shrub. Ecological Applications 13(3):565-574. 82

PAGE 95

83 Brown, C. S., and R.L. Bugg 2001. Effects of established perennial grasses on introduction of native forbs in California. Restoration Ecology 9:38-48. Brown, D. 1992. Estimating the composition of a forest seed bank: a comparison of the seed extraction and seedling emergence methods. Canadian Journal of Botany 70:1603-1611. Buchanan, K. S. 1999. Restoration of sandhill vegetation on abandoned agricultural lands. Master of Science. University of Central Florida, Orlando, Florida. Buhler, D. D., and B. D. Maxwell. 1993. Seed separation and enumeration from soil using K2CO3 centrifugation and image analysis. Weed Science 41:298-302. Butler, B. J., and R. L. Chazdon. 1998. Species richness, spatial variation, and abundance of the soil seed bank of a secondary tropical rain forest. Biotropica 30:214-222. Carrington, M. E. 1997. Soil seed bank structure and composition in Florida sand pine scrub. The American Midland Naturalist 137:39-47. Chazdon, R. L., R. K. Colwell, J. S. Denslow, and M. R. Guariguata. 1998. Statistical methods for estimating species richness of woody regeneration in primary and secondary rain forests of northeastern Costa Rica. Pages 285-309 In F. Dallmeier and J. A. Comiskey, editors. Forest Biodiversity Research, Monitoring and Modeling: Conceptual Background and Old World Case Studies. Parthenon Publishing, Paris, France. Clewell, A. F. 1989. Natural history of wiregrass (Aristida stricta Michx., Gramineae). Natural Areas Journal 9:223-233. Colwell, R. K. 1997. EstimateS: Statistical estimation of species richness and shared species from samples. Version 6. User's Guide and application published at: http://viceroy.eeb.uconn.edu/estimates. Corbett, E. A., R.C. Anderson, and C.S. Rodgers. 1996. Prairie revegetation of a strip mine in Illinois: fifteen years after establishment. Restoration Ecology 4:346-354. Cottam, G. 1987. Community dynamics on an artificial prairie. Pages 257-270 In W. R. Jordan, M. E. Gilpin, and J. D. Aber, editors. Restoration Ecology: A Synthetic Approach to Ecological Research. Cambridge University Press, New York, New York. Cox, A. C., D. R. Gordon, G. S. Seamon, and J. L. Slapcinsky. 2003. Understory restoration in longleaf pine sandhills. Natural Areas Journal. In press.

PAGE 96

84 D'Antonio, C. D., and L. A. Meyerson. 2002. Exotic plant species as problems and solutions in ecological restoration:A synthesis. Restoration Ecology 10:703-713. Dhillion, S. S., and J. C. Zak. 1993. Microbial dynamics in arid ecosystems: desertification and the potential role of mycorrhizas. Revista Chilena De Historia Natural 66:253-270. Ebbers, B. C., R. C. Anderson, and A. E. Liberta. 1987. Aspects of the mycorrhizal ecology of prairie dropseed, Sporobolus heterolepis (Poaceae). American Journal of Botany 74:564-573. Eom, A., D.C. Hartnett, and G.W.T. Wilson. 2000. Host plant species effects on arbuscular mycorrhizal fungal communities in tallgrass prairie. Oecologia 122:435-444. Fisher, J. B., and K. Jayachandran. 1999. Root structure and arbuscular mycorrhizal colonization of the palm Serenoa repens under field conditions. Plant and Soil 217:229-241. Florida Exotic Pest Plant Council. 2003. List of Florida's Invasive Species. Florida Exotic Pest Plant Council. URL: http://www.fleppc.org/03list.htm, February 2003. Florida Forever. 2003. Florida Forever Act. Florida Department of Environmental Protection. URL: http://www.dep.state.fl.us/lands/carl_ff/, January 2003. Florida Natural Areas Inventory and Florida Department of Natural Resources. 1990. Guide to the natural communities of Florida. Florida Natural Areas Inventory and Department of Natural Resources, Tallahassee, FL. Godfrey, R. K., and J. M. Wooten. 1979. Aquatic and Wetland Plants of Southeastern United States: Monocotyledons. The University of Georgia Press, Athens, Georgia. Godfrey, R. K., and J. M. Wooten. 1981. Aquatic and Wetland Plantsof Southeastern United States: Dicotyledons. The University of Georgia Press, Athens, Georgia. Gordon, D. R., A. C. Cox, G. S. Seamon, J. L. Slapcinsky, and D. A. Jones. 2001. Native sandhill species revegetation techniques. Final Report to the Florida Department of Transportation Contract # BB 937, University of Florida and The Nature Conservancy, Gainesville, FL. Gordon, D. R., M. J. Hattenbach, G. S. Seamon, K. Freeman, and D. A. Jones. 2000. Establishment and Management of upland native plants on Florida roadsides. Final report to the Florida Department of Transportation Contract # BA 523, University of Florida and The Nature Conservancy, Gainesville, Florida.

PAGE 97

85 Gotelli, N.J. 2001. The Primer of Ecology, Third Edition. Sinauer Associates, Inc. Sunderland, MA. Gross, K. L. 1990. A comparison of methods for estimating seed numbers in soil. Journal of Ecology 78:1079-1093. Gross, K. L. and K. A. Renner. 1989. A new method for estimating seed numbers in the soil. Weed Science 37:836-839. Hardin, E. D., and D. L. White. 1989. Rare vascular plant taxa associated with wiregrass (Aristida stricta) in the southeastern United States. Natural Areas Journal 9:234-245. Harper-Lore, B. L. 1998. Do native grasslands have a place on roadsides? Land and Water July/August:29-31. Hart, M. M., R. J. Reader, and J. N. Klironomos. 2001. Life-history strategies of arbuscular mycorrhizal fungi in relation to their successional dynamics. Mycologia 93:1186-1194. Hartel, P. G. 1999. The soil habitat. In D. M. Sylvia, J. H. Fuhrmann, P. G. Hartel, and D. A. Zuberer, editors. Principles and applications of soil microbiology. Prentice Hall, Upper Saddle River, New Jersey, USA. Hartnett, D. C., and G. W. T. Wilson. 1999. Mycorrhizae influence plant community structure and diversity in tallgrass prairie. Ecology 80:1187-1195. Hattenbach, M. J., D. R. Gordon, G. S. Seamon, and R. G. Studenmund. 1998. Development of direct seeding techniques to restore native groundcover in a sandhill ecosystem. Pages 64-70 In Proceedings, Longleaf Pine Ecosystem Restoration Symposium, Auburn, AL. Heady, H. F. 1958. Vegetational changes in the California annual type. Ecology 39:402-416. International Culture Collection of Vesicular Arbuscular Mycorrhizal Fungi. 2003. International Culture Collection of (Vesicular) Arbuscular Mycorrhizal Fungi. West Virginia University. URL: http://invam.caf.wvu.edu/methods/assays /MIP.htm, March 2003. Jakobsen, I. 1992. Phosphorus transport by external hyphae of vesicular-arbuscular mycorrhizas. Pages 48-58 In D. J. Read, D. H. Lewis, A. H. Fitter, and I. J. Alexander, editors. Mycorrhizas in Ecosystems. CAB International, Wallingford, UK.

PAGE 98

86 Johnson, N. C., D. R. Zak, D. Tilman, and F. L. Pfleger. 1991. Dynamics of vesicular-arbuscular mycorrhizae during old field succession. Oecologia 86:349-358. Jordan, C. L. 1984. Florida's weather and climate: implications for water. Pages 18-35 In E. A. Fernald and D. J. Patton, editors. Water Resources Atlas of Florida. Florida State University, Tallahassee, Florida. Kabir, Z., I.P. O'Halloran, and C. Hamel. 1999. Combined effects of soil disturbance and fallowing on plant and fungal components of mycorrhizal corn (Zea mays L.). Soil Biology and Biochemistry 31:307-314. Kesler, T. R., L. C. Anderson, and S. H. Hermann. 2003. A taxonomic reevaluation of Aristida stricta (Poaceae) using anatomy and morphology. Southeastern Naturalist 2:1-10. Kitajima, K., and D. Tilman. 1996. Seed banks and seedling establishment on an experimental productivity gradient. Oikos 76:381-391. Kltzli, F, and A. P. Grootjans. 2001. Restoration of semi-natural wetland systems in central Europe: Progress and predictability of developments. Restoration Ecology 9(2):209-219. Landers, J. L., and A. S. Johnson. 1976. Bobwhite quail food habits. Miscellaneous Report # 4, Tall Timbers Research Station, Tallahassee, FL. Langeland, K. A., and K. C. Burks, Editors. 1998. Identification and Biology of Non-native Plants in Florida's Natural Areas. University of Florida, Gainesville, FL. Laughlin, D. C. 2003. Lack of native propagules in a Pennsylvania, USA, limestone prairie seed bank: Futile hopes for a role in ecological restoration. Natural Areas Journal 23:158-164. Leckie, S., M. Vellend, G. Bell, M. J. Waterway, and M. J. Lechowicz. 2000. The seed bank in an old-growth, temperate deciduous forest. Canadian Journal of Botany 78:181-192. Magurran, A. E. 1988. Ecological Diversity and its Measurements. Princeton University Press, Princeton, New Jersey. Maliakal, S. K., E. S. Menges, and J. S. Denslow. 2000. Community composition and regeneration of Lake Wales Ridge wiregrass flatwoods in relation to time since fire. Journal of Torrey Botanical Society 127:125-138. Malone, C. R. 1967. A rapid method for enumeration of viable seeds in soils. Weeds 15:381-382.

PAGE 99

87 Martin, A. C., and W. D. Barkley. 1961. Seed Identification Manual. University of California Press, Berkeley, California. McGonigle, T. P., and A. H. Fitter. 1990. Ecological specificity of vesicular arbuscular mycorrhizal associations. Mycological Research 94:120-122. McGonigle, T. P., M.H. Miller, D.G. Davis, G.L. Fairchild, and J.A. Swan. 1990. A new method which gives an objective measure of colonization of roots by vesicular-arbuscular mycorrhizal fungi. New Phytologist 115:495-501. Medina, O. A., A. E. Kretschmer Jr., and D. M. Sylvia. 1988. The occurence of vesicular-arbuscular mycorrhizal fungi on tropical forage legumes in South Florida. Tropical Grasslands 22:73-78. Miller, R. M. 1987. Mycorrhizae and succession. Pages 205-219 in C. L. Jordan, M. E. Gilpin, and J. D. Aber, editors. Restoration Ecology: A Synthetic Approach to Ecological Research. Cambridge University Press, New York, New York. Miller, R. M. 1997. Prairie Underground. Pages 23-27 in S. Packard and C. F. Mutel, editors. The Tallgrass Restoration Handbook for Prairies, Savannas, and Woodlands. Island Press, Washington, D.C. Moynahan, O. S., and C. A. Zabinski. 2002. Microbial community structure and carbon-utilization diversity in a mine tailings revegetation study. Restoration Ecology 10:77-87. Mullahey, J. J., and C. S. Speed. 1991. The occurence of vesicular-arbuscular mycorrhizae on Florida range grasses. Soil and Crop Science Society of Florida Proceedings 50:44-47. Mulligan, M. K., L. K. Kirkman, and R. J. Mitchell. 2002. Aristida beyrichiana (wiregrass) establishment and recruitment: Implications for restoration. Restoration Ecology 10:68-76. The Nature Conservancy. 1996. Upland Restoration Plan: Conceptual Plan and Assessmant of Pasture Restoration Sites. The Nature Conservancy, Kissimmee, Florida. The Nature Conservancy. 1997. Fourth Annual Monitoring Report for the Disney Wilderness Preserve, The Nature Conservancy. Kissimmee, FL. The Nature Conservancy. 1999. The Analysis of Data from the First Two Years of Post-treatment Monitoring: Disney Wilderness Preserve Upland Restoration Program. Subcontract Agreement # NRA63, The Nature Conservancy, Disney Wilderness Preserve, Kissimmee, FL.

PAGE 100

88 The Nature Conservancy. 2000. Proceedings of the Upland Restoration Workshop. The Nature Conservancy. Kissimmee, FL. Noss, R. F. 1989. Longleaf pine and wiregrass: Keystone components of an endangered ecosystem. Natural Areas Journal 9:211-213. Oosting, H. L., and M. E. Humphreys. 1940. Buried viable seeds in a successional series of old field and forest soils. Bulletin of the Torrey Botanical Club 67:253-273. Packard, S., and C. F. Mutel. 1997. The Tallgrass Restoration Handbook: For Prairies, Savannas, and Woodlands. Island Press, 1997. Peng, S., D. M. Eissenstat, J. H. Graham, K. Williams, and N. C. Hodge. 1993. Growth depression in mycorrhizal citrus at high-phosphorus supply. Plant Physiology 101:1063-1071. Platt, W. J. 1988. Fire ecology of the southeastern longleaf pine flatwoods. Pg 5 in L. C. Duever and R. F. Noss, editors. Wiregrass Biology and Management: Maintaining Groundcover Integrity in Longleaf: Proceedings of the Symposium on. Valdosta State College, Gainesville, FL. Putwain, P. D., and D. A. Gillham. 1990. The significance of the dormant viable seed bank in the restoration of heathlands. Biological Conservation 52:1-16. Pyne, S. J., P.L. Andrews, and R.D. Laven. 1996. Introduction to Wildland Fire. John Wiley and Sons, Inc., New York. Rabinowitz, D. 1981. Buried viable seeds in a North American tallgrass prairie: the resemblance of their abundance and composition to dispersing seeds. Oikos 36:191-195. Read, D. J., H. K. Koucheki, and J. Hodgson. 1976. Vesicular-arbuscular mycorrhiza in natural vegetation systems: 1. The occurence of infection. New Phytologist 77:641-653. Richter, B. S., and J. C. Stutz. 2002. Mycorrhizal inoculation of big sacaton: Implications for grassland restoration of abandoned agricultural fields. Restoration Ecology 10:607-616. Roberts, H. A. 1981. Seed banks in soils. In T. H. Coaker, editor. Advances in Applied Biology. Academic Press. SAS. 1999. The SAS System for Windows. SAS Institute, Cary, NC. Schenck, N. C., and R. A. Kinloch. 1980. Incidence of mycorrhizal fungi on six field crops in monoculture on a newly cleared woodland site. Mycologia 72:445-456.

PAGE 101

89 Schott, G. W., and S. P. Hamburg. 1997. The seed rain and seed bank of an adjacent native tallgrass prairie and old field. Canadian Journal of Botany 75:1-7. Schwandes, L. P., M. Chen, and J. Galbraith. 2001. Total and extractable soil phosphorus in six ecological communities of Florida. Soil and Crop Science Society of Florida Proceedings 60:53-56. Seamon, P. A., and R. L. Myers. 1992. Propagating wiregrass from seed. The Palmetto Winter:6-7. Siguenza, C., I. Espejel, and E. B. Allen. 1996. Seasonality of mycorrhizae in coastal sand dunes of Baja California. Mycorrhiza 6:151-157. Smith, M. R., I. Charvat, and R. L. Jacobson. 1998. Arbuscular mycorrhizae promote establishment of prairie species in a tallgrass prairie restoration. Canadian Journal of Botany 76:1947-1954. Smith, S. E., and D. Read. 1997. Mycorrhizal Symbiosis. Academic Press, London. Soil Conservation Service. 1990. Soil survey of Polk County, Florida. U.S. Department of Agriculture. Soil Conservation Service. Sylvia, D. M. 1986. Spatial and temporal distribution of vesicular-arbuscular mycorrhizal fungi associated with Uniola paniculata L. in Florida foredunes. Mycologia 67:734-740. Sylvia, D. M. 1994. Vesicular-arbuscular mycorrhizal fungi. Pages 351-378 in R. W. Weaver, S. Angle, P. Bottomley, D. Bezdicek, S. Smith, A. Tabatabai, and A. Wollum (eds.), editor. Methods of soil analysis, Part 2. Microbiological and biochemical properties. Soil Science Society of America, Madison, WI. Sylvia, D. M. 1999. Mycorrhizal symbioses. Pages 408-446 in D. M. Sylvia, J. H. Fuhrmann, P. G. Hartel, and D. A. Zuberer, editors. Principles and applications of soil microbiology. Prentice Hall, Upper Saddle River, New Jersey, USA. Sylvia, D. M., A. Alagely, D. O. Chellemi, and L. W. Demchenko. 2001. Arbuscular mycorrhizal fungi influence tomatoes competition with bahiagrass. Biology and Fertility of Soils 34:448-452. Sylvia, D. M., and J. N. Burks. 1988. Selection of a vesicular-arbuscular mycorrhizal fungus for a practical inoculation of Uniola paniculata. Mycologia 80:565-568. Thomas, G. W. 1996. Soil pH and soil acidity. Pages 475-490 in D. L. Sparks, editor. Methods of Soil Analysis: Chemical Methods. Part 3. Soil Science Society of America, Inc. and American Society of Agronomy, Inc., Madison, Wisconsin.

PAGE 102

90 Thompson, K. 2000 The functional ecology of soil seed banks. Pages 215-235 in M. Fenner, editor. Seeds: The Ecology of Regeneration in Plant Communities. 2nd Edition. CABI Publishing, Wallingford, UK. Underwood, A. J., and M. G. Chapman. 1998. A method for analyzing spatial scales of variation in composition of assemblages. Oecologia 117:570-578. van der Heidjen, M. G. A., J. N. Klironomos, M. Ursic, P. Moutoglis, R. Streitwolf-Engel, T. Boller, A. Wiemken, and I. R. Sanders. 1998. Mychorrhizal fungal diversity determines plant biodiversity, ecosystem variability and productivity. Nature 396:69-72. van der Valk, A. G., and R. L. Pederson. 1989. Seed banks and the management and restoration of natural vegetation. Pages 329-346 in M. A. Leck, V. T. Parker, and R. L. Simpson, editors. Ecology of Soil Seed Banks. Academic Press, San Diego. Violi, H. 1999. Evaluation of restoration techniques in an abandoned bahiagrass pasture. Archbold Biological Station, Lake Placid, FL. Wunderlin, R. P. 1998. Guide to the Vascular Plants of Florida. University Press of Florida, Gainesville, FL.

PAGE 103

BIOGRAPHICAL SKETCH Amy Miller Jenkins was born in Fort Lauderdale, Florida. She graduated from the Florida State University in 1995 with a B.S. degree in biology. She then set out on a series of ecological field jobs and lots of adventuring around the United States, which led her to California, Georgia, Arizona; and then back to her homeland in Florida, where she has been ever since. During this period she found her love for native flora, especially the longleaf pine/wiregrass community of the southeast; and the idea of land restoration. In 1998, she had the great opportunity to work with Dr. Doria Gordon on a research project in south and central Florida investigating native grasses for use on roadsides in Florida. This job gave her great experiences that pushed her intellectually. It also introduced her to the many projects and people in Florida restoring pine flatwoods vegetation in pastures; and to the opportunity to pursue graduate research on pasture-restoration projects underway at The Nature Conservancys Disney Wilderness Preserve. In August 2001, Amy entered into a masters program in the University of Florida, College of Natural Resources and Environment advised by Dr. Gordon. She plans to pursue a career in restoration and land management. 91


Permanent Link: http://ufdc.ufl.edu/UFE0001399/00001

Material Information

Title: Seed Banking and Vesicular-Arbuscular Mycorrhizae in Pasture Restoration in Central Florida
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0001399:00001

Permanent Link: http://ufdc.ufl.edu/UFE0001399/00001

Material Information

Title: Seed Banking and Vesicular-Arbuscular Mycorrhizae in Pasture Restoration in Central Florida
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0001399:00001


This item has the following downloads:


Full Text












SEED BANKING AND VESICULAR-ARBUSCULAR MYCORRHIZAE IN
PASTURE RESTORATION IN CENTRAL FLORIDA















By

AMY MILLER JENKINS


A THESIS PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
MASTER OF SCIENCE

UNIVERSITY OF FLORIDA


2003

































Copyright 2003

by

Amy Miller Jenkins

































To my husband, Michael, who is always an inspiration!















ACKNOWLEDGMENTS

This study would not have been possible without the help of many friends and

advisors. I would like to acknowledge my two committee members, Dr. Doria Gordon

and Dr. Kaoru Kitajima for their guidance, wisdom, and inspiration. I would especially

like to thank Dr. Doria Gordon for being a wonderful mentor over the past five years. I

also thank Dr. Kaoru Kitajima for her valuable insight and use of her laboratory space

and equipment for all of my lab work.

I thank the graduate students and staff of the School of Natural Resources and

Environment (SNRE) and the Botany Department as well as Dr. Humphrey, the Dean of

SNRE for his insightfulness and advice. SNRE provided a research assistantship which

made my graduate program possible. The Nature Conservancy (TNC) and the staff at the

Disney Wilderness Preserve and Gainesville Office have made this research possible

through generous use of their land and facilities for my research purposes. Dr. David

Sylvia and Dr. James Graham provided invaluable help with the mycorrhizae chapter in

experimental design and results interpretation. Abid Alagely and Sarah Bray provided

technical assistance in mycorrhizae structural identification and methods. Nancy Bissett

(The Natives Inc.) and Michael Byrne (TNC) provided tremendous help with plant

identification. I thank Richard Abbott and Kent Perkins for help with seed identification

and use of the UF herbarium facilities. Dr. Doug Levey gave advice on the direct count

seed bank methods. The University of Florida School of Forest Resources provided

shadehouse space that made the seed bank germination trials possible. Sarah Bray,









Catherine Cardelus, and Michael Jenkins provided helpful editorial comments on this

thesis. I thank Clea Paz and Sarah Bray for their support and friendship throughout this

graduate experience.

I would particularly like to thank my parents for their love, support, and

encouragement to pursue graduate school. Mostly I thank Michael, my husband, for his

endless love, support, and assistance both in my coursework and with my research. He

happily provided countless hours of fieldwork, laboratory assistance, editing,

brainstorming, shadehouse construction, coffee making, and much more!
















TABLE OF CONTENTS
Page

A C K N O W L E D G M E N T S ................................................................................................. iv

LIST OF TABLES ................................................................ ........ viii

LIST OF FIGURES ......... ......................... ...... ........ ............ ix

ABSTRACT .............. ................. .......... .............. xi

CHAPTER

1 GENERAL INTRODUCTION AND SITE DESCRIPTION .....................................1

2 SEED BANKING IN A PASTURE RESTORATION SITE AND MESIC PINE
FLATW OODS IN CENTRAL FLORIDA........................................ .....................9

Introduction ................. ......... ......... ...... ............. ............. 9
M eth o d s .............................................................................. 13
G erm nation A ssay ...................... ...... ............ ........................... 14
D direct Count A ssay ......... ... ................ ....... .. ............. ......... .. 17
V vegetation Survey .................................. .. .. ...... ............. 18
Data Analysis ........................................................................... .. ......................... 18
R e su lts ........... ........................................................................... 2 0
G erm nation A ssay ...................... ...... .......... ................. .... ..... .. 20
Direct Count Assay .......... ........ ....... .. ................ .... ......22
V vegetation Survey .................................. .. .. ...... ............ 24
A ristida strict R ecruitm ent ........................................ .......................... 25
M ethods C om parison................................................. .............................. 25
Discussion ............... ............. ......... .................... 26
M ethods C om prison ......................................... ............................ ...........30
M anagem ent Im plications ...........................................................................32

3 VESICULAR-ARBUSCULAR MYCORRHIZAE IN UPLAND RESTORATION
SITES, MESIC PINE FLATWOODS, AND PASPALUMNOTATUMPASTURES
IN CEN TRAL FLORID A ............................................... ............................... 40

Introdu action ...................................... ................................................. 4 0
M eth o d s ................................................. ...... .... .......... ................ 4 5
Experiment 1: First-year and Pre-restoration Complexes ...................................46
Experim ent 2: Five-year Com plex ........................................... ............... 49









D ata A n a ly sis ................................................................................................. 5 1
E x p erim ent 1 ............................................................. 5 1
E xperim ent 2 ...................... .................... ........... ..... .... 52
R e su lts ...........................................................................................5 3
E x p e rim e n t 1 ................................................................................................. 5 3
E x p e rim e n t 2 ................................................................................................. 5 4
D isc u ssio n ............................................................................................................. 5 5
E x p e rim e n t 1 ................................................................................................. 5 5
E x p e rim e n t 2 ................................................................................................. 5 8
Conclusions/Applications ...................................................... 61

A PPEN D IX SPE C IE S L ISTS ..................................................................................... 72

L IST O F R E FE R E N C E S ............................................................................... 82

B IO G R A PH IC A L SK E T C H ....................................................................................... 91
















LIST OF TABLES


Table p

2-1 Species in common between the restoration and flatwoods soils and unique
species in the flatwoods and restoration soils in the germination assay ..................34

3-1 Composition of the modified Hoaglands fertilizer solution...................................63

3-2 Spearman correlation matrix for MIP and soil chemistry in the pre-restoration
co m p lex .......................................................... ................ 6 3

3-3 Spearman correlation matrix for MIP and soil chemistry variables in the one-year
com p lex .......................................................... ................ 64

3-4 Results of two-way nested ANOVA for each soil chemistry variable on
restoration treatment, elevation and their interaction within the one-year
co m p lex .. ................................................................................. 6 4

3-5 Results of Tukey mean comparisons of the three treatments in Experiment 1 for
each significant variable........................................................................... .... ..... 65

3-6 Results of the Tukey mean comparisons of the five elevation strata in
Experim ent 1 for significant variables ........................................ .....................65

3-7 P-values from the Spearman's correlation matrix of eight soil variables in
E xperim ent 2 ...................................................... ................. 65

A-i Species list for the germination assay with life form and category..........................73

A-2 Species list for the direct count assay with life form and category..........................75

A-3 Species list for the field vegetation sampling by elevation strata (n=5 strata) and
frequency by plot size. ..................... ................ .............................76

A-4 Species observed in fifteen healthy pine flatwoods surveyed at DWP in 1997.......80















LIST OF FIGURES


Figure pge

1-1 Location of the The Nature Conservancy's Disney Wilderness Preserve within
the central Florida region. ........................................ ................................. 8

2-1 The mean number of seedlings per sample (+1 SE) in the germination assay.........35

2-2 Mean species accumulation curves of observed species richness for the
restoration and flatw oods soils......................................... ............................ 35

2-3 Relative abundance of species categories at the 5 elevation strata ........................36

2-4 Mean (+1 SE) density ofAristida strict (m-2, n=5) for each of the 5 elevation
strata six months post-seeding in the first-year restoration site. ...........................37

2-5 Comparison of the mean species accumulation curves of observed species
richness for the restoration soils........................................ ........................... 38

2-6 Number of seeds/seedlings per m2. ............................... .................................. 39

3-1 Correlation between MIP (n = 5) and pH (n = 3) for soils from flatwoods and
one-year unit restoration soils in June 2002 .........................................................66

3-2 Mean mycorrhizal inoculation potential (+1 SE, n = 25) of the one-year unit,
and adjacent flatwoods and pasture.................................................. ............... 66

3-3 Mean inoculation potential (+1 SE, n = 15) of the 5 elevation strata in June
2 0 02 .................................................................................6 7

3-4 Mean gravimetric soil moisture content (+1 SE, n = 5) for each treatment by
elevation strata in the one-year complex ..................................... ....... ........... 67

3-5 Mean pH (+1 SE, n = 3) for each treatment by elevation in the one-year
co m p lex .......................................................... ................ 6 8

3-6 Mean soil P content (mg/kg) (+1 SE, n=3) for each treatment in the one-year
co m p lex .......................................................... ................ 6 8

3-7 Mean soil total Kjeldahl nitrogen (g/kg) (+1 SE, n=3) for each treatment in the
one-year com plex. ....................... ...................... ......................... 69









3-8 Percent colonization (n = 1) ofAristida strict roots by VAM fungi in the six
restoration sites and adjacent flatwoods in the Experiment 2...............................69

3-9 Mean inoculation potential (+1 SE, n = 30) of the soils in the five treatments of
the fiv e-y ear com plex ...................................................................... ...................70

3-10 Mean soil total Kjeldahl nitrogen (+1 SE, n = 18) of the restoration and adjacent
flatwoods and pasture soils in Experim ent 2................................. ............... 70

3-11 Mean soil gravimetric moisture content (+1 SE, n = 30) of the restoration and
adjacent flatwoods and pasture soils in Experiment 2 ...........................................71

3-12 Mean soil extractable P content (+1 SE, n = 18) of the restoration and adjacent
flatwoods and pasture soils in Experim ent 2................................. ............... 71















Abstract of Thesis Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Master of Science

SEED BANKING AND VESICULAR-ARBUSCULAR MYCORRHIZAE IN
PASTURE RESTORATION IN CENTRAL FLORIDA

By

Amy Miller Jenkins

December 2003

Chair: Doria R. Gordon
Major Department: Natural Resources and Environment

Increasing efforts to restore longleaf pine systems that have been converted to

non-native pasture grasses are underway in Florida. Native vegetation does not recover

quickly in abandoned pastures unless restoration efforts are introduced. The goal of this

study was to examine the contribution of relict seed banks and vesicular-arbuscular

mycorrhizae (VAM) fungi in pasture restoration sites in central Florida. I compared the

seed bank in a restoration site and an adjacent flatwoods and evaluated the mycorrhizal

inoculation potential (MIP) of three restoration areas in different stages of progress.

In Chapter 2, I investigated the species composition of the soil seed bank in a

pasture restoration area and adjacent pine flatwoods. In this study, I examined the seed

bank with germination and direct count assays and evaluated the aboveground vegetative

species composition of the restoration area 6 months post-sowing. The restoration and

flatwoods seed banks differed in species composition. The restoration soils had more

non-native and weedy native species, while the flatwoods had more native species









characteristic to pine flatwoods. The flatwoods soils had more unique species than the

restoration soils. Although several perennial grasses were recorded, none of the perennial

grasses most characteristic of flatwoods (e.g., Aristida strict) were found in the seed

bank with my methods. The restoration area seed bank is rich with native species.

However, the seedbank will not regenerate a natural flatwoods community because most

of the characteristic flatwoods species are absent. Seed ofA. strict and other perennial

grasses must be reintroduced and weedy species controlled in restoration projects.

In Chapter 3, I examined the relationship of VAM fungi to soil chemistry,

restoration efforts, and temporal variation. Mycorrhizal inoculation potential was

measured in pre-restoration, first-year restoration and five-year post restoration sites

coupled with their adjacent pasture and natural flatwoods. Mycorrhizal inoculation

potential of the restoration area before restoration seeding with a native seed mix was low

(-3%) and similar to that found in pasture. There was a significant increase in MIP in the

restoration area just 6 months post seeding not seen in the pasture or flatwoods. The

temporal variation in MIP followed the phenology of the aboveground vegetation. The

MIP in natural flatwoods was low in all tests (-2%). I also compared the VAM fungi

colonization of A. strict roots in 5-year post restoration sites with plants in adjacent

flatwoods and found no significant difference. The MIP data coupled with the A. strict

results, suggest that inoculum is not lacking in restoration areas in central Florida; and

that VAM fungi appear to infect native plants in restoration areas at a rate similar to that

found in pine flatwoods. Overall, these results suggest that seed, but not mycorrhizae,

will need to be augmented in pasture restoration efforts.














CHAPTER 1
GENERAL INTRODUCTION AND SITE DESCRIPTION

The natural environment of Florida has been dramatically changed through

development pressures, habitat fragmentation, and agricultural practices. Pastures planted

with Paspalum notatum Fligge (bahia grass), a non-native species, are widespread across

Florida and the Southeast for cattle forage and sod production. Restoration of degraded

upland landscapes such as abandoned pastures and agricultural fields may be an

important mode for reintroducing many diminishing community types with increasing

development pressures and habitat fragmentation in Florida. Native vegetation does not

recover quickly in abandoned pastures dominated by P. notatum unless restoration efforts

reintroduce native species (Violi 1999). Many land managers in Florida are restoring

pastures to native upland habitat for various other reasons, including upland mitigation

and phosphate-mine reclamation (The Nature Conservancy 2000). Additionally, with the

continuing programs in the State of Florida to purchase conservation lands through bond

issues (Florida Forever 2003), more former agricultural lands and semi-degraded

properties that may have associated restorable pastures are being purchased. Many upland

restoration projects are underway in Florida today. State agencies such as St. Johns River

Water Management District are restoring old agricultural fields; phosphate companies

such as CF Industries Inc. (Plant City, FL) and IMC Agrico Inc. (Fort Green, FL) are

incorporating restoration of pine flatwoods into their mine-reclamation process; and The

Nature Conservancy is restoring both sandhill in old windrowed pine plantations and pine

flatwoods in pastures (The Nature Conservancy 2000).









Are there certain ecological features of these pastures and restoration sites that, if

better understood, would aid in restoration success? In this project, I investigated the

contribution of seed banks and mycorrhizae to restoration of improved pastures at The

Nature Conservancy's Disney Wilderness Preserve (DWP) in Kissimmee, Florida.

The Disney Wilderness Preserve is a 4,797-ha preserve owned and managed by

The Nature Conservancy. The preserve is located just south of Orlando, Florida on the

north side of Lake Hatchineha, at the base of the Reedy Creek/Lake Marion Creek

watershed in Osceola and Polk Counties in central Florida (Figure 1-1). Many native

terrestrial plant communities exist at the preserve. Mesic pine flatwoods is the most

common among a mosaic of community types such as scrub, upland hardwood forest, dry

prairie, wet prairie, baygall, cypress swamps, and depressional marsh, (FNAI and FDNR

1990). The average for the maximum daily temperature for this region is 31 C in

summer and 18 C in winter. Freezing temperatures do occur but are infrequent and of

short duration (Jordan 1984). This region of Florida has pronounced wet and dry seasons

and long-term periods of flooding and drought with an average annual rainfall of 1270

mm (Jordan 1984). Soil types at our study site were Myakka and Smyrna series

belonging to the order spodosols, which is common in pine flatwoods and characterized

by a spodic horizon and sandy poorly drained conditions (Soil Conservation Service

1990, Brady and Weil 2002). These soils in Central Florida have low nutrients and

organic matter and variable pH (Abrahamson and Hartnett 1990).

The Disney Wilderness Preserve, previously a cattle ranch, has approximately 486

ha of improved and unimproved pastures scattered throughout the property in a matrix of

natural flatwoods and wetlands. These pastures can alter ecosystem function by impeding









movement of natural processes such as naturally occurring fire (The Nature Conservancy

1996). Additionally, the pastures are composed of non-native pasture grasses such as

Paspalum notatum, Cynodon dactylon (Bermuda grass) Panicum repens (torpedo grass)

and Digitariapentzii (pangolagrass), and also non-native invasive species such as

Solanum viarum (tropical soda apple) (Langeland and Burks 1998) and Imperata

cylindrica (cogongrass). At DWP, old aerial photographs show that pine flatwoods

occupied the sites before they were converted to pasture (The Nature Conservancy 1996).

For these reasons, The Nature Conservancy sought to restore the pastures on DWP to a

plant community assemblage that more closely resembled the pine flatwoods community.

The goal of the restoration is to restore the structure and function of the ecosystem.

The pine flatwoods community is the most widespread community type in Florida

and historically extended over much of the southeastern coastal plain (Abrahamson and

Hartnett 1990). Unfortunately pine flatwoods are one of the communities most impacted

by human activities (Abrahamson and Hartnett 1990). Pine flatwoods are characterized

by scattered longleaf pine (Pinuspalustris) trees with a dense, species-rich ground layer

of perennial herbaceous plants, which historically were maintained by naturally occurring

high-frequency (5-10 years, Platt 1988), low-intensity forest fires (Abrahamson and

Hartnett 1990, Pyne et al. 1996). Such fires were carried by the fuels of the herbaceous

understory, especially Aristida strict Trin. & Rupr. (wiregrass (Kesler et al. 2003)).

Presently, many landowners and land managers manage the pinelands with prescribed

fire to maintain community species composition and structure.

Upland restoration in Florida, in the pine flatwoods and sandhill communities, is

often modeled after restoration that has occurred in the prairie states for many years









(Cottam 1987, Packard and Mutel 1997). Groundcover in the pine flatwoods strongly

resembles the tallgrass prairie of the Midwest both in structure and function. This

restoration has focused primarily on the native groundcover under the assumption that the

groundcover (dominated by A. strict in Florida) is what fuels the growing-season fires

that maintain pine flatwoods and sandhills (Pyne et al. 1996). Restoring A. strict cover is

a primary goal of pasture restoration. Restorers also assume that longleaf pine trees can

always be added after groundcover is established (N. Bissett, The Natives 2002, pers.

comm.).

Restoration begins with site preparation to remove the non-native cover and reduce

the weedy seed bank. Many practitioners recommend that preparation occur over a period

of about one year (N. Bissett, The Natives, Inc. 2002 pers. comm.; B. Wertschnig, CF

Industries, Inc. 2000 pers. comm.). Removal of non-native pasture grasses is through

herbicide application and soil disking (Bissett 1996). Soil disking is a type of plowing

that shallowly turns the soil to about 15 cm depth. Disking breaks up dense root mats of

pasture grasses and tends to scarify and raise seeds in the seed bank to the surface. Soil

disking is done for various reasons and is usually repeated often throughout site

preparation. Disking repeatedly surfaces new seeds so they can germinate and then turns

the seedlings back under to kill them (The Nature Conservancy 1996). This technique

effectively kills P. notatum and exhausts the seed bank's weedy component that tends to

dominate restoration efforts in pastures (N. Bissett, 2002, pers. comm.; The Nature

Conservancy 1996). These sites are then sown with native understory/herbaceous seed

collected from as close a source as possible.









Seed-collection sites are burned in the growing season and the seed is sown in the

winter (Bissett 1996). Seeds are collected in the fall in central Florida (using various

methods, but most commonly with a silage cutter mounted on the front of a tractor). The

tractor is driven through a flatwoods and it cuts and collects everything above 31-45 cm

above the ground (Bissett 1996). Alternatively, a Woodward Flail-vac (used at DWP; Ag

Renewal Inc., Manhattan, KS) is driven through the flatwoods and strips the seeds off the

stalks of mostly grasses and forbs with a large rotating brush (approximately 45 cm above

the ground). The seed then accumulates in a holding chamber. Once collected, the seeds

are stored and sown onto the site in the winter. The seed is sown with a hay blower onto

the prepared soil and then pressed into the soil with a roller. Rolling improves seed

contact with soil and increases seed germination in A. strict (Gordon et al. 2000).

Invasive non-native species control is required for many years in these sites after seeding.

Upland restoration at DWP began in 1995 with the creation of an experimental

pilot study (The Nature Conservancy 1996). This study aimed to identify the best method

of site preparation for restoration. Six randomly placed sites (one within each of six P.

notatum pastures) were chosen; and 5 methods of P. notatum removal (single disk, single

herbicide, disk and herbicide, multiple disk, and multiple herbicide) were administered to

each site. Then the sites were sown with native flatwoods seed mix collected on-site (The

Nature Conservancy 1996). Vegetation was monitored for 3 years after seeding to

determine if significant differences existed among the site-preparation methods in the

percent cover of native and non-native species. After 2 years of monitoring, the multiple

herbicide method had a significantly higher percent cover of native species and

significantly lower percent cover ofP. notatum, while the other three methods were not









significantly different from one another in percent cover of P. notatum (The Nature

Conservancy 1999). There was tremendous native species richness that became

established in the sites in all treatments (The Nature Conservancy 1999). Total species

richness recorded was 219 of which 141 were native species characteristic to pine

flatwoods, 38 were native weedy species, 37 were non-native weedy species, and 3 were

non-native invasive species (The Nature Conservancy 1999). Species richness ranged

from 34 to 47 species per 450 m2 in 1997, 35 to 61 per 450 m2 species in 1998 (Gordon et

al. 2000), and 30 to 45 per 450 m2 in 1999 (Gordon et al. 2001). Average species richness

per site was 37 species in 1997, 46 species in 1998, and 39 species in 1999 (Gordon et al.

2001). However, no further site management occurred and many of the characteristic

native species were no longer present by 2001 (A. Jenkins 2001, pers. obs.). Management

methods for maintaining native species in such restoration sites for more than a few

years, therefore, requires further investigation.

Results of this experiment led to the implementation of larger restoration units on

DWP. A 10.7 ha restoration effort at DWP began with site preparation (disking and

herbicide application) in 2001 and seeding occurred in early February 2002. Another

adjacent unit of approximately 26.39 ha was sown in January 2003 while a 5.26 ha unit

was site prepared during 2002-2003 and will be sown in 2004.

The results of the pilot experiment and other restoration projects around the state

raised questions that led to the work in this thesis. Those original experimental pilot sites

and the new 10.7 ha and 5.26 ha restoration units at DWP provided suitable locations to

investigate whether the tremendous species richness recorded in the DWP study could be

attributed to a persistent seed bank of native flatwoods species, a seed bank that remained









in the soil since pasture conversion. Gordon et al. (2000) concluded that "existing seed

banks or dispersal from adjacent sites clearly contributed significantly to the high species

richness that resulted." They did not attribute the high species richness to the applied seed

mix alone (Gordon et al. 2000). In Chapter 2, I compare the composition of the soil seed

bank in the 18-ha pasture restoration unit and an adjacent flatwoods at DWP. The other

puzzling result from the experimental pilot study was the lack of persistence by many

native species that germinated. This result led me to question whether some soil factor

was responsible. Factors such as the lack of beneficial mycorrhizal fungi may reduce the

competitive advantage of the newly established native species. In Chapter 3, I examine

the effects of restoration on soil mycorrhizal inoculum potential by comparing restoration

soils with soils in flatwoods and P. notatum pastures. I also compared mycorrhizal field

colonization of A. strict plants growing in the previously mentioned experimental sites

and adjacent plants in natural flatwoods. Addressing these questions will lead us to better

understand the ecology of pasture restoration and to more successfully establish the

native pine flatwoods community in restored pastures in Florida.













































Figure 1-1. Location of the The Nature Conservancy's Disney Wilderness Preserve
within the central Florida region.














CHAPTER 2
SEED BANKING IN A PASTURE RESTORATION SITE AND MESIC PINE
FLATWOODS IN CENTRAL FLORIDA

Introduction

Restoration efforts may be simplified when a seed bank of native species from the

historical native community remains in the sites involved. A seed bank is a pool of seeds

in the soil profile that can germinate at some time and replace adult plants of the

aboveground floristic community (Baker 1989). Restoration of native communities by

using a "relict" native seed bank would be ideal since collecting native seed is extremely

time consuming and expensive, and seed is often unavailable. It is unknown, however,

how long native seeds remain viable in the seed bank in soils that have been converted to

agriculture (van der Valk and Pederson 1989). Seed banks have been successfully used in

certain areas such as heathlands in Britain where native propagule banks (seeds and buds)

were used as donors to reestablish heathlands in restoration sites (Putwain and Gillham

1990). However, persistent seed banks are not always present. For example, research in

limestone prairie revealed that the dominant species were absent in the seed bank and

needed to be actively reintroduced (Laughlin 2003). This research investigates the

composition of the native pine flatwoods relict seed bank in the soils of pastures in

central Florida in an applied context for future restoration. By comparing the seed bank

composition of native pine flatwoods soils to pasture restoration areas, seeding needs of

the site can be better evaluated.









Pastures in Florida are usually vegetatively dominated by non-native grass species

such as Paspalum notatum, Cynodon dactylon, Panicum repens, and Digitariapentzii,

and harbor other non-native invasive species as well as weedy and aggressive native

species such as Eupatorium capillifolium (dogfennel). Efforts to restore such degraded

landscapes to the natural assemblage of native pine flatwoods species is often hindered

by competition from these non-native and weedy native species (Harper-Lore 1998,

Brown and Bugg 2001) that are present in the seed bank (D'Antonio and Meyerson

2002). As a result, extensive site preparation is essential to restoration (Cottam 1987),

especially in pastures because it removes competing non-native seeds from the seed bank

through continued herbicide application and disking of the soil which promotes flushes of

seedling growth before native seeds are introduced to the site (Harper-Lore 1998).

The pine flatwoods in Florida are dominated by a species-rich herbaceous ground

cover (Hardin and White 1989), which is the target suite of species for The Nature

Conservancy's pasture restoration efforts. Pine flatwoods can also have a non-continuous

shrub layer consisting of Serenoa repens (saw palmetto), Ilex glabra gallberryy), various

species of Ericaceous plants, and a pine canopy which is primarily Pinuspalustris

(longleaf pine) with occasional Pinus elliottii (slash pine) along wetland margins.

Aristida strict (wiregrass) (Kesler et al. 2003) is the dominant perennial grass species, a

keystone in the pine flatwoods community because it produces fine fuel which carries

low-intensity community-maintaining fires (Platt 1988, Clewell 1989, Noss 1989,

Abrahamson and Hartnett 1990). Many pine flatwoods species rely on these low-intensity

fires for their existence (Clewell 1989, Noss 1989). For example, Pinuspalustris is a

poor competitor and needs the low-intensity fires to control hardwoods while many









herbaceous species reproduce only after such low-intensity fires (Clewell 1989). As a

result, A. strict is primarily targeted as a target species to reestablish in upland

restoration of both pine flatwoods and in the similar community type, sandhill, because it

aids in reestablishing community structure and function. Extensive research has been

dedicated to seed germination and establishment of A. strict (Seamon and Myers 1992,

Mulligan et al. 2002). Sandhill restoration studies found that A. strict only becomes

established when directly sown in revegetation efforts (Hattenbach et al. 1998, Cox et al.

2003). For this reason, A. strict seed is collected and directly sown into restoration sites

across the state. Other species are sometimes hand-collected and added to the seed mix

but the quantities are far lower than that for A. strict. Earlier work has demonstrated that

A. strict is often the only species introduced from the seed mix restoration efforts

(Gordon et al. 2000).

Seed bank composition can influence restoration approach and objectives (van der

Valk and Pederson 1989). The composition of the relict native seed bank in pastures that

were formerly flatwoods has been little studied (Violi 1999), particularly comparing the

composition of the pasture seed bank to that of intact flatwoods soils. Knowing the

differences and similarities in the seed bank between pastures and flatwoods will be

useful to upland restoration efforts in pastures because practitioners will better be able to

evaluate seeding and future non-native species management requirements. Differences

among agricultural and disturbed land uses could be an important influence on the relict

seed bank composition for example, abandoned pastures in Lake Placid, Florida, were

found to have a species rich native seed bank (Violi 1999), while very few "obligate"

sandhill species were found in the seed bank of abandoned citrus groves in central Florida









(Buchanan 1999). Hattenbach et al. (1998) and Cox et al. (2003) both found species rich

native seed banks in disturbed sandhill sites in north Florida.

Many different methods for quantifying the presence and abundance of seeds have

been used in studies of soil seed banks. Most studies have quantified the seed bank by

spreading soil samples into germination flats, placing them in a greenhouse or

shadehouse, and identifying and counting any germinated seedlings (Rabinowitz 1981,

Kitajima and Tilman 1996, Carrington 1997, Schott and Hamburg 1997, Butler and

Chazdon 1998, Leckie et al. 2000). However, only viable seeds whose specific

germination cues are met in the environmental conditions of the test are identified by this

method. As a result, some seeds and species are likely underestimated or missed. Other

studies have used direct separation of seeds from soil with a salt solution method (Malone

1967, Brown 1992, Buhler and Maxwell 1993), sieving (Carrington 1997), or elutriation

(Gross 1989). These direct separation methods may overestimate the viable portion of the

seed bank unless seed viability is also measured. At least two studies have compared

more than one method of estimating the soil seed bank (Gross 1990, Brown 1992), with

inconsistent results. Germination trials with cold-stratified soil resulted in more species

than direct germination or elutriation in a ploughed field in Michigan (Gross 1990). In a

woodlot in Ontario, 102 species were detected by elutriation and only 60 by the

germination technique (Brown 1992). A combination of methods might be most suitable

to accurately quantify the composition of a soil seed bank.

One objective of this study was to investigate whether there were significant

differences in species composition of the soil seed bank between a pasture restoration

area (prior to sowing) and the adjacent, relatively undisturbed flatwoods. Another









objective was to compare the seed bank composition in the restoration soils with that of

the aboveground vegetation in the restoration area 6 months post restoration seeding to

determine similarity. I also wanted to determine which species germinated in the

restoration field but were not in the seed bank assays to evaluate which species could be

attributed to the seed mix. In this study, I examined the composition of the seed bank

with two methods. The seed bank species composition was evaluated through the

germination method and a direct counting method sievingg) and frequency of plant

species in the restoration area was examined. I hypothesized that the seed bank in the

restoration and flatwoods soils would be similar in species composition, that both would

be dominated by weedy native species, and that the restoration soils would have more

non-native species than the flatwoods soils. I also hypothesized that the vegetation

emerging in the restoration area would be composed of similar species as in the two seed

bank assays, with the addition of the sown A. strict. I predicted that the two methods of

evaluating seed bank species compositions would yield differing results based on the

findings of Gross (1990) and Brown (1992).

Methods

The Nature Conservancy initiated a 10.7 ha upland restoration project at the Disney

Wilderness Preserve (DWP; see site description in Chapter 1) in June 2001. The site was

treated with herbicide and disked six times over the course of the year to deplete the

weedy seed bank and control non-native species. Seed collection of A. strict and other

native herbaceous species (as described in Chapter 1) occurred in November 2001 on-site

in pine flatwoods stands (121.5 ha) with dense A. strict plants that were burned in

March and July 2001. Several species were also hand-collected and added to the seed

mix. These species include: Buchnera americana, Panicum rigidulum, Rhexia nashii,









Sorghastrum secundum, Pterocaulon virgatum, Dichanthelium spp., Pityopsis

graminifolia, Panicum anceps, Solidagofistulosa, Axonopus spp., L/i uni'he\

caroliniana, Dichanthelium portoricense, Saccharum giganteum, and Eragrostis spp.

Total seed mix collected was approximately 4.536 m3 (40, 30 gallon bags), which was

stored in a cool, dry place until seeding occurred (K. Kosel 2001, pers. comm.). The seed

was sown with a modified hay blower on 15 February 2002 onto the prepared 10.7 ha

restoration site. Just before and after seeding, the soil was compacted with a cultipacker.

Nomenclature throughout follows Wunderlin (Wunderlin 1998) except A. strict

which follows Kesler et al. (2003). Species were also separated into categories to better

distinguish those that are characteristic of native pine flatwoods and represent the long-

term restoration goal (hereafter "native characteristic") from more weedy or early

successional species (hereafter "native weedy") (The Nature Conservancy, unpublished

data). We also separated non-native weedy species and non-native invasive species, the

latter identified through a listing of species invasive in natural areas in Florida (Florida

Exotic Pest Plant Council 2003).

Germination Assay

To compare the seed banks in the restoration area and the adjacent pine flatwoods,

a seed bank germination experiment was initiated. The restoration area was classified into

five elevation strata to capture variability across the site. Using a topographical map and

aerial photographs; one 50 m X 50 m quadrat was permanently marked within each

stratum. Five strata at the same elevations were established in the adjacent native

flatwoods. These flatwoods were chosen based on several criteria: proximity, species

composition, and the same soil type and elevation range as those in the restoration area.









On 9 February 2002, in each of the 5 strata in each site (restoration and flatwoods),

10 randomly placed soil samples were collected. Each sample was a composite of 2 soil

cores taken 0.5 m apart and well mixed. Each soil core was 12 cm deep x 8 cm diameter

(surface 0.0050 m2; 602 cm3). Composited samples were therefore 0.01 m2 (1204 cm3).

We chose to sample at this depth because the soil in the restoration area had been

repeatedly disked to roughly 12 cm (K. Kosel 2001, pers. comm.). Similar depths have

been used in other seed bank studies (Rabinowitz 1981, Butler and Chazdon 1998). All

samples were immediately placed in a cold room (10 C) and stored there until the

shadehouse experiment was established on 20 April 2002. Soil was collected immediately

prior to restoration seed sowing in the field (15 February 2002).

The experiment was established in an open air shadehouse on the University of

Florida campus in Gainesville (Alachua Co., FL). The shadehouse has benches on which

the flats for the experiment were placed and a wooden frame, which supported 50%

shade cloth. Plastic sheeting (4 mil polyethelene film) was also installed just under the

shade cloth to prevent direct hard rain contact with the soil in the flats. This was done to

avoid unintentional mixing of soil and seeds from flat to flat in heavy rain storms. A

misting-style sprinkler was installed and the flats were irrigated daily for one hour. Flats

were 25 cm X 50 cm in dimension and divided by a barrier into two 25 cm x 25 cm

sections. Newspaper was placed in the bottom of each flat section to prevent soil loss and

to aid in moisture retention.

The soil used for this experiment was 50% sand and 50% vermiculite. The sand

was autoclaved for 45 minutes at 121 OC to kill any contaminating seeds. Each flat

section was filled with 3 cm of the soil mix and watered heavily. Field soils were









randomly assigned to the flat sections (50 flatwoods, 50 restoration area, and 20 control).

The control flat sections were included to account for germinated seeds that were

dispersed from the local area instead of the field site. The control soil was 50% store

bought potting soil and 50% sand (both autoclaved for 45 minutes at 1210 C to kill any

contaminating seeds). In each flat section, 236.5 cm3 of the field soil (restoration or

flatwoods) or control soil was spread to a depth of 0.38 cm over the moistened

sand/vermiculite mixture. This amount equates to 39.4 % of each composite soil sample

and therefore represents 0.002 m2 of soil at 0-12 cm depth. After the experiment was

established, the unused field soils were air-dried and stored for use in the direct seed

count assay (below).

Germination was checked bi-weekly for the first 10 weeks and then weekly until

germination ceased, around 15 August 2002. The arrangement of the flats on the benches

was randomly repositioned weekly to minimize any irrigation or sunlight variability

caused by bench position. Data on germinated seedlings were recorded each monitoring

day for date of emergence and morphological characteristics. Most seedlings were also

marked with color coded toothpicks for easier relocation. Most species of grasses and

sedges were not identified until they produced reproductive structures which did not

occur until approximately August/September 2002 or later. In many flat sections

(primarily restoration soils) numerous Cyperus spp. seedlings emerged simultaneously.

Because tracking that many seedlings was too difficult and competition among seedlings

would reduce survival, I randomly removed a known number of the Cyperus spp.

seedlings. As a result, all species in the genus Cyperus are listed throughout as Cyperus

spp. However, at least 5 species were represented in each of the restoration and flatwoods









soils in the germination assay: C. compressus, C. globulosus, C. polystachyos, C.

retrorsus, and C. surinamensis were identified among the remaining Cyperus spp.

seedlings left in the flats.

On 15 September 2002, a heavy rain severely damaged the plastic sheeting,

allowing all of the flats to be subjected to a heavy rain event. Several of the flats (8

flatwoods and 3 restoration) were completely destroyed by the water. All data from these

flats were eliminated from the experimental analysis because many seedlings had not yet

been identified. The plastic sheeting was replaced. No germination was observed after

this date.

We identified most of the species that germinated in the germination assay with the

help of field keys (Wunderlin 1998) (Godfrey and Wooten 1979, Godfrey and Wooten

1981), field comparison, and experts (N. Bissett, The Natives Inc. pers comm.2002).

However, individuals that produced no reproductive structures by the end of the

experiment were not identified beyond monocots and dicots. They were included in

estimates of seed density.

Direct Count Assay

I initiated a direct count seed bank investigation on 28 January 2003. The same soil

samples that were used for the shadehouse experiment had been air-dried and stored in

plastic bags for use in this investigation. Four samples from each elevation strata were

randomly chosen and three 20 cm3 subsamples were taken from each. The three

subsamples together represent 0.0005 m2 (60 cm3) of soil (at 12 cm depth). Each sub-

sample was poured through a series of three sieves. The sieves were stacked together in

size order with the largest on top (3.5 mm, 0.9 mm, 0.5 mm). The soil caught by each

sieve was transferred to a single layer in a petri dish and examined under a dissecting









microscope. The soil that fell through the 0.5 mm sieve was discarded following

examination of 10 subsamples that yielded no seeds. Seeds were extracted from the soil

under microscope examination, labeled, and set aside for identification. The seeds were

identified when possible by comparison to herbarium specimens at the University of

Florida, Florida Museum of Natural History herbarium, published line drawings (Godfrey

and Wooten 1979, Godfrey and Wooten 1981), photographs (Martin and Barkley 1961,

Landers and Johnson 1976), and expert consultation (R. Abbott 2003, pers. comm.).

Identification was to species where possible and to morphospecies where not possible.

Vegetation Survey

To accompany the seed bank data, I also studied species assemblages in the

restoration site which incorporated any plants that emerged in the restoration area from

the seed bank and sown seed mix. I monitored the vegetation in the one-year restoration

field in August 2002 (6 months post restoration seeding) in the same 5 strata that were

used for the seed bank sampling. Each of the 50 X 50 m quadrats in the restoration area

was divided into 100, 5 X 5 m subquadrats. Ten of these squares were randomly selected

in each quadrat for vegetation sampling. In the northeast corner of each randomly-chosen

subquadrat, I placed a nested 20 cm X 20 cm and 1 X 1 m frame (nested frequency

sampling). Presence of any species within each 20 x 20 cm, 1 x 1 m, and 5 x 5 m quadrat

was recorded. Additionally, density ofA. strict plants was recorded in the 1 x 1 m frame

to examine A. strict recruitment. All plants were keyed to species in the field when

possible and unknown plants were collected and subsequently identified.

Data Analysis

Species accumulation curves were generated to compare the species richness

between treatments (restoration and flatwoods) in the germination and direct count









assays, and among the three types of data collected in the restoration soils (germination,

count, and field emergence) (Colwell 1997, Chazdon et al. 1998). Species accumulation

curves were generated with the computer software, EstimateS (Colwell 1997), from 50

randomizations of the samples to compare observed species richness among treatments.

A non-parametric estimator of species richness was used to estimate true species richness

based on the observed species (Colwell 1997, Chazdon et al. 1998). The Incidence-based

Coverage Estimator (ICE) was calculated with EstimateS (Colwell 1997), and chosen

based on the results of Chazdon et al. (1998) that compared multiple non-parametric

estimators. Coleman curves were also generated with EstimateS (Colwell 1997), which

evaluate patchiness in the distribution of species in the dataset by sub-sampling from the

50 randomizations. If the Coleman Curve overlaps the species accumulation curve, the

distribution of the species is not considered patchy. Unique species are defined as species

that only occur in one sample.

The Jaccard (Cj) incidence-based similarity index (Magurran 1988) was calculated

to compare seed bank species composition between the flatwoods soils and restoration

soils in the germination and direct count assay. The Bray-Curtis quantitative dissimilarity

index (Underwood and Chapman 1998), that accounts for abundance, was calculated to

determine dissimilarity of seed bank species composition and abundance between the

flatwoods and restoration soils in both laboratory assays.

Differences in species richness among the five elevation strata in the vegetation

survey for each of seven vegetation categories (native characteristic graminoid, native

weedy graminoid, non-native invasive graminoid, non-native weedy graminoid, native

characteristic forb, native weedy forb, and non-native weedy forb) were tested using









ANOVA (SAS 1999). ANOVA was also used to determine if the density of A. strict

plants in the restoration field was dependent on the elevation strata. Tukey tests (p <

0.05) were used to determine post-hoc differences among means.

Results

Germination Assay

Most seedlings germinated within 70 days after sowing with a total of 40 species

recorded in the shadehouse experiment (see Table A-1). The flatwoods and restoration

soils each had 31 species, 20 of them shared among the two soil types. Of the 40 species,

21 were native species characteristic to flatwoods, 10 were native weedy species, and 5

were non-native weedy species (no category determined for four unknowns). Several

individuals were only determined to genus, one species was only determined to family

(Asteraceae), and unknown dicot and sedge groups were used. Two of the flatwoods

samples showed no germination.

Species composition was 35% similar (Cj = 0.359) between the flatwoods and

restoration soils. Of the 40 total species found, 55% were forb and 45% were graminoid

species (Table A-i). Nine species were unique to each treatment and three of the species

unique to the restoration soils were non-native species (Table 2-1). Several perennial

grasses were recorded in the restoration soils (i.e., Andropogon sp. and Dichanthelium

portoricense), however none of the perennial bunchgrasses most characteristic of mature

flatwoods such as A. strict, Sorghastrum secundum, and Schizacharium scoparium were

found (Table A-4).

The relative abundances of species in the two treatments were significantly

different. The Bray-Curtis quantitative dissimilarity index revealed that the two

treatments were 92.3% dissimilar in species composition when abundance was accounted









for. Only two species, Cyperus spp. and Hedyotis uniflora, were among the top most

abundant in both soil types (Figure 2-1). The five most common species in the restoration

soils were Cyperus sp., Hedyotis uniflora, Scoparia dulcis, Ludwigia octavalvis, and

Kyllinga brevifolia (Figure 2-la). Cyperus sp. dominated all other species in these soils

and consisted of at least 5 species (C. compressus, C. globulosus, C. polystachyos, C.

retrorsus, C. surinamensis). These species together averaged 71.5 seedlings per sample

and accounted for 83% of all seedlings in the restoration soils (Figure 2-la). Cyperus spp.

averaged only 2.05 seedlings per sample in the flatwoods soils (Figure 2-1b) and only

accounted for 6% of the total seedlings germinated. The five most abundant species in the

flatwoods soils were Eleocharis sp. filiformm type), Hedyotis uniflora, Dichanthelium sp.,

Cyperus spp., and Polypremum procumbens (Figure 2-1b). The most common species in

the flatwoods soils was Eleocharis sp., which averaged 16.9 seedlings per flat section and

accounted for 51% of all seedlings in the flatwoods soils, followed by Hedyotis uniflora

which averaged 6.52 seedlings per plot (20% of all flatwoods seedlings) (Figure 2-1b).

The mean species accumulation curve approached its asymptote more rapidly for

the restoration soils than the flatwoods soils (Figure 2-2a). The ICE estimator for

incidence-based species richness predicted higher species richness in the flatwoods (40.7)

than in the restoration soils (36.4). The curves indicated the sufficiency of the sampling,

since the observed species richness is close to reaching an asymptote (particularly for the

restoration soils) and the ICE predicted values. The Coleman Curves (not shown here) for

each soil type overlapped the species accumulation curves and therefore did not predict a

patchy spatial distribution for either soil type. In flatwoods samples, 9 of the 31 species

occurred only once, whereas in the restoration site samples 7 of the 31 species only









occurred once. The seed bank in the restoration soils seems to be more homogeneous

because of the steeper slope on the species accumulation curve (Figure 2-2) and fewer

unique species. This pattern likely resulted from the soil mixing that occurred during

pasture conversion and restoration site preparation. Greater patchiness would be expected

in the undisturbed flatwoods soils.

Estimated seed density in the restoration and flatwoods soils differed by almost

40% when numbers are extrapolated to seedlings per m2. Three times more seedlings

emerged from the restoration soils (4045 seedlings per 0.094 m2 of soil = 43,032 seeds m

2) than the flatwoods soils (1402 seedlings per 0.084 m2 of soil = 16,690 seeds m-2).

Direct Count Assay

Direct seed count resulted in a total of 22 species. Thirteen species were found in

the flatwoods soils and 15 in restoration soils with seven overlapping species (see Table

A-2). Species were labeled as unknown when genus and family could not be determined.

Many more seeds could not be identified in this experiment compared to the germination

assay. In addition, the condition of many soil-extracted seeds was poor (due to herbivory,

soil scarification, broken parts, etc.) and vital morphological features were often missing.

Both the flatwoods and restoration soils were dominated by native graminoid seeds.

However, the flatwoods graminoid species were native and characteristic of pine

flatwoods whereas those in the restoration soils were dominated by native weedy species.

Non-native species were only found in the restoration soils: Desmodium triflorum,

Fimbristylis schoenoides, and Kummerowia striata. The two soil types were 29% similar

in species composition according to the Jaccard index (Cj = 0.29). Again, as in the

germination assay, none of the perennial bunchgrasses most characteristic of native

flatwoods were found in either soil type.









Although observed species number was similar in both the flatwoods and

restoration soils, ICE estimated species richness was much higher for the flatwoods soils

(45 species) than for the restoration soils (16 species). This large difference in species

richness reflected their difference in accumulation curves (Figure 2-2b). The cumulative

number of species in the restoration soils almost reached an asymptote within the 20

samples while those in the flatwoods did not (Figure 2-2b). The large difference in the

ICE estimated species richness for the restoration and flatwoods for this direct seed

counting reflects differences between the soil types in incidence of rare and common

species. For the total area of 0.01 m2, 15 species were found in the restoration soils, two

of which occurred only once, whereas 13 species were found in the flatwoods soils, and

nine of them occurred only once. Although the Coleman Curves generated (not shown

because they overlapped the species accumulation curve) did not predict patchy spatial

distribution, the species accumulation curves and the number of unique species found

demonstrate greater homogeneity in restoration site than flatwoods seed banks. Again,

this difference is likely due to loss of species and the soil mixing that occurred in the

restoration soils during pasture management and site preparation.

Consistent with the germination study, only a few species accounted for the

majority of the seeds extracted from the soil, and these dominant species were different

for the two soil types. The Bray-Curtis dissimilarity index revealed the two treatments to

be 91.3% dissimilar. The dominant species in the flatwoods soils were Fimbristylis

autumnalis (present in the germination assay but not among the top five species) and

Dichanthelium portoricense. Those two species accounted for 87% of all the seeds

extracted from the flatwoods soil but only 4% of the seeds extracted from the restoration









soils. Cyperus compressus, Cyperus globulosus, Cyperus polystachyos, Kyllinga

brevifolia, and Fimbristylis dichotoma (all Cyperaceae) dominated the restoration soils

and accounted for 91% (15%, 16%, 17%, 25%, and 18% respectively) of the total seeds

extracted. None of these five species were found in the flatwoods soils. Again, many

more seeds were extracted from the restoration (1258 in 0.01 m2 = 2,516,000 seeds m-2),

than from the flatwoods soils (202 in 0.01 m2 of soil = 404,000 seeds m-2). There were

four species found in the direct count that did not germinate in the germination assay but

none of them were dominant components of the directly counted seeds and none were

identified to species.

Vegetation Survey

The restoration area at the time of the vegetation survey was extremely dense with

vegetation approximately 60 cm tall with scattered taller plants such as Phytolaca

Americana (pokeweed) and Eupatorium capillifolium (dogfennel). Most species were

annuals and producing reproductive structures at the time of sampling with the exception

of several perennial species such as A. stricta plants which generally had 1-2 leaves about

15 cm tall. I found 84 species in the vegetation sampling (Table A-3), 31 of which were

also found in the two seed bank assays. Of those 84 species, 52 were native species

characteristic to flatwoods, 16 were native weedy species, 13 were non-native weedy

species, and 3 were non-native invasive species (Table A-3). While the species varied

across the elevation strata, the species richness observed in the vegetation survey were

consistently higher than in either of the seed bank assays (Figure 2-3a). Of the 49 species

that were observed in the vegetation survey but never in either seed bank assay, 67%

were native characteristic species, 10% were native weedy species, 14% were non-native









weedy species, and 6% were non-native invasive species (which were absent from all

seed bank assays).

The elevation strata differed in species richness (Figure 2-3b). Stratum 1 was at the

lowest elevation, close to a cypress dome, and had the highest species richness. This

richness was probably due to the additional wetland species that were specific to that

stratum (i.e., Amphicarpum muhlenbergianum, Rhynchospora microcephala, Sabatia

grandiflora, Rhexia nashii). The remaining strata had variable levels of species only

occurring in one stratum (Figure 2-3b). All strata shared 21 species in and many species

were sporadic (shared by at least two strata) (Figure 2-3b, see Table A-3).

The species richness varied among the categories (F = 73.87, p< 0.0001) and

elevation strata (F = 7.04, p < 0.0001) (Table A-3). While not consistent along the

elevation gradient, I found significantly more native characteristic graminoids in Strata 1

and 4 than in Strata 5 and 3 (F = 7.18, p < 0.0009), non-native weedy graminoids in

Stratum 1 than in the other strata (F = 25.58, p < 0.0001), and native characteristic forbs

in Strata 1 and 2 higher than in Stratum 3 (F = 5397, p < 0.0025).

Aristida stricta Recruitment

Aristida strict was found in all but four of the 25, 1 X 1 m plots monitored. The

average density was 2.56 m2 per plot and ranged from 0 to 15 m2 per plot with no

significant differences among the elevation strata (Figure 2-4). Differences may be

attributable to uneven seed density.

Methods Comparison

Species richness per area of soil sampled for the restoration soils could be

compared through the species accumulation curves for all trials conducted, scaled for the

area sampled (Figure 2-5). As discussed earlier, the germination assay yielded many









more species than did the direct count assay for the restoration soils (Figure 2-5a). For the

restoration soils, the species richness was greatest in the vegetation survey which

sampled far more soil area (625 m2 total) and incorporated both seed bank and sown

species (Fig. 2-5b).

Another comparison of the different methods for seed bank quantification is to

examine the number of seeds/seedlings per m2. The species richness per sample in the

restoration and flatwoods soils is fairly consistent in the germination assays when

calculated for equivalent area (m2), but differs in the direct counting assay (Figure 2-6).

Fewer seeds were identified in the direct count than in the germination assay from the

flatwoods soils (Figure 2-6). This discrepancy is because Eleocharis sp. and Hedyotis

uniflora, which were the most abundant species in the germination assay in the flatwoods

soils, were never extracted from the soils in the direct count assay.

Discussion

I found clear differences in the seed bank composition of the flatwoods and

restoration sites in both germination and direct counting assays. The two soil types were

only 35% and 29% similar in species composition in the germination and direct count

assays, respectively. The higher abundance of seeds in the restoration soils was also a

noticeable difference (92.3 % and 91.3% dissimilar to flatwoods soils when the

abundance of seeds are accounted for in the germination and direct count assays

respectively). The flatwoods soils had higher incidence of unique species than the

restoration soils. Additionally, the species that dominated the seed bank differed by soil

type. The restoration soils had more seeds of non-native and weedy native species, while

the flatwoods had more native characteristic flatwoods species (Tables 1, A-i, A-2). It is

probable that disking during site preparation, which is meant to eliminate the weedy seed









bank in the former pastures, also eliminated relict native characteristic species from the

seed bank.

Another distinction that can be made between the seed bank composition of the

flatwoods and restoration soils is that of permanent and transient seed bank. Transient

seed banks exist when certain seeds bank in the soil for a short period of time such as just

until the next favorable season (Thompson 2000). Permanent seed banks exist in the soils

for longer periods of time and have many reasons and mechanisms for their persistence

(Thompson 2000). The seed bank composition in the restoration soils is primarily

permanent seed bank since there were no plants reproducing in the field for about one

year before the samples were taken. However, the flatwoods soils likely contained both

seed bank types. Germination assays with soil taken at different times of year in the

flatwoods could reveal which species might be only transiently banking in the soil.

Elimination of any transient species from the flatwoods seed bank species composition

would yield a more valuable comparison between the two soil types.

The seed bank species composition results confirm what many restoration

practitioners already suspected: native sedges, mostly in the genera Cyperus and

Fimbristylis, and annual dicots, such as Hedyotis uniflora, Polypremumprocumbens and

Ludwigia spp., dominate the seed bank in restoration areas in central Florida. These

results are similar to seed bank species composition found in pastures in Lake Placid, FL

(Violi 1999), despite the greater estimated seed density in my study. The Cyperaceae

family is frequently dominant in the seed banks of grasslands (Roberts 1981). The

restoration area seed bank at DWP is rich and diverse with native species. Some are early

successional "weedy" and some are characteristic of healthy, mature pine flatwoods









systems. Restoration areas that were formerly sandhill in north Florida have also been

found to have rich, diverse native seed banks (Cox et al. 2003).

Many dominant perennial bunchgrasses were absent from all seed bank assays,

most importantly the keystone species, A. strict. Similar findings were reported from

three sandhill studies in North Florida (Hattenbach et al. 1998, Cox et al. 2003), (G.

Parks, unpublished data), a sandhill study in central Florida (Buchanan 1999), and a

flatwoods study in south central Florida (Maliakal et al. 2000). Aristida strict has only

been found in sandhill restoration plots when directly sown (Cox et al. 2003).

Apparently A.stricta must be sown in order to become established in restoration

sites in central Florida. Because of the importance of this species for system function

(Platt 1988, Clewell 1989, Noss 1989, Abrahamson and Hartnett 1990), where lost from

the system through habitat degradation or destruction, A. strict must be actively

reintroduced. Others have observed that supplementation of the seed bank is necessary in

restoration efforts when the dominant plant species is unlikely to be found in the seed

bank (Laughlin 2003). Many guidelines have been presented for the best methods of

reestablishing A. strict both in pine flatwoods (Bissett 1996, Conservancy 2000) and in

sandhill (Seamon and Myers 1992, Hattenbach et al. 1998, Cox et al. 2003).

Species composition of the two seed bank assays and the restoration vegetation

survey were different. (Figures 2-3a and 2-5). The vegetation survey contained many

more native species characteristic to flatwoods and the only account of non-native

invasive species in the study. Therefore, the majority of the species that the vegetation

survey and the seed bank assays had in common were native weedy species. This

difference likely results from several factors. Firstly, when the vegetation field site was









sown, other species were also added through the seed mix (e.g., A. strict, Andropogon

virginicus var. glaucus, and Pityopsis graminifolia). Secondly, the presence of non-native

species also contributes to this difference. Species such as Panicum repens, Cynodon

dactylon, and Solanum viarum likely were not sown in with the seed mix and came from

the seed bank or vegetative propagules in the restoration area. Non-native and weedy

species were probably introduced during pasture creation and cattle or agricultural

management, as the adjacent non-converted flatwoods did not contain these species.

Similar findings have been reported from an abandoned agricultural area on former

sandhills in central Florida in which a weedy seed bank had no native sandhill species

(Buchanan 1999). Thus, historical land use significantly influences seed bank

composition, with important implications for restoration (Roberts 1981, Bekker et al.

1997).

While some species found in the vegetation survey plots may differ from species in

soil collected prior to sowing because they were introduced in the seed mix, this

explanation is not likely the primary source of species difference. Many of the additional

species found in the field probably originated from the seed bank itself or as vegetative

propagules (i.e., Paspalum notatum) in soil because of their absence from healthy

flatwoods stands, their reproductive phenology (unlikely to have been in seed when seed

was collected), and/or height (i.e., seed stalk is lower than the seed harvester could

access). Dominant species in the vegetation survey such as Eupatorium capillifolium,

were absent in the two seed bank assays. Post-seeding dispersal could be playing a large

role in this difference as well.









Finally, there were large differences in total soil area sampled among the three

sampling methods: vegetation sampling plots were 625 m2; the direct count assay, 0.01

m2; and the germination assay, 0.094 m2 (Figure 2-5). Species area relationships (Gotelli

2001) suggest that area is likely also responsible for the difference in the species richness

observed in the three methods. For all the methods, a larger sample area would have

yielded a greater sampled species richness.

Differences in richness and composition across the elevation strata found in the

vegetation survey suggest that species distribution is patchy and likely to vary within and

among restoration sites. However, aside from higher richness in the wettest stratum, no

clear patterns with elevation were observed. While I am unable to extrapolate to other

flatwoods restoration areas, these results suggest that prediction of species composition of

a restored site, even among relatively homogeneous areas, may not be possible (Klotzli

and Grootjans 2001).

Methods Comparison

I found detectable differences in the seed bank composition between the

germination and direct count methods. Brown (1992) found similar differences in the

estimates of the soil seed bank from the solution separation method and the seedling

germination method. One of the most common species in Brown's germination trial,

Verbascum ithq/i,'t L., only represented 1 % of the seeds directly extracted from the soil

(Brown 1992). Contrasting the two seed bank assays in the flatwoods soils of my study,

the most common species found in the germination assay, Eleocharis sp., was not found

in the direct count assay (Figure 2-6). Clearly this is due to an extraction error since the

soils used for both assays originated from the same soil samples. Perhaps the 0.5 2.5









mm (Godfrey and Wooten 1979) Eleocharis seeds were too small to directly extract from

the soil.

I found far more seeds per m2 through the direct count method (404,000 -

2,516,000 m-2) than the germination method (16,690 43,032 m-2) but fewer species, 22

and 41 respectively. Here, the seedling germination assay underestimated the seed bank

density, whereas the direct count method may have overestimated it. Seeds extracted

from the soils in the direct count assay may not all be viable. Thus, the extrapolated

numbers likely overestimate the real seed bank, particularly as the estimated seed density

is considerably higher than other studies in Florida pasture (Violi 1999) and sand pine

scrub (20 seeds m-2) (Carrington 1997), old-field pine in North Carolina (1470 seeds m-2)

(Oosting and Humphreys 1940), tall-grass prairie (6368 seeds m-2) (Rabinowitz 1981),

pastures in Great Britain (400 70,000 seeds m-2) (Roberts 1981) and an annual

grassland in California (5156 54,687 seeds m-2) (Heady 1958) all of which used the

germination method. Results from this and other studies suggest that the use of only one

method inadequately estimates species composition and seed abundance present in the

soil seed bank. However, the time investment for the direct counting assay was great and

I only found a few species in the direct count that were not represented in the germination

assay. Therefore, based on my results for species richness, direct count assays may not be

worthwhile.

Composition of the aboveground vegetation in the flatwoods was not examined in

this study. However, results from my two seed bank assays may be qualitatively

compared with the aboveground species composition measured by The Nature

Conservancy (The Nature Conservancy 1997) in 15 pine flatwoods in 1997. Most of the









species from my seed bank assays are also in the species list observed by The Nature

Conservancy in the flatwoods vegetation (Table A-4). However, the species that

characterize the flatwoods community are not represented in the seed bank of either the

flatwoods or restoration soils. These include: A. strict, Serenoa repens, and Ilex glabra,

more grass species than forbs (Abrahamson and Hartnett 1990), and many species in the

Ericaceae (FNAI and FDNR 1990) (Tables A-i, A-2). Most of the perennial

bunchgrasses, absent in my seed bank samples, were well represented in flatwoods

vegetation. Maliakal et al. (2000) found that 54% of the species in A. strict flatwoods in

south-central Florida rely only on vegetative regeneration rather than on seed following

fire. Thus, the species composition found in the two seed bank assays are not

representative of the aboveground vegetation in a mature pine flatwoods. Similar findings

were reported in sand pine scrub (Carrington 1997), flatwoods in south-central Florida

(Maliakal et al. 2000), and limestone prairies of Pennsylvania (Laughlin 2003).

Management Implications

Understanding the composition of the seed bank in restoration sites can help

restoration practitioners better gauge the direct seeding needs of their site. Most obvious

is the finding that the pine flatwoods dominant perennial bunchgrasses are absent from

the seed bank both in the flatwoods and restoration soils; thus direct seeding of these

species into restoration projects is necessary, especially if there are no adjacent seed

sources from which they could naturally disperse. These results also suggest that

recruitment of these species may depend on specific factors influencing seed production

or subsequent seedling establishment.

Another important finding based on the vegetation survey was that many non-

native species emerged from the restoration soils and likely came from the seed bank or









vegetative propagules. These species were not thoroughly eliminated by the site

preparation techniques (herbicide and disking) and will continue to require active

management.

This research is vital for the restoration of pine flatwoods communities because

conversion of this land to pasture eliminates native bunchgrasses from the vegetation.

While restoration of a diverse groundcover may not require direct seeding, many of the

species present are early successional or weedy natives that persist in the seed bank under

abandoned pastures. Conversely, seed of the dominant pine flatwoods species must be

reintroduced into restoration projects as they will not naturally regenerate through relict

seed banks.












Table 2-1. Species in common between the restoration and flatwoods soils and unique
species in the flatwoods and restoration soils in the germination assay. Non-
native species are indicated with an asterisk.


Common Species
Andropogon sp.
Cyperus surinamensis
Cyperus compressus
Cyperus polystachyos
Cyperus retrorsus
Cyperus sp.
Dichanthelium sp.
Dichanthelium portericense
Desmodium triflorum
Eleocharis filiform type
Fimbristylis autumnalis
Fimbristylis dichotoma
Gnaphalium sp.
Hedyotis uniflora
Kyllinga brevifolia
Ludwigia round leaf
Murdannia nudiflora
Oxalis corniculata
Ludwigia octavalvis
Polypremum procumbens
Scoparia dulcis
Xyris-like


Flatwoods only
Asterceae basal rosette
Crotolaria rotundifolia
Drosera brevifolia
Eleocharis viviparous
Ludwigia maritima
Rhynchospora fasicularis
Scleria reticularis
Scleria sp.
Toxicodendron radicans


Restoration only
Chamaecrista nictitans
Cyperus globulosus
Erechtites hieracifolia
Fimbristylis .c h/e i, li i"/'"
Hedyotis c ,i ,yinih, %t"
Kummerowia %I ii i/'
Lindernia sp.
Ludwigia repens
Ludwigia sp.











80 80


E 75 75
75 ^H Restoration
-- Flatwoods
0)









CYSP HEUN SCDU LUOC KYBR ELSP HEUN DISP CYSP POPR



Figure 2-1. The mean number of seedlings per sample (+1 SE) in the germination assay
)70 70 U













for the five most common species in the a) restoration soils (n = 47) compared
with the same species in the flatwoods soils and b) flatwoods soils (n = 42)
a)- *25,-
















and species on axes. Species codes as follows: CYSP = Cyperus sp., HEUN =
Hedyotis uniflora, SCDU = Scoparia dulcis, LUOC = Ludwigia octavalvis,
KYBR = Kyllinga brevifolia, ELSP = Eleocharis sp., DISP = Dichanthelium
portoricense, POPR = Polypremum procumbens.
10 20 -






























. 40 40
3015
















S-10 10-























0 10 20 30 40 50 0 10 15 20 25
Figure 2-2. Mean species in the restoration curves ofFive most common species richness for thewoods soils














Figure 2-1. The mean number of seedlings per sample (+ SE) in the germination assay
for the five most common species in the a) restoration soils (n = 47) compared
with the same species in the flatwoods soils and b) flatwoods soils (n = 42)
compared with the same species in the restoration soils. Note different scales
and species on axess. Species codes as follows: CYSP Cyperus sp., HEUN
Hedyotis uniflora, SCDU = Scoparia dulcis, LUOC = Ludwigia octavalvis,
KYBR = Kyllinga brevifolia, ELSP = Eleocharis sp., DISP Dichanthelium
portoricense, POPR Polypremum procumbens.






50 50


S40 40







20 0 12 20 30 4-0 5-12

a b


Cumulative number of samples b Cumulative number of samples


Figure 2-2. Mean species accumulation curves of observed species richness for the
restoration and flatwoods soils in the a) germination assay and b) direct count
assay. Isolated points set to the right are the ICE incidence-based estimators of
species richness.









Species found in seedbank assays
|yw Species only found in field trial


I


0 o


Common species
Sporadic species
Unique species


0


2 3


4 5


Elevation strata

Figure 2-3. Relative abundance of species categories at the 5 elevation strata: a) species
whether found in both the seed bank assays and vegetation survey, or species
found only in vegetation survey and b) species whether common to all strata,
found in between 2 and 4 strata (sporadic species), or species unique to each
stratum.


I


I



















Mean = 2.56 m-2


1 2


Lowest <-


3

Elevation Strata


-> Highest


Figure 2-4. Mean (+1 SE) density ofAristida stricta (m-2, n=5) for each of the 5 elevation
strata six months post-seeding in the first-year restoration site.


2 -




0
0

















()
a,
0 30
U,
0_
(0)
- 25
..,
-o
E 20

c,
._ 15


E 10
0
5


a 0


(n

80
(n

.C

E 60
c,


E 40
0



b 20


S
9f


-*- Germination Assay
o-- Direct Count Assay


0.08 0.10


*


0.00 0.02 0.04 0.06

Cumulative area sampled (m2)


-- Field sampling


200 400


Figure 2-5. Comparison of the mean species accumulation curves of observed species

richness for the restoration soils in the (a) direct count and germination trials,

(b) field sampling scaled to the area of soil sampled. Isolated points set to the

right are the ICE predicted species richness. Note the difference in scale of

both axes.

























-4


Flatwoods Restoration


0
S
S




S






S
S
S
S


Flatwoods Restoration


b


Figure 2-6. Number of seeds/seedlings per m2 for a) direct count assay (n = 40), b)
germination assay (n = 89) in the flatwoods and restoration soils. Each point
represents one sample.














CHAPTER 3
VESICULAR-ARBUSCULAR MYCORRHIZAE IN UPLAND RESTORATION
SITES, MESIC PINE FLATWOODS, AND PASPALUMNOTA TUMPASTURES IN
CENTRAL FLORIDA

Introduction

Increasing efforts to restore longleaf pine systems that have been converted to non-

native pasture grasses ("improved pasture") are underway in Florida. The goal of this

research is to understand whether restoration of vesicular arbuscular-mycorrhizae (VAM)

fungi needs to accompany that of native vegetation to ensure successful establishment of

native plant species in upland restoration sites in central Florida. Understanding how

VAM fungi respond to restoration efforts and the mycorrhizal role in the establishment of

native pine flatwoods plant species such as Aristida strict (wiregrass), can improve the

success of pasture restoration efforts and facilitate future native plant restoration projects

in the state.

Vesicular-arbuscular mycorrhizae are soil-borne fungi that form symbiotic

relationships with plants. VAM fungi associated with plant roots receive carbon from the

plant while they aid plants in acquiring nutrients and water (Smith and Read 1997). This

symbiosis is especially important for immobile nutrients like phosphorus because these

fungi can bridge the phosphorus-depletion zone around plant roots and transport P to the

plant (Jakobsen 1992, Sylvia 1999). Vesicular-arbuscular mycorrhizae fungi are vital

ecological elements of ecosystems, can influence and be influenced by their aboveground

plant community (Eom et al. 2000, Hart et al. 2001), and play a crucial role in plant

community structure (van der Heidjen et al. 1998). Hart et al. (2001) suggest that VAM









fungi are phenotypically variable and that host plant identity and environmental

heterogeneity may be playing a major role in their behavior. Vesicular-arbuscular

mycorrhizae fungi also aid in soil aggregation (Smith and Read 1997).

Symbiotic relationships between plants and VAM fungi are widespread (Smith and

Read 1997). Many plant species in the pine flatwoods including Serenoa repens (saw

palmetto) (Fisher and Jayachandran 1999), A. strict (Mullahey and Speed 1991),

Andropogon virginicus var. glauca (chalky bluestem, listed as A. capillipes in Mullahey

and Speed 1991), Andropogon virginicus (broomsedge bluestem) (Mullahey and Speed

1991), Liatris tenuifolia var. laevigata (Anderson and Menges 1997), Pityopsis

graminifolia (Anderson and Menges 1997), and Balduina angustifolia (Anderson and

Menges 1997) have symbiotic associations with VAM fungi. Restoration practitioners in

Florida hope to introduce many of these native grasses and forbs to their upland

restoration sites. Such herbaceous species, which produce ground fuels for the fast

moving surface fires that are natural and essential in southeastern pine flatwoods (Pyne et

al. 1996), aid in the restoration of the structure and function of the community.

Vesicular-arbuscular mycorrhizae fungi presence in soils is closely related to soil

chemistry. Vesicular-arbuscular mycorrhizae fungi tend to be more prevalent in low

nutrient (Miller 1987) and acidic soils (Read et al. 1976), and less important in soils with

high P availability (Read et al. 1976). The soils of the pine flatwoods are both acidic and

low in nutrients (Abrahamson and Hartnett 1990).

Land use and agricultural practices can also affect VAM fungi. Soil management

such as tillage, soil disturbance, and fallow treatments decrease soil inoculum in some

cases (Kabir 1999). Additions of phosphorus fertilizers can also decrease the efficacy of









VAM fungi (Peng et al. 1993). Therefore in pastures where severe soil management has

occurred and fertilizers are abundant, the presence of VAM fungi may be reduced.

Vesicular-arbuscular mycorrhizae fungi have played an important role in land

restoration projects such as coastal dune restoration (Sylvia and Burks 1988), restoration

of mine spoil sites (Corbett 1996, Moynahan and Zabinski 2002), restoration of

Sporobolus wrightii (big sacaton) grass in the desert southwest (Richter and Stutz 2002),

and restoration of tallgrass prairie (Smith et al. 1998). Mycorrhizal fungi have been well

studied in the tallgrass prairie community, which is analogous to the pine flatwoods

understory in ecology and herbaceous species diversity and structure. The majority of

tallgrass prairie species have symbiotic relationships with mycorrhizae, especially the

warm-season grasses like Andropogon gerardii (big bluestem) and Sorghastrum nutans

(Indian grass) (Miller 1997). At a tallgrass prairie restoration site, Smith et al. (1998)

found increased percent cover and percent colonization of native species when VAM

fungi inoculum was placed below the sown seeds. Miller (1997) suggests that in mesic

sites with rich soils, the below-ground mycorrhizal changes that take place when an

agricultural field is restored back to prairie mirror the changes in the above-ground plant

community. Weedy mycorrhizal species are succeeded by a more diverse suite of non-

weedy species that already exist in the soil and do not need to be reintroduced (Miller

1997). These generalizations may not be true for more xeric, nutrient poor, or disturbed

sites (Miller 1997).

Studies in the greenhouse (Kabir 1999), prairie restoration (Smith et al. 1998), arid

regions (Allen 1989), and ultisols of Indonesia (Boddington and Dodd 2000) have shown

that mycorrhizal fungi populations are significantly reduced by soil disturbance and/or









fallow treatments. Reductions in soil inoculum could lead to early colonization,

persistence, and competition by nonmycotropic plants. This response could have serious

implications for pasture restoration in Florida since site preparation consisting of repeated

cycles of soil disturbance and fallow periods is often required to reduce populations of

non-native and weedy species (Bissett 1996, Harper-Lore 1998). Reductions in inoculum

could also impact the ability of native plant species to persist in the restoration site.

Mycorrhizae also show temporal variation in activity such as sporulation (Siguenza

et al. 1996, Eom et al. 2000). Several researchers have shown that spore production tends

to increase at the end of the growing season for the host plant (Ebbers et al. 1987,

Siguenza et al. 1996). Phenology of VAM fungi have been shown to vary with the

phenology of the plants with which they associate (Siguenza et al. 1996, Hartnett and

Wilson 1999).

During the course of pasture restoration, the floral species composition changes

from a managed perennial monoculture, to a transient community dominated by weedy

and annual species, to the target pine flatwoods natural community, characterized by

diverse, perennial, herbaceous species. Because mycorrhizae can differ in their life

history strategies (Bever et al. 2001), which can be influenced by their host plants, the

mycorrhizal community is likely to reflect changes in the aboveground vegetative

community. Therefore, in long-lived perennial dominated community types, the turnover

and sporulation of VAM fungi would be much reduced compared to that of an annually

dominated plant community (J. Graham 2002, pers. comm.). Additionally, diverse plant

communities generally support diverse VAM fungi communities (Bever et al. 2001).









It is not currently known whether the soil disturbance that is necessary for pasture

restoration in Florida alters the soil mycorrhizal community and inoculation potential as

has been shown in other studies. We also have no information on whether the suite of

mycorrhizae species present in the soils of pastures and restoration areas are beneficial to

native flatwoods plant species. If additions of mycorrhizae during restoration can increase

percent cover and give the native species in Florida a competitive advantage over re-

colonizing non-native species, the success of upland restoration efforts could be greatly

improved. Understanding whether this important component of the floral community

balance is negatively affected by restoration efforts seems vital to restoration success and

persistence of the reintroduced floral community.

In this study I examined VAM fungi in relation to soil chemistry, restoration

activities, and temporal variation in pastures, pine flatwoods, and restoration areas in

central Florida. By comparing mycorrhizae in each of three upland restoration areas with

different starting conditions, I could evaluate immediate and short-term responses to

restoration site preparation and native species planting efforts. The upland restoration

complexes were: pre-restoration improved pasture, first-year restoration (hereafter "one-

year unit") and five-years following restoration (hereafter "five-year unit") each coupled

with an adjacent pasture and pine flatwoods. I hypothesized that the one-year unit would

initially have significantly reduced MIP than the adjacent pasture and flatwoods, but

would recover with time after seeding takes place. Therefore, the five-year unit should

have higher MIP than the one-year unit, but lower MIP than the adjacent pasture and

flatwoods. Further, I hypothesized that the VAM fungal root colonization of A. strict

plants in the five-year unit would be significantly less than that of A. strict plants in









natural flatwoods. Corollary predictions based on soil chemistry and temporal variation

are as follows:

* Pastures and restoration sites, having been fertilized and limed in the past (The
Nature Conservancy 1996), would have elevated concentrations of soil nutrients
and pH relative to the flatwoods soils.

* Variations in soil chemistry, especially phosphorus and pH, would alter soil
mycorrhizae inoculation potential (MIP) and A. strict root colonization rates
especially in the restoration and pasture soils.

* I predicted that there would be temporal variation among the three sampling dates
for the MIP in the one-year complex, with higher soil inoculum in the fall and
winter months because this is the time of greatest sporulation (Siguenza et al.
1996).



Methods

I evaluated the potential of the soil to colonize plants with VAM fungi through

spores, root fragments or hyphae fragments using soil mycorrhizae inoculation potential

(MIP) tests. MIP assays are a widely used method to evaluate soil inoculum (Sylvia 1994,

Corbett 1996, Anderson and Menges 1997, Bray et al. 2003). The MIP method was

chosen because I was interested in relative differences in different soils rather than the

absolute number of propagules. I compared the MIP of soils in three restoration areas,

each to their adjacent flatwoods, and pasture. In these sites field soils were used as

inoculum in laboratory assays with host plants to estimate inoculation potential of the

soils. In Experiment 1, I evaluated the MIP of a pre-restoration area and the one-year

unit, each compared to adjacent pastures and flatwoods (Pre-restoration and First-year

restoration complexes respectively). In Experiment 2, the five-year unit allows me to

examine the longer-term effects of restoration on the MIP of the soils in restoration plots

that were sown in 1998 compared to adjacent flatwoods and pastures. I also examined the









VAM fungal colonization of A. strict in the five-year unit and adjacent flatwoods. Soil

chemistry, including nutrient content, pH, organic matter content, and soil moisture

content, were also measured at all of the sites. The soils of the preserve have been

mapped and all experiments for this study took place in Symrna or Myakka fine sands

soil series.

Experiment 1: First-year and Pre-restoration Complexes

In the one-year unit, I established an experiment to determine whether soil

disturbance during site preparation had a negative impact on the mycorrhizae inoculation

potential (MIP) of the soil. Site preparation was initiated in the 10.7 ha (26.5 acre) one-

year unit in June 2001. Glyphosate (3% concentrate) herbicide was broadcast on the

pasture grasses, followed by soil disking (6 times over 6 months) to deplete the weedy

seed bank. Seed collection (see chapter 1) occurred in November 2001. Immediately prior

to seeding (February 2002), the site was prepared with a cultipacker to compact the

disked soil. The seed was sown with a hay blower onto the prepared site. Non-native,

invasive plant species were treated with herbicides, the only management performed on

the site after seeding.

The one-year unit was divided by elevation differences into five strata (1 to 5,

lowest to highest elevation respectively). The overall elevation gradient in the restoration

plot was only 0.45 m (1.5 ft.) The stratification was done by approximation using a

topographical map and aerial photographs; 50 m X 50 m plot was marked at each

elevation strata. Five plots were established in the adjacent flatwoods and pasture at

corresponding elevations. The soils in these elevation strata were sampled several times

throughout 2002. At each sampling period, in each stratum (15 total: 5 pasture, 5

restoration, 5 flatwoods) two soil samples (12 cm deep x 8 cm diameter) were









composite and kept cold for one day during transportation back to the laboratory. I

sampled the flatwoods soils on 28 June and 5 December 2002. I sampled the restoration

area and the pasture soils on 2 March, 28 June, and 5 December 2002.

The same sampling methods were used in a second restoration site at DWP, the

pre-restoration complex. This area was scheduled for site preparation in 2002. As a result,

I could compare the pre-restoration mycorrhizal community of this site and the

experimental pasture (-9 km distant). In the pre-restoration complex, soils were collected

on 30 May 2002 to evaluate the pre-restoration mycorrhizal conditions.

The methods for the laboratory assays were from Sylvia (1994), International

Culture Collection of Vesicular Arbuscular Mycorrhizal Fungi (2002), and personal

communication with A. Alagely. For the laboratory assays, planting tubes were filled 3A

full with a sterilized soil mix composed of equal parts of sand, peat moss, and

vermiculite. The sand was treated with 25% Muriatic acid for 24 hours and then rinsed

for 24 hours with water to remove any phosphorus. A 20 ml sample of field soil from

each elevation stratum was placed on the surface of the soil mix in each of 5 tubes (5

replicates). A 10 ml layer of the soil mix was then placed on top of the field soil and three

'Silver Queen' corn seeds were placed on the soil. The corn seeds were purchased from a

local feed store and were washed with soapy water for 30 minutes to remove the

fungicide present. The seeds were covered with 15 ml of the soil mix. The sown tubes

were placed in a growth chamber with a day/night cycle of 230 C night cycle for 8 h and

a 280 C day cycle for 16 h. The light intensity inside the chamber for the day cycle was

measured with a LI-250 (Li-Cor, Lincoln, Nebraska) and was 181.5 [tmol/m2/s. The

plants were grown for 28 days (to detect only primary colonization, Sylvia 1994). They









were fertilized on the day 10, 17, and 24 post-sowing with a modified Hoaglands solution

(Table 3-1) (Sylvia et al. 2001). Occasionally, due to the vertical space limitations of the

growth chamber the tips of the corn leaves had to be trimmed off to avoid growth into the

lights.

After 28 days the corn plants were removed from the growth chamber and root

samples were taken from each plant. The root samples were placed into labeled tissue

cassettes (OmniSette tissue cassettes, Fisher Scientific, Pittsburgh, PA). Root samples

were cleared for 30 minutes in 1.8 M KOH to remove root pigments and then stained

with Trypan Blue stain to enhance mycorrhizal structures according to methods in Sylvia

(1994). Samples were stored in the refrigerator until they were examined

microscopically.

The Gridline-Intersect method was used to estimate the proportion of colonized

root in the corn plants (Sylvia 1994). Stained root samples were spread out in a petri dish

with scribed gridlines 1.27 cm (0.5 in) apart. The gridlines were scanned under a

dissecting microscope and the total number of intersections of the gridlines with roots

was counted as well as the number of roots intersected that were colonized by VAM.

Colonization was determined when arbuscules, vesicles, or spores were seen. Hyphae that

could not be traced to one of these structures were not counted. Any structures that could

not be clearly identified through the dissecting scope were examined closely with a

compound microscope. The root segment involved was mounted on a microscope slide in

Polyvinyl-Lacto-Glycerol (PVLG) for future reference (International Culture Collection

of Vesicular Arbuscular Mycorrhizal Fungi 2002), and examined with a compound









microscope. Two hundred intersections were examined and the percentage of colonized

to total intersections was calculated.

Soil chemistry was evaluated at the same locations of the mycorrhizae tests for the

one-year and pre-restoration complexes. Three randomly placed soil samples (12 cm

depth x 4 cm diameter, from each stratum in each of the three locations (restoration,

flatwoods, pasture)) were collected on 12 October 2002, air dried and sent to the

Analytical Research Lab at the University of Florida for analysis of K, extractable P, Ca,

Mg, total Kjeldahl nitrogen (TKN), and soil organic matter content (OM) (Analytical

Research Laboratory 2003). pH of the same sample locations was measured using the

methods for pH determination in water (Thomas 1996) on air dry soil. Soil samples were

collected for soil moisture content analysis on 1 February 2003. Five randomly placed

replicates within each elevation stratum were collected. Fresh mass was taken from each

sample and recorded. All samples were dried to a constant mass at 600C. Dry mass was

recorded and percent water content was calculated (% water content = mass of water/dry

mass).

Experiment 2: Five-year Complex

Six restoration sites in the five-year complex (A-F) were established in 1997

throughout the preserve in six different pastures (with differing conversion and

fertilization histories as pastures) (see Chapter 1) (The Nature Conservancy 1996).

Paspalum notatum had been reduced by multiple herbicide (3% glyphosate concentrate

(51% glyphosate, 49% inert ingredients)) applications prior to sowing with native

flatwoods seed (The Nature Conservancy 1999). At the time of my sampling, these six

restoration sites were five-years post restoration. All of the sites contained at least one A.









strict plant and high cover ofPaspalum notatum and other non-native species (see

Chapter 1).

Roots from one A. strict plant were harvested in each of the six restoration sites

and their adjacent flatwoods on 30 July 2002. Three root sub-samples from each plant

sample were separated, washed, cleared for 30 minutes in 1.8 M KOH, and stained

(Sylvia 1994). A modified magnified intersections method described by McGonigle et al.

(1990) was used to estimate percent colonization of A. strict roots by VAM fungi. In

this method, each sub-sample was spread thinly on a petri dish that was scored with a 1

cm grid. The roots were cut into one-cm segments and 30 one-cm segments were

randomly chosen from the petri dish. These segments were mounted on microscope slides

in PVLG. On each slide, two sets of five root samples were lined up parallel to the long

side of the slide and covered with cover slips, resulting in three slides for each sub-

sample. Data from the three slides were added together for each sub-sample. The slides

were examined with a compound microscope at 40x. The microscope was fitted with a

crosshair ocular that visually produced two thin crosshair lines in the field of view. The

field of view was moved perpendicular to the root segments; colonization by VAM fungi

was recorded and at each intersection of root with the center of the crosshairs

(McGonigle et al.1990). Positive colonization was only recorded if arbuscles, vesicles, or

coils were visible.

On 12 October 2002, two composite soil cores were collected in five contrasting

vegetation locations for each five-year unit for use in mycorrhizal inoculation procedure

assays. The five vegetative locations including the following: random within restoration

site, directly next to a random A. strict plant within restoration site, random in improved









pasture, random in flatwoods, and directly next to a random A. strict plant in flatwoods.

These soils were used to evaluate the mycorrhizal inoculation potential of the soil at those

sites and vegetation locations to complement the A. strict colonization data. From each

soil sample (n=30) five 20 ml samples were assayed for MIP as described for Experiment

1 (See above, Sylvia 1994)). Soil chemistry was also evaluated as described for

Experiment 1.

Data Analysis

Multiple regressions were performed on the one-year unit (June data only), pre-

restoration, five-year unit data to see if the MIP and wiregrass colonization (for five-year

unit) data were significantly related to the soil chemistry variables (SAS 1999).

Spearman's correlations were included in the regression to determine whether any of the

soil chemistry variables were correlated with each other (SAS 1999). For these analyses,

means for each elevation stratum (for pre-restoration area and one-year unit) and each

site (five-year unit) within restoration and flatwoods were used for the soil chemistry

variables and MIP data. Raw data were used for A. strict colonization. Pasture data was

not included for the analysis of the one-year unit or the five-year unit because I expected

pastures would be more affected by past fertilization activities but was included in the

pre-restoration site because the measurements represent pre-restoration conditions (i.e.,

both the restoration and the pasture were bahia grass pastures).

Experiment 1

Due to the skewness of the MIP dataset, normality was not achieved even with

various transformations. A Kruskal-Wallace non-parametric test was used on the MIP

data to determine if there were statistical differences among the elevation strata and

treatments for each sampling date ( SAS 1999). Each soil chemistry variable was either









normally distributed or transformed to approach normality. A two-way nested ANOVA

was performed on each environmental variable to determine if there were statistical

differences among the elevation strata and treatments for each of the sampling dates

(SAS 1999). For these Kruskal-Wallace tests and ANOVA's, the five replicates were

nested within elevation strata and Tukey's means comparisons (p < 0.05) were used to

determine differences among all pairs of elevations and sites.

A repeated measure ANOVA was also performed on the MIP data separately for

the pasture and restoration treatments to see if there were significant differences among

the three sampling dates (March, June, and December 2002) (SAS 1999). Since sampling

only occurred in the flatwoods in June and December 2002, I performed an one-way

ANOVA to see if date was a significant factor and used Tukey means comparisons (p <

0.05) to determine which date had higher MIP (SAS 1999).

Experiment 2

A t-test was performed on the A. strict colonization data to see if there were

significant differences between those plants growing in natural flatwoods conditions and

the restored five-year unit.

As mentioned for Experiment 1, normality was not achieved in the MIP data and

Kruskal-Wallace tests were used to test for differences in vegetation location (SAS

1999). ANOVA's were performed on each soil chemistry variable (P, K, Ca, Mg, TKN,

pH, organic matter, and percent water content) to determine if there were statistical

differences among the sites and vegetation locations (SAS 1999). Phosphorus and pH

were log transformed to improve normality. For these soil chemistry ANOVA's,

vegetation location was nested within site. Because I found no significant difference

among the sites, all sites were lumped in order to test differences among vegetation









locations. One-way ANOVA's were performed on each soil chemistry variable and a

Kruskal-Wallace test was performed on the MIP data and means comparisons were made

between all pairs of means using Tukey tests (p < 0.05) (SAS 1999).

Results

Experiment 1

I found no significant relationship between the MIP data and the soil environmental

variables in the pre-restoration complex (restoration and flatwoods only) (Table 3-2).

There were significant correlations among some of the soil chemistry variables, however

(Table 3-2). In the one-year unit MIP was only positively correlated with soil water

content and pH and many of the soil environmental variables were correlated with each

other (Table 3-3). There was a significant positive relationship between MIP and pH in

the one-year complex (restoration and flatwoods only) when restoration and flatwoods

soils were combined (Regression: F = 7.04, p < 0.0291, r2 = 0.47) (Figure 3-1).

Restoration site soil MIP was consistently higher than that of flatwoods site soil at

equivalent pH (Figure 3-1).

In the March 200 sampling of the one-year complex, I found no significant

differences in MIP between pasture and restoration site or among elevation strata. In

June, the restoration area soils had significantly higher MIP than the pasture and

flatwoods soils (Kruskal-Wallace: F = 22.88, p < 0.0001, Figure 3-2). Additionally,

elevation Stratum 2 (19.6 m) had a significantly higher MIP than did Stratum 4 (19.98 m)

(Kruskal Wallace: F = 3.37, p < 0.03, Figure 3-3) In December there was no significant

difference among the elevation strata but the pasture had significantly higher MIP levels

than the flatwoods (Kruskal-Wallace: F = 6.35, p < 0.03, Figure 3-2).









Results of the repeated measures analysis of variance indicate a significant

difference among the three sampling dates for the restoration soils (over time: F = 13.56,

p < 0.0002) but not for the pasture soils. In June, MIP of the restoration soils was

significantly higher than on the other two sampling dates (Figure 3-2). In the flatwoods

soils, there was no significant difference between the two sampling dates (Figure 3-2).

All soil environmental variables except for TKN and organic matter varied

significantly by restoration treatment (Tables 3-4, 3-5). Flatwoods soils had significantly

higher soil water content than did pasture and restoration soils (Two-way nested

ANOVA: p < 0.05, Figure 3-4). Pasture soils had significantly higher pH than restoration

and flatwoods soils (Figure 3-5). The pasture soils had the highest means among the three

treatments for all variables except for water content and organic matter; but differences

were not always significant (Figure 3-4, 3-5, 3-6, Table 3-5). Similarly, variables except

for K, pH, and organic matter varied significantly by elevation (Table 3-4). There was no

consistent relationship between elevation stratum and concentration of soil nutrients,

however, the lower elevations (strata 1 and 2) tended to have high means for soil

chemistry compared to the mid elevations (strata 3 and 4) (Table 3-6). The interaction

between treatment and elevation strata was significant for P, Ca, and TKN (Table 3-4)

with no consistent pattern (Figures 3-4, 3-6, 3-7).

Experiment 2

Aristida strict root colonization by VAM fungi was 25.8% ( 6.1 SE) in the

flatwoods and 25.5% ( 4.4 SE) in the five-year unit. However these data are extremely

variable on a per plant and per site basis (Figure 3-8).

The colonization level was dependent on none of the combinations of variables

examined. The correlation matrix among the 10 soil variables (A. strict colonization,









MIP, P, K, Ca, Mg, TKN, pH, organic matter, and water content) resulted in several

significant correlations among the variables. As seen in the one-year site, many of the

nutrients were correlated with each other. Phosphorus (p < 0.02) and Ca (p < 0.01) were

both positively correlated with organic matter. Potassium was positively correlated with

TKN (p < 0.05). Water content was positively correlated with K (p < 0.04) and TKN (p <

0.01). I similarly found no dependence of MIP on the soil chemistry variables (Table 3-

7).

Because I found no significant difference among the 6 sites when vegetation

locations were nested, location was directly tested across sites. Soil chemistry was not

significantly different across vegetation locations for Ca, pH, K, Mg, and organic matter.

Flatwoods soils had significantly higher TKN content than restoration soils (One-way

ANOVA: F = 3.22, p < 0.05, Figure 3-10) and significantly higher water content than

pasture and restoration soils (One-way ANOVA: F = 8.65, p < 0.01, Figure 3-11). Pasture

soils had significantly higher P content than the restoration area soils (One-way ANOVA:

F = 4.97, p < 0.01, Figure 3-12).

The MIP differed significantly by vegetation location, (Kruskal-Wallace: F = 3.02,

p < 0.0001; Figure 3-9) but no significant differences in MIP were found among the sites

in the five-year unit. When sites are lumped and vegetation locations are tested, the soils

from restoration wiregrass and pasture had significantly higher soil MIP than the

flatwoods wiregrass soils (Kruskal-Wallace: F = 4.56, p = 0.0017) (Figure 3-9).

Discussion

Experiment 1

Results of the MIP data coupled with the A. strict colonization rates, suggest that

inoculum is not lacking in the restoration areas at DWP and that site-preparation









activities are not resulting in long term decreases in soil inoculum. There were no

important relationships found between the MIP and soil chemistry variables tested and

variation in the soil inoculum at the dates sampled in these study sites seem to follow the

temporal variation seen in the aboveground vegetation.

As hypothesized, I found elevated levels of soil nutrients in the pasture soils.

Although not always significant, there was a general trend of the pastures having the

highest concentration of soil chemistry variables and the highest pH of the three

treatments followed by the restoration soils and then the flatwoods soils (Tables 3-4 and

3-5). There was a significant interaction between treatment and elevation strata for

several soil chemistry variables (water content, TKN, and Ca) but no consistent pattern

for the interaction. pH was one of the only factors that was correlated with MIP in the

one-year unit. While fungi are generally more dominant in acidic soils (Hartel 1999), my

MIP levels were higher in the higher pH soils (Figure 3-1). However, the small pH range

(4.1 4 .5) involved could minimize the importance of this result.

The MIP of all the sites, including flatwoods dominated by perennial native plants,

was low (Figures 3-2, 3-9), and similar to that of other studies in Florida sandhill

(Anderson and Menges 1997) and warm-season grass areas of a prairie in Illinois

(Corbett 1996). Results of the MIP data at the one-year unit suggest that mycorrhizae

recovered from the soil disturbance caused by the restoration site preparation. The VAM

fungi were not depleted by the soil disturbance and fallow treatment (Figure 3-2) as in

other studies (Smith et al. 1998, Kabir 1999). The amount and severity of soil disturbance

in our restoration methods may not be sufficient to result in long-term decreases in soil









inoculum. These results indicate that mycorrhizal inoculum may not be needed in upland

restoration efforts in central Florida.

The significant increase in the MIP of the soil at the one-year unit six months post-

sowing, however, provides no information about the species and diversity of mycorrhizae

present in that soil. Studies in many areas have recently found that VAM fungi can be

host specific (McGonigle and Fitter 1990, Hartnett and Wilson 1999, Eom et al. 2000,

Bever et al. 2001). Therefore, just having a significant increase in VAM fungi post-

restoration does not imply that reintroduced native species can utilize these beneficial

fungi. Sylvia (1986) found that a beach dune restoration site also showed an increase in

VAM fungi just after restoration. After further investigation, however, the increase was

partially due to abundance of a non-symbiotic fungus (Sylvia 1986, Sylvia and Burks

1988). In an Illinois prairie restoration, Miller (1997) found that weedy mycorrhizal

species are succeeded by a more diverse suite of species that already exist in the soil and

do not need to be introduced.

However, restoration of disturbed lands to diverse plant communities relies on the

presence and diversity (both functional and taxonomic) of VAM fungi (Bever et al.

2001). High VAM fungi can lead to high plant diversity (van der Heidjen et al. 1998).

Further work should investigate whether the VAM fungi present in our disturbed

restoration site soil are beneficial to the native plant species. Root colonization of native

desirable plant species growing in the restoration area throughout the first years of the

restoration process (compared to those growing in undisturbed natural areas) would also

be beneficial information to complement my results.









Zea mays is the standard host plant for use in MIP assays because it is very

mycotropic and the roots are cleared easily and fungal structures stain well ( International

Culture Collection of Vesicular Arbuscular Mycorrhizal Fungi 2003). However, it is

possible that the fungi colonizing the Zea mays plants in the soils I tested was only a

subset of the total fungi in the soil or that Zea mays is not the correct assay plant for these

flatwoods soils. Future studies could possibly use native plants as hosts in laboratory

assays to investigate these possibilities.

The temporal variation in the soil inoculum supports the observations that

sporulation increases in the late growing season and the fungal phenology is related to

plant species phenology (Siguenza et al. 1996, Eom et al. 2000). Although not significant

over time in the repeated measures ANOVA, I saw a trend of increasing soil MIP in the

pasture and flatwoods in the December sampling date which corresponds to the end of the

growing season for P. notatum and the perennial flatwoods species (Figure 3-2). In the

restoration soils, temporal variation was significant and MIP was highest in June, which

corresponds to the phenology of the multitude of annual species at the end of their life

cycle in the restoration area at that time (Figure 3-2).

Experiment 2

I found no significant relationship between the A. strict colonization and soil

environmental variables in the five-year unit. Mullahey and Speed (1991) similarly found

no correlation between root colonization of four Florida native grasses including A.

strict and soil nutrients. In another study, Medina et al. (1988) also found no

relationship between root colonization of tropical forage legumes and soil nutrients in

South Florida. The MIP results for the five-year complex were consistent with the one-









year complex. The flatwoods soils had low mycorrhizal inoculum while the restoration

and pastures had higher levels.

Since there was a significant difference among the vegetation locations in P

content, I expected to see that difference influence the MIP data. However, P may have

been at such low concentrations or over such a small range of values in both experiments

that the correlation could not be determined (Schwandes et al. 2001) (D. Sylvia 2003,

pers. comm.). In this study the flatwoods soils averaged 2.55 mg/kg, the restoration

averaged 2.3 mg/kg, and the pasture soils averaged 3.15 mg/kg, which is substantially

lower than the 4.7 mg/kg and 9.9 mg/kg reported for north and south Florida flatwoods

soils respectively (Schwandes et al. 2001).

The TKN content in the soils of the flatwoods was significantly higher than in the

five-year unit restoration soils where considerable soil disturbance and fallow period had

taken place. Perhaps the nitrogen in the restoration soils was leached out of the system

during site preparation because of the low nitrogen demand during the period with no or

few plants growing in the system. This result was not consistent across experiments.

Aristida strict plants growing in the restoration areas for five years reach the same

colonization rates by mycorrhizae as those growing in natural flatwoods systems. This

result suggests that the mycorrhizae present in the soils of the restoration areas can

colonize A. strict. Larger sample sizes were not possible in this experiment due to the

lack of A. strict plants in some restoration sites but would potentially clarify the

variation seen on a per plant basis. However, while colonization rates are similar we have

no information about species identity or richness and other native plants species were not

tested in this study. Further evaluation of other native plant species and VAM fungi









identity and diversity could increase our knowledge of how mycorrhizae are responding

after restoration and how well other native plants utilize the mycorrhizal species present.

This information could help explain the persistence problem in the five-year restoration

unit.

I found higher colonization rates in A. strict than has been found in other studies

of this species (Mullahey and Speed 1991, Anderson and Menges 1997). Anderson and

Menges (1997) found no colonization of A. strict by VAM fungi in a sandhill in

Highlands County, FL Mullahey and Speed (1991) collected twelve A. strict plants in

native range in South Florida (three each from four locations) and concluded an average

root colonization percentage by VAM fungi of 8%. Our average of 25% colonization is

considerably higher. Therefore, A. strict colonization rates may vary by soil or plant

community type.

The results of earlier studies at DWP indicated that diversity and abundance of

native plant species in the five-year unit decreased over the three years of post-restoration

vegetation monitoring (see Chapter 1) (The Nature Conservancy 1999). Many native

species that emerged in the restoration did not persist over time while there was an

increase in the cover of non-native species (The Nature Conservancy 1999). One possible

explanation for this lack of persistence could be the lack of the correct suite of

mycorrhizal species in the restoration soils. VAM fungi host specificity can be related to

plant species specificity (Dhillion and Zak 1993). Significant levels of host specificity

have been noted in the tallgrass prairie ecosystem (Hartnett and Wilson 1999, Eom

2000). It is therefore important to know whether the suite of VAM fungi species in pre-

restoration sites are specialists or generalists and whether the restored vegetation can









utilize the suite of VAM species present in the restoration site. Johnson et al (1991) found

that forests and old fields were dominated by different species of VAM fungi. A. strict

can utilize the species of mycorrhizae present in the restoration soils (see Experiment 2)

at DWP and persisted in the five-year restoration units but it is possible that other native

plants cannot, which could affect persistence. Testing other native plant species

colonization rates by VAM fungi at the early stages of restoration would clarify host

specificity questions further.

Conclusions/Applications

Important distinctions that can be drawn from the data in both experiments are the

differences between the MIP of the managed and unmanaged soils and of perennially and

annually dominated plant assemblages. In all experiments the native pine flatwoods soil

had lower inoculum than soils in the pastures or restoration areas, both areas under

management activities. Lower spore yields were reported in soils of native woodlands

compared to that of six agronomic crops that were planted in adjacent cleared woodlands

in Northwest Florida (Schenck and Kinloch 1980). Above ground vegetation has been

noted to influence mycorrhizal community (Bever et al. 2001) and seems to be the case

for my study as well. The native pine flatwoods is also a system dominated by long-lived

perennial plant species and therefore may not have high turnover of plants and VAM

fungi. Low incidence of sporulation by native perennial plants in England (Read et al.

1976) and in a north Florida woodland (Schenck and Kinloch 1980) have been reported.

The consistently higher MIP in the restoration soils than the perennially dominated

flatwoods in this study may, therefore, have reflected the dominance of annual plants in

the aboveground vegetation, high plant turnover, and relatively high soil inoculum.









Clearly VAM fungi inoculum is more dynamic in the restoration area than in the

native flatwoods in this central Florida site. This mirrors the substantial changes that are

occurring with the aboveground vegetation in the site. This study did not clarify whether

the suite of VAM fungal species in the restoration is appropriate for use by the native

flatwoods species being introduced in the restoration. However, there is convincing

evidence from the MIP data and the A. strict root colonization data that mycorrhizae

inoculum is not limiting in this restoration. This should come as good news to

practitioners of restoration in Florida.






63




Table 3-1. Composition of the modified Hoaglands fertilizer solution (Sylvia et al 2001).

Stock Chemical ml/L
0.01M KH2PO4 0.3
1M KNO3 1.5
NaFeEDTA 0.3
0.1M NaCl 0.45
1M Ca(N03)2 1.5
1M MgSO4 0.6
Micronutrient Stock 0.3
H3B03-2.86 g/L
MnCl2-1.81 g/L
ZnS04-0.22 g/L
CuSO4-0.08 g/L
NaMoO4-0.02 g/L

Table 3-2. Spearman correlation matrix for MIP and soil chemistry flatwoods and pre-
restoration treatments only; only p < 0.05 shown. Non-correlated variables are
indicated with dashes (--).

MIP P K Ca Mg TKN OM water
MIP -- -- -- -- -- -- -- --
P -- 0.0066 0.0191 0.0042 -- -- --
K -- 0.0066 -- 0.0001 0.0313 -- --
Ca -- 0.0191 -- -- -- -- -- --
Mg -- 0.0042 0.0001 -- 0.0300 -- --
TKN -- 0.0313 -- 0.0300 -- 0.0220 --
OM -- -- -- 0.0220 -- --
water -- -- -- -- -- -- -- --












Table 3-3. Spearman correlation matrix for MIP and soil chemistry variables in the one-year complex (restoration and flatwoods only;
significant p-values shown) in June. Non-correlated variables are indicated with dashes (--).

MIP P K Ca Mg TKN pH OM water
MIP -- 0.0089 -- 0.0245
P -- 0.0061 -- 0.0111 -- 0.0072 --
K -- 0.0061 ---- -- --
Ca -- -- 0.0008 -- -- 0.0005
Mg -- -- 0.0008 -- -- 0.0022
TKN -- 0.0111 -- -- -- 0.0392 --
pH 0.0089 -- -- --
organic matter -- 0.0072 -- -- 0.0392 --
water content 0.0245 -- -- 0.0005 0.0022 -- -- -- --


Table 3-4. Results of two-way nested ANOVA for each soil chemistry variable on restoration treatment (pasture, restored pasture,
native flatwoods), elevation (5 strata) and their interaction (Treatment x Elevation) within the one-year complex. Non-
significant tests are indicated with dashes (--).


Treatment Elevation Treatment x Elevation
Variable df F-value P-value df F-value P-value df F-value P-value
Water content 2 21.99 0.0001 4 5.88 0.0027 8 4.44 0.0032
P 2 5.12 0.0160 4 5.97 0.0101 8
K 2 7.07 0.0048 4 -- 8 -
Ca 2 38.99 0.0001 4 10.36 0.0014 8 4.49 0.0151
Mg 2 11.25 0.0005 4 4.64 0.0223 8 --
TKN 2 -- 4 3.99 0.0345 8 3.52 0.0333
pH 2 50.31 0.0001 4 -- 8 -
Organic matter 2 -- -- 4 -- --8 -- --









Table 3-5. Results of Tukey mean comparisons of the three treatments in Experiment 1
for each significant variable. Treatments for each variable are listed from
highest to lowest mean (left to right respectively). Different letters indicate
significant differences (p < 0.05).


Variable Treatments
water content Flatwoodsa Pasture Restoration
P Pasturea Flatwoodsab Restorationb
K Pasturea Restorationb Flatwoodsb
Ca Pasturea Restorationa Flatwoodsb
Mg Pasturea Restorationab Flatwoodsb
pH Pasturea Restorationb Flatwoodsb


Table 3-6. Results of the Tukey mean comparisons of the five elevation strata in
Experiment 1 for significant variables. Elevation strata (numerically from
lowest to highest elevation) for each variable are listed from highest to lowest
mean (left to right respectively). Different letters indicate significant
differences (p < 0.05).


Variable
water content
P
TKN
Ca
Mg


Elevation Strata
5ab 2b 4b
lab 4ab 3b
5ab 4ab 3b
5ab 3b 4b
5ab 3ab 4b


Table 3-7. P-values from the Spearman's correlation matrix of eight soil variables in
Experiment 2 (significant p-values shown). Non-correlated variables are
indicated with dashes (--).

MIP P K Ca Mg TKN pH OM water
content
M IP ....... .
P -- -- -- -- 0.0202 --
K -- -- -- 0.0548 -- 0.0415
Ca -- -- -- -- 0.0102 --
Mg -- -- -- -- -- -- -- -- --
TKN -- 0.0548 -- -- -- 0.0139
pH -- 0.0202 -- -- -
OM -- -- 0.0102 -- -- --
Water -- 0.0415 -- 0.0139 -- --
content














SFlatwoods
Restoration



p < 0.0291
R2 = 0.47


Figure 3-1. Correlation between MIP (n = 5) and pH (n
one-year unit restoration soils in June 2002.





14

12 a

10 -
S10

o A
a. 8


CU 6 -
3 b
0

CR




Restoration Pasture


3) for soils from flatwoods and


- March
--- IJune
December


Flatwoods


Treatments


Figure 3-2. Mean mycorrhizal inoculation potential (+1 SE, n = 25) of the one-year unit,
and adjacent flatwoods and pasture. Different letters indicate significant
differences and capital and lower case letters indicate separate univariate tests.







































Elevation strata



Figure 3-3. Mean inoculation potential (+1 SE, n = 15) of the 5 elevation strata in June
2002 (1 = lowest, 5 = highest) for restoration, flatwoods and pasture together
in the Experiment 1. Different letters indicate significant differences.


-0- Flatwoods
-0- Pasture
SRestoration


0.40
E
C 0.35
8
0

0
) 0.30
E

- 0.25
E

0 0.20


0.15


2 3
Elevation Strata


LOWEST


Figure 3-4. Mean gravimetric soil moisture content (+1 SE, n
elevation strata in the one-year complex.


5) for each treatment by


0.45
















-- Flatwoods
-0- Pasture
--- Restoration


Evelation strata


Figure 3-5. Mean pH (+1 SE, n
complex.


0)
133
E 3.5

-1-
1 3.0-
o
0
C-
. 2.5
0'
2.0-


1.5


1.0


3) for each treatment by elevation in the one-year


- Flatwoods
-4 Pasture
- Restoration


3 4

Blocks by elevation


Figure 3-6. Mean soil P content (mg/kg) (+1 SE, n=3) for each treatment in the one-year
complex.


7-





6-





I 5-





4-





3 -














-- Flatwoods
-0- Pasture
-y- Restoration


Blocks by elevation


Figure 3-7. Mean soil total Kjeldahl
the one-year complex.


60


50


40
N
0
o 30
20
a 20


nitrogen (g/kg) (+1 SE, n=3) for each treatment in


SRestoration area
I I Flatwoods


A B C D E F


Sites


Figure 3-8. Percent colonization (n = 1) ofAristida strict roots by VAM fungi in the six
restoration sites and adjacent flatwoods in the Experiment 2.















7-


6-


5-


S4-


r 3-


2-


1-


0-


Flatwoods Flatwiregrass Pasture Restoration Restwiregrass


Treatments



Figure 3-9. Mean inoculation potential (+1 SE, n = 30) of the soils in the five treatments
of the five-year complex. Different letters indicate significant differences
among the five treatments.


2.5



2.0


--
1.5
z
I-
0
o



0.5



0.0


Flatwoods


Pasture

Treatments


Restoration


Figure 3-10. Mean soil total Kjeldahl nitrogen (+1 SE, n = 18) of the restoration and
adjacent flatwoods and pasture soils in Experiment 2. Different letters indicate
significant differences.






















S 20
0
E
=0 15
0)


E 10
(0


Flatwoods


Pasture


Restoration


Treatments


Figure 3-11. Mean soil gravimetric moisture content (+1 SE, n = 30) of the restoration
and adjacent flatwoods and pasture soils in Experiment 2. Different letters
indicate significant differences.


4 a




ab

) b
E
a_
-2
2 -

X
LU
o
0
U)
I 1


0 ----------------------


Flatwoods


Pasture


Restoration


Treatments


Figure 3-12. Mean soil extractable P content (+1 SE, n = 18) of the restoration and
adjacent flatwoods and pasture soils in Experiment 2. Different letters indicate
significant differences.















APPENDIX
SPECIES LISTS













Table A-1. Species list for the germination assay with life form and category.
Flatwoods Elevation Strata Restoration Elevation Strata
species Life form category 1 2 3 4 5 mean 1 2 3 4 5 mean


Andropogon sp.
Chamaecrista nictatans
Cyperus surinamensis
Dichanthelium sp.
Eleocharis viviparous
Hedyotis uniflora
Lindernia sp.
Ludwigia maritima
Ludwigia repens
Polypremum procumbens
Toxicodendron radicans
Dichanthelium portericense
-a Crotolaria rotundifolia
Cyperus compressus
Cyperus polystachyos
Drosera brevifolia
Eleocharis filiform
Rhynchosporafascicularis
Scleria reticularis
Scleria sp.
Xyris like
Gnaphalium sp.
Ludwigia octavalvis
Oxalis corniculata
Scoparia dulcis


grass
forb
forb
grass
sedge
forb
forb
forb
forb
forb
forb
grass
forb
sedge
sedge
forb
sedge
sedge
sedge
sedge
sedge
forb
forb
forb
forb


0 0 0.17 0.1 0.22 0.10
0 0 0 0 0 0.00
0 0.13 0 0.2 0 0.07
0.67 0.75 0.67 0.7 0.56 0.67
0 0 0 0.1 0.11 0.04
0.78 0.63 0.5 0.6 0.44 0.59
0 0 0 0 0 0.00
0 0 0 0 0.11 0.02
0 0 0 0 0 0.00
0 0.25 0 0 0 0.05
0 0 0.17 0 0 0.03
0.22 0 0.33 0.1 0.33 0.20
0 0 0 0 0.11 0.02
0.11 0.13 0 0.1 0.22 0.11
0 0.13 0.17 0 0.22 0.10
0 0 0.33 0 0 0.07
0.89 0.25 1 0.2 0.56 0.58
0.11 0 0 0 0 0.02
0.11 0 0 0 0 0.02
0 0 0.17 0.1 0 0.05
0.11 0 0.5 0 0 0.12
0.22 0.38 0.17 0.2 0.33 0.26
0.11 0.13 0 0 0 0.05
0.44 0 0.5 0.2 0.11 0.25
0 0.13 0 0 0.11 0.05


0 0 0.2 0
0 0.11 0 0
0 0 0 0.1
0 0.11 0.1 0
0 0 0 0
0.3 0.56 0.8 0.2
0 0.11 0.2 0.2
0 0 0 0
0.2 0 0.1 0.1
0.2 0.44 0.7 0.2
0 0 0 0
0 0 0 0.1
0 0 0 0
0.5 0.89 0.6 0.9
0.6 1 0.8 0.8
0 0 0 0
0 0.11 0.4 0.3
0 0 0 0
0 0 0 0
0 0 0 0
0.1 0 0 0
0 0.11 0.2 0
0.1 0 0 0.1
0.3 0.33 0.2 0.3
0 0.33 0.6 0.2


0
0
0
0.13
0
0
0.38
0
0.13
0.25
0
0
0
0.88
1
0
0
0
0
0
0
0.38
0.13
0.5
0.38
o















0.38


0.04
0.02
0.02
0.07
0.00
0.37
0.18
0.00
0.11
0.36
0.00
0.02
0.00
0.75
0.84
0.00
0.16
0.00
0.00
0.00
0.02
0.14
0.07
0.33
0.30












Table A-1. Continued
Flatwoods Elevation Strata Restoration Elevation Strata
species Life form category 1 2 3 4 5 mean 1 2 3 4 5 mean
Cyperus glogulosus sedge nw 0 0 0 0 0 0.00 0 0 0 0.1 0 0.02
Cyperus retrorsus sedge nw 0 0 0.17 0.1 0 0.05 0.2 0.22 0.4 0.4 0.25 0.29
Erechtites hieracifolia forb nw 0 0 0 0 0 0.00 0 0 0 0 0.13 0.03
Fimbristylis autumnalis sedge nw 0.11 0.13 0 0 0.11 0.07 0.2 0.11 0.1 0.4 0.63 0.29
Fimbristylis dichotoma sedge nw 0 0 0.17 0 0 0.03 0.3 0.11 0.2 0.4 0.13 0.23
Kyllinga brevifolia sedge nw 0.11 0 0.17 0 0 0.06 0.4 0 0.2 0.4 0.13 0.23
Hedyotis corymbosa forb nnw 0 0 0 0 0 0.00 0.1 0 0 0 0.25 0.07
Kummerowia striata forb nnw 0 0 0 0 0 0.00 0.1 0 0 0 0 0.02
Murdannia nudiflora forb nnw 0 0.13 0 0 0 0.03 0 0.11 0.5 0.1 0.25 0.19
Desmodium triflorum forb nnw 0 0 0.17 0 0 0.03 0.2 0.22 0.1 0.2 0.13 0.17
Fimbristylis schoenoides sedge nnw 0 0 0 0 0 0.00 0.3 0 0 0.1 0 0.08
Asteraceae basal rosette forb 0 0 0 0.1 0 0.02 0 0 0 0 0 0.00
Ludwigia round leaf forb 0.11 0.13 0.33 0.1 0 0.13 0.1 0 0.1 0 0 0.04
Ludwigiasp. forb 0 0 0 0 0 0.00 0.1 0.22 0.2 0.3 0.13 0.19
Cyperus sp. sedge 0.33 0.13 0.17 0 0 0.13 0.1 0 0 0 0 0.02
# flats destroyed 1 2 4 0 1 1.60 0 1 0 0 2 0.60
Categories include: native characteristic = nc, native weedy = nw, non-native weedy = nnw, non-native invasive = nni (D. Gordon, unpublished data). Species
without a category were not identified clearly enough to determine which category they belong to. Native species are list first.










Table A-2. Species list for the direct count assay with life form and category


Restoration


species

Caryophylaceae/Portulaca/
Chenopod
Dichanthelium
portoricense
Fimbristylis autumnalis
legume small tan
Paspalum/panicum round

Poaceae huge seed
Poaceae long awn
Rhynchospora
microcephala?
Rhynchospora sp.
Rubiaceae
Scleria reticularis
Scleria sp.
unknown


life
form
forb

grass

sedge
forb
grass

grass
grass
sedge

sedge
forb
sedge
sedge


category species

Cyperus compressus

nc Cyperus globulosus

nw Cyperus polystachyos
Desmodium triflorum
Dichanthelium
portoricense
Fimbristylis autumnalis
Fimbristylis dichotoma
nc Fimbristylis
schoenoides?
nc Kummerowia striata
Kyllinga brevifolia
nc legume small tan
nc Rhynchospora sp.
Rhynchospora tracyii?
Scleria reticularis
unknown


life
form
sedge

sedge

sedge
forb
grass

sedge
sedge
sedge

forb
sedge

sedge
sedge
sedge


category

nc

nw

nc
nnw
nc

nw
nw


nnw
nw

nc
nc
nc


Categories include: native characteristic = nc, native weedy = nw, non-native weedy = nnw, non-native
invasive = nni (D. Gordon, unpublished data). Species without a category were not identified clearly
enough to determine which category they belong to.


Flatwoods











Table A-3. Species list for the field vegetation sampling by elevation strata (n=5 strata) and frequency by plot size.
Elevation Strata- 1 1 1 2 2 2 3 3 3 4 4 4 5 5 5 mean mean mean
life category 0.2x0.2 lxl 5x5 0.2x0.2 lxl 5x5 0.2x0.2 lxl 5x5 0.2x0.2 lxl 5x5 0.2x0.2 lxl 5x5 .2x.2 lxl 5x5


Species form
Aster fistifolius Forb
Centella asiatica Forb
Chamaecrista Forb
nictatans
Crotolaria Forb
rotundifolia
Diodia virginiana Forb
Euthamia minor Forb
Hedyotis uniflora Forb
Hydrocotyle Forb
umbellata
Hypericum Forb
cistifolium
Hypericum Forb
tetrapetalum
Lachnanthes Forb
caroliniana
Lindernia group? Forb
Ludwigia arcuata Forb
Ludwigia maritima Forb
Ludwigia repens Forb
Phyla nodiflora Forb
Phytolaca Forb
americana
Pityopsis Forb
graminifolia
Pluchea odorata Forb
Pluchea rosea Forb
Polygonum Forb
punctatum


0 0 0 0 0 0.2 0 0 0 0 0 0 0 0 0 0 0 0.04
0.6 0.6 0.6 0.4 0.6 1 0 0.2 0.6 0 0.2 0.8 0 0 0 0.2 0.32 0.6
0 0 0.6 0.4 0.6 1 0 0.8 1 0.4 0.8 1 0.2 1 1 0.2 0.64 0.92


nc 0 0 0 0


0 0.2 0 0.2 0.4 0 0 0.2 0 0 0 0 0.04 0.16


0 0.2 1 0.4 0.4 0.8 0 0 0 0 0.2 0.4 0.2 0.2 0.4 0.1 0.2 0.52
0 0 0.2 0 0 0 0 0 0.4 0 0 0 0 0 0 0 0 0.12
0.6 1 1 0.2 0.6 0.8 0 0.2 0.4 0.2 1 1 0.2 0.6 0.8 0.2 0.68 0.8
0 0.4 0.8 0.2 0.2 0.4 0 0.2 0.2 0 0 0.2 0.4 0.4 0.6 0.1 0.24 0.44


nc 0 0 0.2 0

nc 0 0 0.2 0

nc 0 0 0.2 0


0 0 0

0 0 0

0 0 0


0 0 0 0 0 0 0 0 0 0 0.04

0 0 0 0 0 0 0 0 0 0 0.04

0 0 0 0 0 0 0 0 0 0 0.04


0 0 0 0 0 0 0 0 0 0 0.2 0.2 0 0 0 0 0.04 0.04
0 0 0.2 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.04
0 0 0.4 0 0.2 1 0 0.4 0.6 0 0.2 0.2 0 0.2 1 0 0.2 0.64
0 0.6 1 0 0.6 0.6 0 0 0 0 0.4 0.8 0 0.6 0.8 0 0.44 0.64
0 0 0 0 0 0.2 0 0 0 0 0 0 0 0.2 0.2 0 0.04 0.08
0 0 0 0 0.2 0.4 0 0 0.2 0 0 0 0 0 0.4 0 0.04 0.2


nc 0 0 0 0


0 0 0.2 0
0 0 0.4 0
0 0 0 0


0 0 0

0 0 0
0 0 0
0 0.2 0


0 0 0 0 0.2 0 0 0 0 0 0.04


0 0
0 0
0 0


0 0 0.2 0 0 0 0 0 0.08
0 0 0 0 0 0 0 0 0.08
0 0 0 0 0 0 0 0 0.04











Table A-3. Continued
Elevation Strata- 1 1 1 2 2 2 3 3 3 4 4 4 5 5 5 mean mean mean
life category 0.2x0.2 lxl 5x5 0.2x0.2 lxl 5x5 0.2x0.2 lxl 5x5 0.2x0.2 lxl 5x5 0.2x0.2 lxl 5x5 .2x.2 lxl 5x5


Species form
Polypremum Forb nc
procumbens
Rhexia nashii Forb nc
Sabatia griaidrlora Forb nc
Amphicarpum grass nc
muhlenbergianum
Andropogon
glomeratus grass nc
glaucopsis
Andropogon
glomeratus grass nc
hirsuitior


Andropogon grass nc
virginicus glauca
Aristida beyrichiana grass nc
Axonopusfurcatus grass nc
Dichanthelium
ensifolium grass nc
Dichanthelium grass nc
portoricense
Leersia hexandra grass nc
Paspalum laeve grass nc
Paspalum setaceum grass nc
Cyperus compressus sedge nc
Cyperus haspan sedge nc
Cyperus sedge nc
surinamensis
Eleocharis filiform sedge nc


0.2 1 1 0.6 0.8 1 0.6 1 1 0.6 0.8 1 0.2 1 1 0.4 0.92 1


0 0 0.2 0
0 0 0.2 0
0 0.2 0.4 0


0 0.2 0.4 0


0.2 0.2 0.4 0

0 0 0 0

0 0 0 0


0 0 0
0 0 0
0 0 0


0 0 0


0 0 0

0 0 0

0 0 0


0 0
0 0
0 0


0.4 0.8 0.8 0 0.8 1 0.4 0.6 1
0 0.2 0.4 0 0 0.2 0 0 0


0 0 0 0
0 0 0 0


0 0 0 0 0 0 0 0 0.04
0 0 0 0 0 0 0 0 0.04
0 0 0.2 0 0 0 0 0.04 0.12


0 0 0 0 0 0 0 0.04 0.08


0 0.2 0.4 0 0 0 0 0.08 0.16

0 0 0.4 0 0 0 0 0 0.08

0 0 0.4 0 0 0 0 0 0.08

0 1 1 0.4 1 1 0.2 0.84 0.96
0 0 0.2 0 0 0 0 0.04 0.16


0 0 0 0 0 0 0 0.4 0 0 0 0 0.04 0.08
0 0 0 0.2 0.2 0.2 0.4 0.4 0 0 0.4 0 0.12 0.2


0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.2 0 0 0.04
0 0 0 0 0 0 0 0 0.2 0 0 0 0 0 0 0 0 0.04
0 0 0.2 0 0 0.2 0 0 0 0 0 0 0 0 0 0 0 0.08
0.2 0.2 0.4 0.4 1 1 0.6 0.8 1 0.4 0.8 1 0.6 1 1 0.4 0.76 0.88
0 0 0.4 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.08
0 0 0.6 0 0 0.4 0 0 0 0.2 0.6 0.6 0 0 0 0 0.12 0.32


0 0.6 1 0.2 0.2 0.6 0 0 0 0.2 0.4 0.6 0 0 0 0.1 0.24 0.44


Andropogon
virginicus


grass nc


0 0.6 1 0.2 0.2 0.6 0


0 0 0.2 0.4 0.6 0 0 0 0.1 0.24 0.44











Table A-3. Continued
Elevation Strata- 1 1 1 2 2 2 3 3 3 4 4 4 5 5 5 mean mean mean
life category 0.2x0.2 lxl 5x5 0.2x0.2 lxl 5x5 0.2x0.2 lxl 5x5 0.2x0.2 lxl 5x5 0.2x0.2 lxl 5x5 .2x.2 lxl 5x5


Species
Fimbristylis
caroliniana
Fimbristylis
puberula
Fimbristylis
tomentosa?
Rhynchospora
fascicularis
Rhynchospora
microcephala
Rhynchospora
rariflora


form
sedge nc

sedge nc

sedge nc

sedge nc

sedge nc

sedge nc


Scleria retucularis sedge nc
Scleria sp. sedge nc
Scleria triglomerata sedge nc
Xyris caroliniana sedge nc
Xyris sp. sedge nc
Circium nutallii Forb nw
Commelina dittl.sa Forb nw
Erechtites forb nw
hieracifolia
Eupatorium forb nw
capillifolium
Ludwigia octavalis Forb nw
Oxalis corniculata Forb nw
Portulaca amilis Forb nw
Scoparia dulcis Forb nw
Sesbannia herbacea Forb nw
Digitaria serotina grass nw
Cyperus globulosus sedge nw


0 0 0 0

0 0 0 0


0 0 0

0 0 0


0 0 0 0 0.6 0.6 0


0 0 0.4 0

0.2 0.4 0.6 0

0 0.2 0.2 0


0 0 0

0 0 0

n0 n


0.2 0.6 1 0 0 0 0
0 0.2 0.2 0 0 0 0
0 0.2 0.2 0 0 0 0
0 0 0.4 0 0 0 0
0 0 0.2 0 0 0 0
0 0.2 0.4 0 0.2 0.4 0
0 0 0 0 0 0.2 0
0 0 0.6 0 0 0.2 0


0.2 0.4 1


0 0.4 0.8 0.2


0.4 0.8 1 0.2 0.4 0.8 0
0 0.2 0.2 0 0 0 0
0 0 0 0 0 0 0
0.4 0.8 0.8 1 1 1 0.6
0 0 0 0 0 0 0
0.4 0.6 0.8 0 0.4 0.8 0.2
0 0 0.4 0 0.2 0.4 0.2


0 0 0 0.2 0.2 0 0 0 0 0.04 0.04

0 0 0 0.4 0.8 0 0.2 0.2 0 0.12 0.2

0 0 0 0 0.6 0 0 0 0 0.12 0.24

0 0 0 0 0 0 0 0 0 0 0.08

0 0 0 0 0 0 0 0 0 0.08 0.12

0 0 0 0 0 0 0 0 0 0.04 0.04


0 0
0 0
0 0
0 0
0 0
0 0
0 0.4
0.2 0.4

0.8 1

0 1
0 0.2
0 0
0.8 1
0 0
0.6 0.8
0.4 0.8


0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0


0 0 0 0.12 0.2
0.2 0.2 0 0.08 0.08
0 0 0 0.04 0.04
0 0 0 0 0.08
0 0 0 0 0.04
0 0 0 0.08 0.16
0 0.2 0 0 0.16
0 0 0 0.04 0.24


0 0.2 1 0.2 0.8 1 0.1 0.52 0.96


0.6 0.8
0.2 0.4
0 0.2
0.8 1
0 0
0 0.6
0.4 0.4


0.2 0.4 0.2 0.4 0.8
0.4 0.8 0 0.16 0.32
0 0.2 0 0 0.08
1 1 0.5 0.88 0.96
0 0 0 0 0
0.6 1 0.2 0.44 0.8
0.2 0.6 0 0.24 0.52











Table A-3. Continued
Elevation Strata--
life catego
Species form
Cyperus polystachos sedge nw
Cyperus retrorsus sedge nw
Fimbristylis sedge nw
autumnalis
Fimbristylis sedge nw
dichotoma
Kyllinga brevifolia sedge nw
Cuphea Forb nnw
CL, i/ i .^. ,1 1i
Desmodium Forb nnw
triflorum
Hedyotis corymbosa Forb nnw
Indigofera hirsuta Forb nnw
Kummerowia striata Forb nnw
Macroptilium Forb nnw


Forb nnw


Solanum viarum Forb nnw
Eleusine indica grass nnw
Paspalum urvellii grass nnw
Sacciolepis indica grass nnw
Fimbristylis sedge nnw
shoenoides
Cynodon dactylon grass nni
Panicum repens grass nni
Paspalum notatum grass nni
Bulbostylis sp. sedge


1 1 1 2 2 2 3 3 3 4 4 4 5 5 5 mean mean mean
y 0.2x0.2 lxl 5x5 0.2x0.2 lxl 5x5 0.2x0.2 lxl 5x5 0.2x0.2 lxl 5x5 0.2x0.2 lxl 5x5 .2x.2 lxl 5x5

0.2 0.6 0.8 0.8 1 1 0.6 1 1 1 1 1 0.8 1 1 0.7 0.92 0.96
0 0.2 0.6 0 0.4 0.8 0 0.6 0.8 0.2 0.6 0.8 0.2 0.8 0.8 0.1 0.52 0.76
0 0 0.2 0 0 0 0 0 0 0 0 0 0 0 0.2 0 0 0.08

0.2 0.6 1 0.8 1 1 0.4 0.8 0.8 0.2 0.8 0.8 0.4 0.8 1 0.4 0.8 0.92

0 0.2 0.2 0 0.4 0.4 0.2 0.4 0.4 0.2 0.8 0.8 0 0 0.2 0.1 0.36 0.4
0 0 0 0 0 0 0 0 0 0 0 0.4 0 0 0 0 0 0.08


0.2 0.6 0.8 0.2 0.8 1 0.6 0.8 1


0 0 0 0.4 0.6 0.8 0.3 0.56 0.72


0 0 0 0.2 0.2 0.2 0 0.2 0.2 0 0.2 0.6 0 0 0.4 0 0.12 0.28
0 0 0 0 0 0 0 0 0 0 0 0 0 0.2 0.2 0 0.04 0.04
0.4 0.8 1 0.2 1 1 0.2 0.8 1 0.4 0.8 1 0 0.4 0.6 0.2 0.76 0.92
0 0 0 0 0 0.2 0 0 0.4 0 0 0.2 0 0 0 0 0 0.16


0 0.2 0.6 0.2 0.4 0.8 0.4 1 1


0 0 0.2 0
0 0.2 0.2 0
0.2 0.2 0.6 0
0 0.6 1 0
0.8 1 1 0


0 0.2 0
0 0.2 0
0 0 0
0 0.2 0
0 0 0


0 0.4 0.6 0.4 0.8 0.8 0.2 0.56 0.76


0 0.2 0 0 0.4 0 0 0.2 0 0 0.24
0 0.2 0 0 0.2 0 0.2 0.2 0 0.08 0.2
0 0.2 0 0 0 0 0 0 0 0.04 0.16
0 0.2 0 0 0 0 0 0 0 0.12 0.28
0 0 0 0 0 0 0 0 0.2 0.2 0.2


0 0 0 0 0 0 0 0 0.2 0 0 0 0 0 0 0 0 0.04
0 0 0 0 0 0 0 0 0 0 0 0.2 0 0 0.2 0 0 0.08
0 0.2 0.8 0 0.2 0.6 0.2 0.2 0.8 0 0.2 0.8 0.2 0.4 1 0.1 0.24 0.8
0 0 0.2 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.04


Life form and category included for each species where possible (native characteristic = nc, native weedy = nw, non-native weedy = nnw, non-native
invasive = nni (D. Gordon, unpublished data)). Native species are listed first.


lathroides
Murdannia
nudiflora









Table A-4. Species observed in fifteen healthy pine flatwoods surveyed at DWP in 1997


Acalypha graciliens
Agalinis sp.
Amphicarpum
muhlenbergianum
Andropogon brachystachys
Andropogon glomeratus
glaucopsis
Andropogon glomeratus
glomeratus
Andropogon glomeratus
hirsutior
Andropogon gyrans gyrans
Andropogon ternarius
Andropogon virginicus
Andropogon virginicus
brachystachys
Andropogon virginicus
glaucus
Andropogon virginicus
virginicus
Aristida spiciformis
Aristida stricta
Asclepias pedicellata
Asimina reticulata
Aster tortifolius
Axonopusfurcatus
Balduina angustifolia

Befaria racemosa

Bigelowia nudata/Balduina
uniflora
Bulbostylis stenophylla
Carphephorus carnosus
Carphephorus corymbosus
Carphephorus paniculatus
Chamaecrista nictitans
Chamaecrista sp.
Chapmannia floridana
Chaptalia tomentosa
Circium horridulum
Cnidoscolus stimulosus
Composite unknown
Crotolaria rotundifolia
Ctenium aromaticum


Elephantopus elatus
Elephantopus sp.
Erigeron vernus

Eupatorium morhii
Eupatorium rotundifolium

Euphorbia polyphylla

Euthamia minor

Fimbristylis puberula
Galactia elliottii
Galactia sp.
Gaylussacia dumosa

Gaylussacia fondosa nana

Gratiola hispida

Gymnopogon chapmanianus
Hedyotis procumbens
Hedyotis uniflora
Hypericum cistifolium
Hypericum fasciculatum
Hypericum hypericoides
Hypericum
myrtifolium/tetrapetalum
Hypericum reductum

Hypericum tetrapetalum

Hypoxisjuncea
Ilex glabra
Juncus dichotomus
L // h 11/uthe1' caroliniana
Lachnocaulon beyrichiana
Lachnocaulon anceps
Lechea torreyi
Liatris laevigata
Liatris tenuifolia
Licania michauxii
Ludwigia maritima
Ludwigia not peruviana
Ludwigia sp


Pinus palustris
Pityopsis graminifolia
Pluchea rosea

Pluchea sp.
Polygala rugelii

Polygala setacea

Pterocaulon virgatum

Quercus chapmanii
Quercus minima
Quercus myrtifolia
Quercus virginianum

Rhexia mariana

Rhexia nutallii

Rhus copallina
Rhynchospora fernaldii
Rhynchospora cephalantha
Rhynchospora fascicularis
Rhynchospora fernaldii
Rhynchospora filiform
Rhynchospora intermedia

Rhynchospora
microcephala
Rhynchospora sp.

Rubus trivialis
Sabalpalmetto
Schizacharium scoparium
Scleria pauciflora
Scleria sp.
Scleria triglomerata
Sesbania vesicaria
Setaria geniculata
Solidagofistulosa
Solidago strict
Sorghastrum secundum
Styllingia sylvatica
S) g I 11at ll u/ Jflavidulus









Table A-4. Continued
Cuthbertia striped leaves Lygodesmia aphylla Tephrosia sp.
Cyperus globulosus Lyonia fruiticosa Utricularia sp.
Cyperus nashii Lyonia lucida Vaccinium myrsinites
Cyperus polystachyos Myrica cerifera Vitus rotundifolia
Cyperus retrorsus Opuntia sp. Xyris brevifolia
Desmodium incanum Panicum anceps Xyris caroliniana
Dichanthelium ensifolium Panicum longifolium Xyris difformis curtissii
Dichanthelium ensifolium var. Paspalum laeve Xyris elliottii
ensifolium
Dichanthelium portoricense Paspalum setaceum
Dichanthelium strigosum Phoebanthus grandiflora
Eleocharis filiform Physalis sp.
Eleocharis viviparous Pinus elliottii
Reprinted with permission from The Nature Conservancy 1997. Fourth Annual Monitoring Report. The
Nature Conservancy, Kissimmee, Florida. Table 4, Page 31.
















LIST OF REFERENCES


Abrahamson, W. G., and D. C. Hartnett. 1990. Pine flatwoods and dry prairies. Pages
103-149 in R. L. Myers and J. J. Ewel, editors. Ecosystems of Florida. University
of Central Florida Press, Orlando, Florida.

Allen, E. B. 1989. The restoration of disturbed arid landscapes with special reference to
mycorrhizal fungi. Journal of Arid Environment 17:279-286.

Analytical Research Laboratory. 2003. Analytical Research Laboratory, University of
Florida, Institute of Food and Agriculture Sciences URL: http://arl.ifas.ufl.edu/,
January 2003.

Anderson, R. C., and E.S. Menges. 1997. Effects of fire on sandhill herbs: nutrients,
mycorrhizae, and biomass allocation. American Journal of Botany 84:938-948.

Baker, H. G. 1989. Some aspects of the natural history of seed banks. In M. A. Leck, V.
T. Parker, and R. L. Simpson, editors. Ecology of soil seed banks. Academic
Press, Inc., San Diego, California, USA.

Bekker, R. M., G. L. Verweij, R. E. N. Smith, R. Reine, J. P. Bakker, and S. Schneider.
1997. Soil seed banks in European grasslands: does land use affect regeneration
perspectives? Journal of Applied Ecology 34:1293-1310.

Bever, J. D., A. P. Schultz, A. Pringle, and J. P. Morton. 2001. Arbuscular mycorrhizal
fungi: More diverse than meets the eye, and the ecological tale of why.
BioScience 51:923-931.

Bissett, N. J. 1996. Upland restoration challenge: Direct seeding of wiregrass and
associated species. The Palmetto Summer:8-11.

Boddington, C. L., and J. C. Dodd. 2000. The effect of agricultural practices on the
development of indigenous arbuscular mycorrhizal fungi. I. Field studies in an
Indonesian ultisol. Plant and Soil 218:137-144.

Brady, N. C., and R. R. Weil. 2002. The Nature and Properties of Soils, Thirteenth
edition. Prentice Hall, Upper Saddle River, New Jersey.

Bray, S. R., K. Kitajima, and D. M. Sylvia. 2003. Mycorrhizae differentially alter growth,
physiology, and competitive ability of an invasive shrub. Ecological Applications
13(3):565-574.










Brown, C. S., and R.L. Bugg 2001. Effects of established perennial grasses on
introduction of native forbs in California. Restoration Ecology 9:38-48.

Brown, D. 1992. Estimating the composition of a forest seed bank: a comparison of the
seed extraction and seedling emergence methods. Canadian Journal of Botany
70:1603-1611.

Buchanan, K. S. 1999. Restoration of sandhill vegetation on abandoned agricultural
lands. Master of Science. University of Central Florida, Orlando, Florida.

Buhler, D. D., and B. D. Maxwell. 1993. Seed separation and enumeration from soil
using K2CO3 centrifugation and image analysis. Weed Science 41:298-302.

Butler, B. J., and R. L. Chazdon. 1998. Species richness, spatial variation, and abundance
of the soil seed bank of a secondary tropical rain forest. Biotropica 30:214-222.

Carrington, M. E. 1997. Soil seed bank structure and composition in Florida sand pine
scrub. The American Midland Naturalist 137:39-47.

Chazdon, R. L., R. K. Colwell, J. S. Denslow, and M. R. Guariguata. 1998. Statistical
methods for estimating species richness of woody regeneration in primary and
secondary rain forests of northeastern Costa Rica. Pages 285-309 In F. Dallmeier
and J. A. Comiskey, editors. Forest Biodiversity Research, Monitoring and
Modeling: Conceptual Background and Old World Case Studies. Parthenon
Publishing, Paris, France.

Clewell, A. F. 1989. Natural history of wiregrass (Aristida strict Michx., Gramineae).
Natural Areas Journal 9:223-233.

Colwell, R. K. 1997. EstimateS: Statistical estimation of species richness and shared
species from samples. Version 6. User's Guide and application published at:
http://viceroy.eeb.uconn.edu/estimates.

Corbett, E. A., R.C. Anderson, and C.S. Rodgers. 1996. Prairie revegetation of a strip
mine in Illinois: fifteen years after establishment. Restoration Ecology 4:346-354.

Cottam, G. 1987. Community dynamics on an artificial prairie. Pages 257-270 In W. R.
Jordan, M. E. Gilpin, and J. D. Aber, editors. Restoration Ecology: A Synthetic
Approach to Ecological Research. Cambridge University Press, New York, New
York.

Cox, A. C., D. R. Gordon, G. S. Seamon, and J. L. Slapcinsky. 2003. Understory
restoration in longleaf pine sandhills. Natural Areas Journal. In press.









D'Antonio, C. D., and L. A. Meyerson. 2002. Exotic plant species as problems and
solutions in ecological restoration:A synthesis. Restoration Ecology 10:703-713.

Dhillion, S. S., and J. C. Zak. 1993. Microbial dynamics in arid ecosystems:
desertification and the potential role of mycorrhizas. Revista Chilena De Historia
Natural 66:253-270.

Ebbers, B. C., R. C. Anderson, and A. E. Liberta. 1987. Aspects of the mycorrhizal
ecology of prairie dropseed, Sporobolus heterolepis (Poaceae). American Journal
of Botany 74:564-573.

Eom, A., D.C. Hartnett, and G.W.T. Wilson. 2000. Host plant species effects on
arbuscular mycorrhizal fungal communities in tallgrass prairie. Oecologia
122:435-444.

Fisher, J. B., and K. Jayachandran. 1999. Root structure and arbuscular mycorrhizal
colonization of the palm Serenoa repens under field conditions. Plant and Soil
217:229-241.

Florida Exotic Pest Plant Council. 2003. List of Florida's Invasive Species. Florida Exotic
Pest Plant Council. URL: http://www.fleppc.org/031ist.htm, February 2003.

Florida Forever. 2003. Florida Forever Act. Florida Department of Environmental
Protection. URL: http://www.dep.state.fl.us/lands/carlff/, January 2003.

Florida Natural Areas Inventory and Florida Department of Natural Resources. 1990.
Guide to the natural communities of Florida. Florida Natural Areas Inventory and
Department of Natural Resources, Tallahassee, FL.

Godfrey, R. K., and J. M. Wooten. 1979. Aquatic and Wetland Plants of Southeastern
United States: Monocotyledons. The University of Georgia Press, Athens,
Georgia.

Godfrey, R. K., and J. M. Wooten. 1981. Aquatic and Wetland Plantsof Southeastern
United States: Dicotyledons. The University of Georgia Press, Athens, Georgia.

Gordon, D. R., A. C. Cox, G. S. Seamon, J. L. Slapcinsky, and D. A. Jones. 2001. Native
sandhill species revegetation techniques. Final Report to the Florida Department of
Transportation Contract # BB 937, University of Florida and The Nature
Conservancy, Gainesville, FL.

Gordon, D. R., M. J. Hattenbach, G. S. Seamon, K. Freeman, and D. A. Jones. 2000.
Establishment and Management of upland native plants on Florida roadsides.
Final report to the Florida Department of Transportation Contract # BA 523,
University of Florida and The Nature Conservancy, Gainesville, Florida.









Gotelli, N.J. 2001. The Primer of Ecology, Third Edition. Sinauer Associates, Inc.
Sunderland, MA.

Gross, K. L. 1990. A comparison of methods for estimating seed numbers in soil. Journal
of Ecology 78:1079-1093.

Gross, K. L. and K. A. Renner. 1989. A new method for estimating seed numbers in the
soil. Weed Science 37:836-839.

Hardin, E. D., and D. L. White. 1989. Rare vascular plant taxa associated with wiregrass
(Aristida strict) in the southeastern United States. Natural Areas Journal 9:234-
245.

Harper-Lore, B. L. 1998. Do native grasslands have a place on roadsides? Land and
Water July/August:29-31.

Hart, M. M., R. J. Reader, and J. N. Klironomos. 2001. Life-history strategies of
arbuscular mycorrhizal fungi in relation to their successional dynamics.
Mycologia 93:1186-1194.

Hartel, P. G. 1999. The soil habitat. In D. M. Sylvia, J. H. Fuhrmann, P. G. Hartel, and D.
A. Zuberer, editors. Principles and applications of soil microbiology. Prentice
Hall, Upper Saddle River, New Jersey, USA.

Hartnett, D. C., and G. W. T. Wilson. 1999. Mycorrhizae influence plant community
structure and diversity in tallgrass prairie. Ecology 80:1187-1195.

Hattenbach, M. J., D. R. Gordon, G. S. Seamon, and R. G. Studenmund. 1998.
Development of direct seeding techniques to restore native groundcover in a
sandhill ecosystem. Pages 64-70 In Proceedings, Longleaf Pine Ecosystem
Restoration Symposium, Auburn, AL.

Heady, H. F. 1958. Vegetational changes in the California annual type. Ecology 39:402-
416.

International Culture Collection of Vesicular Arbuscular Mycorrhizal Fungi. 2003.
International Culture Collection of (Vesicular) Arbuscular Mycorrhizal Fungi.
West Virginia University. URL: http://invam.caf.wvu.edu/methods/assays
/MIP.htm, March 2003.

Jakobsen, I. 1992. Phosphorus transport by external hyphae of vesicular-arbuscular
mycorrhizas. Pages 48-58 In D. J. Read, D. H. Lewis, A. H. Fitter, and I. J.
Alexander, editors. Mycorrhizas in Ecosystems. CAB International, Wallingford,
UK.









Johnson, N. C., D. R. Zak, D. Tilman, and F. L. Pfleger. 1991. Dynamics of vesicular-
arbuscular mycorrhizae during old field succession. Oecologia 86:349-358.

Jordan, C. L. 1984. Florida's weather and climate: implications for water. Pages 18-35 In
E. A. Femald and D. J. Patton, editors. Water Resources Atlas of Florida. Florida
State University, Tallahassee, Florida.

Kabir, Z., I.P. O'Halloran, and C. Hamel. 1999. Combined effects of soil disturbance and
fallowing on plant and fungal components of mycorrhizal corn (Zea mays L.).
Soil Biology and Biochemistry 31:307-314.

Kesler, T. R., L. C. Anderson, and S. H. Hermann. 2003. A taxonomic reevaluation of
Aristida strict (Poaceae) using anatomy and morphology. Southeastern
Naturalist 2:1-10.

Kitajima, K., and D. Tilman. 1996. Seed banks and seedling establishment on an
experimental productivity gradient. Oikos 76:381-391.

Klotzli, F, and A. P. Grootjans. 2001. Restoration of semi-natural wetland systems in
central Europe: Progress and predictability of developments. Restoration Ecology
9(2):209-219.

Landers, J. L., and A. S. Johnson. 1976. Bobwhite quail food habits. Miscellaneous
Report # 4, Tall Timbers Research Station, Tallahassee, FL.

Langeland, K. A., and K. C. Burks, Editors. 1998. Identification and Biology of Non-
native Plants in Florida's Natural Areas. University of Florida, Gainesville, FL.

Laughlin, D. C. 2003. Lack of native propagules in a Pennsylvania, USA, limestone
prairie seed bank: Futile hopes for a role in ecological restoration. Natural Areas
Journal 23:158-164.

Leckie, S., M. Vellend, G. Bell, M. J. Waterway, and M. J. Lechowicz. 2000. The seed
bank in an old-growth, temperate deciduous forest. Canadian Journal of Botany
78:181-192.

Magurran, A. E. 1988. Ecological Diversity and its Measurements. Princeton University
Press, Princeton, New Jersey.

Maliakal, S. K., E. S. Menges, and J. S. Denslow. 2000. Community composition and
regeneration of Lake Wales Ridge wiregrass flatwoods in relation to time since
fire. Journal of Torrey Botanical Society 127:125-138.

Malone, C. R. 1967. A rapid method for enumeration of viable seeds in soils. Weeds
15:381-382.









Martin, A. C., and W. D. Barkley. 1961. Seed Identification Manual. University of
California Press, Berkeley, California.

McGonigle, T. P., and A. H. Fitter. 1990. Ecological specificity of vesicular arbuscular
mycorrhizal associations. Mycological Research 94:120-122.

McGonigle, T. P., M.H. Miller, D.G. Davis, G.L. Fairchild, and J.A. Swan. 1990. A new
method which gives an objective measure of colonization of roots by vesicular-
arbuscular mycorrhizal fungi. New Phytologist 115:495-501.

Medina, O. A., A. E. Kretschmer Jr., and D. M. Sylvia. 1988. The occurence of vesicular-
arbuscular mycorrhizal fungi on tropical forage legumes in South Florida.
Tropical Grasslands 22:73-78.

Miller, R. M. 1987. Mycorrhizae and succession. Pages 205-219 in C. L. Jordan, M. E.
Gilpin, and J. D. Aber, editors. Restoration Ecology: A Synthetic Approach to
Ecological Research. Cambridge University Press, New York, New York.

Miller, R. M. 1997. Prairie Underground. Pages 23-27 in S. Packard and C. F. Mutel,
editors. The Tallgrass Restoration Handbook for Prairies, Savannas, and
Woodlands. Island Press, Washington, D.C.

Moynahan, O. S., and C. A. Zabinski. 2002. Microbial community structure and carbon-
utilization diversity in a mine tailings revegetation study. Restoration Ecology
10:77-87.

Mullahey, J. J., and C. S. Speed. 1991. The occurence of vesicular-arbuscular
mycorrhizae on Florida range grasses. Soil and Crop Science Society of Florida
Proceedings 50:44-47.

Mulligan, M. K., L. K. Kirkman, and R. J. Mitchell. 2002. Aristida beyrichiana
(wiregrass) establishment and recruitment: Implications for restoration.
Restoration Ecology 10:68-76.

The Nature Conservancy. 1996. Upland Restoration Plan: Conceptual Plan and
Assessment of Pasture Restoration Sites. The Nature Conservancy, Kissimmee,
Florida.

The Nature Conservancy. 1997. Fourth Annual Monitoring Report for the Disney
Wilderness Preserve, The Nature Conservancy. Kissimmee, FL.

The Nature Conservancy. 1999. The Analysis of Data from the First Two Years of Post-
treatment Monitoring: Disney Wilderness Preserve Upland Restoration Program.
Subcontract Agreement # NRA63, The Nature Conservancy, Disney Wilderness
Preserve, Kissimmee, FL.









The Nature Conservancy. 2000. Proceedings of the Upland Restoration Workshop. The
Nature Conservancy. Kissimmee, FL.

Noss, R. F. 1989. Longleaf pine and wiregrass: Keystone components of an endangered
ecosystem. Natural Areas Journal 9:211-213.

Boosting, H. L., and M. E. Humphreys. 1940. Buried viable seeds in a successional series
of old field and forest soils. Bulletin of the Torrey Botanical Club 67:253-273.

Packard, S., and C. F. Mutel. 1997. The Tallgrass Restoration Handbook: For Prairies,
Savannas, and Woodlands. Island Press, 1997.

Peng, S., D. M. Eissenstat, J. H. Graham, K. Williams, and N. C. Hodge. 1993. Growth
depression in mycorrhizal citrus at high-phosphorus supply. Plant Physiology
101:1063-1071.
Platt, W. J. 1988. Fire ecology of the southeastern longleaf pine flatwoods. Pg 5 in L. C.
Duever and R. F. Noss, editors. Wiregrass Biology and Management: Maintaining
Groundcover Integrity in Longleaf: Proceedings of the Symposium on. Valdosta
State College, Gainesville, FL.

Putwain, P. D., and D. A. Gillham. 1990. The significance of the dormant viable seed
bank in the restoration of heathlands. Biological Conservation 52:1-16.

Pyne, S. J., P.L. Andrews, and R.D. Laven. 1996. Introduction to Wildland Fire. John
Wiley and Sons, Inc., New York.

Rabinowitz, D. 1981. Buried viable seeds in a North American tallgrass prairie: the
resemblance of their abundance and composition to dispersing seeds. Oikos
36:191-195.

Read, D. J., H. K. Koucheki, and J. Hodgson. 1976. Vesicular-arbuscular mycorrhiza in
natural vegetation systems: 1. The occurence of infection. New Phytologist
77:641-653.

Richter, B. S., and J. C. Stutz. 2002. Mycorrhizal inoculation of big sacaton: Implications
for grassland restoration of abandoned agricultural fields. Restoration Ecology
10:607-616.

Roberts, H. A. 1981. Seed banks in soils. In T. H. Coaker, editor. Advances in Applied
Biology. Academic Press.

SAS. 1999. The SAS System for Windows. SAS Institute, Cary, NC.

Schenck, N. C., and R. A. Kinloch. 1980. Incidence of mycorrhizal fungi on six field
crops in monoculture on a newly cleared woodland site. Mycologia 72:445-456.