Citation
Variation in Functional Morphology and Physiological Responses of Dicerandra (lamiaceae) Congeners Native to Sandhill Habitats and Florida Scrub

Material Information

Title:
Variation in Functional Morphology and Physiological Responses of Dicerandra (lamiaceae) Congeners Native to Sandhill Habitats and Florida Scrub
Creator:
MERCHANT, AMETHYST GAIL
Copyright Date:
2008

Subjects

Subjects / Keywords:
Annuals ( jstor )
Biomass ( jstor )
Growth habit ( jstor )
Leaves ( jstor )
Nutrients ( jstor )
Perennial growth ( jstor )
Perennials ( jstor )
Plants ( jstor )
Species ( jstor )
Transpiration ( jstor )
City of Lake Wales ( local )

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright Amethyst Gail Merchant. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Embargo Date:
12/31/2008
Resource Identifier:
659561101 ( OCLC )

Downloads

This item is only available as the following downloads:


Full Text

PAGE 1

VARIATION IN FUNCTIONAL MORPHOLOGY AND PHYSIOLOGICAL RESPONSES OF Dicerandra (LAMIACEAE) CONGENERS NATIVE TO SANDHILL HABITATS AND FLORIDA SCRUB By AMETHYST GAIL MERCHANT A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2006

PAGE 2

Copyright 2006 by Amethyst Gail Merchant

PAGE 3

To my grandmother, Grovene Merchant

PAGE 4

iv ACKNOWLEDGMENTS I wish to extend gratitude to my committee members (Drs. Stephen Mulkey, Kenneth Boote, Kaoru Kitajima, Timothy Mar tin, and Walter Judd) who have contributed to the completion of this dissertation and for their patience throughout the process. It was a privilege to work with them at the University of Florida. I thank them for their guidance and support. I thank Dr. Stephen Mulkey, my advisor, for giving me the freedom to pursue my own research interests. I greatly appreciate the support that allowed me to visit the Smithsonian Tropical Research Institute and Ba rro Colorado Island. I am thankful for the experiences I had on the canopy cranes in Panama. Special thanks go to Dr. Walter Judd for introducing me to the genus Dicerandra . He inspired me to question the reasons behind the variance in trait expression of these species. His enthusiasm is contagious. Completion of this dissertation would not have been possible without the help of several individuals. I would like to sincerely thank Drs. Kenneth Boote and Alison Fox for helping me acquire study sites for this research at Irrigation Park and Bivens Arm in Gainesville, Florida. The help I received from Dr. Shirley West was essential for performance of these studies. I thank him for allowing me to use his seed germination chamber for an extended period of time. I thank Dr. Kaoru Kitajima for allowing me to use her equipment for measurements in the field and germination of seeds. I deeply

PAGE 5

v appreciate the financial and technical support for plant tissue CHN analysis that I received from Dr. Timothy Martin. I thank Dr. Eric Menges at Archbold Biol ogical Research Station along the Lake Wales Ridge in southern Florida for his guidance concerning research utilizing Dicerandra and the seeds he supplied. I would like to thank the park service in charge of OÂ’Leno State Park in High Springs, Florida, and the management team of the Marjorie Harris Carr Greenway and Trails property located southwest of Ocala, Florida, for allowing me to setup field plots, take measurements, collect tissue samples, and collect seeds from Dicerandra . Funding was provided by a grant from the Florida Native Plant Society. I thank my parents, Richard and Rhonda Merchant, for their love and devotion. Their support has helped me to reach my goals and fulfill my dreams throughout my life. I extend my dearest appreciation to my husband, Dr. Carlos Messina, for the patience he showed and daily encouragement he gave me. His love and friendship were essential for the completion of this dissertation.

PAGE 6

vi TABLE OF CONTENTS page ACKNOWLEDGMENTS.................................................................................................iv LIST OF TABLES.............................................................................................................ix LIST OF FIGURES.............................................................................................................x ABSTRACT....................................................................................................................... xi CHAPTER 1 INTRODUCTION........................................................................................................1 The Role of Ecophysiological Traits in Evolutionary Adaptation...............................1 The Genus Dicerandra .................................................................................................3 Abiotic Conditions of Scrub and Sandhill Habitats......................................................3 Florida Scrub Ecosystems.............................................................................................5 Research on Species of Dicerandra .............................................................................7 Objectives and Organiza tion of Dissertation................................................................8 2 FORM AND FUNCTION WITHIN Dicerandra (LAMIACEAE): VARIATION IN PLANTAND LEAF-LEVEL TRAITS OF CONGENERS NATIVE TO FLORIDA SCRUB AND SANDHILL COMMUNITIES IN THE SOUTHEASTERN UNITED STATES.....................................................................12 Introduction.................................................................................................................12 Materials and Methods...............................................................................................15 Species and Seed Source.....................................................................................15 Light Availability and Leaf Traits w ithin the Native Habitat of Two Species...16 Plant Culture........................................................................................................16 Experimental Design...........................................................................................17 Biomass Allocation and Shoot Architectural Parameters...................................18 Leaf Gas Exchange and Morphology Measurements..........................................19 Leaf Anatomical Measurements..........................................................................20 Statistical Methods..............................................................................................21 Results........................................................................................................................ .22 Light Availability and Leaf Traits within Native Habitats..................................22 Growth Habit.......................................................................................................22

PAGE 7

vii Plant Biomass Allocation....................................................................................23 Shoot Architecture...............................................................................................24 Leaf Morphology and Physiology.......................................................................25 Leaf Anatomy......................................................................................................26 Discussion...................................................................................................................28 Divergence in Growth Habit, Biomass Allocation, and Shoot Architecture.......29 Variation in Leaf Morphology, Physiology and Anatomy..................................30 Conclusions.........................................................................................................36 3 RESPONSES TO VARIATION IN NUTRIENT AVAILABILITY OF AN ANNUAL AND A PERENNIAL CONGENER NATIVE TO DIFFERENT FLORIDA HABITATS..............................................................................................47 Introduction.................................................................................................................47 Materials and Methods...............................................................................................50 Species and Seed Source.....................................................................................50 Plant Culture........................................................................................................51 Garden Experiment..............................................................................................51 Harvesting............................................................................................................52 Tissue Analysis....................................................................................................53 Trait Expression of Species in Native Habitats...................................................53 Statistical Analysis..............................................................................................54 Results........................................................................................................................ .55 Overall Growth....................................................................................................55 Total Plant N........................................................................................................55 Effect of Nutrient Availab ility on Annual Reproduction....................................56 Biomass Allocation.............................................................................................56 Plant Nitrogen-use Efficiency.............................................................................57 N Concentration of Plant Tissues........................................................................58 Discussion...................................................................................................................59 Growth and Biomass Allocation.........................................................................59 Nutrient Uptake and Use.....................................................................................61 Growth of Species in Native Habitats.................................................................63 Ecological Implications.......................................................................................65 Conclusions.........................................................................................................66 4 INFLUENCE OF INCREASING WATER-DEFICIT ON RESPONSES OF Dicerandra CONGENERS DIFFERING IN NATIVE HABITAT AND GROWTH HABIT......................................................................................................72 Introduction.................................................................................................................72 Materials and Methods...............................................................................................75 Species and Seed Source.....................................................................................75 Plant Culture........................................................................................................76 Characterization of Drought Stress.....................................................................78 Experimental Design...........................................................................................78 Leaf Gas Exchange and Physiological Measurements........................................81

PAGE 8

viii Results........................................................................................................................ .83 Plant-level Responses to Water-deficit...............................................................83 Leaf-level Responses to Water-deficit................................................................85 Discussion...................................................................................................................87 5 CONCLUSIONS......................................................................................................100 LIST OF REFERENCES.................................................................................................105 BIOGRAPHICAL SKETCH...........................................................................................115

PAGE 9

ix LIST OF TABLES Table page 1-1 Environmental factors and life-history traits............................................................11 2-1 Aboveground biomass accumulation and allocation of Dicerandra species...........38 2-2 Architectural parameters of Dicerandra species......................................................39 2-3 Leaf ecophysiological characteristics of Dicerandra species..................................40 2-4 Light response curve components of Dicerandra species........................................41 2-5 Leaf anatomical characteristics of Dicerandra species (part 1)...............................42 2-6 Leaf anatomical characteristics of Dicerandra species (part 2)...............................43 4-1 Duration of dry-down and length of survival period for D. densiflora and D. cornutissima plants in the water-deficit treatment...................................................92 4-2 Shoot (leaf and stem) and leaf mass of D. densiflora and D. cornutissima plants in the water-deficit treatment...................................................................................93 4-3 FTSW threshold values for plantand leaf-level responses.....................................94

PAGE 10

x LIST OF FIGURES Figure page 2-1 Distribution of Dicerandra species utilized in this study........................................44 2-2 Aboveand belowground biomass allocation of an annual ( D. densiflora ) and a perennial ( D. cornutissima ) species grown in a common garden............................45 2-3 Photographs of leaf transverse sections of each species at 200x.............................46 3-1 Total (a) plant biomass, (b) plant N, and (c) number of flowers produced by D. densiflora (DEN) and D. cornutissima (COR) when grown under each combination of N and P availability.........................................................................68 3-2 Plant tissue mass ratios of (a) root, (b) stem, (c) leaf, and (d) floral structure tissues produced by D. densiflora (DEN) and D. cornutissima (COR) when grown under each combination of N and P availability...........................................69 3-3 Nitrogen-use efficiency of D. densiflora (DEN) and D. cornutissima (COR) when grown under each combination of N and P....................................................70 3-4 The %N of (a) root, (b) stem, (c) leaf, and (d) floral tissue of D. densiflora (DEN) and D. cornutissima (COR) when grown under each combination of N and P availability......................................................................................................71 4-1 Time course of plant transpiration per unit leaf mass for D. densiflora (a) and D. cornutissima (b) plants in the wa ter-deficit treatment..............................................95 4-2 NTR – FTSW response curves of D. densiflora (a) and D. cornutissima (b)...........96 4-3 Normalized Amax-area, E, and gs responses of D. densiflora and D. cornutissima to changing FTSW ........................................................................................................97 4-4 Normalized RWC – FTSW response curves of D. densiflora (a) and D. cornutissima (b) plants in the wa ter-deficit treatment..............................................98 4-5 Normalized Amax-area and E responses of D. densiflora and D. cornutissima to changing RWC.........................................................................................................99

PAGE 11

xi Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy VARIATION IN FUNCTIONAL MORPHOLOGY AND PHYSIOLOGICAL RESPONSES OF Dicerandra (LAMIACEAE) CONGENERS NATIVE TO SANDHILL HABITATS AND FLORIDA SCRUB By Amethyst Gail Merchant December 2006 Chair: Stephen S. Mulkey Major Department: Botany The genus Dicerandra contains perennial species endemic to FloridaÂ’s unique scrub communities and annual species inhabiting sandhill habitats. Variation in plantand leaf-level ecophysiological attributes of th ese closely related species were studied to gain insight into the significance of their divergence in form and function in relation to environmental conditions of their native habitats. Two sandhill and four scrub species were grown within a common garden. Divergent differences in growth habit, ar chitecture, leaf morphology, and leaf anatomy were found. Perennials native to scrub gaps exhibited traits typical of plants adapted to high-light. Annuals native to sandhill understories expressed traits of plants adapted to shaded environments. Dicerandra densiflora , sandhill annual, and D. cornutissima , scrub perennial, were grown under various nutrient availabilities to examine responses in relation to their

PAGE 12

xii growth habit. Trade-offs between traits optimizing fecundity in the annual and those promoting survival in the perennial were found. Dicerandra cornutissima exhibited biomass allocation plasticity as resources shifted from root to leaf and stem production with increased N availability. Ontogenetic shifts to seed production may have constrained biomass allocation plasticity in D. densiflora . However, increases in N resulted in an increase in flower number due to increases in mass. The annual also exhibited greater nitrogen-use efficiency per unit leaf N. Gas exchange responses of these species were compared as water-deficit increased. Dicerandra cornutissima expressed greater drought tolerance. Its stomata stayed open and carbon gain continued at lower water availability. This allows the species to be less sensitive to short-term, small decreases in water availability that periodically occur in scrub. Dicerandra densiflora exhibited greater sensitivity with its faster decline in stomatal conductance. This may be a mechanism to conserve water to successfully complete seed fill. Each speciesÂ’ strategy appears to be influenced by its growth habit and water availability within its native habitat. These are the first ecophysiological studies conducted on Dicerandra species. Results suggest that evolution under the environmental conditions of their native habitats led to divergence in traits of these congeners. Advantages of variation in their functional morphology and physiological responses could only be understood in the context of their growth habit.

PAGE 13

1 CHAPTER 1 INTRODUCTION The Role of Ecophysiological Traits in Evolutionary Adaptation Plant species inhabit a diverse array of ecosystems that exhibit great variation in resource availability and biotic interactions between species. Evolution has led to adaptation, acclimation, and speciation in plants within these natural systems in response to selective pressures of the environment (Nilsen and Orcutt, 1996). These long-term processes have created trait combinations that allow plants to cope with the abiotic and biotic environmental conditions of their native habitats. The phenotype of a plant is an “integrated function” of its growth, morphology, physiology, and life history (Arntz and Delph, 2001). Ecophysiological traits such as growth rate, allocation of biomass and nutri ents, shoot and root architecture, phenology, gas exchange, and leaf structure influence plant fitness (Ackerly et al., 2000). Natural selection acts to select for genotypes (individuals) expressing characteristics that increase fitness given the environment’s resource availability and climate (Ackerly et al., 2000; Arntz and Delph, 2001). Variation in trait expr ession leads to differences in allocation of resources to growth, maintenance, and reproduction (Cody, 1966; Gadgil and Bossert, 1970; Bazzaz et al., 1987). Since resource limitation can vary in frequency and intensity within the year, trade-offs between traits promoting fecundity and those promoting survival cause variation in development of life-history traits (Pitelka, 1977; Bazzaz et al., 1987; Crawley, 1997). Plants can adapt to sources of stress by altering their life history

PAGE 14

2 to complete their life cycle before stressful conditions occur or by increasing their ability to tolerate stress (Griffith and Watson, 2005). A number of environmental factors can influence plant growth and development. Temperature regime and photoperiod are the be st-known factors influencing growth form expression and development (Kim and Okubo, 1995). Plant responses to variation in these factors lead to latitudinal range limits of species (Griffith and Watson, 2005). Resource availability can affect expressed growth habits by controlling the growth rate and therefore size of an individual (Hirose and Kachi, 1982; Bazzaz et al., 1987). Attainment of a critical plant size is required to induce flowering in some species (Hirose and Kachi, 1982). When resources are limited, the ability of an annual plant to optimally allocate resources between life history proce sses for the successful completion of its life cycle within a year is decreased (Hirose and Kachi, 1982; Kachi and Hirose, 1983a; Bazzaz and Morse, 1991). The expression of perennial life-history traits, and corresponding morphological and physiological attr ibutes, may allow for an advantage in growth, maintenance, and reproduction under resource-limited conditions (Schaffer and Gadgil, 1975). However, in a resource-rich environment supporting numerous competitors, the comparatively faster resource sequestration of an annual growth habit may allow for optimal allocation and successful reproduction. Disturbance regimes also shape growth form expression. In nutrient-rich, unpredictable environments where disturban ces such as fire are common, fast-growing short-lived species predominate (Schaffer and Gadgil, 1975; Grime, 1977; Pitelka, 1977; Lambers et al., 1998). However, slow-growing longer-lived species dominate more

PAGE 15

3 predictable environments where disturbances such as fire are less frequent, especially when resources are limited. The Genus Dicerandra The genus Dicerandra (Family Lamiaceae; subfamily Nepetoideae) consists of species endemic to FloridaÂ’s unique scrub communities and various sandhill habitats within the southeastern coastal plain of the United States (Huck, 1987; Huck et al., 1989). Each of five suffrutescent (having only a woody base) perennial species is limited to a small, geographically isolated, and increasingly rare scrub community within FloridaÂ’s peninsula. All perennial species are listed federally as endangered with global 1 state 1 (G1S1) status (greatest threat level). These chamaephytes produce determinate reproductive shoots and indeterminate, overwintering vegetative shoots which extend from a ramose (having many branches) base. Four herbaceous annual species are therophytes, which inhabit recently disturbed sandhill habitats throughout portions of northern Florida and southern Georgia with smaller ranges in Alabama and South Carolina. One annual species, Dicerandra radfordiana , is considered endangered with G1S1 status. Observations of Dicerandra within their native habitats find that species exhibit variations in plantand leaf-level traits although all members of this genus are closely related. Study of these closely related species of Dicerandra offers a unique opportunity to elucidate the importance of divergence in plant form and function in relation to the environmental conditions experienced by these species within their native habitats. Abiotic Conditions of Scrub and Sandhill Habitats Scrub and sandhill habitats, where Dicerandra species are native, differ in climate and resource availability (Table 1-1). Perennial species experience sub-tropical

PAGE 16

4 temperatures in scrub habitats along the ridge systems of central and southern portions of FloridaÂ’s peninsula (Chen and Gerber, 1990; Menges, 1999). Sandhill communities inhabited by annual species experience a more temperate climate where lower winter temperatures may shorten the growth season. Microhabitats occupied by Dicerandra differ in light availability. Perennial species of Dicerandra grow under high-light conditions on bare sand within gaps or along the edges of scrub. However, annual species are typically found beneath pine canopies competing with other understory species. Both scrub and sandhill habitats inhabited by Dicerandra have nutrient-poor soils. Relict beach ridges and dunes supporting scrub habitats are composed of predominantly quartz sand entisols with no organic horizon and are considered low in nutrients, notably phosphorus (Willis and Yemm, 1961; Kachi and Hirose, 1983b; Abrahamson et al., 1984; Christman and Judd, 1990; Myers, 1990; Menges and Gallo, 1991; Anderson and Menges, 1997; Brady and Weil, 1999; Menges, 1999). Ground cover and litter are sparse in these scrub habitats where herbaceous species are rare. Areas supporting sandhill vegetation consist of ultisols and spodosols with a layer of nutrient-rich organic material at the soil surface (Myers, 1990). These habitats have a great diversity of herbaceous species, including many grasses, in the understory and an overstory of pine species (Myers, 1990; Carrington and Keeley, 1999). Therefore, litter accumulation is greater in these sandhill habitats. Although both ecosystems are pyrogenic, low intensity fires occur within sandhill every 1 to 10 years while high intensity fires move through scrub every 10 to 100 years (Myers, 1990). An increase in soil phosphorus availability and

PAGE 17

5 plant tissue phosphorus concentrations has been found in Florida sandhill habitats following controlled burns (Anderson and Menges, 1997). Scrub and sandhill species of Dicerandra experience disparate levels of water availability in their native habitats during their lifetime due to differences in soil characteristics and expressed growth habits. The pure and deep sand soils of scrub are excessively well-drained and have low wa ter-holding capacity (Myers, 1990; Menges and Gallo, 1991; Menges, 1999). A litter layer is typically not found in areas inhabited by perennial Dicerandra . Plants growing in scrub periodically contend with limited water availability due to soil properties and high evaporative demand. Overwintering vegetative shoots of perennial species survive the dry season, which begins in November and continues into late spring (Huck, 1987; Chen and Gerber, 1990; Myers, 1990; Menges and Gallo, 1991). All annual species of Dicerandra avoid significant drought by completing their life cycle as the dry season begins. Populations of annual Dicerandra establish in recently disturbed areas of sandhill within a few hundred meters of streams or rivers. A layer of organic material along the surface increases the soil water-holding capacity of their microhabitats (Myers, 1990). One annual species, Dicerandra densiflora , is native only to sandhill habitats where clay layers with high soil waterholding capacity are present approximately 20 cm below the surface (observation; Huck, 1987). Florida Scrub Ecosystems The scrub habitats, where perennial Dicerandra are endemic, support unique communities found only on ancient dune ridges within the interior of the Florida peninsula and in dune systems along FloridaÂ’s Atlantic coastline (Christman and Judd, 1990; Myers, 1990). This shrubland is dominated by xeromorphic plants with

PAGE 18

6 characteristics such as short stature, small evergreen leaves, and high allocation to root systems (Menges, 1999). The proportion of endemism in Florida scrub is one of the highest in North American plant communities (Estill and Cruzan, 2001). An estimated 40 to 60% of extant scrub species are endemic. However, more than 90% of areas supporting these habitats have been lost due to urbanization and agricultural use (Christman and Judd, 1990; Myers, 1990). Scrub habitats are geographically isolated ecosystems surrounded by other plant communities along these ridges in Florida (Christman and Judd, 1990). Distribution of endemic species is often limited to one or a few disjunct scrub communities. No single scrub supports all endemics. Few studies have been concerned with characterizing and understanding the importance of ecophysiological traits expressed by species native to these unique plant communities. Given the rapid loss of these habitats, studies of the physiological ecology of scrub species would benefit conservation and restoration efforts. Several studies have focused on determining the causes of variation in seasonal and daily water relations of three Quercus species ( Q. myrtifolia, Q. chapmanii, and Q. inopina ; Menges and Gallo, 1991; Abrams and Menges, 1992; Menges, 1994). Measurements of Quercus predawn water potentials suggested that these species experienced drought-stress during the dry season; however, these values were not correlated with surface soil moisture or the distance to the water table (Menges and Gallo, 1991). These species were found to exhibit high bulk modulus of elasticity values and lower osmotic potentials at full and zero turgor when compared to average drought responses of Quercus species in North America (Abrams and Menges, 1992). However, significant seasonal variation in osmotic potentials was correlated with fluctuations in

PAGE 19

7 temperature and leaf age. During drought, sc rub oak species were able to significantly increase predawn water potentials by uptaking fog through their foliage (Menges, 1994). Many other studies have measured the long-term effects of elevated CO2 levels on traits of scrub oak species and on ecosystem level responses (Li et al., 1999; Hungate et al., 2002; Hymus et al., 2002; Hymus et al., 2003; Li et al., 2003; Stiling et al., 2004). However, studies have not measured responses of other native scrub species to variation in resource availability to increase our knowledge of the functional morphology and physiological responses of species endemic to these unique communities. Research on Species of Dicerandra A study of the evolution and systematics of Dicerandra was conducted by Robin B. Huck (1987). Floral morphology and pollinati on biology of the species were observed. Crossability and pollen fertility of several interspecific crosses were determined through breeding experiments. The presented phylogenetic analysis divided this genus into two sections based on differences in floral characteristics. Section Lecontea contained two annual species, Dicerandra odoratissima and D. radfordiana . The other two annual species, Dicerandra linearifolia and D. densiflora , and all perennial species were placed in Section Dicerandra . A revised cladistic analysis utilizing many morphological characters supported division of Dicerandra species into the two sections presented by Huck (1987) although differences in species re lationships were found (Huck et al., 1989). However, a recent DNA-based cladistic analysis concluded that the genus consists of two sister clades, one containing the shrubby perennial species (subgenus Kralia ) and the other containing all of the herbaceous annual species (subgenus Dicerandra ; Oliveira et al., in press). A cytological study of this genus also found Dicerandra species to be either tetraploid or hexaploid (Huck and Chambers, 1997).

PAGE 20

8 Ecophysiological characteristics of Dicerandra species native to scrub or sandhill habitats have not been determined. Studies of scrub Dicerandra have examined their pollination biology (Deyrup and Menges, 1997; Evans et al., 2004), identified chemical defenses of leaf tissue (Eisner et al., 1990; McCormick et al., 1993), measured genetic variation within remaining populations (McDonald and Hamrick, 1996; Menges et al., 2001), and discerned microhabitat preferences (Menges, 1992; Menges et al., 1999; Menges et al., 2006). The annual, Dicerandra linearifolia , was utilized to study seasonal, within-individual changes in a few leaf traits (Winn, 1996a, 1996b, 1999). WinnÂ’s results showed that variation in the measured leaf traits in the field was caused by programmed developmental change, and was not due to changes in experienced temperature regimes. Objectives and Organization of Dissertation The overall objective of this dissertation is to study the variation in plantand leaflevel ecophysiological attributes of these closely related species of Dicerandra to gain insight into the significance of their divergence in form and function in relation to the environmental conditions of their native habitats. This research is divided into three chapters that determine inherent variation in traits of Dicerandra species within a common garden, examine responses of a sandhill annual and a scrub perennial to various levels of nutrient availability, and measure plantand leaf-level gas exchange of a sandhill annual and a scrub perennial as water availability is decreased. Although this research collected measurements solely from species of Dicerandra , information compiled from this group of species increases our knowledge of functional morphology and physiological responses expressed by vege tation endemic to FloridaÂ’s unique scrub communities.

PAGE 21

9 Observations of Dicerandra within their native habitats find that species exhibit differences in plantand leaf-level traits although all members of this genus are closely related. For example, the scrub perennials have a more prostrate canopy with shorter internodes and smaller leaves than the sandhill annuals. It is not known if these divergences in traits are due to differences in genotype among the species or are phenotypic responses to each environmentÂ’s resource availability. In Chapter 2, two sandhill annuals and four scrub perennials were grown within a common garden under non-limiting light, nutrient, and water availability to determine the inherent differences among these closely related species. A comparison of the variation in plantand leaflevel attributes of these congeners provides insight into the traits important for plant fitness in the unique scrub community of Florida. Further studies were implemented to compare responses of a sandhill annual and a scrub perennial to limitations in nutrient and water availability to identify traits of the scrub species that may confer a greater advantage ecologically under limited resource availability. In Chapter 3, Dicerandra densiflora and D. cornutissima were grown under various levels of nutrient availability to examine the responses of these closely related species in relation to their different growth habits. Trade-offs between expression of traits optimizing greater seed production in the annual versus traits promoting greater growth and maintenance to ensure survival and future fecundity in the perennial were expected. To test these predictions, a factorial experiment was conducted utilizing three levels of nitrogen (N) and three levels of phosphorus (P). Measurements were taken at the end of the life cycle of D. densiflora . This allowed for determination of more long-term effects

PAGE 22

10 of nutrient availability on reproduction within the annual and resource allocation within the perennial. In Chapter 4, plantand leaf-level gas exchange responses of Dicerandra densiflora and D. cornutissima were measured as water availability decreased due to an imposed dry-down. These species experience different levels of water availability due to differences in their expressed growth habits and the soil characteristics of their native habitats. Variation in these speciesÂ’ responses to increasing water-deficit stress provides insight into the value of these physiological mechanisms as strategies of specific adaptations to drought. This is the first study to measure the effects of water availability on carbon gain and water loss in a species native to the unique scrub habitats of Florida. The sandhill annual was expected to quickly decrease gas exchange at higher soil water availability to avoid drought while the scrub perennial exhibits drought tolerance by maintaining high levels of gas exchange at lower water availability. Chapter 5 presents a summary and an integration of the conclusions from each study.

PAGE 23

11 Table 1-1. Environmental factors and life-hist ory traits of perennial and annual species of Dicerandra . Growth habit Perennial Annual Native habitat scrub sandhill Climate sub-tropical temperate Microhabitat gaps beneath pine canopies Substrate entisols spodosols, ultisols pH of substrate acidic acidic Fire high intensity, infrequent low intensity, frequent During dry season roots and vegetative shoots seed

PAGE 24

12 CHAPTER 2 FORM AND FUNCTION WITHIN Dicerandra (LAMIACEAE): VARIATION IN PLANTAND LEAF-LEVEL TRAITS OF CONGENERS NATIVE TO FLORIDA SCRUB AND SANDHILL COMMUNITIES IN THE SOUTHEASTERN UNITED STATES Introduction FloridaÂ’s unique scrub communities are found only on ancient dune ridges within the interior of the peninsula and in dune systems along the Atlantic coastline (Christman and Judd, 1990; Myers, 1990). This shrubland is dominated by xeromorphic plants with characteristics such as short stature, small evergreen leaves, and high allocation to root systems (Menges, 1999). The proportion of endemism in Florida scrub is one of the highest in North American plant communities (Estill and Cruzan, 2001). An estimated 40 to 60% of extant scrub species are endemic. However, more than 90% of areas supporting these habitats have been lost due to urbanization and agricultural use (Christman and Judd, 1990; Myers, 1990). Scrub habitats are isolated ecosystems surrounded by other plant communities along these ridges in Florida (Christman and Judd, 1990). Distribution of endemic species is often limited to one or a few disjunct scrub communities. No single scrub supports all endemics. Few studies have been concerned with characterizing and understanding the importance of ecophysiological traits expressed by species native to these unique plant communities (Menges and Gallo, 1991; Abrams and Menges, 1992; Menges, 1994; Li et al., 1999). Given the rapid loss of these habitats, studies of the physiological ecology of scrub species would benefit conservation and restoration efforts.

PAGE 25

13 The genus Dicerandra (Lamiaceae) consists of species endemic to Florida scrub and various sandhill habitats within the southeastern United States (Fig. 2-1; Huck, 1987; Huck et al., 1989). Each of five suffrutescent perennial species is limited to a small, geographically isolated, and increasingly rare scrub community within FloridaÂ’s peninsula. All perennial species are listed federally as endangered with global 1 state 1 (G1S1) status. Four herbaceous annual species inhabit recently disturbed sandhill habitats throughout southern Georgia and northern Florida with smaller ranges in Alabama and South Carolina. Observations of Dicerandra within their native habitats find that species exhibit variations in plantand leaf-level traits although all members of this genus are closely related. Study of the Dicerandra system offers a unique opportunity to elucidate the importance of divergence in plant form and function. Dicerandra species growing in scrub or sandhill habitats encounter differences in climate and resource availabilities. Sandhill communities inhabited by annual species experience a temperate climate where lower winter temperatures may shorten the growth season. Perennial species experience sub-tropical temperatures within scrub along the ridge systems of central and southern portions of FloridaÂ’s peninsula (Chen and Gerber, 1990; Menges, 1999). Microhabitats occupied by Dicerandra differ in light availability. Annual species are typically found beneath pine canopies competing with other understory species. However, perennial species grow in open areas within gaps or along the edges of scrub. Both habitats inhabited by Dicerandra have nutrient-poor soils. Substrate supporting sandhill vegetation consists of ultisols and spodosols with a layer of nutrient-rich organic material at the surface (Myers, 1990). Relict beach ridges and dunes supporting scrub habitats are composed of predominantly quartz sand entisols with

PAGE 26

14 no organic horizon and are considered low in nutrients, notably phosphorus (Willis and Yemm, 1961; Abrahamson et al., 1984; Chri stman and Judd, 1990; Myers, 1990; Menges and Gallo, 1991; Anderson and Menges, 1997; Brady and Weil, 1999; Menges, 1999). The pure and deep sand soils of scrub are also excessively well-drained given their low water holding capacity although annual rainfall totals approximately 1270 mm (Myers, 1990; Menges and Gallo, 1991; Menges, 1999). Overwintering vegetative shoots of the perennials survive the dry season, which begins in November and continues into late spring (Huck, 1987; Chen and Gerber, 1990; Myers, 1990; Menges and Gallo, 1991). Therefore, plants growing in scrub contend with limited nutrient and water availability (Myers, 1990). Annual Dicerandra avoid significant drought by completing their life cycle before the dry season begins (Chen and Gerber, 1990). Observations of all species in their native habitats find Dicerandra congeners expressing differences in growth habit, shoot architecture, and morphological leaf traits. The sandhill annuals exhibit characteristics such as greater plant height within one growth season, longer internodes, and larger leaves than the scrub perennials. However, traits expressed by the perennial species in scrub do appear to vary along a continuum. It is not known if divergences in traits such as these are due to differences in genotype or are phenotypic responses to each environmentÂ’s resource availability. In this study, annual and perennial species of Dicerandra were grown within a common garden under non-limiting light, water, and nutrient availabilities to determine the inherent differences among these closely related species. A comparison of the variation in plantand leaflevel attributes of these congeners provides insight into the traits important for plant fitness in the unique scrub communities of Florida.

PAGE 27

15 Materials and Methods Species and Seed Source Two annual and four perennial species were utilized for this experiment. The herbaceous annuals were D. densiflora and D. linearifolia var. robustior. The suffruticose, shrubby perennials were D. cornutissima , D. frutescens subsp . frutescens , D. thinicola , and D. christmanii . Seeds of all species were collected from multiple individuals growing in their respective native habitats. Dicerandra densiflora seeds were collected in late 2000 and 2001 from sites near the Santa Fe River within OÂ’Leno State Park in High Springs, Florida with permission of the Florida Depa rtment of Environmental Protection, Division of Recreation and Parks (permit # 10160012). Seeds of D. linearifolia var. robustior were collected from individuals growing along the roadway near Suwannee River State Park in Live Oak, Florida in 2000 and 2001. Pe rmits from the Division of Plant Industry within the Florida Department of Agriculture and Consumer Services were required to collect seeds of the following endangered species: D. cornutissima and D. frutescens subsp. frutescens . Seeds of D. cornutissima were collected from populations growing on the Marjorie Harris Carr Greenway and Trails property managed by the Department of Environmental Protection, Office of Greenways and Trails located southwest of Ocala, Florida in late 2000 (permit # 413). Seeds of D. frutescens subsp. frutescens were collected from populations growing within the Archbold Biological Research Station along the Lake Wales Ridge of Lake Placid, Florida in late 2000 and 2001 (permit # 414 and 448). Dicerandra christmanii seeds were supplied by Dr. Eric Menges, researcher at Archbold Biological Station. Dicerandra thinicola seeds were collected by permission in

PAGE 28

16 late 2000 from an area managed by the Brevard Environmentally Endangered Lands Program in Titusville, Florida. Light Availability and Leaf Traits within the Native Habitat of Two Species A basic assessment of light availability and leaf traits of two Dicerandra in their native habitats was performed on D. densiflora (annual) and D. cornutissima (perennial) individuals located throughout the areas utilized for seed collection. Hemispherical canopy photographs were taken in September of 2001 above 10 D. densiflora and 14 D. cornutissima plants with a digital camera (Nikon) equipped with a fish-eye lens. The camera lens was positioned one meter above the surface to insure that leaf and stem material from each Dicerandra was not included in the photos. The % of canopy openness and the leaf area index (LAI) of the canopy above these Dicerandra were determined using the Gap Light Analyzer (GLA) software program, version 2.0. Leaf size (cm2) and specific leaf area (cm2 g-1) were measured from three outercanopy leaves from each of 22 D. densiflora plants and from six outer-canopy leaves from each of 24 D. cornutissima plants. Leaves were collected in the field approximately two hours after sunrise and placed in humidified plastic bags to allow for preservation during transport to the lab. Leaf size was measured using a leaf area meter (LI-COR 3100C, Lincoln, NE, USA). Leaf tissue was dried at 60 C for one week before mass was measured. Plant Culture Seeds were germinated beginning mid-May of 2002 over a four-week period in plastic Petri dishes on moistened heavy-weight paper placed in a growth chamber supplying 10 hours of light and 14 hours of darkness at 25 C day and night. Seedlings

PAGE 29

17 were planted in 142 ml pots containing the subs trate mixture explained in the next section and allowed to grow for four weeks under shaded conditions within the Botany Department, University of Florida, greenhous e. Nutrients were applied twice using water-soluble PeterÂ’s fertilizer with 20-20-20 % NPK concentrations. After four leaves had expanded, seedlings were transferred to an unshaded greenhouse in Irrigation Park (University of Florida, Gainesville) to allow for acclimation (2-3 weeks) to higher light and temperature conditions. Experimental Design A common garden experiment was setup in an open field at Irrigation Park the first week of August 2002. Approximately 20 seedlings of each species were transplanted into 2.84 l pots containing the substrate mixture and nutrient material described below. Pots were placed over weed cloth and frequently rotated. Components of the substrate mixture were chosen to mimic characteristics of the substrate within the speciesÂ’ native habitats. The mixture was 60% sand, 20% bark from Pinus species, 10% vermiculite, and 10% perlite. Sand was purchased from the Feldspar Corporation in Edgar, Florida where it was mined, heated, and pressure washed. Vermiculite was added to increase water and nutrient retention. Perlite and bark were used to prevent soil compaction and allow for airflow through the substrate. Pine bark introduced acidic elements that are present in the speciesÂ’ native habitats. A piece of weed cloth was placed inside each pot along its base before the mixture was added to prevent the components from flowing out of the pot when water was applied. Nutrients and water were abundantly supplied to ensure non-limiting conditions. One gram of CaSO4 and 2.5 g of micronutrients were mixed within the substrate of the pots 5 to 8 cm below the surface. A 2.5 g application of a slow release fertilizer

PAGE 30

18 (Osmocote) with 14-14-14 % NPK concentrations was placed in each pot at the beginning of the experiment to allow a dose of macronutrients with each water application. Every three weeks, one additional gram of this fertilizer was added to each pot to replace material that was blown or washed away. All individuals were wellwatered (every 2 days) unless sufficient rainfall occurred. Water was withheld once, for a three-day period, to obtain leaf relative water content measurements. Biomass Allocation and Shoot Architectural Parameters At the beginning of the experiment, five i ndividuals of each species were selected at random and designated for future harvest to obtain biomass allocation and architecture data. These plants were transplanted into 8.83 l pots after one month of growth within the common garden conditions to allow for unhindered root growth as plant size increased. After four months of growth with in the common garden, harvesting of aboveground biomass from the five previously selected individuals was performed over a two-week period on all species. Annual species were in the process of seed formation. Shoots were separated into leaf, stem, and floral tissues to determine the proportion of aboveground resources allocated to each. The floral mass ratio of the plantsÂ’ included inflorescences, seeds, and floral support structur es. All annual plants flowered. None of the perennial species flowered during this e xperiment. Measurements of architectural parameters such as the length of the main stem and the number of branches were taken during the harvest. Multiple measurements ( N = 10) of internode lengths were taken throughout the structure of each individual. Belowground biomass was also collected from an annual, D. densiflora , and a perennial, D. cornutissima , for a comparison of whole-plant allocation patterns. Root

PAGE 31

19 tissue was carefully separated from the substrate by sifting with plastic mesh screens. Roots were not observed along the base or sides of the pots during the harvest; therefore, no individuals were pot-bound. Collected root tissue was rinsed with water to remove remaining substrate material. Plant tissues were dried at 60 C for one week before mass was measured. Leaf Gas Exchange and Morphology Measurements Measurements of ecophysiological traits were begun on all species after two months of growth (beginning of October) under common garden conditions. Data were obtained first from annual species to insure completion of measurements before flowering. Gas exchange measurements for light response curves were collected from fully expanded, high-light acclimated leaf tissue utilizing a portable infrared gas analyzer (LI-COR 6400 gas exchange system, Lincoln, NE, USA). An LED light source was used to span photosynthetic photon flux densities (PPFD) from 0 to 1800 mol m-2 s-1. Measurements began at the highest light level and were decreased 200 mol m-2 s-1 for each data point. The light-saturated rate of photosynthesis (Amax) was measured at 1800 mol m-2 s-1. Five to six measurements were taken as light intensity decreased from 120 mol m-2 s-1 for calculation of light compensation point, apparent quantum efficiency, and dark respiration. During all measurements, the airflow rate was 300 mol s-1 with a supplied CO2 concentration of 400 mols mol-1 and a relative humidity between 50 and 60%. Leaf temperature was maintained between 29 and 32 C. All measurements were completed during the morning hours. The curve-fitting program, Photosyn Assistant (Dundee Scientific, Scotland, UK), was utilized to estimate the light saturation point ( mol m-2 s-1), light compensation point ( mol m-2 s-1), and dark respiration rate ( mol

PAGE 32

20 m-2 s-1) of each light response curve using a non-rectangular hyperbolic equation. Three to five curves were determined for each species. For the perennial species, three to six leaves were measured together for each curve due to their small size. Measurements of leaf size (cm2), specific leaf area (cm2 g-1; SLA), and relative water content (RWC) were taken on six outer-canopy leaves from each of five plants for each species on October 23. After three days without rain or a water application, leaf tissue was collected late in the afternoon, placed in humidified plastic bags, and fresh weight was measured immediately. For RWC calculations, leaves were then placed in a dark, humidified chamber (cooler) for eight hours. Immediately after obtaining the saturated weights of these leaves, leaf size was measured using a leaf area meter (LICOR 3100C, Lincoln, NE, USA). Leaf tissue was dried at 60 C for one week before the dry weight was measured. Additional leaf tissue was collected from each plantÂ’s outer canopy for tissue analysis. Individual leaf tissues of each plant were combined, as this material was ground for tissue analysis. The % nitrogen (% of dry mass; %N) of leaf tissue for each plant was measured with an NCS 2500 automatic elemental analyzer (Carlo Erba Instruments, Thermo Quest Italia SpA, Milan, Italy). Leaf N was calculated on a mass (g N g-1 leaf) and area (g N m-2) basis. Leaf Anatomical Measurements Leaves from different positions within the canopy were randomly collected from each plant at the end of October and preserve d in sealed glass vials containing a solution of FAA (9:0.5:0.5, 70% ethanol:glacial acetic acid:40% formalin). This tissue was prepared for use in acquiring anatomical measurements 1.5 years later. Leaves were

PAGE 33

21 soaked in water to remove excess FAA. Transverse sections of individual leaves were made carefully by hand using a razor blade. Leaf tissues were stained with phloroglucinol and HCl (12N solution). Ex cess liquid was removed. These sections were positioned on a microscope slide. Several drops of glycerol were added and a coverslip was placed over the immersed tissue. Clear fingernail polish was used to form a seal around the edges of the cover slips to create temporary slides. An ocular micrometer calibrated with a stage micrometer was used in conjunction with a compound microscope to view and measure leaf anatomical traits of the species at 400x. Photographs were taken using an adapter and a Nikon digital camera. Data was collected from five leaves from each of five individuals per species for D. densiflora , D. linearifolia var. robustior , D. frutescens subsp. frutescens , and D. thinicola . However, measurements were obtained from five l eaves from each of four individuals for D. cornutissima and three individuals for D. christmanii . Statistical Methods Analyses of plant characteristics measured in the common garden were performed using the GLM procedure for simple ANOVAs and the MIXED procedure for nested design ANOVAs in SAS (Version 8.2; SAS Institute; NC, USA). The level of significance ( alpha ) for all tests was 0.05. When a significant difference in speciesÂ’ responses was found, TukeyÂ’s HSD (for equal sample sizes) or Tukey-Kramer (for unequal sample sizes) multiple comparison tests were utilized to determine which species expressed significantly different values.

PAGE 34

22 Results Light Availability and Leaf Traits within Native Habitats SpeciesÂ’ native habitats exhibited 28.95% ( 5.05 SD) canopy openness in the sandhill microhabitat occupied by D. densiflora versus 46.49% ( 10.69 SD) openness in the scrub microhabitat occupied by D. cornutissima . Areas inhabited by D. cornutissima exhibit overall higher values and a greater range in the % of canopy openness. Leaf area index (LAI; plant leaf area : ground area) of the canopy above individuals of these species was 1.22 ( 0.22 SD) for D. densiflora and 0.65 ( 0.27 SD) for D. cornutissima . These differences in light availability are reflected in the leaf size and specific leaf area (SLA; cm2 g-1) of these species within their native habitats. The leaves of D. densiflora were 1.21 cm2 ( 0.37 SD) while the leaves of D. cornutissima were 0.21 cm2 ( 0.05 SD). SLA was found to be 286.63 cm2 g-1 ( 32.26 SD) for D. densiflora and 126.20 cm2 g-1 ( 25.27 SD) for D. cornutissima . Growth Habit Annual species native to the sandhill habitat ( Dicerandra densiflora and D. linearifolia var. robustior ) flowered and set seed during the experiment. All stems of the annual species were determinate. Individuals of these species that were not harvested died after completion of seed formation. All perennials native to scrub ( Dicerandra cornutissima , D. frutescens subsp. frutescens , D. thinicola , and D. christmanii ) did not flower during this first growth season. All stems of the perennial species were indeterminate. Three individuals of each perennial were maintained within a greenhouse through the winter and placed back in the fiel d plot at the beginning of the next growth season. All perennials did flower at the end of the second growth season.

PAGE 35

23 Plant Biomass Allocation Dicerandra species exhibited significant differences in biomass allocation to various plant tissues ( P < 0.002; Table 2-1) while aboveground biomass was similar for all species ( P = 0.0890). Highly significant differences in leaf allocation were found ( P < 0.0001). All perennials allocated about 60% of their biomass to leaves. Annuals allocated significantly less to leaf ti ssue. The leaf mass ratio (LMR) of D. linearifolia var. robustior (0.1280 g g-1) was half that of D. densiflora (0.2278 g g-1). Stem mass ratio (SMR) did vary between the species ( P = 0.0011). The least allocation to stem tissue was found in D. densiflora (0.2649 g g-1). Among the perennial species, similar stem allocation was found in D. thinicola (0.3384 g g-1). Dicerandra linearifolia var. robustior exhibited a higher stem allocation (0.3492 g g-1) that was comparable to that of the perennial species. However, the ratio of leaf to stem tissue in annuals was 1:1 for D. densiflora and 1:3 for D. linearifolia var. robustior while it was 3:2 for D. cornutissima and 2:1 for all other perennials. Annual species allocated approximately 50% of their biomass to reproductive tissues (floral mass ratio; FMR) while perennial species only allocated resources to vegetative tissue production during this first growth season ( P < 0.0001). Aboveand belowground tissues were collected from D. densiflora and D. cornutissima to allow for a comparison of whole-plant biomass allocation (Fig. 2-2). Differences in leaf, stem, and floral mass ratios at the whole-plant level were similar to results of the aboveground comparison ( P < 0.05). Dicerandra densiflora allocated overall significantly less to leaf and stem tissue due to its large allocation to reproduction. Unexpectedly, the root mass ratios (g g-1; RMR) of this annual and perennial were not significantly different ( P = 0.4304) under the abundant supply of water and nutrients in

PAGE 36

24 this common garden study. The RMR of D. densiflora was 0.2165 g g-1 versus 0.2438 g g-1 in D. cornutissima . Shoot size (height and width) of D. densiflora at the end of this study was similar to larger individuals within the sandhill, while shoot size of D. cornutissima was noticeably much greater under the conditions within this common garden than shoot size obtained after one year of growth in scrub (observation). Shoot Architecture Significant differences in plant architecture were found among these species of Dicerandra (Table 2-2). Annuals were significantly taller than perennial species and significantly different from each other ( P < 0.0001). Comparison of main stem lengths found D. linearifolia var. robustior to be the tallest (52.6 cm) and D. cornutissima to be the shortest (10.1 cm). Annual species produced fewer, but longer (data not shown) branches in comparison to perennial species. Annuals had a more erect canopy structure consisting of only primary and secondary branches. Perennials exhibited a more prostrate growth with their large number of shorter branches. Tertiary branches were found in all perennials. A few quaternary bran ches were also formed in individuals of D. cornutissima and D. frutescens subsp. frutescens . A highly significant difference ( P < 0.0001) between annual and perennial species was found in the number of branches produced per gram of stem tissue. Growth during this study led to the creation of 5.7 branches g-1 stem tissue in D. linearifolia var. robustior and 10.5 branches g-1 stem tissue in D. densiflora . However, all perennial species produced approximately 100 branches g1 stem tissue. Internode length in annual species was significantly greater than perennial species and significantly different from each other ( P < 0.0001). Internode length was longest in D. linearifolia var. robustior (1.91 cm) and shortest in D. cornutissima (0.77 cm).

PAGE 37

25 Leaf Morphology and Physiology Significant differences in leaf ecophysiological characteristics were found among the Dicerandra species (Table 2-3). Leaf size (cm2) differed among species in a similar manner as found for individuals growing within their native habitats. There were highly significant differences in leaf size ( P < 0.0001). The annual species, D. densiflora (0.81 cm2) and D. linearifolia var. robustior (0.95 cm2), had the largest leaves. Perennial species displayed varying degrees of smaller leaf size. Leaves of D. frutescens subsp. frutescens (0.45 cm2) and D. thinicola (0.49 cm2) were approximately half the size of the annualsÂ’ leaves. However, leaves of D. cornutissima (0.26 cm2) and D. christmanii (0.21 cm2) were half the size of the other perennialsÂ’ leaves. Significant differences in specific leaf area (cm2 g-1; SLA; P < 0.0001) were found amongst the continuous range of values displayed by the species. Dicerandra linearifolia var. robustior (193.56cm2 g-1) had the largest SLA while D. frutescens subsp. frutescens (118.53cm2 g-1) had the smallest SLA. Mass-based nitrogen content of leaf tissue (g N g-1 leaf; Nmass) was similar for all species ( P = 0.2840). However, area-based nitrogen content (g N m-2 leaf; Narea) differed significantly among species ( P < 0.0001) with the lowest values in the annual species and the highest values in D. cornutissima and D. frutescens subsp. frutescens . Relative to the annual species, all perennials exhibited significantly higher relative water content (RWC) of leaf tissue ( P < 0.0001). Light response curve parameters were very similar among these species (Table 24). Amax-area ( mol m-2 s-1) was not significantly different ( P = 0.3246) among species, but Amax-mass ( mol kg-1 s-1) of D. linearifolia var. robustior was significantly greater than that of D. christmanii ( P = 0.0495). Light saturation point ( mol m-2 s-1) was similar ( P

PAGE 38

26 = 0.5688) for all species. Although perennial species tended to express higher light compensation points ( mol m-2 s-1) and dark respiration rates ( mol m-2 s-1), there were few significant differences. Only D. thinicola had a light compensation point higher ( P < 0.0145) than the annual species. Dark respiration rates of D. frutescens subsp. frutescens and D. christmanii were significantly greater ( P < 0.0126) than D. linearifolia var. robustior . Leaf Anatomy Dicerandra species exhibited significant differences in leaf anatomical characteristics (Fig. 2-3; Tables 2-5 and 2-6). Perennial species produced significantly thicker leaves than the annual species ( P < 0.0001; Table 2-5). Leav es of perennials were approximately 100 m thicker than those of annuals. The adaxial palisade mesophyll consisted of two cell layers in perennials and one cell layer in annuals ( P < 0.0001). Thickness of this tissue varied significantly ( P < 0.0001) throughout the species with the thinnest adaxial palisade formed in D. linearifolia var. robustior (98 m) and the thickest in D. cornutissima (151 m). Annual species exhibited bilateral leaf stru cture with leaf tissue consisting of an adaxial layer of palisade mesophyll and an abaxial layer of spongy mesophyll (Table 25). However, perennial species displayed diffe rent degrees of isobilateral leaf structure due to the formation of an abaxial layer of palisade-like tissue, also two cell layers thick. Mesophyll cells along the abaxial leaf surface of perennials extended vertically (columnar) with fewer intercellular spaces in comparison to the typical irregularly-shaped spongy mesophyll cells present at the abaxial surface of a bilateral annual leaf (observation). Dicerandra thinicola , which had a high SLA similar to the annual D.

PAGE 39

27 densiflora , displayed the greatest variation in mesophyll development. Some leaves of D. thinicola had well-developed abaxial palisade-like mesophyll while others displayed somewhat horizontally extended mesophyll cells along the abaxial surface with larger intercellular spaces that began to resemble the spongy mesophyll of the annuals. However, adaxial and true abaxial palisade were virtually indistinguishable in all leaves of Dicerandra christmanii . The thickness of this abaxial palisade-like or true palisade layer did not vary significantly between the species, but was thinnest in D. thinicola . The amount of palisade tissue in perennial leaf structure was twice that of the annuals due to the presence of abaxial palisade-like tissue or true palisade. Small variations among the species in spongy mesophyll thickness were significantly different ( P = 0.0063). Dicerandra densiflora leaves displayed the thickest (113 m) and D. frutescens subsp. frutescens (86 m) and D. thinicola (82 m) displayed the thinnest layers of spongy mesophyll (Table 2-6). These differences in mesophyll structure led to a total palisade:spongy mesophyll ratio of 2.4 ( D. christmanii ) to 2.9 ( D. cornutissima ) for perennial species versus 1 in annual species ( P < 0.0001). Cuticle thickness varied among the species (Table 2-6). Adaxial cuticle thickness ranged from thinnest to thickest as follows: D. cornutissima < D. densiflora < D. linearifolia var. robustior < D. christmanii < D. thinicola , D. frutescens subsp. frutescens ( P < 0.0001). Abaxial cuticle thickness of D. densiflora was significantly less than that of all other species ( P < 0.0001). The abaxial cuticle thicknesses of D. linearifolia var. robustior and D. cornutissima were similar but less than the thicknesses of the other perennial species. Dicerandra densiflora had an adaxial cuticle thickness (6 m) three times that of its abaxial cuticles (2 m). Dicerandra cornutissima and D. christmanii had

PAGE 40

28 similar cuticle thicknesses along both surfaces. The other species had substantial abaxial cuticles, but slightly thicker adaxial cuticles. Thickness of the epidermis (excluding the cuticle) did vary among the species. Epidermal layers were one cell thick in all species. Variation in speciesÂ’ adaxial and abaxial epidermal cell thickness ( P < 0.0001) and width ( P < 0.0001; data not shown on table; contact author for information) displayed a trend toward smaller epidermal cells in perennial species. Discussion Whole plantand leaf-level traits significantly varied among these Dicerandra congeners. Differences among species within this common garden were significant although light, water, and nutrients were equally and abundantly supplied to all individuals. Therefore, these variations in ecophysiological traits are due to genotypic differences among species, not acclimation responses. Plants are affected by a combination of environmental factors in their native habitats (Chapin et al., 1987; Dickison, 2000) . Light is more abundant in scrub communities where perennials are found; however, nutrients are scarce, and water availability is spatially and temporally limited over the lifetime of these species in comparison to sandhill resource availability during the annual speciesÂ’ single growth season. Overall, traits of the Dicerandra species native to scrub in this study are common adaptations to open, high-light environm ents when compared to responses of the species native to sandhill (Boardman, 1977; Bjorkman, 1981; Smith, 1982; Givnish, 1987; Ballare et al., 1988; Givnish, 1988; Sc hmitt and Wulff, 1993; Hutchings and de Kroon, 1994). Many scrub speciesÂ’ traits are also considered adaptations to the limited

PAGE 41

29 availability of water or nutrients (Givnish and Vermeij, 1976; Esau, 1977; Cunningham et al., 1999; Dickison, 2000; Arntz and Delph, 2001). Divergence in Growth Habit, Biomass Allocation, and Shoot Architecture Growth habits of the species in their native habitat were also exhibited in the common garden. Individuals of D. densiflora and D. linearifolia var. robustior produced determinate shoots and flowered during this study. In sandhill, all individuals flower regardless of size at the end of their only growth season (observation). Only indeterminate shoots were produced by the other species during this first growth season. Smaller, yet older, individuals of these pe rennial species growing within scrub produce flowers (observation). However, all perennials within this common garden did flower the second year. An age, rather than size, requirement may exist for flowering in these perennial scrub species of Dicerandra . Measurements above plants in their native habitats did find relatively greater light availability in the open scrub habitat of D. cornutissima relative to the sandhill understory of D. densiflora . However, light quality as well as quantity may be different in these habitats. As part of the sandhill understory, annuals compete with neighboring plants and each other for light that should have lower red:far red (R:FR) availability. Contrary to annuals, perennials establish in scrub gaps with few or no competitors; therefore, light quantity and R:FR should be greater. Differences in plant morphology and shoot architecture were found between annual and perennial species. Annual traits resembled those of plants acclimated or adapted to understory environments with low R:FR light availability although individuals in this experiment were grown in an open field (Smith, 1982; Corre, 1983; Ballare et al., 1988; Schmitt and Wulff, 1993; Hutchings and de Kroon, 1994; Anten, 2005). Annuals

PAGE 42

30 allocated proportionally more of their vegetative mass to stem production than perennials when the ratio of leaf to stem tissue of each species was compared (Table 2-1; Smith, 1982; Anten, 2005). In particular, Dicerandra linearifolia var. robustior produced three times more stem than leaf tissue. Future studies should determine biomass allocation patterns when both species are producing only vegetative tissue to ensure that this difference was not due to growth of stem support structures for reproduction in the annuals. Greater height (main stem length) and internode lengths were found in annuals with expression of these traits being significantly greater in D. linearifolia var. robustior (Table 2-2; Ballare et al., 1988; Anten, 2005). Less branching occurred in these sandhill species (Schmitt and Wulff, 1993; Anten, 2005). These differences were apparent between the annuals and perennials in the first few months of growth (observation) while all species were producing only vegetative tissues. Competition with understory neighbors for light within sandhill habitats may have lead to the evolution of these traits in annual species of Dicerandra (Morgan and Smith, 1981; Anten, 2005). The increase in canopy height of annuals caused by expression of these traits may allow for increased light interception along the forest floor. Gr owth of perennials with their many shorter branches and shorter internodes allows for more allocation to leaves and greater canopy extension along the ground. Variation in Leaf Morphology, Physiology and Anatomy Ecophysiological leaf traits of Dicerandra congeners differed along a continuum with perennials exhibiting, overall, varying degrees of sun-adapted traits (Clements, 1905; Hanson, 1917; Boardman, 1977; Bjorkma n, 1981; Givnish, 1988). Leaves of all perennial species were smaller (less leaf area) and thicker than leaves of annual species (Tables 2-3 and 2-5; Clements, 1905; Hanson, 1917; Garnier, 1992; Givnish, 1988;

PAGE 43

31 Lambers et al., 1998; Dickison, 2000). Lower SLA (more mass per unit area) and higher Narea values have been found in sunversus shade-acclimated leaves of many species (Bjorkman, 1981; Lambers et al., 1998) and in comparisons of perennial versus annual grass species (Garnier, 1992; Garnier et al., 1997). Differences in SLA and Narea were not exhibited as clearly between these closely-related species of Dicerandra with different native habitats and growth habits (Table 2-3). A trend of lower SLA was displayed by Dicerandra perennials with the exception of D. thinicola , possibly due to variability in its development of an abaxial layer of palisade-like tissue. With similar Nmass for all species, a trend of higher Narea was exhibited by all perennials given their thicker leaves. Smaller and thicker leaves with lower SLA are also considered adaptations to limited water and nutrient availability (Givnish, 1987; Esau, 1977; Cunningham et al., 1999; Dickison, 2000; Wright et al., 2002). Higher Narea (i.e. greater Rubisco) appears to enhance water conservation during photosynthesis in species adapted to drier, high-light habitats (Gutschic k, 1999; Cunningham et al., 1999; Wright et al., 2001). Comparison of annual and perennial light-response curves did not find clear expression of contrasting adaptations to lowand high-light availability. Higher Narea in leaves of sun-adapted species typically leads to a greater Amax-area and a higher dark respiration rate (per area; Chapin et al., 1987; Lambers et al., 1998; Wright et al., 2001). Measurements from Dicerandra congeners found greater dark respiration rates and light compensation points in some perennials, but not greater Amax-area (Table 2-4). These results may be due to a combination of differences in speciesÂ’ leaf traits and capability of the gas exchange equipment. The LI-COR 6400 gas exchange system and opaque

PAGE 44

32 chamber used to obtain these measurements illuminates only one leaf surface (adaxial) with an LED. Unlike the bilateral leaf structure of annual species, perennial Dicerandra have isobilateral leaves with palisade-like or true palisade mesophyll along both leaf surfaces. In this common garden, perennials held their leaves somewhat vertically at varying angles so that abaxial leaf surfaces were illuminated while annuals held leaves more horizontally (observation). A few studi es have compared photosynthetic rates of horizontally and vertically held leaves (De Lucia et al., 1991; Evans et al., 1993). When only adaxial surfaces are illuminated, horizontal leaves consistently have slightly higher Amax-area than vertical leaves. When responses of adaxial and abaxial illumination are compared, vertical leaves exhibit similar Amax-area for both surfaces while substantially lower values are found for horizontal leaves illuminated abaxially. These responses occur in both bilateral and isobilateral leaves, but isobilateral leaves exhibit the greatest similarity in Amax-area of both surfaces. Therefore, illumination of one surface of the perennialsÂ’ isobilateral leaves may not allow accurate measurement of maximum photosynthetic rates at each light level. Illumination of both leaf surfaces may yield significantly greater Amax-area, light saturation points, and light compensation points in the perennial species as expected in sun-adapted species (Boardman, 1977; Givnish, 1988; Lambers et al., 1998). Species of Dicerandra exhibited divergent differences in leaf morphology and anatomy. Annual species of Dicerandra , which are native to the understory of sandhill habitats, displayed traits typical of shade-adapted species (Clements, 1905; Hanson, 1917; Givnish, 1987; Lambers et al., 1998). Th e greater surface area relative to volume of leaf tissue found in the annual species due to their larger leaf size, but thinner lamina,

PAGE 45

33 enhances light interception (Tables 2-5 and 2-6). The bilateral (dorsiventral) leaf structure observed in annuals allows for proportionally greater amounts of spongy mesophyll tissue whose irregularly shaped cells increase light scattering to enhance light absorption (Fig. 2-3; Esau, 1977; Terashima and Saeki, 1983; Vogelmann et al., 1996). Annual species maximized light interception of adaxial leaf surfaces by holding leaves horizontally (observation; Givnish, 1987; Givnish, 1988). These traits are also commonly exhibited by species native to areas with higher rainfall and greater nutrient availability (Givnish, 1987). Perennial species, which are native to gaps within scrub habitats, displayed leaf traits of sun-adapted species (Clements, 1905; Hanson, 1917; Valladares and Pearcy, 1998). The relatively smaller and thicker leaves of perennial species native to high-light environments may allow for maximization of carbon gain and minimization of photoinhibition by avoidance of interception of excess radiation (Valladares and Pearcy, 1998). The comparatively thinner boundary layers of these smaller leaves decrease differences between leaf and air temperature due to greater heat loss through convection and evaporation (Givnish and Vermeij, 1976; Schuepp, 1993). Positioning leaves within the canopy at an angle as seen in Dicerandra perennials (observation) also protects plants from experiencing photoinhibition or over-hea ting by decreasing the proportion of leaf surface area receiving direct sunlight during the warmest part of the day (Geller and Smith, 1982; Valladares and Pearcy, 1998). In full sun, photoinhibition may occur and lead to a reduction in photosynthetic carbon gain especially when other stresses (i.e. drought and limited nutrients) are present. Therefore, the presence of these traits in scrub natives should be functionally advantageous.

PAGE 46

34 Perennial Dicerandra displayed leaf structural characteristics typically found in plants adapted or acclimated to high-light environments (Clements, 1905; Hanson, 1917; Ashton and Berlyn, 1992; Ashton and Berlyn, 1994; Smith et al., 1998; Dickie and Gasson, 1999; Dickison, 2000). Greater leaf thickness has been correlated with increases in palisade tissue in many species from a diversity of high-light habitats. The comparatively thicker leaves of perennial Dicerandra exhibited a significantly greater amount of palisade tissue due to a larger number of cells making up the adaxial palisade and formation of an abaxial palisade layer (Fig. 2-3; Tables 2-5 and 2-6). These vertically extended, columnar-shaped palisade cells facilitate light penetration so that light is distributed more evenly throughout thicker leaf tissue (Vogelmann and Martin, 1993; Vogelmann et al., 1996; Evans, 1999). Smith et al. (1998) presented evidence that isobilateral leaf structure and vertical leaf orientation have evolved in concert in species belonging to multiple plant communities in Western Australia. Dicerandra christmanii exhibited the most highly developed abaxial palisade layer and the greatest leaf angles in this common garden (observation). The isobilateral leaf structure of perennial Dicerandra allows leaves to gather incident or re flected light at any angle from both leaf surfaces. This may allow these scrub endemics to absorb light reflected off of the white or yellow sands of scrub habitats. Observations during anatomical measurements found that Dicerandra annuals have hypostomatic leaves (abaxial stomata) while perennials have amphistomatic leaves (stomata on both leaf surfaces). Studies revealed a correlation between development of stomata on both leaf surfaces and increasing leaf thickness (Parkhurst, 1978; Mott et al., 1982). The most important resource influencing leaf thickness is light (Nobel and

PAGE 47

35 Hartsock, 1981). Adaptation or acclimation to high-light results in an increased leaf thickness (Clements, 1905; Hanson, 1917; Bjorkman, 1981; Mott and Michaelson, 1991; Smith et al., 1998). Species adapted to partial or full shade typically have thinner, hypostomatic leaves while species adapted to high-light habitats have thicker, amphistomatic leaves (Mott et al., 1982). A study of leaf acclimation in Ambrosia cordifolia to various light levels found that abaxial stomatal density decreased as adaxial stomatal density increased under greater light availability; therefore, amphistomatic leaves do not necessarily have greater stomatal density (Mott and Michaelson, 1991; Beerling and Kelly, 1996). Parkhurst (1978) pr oposed that the development of stomata on both leaf surfaces is an adaptation to reduce the internal diffusion distance (mesophyll resistance) of CO2 in thick leaves. Many adaptations of leaf traits to high-light environments are correlated with drought avoidance (Clements, 1905; Esa u, 1977; Ashton and Berlyn, 1992; Dickison, 2000). Clements (1905) found that an increase in light or a decrease in water availability led to a reduction in leaf surface area and an increase in leaf thickness. Greater development of isobilateral leaf structure, fewer intercellular spaces, and proportionally less irregularly-shaped spongy mesophyll cells are found in sun-adapted or sunacclimated leaves. This leaf architecture promotes greater water-use-efficiency by reducing transpirational water-loss (Clements, 1905; Dickison, 2000). Although resource availability was the same for all species in this common garden, perennial Dicerandra , which are native to high-light microhabitats within scrub communities, displayed these traits and exhibited significantly greater RWC of leaf tissue in contrast to measurements from the annuals (Table 2-3). High-light has the greatest influence on development of

PAGE 48

36 isobilateral leaf structure; however, limited water availability is required for extreme development of the abaxial palisade layer (Clements, 1905; Hanson, 1917; Nobel and Hartsock, 1981). Dicerandra thinicola is the only species in this study that is native to a coastal area. The higher humidity of its native habitat may explain why this species displays the greatest variability in development of the abaxial pseudopalisade. The presence of other leaf anatomical characteristics may confer drought tolerance. Smaller cells, such as the smaller epidermal cells found in the perennials, have been linked to drought tolerance in plants (Lee et al., 2000). Debate still occurs around whether or not thicker cuticles reduce water loss (Dickie and Gasson, 1999). Not all species native to dry environments express this trait. Substantial cuticle thickness is present on adaxial and abaxial leaf surfaces of all perennial species (Table 2-6). The two perennial Dicerandra displaying the greatest leaf angles, D. christmanii and D. cornutissima , have similar cuticle thicknesses along both leaf surfaces. This pattern is not consistent in the annual congeners. While D. linearifolia var. robustior has a thick protective layer on both leaf surfaces, D. densiflora lacks a thick cuticle on its abaxial surface. This variation seems associated with predominant environmental conditions in their native habitats. Dicerandra linearifolia var. robustior grows in sandhill habitats with relatively well-drained sandy soils. However, Dicerandra densiflora is native only to sandhill habitats where a layer of clay with high soil water-holding capacity is present approximately 20 cm below the soil surface (observation; Huck, 1987). Conclusions Growth of Dicerandra species within this common garden found inherent differences in their growth habits. Examina tion of plantand leaf-level traits within species of Dicerandra revealed a strategy shift between species native to scrub versus

PAGE 49

37 species native to sandhill habitats. The perennial species which are native to scrub gaps exhibit biomass allocation patterns, shoot architecture, leaf morphology, and leaf anatomy typical of plants acclimated and adapted to open, high-light environments (Clements, 1905; Hanson, 1917; Lambers et al., 1998). Corresponding traits of the annual species native to the understory of sandhill habitats are similar to those found in plants acclimated and adapted to understory, shaded environments. These inherent variations in form may confer functional advantages to species under the environmental conditions of their native habitat. Many of the high-light traits of Dicerandra scrub natives also correspond to evolutionary divergences in species in response to decreasing rainfall and nutrient availability (Givnish and Vermeij, 1976; Esau, 1977; Cunningham et al., 1999; Dickison, 2000; Arntz and Delph, 2001). Water and nutrients were abundantly supplied to individuals within this common garden study. Future work with these closely related congeners from contrasting environments will examine how variation in water and nutrient availability affects speciesÂ’ responses and trait expression.

PAGE 50

38Table 2-1. Aboveground biomass accumulation and allocation of Dicerandra species grown in a common garden. Mean and SE values were calculated using data collected from 5 individua ls of each species except for D. densiflora whose N = 4. Means followed by different letters within a column indicate significant differences at = 0.05 utilizing Tukey-Kramer. Species Aboveground dry weight (g plant-1) Leaf mass ratio (g g-1) Stem mass ratio (g g-1) Floral mass ratio (g g-1) Annual mean SE mean SE mean SE mean SE D. densiflora 9.115 a 1.691 0.2278b 0.0209 0.2649 a 0.0132 0.5073a 0.0309 D. linearifolia v. robustior 7.966 a 1.348 0.1280a 0.0104 0.3492 b 0.0155 0.5228a 0.0169 Perennial D. cornutissima 11.838 a 0.505 0.5981c 0.0160 0.4019 b 0.0160 0.0000b 0.0000 D. frutescens 9.359 a 1.132 0.6302c 0.0194 0.3698 b 0.0194 0.0000b 0.0000 D. thinicola 10.703 a 1.772 0.6616c 0.0176 0.3384 ab 0.0176 0.0000b 0.0000 D. christmanii 6.723 a 0.772 0.6510c 0.0212 0.3489 b 0.0212 0.0000b 0.0000 P value 0.0890 < 0.0001 0.0011 < 0.0001

PAGE 51

39Table 2-2. Architectural parameters of Dicerandra species grown in a common garden. Means followed by different letters within a column indicate significant differences at = 0.05 utilizing Tukey-Kramer. # of branches Species Main stem length (cm) primary secondary tertiary quaternary Internode length (cm) # of branches per gram stem dry weight Annual mean SE mean SEmean SE mean SE mean SE mean SE mean SE D. densiflora 31.7 b 1.7 14a 2 12a 7 0a 0 0a 0 1.37b 0.09 10.5a 2.6 D. linearifolia v. robustior 52.6 a 4.0 11a 2 5a 3 0a 0 0a 0 1.91a 0.08 5.7a 1.0 Perennial D. cornutissima 10.1 c 1.3 47c 3 294c 30 156b 47 1a 1 0.77c 0.06 106.3b 13.0 D. frutescens 12.4 c 2.0 26ab 4 177bc 7 128ab 45 7a 7 0.88c 0.02 105.0b 10.0 D. thinicola 12.6 c 1.7 32bc 6 227bc 37 92ab 32 0a 0 0.87c 0.07 98.1b 14.2 D. christmanii 11.3 c 1.3 35bc 3 161b 36 33ab 19 0a 0 0.79c 0.05 100.1b 15.7 P value < 0.0001 < 0.0001 < 0.0001 0.0045 0.3533 < 0.0001 < 0.0001

PAGE 52

40Table 2-3. Leaf ecophysiological characteristics of Dicerandra species grown in a common garden. Mean and SE values were calculated using data collected from 6 leaves of each of 5 individuals per species ( N = 30) except for nitrogen content of leaf tissue. For that trait N = 8 for all species except D. linearifolia v. robustior ( N = 10) and D. christmanii ( N = 5). Means followed by different letters within a column indicate significant differences at = 0.05 utilizing Tukey-Kramer. Species Leaf size (cm2) SLA (cm2 g-1) Nmass (g N g-1 leaf) Narea (g N m-2 leaf) Relative water content Annual mean SE mean SE mean SE mean SE mean SE D. densiflora 0.81 a 0.20 160.14b 5.27 0.0241a 0.0019 1.502a 0.121 0.887a 0.017 D. linearifolia v. robustior 0.95 a 0.29 193.56a 8.92 0.0282a 0.0022 1.455a 0.114 0.910a 0.017 Perennial D. cornutissima 0.26 c 0.01 130.73bc 5.74 0.0306a 0.0026 2.339bc 0.202 0.945b 0.009 D. frutescens 0.45 b 0.02 118.53c 5.69 0.0287a 0.0017 2.419c 0.143 0.946b 0.008 D. thinicola 0.49 b 0.02 160.71b 8.07 0.0293a 0.0014 1.821ab 0.088 0.962b 0.010 D. christmanii 0.21 c 0.02 136.45bc 7.60 0.0259a 0.0016 1.900abc 0.117 0.950b 0.013 P value < 0.0001 <0.0001 0.2840 < 0.0001 < 0.0001

PAGE 53

41Table 2-4. Light response curve components of Dicerandra species grown in a common garden. Means followed by different letters within a column indicate significant differences at = 0.05 utilizing Tukey-Kramer. Species Amax-area ( mol m-2 s-1) Amax-mass ( mol kg-1 s-1) Saturation point ( mol m-2 s-1) Light compensation point ( mol m-2 s-1) Dark respiration ( mol m-2 s-1) Annual n mean SE mean SE mean SE mean SE mean SE D. densiflora 3 17.7 a 1.3 301.4ab 26.2 768a 40 60.4a 6.5 2.20ab 0.57 D. linearifolia v. robustior 5 19.5 a 1.2 377.1a 25.8 951a 102 67.5a 8.5 2.02a 0.22 Perennial D. cornutissima 3 17.8 a 2.2 304.0ab 54.5 930a 123 95.5ab 7.8 3.07ab 0.25 D. frutescens 3 21.6 a 1.6 367.4ab 41.3 849a 71 92.1ab 18.0 3.67b 0.55 D. thinicola 4 16.5 a 1.8 265.9ab 32.5 971a 31 124.2b 20.3 3.31ab 0.41 D. christmanii 3 16.9 a 2.2 220.7b 36.9 977a 103 118.3ab 13.9 3.70b 0.24 P value 0.3246 0.0495 0.5688 0.0145 0.0126

PAGE 54

42Table 2-5. Leaf anatomical characteristics of Dicerandra species grown in a common garden (part 1). Mean and SE values were calculated using data collected from 5 leaves of each of 5 individuals per species (N = 25) except for D. cornutissima (4 individuals, 5 leaves from each) and D. christmanii (3 individuals, 5 leaves from each). Means followed by different letters within a column indicate significant differences at = 0.05 utilizing Tukey-Kramer. Species Lamina thickness ( m) Adaxial # of palisade layers Abaxial # of palisade layers Adaxial palisade thickness ( m) Abaxial palisade thickness ( m) Annual mean SE mean SE mean SE mean SE mean SE D. densiflora 261 a 8 1.00a 0.00 0.00a 0.00 106 ab 4 0a 0 D. linearifolia v. robustior 245 a 9 1.00a 0.00 0.00a 0.00 98 a 3 0a 0 Perennial D. cornutissima 382 b 11 2.19b 0.04 2.10bc 0.03 151 c 6 100b 5 D. frutescens 358 b 11 2.00b 0.00 2.08bc 0.07 134 c 5 94b 4 D. thinicola 346 b 10 2.01b 0.01 2.01b 0.01 135 c 5 88b 4 D. christmanii 374 b 11 2.07b 0.07 2.29c 0.15 127 bc 5 104b 5 P value < 0.0001 < 0.0001 < 0.0001 < 0.0001 < 0.0001

PAGE 55

43Table 2-6. Leaf anatomical characteristics of Dicerandra species grown in a common garden (part 2). Mean and SE values were calculated using data collected from 5 leaves of each of 5 individuals per species ( N = 25) except for D. cornutissima (4 individuals, 5 leaves from each) and D. christmanii (3 individuals, 5 leaves from each). Means followed by different letters within a column indicate significant differences at = 0.05 utilizing Tukey-Kramer. Species Spongy mesophyll thickness ( m) Total palisade:spongy mesophyll tissue Adaxial cuticle thickness ( m) Abaxial cuticle thickness ( m) Adaxial epidermal thickness ( m) Abaxial epidermal thickness ( m) Annual mean SE mean SE mean SE mean SE mean SE mean SE D. densiflora 113 a 6 0.950a 0.052 6ab 0.2 2 a 0.2 24a 1.2 16a 1.1 D. linearifolia v. robustior 108 ab 5 0.911a 0.030 7bc 0.4 5 b 0.3 20ab 1.2 16a 1.1 Perennial D. cornutissima 89 ab 5 2.866b 0.149 5a 0.2 5 b 0.2 15c 0.6 12bc 0.5 D. frutescens 86 b 6 2.739b 0.157 9c 0.5 7 c 0.6 16bc 0.9 12bc 0.6 D. thinicola 82 b 7 2.830b 0.202 9c 0.6 7 c 0.4 14c 0.8 10c 0.7 D. christmanii 97 ab 6 2.401b 0.142 8c 0.6 7 c 0.4 16bc 0.8 15ab 1.3 P value 0.0063 < 0.0001 < 0.0001 < 0.0001 < 0.0001 < 0.0001

PAGE 56

44 Annual species: Annual species: D. D. linearifolia linearifolia var. var. robustior robustior D. D. densiflora densiflora Perennial species: Perennial species: D. D. cornutissima cornutissima D. D. frutescens frutescens subsp subsp . . frutescens frutescens D. D. christmanii christmanii D. D. thinicola thinicola Figure 2-1. Distribution of Dicerandra species utilized in this study (Compiled from Huck 1987).

PAGE 57

45 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1D. densifloraD. cornutissimaOrgan weight ratio Flower and Seed Leaf Stem Root Figure 2-2. Aboveand belowground biomass allocation of an annual ( D. densiflora ) and a perennial ( D. cornutissima ) species grown in a common garden.

PAGE 58

46 ab cd e f Figures 2-3. Photographs of leaf transverse sections of each species at 200x. Scale bars = 100 m. a. D. densiflora . b. D. linearifolia var. robustior . c. D. cornutissima . d. D. frutescens subsp. frutescens . e. D. thinicola . f. D. christmanii.

PAGE 59

47 CHAPTER 3 RESPONSES TO VARIATION IN NUTRIENT AVAILABILITY OF AN ANNUAL AND A PERENNIAL CONGENER NATIVE TO DIFFERENT FLORIDA HABITATS Introduction Growth, maintenance, and reproduction are the three component processes of an organismÂ’s life-history (Cody, 1966; Gadgil a nd Bossert, 1970; Bazzaz et al., 1987). To increase its fitness in a specific environment, an organism must optimally allocate limited resources among these competing processes (Cody, 1966; Bazzaz et al., 2000). Tradeoffs between traits promoting fecundity and those promoting survival exists (Crawley, 1977; Pitelka, 1977; Bazzaz et al., 2000). Responses of species to environmental (abiotic and biotic) factors affect their growth and development. Semelparity is favored when juvenile survivorship is high while iteroparity is selected if juvenile survivorship is low (Sibly and Calow, 1983). A number of factors including temperature regime, photoperiod (Kim and Okubo, 1995), and water availability (Morishima et al., 1984) exert influence over growth form expression. Studies have also found that nutrient availability can affect plant characteristics and expressed life-history traits (Hirose and Kachi, 1982; Kachi and Hirose, 1983a; Muller and Garnier, 1990). Perennial, rather than annual, species dominate infertile habitats (Grime and Hunt, 1975; Chapin, 1980). A plantÂ’s ability to optimally allocate resources between life history processes to successfully complete its life cycle within a single growth season is d ecreased when nutrients limit growth (Hirose and Kachi, 1982; Kachi and Hirose, 1983a; Bazzaz et al., 1987; Bazzaz and Morse,

PAGE 60

48 1991). Under nutrient-poor conditions, allocating resources to increase plant survival can be of greater benefit to overall plant fitness if seedling establishment is rare (Schaffer and Gadgil, 1975). Perennial species usually express lower relative growth rates than annual species when congeners are compared in controlled environments (Grime and Hunt, 1975; Pitelka, 1977; Muller and Garnier, 1990; Garnier, 1992; Garnier and Vancaeyzeele, 1994). The expression of perennial life-hi story traits, and corresponding morphological and physiological attributes, may allow for an advantage in growth, maintenance, and reproduction under resource-limited conditions (Schaffer and Gadgil, 1975). However, in a resource-rich environment supporting numerous competitors, the comparatively faster resource sequestration of an annual growth habit may allow for optimal allocation and greater reproduction. For this study, trait expression of two congeners from the genus Dicerandra (Benth.) was measured at various nutrient availabilities. Dicerandra (Lamiaceae) consists of closely related species displaying different growth habits in specific ecosystems (Huck, 1987; Huck et al., 1989). Herbaceous annual species range throughout sandhill habitats of the southeastern United States. Perennial species are suffrutescent evergreens native to unique scrub habitats in Florida along ancient dune ridges within the interior of the peninsula and in dune systems along the Atlantic coastline (Huck, 1987; Huck et al., 1989; Christman and Judd, 1990). The annual D. densiflora and the perennial D. cornutissima were utilized in this study. Recent molecular studies have found that the annual and perennial taxa constitute sister clades; however, the perennial species, D. cornutissima , has chloroplast DNA sequences more closely related to annual than other perennial taxa (Oliveira et al., in press).

PAGE 61

49 Both scrub and sandhill habitats inhabited by these species of Dicerandra have nutrient-poor soils. Relict beach ridges supporting scrub habitats with D. cornutissima are composed predominately of pure quartz sand entisols with no organic horizon and are considered low in nutrients, notably phosphorus (Willis and Yemm, 1961; Kachi and Hirose, 1983b; Abrahamson et al., 1984; Christman and Judd, 1990; Myers, 1990; Menges and Gallo, 1991; Anderson and Menges, 1997; Brady and Weil, 1999; Menges, 1999). Ground cover and litter are sparse in these scrub habitats where herbaceous species are rare. Areas supporting sandhill vegetation where D. densiflora is native consist of ultisols with a greater amount of organic material and a layer of clay approximately 20cm below the soil surface (Myers, 1990). These habitats have a great diversity of herbaceous species, including many grasses, in the understory and an overstory of widely spaced longleaf pine ( Pinus palustris ) (Myers, 1990; Carrington and Keeley, 1999). Therefore, litter accumulation is greater in these sandhill habitats. Although both ecosystems are pyrogenic, low inte nsity fires occur within sandhill every 1 to 10 years while high intensity fires move through scrub every 10 to 100 years (Myers, 1990). An increase in soil phosphorus availability and tissue phosphorus concentrations has been found in Florida sandhill species following controlled burns (Anderson and Menges, 1997). Dicerandra densiflora and D. cornutissima were grown under various levels of nutrient availability in this study to examine the responses of these closely related species in relation to their different life-history st rategies. Trade-offs between expression of traits optimizing greater seed production in the annual versus traits promoting greater growth and maintenance to ensure survival and future fecundity in the perennial were

PAGE 62

50 expected. To test these predictions, a factorial experiment was conducted utilizing three levels of nitrogen (N) and three levels of phosphorus (P). To include effects of the annualÂ’s reproductive phase on growth and allocation, this experiment was designed to continue until the annual species was near comp letion of its life cycle. Comparison of the variation in attributes of these closely related species is expected to provide insight into the significance of their divergence in form and function. Materials and Methods Species and Seed Source The most southern annual species, Dicerandra densiflora , and the most northern perennial species, D. cornutissima , were utilized for this research to control for the influence of climate on speciesÂ’ traits (Pitelka, 1977). Dicerandra densiflora is an herbaceous annual with determinate shoots. Dicerandra densiflora grows in northcentral Florida within the understory of sa ndhill communities and along the edges of this habitat next to roadways and trails. Seedlings are more abundant in sections recently disturbed (within 1 or 2 years) by fire and road/trail construction and maintenance. Dicerandra cornutissima is a suffruticose, shrubby perennial with determinate reproductive shoots and indeterminate, overwintering vegetative shoots extending from a ramose base (Huck et al., 1989). These shrubs typically grow along the edges and within gaps of an oak scrub community limited to a small area within Marion County, Florida. Seeds of D. densiflora were collected in the fall of 2000 and 2001 from sites near the Santa Fe River within OÂ’Leno State Park in High Springs, Florida with permission of the Florida Department of Environmental Protection, Division of Recreation and Parks (permit # 10160012). Seeds of D. cornutissima were collected in the fall of 2000 from the Marjorie Harris Carr Greenway and Trails property located southwest of Ocala,

PAGE 63

51 Florida on the Sumter Upland in Marion County. A permit from the Division of Plant Industry (permit# 413) within the Florida Department of Agriculture and Consumer Services was required to collect seeds of this species due to its endangered status. Plant Culture Seeds were germinated over a three-week period in plastic petri dishes on moistened heavy-weight germination paper placed in a growth chamber supplying 10 hours of light and 14 hours of darkness at 25 C day and night. Seedlings were planted in 142 ml pots containing the substrate mixture described below and allowed to grow for four weeks in a greenhouse. Nutrients were applied twice using a water-soluble fertilizer. Seedlings were transferred to the field site and placed under neutral density shade cloth for two weeks to allow for acclimation to differences in humidity and light. Components of the substrate mixture were chosen to mimic characteristics of the substrate found within the speciesÂ’ native habitats. The mixture contained 60% sand, 20% bark from Pinus species, 10% vermiculite, and 10% perlite. Sand, virtually free of nutrients, was purchased from the Feldspar Corporation in Edgar, Florida where it was mined, heated, and pressure washed. Chemical analysis after these treatments found 99.5% SiO2, 0.20% Al2O3, and 0.05% Fe2O3. Vermiculite was included in the mix to allow for greater water and nutrient retention. Perlite and bark were used to prevent soil compaction and allow for airflow through the substrate. Pine bark introduced acidic elements that are present in the speciesÂ’ native habitats. Garden Experiment A factorial randomized complete block design with three levels of N and three levels of P was utilized. Seedlings were randomly assigned to a block and a treatment. Each treatment within the two blocks contained four D. densiflora and three D.

PAGE 64

52 cornutissima seedlings. The experiment was conducted at the Bivens Arm research site in Gainesville, Florida during the fall of 2002. Each seedling was transferred into a 2.84 l pot containing the substrate mixture with a specific N and P availability. Seedlings were randomly placed throughout the field site 65 to 70 cm apart. Plants were well-watered throughout the experiment every two or three days. To slow nutrient leaching, care was taken to not over-saturate pots with water. Levels of N and P were selected for each treatment to allow for 1.0, 2.5, or 5.0 g of plant growth assuming 50% uptake efficiency. These calculations of N and P requirements for this amount of plant tissue growth were made assuming that 1.5% of dry plant tissue is N and 0.2% is P (Salisbury and Ross, 1992). Nitrogen quantities for the various levels were 88, 221, and 441 mg. Phosphorus level quantities were 9, 22, and 44 mg. These availabilities correspond to 27, 69, and 137 ppm of N, and 3, 7, and 14 ppm of P. Aqueous solutions of anhydrous NH4NO3 were applied five times, every two weeks, during the course of the study. Solutions with the proper concentrations for each N level were made the day of application. Pots were watered prior to the N application to prevent immediate leaching of this applied nutrient. The first application was given after seedlings were transferred into 2.84 l pots. P2O5, other macronutrients, and micronutrients were added to the substrate at the beginning of the experiment. Potash (K2O), CaSO4, and minor elements were supplied at non-limiting levels, assuming 7.5 g of plant growth with 50% uptake efficiency. Harvesting Plants of both species grew under nutrient treatments for 14 weeks. The harvest started when D. densiflora began to set seed. Plant material was separated into different tissues. The floral mass ratio (g g-1) included inflorescences and seeds. Basal diameter

PAGE 65

53 measurements were collected from all plants using digital calipers. The leaf area of a subsample of randomly chosen leaves from each plant was measured using a leaf-area meter (LI-COR Inc., Model 3100C, Lincoln, Nebraska). Leaf subsamples consisted of 10 leaves for D. densiflora and 15 leaves for D. cornutissima . This information was used to calculate the specific leaf area (cm2 g-1; SLA) after the leaf subsamples were dried and weighed. Root tissue was separated from the substrate by sifting with plastic mesh screens. No individuals were pot-bound. Coll ected root tissue was rinsed with water to remove remaining substrate material. Plant tissues were dried at 60 C for a week before mass was measured. Tissue Analysis Percent nitrogen of each tissue was measured on ground material of each plant with a NCS 2500 automatic elemental analyzer (Carlo Erba Instruments, Thermo Quest Italia SpA, Milan, Italy). The instrument was calibrated using a pine needle standard (National Institute for Standards and Testing, Gaithersburg, MD, USA). Calculations utilizing the %N and dry biomass of each organ allowed for determination of total plant N content (mg) and nitrogen-use efficiencies relative to whole plant N (mg biomass produced mg-1 plant N; NUEN-plant) and only leaf N (mg biomass produced mg-1 leaf N; NUEN-leaf) for each plant. Trait Expression of Species in Native Habitats The relative growth rates (g g-1 d-1; RGR) of individuals gr owing within their native habitats were determined nondestructively. Basal diameter measurements of individuals were collected from 28 D. densiflora and 38 D. cornutissima plants using digital calipers at the beginning and end of a three-week period in September. Basal diameter measurements of plants in our nutrient study and their corresponding total dry plant

PAGE 66

54 biomass (g) were utilized to calculate the allometric relationship between these two parameters for each species using a power equation. Utilizing each speciesÂ’ allometric relationship, the change in basal diameter measurements of individuals growing in their native habitats over time allowed for estimation of each individualÂ’s increase in dry mass. Leaf subsamples were collected from several individuals in their native habitats to allow for determination of each speciesÂ’ SLA. Leaves were collected from 22 D. densiflora and 24 D. cornutissima plants. Leaf areas were measured and dry mass was obtained. Statistical Analysis Analyses of plant characteristics expressed by individuals grown under the nutrient treatments were performed using the MI XED procedure in SAS (Version 8.2; SAS Institute; NC, USA). Fixed main effects of the model included species, N level, P level, and the interactions N x P, species x N, species x P, and species x N x P. Random effects included block and the interaction of block x species x N x P. Alpha level was set at 0.05. When a significant difference in speciesÂ’ responses were found, TukeyÂ’s HSD (equal sample sizes) or Tukey-Kramer (unequal sample sizes) multiple comparison tests were utilized to determine which species expressed significantly different values. Profile analyses of continuous variables displayed in this chapterÂ’s figures were used to show fixed effects of N and P. Linear regr ession analysis was conducted using the REG procedure in SAS to determine the number of flowers produced per gram of plant biomass. An F -test was used to discern if regression coefficients were significantly different.

PAGE 67

55 Results Overall Growth Increases in N availability led to greater biomass accumulation for both species ( F = 48.09, df = 2, 17, P < 0.0001; Fig. 3-1a). However, species responses were different ( F = 42.46, df = 1, 17, P < 0.0001) as Dicerandra cornutissima plants gained more mass than D. densiflora . An increase in biomass was found when N availability increased from 88 mg (low level) to 221 mg (medium level) for D. cornutissima (Tukey-Kramer multiple comparison P = 0.0003). However, a significant difference for D. densiflora occurred only when N increased from 88 mg to 441 mg (high level; Tukey-Kramer P = 0.0014). Dicerandra cornutissima plants grown at high N gained more mass than D. densiflora plants at any N level (Tukey-Kramer P 0.0005). Overall, these differences in species responses to N availability were indicated by a significant species x N interaction ( F = 4.76, df = 2, 17, P = 0.0229) . The significant effect of phosphorus ( F = 3.92, df = 2, 17, P = 0.0399) was due to the total biomass of D. cornutissima plants grown under medium N and high P being greater than plants of D. cornutissima grown with low N, but similar to plants grown with high N. This result may have been an artifact due to random variation in plant sizes at the beginning of the study. Total Plant N Total plant N (mg) varied significantly between species ( F = 14.70, df = 1, 17, P = 0.0013; Fig. 3-1b). Increases in N availability ( F = 78.93, df = 2, 17, P < 0.0001) lead to increases in N uptake of both species. An increase in total plant N of D. cornutissima was found with each increase in N level (Tukey-Kramer P < 0.005). Plant N of D. densiflora only increased at high N availability (Tukey-Kramer P < 0.005). Although no

PAGE 68

56 differences between species were apparent at low N, total plant N of D. cornutissima was greater than that of D. densiflora at high N (Tukey-Kramer P < 0.0281). Therefore, D. cornutissima was able to uptake significantly more N as availability increased. Effect of Nutrient Availability on Annual Reproduction The number of flowers produced by D. densiflora did significantly increase as N availability increased ( F = 13.78, df = 2, 17, P = 0.0003; Fig. 3-1c). However, the number of flowers produced per gram of plant dry mass did not change with variation in nutrient availability (data not shown); therefore, N and P availability did not affect reproductive efficiency. Flower number was positively correlated with total plant biomass in a linear relationship ( y = 145.28 x + 9.1984; r2 = 0.9063; RMSE = 18.10). The perennial species, Dicerandra cornutissima , did not reproduce during this first growth season. Biomass Allocation Significant shifts in biomass allocation occurred in D. cornutissima , but not D. densiflora , as N availability changed (Figs. 3-2). Species ( F = 726.21, df = 1, 17, P < 0.0001) exhibited different root mass ratios (g g-1, RMR; Fig. 3-2a). At low N 50% of D. cornutissima biomass was root tissue while the RMR of D. densiflora was only 25%. Increases in N availability ( F = 38.10, df = 2, 17, P < 0.0001) led to decreases in the RMR of D. cornutissima , but not D. densiflora . This difference in speciesÂ’ RMR response to increases in N was indicated by a significant species x N interaction ( F = 14.20, df = 2, 17, P = 0.0002). An increase in stem mass ratio (g g-1, SMR; Fig. 3-2b) was found only between the responses of D. cornutissima at low versus high levels of N (Tukey-Kramer P = 0.0120). This small difference in the SMR of the species due to variability in N level was evidence of a significant species x N interaction ( F = 7.26, df =

PAGE 69

57 2, 17, P = 0.0053). SpeciesÂ’ leaf mass ratios (g g-1; LMR; Fig. 3-2c) were different ( F = 640.30, df = 1, 17, P < 0.0001). The greatest LMR of D. cornutissima was twice the leaf tissue allocation of D. densiflora under all treatments. Increases in N availability ( F = 9.11, df = 2, 17, P < 0.0020) did lead to an increase in the LMR of D. cornutissima . Growth of D. cornutissima under medium (0.42 g g-1) and high N (0.45 g g-1) levels allowed for production of more leaf tissue than under low N (0.36 g g-1; Tukey-Kramer P < 0.0200). However, no differences were found in the LMR of D. densiflora (0.18 g g-1). These differences in speciesÂ’ LMR were indicated by a significant species x N interaction ( F = 9.30, df = 2, 17, P = 0.0019). Floral mass ratios (g g-1; FMR; Fig. 3-2d) of the species were different since no reproductive structures were produced by the perennial species ( F = 1479.42, df = 1, 17, P < 0.0001). However, Dicerandra densiflora , the annual species, allocated about 45% of its biomass to reproduction regardless of N and P availability. Further analysis utilizing only data from this annual species determined that the trend of increase in FMR with increasing N availability was not significant ( F = 2.35, df = 2, 8, P = 0.1572). Plant Nitrogen-use Efficiency Nitrogen-use efficiency calculations were based on biomass produced per unit of total plant N (mg biomass mg-1 plant N; NUEN-plant) and on biomass produced per unit leaf N (mg biomass mg-1 leaf N; NUEN-leaf) to determine differences in resource-use of the annual and perennial species. The length of this experiment did not allow for incorporation of nutrient loss and leaf lif espan into the calculation of nitrogen-use efficiency since leaf senescence was beginning in the annual at the end of the experiment and did not occur in the perennial (Aerts and Chapin, 2000). Significant differences in NUEN-plant were found between species ( F = 36.95, df = 1, 17, P < 0.0001) with D.

PAGE 70

58 cornutissima exhibiting overall greater NUEN-plant than D. densiflora (69.46 1.27 versus 58.83 1.20 mg biomass mg-1 plant N; Fig. 3-3a). NUEN-plant significantly decreased with each increase in N availability ( F = 111.81, df = 2, 17, P < 0.0001) for both species. Significant differences in NUEN-leaf were also found between species ( F = 53.12, df = 1,17, P < 0.0001). Dicerandra densiflora exhibited NUEN-leaf values more than twice the NUEN-leaf of D. cornutissima (338.47 19.96 versus 127.97 20.88 mg biomass mg-1 leaf N; Fig. 3-3b). These differences be tween species were significant at low and medium N availability (Tukey-Kramer P < 0.0200). Availability of N did affect NUENleaf ( F = 15.37, df = 2, 17, P = 0.0002). NUEN-leaf was significantly greater for D. densiflora at low N (Tukey-Kramer P < 0.0300); however, variations in NUEN-leaf with changes in N availability were not significant for D. cornutissima . N Concentration of Plant Tissues Nitrogen concentration (%N) of each plan t tissue significantly increased with increased N availability in both species. Differences were greater for D. densiflora than for D. cornutissima (Fig. 3-4). Differences in %N of root tissue were found between the species ( F = 13.91, df = 1, 17, P = 0.0017) and due to variation in N availability ( F = 49.76, df = 2, 17, P < 0.0001; Fig. 3-4a). High N availability allowed for an increase in %N of root tissue for both species. However, the overall N concentration of root tissue at high N in D. densiflora was greater than that of D. cornutissima (Tukey-Kramer P = 0.0056). This greater increase in %N of D. densiflora root tissue as N availability increased was indicated by a significant species x N interaction ( F = 3.81, df = 2, 17, P = 0.0430). Stem %N varied between species ( F = 6.77, df = 1, 17, P = 0.0186) and with N levels ( F = 16.41, df = 2, 17, P = 0.0001; Fig. 3-4b). Nitrogen concentration of D.

PAGE 71

59 densiflora stem tissue was greater at high N than for plants under low and medium N (Tukey-Kramer P < 0.0350). Nitrogen concentration of D. cornutissima stem tissue at high N availability was only greater than low N level plants (Tukey-Kramer P < 0.0500). Greater N availability ( F = 49.69, df = 2, 17, P < 0.0001) led to similar increases for both species in %N of leaf tissue (Fig. 3-4c). Higher N concentration of leaf tissue at high N was greater than the lower values of individuals at low and medium N levels (TukeyKramer P < 0.0300). Leaf %N doubled over this range of N availability (1.5 to 3.0%N). The %N of floral tissue in D. densiflora did increase with N levels ( F = 13.63, df = 2, 17, P = 0.0003; Fig. 3-4d). High N availability allowed for greater N accumulation in reproductive tissue (Tukey-Kramer P < 0.0040). Discussion An annual and a perennial congener from the genus Dicerandra were grown under various availabilities of N and P. Both species are native to nutrient-poor habitats. This research compared the traits of these closely related species to determine the significance of variation in their form and function. The difference in the growth habit of these species often requires expression of contrasting responses to maximize the fitness of each species under various nutrient availabilities. Trade-offs in traits expressed by these species increase overall plant fitness given the requirements of their growth habits. Growth and Biomass Allocation The perennial species exhibited greater biomass accumulation. Although growth of both species was similar under low N availability, Dicerandra cornutissima gained significantly more biomass as N availability increased (Fig. 3-1a). Ontogenetic shifts to seed formation may have constrained overall biomass production in D. densiflora , an annual that produces only deterministic shoots (Gadgil and Bossert, 1970; Crawley,

PAGE 72

60 1977; Pitelka, 1977; Abrahamson and Caswell, 1982; Bazzaz et al., 2000). As plants enter the adult reproductive phase, resources are transferred from growth of vegetative tissues to the development and growth of reproductive tissues. In habitats with a fixedlength growth season, the optimal allocation strategy for an annual requires a switch from purely vegetative to purely reproductive ti ssue production (Cohen, 1971). A fixed-length growth season may be experienced by D. densiflora in its native sandhill habitat given the environment’s climate and resource availability. All annual plants flowered during the course of this study regardless of size and nutrient availability. However, perennial species of Dicerandra do not flower in the first year even when nutrients are non-limiting (previous chapter; common garden study). An age requirement for flowering may exist for these species (observation in common garden study). Perennial Dicerandra are able to utilize all resources during their first growth season to maximize growth and maintenance to ensure survival and future reproductive outputs. Therefore, the perennial’s root syst em continues to expand and collect nutrients as more leaves develop and perform photosynt hesis. This uninterrupted growth of “source” tissues in the perennial versus the annual may explain the difference in biomass accumulation between these species. Similar results may be found in other comparisons of species with different growth habits if measurements are taken near the end of the annual’s life cycle. Only the perennial displayed plasticity in biomass allocation (Fig. 3-2). Dicerandra cornutissima exhibited significant shifts in biomass allocation with changes in N level. This perennial transferred res ources from root production to leaf and stem production as N availability increased (Hirose and Kitajima, 1986). With the increase in

PAGE 73

61 LMR and %N of leaf tissue, D. cornutissima was able to produce more biomass through greater photosynthesis as N availability increas ed (Figs. 3-1a, 3-2, and 3-4; Hirose and Kitajima, 1986). These changes in allocation signify plastic responses of D. cornutissima to continue growth while ensuring survival and future fecundity at various availabilities of N. No significant shifts in biomass allocation occurred in D. densiflora , the annual (Fig. 3-2). Dicerandra densiflora accumulated more N in its tissues and did not consistently increase biomass production (Figs. 3-1a and 3-4). However, flower number did increase with increases in N availability (Fig. 3-1c). For an annual, a plantÂ’s fitness is maximized when the greatest possible number of viable seeds is produced by the end of the growing season. Many studies have f ound that variation in reproductive output is more closely correlated to variation in plant size (mass), not variation in proportion of biomass allocated to reproduction (i.e. FMR; Bazzaz et al., 2000). As observed in previous studies of other annual species, the FMR of D. densiflora was relatively high, but not affected by changes in nutrient availability. However, increases in N availability resulted in an increase in the number of fl owers produced by each plant that corresponded to the increase in plant mass (Figs. 3-1a and c; Hirose and Kachi, 1982). Larger plants produced more seeds. Increases in N availability led to increases in seed production per plant through increases in plant biomass (Bazzaz et al., 2000). Nutrient Uptake and Use Dicerandra cornutissima was able to uptake larger quantities of N at high N availability given its greater allocation to root growth (Figs. 3-1b and 3-2a). Greater allocation to root growth in D. cornutissima was expected. In comparison to D. densiflora , Dicerandra cornutissima has a taproot that is more extensively branched with

PAGE 74

62 many spreading fibrous secondary roots (Huck et al., 1989). This difference may confer an ecological advantage to this species in its native habitat. Nutrients are limiting and leaching is high within the deep, sandy soils of scrub communities. Dicerandra cornutissima may increase nutrient absorption in scrub by providing a larger absorptive system that allows for interception of spatially unpredictable and relatively short-lived nutrient pulses year-round (Chapin, 1980; Crick and Grime, 1987). The smaller root system of D. densiflora may be adequate for its native sandhill habitat where an organic horizon may provide greater nutrient availability and the presence of a clay layer approximately 20 cm below the soil surface (observation) slows nutrient leaching (Chapin, 1980; Crick and Grime, 1987). Chapin postulated that species from infertile habitats would accumulate and store nutrients as availability increased rather than increase tissue growth (Chapin, 1980). This ‘luxury consumption’ did not occur in the scrub perennial. Although the %N of D. cornutissima tissues did increase with N availability, so did the growth of its indeterminate shoots (Figs. 3-1a and 3-4). Slightly greater N concentrations were found in D. densiflora as growth did not continually increase with increasing N availability. Growth may have been limited in this annual due to the production of only determinate shoots. In comparisons of other congeners, annual species have been found to have higher N concentrations in shoot tissues (Garnier and Vancaeyzeele, 1994; Padgett and Allen, 1999). NUEN-plant increased under low N availability for both species; however, the perennial exhibited higher NUEN-plant than the annual (Fig. 3-3a). Differences between the species’ NUEN-plant were caused by differences in tissue allocation (Fig. 3-2). At low

PAGE 75

63 N availability, Dicerandra cornutissima allocated about 50% of its resources to root tissue that has a low N concentration (1 to 1.5%; Figs. 3-2a and 3-4a). Developmental changes required in the annual for completion of its life-cycle led to Dicerandra densiflora allocating about 45% of its resources to reproductive tissues that have a higher N concentration (2 to 2.5%; Figs. 3-2d and 3-4d). NUEN-plant for both species was relatively low (Hiremath et al., 2002). Given the function of N in leaf tissue for performance of photosynthesis, leaf N is the most important to plant growth. Although the perennial exhibited greater NUEN-plant, the annual species was found to produce significantly more biomass per unit N invested in leaves (i.e. greater NUEN-leaf). The NUEN-leaf of D. densiflora was more than twice that of D. cornutissima . Only D. densiflora displayed plasticity in this physiological trait with significantly greater NUEN-leaf at lower N availability. Given the determinate growth of shoots in annual species, this plasticity in a physiological response of leaf tissue is highly beneficial for maximizing plant fitness. Growth of Species in Native Habitats Although Dicerandra cornutissima exhibited greater biomass accumulation in this experiment, its growth under natural conditions is slow compared to that of D. densiflora . Measurements of scrub plants found RGR of D. cornutissima in its native habitat to be extremely low at 0.0039 ( 0.0056 SD) g g-1 d-1. Measurements of sandhill plants found D. densiflora growing quickly in its native habitat at 0.0319 ( 0.0095 SD) g g-1 d-1. Leaf samples collected from individuals in the field found D. cornutissima to have a low SLA of 126.2 ( 25.3 SD) cm2 g-1 in scrub habitats and D. densiflora to have a high SLA of 286.6 ( 32.3 SD) cm2 g-1 in sandhill habitats. Past research has found the SLA of annual

PAGE 76

64 species to be consistently greater than that of perennial species (Muller and Garnier, 1990; Garnier, 1992; Garnier et al., 1997). Many studies have determined that variation in speciesÂ’ SLA explains their differences in RGR (Muller and Garnier, 1990; Poorter and Remkes, 1990; Garnier, 1992; Van der Werf et al., 1993; Atkin et al., 1996). Although RGR data was not collected duri ng this study, biomass accumulation and SLA were measured. Unexpectedly, Dicerandra cornutissima exhibited greater aboveground growth in this experiment than it does in its native habitat while the opposite was true for D. densiflora . The size of D. cornutissima plants grown under low N over the course of a few months resembled those in the scrub habitat after approximately two years of growth (observation). The size of D. densiflora plants under high N resembled smaller plants in sandhill habitats during the seed fill stage (observation). However, in this experiment, the SLAs were similar with 187.7 ( 13.2 SE) cm2 g-1 for D. cornutissima and 184.4 ( 13.2 SE) cm2 g-1 for D. densiflora . Variation in N and P availability did not cause significant changes in the SLA of these closely related species (data not shown; contact author). However, both species are able to respond to increases in nutrient availability by increasing their growth. Dicerandra cornutissima and D. densiflora grown under unlimited availability of all macroand micronutrients in a common garden during the same time period as this study were ten times the biomass of individuals under the greatest N level in this experiment with no significant differences in biomass accumulation or SLA of these species (previous chapter; common garden study). Differences in availability of other resources (i.e. water and light) within their native habitats may be the cause of the divergence in their field SLAs.

PAGE 77

65 Ecological Implications Within its native habitat, Dicerandra cornutissima establishes in gaps and along the edges of scrub where competition is low (Huck, 1987; Myers, 1990). Allocation in D. cornutissima during this study allowed for a RMR about two times greater than that of D. densiflora under low N availability. Given the low availability of nutrient and water resources within this community, this greater allocation to root growth seems ecologically advantageous for the survival of a perennial (Abrahamson et al., 1984; Myers, 1990; Menges and Gallo, 1991; Menges, 1999). Expression of a perennial growth habit may be another ecological advantage for D. cornutissima . This perennialÂ’s extensive root system and evergreen canopy allows for nutrient uptake, greater drought tolerance, and light interception throughout the year. The allocation and trait plasticity displayed by this perennial species may allow individuals to respond to the temporal variation in resource availability of scrub habitats with each change of season (Zangerl and Bazzaz, 1983). With low resource availability and persistence of gaps, this plastic perennial growth habit may confer greater growth, survival, and reproductive output in scrub from one plant over time versus that of an annual beginning each year as seed. Dicerandra densiflora grows in sandhill habitats with abundant understory species (Huck, 1987; Myers, 1990; Carrington and Keeley, 1999). Light is more limiting in sandhill relative to scrub (data presented in common garden chapter). The determinate growth habit of annuals may limit their ability to acclimate to changes in resources. However, steady tissue mass ratios would allow for consistent leaf and stem growth before flowering. This should be an advantage when competing against other understory species for light. Expression of the annual growth habit may be advantageous in sandhill where D. densiflora establishes in recently disturbed areas with few or no competitors. If

PAGE 78

66 these individuals were perennial, they would be shaded by other species by the end of their first growth season. An annual growth habit also allows this species to complete seed set before the drought season begins in November and avoid the freezing winter temperatures. Greater reproductive output may be more assured through this opportunistic annual growth habit in their native sandhill habitat. Conclusions Availability of N in this experiment significantly affected the growth of both species and led to significant differences in most traits. Nitrogen was more limiting than P given the lack of significant P effects or in teractions between N and P. Therefore, both the scrub and sandhill species of Dicerandra appear to be adapted to growth under low P availability. Future studies should determine if mycorrhizae colonize the roots of Dicerandra species, and what effects their presence has on plant responses to nutrient availability. The scrub perennial exhibited plasticity in its biomass accumulation, N uptake, biomass allocation, and NUEN-plant. These responses may be due in part to the indeterminate growth of the perennial form. It appears that this perennial gained more biomass than the annual at high N availability because the developmental changes (i.e. production of determinate shoots) required by the annualÂ’s life cycle constrained biomass production and plasticity. Differences in biomass allocation due to growth habit even led to a slightly greater NUEN-plant for D. cornutissima given its greater allocation to tissues with low N concentrations (i.e. roots). However, when nitrogen-use efficiency was expressed as biomass produced per unit leaf N (NUEN-leaf), only D. densiflora exhibited plasticity in this physiological trait with greater NUEN-leaf at lower N availability.

PAGE 79

67 Plasticity in this trait by the annual should allow for a greater increase in its fitness (seed production) even though its shoots are determinate. Both species exhibited plasticity in some plant traits that should result in increased plant fitness. Traits promoting greater growth and maintenance to ensure survival and future fecundity were found in the perennial. Flowers and seeds were produced by individuals of the annual species at all levels of N availability. The annual exhibited greater seed production at higher N availability. A perennial growth habit appears to be advantageous in scrub communities while an annual growth habit may be better suited for sandhill habitats. This experiment compared their traits later in life when ontogenetic differences were apparent to determine the ecological significance of trait variation in these species displaying different growth habits. Further study of these species should include a shortterm experiment with the same nutrient treatments to discern if the annualÂ’s juvenile (i.e. vegetative) stage exhibits the faster growth and plasticity of biomass allocation similar to that of the perennial. However, plant ecologists should compare closely related species with different growth habits at multiple times throughout their life cycle to document and understand the effects of changes due to ontogeny on their responses to resource availability.

PAGE 80

68 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 0100200300400500 0 5 10 15 20 25 30 35 0100200300400500Total plant biomass (g) Number of flowersTotal plant N (mg)Applied Nitrogen (mg) (a) (b) 0 50 100 150 200 250 0100200300400500(c) DEN 9mg P DEN 22mg P DEN 44mg PCOR 44mg P COR 22mg P COR 9mg P Figure 3-1. Total (a) plant biomass, (b) plant N, and (c) number of flowers produced by D. densiflora (DEN) and D. cornutissima (COR) when grown under each combination of N and P availability. Lsmeans are shown with error bars denoting ± 1 SE.

PAGE 81

69 0.00 0.10 0.20 0.30 0.40 0.50 0.60 0.00 0.05 0.10 0.15 0.20 0.25 0100200300400500 0.00 0.10 0.20 0.30 0.40 0.50 0.60 0.30 0.35 0.40 0.45 0.50 0.55 0.60 0100200300400500RMR (g g-1) FMR (g g-1)LMR (g g-1) SMR (g g-1)Applied Nitrogen (mg) DEN 9mg P DEN 22mg P DEN 44mg PCOR 44mg P COR 22mg P COR 9mg P(a) (b) (c) (d) Figure 3-2. Plant tissue mass ratios of (a) root, (b) stem, (c) leaf, and (d) floral structure tissues produced by D. densiflora (DEN) and D. cornutissima (COR) when grown under each combination of N and P availability. Lsmeans are shown with error bars denoting ± 1 SE.

PAGE 82

70 30 40 50 60 70 80 90 100 0100200300400500 0 100 200 300 400 500 600 700 0100200300400500 DEN 9mg P DEN 22mg P DEN 44mg P COR 9mg P COR 22mg P COR 44mg PApplied Nitrogen (mg)NUEN-leaf(mg mg-1N) NUEN-plant(mg mg-1N)(b) (a) Figure 3-3. Nitrogen-use efficiency of D. densiflora (DEN) and D. cornutissima (COR) when grown under each combination of N and P: (a) NUEN-plant mg biomass mg-1 plant N and (b) NUEN-leaf mg biomass mg-1 leaf N. Lsmeans are shown with error bars denoting ± 1 SE.

PAGE 83

71 0 0.5 1 1.5 2 2.5 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 0100200300400500 0 0.5 1 1.5 2 2.5 3 3.5 4 0 0.5 1 1.5 2 2.5 3 0100200300400500% N of root tissue % N of stem tissue % N of leaf tissue % N of floral tissueApplied Nitrogen (mg) DEN 9mg P DEN 22mg P DEN 44mg P COR 9mg P COR 22mg P COR 44mg P(b) (a)(c) (d) Figure 3-4. The %N of (a) root, (b) stem, (c) leaf, and (d) floral tissue of D. densiflora (DEN) and D. cornutissima (COR) when grown under each combination of N and P availability. Lsmeans are shown with error bars denoting ± 1 SE.

PAGE 84

72 CHAPTER 4 INFLUENCE OF INCREASING WATER-DEFICIT ON RESPONSES OF Dicerandra CONGENERS DIFFERING IN NATIVE HABITAT AND GROWTH HABIT Introduction FloridaÂ’s unique scrub communities are found only on ancient dune ridges within the interior of the peninsula and in dune systems along the Atlantic coastline (Christman and Judd, 1990; Myers, 1990). This shrubland is dominated by xeromorphic plants with characteristics such as short stature, evergreen canopies, small thick leaves, and high allocation to root systems (Menges, 1994; Menges, 1999). The proportion of endemism in Florida scrub is one of the highest in North American plant communities (Estill and Cruzan, 2001). An estimated 40 to 60% of extant scrub species are endemic (Christman and Judd, 1990; Myers, 1990). Few studies have been concerned with characterizing and understanding the importance of ecophysiological traits expressed by species native to these unique plant communities. Past studies of water availability effects on scrub vegetation have focused only on understanding the water relations of three sclerophyllous Quercus species. Their predawn water potentials indicated drought-stress during the dry season (Menges and Gallo, 1991). These species exhibited high bulk modulus of elasticity values and lower osmotic potentials at full and zero turgor when compared to average drought responses of Quercus species in North America (Abrams and Menges, 1992). During drought, oak species were able to uptake fog through their foliage and significantly increase predawn water potentials (Menges, 1994).

PAGE 85

73 The genus Dicerandra (Family Lamiaceae) consists of closely related species endemic to Florida scrub and various sandhill habitats within the southeastern United States (Huck, 1987; Huck et al., 1989). Each of five suffrutescent perennial species is limited to a small, geographically isolated scrub community within FloridaÂ’s peninsula. Dicerandra cornutissima is the northern-most ranging perennial. It is native to gaps within the matrix of an oak scrub community in central Florida. Four herbaceous annual species inhabit recently disturbed sandhill habitats throughout southern Georgia and northern Florida with smaller ranges in Alabama and South Carolina. Dicerandra densiflora is the southern-most ranging annual. It is native to the understory of sandhill habitats in north-central Florida. Scrub and sandhill habitats inhabited by these Dicerandra species differ to some extent in resource availability. Perennials grow in open, high-light environments within gaps of scrub while annuals inhabit more shaded conditions beneath pine canopies where they compete with other understory species. Both habitats have nutrient-poor soils. Relict beach ridges and dunes supporting scrub habitats are composed of predominantly quartz sand entisols with no organic horizon and are considered low in nutrients (Willis and Yemm, 1961; Kachi and Hirose, 1983b; Abrahamson et al., 1984; Christman and Judd, 1990; Myers, 1990; Menges and Gallo, 1991; Anderson and Menges, 1997; Brady and Weil, 1999; Menges, 1999). Substrate supporting sandhill vegetation consists of ultisols or spodosols with a layer of nutrient-rich organic material at the surface (Myers, 1990). Scrub and sandhill species of Dicerandra experience disparate levels of water availability in their native habitats during their lifetime due to differences in soil

PAGE 86

74 characteristics and expressed growth habits. The relatively pure and deep sand soils of scrub are excessively well-drained and ha ve low water-holding capacity (Myers, 1990; Menges and Gallo, 1991; Menges, 1999). A litter layer is typically not found in areas inhabited by perennial Dicerandra . Therefore, plants growing in scrub periodically contend with limited water availability due to soil properties and high evaporative demand. Populations of annual Dicerandra establish in recently disturbed areas of sandhill within a few hundred meters of streams or rivers. A layer of organic material along the surface increases the soil water-holding capacity of their microhabitats (Myers, 1990). The annual D. densiflora is native only to sandhill habitats where clay layers with high soil water-holding capacity are present approximately 20 cm below the surface (observation; Huck, 1987). Seeds of D. cornutissima and D. densiflora typically germinate after heavy rains at the end of spring (April or May; observation). Rainfall is steady in Florida from June until it begins to decline in October (Chen and Gerber, 1990). For habitats in north and central Florida, rainfall is lowest in Novemb er. These species begin flowering at the end of September or the beginning of October (observation). Seeds are filled by midNovember in the annual and as late as the end of November in the perennial (observation). The dry season begins in November and continues into late spring. Annual species of Dicerandra avoid this significant drought by completing their life cycle under terminal drought as the dry season begins. However, perennial species with evergreen vegetative shoots survive this dry season within the excessively, well-drained sandy soils (Huck, 1987; Chen and Gerber, 1990; Myers, 1990; Menges and Gallo, 1991).

PAGE 87

75 The study of Dicerandra species offers a unique opportunity to elucidate the importance of divergence in plant form and function. Physiological mechanisms explaining the survival and success of D. cornutissima in drought-prone scrub have not been determined. The objective of this study was to examine plantand leaf-level gas exchange responses of D. cornutissima and D. densiflora through an imposed dry-down. This is the first study to measure the effects of water availability on carbon gain and water loss in a species native to the unique scrub habitats of Florida. Since these closely related species experience different levels of water availability in their native habitat during their life cycle, variation in their responses to increasing water-deficit stress may provide insight into the value of these physiological mechanisms as strategies of specific adaptations to drought. Comparatively, the sandhill annual is expected to exhibit fast responses to decreasing water availability to avoid drought while the scrub perennial is expected to exhibit greater tolerance to drying soil. Decreases in plantand leaf-level gas exchange were expected to occur at higher soil water availability in D. densiflora while D. cornutissima will maintain high levels of gas exchange at lower water availability. Materials and Methods Species and Seed Source The most southern annual species, Dicerandra densiflora , and the most northern perennial species, D. cornutissima , were utilized for this research to control for the influence of climate on speciesÂ’ traits (Pitelka, 1977). Dicerandra densiflora is an herbaceous annual with determinate shoots. Dicerandra densiflora grows in northcentral Florida within the understory of sa ndhill communities and along the edges of this habitat next to roadways and trails. The substrate is composed of sand with an organic layer at the soil surface and a clay layer approximately 20 cm below the soil surface that

PAGE 88

76 allows for higher soil water-holding capacity (Huck, 1987). Seedlings are more abundant in areas recently disturbed (within 1 or 2 years) by fire and road/trail construction and maintenance. Dicerandra cornutissima is a suffruticose, shrubby perennial with determinate reproductive shoots and indeterminate, overwintering vegetative shoots extending from a ramose base (Huck et al., 1989). These shrubs typically grow along the edges and within gaps of an oak scrub community limited to a small area within Marion County, Florida. The pure sand soils of this habitat are excessively well-drained and have a lower water-holding capacity (Menges and Gallo, 1991). Seeds of D. densiflora were collected in the fall of 2000 and 2001 from populations near the Santa Fe River within OÂ’Leno State Park in High Springs, Florida with permission of the Florida Department of Environmental Protection, Division of Recreation and Parks (permit # 10160012). Seeds of D. cornutissima were collected in the fall of 2000 from populations within the Marjorie Harris Carr Greenway and Trails property located southwest of Ocala, Florida on the Sumter Upland in Marion County. A permit from the Division of Plant Industry (permit# 413) within the Florida Department of Agriculture and Consumer Services was required to collect seeds of this species due to its endangered status. Plant Culture Seeds were germinated in June of 2003 over a four-week period in plastic Petri dishes on moistened heavy-weight germination paper placed in a growth chamber supplying 12 hours of light and 12 hours of darkness at 25 C day and night. Seedlings were planted in 142 ml pots containing the substrate mixture described below and allowed to continue to grow in the growth chamber. A 25 ml solution of a water-soluble

PAGE 89

77 fertilizer (PeterÂ’s) with 20-20-20 % NPK concentrations was applied twice during this time. After expansion of eight to twelve leaves, seedlings were transplanted into 655 ml pots containing the substrate mixture and placed in a greenhouse maintained by the Botany Department of the University of Florida in Gainesville, Florida. Incandescent lamps were positioned above the plants to extend the photoperiod to 14 hours throughout the experiment to prevent reproductive tissue growth by the species. A 1.0 g application of a slow release fertilizer (Osmocote) with 14-14-14 % NPK concentrations was added to each pot to allow a dose of macronutrients with each water application. These seedlings were positioned beneath a 30% neutral density shade cloth for two weeks to allow for acclimation to the higher light and temperature conditions. After removal of the shade cloth, seedlings were allowed to further acclimate to full light conditions (1200 mol m-2 s-1) within the greenhouse for six weeks before beginning the experiment. Relative humidity in the greenhouse was on average 75%. All plants were well-watered during this growth period. Components of the substrate mixture were chosen to mimic characteristics of the substrate found within the speciesÂ’ native habitats. The mixture contained 60% sand, 30% vermiculite, and 10% perlite. Vermiculite was included in the mix to increase water and nutrient retention of the substrate. Perlite was used to prevent soil compaction and allow for airflow through the substrate. A layer of weed cloth was placed at the bottom of each pot to prevent substrate loss through the drainage holes.

PAGE 90

78 Characterization of Drought Stress Physiological ecologists have traditionally used leaf water potential as a measurement of plant water status (Nilsen and Orcutt, 1996; Sadras and Milroy, 1996). However, changes in stomatal response and conductance are more closely coupled to soil drying than leaf water potential particularly under moderate drought stress (Chaves, 1991). Studies of annual crop responses to increasing water-deficit conditions have utilized the fraction of transpirable soil water ( FTSW ) as an indicator of the soil waterdeficit (dryness) experienced by a plant (Sinclair and Ludlow, 1986; Muchow and Sinclair, 1991; Sadras and Milroy, 1996; Ray and Sinclair, 1997; Ray and Sinclair, 1998; Ray et al., 2002). Changes in physiological processes such as leaf expansion, wholeplant transpiration, leaf gas exchange, and nitrogen fixation rates have been related to decreases in FTSW (Sinclair, 1986; Sadras and Milroy, 1996). Use of FTSW in studies provides a non-destructive method to determine soil water availability and ensures that responses of experimental groups are compared at the same transpirable soil water content. Using FTSW to characterize the soil water-deficit that a plant experiences is comparable to predawn leaf water potential w ith only some variability between the values in the upper range of predawn values (Pellegrino et al., 2004). This technique was employed in this study to analyze the effect of increasing water-deficit conditions on gas exchange responses of D. cornutissima and D. densiflora . Experimental Design The dry-down experiment began the second week of November. Seedlings were randomly assigned to control and experimental groups. The well-watered control group consisted of three individuals for each species. The water-deficit experimental group contained three D. densiflora and four D. cornutissima individuals. A fresh 1.0 g

PAGE 91

79 application of the slow release fertilizer (Osmocote) was added to each pot. On November 8, all pots were saturated with water and allowed to drain overnight. The next morning before sunrise, all pots were sealed by placing each one in a white plastic bag, bunching up the plastic material around the main stem of each plant, and sealing the opening around the main stem by using a twist tie so as to permit water loss only by transpiration of shoot tissue. Pots were weighed at this time to obtain initial pot weight. Water was withheld from plants belonging to the water-deficit experimental group throughout the experiment. To maintain the well-watered conditions in control plants, the total amount of water lost every three days through transpiration was added back to the pots through a pipette. Pots were weighed each evening at approximately 18:00 h. Daily transpiration was calculated as the decrease in pot weight from one day to the next. These data were analyzed utilizing the procedure of Sinclair (Sinclair and Ludlow, 1986; Ray and Sinclair, 1997; Ray and Sinclair, 1998). For each species, daily transpiration rates of the waterdeficit plants were normalized against those of the well-watered plants to minimize the effects of variations due to fluctuations in environmental factors. The daily transpiration of each water-deficit plant was divided by the da ily average transpiration of the speciesÂ’ well-watered plants to obtain normalization. plants watered well of ion transpirat average plant deficit water of ion transpirat TR The transpiration rate of each plant varied due to differences in plant size; therefore, another normalization was necessary. The normalized transpiration of each water-deficit plant was fixed at a value of 1.0 during the first few days of the well-watered stage. The

PAGE 92

80 daily TR for each water-deficit plant was divided by itsÂ’ average TR during the wellwatered stage to calculate a daily normalized transpiration rate (NTR). Pot weight of water-deficit plants was measured every day, as water was lost through transpiration until NTR dropped below 0.10. An NTR of 0.10 was defined as zero transpirable soil water because it signified that stomatal conductance had decreased an order of magnitude below the value of well-watered control plants (Sinclair and Ludlow, 1986). The total amount of transpirable soil water available for each waterdeficit plant was determined as the difference between the initial pot weight and this ending pot weight. To facilitate comparisons between speciesÂ’ responses, the available soil water for each water-deficit plant on each day was expressed as the fraction of transpirable soil water. The daily FTSW was determined for each water-deficit plant as the daily pot weight minus the final pot weight divided by the total amount of transpirable soil water. weight pot final weight pot initial weight pot final weight pot daily FTSW daily The relationship between FTSW and NTR for each speciesÂ’ throughout the dry-down experiment was analyzed using non-linear re gression procedures in SAS (Version 8.2; SAS Institute; NC, USA) to fit the following equation: ) exp( 1 1 FTSW B A NTR This equation was found to fit well in studies using the same techniques (Muchow and Sinclair, 1991; Ray and Sinclair, 1998). Diffe rences between species were compared by means of the 95% confidence intervals of coefficients A and B. The FTSW threshold value (i.e. when NTR began to decline) was defined as the FTSW value when NTR had

PAGE 93

81 decreased to 0.95. Species’ differences in the B coefficient signify major differences in their FTSW thresholds and in how quickly their NTR decreases after that point. Species’ differences in the A coefficient only indicate minor variations in species’ FTSW threshold values. The amount of water transpired during the water-deficit experiment until NTR decreased to 0.10, the length of this dry-down period, and the number of days before death (survival period) after NTR decreased below 0.10 was determined for all plants of each species. Aboveground biomass was collected from each plant at the end of this experiment. This material was dried at 60 C for a week before measurements of mass were obtained. The GLM procedure for simple ANOVAs in SAS was utilized to determine if species’ responses were significantly different at the 0.05 ( ) level. Leaf Gas Exchange and Physiological Measurements The light-saturated rate of net leaf photosynthesis (Amax-area) was measured on three individuals of both species at several CO2 concentrations to estimate gas phase resistance (rg) and mesophyll resistance (rm) to CO2 diffusion. A portable infrared gas analyzer (LICOR 6400 gas exchange system, Lincoln, NE, USA) with an LED light source supplied a photon flux density of 1200 mol m-2 s-1, an airflow rate of 200 mol s-1, and 60-75% relative humidity within the chamber for all measurements. Jones (1985) Method V “differential” approach was used to determine rg and rm. Measurements were taken at ambient CO2 (380 mol mol-1) and at CO2 concentrations 25 and 50 mol mol-1 above and below the ambient CO2 concentration. Leaf temperature was kept between 28.0 and 29.5 C. The GLM procedure for ANOVAs in SAS was utilized to determine if species’ resistances were significantly different at the 0.05 ( ) level.

PAGE 94

82 Gas exchange measurements of leaf tissue from water-deficit and well-watered plants were taken every other day during the course of the water-deficit treatment. Data was collected from two D. densiflora and four D. cornutissima water-deficit plants. Outer-canopy, fully expanded leaves were marked the first day of data collection on all plants to insure that measurements were taken from the same leaves throughout the experiment. A portable infrared gas analyzer (LI-COR 6400 gas exchange system, Lincoln, NE, USA) with an LED light sour ce was used to obtain measurements at a photon flux density of 1200 mol m-2 s-1. During all measurements, the airflow rate was set at 200 mol s-1 with a supplied CO2 concentration of 380 mols mol-1. Leaf temperature was kept between 27.5 and 30.0 C. Relative humidity fluctuated between 50 and 60%. All measurements were completed in the morning. Photosynthesis ( mol CO2 m-2 s-1, Amax-area), transpiration (mmol H2O m-2 s-1, E ), and stomatal conductance (mol H2O m-2 s-1, gs) measurements were normalized twice and evaluated using nonlinear regression following the same protocol used to normalize plant transpiration rate. The FTSW thresholds for these measurements were calculated utilizing the method used for NTR . Leaves were collected from all plants at approximately 10:30 am multiple times throughout the experiment to estimate leaf relative water content (RWC) at different levels of FTSW . Each collection consisted of 2-3 leaves from D. densiflora plants and 35 leaves from D. cornutissima plants. Fresh leaves were placed in humidified plastic bags and taken to the lab where fresh weight was immediately measured. Leaves were then placed in a dark, humidified chamber (cooler) for six hours. After obtaining the saturated weight of these leaves, leaf tissue was dried at 60 C for one week before

PAGE 95

83 measuring dry weight. The previously descri bed protocol was followed to normalize leaf RWC responses to decreasing FTSW and determine the FTSW threshold of RWC for each species. The condition of each water-deficit plant was monitored daily. The onset and severity of leaf wilting was noted. Additional leaf samples from water-deficit plants were collected when leaves wilted, as well as from well-watered controls. This material was dried at 60 C and ground. Measurements of leaf 13C were obtained using a Costech elemental analyzer coupled with a Finnigan Delta XL Plus isotope ratio mass spectrometer. The 13C of well-watered controls and water-deficit plants after 2-3 days of wilting were compared. The GLM procedure for ANOVAs in SAS was utilized to determine if speciesÂ’ traits were significantly different at the 0.05 ( ) level. Results Plant-level Responses to Water-deficit A time course of plant transpiration (g) per unit leaf mass (g) found a trend of greater water-loss per unit leaf mass in D. densiflora at the beginning of the dry-down (Fig. 4-1). Transpiration per unit leaf mass did decrease rapidly over time for all D. densiflora individuals (Fig. 4-1a). Plant transpiration responses of D. cornutissima individuals did vary over time (Fig. 4-1b). However, D. cornutissima exhibited overall less water-loss per unit leaf mass rates at the beginning of the dry-down which decreased more slowly during the course of the dr y-down. The duration of the dry-down was shorter for D. densiflora ; however, the differences between species were not significant ( F = 2.91, df = 1, 5, P = 0.3890; Table 4-1). Significant differences were not found in the

PAGE 96

84 leaf mass ( F = 0.03, df = 1, 5, P = 0.8633) or shoot mass ( F = 0.02, df = 1, 5, P = 0.8924) of water-deficit plants belonging to the two species (Table 4-2). The NTR and corresponding FTSW values throughout the dry-down period were calculated for each plant. Use of these values for data analysis ensured that responses of individuals belonging to the experimental gr oups of both species were compared at the same transpirable soil water content ( FTSW ) with the effects of plant size and differences in daily transpiration responses due to variability in environmental factors removed through normalizations ( NTR ). Differences in the response of NTR to decreases in FTSW were found between the species (Fig. 4-2). The overlap of the 95% confidence intervals for A was substantial while the overlap for B was relatively small. Given this result and the correlation of A and B, another analysis was performed with the A coefficient fixed at a value of 5 for both species. A significant difference was found between the species’ B coefficients with D. cornutissima exhibiting a significantly smaller value ( D. densiflora , B = –8.21, 95% CI = –8.89 to –7.53; D. cornutissima , B = –10.60, 95% CI = –11.46 to –9.74; r2 = 0.99). The importance of this difference was apparent when the FTSW threshold of each species was calculated utilizing the values for A and B found in the first analysis. The FTSW threshold at which whole-plant relative transpiration begins to decrease was determined to be 0.547 for D. densiflora and 0.445 for D. cornutissima (Table 4-3). Other plant-level characteristics concerned with water-use and its overall affects revealed trends of higher values exhibited by D. cornutissima ; however, differences were not significant. The amount of water transpired by plants until NTR declined below 0.10 totaled 209 ( 11 SE) g for D. cornutissima and 195 ( 8 SE) g for D. densiflora ( F =

PAGE 97

85 0.89, df = 1, 5, P = 0.3890). After NTR declined below 0.10, Dicerandra cornutissima plants survived for 5.0 ( 1.00 SE) days and D. densiflora plants survived for 4.33 ( 0.33 SE) days (Table 4-1). Leaf-level Responses to Water-deficit Measurements from the two Dicerandra species were taken before the water-deficit treatment began to estimate gas phase resistance (rg) and mesophyll resistance (rm) to CO2 diffusion. Gas phase resistances were not si gnificantly different between the two species as rg was 10.57 ( 1.04 SE) m2 s mol-1 for D. cornutissima and 11.64 ( 1.61 SE) m2 s mol-1 for D. densiflora ( F = 0.31, df = 1, 4, P = 0.6085). Significantly greater mesophyll resistance was found in D. cornutissima ( F = 12.98, df = 1, 4, P = 0.0227). The mesophyll resistance of D. cornutissima was 17.42 ( 0.39 SE) m2 s mol-1 while that of D. densiflora was 10.27 ( 1.95 SE) m2 s mol-1. Absolute values of leaf Amax-area, E , and gs for each species under well-watered conditions were not significantly different (d ata not shown; P > 0.2400). Differences in normalized leaf-level gas exchange responses to decreasing FTSW were found between the species (Fig. 4-3). The 95% confidence intervals for speciesÂ’ A coefficients overlapped for normalized Amax-area, E , and gs; therefore, these coefficients were not significantly different. However, the B coefficients of D. cornutissima for normalized Amax-area and E were significantly greater than D. densiflora . No significant difference was found in the B coefficients for regression of normalized gs. The FTSW thresholds for each leaf-level gas exchange response were approximately 0.5 for D. densiflora and 0.2 for D. cornutissima (Table 4-3).

PAGE 98

86 Species’ FTSW thresholds at which leaf RWC began to decrease were not significantly different although species’ A and B coefficients were different (Fig. 4-4 and Table 4-3). Decreases in leaf RWC began at 0.086 FTSW for D. densiflora and 0.070 FTSW for D. cornutissima . Observations during the experiment noted that leaf wilting began in water-deficit plants at FTSW values just below the RWC threshold of each species (0.074 0.002 SE for D. densiflora and 0.055 0.008 SE for D. cornutissima ; F = 3.59, df = 1, 5, P = 0.1165). An exponential equation was used to model normalized Amax-area and E responses to RWC (Fig. 4-5). Species were compared using the fitted parameters for the exponents. The higher values for parameters of D. densiflora signify a greater decrease in gas exchange of this species with each decrease in RWC. Gas exchange of D. cornutissima was relatively less responsive to decreases in RWC. Dicerandra densiflora reduced both Amax-area and E at a higher RWC. When RWC was 0.8, the normalized rate of both gas exchange parameters was 0.6 for D. densiflora and 0.8 for D. cornutissima . These results are consistent with the species’ responses observed for normalized Amax-area and E to FTSW in which D. densiflora reduced Amax-area and E at higher FTSW . Leaves of Dicerandra densiflora plants under well-watered and water-deficit conditions had significantly different 13C (‰; F = 18.52, df = 1, 4, P = 0.0126). The 13C for plants under well-watered conditions was –30.47 ( 0.35 SE) ‰ while that of water-deficit plants was –28.86 ( 0.14 SE) ‰. Leaves from well-watered and waterdeficit D. cornutissima plants did not differ significantly in 13C composition ( F = 3.22, df = 1, 5, P = 0.1326). The 13C for leaf tissue of well-watered plants was –29.97 ( 0.60 SE) ‰ and that of water-deficit plants was –28.67 ( 0.51 SE) ‰. Species’ differences

PAGE 99

87 in the values of leaf 13C under each condition were not significantly different ( P > 0.0500). Discussion The main objective of this study was to compare physiological mechanisms utilized by two closely related species, D. densiflora and D. cornutissima , to cope with increasing water-deficit conditions. Differences in whole-plant transpiration relative to soil water availability were found between these closely related congeners. Compared to the responses of D. cornutissima, Dicerandra densiflora exhibited a decline in whole-plant transpiration at a higher level of water availability (i.e. higher FTSW threshold; Fig. 4-2 and Table 4-3). As soil dries, the supply rate of water to roots decreases. Plants typically increase stomatal resistance to decrease shoot transpiration and prevent dessication (Chaves, 1991; Ray et al., 2002). Measurements of leaf-level gas exchange responses in D. densiflora showed that stomatal conductance and corresponding transpiration and photosynthesis rates began to decline at the same FTSW as whole-plant transpiration, unlike the overall responses of D. cornutissima (Fig. 4-3). Dicerandra densiflora appears to exhibit greater sensitivity of leaves to decreases in water availability with its faster decline in stomatal conductance. This earlier stomata closure may have caused development of the significant difference between 13C composition of water-deficit and well-watered plants by increasing the water-use efficiency of the water-deficit plants. Plantand leaf-level gas exchange of D. cornutissima was unhindered by the drydown until reaching lower levels of water availability (i.e. lower FTSW thresholds; Figs. 4-2 and 4-3). Overall, decreases in planta nd leaf-level gas exchange did not occur at the same FTSW . One of the four plants did exhibit decreases in plantand leaf-level gas

PAGE 100

88 exchange at approximately the same FTSW value; however, average responses of D. cornutissima plants show that leaf-level measurements began to decline at substantially lower water availability (Table 4-3). Overall, Dicerandra cornutissima appears to be more tolerant of water-deficit conditions imposed by this dry-down experiment. This species kept stomata open and continued to gain carbon at lower FTSW values than D. densiflora . A possible hypothesis to explain these results is that turgor pressure was maintained at lower FTSW values in D. cornutissima through osmotic adjustment (McCree and Richardson, 1987; Chaves, 1991; Mulkey et al., 1991; Nilsen and Orcutt, 1996). In future studies, data for pressure-volume curves from well-watered and waterdeficit plants should be collected near the e nd of the dry-down period just before leaves wilt. Characterization of plantand leaf-level gas exchange responses to declining soil water availability has been performed on many annual crops (Sadras and Milroy, 1996). FTSW threshold values for these speciesÂ’ responses range from approximately 0.2 to 0.8. Usually, plant transpiration declines between 0.3 and 0.4 FTSW (Muchow and Sinclair, 1991; Sadras and Milroy, 1996; Ray and Sinc lair, 1997; Ray and Sinclair, 1998; Ray et al., 2002). There is debate in the literature as to whether thresholds are higher in coarser, sandy soils (Sadras and Milroy, 1996). Defining the lower limit of FTSW is considered difficult for sandy substrates. The substrate used in this study of Dicerandra congeners was composed of 60% sand to mimic the soil composition within their native habitats. Dicerandra densiflora exhibited higher FTSW thresholds of 0.547 for whole-plant transpiration and approximately 0.5 for leaf gas exchange parameters. The FTSW thresholds for D. cornutissima grown in the same substrate were lower at 0.445 for

PAGE 101

89 whole-plant transpiration and approximately 0.2 for leaf gas exchange parameters. Although the sandy substrate may be artificially increasing their FTSW thresholds, these closely-related species, which are native to habitats with contrasting water availabilities, respond differently to an extended period of drought. High FTSW thresholds (0.5) for declines in leaf photosynthesis, transpiration, and conductance, similar to those of D. densiflora, have also been reported for another wild species, Nerium oleander , by Gollan et al. (1985). The responses exhibited by D. cornutissima seem ecologically advantageous due to its perennial growth habit and the temporal variation in resource availability it experiences periodically and with each change of season in scrub habitats throughout its lifetime. This speciesÂ’ responses were less sensitive to decreases in soil water availability. Dicerandra cornutissima seedlings must contend with short-term dry conditions periodically in scrub due to the low water-holding capacity of the deep, sandy soils. Although daily transpiration did decline before FTSW fell to 0.40, gas exchange measurements of leaves in early morning hours continued at well-watered levels until FTSW reached 0.2. These perennials must also exhibit traits to promote their survival during the dry season. The carbon gain obtained when conditions were favorable at lower levels of water availability may be used to increase chances of survival when dry conditions continue or intensify. The growth habit of this species would allow utilization of resources for root growth. Resources can be allocated in this way because its growth habit does not require resources to be sent to reproduction. In previous studies (Chapters 2 and 3), Dicerandra cornutissima did not flower during the first year of growth regardless of plant size or resource availability. Greater allocation to root growth was

PAGE 102

90 found in this species as higher RMR values were exhibited by individuals under limited nitrogen availability. Several leaf traits of D. cornutissima have been noted as possible adaptations to dry conditions. Their smaller and thicker leaves with greater Narea enhance water conservation. Their isobilateral leaf structure with its fewer intercellular spaces and less irregularly-shaped spongy mesophyll cells decrease relative transpiration (Chapter 2). This speciesÂ’ significantly greater rm measured in this study may be a consequence of these traits (Taiz and Zeiger, 1998). Therefore, Dicerandra cornutissima exhibits many traits that taken together are advantageous under conditions of shortand long-term water-deficits experienced by this species in scrub habitats. The responses exhibited by D. densiflora seem ecologically advantageous due to the constraints imposed by its annual growth habit and the environmental conditions experienced during its short life cycle. Plantand leaf-level gas exchange declined at relatively high soil water availability. Stomata closure and the corresponding restrictions imposed on the maximum transpiration rate leads to increased water-use-efficiency and soil water conservation. Despite higher water availability in this speciesÂ’ native habitat, Dicerandra densiflora experiences extended periods of drought or extreme decreases in soil water availability during flowering and seed fill (Chen and Gerber, 1990). The growth habit of D. densiflora will not allow carbon gain by individuals in its native habitat to go to increased root growth to improve water uptake once these dry conditions begin during the reproductive phase. The fast onset of water conservation by this annual species may be necessary to ensure that water is available for production of viable seed each year during the seed fill stage. A recent modeling study of the effects of transpiration limitation in sorghum found that incorporation of a restricted maximum-

PAGE 103

91 transpiration-rate trait would increase yield in dry years by conserving soil water for seed fill (Sinclair et al., 2005). A divergence was found in the physiological responses of these closely related species to increasing water-deficit conditions. The strategy employed by each species appears to be influenced by water availability within their native habitats and their growth habit. Both species exhibited some degree of drought tolerance. The traits expressed by D. densiflora and D. cornutissima allowed for maintenance of leaf RWC near saturation and prevention of leaf wilting until FTSW decreased below 0.1 in both species. Overall, Dicerandra cornutissima expressed traits conferring greater drought tolerance as stomata were kept open and carbon gain continued at lower levels of soil water availability.

PAGE 104

92 Table 4-1. Duration of dry-down and length of survival period for D. densiflora and D. cornutissima plants in the water-deficit treatment. Means followed by different letters within a column indicate significant differences at = 0.05 utilizing TukeyÂ’s HSD. Duration of dry-down (d) Length of survival period (d) Species n mean SE mean SE D. densiflora 3 17.3a 0.6 4.33a 0.33 D. cornutissima 4 32.5a 7.5 5.00a 1.00

PAGE 105

93 Table 4-2. Shoot (leaf and stem) and leaf mass of D. densiflora and D. cornutissima plants in the water-deficit treatment. Means followed by different letters within a column indicate significant differences at = 0.05 utilizing TukeyÂ’s HSD. Shoot mass (g) Leaf mass (g) Species n mean SE mean SE D. densiflora 3 1.177a 0.008 0.820a 0.013 D. cornutissima 4 1.132a 0.268 0.864a 0.208

PAGE 106

94 Table 4-3. FTSW threshold values for plantand leaf-level responses. FTSW thresholds D. densiflora D. cornutissima Plant-level responses transpiration rate 0.547 0.445 Leaf-level responses photosynthesis 0.514 0.175 transpiration 0.560 0.167 stomatal conductance 0.553 0.162 RWC 0.086 0.070

PAGE 107

95 0 5 10 15 20 25 30 35 01020304050 Plant 1 Plant 2 Plant 3 0 5 10 15 20 25 30 35 01020304050 Plant 1 Plant 2 Plant 3 Plant 4Days under Dry-downTranspiration / Leaf Mass (g H2O g-1leaf) Transpiration / Leaf Mass (g H2O g-1leaf)a) D. densiflora b) D. cornutissima Figure 4-1. Time course of plant transpiration per unit leaf mass for D. densiflora (a) and D. cornutissima (b) plants in the water-deficit treatment.

PAGE 108

96 0 0.2 0.4 0.6 0.8 1 1.2 1.4 0 0.2 0.4 0.6 0.8 1 0 0.2 0.4 0.6 0.8 1 1.2 1.4 0 0.2 0.4 0.6 0.8 1Normalized Transpiration RatioFraction of Transpirable Soil Water1/(1+A*exp(B*X)) r2= 0.99 A = 5.28, 95% CI = 3.93 to 6.63 B = -8.43, 95% CI = -9.68 to –7.17 1/(1+A*exp(B*X)) r2= 0.98 A = 4.41, 95% CI = 3.34 to 5.48 B = -9.96, 95% CI = -11.44 to –8.48 a) D. densiflora b) D. cornutissima Figure 4-2. NTR – FTSW response curves of D. densiflora (a) and D. cornutissima (b) plants in the water-deficit treatment. Symbols represent daily values of waterdeficit plants throughout the dry-down period. The solid line represents the fit of all data points to the equation of Muchow and Sinclair (1991).

PAGE 109

97 0 0.2 0.4 0.6 0.8 1 1.2 1.4 0 0.2 0.4 0.6 0.8 1 0 0.2 0.4 0.6 0.8 1 1.2 1.4 0 0.2 0.4 0.6 0.8 1 0 0.2 0.4 0.6 0.8 1 1.2 1.4 0 0.2 0.4 0.6 0.8 1 0 0.2 0.4 0.6 0.8 1 1.2 1.4 0 0.2 0.4 0.6 0.8 1 0 0.2 0.4 0.6 0.8 1 1.2 1.4 0 0.2 0.4 0.6 0.8 1 0 0.2 0.4 0.6 0.8 1 1.2 1.4 0 0.2 0.4 0.6 0.8 1Normalized Amax-areaNormalized E Normalized gsD. densiflora D. cornutissima Fraction of Transpirable Soil Waterr2= 0.99 A = 6.18 95% CI = 3.49 to 8.87 B = -9.27 95% CI = -11.33 to –7.21 r2= 0.97 A = 8.17 95% CI = 1.38 to 14.96 B = -28.91 95% CI = -40.97 to –16.85 r2= 0.99 A = 8.25 95% CI = 3.26 to 13.24 B = -9.02 95% CI = -11.49 to –6.55 r2= 0.94 A = 8.67 95% CI = -2.04 to 19.37 B = -30.64 95% CI = -49.25 to –12.03 r2= 0.98 A = 13.68 95% CI = -0.12 to 27.46 B = -10.06 95% CI = -13.84 to –6.28 r2= 0.94 A = 10.26 95% CI = -3.62 to 24.13 B = -32.49 95% CI = -52.76 to –12.21 Figure 4-3. Normalized Amax-area, E, and gs responses of D. densiflora and D. cornutissima to changing FTSW . Symbols represent daily values of waterdeficit plants throughout the dry-down period. The solid line represents the fit of all data points to the equation of Muchow and Sinclair (1991).

PAGE 110

98 0 0.2 0.4 0.6 0.8 1 1.2 0 0.2 0.4 0.6 0.8 1 0 0.2 0.4 0.6 0.8 1 1.2 0 0.2 0.4 0.6 0.8 1Normalized Relative Water ContentFraction of Transpirable Soil Water1/(1+A*exp(B*X)) r2= 0.99 A = 1.43, 95% CI = 1.12 to 1.74 B = -46.92, 95% CI = -55.92 to –37.93 1/(1+A*exp(B*X)) r2= 0.99 A = 0.41, 95% CI = 0.30 to 0.52 B = -23.86, 95% CI = -36.89 to –10.84a) D. densiflora b) D. cornutissima Figure 4-4. Normalized RWC – FTSW response curves of D. densiflora (a) and D. cornutissima (b) plants in the water-deficit treatment. Symbols represent daily values of water-deficit plants throughout the dry-down period. The solid line represents the fit of all data points to the equation of Muchow and Sinclair (1991).

PAGE 111

99 0 0.2 0.4 0.6 0.8 1 1.2 00.20.40.60.81 0 0.2 0.4 0.6 0.8 1 1.2 00.20.40.60.81 0 0.2 0.4 0.6 0.8 1 1.2 00.20.40.60.81 0 0.2 0.4 0.6 0.8 1 1.2 00.20.40.60.81Relative Water ContentNormalized Amax-areaNormalized ED. densiflora D. cornutissima y = 0.0009e7.2983xr2= 0.80 y = 0.0007e7.3823xr2= 0.79 y = 0.0151e4.4199xr2= 0.79 y = 0.0101e4.8229xr2= 0.74 Figure 4-5. Normalized Amax-area and E responses of D. densiflora and D. cornutissima to changing RWC.

PAGE 112

100 CHAPTER 5 CONCLUSIONS The genus Dicerandra consists of perennial species endemic to various portions of FloridaÂ’s unique scrub communities and annual species inhabiting sandhill habitats within the southeastern coastal plain of the United States. The overall objective of this dissertation was to study the variation in plan tand leaf-level ecophysiological attributes of these closely related species of Dicerandra to gain insight into the significance of their divergence in form and function in relation to the environmental conditions of their native habitats. Previous ecophysiological research of species native to FloridaÂ’s unique scrub habitats has been limited to studies concerning the water relations and long-term effects of elevated CO2 levels on traits of Quercus species. Comparison of attributes of these scrub and sandhill congeners increases our knowledge of plantand leaf-level traits that may be ecologically advantageous in scrub habitats. In the first study (Chapter 2), two sandhill annual and four scrub perennial species of Dicerandra were grown within a common garden under abundant availability of resources to determine genotypic differences among these congeners. Growth of Dicerandra species within this common garden found inherent differences in their lifehistory traits. Divergent differences in s hoot architecture, leaf morphology, and leaf anatomy revealed a strategy shift between these closely related species. Perennials native to scrub gaps exhibited traits typically found in plants acclimated and adapted to open, high-light environments. Annuals native to the understory of sandhill habitats expressed traits of plants acclimated and adapted to more shaded environments. The differences

PAGE 113

101 observed in speciesÂ’ leaf morphology and anatomy may confer functional advantages to these species in their respective native habitats. Many of the high-light traits expressed by Dicerandra scrub endemics also correspond to evolutionary divergences in other species in response to decreasing water and nutrient availability. This may increase the ecological benefit of these traits in scrub habitats since nutrients are scarce and water availability is spatially and temporally limited over the lifetime of the perennials. These endemic scrub species did maintain greater RWC after only three days without additional water in the common garden. However, contrasting differences were not found in SLA, Narea, and light response curve components of sandhill annual and scrub perennial Dicerandra congeners. Expression of these traits differed along a continuum with the perennials exhibiting varying degrees of high-li ght adaptation. Further work with these species should include reciprocal transplant studies to measure the growth, morphology, physiological responses, survival, and reproduction of scrub endemics within sandhill habitats and sandhill natives within scrub habitats. In Chapter 3, Dicerandra cornutissima and D. densiflora were grown under various levels of nutrient availability to examine the responses of these closely related species in relation to their different growth habits. A factorial experiment was conducted utilizing three levels of nitrogen (N) and three levels of phosphorus (P). Trade-offs between expression of traits promoting greater growth and maintenance to ensure survival and future fecundity in the perennial versus traits optimizing greater seed production in the annual were found. Dicerandra cornutissima exhibited greater biomass accumulation. This perennial transferred resources from r oot production to leaf and stem production as N availability increased. These changes in allocation signify plastic responses of this

PAGE 114

102 species to continue growth while ensuring survival and future fecundity at various N availabilities. No significant shifts in biomass allocation occurred in D. densiflora . Ontogenetic shifts to seed formation may have constrained overall biomass allocation plasticity in this annual. However, increases in N availability resulted in an increase in the number of flowers produced by each plant due to an increase in plant mass. The perennial exhibited higher biomass NUEN-plant than the annual. Differences between the speciesÂ’ NUEN-plant were caused by differences in tissue allocation. Dicerandra cornutissima allocated more of its resources to root tissue that has a lower N concentration while D. densiflora allocated most of its resources to reproductive tissue that has a higher N concentration. However, when nitrogen-use efficiency was calculated per unit leaf N, the annual exhibited a significantly greater NUEN-leaf that was more than twice that of D. cornutissima . Plasticity in this physiological trait will increase plant fitness in the annual at lower N availability. Availability of N, but not P, affected the growth and traits of both species. Therefore, this scrub and sandhill species both appear to be adapted to growth under low P availability. Future studies should determine if mycorrhizae colonize the roots of Dicerandra species, and what effects their presence has on plant responses to nutrient availability. In the final study (Chapter 4), gas exchange responses of D. cornutissima and D. densiflora were compared as soil water availability decreased to gain insight into the physiological mechanisms utilized by these closely related species to cope with increasing water-deficit conditions. These species experience disparate levels of water availability due to differences in the soil characteristics of their native habitats and their

PAGE 115

103 growth habits. The perennial must survive shortand long-term water-deficits in scrub. Dicerandra cornutissima expressed traits conferring greater drought tolerance as its stomata were kept open and carbon gain continued at lower levels of soil water availability. This allows the species to be less sensitive to short-term, small decreases in water availability that periodically occur in scrub. Future studies should determine if this species exhibits osmotic adjustment or if its indeterminate growth habit allows carbon gain under increasing water deficits to be utilized for root growth to benefit plants under long-term water deficit. Despite higher water availability in its native habitat, Dicerandra densiflora experiences decreases in water availability during flowering and seed fill. This speciesÂ’ greater sensitivity to decreases in water availability through its faster decline in stomatal conductance may be a mechanism to conserve water to successfully complete the seed fill stage. The strategy employed by each species appears to be influenced by water availability within their native habitat and their growth habit. This dissertation presented findings of the first ecophysiological studies conducted on species of Dicerandra , a genus endemic to the southeastern portion of the United States. Results suggest that evolution under the environmental conditions of their native sandhill habitats and FloridaÂ’s unique scrub communities led to divergence in traits of these closely related species. Compared to the sandhill annuals, scrub perennials in the common garden exhibited the shoot architecture, leaf morphology, and leaf anatomy typical of high-light adapted plants. However, the correlation seen in other studies between traits that are advantageous under high-light, limited water, and poor nutrient availability prevents distinction as to which abiotic factor or combination of factors led to expression of these traits in the scrub endemics. In the other studies evaluating speciesÂ’

PAGE 116

104 responses to variation in nutrient and water availability, the scrub perennial, Dicerandra cornutissima , exhibited many differences in form and function that should be advantageous under conditions of limited nutrient availability and shortor long-term water-deficits that are experienced in scrub habitats. This research found that the advantages of variation in functional morphology and physiological responses of Dicerandra species could only be understood in the context of their growth habit.

PAGE 117

105 LIST OF REFERENCES Abrahamson, W. G., and H. Caswell. 1982. On the comparative allocation of biomass, energy, and nutrients in plants. Ecology 63: 982-991. Abrahamson, W. G., A. F. Johnson, J. N. Layne, and P. A. Peroni. 1984. Vegetation of the Archbold Biological Station, Florida: an example of the southern Lake Wales Ridge. Florida Scientist 47: 209-250. Abrams, M.D., and E. S. Menges. 1992. Leaf ageing and plateau effects on seasonal pressure-volume relationships in three sclerophyllous Quercus species in southeastern USA. Functional Ecology 6: 353-360. Ackerly, D. D., S. A. Dudley, S. E. Sultan, J. Schmitt, J. S. Coleman, C. R. Linder, D. R. Sandquist, M. A. Geber, A. S. Evans, T. E. Dawson, and M. J. Lechowicz. 2000. The evolution of plant ecophysiological traits: recent advances and future directions. BioScience 50: 979-995. Aerts, R., and F. S. Chapin III. 2000. The mineral nutrition of wild plants revisited: A re-evaluation of processes and patterns. Advances in Ecological Research 30: 1-67. 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. Anten, N. P. R. 2005. Optimal photosynthetic characteristics of individual plants in vegetation stands and implications for species coexistence. Annals of Botany 95: 495-506. Arntz, A. M., and L. F. Delph. 2001. Pattern and process: evidence for the evolution of photosynthetic traits in natural populations. Oecologia 127: 455-467. Ashton, P. M. S., and G. P. Berlyn. 1992. Leaf adaptations of some Shorea species to sun and shade. New Phytologist 121: 587-596. Ashton, P. M. S., and G. P. Berlyn. 1994. A comparison of leaf physiology and anatomy of Quercus (section Erythrobalanus – Fagaceae) species in different light environments. American Journal of Botany 81: 589-597. Atkin, O. K., B. Botman, and H. Lambers. 1996. The causes of inherently slow growth in alpine plants: an analysis based on the underlying carbon economies of alpine and lowland Poa species. Functional Ecology 10: 698-707.

PAGE 118

106 Ballare, C. L., R. A. Sanchez, A. L. Scopel, and C. M. Ghersa. 1988. Morphological responses of Datura ferox L. seedlings to the presence of neighbours: Their relationships with canopy microclimate. Oecologia 76: 288-293. Bazzaz, F. A., D. D. Ackerly, and E. G. Reekie. 2000. Reproductive allocation in plants. In M. Fenner [ed.], Seeds: the ecology of regeneration in plant communities, 1-29. CABI Publishing, Wallingford, United Kingdom. Bazzaz, F. A., N. R. Chiariello, P. D. Coley, and L. F. Pitelka. 1987. Allocating resources to reproduction and defense. BioScience 37: 58-67. Bazzaz, F. A., and S. R. Morse. 1991. Annual plants: potential responses to multiple stresses. In H. A. Mooney, W. E. Winner, and E. J. Pell [eds.], Responses of plants to multiple stresses, 283-305. Academic Press, San Diego, California. Beerling, D. J., and C. K. Kelly. 1996. Evolutionary comparative analyses of the relationship between leaf structure and function. New Phytologist 134: 35-51. Bjorkman, O. 1981. Photosynthetic respons es to different quantum flux densities. In O. L. Lange, P. S. Nobel, C. B. Osmond, and H. Ziegler [eds.], Encyclopedia of plant physiology, 12a: 57-107. Springer-Verlag, Berlin, Germany. Boardman, N. K. 1977. Comparative photosynthesis of sun and shade plants. Annual Review of Plant Physiology 28: 355-377. Brady, N. C., and R. R. Weil. 1999. The nature of soil properties. Prentice-Hall, Upper Saddle River, New Jersey. Carrington, M. E., and J. E. Keeley. 1999. Comparison of post-fire seedling establishment between scrub communities in Mediterranean and non-mediterranean climate ecosystems. Journal of Ecology 87: 1025-1036. Chapin III, F. S. 1980. The mineral nutrition of wild plants. Annual Review of Ecology and Systematics 11: 233-260. Chapin III, F. S., A. J. Bloom, C. B. Field, and R. H. Waring. 1987. Plant responses to multiple environmental factors. BioScience 37: 49-57. Chaves, M. M. 1991. Effects of water deficits on carbon assimilation. Journal of Experimental Botany 42: 1-16. Chen, E., and J. F. Gerber. 1990. Climate. In R. L. Myers and J. J. Ewel [eds.], Ecosystems of Florida, 11-34. University of Central Florida Press, Orlando, Florida. Christman, S. P., and W. S. Judd. 1990. Notes on plants endemic to Florida scrub. Florida Scientist 53: 52-73.

PAGE 119

107 Clements, E. S. 1905. The relation of leaf structure to physical factors. Transactions of the American Microscopical Society 26: 19-98. Cody, M. L. 1966. A general theory of clutch size. Evolution 20: 174-184. Cohen, D. 1971. Maximizing final yield when growth is limited by time or by limiting resources. Journal of Theoretical Biology 33: 299-307. Corre, W. J. 1983. Growth and morphogenesis of sun and shade plants II. The influence of light quality. Acta Botanica Neerlandica 32: 185-202. Crawley, M. J. 1977. Life history and environment. In M. J. Crawley [ed.], Plant ecology, 73-131. Blackwell Scientific Publications, Boston, Massachusetts. Crick, J. C., and J. P. Grime. 1987. Morphological plasticity and mineral nutrient capture in two herbaceous species of contrasted ecology. New Phytologist 107: 403-414. Cunningham, S. A., B. Summerhayes, and M. Westoby. 1999. Evolutionary divergences in leaf structure and chemistry. Comparing rainfall and soil nutrient gradients. Ecology 69: 569-588. De Lucia, E. H., H. D. Shenoi, S. L. Na idu, and T. A. Day. 1991. Photosynthetic symmetry of sun and shade leaves of different orientations. Oecologia 87: 51-57. Deyrup, M., and E. S. Menges. 1997. Pollination ecology of the rare scrub mint Dicerandra frutescens (Lamiaceae). Florida Scientist 60: 143-157. Dickie, J. B., and P. E. Gasson. 1999. Comparative leaf anatomy of the Penaeaceae and its ecological implications. Botanical Journal of the Linnean Society 131: 327351. Dickison, W. C. 2000. Integrative plant anatomy. Academic Press, San Diego, California. Eisner, T., K. D. McCormick, M. Sakaino, M. Eisner, S. R. Smedley, D. J. Aneshansley, M. Deyrup, R. L. Myers, and J. Meinwald. 1990. Chemical defense of a rare mint plant. Chemoecology 1: 30-37. Esau, K. 1977. Anatomy of seed plants. John Wiley and Sons, New York, New York. Estill, J. C., and M. B. Cruzan. 2001. Phytogeography of rare plant species endemic to the Southeastern United States. Castanea 66: 3-23. Evans, J. R. 1999. Leaf anatomy enables more equal access to light and CO2 between chloroplasts. New Phytologist 143: 93-104.

PAGE 120

108 Evans, J. R., I. Jakobsen, and E. Ogren. 1993. Photosynthetic light-response curves: 2. Gradients of light absorption and photosynthetic capacity. Planta 189: 191-200. Evans, M. E., E. S. Menges, and D. R. Gordon. 2004. Mating systems and limits to seed production in two Dicerandra mints endemic to Florida scrub. Biodiversity and Conservation 13: 1819-1832. Gadgil, M., and W. H. Bossert. 1970. Life hi storical consequences of natural selection. American Naturalist 104: 1-24. Garnier, E. 1992. Growth analysis of congeneric annual and perennial grass species. Journal of Ecology 80: 665-675. Garnier, E., P. Cordonnier, J.-L. Guillerm, and L. Sonie. 1997. Specific leaf area and leaf nitrogen concentration in annual and perennial grass species growing in Mediterranean old-fields. Oecologia 111: 490-498. Garnier, E., and S. Vancaeyzeele. 1994. Carbon and nitrogen content of congeneric annual and perennial grass species: relationships with growth. Plant, Cell and Environment 17: 399-407. Geller, G. N., and W. K. Smith. 1982. Influence of leaf size, orientation, and arrangement on temperature and transpiration in three high-elevation, large-leafed herbs. Oecologia 53: 227-234. Givnish, T. J. 1987. Comparative studies of leaf form: Assessing the relative role of selective pressures and phylogenetic constraints. New Phytologist 106 (Suppl.): 131-160. Givnish, T. J. 1988. Adaptation to sun and shade: A whole-plant perspective. Australian Journal of Plant Physiology 15: 63-92. Givnish, T. J., and G. J. Vermeij. 1976. Size and shape of liane leaves. American Naturalist 110: 743-778. Gollan, T., N. C. Turner, and E.-D. Schulze. 1985. The responses of stomata and leaf gas exchange to vapour pressure deficits and soil water content. Oecologia 65: 356-362. Griffith, T. M., and M. A. Watson. 2005. Stress avoidance in a common annual: reproductive timing is important for local adaptation and geographic distribution. Journal of Evolutionary Biology 18: 1601-1612. Grime, J. P. 1977. Evidence for the existence of three primary strategies in plants and its relevance to ecological and evolutionary theory. American Naturalist 111: 11691194.

PAGE 121

109 Grime, J. P., and R. Hunt. 1975. Relative growth-rate: its range and adaptive significance in a local flora. Journal of Ecology 63: 393-422. Gutschick, V. P. 1999. Research reviews: Biotic and abiotic consequences of differences in leaf structure. New Phytologist 143: 3-18. Hanson, H. C. 1917. Leaf-structure as related to environment. American Journal of Botany 4: 533-560. Hiremath, A. J., J. J. Ewel, and T. G. Cole. 2002. Nutrient use efficiency in three fastgrowing tropical trees. Forest Science 48: 662-672. Hirose, T., and N. Kachi. 1982. Critical Plant Size for flowering in biennials with special reference to their distribution in a sand dune system. Oecologia 55: 281284. Hirose, T., and K. Kitajima. 1986. Nitroge n uptake and plant growth I. Effect of nitrogen removal on growth of Polygonum cuspidatum . Annals of Botany 58: 479486. Huck, R. B. 1987. Systematics and evolution of Dicerandra (Labiatae). Phanerogamarum Monographiae 19: 1-343. Huck, R. B., and H. L. Chambers. 1997. Polyploidy: a factor in the evolution of Dicerandra Benth. (Labiatae). Edinburgh Journal of Botany 54: 217-229. Huck, R. B., W. S. Judd, W. M. Whitten, J. D. Skean, R. P. Wunderlin, and K. R. Delaney. 1989. A new Dicerandra (Labiatae) from the Lake Wales Ridge of Florida, with a cladistic analysis and discussion of endemism. Systematic Botany 14: 197-213. Hungate, B. A., M. Reichstein, P. Dijkstra, D. Johnson, G. Hymus, J. D. Tenhunen, C. R. Hinkle, and B. G. Drake. 2002. Evaporation and soil water content in a scrub-oak woodland under carbon dioxide enrichment. Global Climate Change 8: 289-298. Hutchings, M. J., and H. de Kroon. 1994. Foraging in plants: the role of morphological plasticity in resource acquisition. Advances in Ecological Research 25: 159-238. Hymus, G. J., D. P. Johnson, S. Dore, H. P. Anderson, C. R. Hinkle, and B. G. Drake. 2003. Effects of elevated atmospheric CO2 on net ecosystem CO2 exchange of a scrub-oak ecosystem. Global Climate Change 9: 1802-1812. Hymus, G. J., T. G. Snead, D. P. Johnson, B. A. Hungate, and B. G. Drake. 2002. Acclimation of photosynthesis and respiration to elevated atmospheric CO2 in two scrub oaks. Global Climate Change 8: 317-328. Jones, H. G. 1985. Partitioning stomatal and non-stomatal limitations to photosynthesis. Plant, Cell and Environment 8: 95-104.

PAGE 122

110 Kachi, N., and T. Hirose. 1983a. Bolting induction in Oenothera erythrosepala Borbas in relation to rosette size, vernalization, and photoperiod. Oecologia 60: 6-9. Kachi, N., and T. Hirose. 1983b. Limiting nutrients for plant growth in coastal sand dune soils. Journal of Ecology 71: 937-944. Kim, S. E., and H. Okubo. 1995. Control of growth habit in determinate lablab bean ( Lablab purpureus ) by temperature and photoperiod. Scientia Horticulturae 61: 147-155. Lambers, H., F. S. Chapin III, and T. L. Pons. 1998. Plant physiological ecology. Springer-Verlag, New York, New York. Lee, D. W., S. F. Oberbauer, P. Johnson, B. Krishnapilay, M. Mansor, H. Mohamad, and S. K. Yap. 2000. Effects of irradiance and spectral quality on leaf structure and function in seedlings of two southeast Asian Hopea (Dipterocarpaceae) species. American Journal of Botany 87: 447-455. Li, J. H., P. Dijkstra, C. R. Hinkle, R. M. Wheeler, and B. G. Drake. 1999. Photosynthetic acclimation to elevated atmospheric CO2 concentration in the Florida scrub-oak species Quercus geminata and Quercus myrtifolia growing in their native environment. Tree Physiology 19: 229-234. Li, J. H., W. A. Dugas, G. J. Hymus, D. P. Johnson, C. R. Hinkle, B. G. Drake, and B. A. Hungate. 2003. Direct and indirect effect s of elevated CO2 on transpiration from Quercus myrtifolia in a scrub-oak ecosystem. Global Change Biology 9: 96-105. McCormick, K. D., M. A. Deyrup, E. S. Menges, S. R. Wallace, J. Meinwald, and T. Eisner. 1993. Relevance of chemistry to conservation of isolated populations: the case of volatile leaf components of Dicerandra mints. Proceedings of the National Academy of Sciences 90: 7701-7705. McCree, K. J., and S. G. Richardson. 1987. Stomatal closure vs. osmotic adjustment: a comparison of stress responses. Crop Science 27: 539-543. McDonald, D. B., and J. L. Hamrick. 1996. Genetic variation in some plants of Florida scrub. American Journal of Botany 83: 21-27. Menges, E. S. 1992. Habitat preferences and response to disturbance for Dicerandra frutescens , a Lake Wales Ridge (Florida) endemic plant. Bulletin of the Torrey Botanical Club 119: 308-313. Menges, E. S. 1994. Fog temporarily increases water potential in Florida scrub oaks. Florida Scientist 57: 65-74. Menges, E. S. 1999. Ecology and conservation of Florida scrub. In R. C. Anderson, J. S. Fralish, and J. M. Baskin [eds.], Savannas, barrens, and rock outcrop plant communities of North America, 7-22. Cambridge University Press, USA.

PAGE 123

111 Menges, E. S., R. W. Dolan, R. Yahr, and D. R. Gordon. 2001. Comparative genetics of seven plants endemic to Florida’s Lake Wales Ridge. Castanea 66: 98-114. Menges, E. S., and N. P. Gallo. 1991. Water relations of scrub oaks on the Lake Wales Ridge, Florida. Florida Scientist 54: 69-79. Menges, E. S., P. J. McIntyre, M. S. Finer, E. Goss, and R. Yahr. 1999. Microhabitat of the narrow Florida scrub endemic Dicerandra christmanii , with comparisons to its congener D. frutescens . Journal of the Torrey Botanical Society 126: 24-31. Menges, E. S., P. F. Quintana Ascencio, C. W. Weekley, and O. G. Gaoue. 2006. Population viability analysis and fire return intervals for an endemic Florida scrub mint. Biological Conservation 127: 115-127. Morgan, D. C., and H. Smith. 1981. Non-photosynthetic responses to light quality. In O. L. Lange, P. S. Nobel, C. B. Osmond, and H. Ziegler [eds.], Physiological plant ecology: Responses to the physical environment, 109-130. Springer-Verlag, Berlin, Germany. Morishima, H., Y. Sano, and H. I. Oka. 1984. Differentiation of perennial and annual types due to habitat conditions in the wild rice Oryza perennis . Plant Systematics and Evolution 144: 119-135. Mott, K. A., A. C. Gibson, and J. W. O’Leary. 1982. The adaptive significance of amphistomatic leaves. Plant, Cell and Environment 5: 455-460. Mott, K. A., and O. Michaelson. 1991. Amphistomy as an adaptation to high light intensity in Ambrosia cordifolia (Compositae). American Journal of Botany 78: 76-79. Muchow, R. C., and T. R. Sinclair. 1991. Water deficit effects on maize yields modeled under current and “greenhouse” climates. Agronomy Journal 83: 1052-1059. Muller, B., and E. Garnier. 1990. Components of relative growth rate and sensitivity to nitrogen availability in annual and perennial species of Bromus . Oecologia 84: 513-518. Mulkey, S. S., S. J. Wright, and A. P. Smith. 1991. Drought acclimation of an understory shrub ( Psychotria limonensis ; Rubiaceae) in a seasonally dry tropical forest in Panama. American Journal of Botany 78: 579-587. Myers, R. L. 1990. Scrub and high pine. In R. L. Myers and J. J. Ewel [eds.], Ecosystems of Florida, 150-193. University of Central Florida Press, Orlando, Florida. Nilsen, E. T., and D. M. Orcutt. 1996. Physiology of plants under stress. John Wiley and Sons, Inc., New York, New York.

PAGE 124

112 Nobel, P. S., and T. L. Hartsock. 1981. Development of leaf thickness for Plectranthus parviflorus – Influence of photosynthetically active radiation. Physiologia Plantarum 51: 163-166. Oliveira, L. O., R. B. Huck, M. A. Gitzendanner, W. S. Judd, D. E. Soltis, and P. S. Soltis. In press. Molecular phylogeny, biogeography, and systematics of Dicerandra (Lamiaceae), a genus endemic to the southeastern United States. American Journal of Botany . Padgett, P. E., and E. B. Allen. 1999. Differential responses to nitrogen fertilization in native shrubs and exotic annuals common to Mediterranean coastal sage scrub of California. Plant Ecology 144: 93-101. Parkhurst, D. F. 1978. The adaptive significance of stomatal occurrence on one or both surfaces of leaves. Journal of Ecology 66: 367-383. Pellegrino, A., E. Lebon, M. Voltz, and J. Wery. 2004. Relationships between plant and soil water status in vine ( Vitis vinifera L.). Plant and Soil 266: 129-142. Pitelka, L. F. 1977. Energy allocation in annual and perennial lupines ( Lupinus : Leguminosae). Ecology 58: 1055-1065. Poorter, H., and C. Remkes. 1990. Leaf area ratio and net assimilation rate of 24 wild species differing in relative growth rate. Oecologia 83: 553-559. Ray, J. D., R. W. Gesch, T. R. Sinclair, and L. H. Allen. 2002. The effect of vapor pressure deficit on maize transpiration response to a drying soil. Plant and Soil 239: 113-121. Ray, J. D., and T. R. Sinclair. 1997. Stomatal closure of maize hybrids in response to drying soil. Crop Science 37: 803-807. Ray, J. D., and T. R. Sinclair. 1998. The e ffect of pot size on growth and transpiration of maize and soybean during water deficit stress. Journal of Experimental Botany 49: 1381-1386. Sadras, V. O., and S. P. Milroy. 1996. Soil-water thresholds for the responses of leaf expansion and gas exchange: A review. Field Crops Research 47: 253-266. Salisbury, F. B., and C. W. Ross. 1992. Plant physiology. Wadsworth Publishing Company, Belmont, California. Schaffer, W. M., and M. D. Gadgil. 1975. Selection for optimal life histories in plants. In M. L. Cody and J. M. Diamond [eds.], Ecology and evolution of communities, 142-157. Belknap Press of Harvard University Press, Cambridge, Massachusetts. Schuepp, P. H. 1993. Tansley Review No. 59: Leaf boundary layers. New Phytologist 125: 477-507.

PAGE 125

113 Schmitt, J., and R. D. Wulff. 1993. Light spectral quality, phytochrome and plant competition. Trends in Ecology and Evolution 8: 47-51. Sibly, R., and P. Calow. 1983. An integr ated approach to life-cycle evolution using selective landscapes. Journal of Theoretical Biology 102: 527-547. Sinclair, T. R. 1986. Water and nitrogen limitation in soybean grain production I. Model development. Field Crops Research 15: 125-141. Sinclair, T. R., G. L. Hammer, and E. J. van Oosterom. 2005. Potential yield and wateruse efficiency benefits in sorghum from limited maximum transpiration rate. Functional Plant Biology 32: 945-952. Sinclair, T. R., and M. M. Ludlow. 1986. Influence of soil water supply on the plant water balance of four tropical grain legumes. Australian Journal of Plant Physiology 13: 329-341. Smith, H. 1982. Light quality, photoperception, and plant strategy. Annual Review of Plant Physiology 33: 481-518. Smith, W. K., D. T. Bell, and K. A. Shepherd. 1998. Associations between leaf structure, orientation, and sunlight exposure in five Western Australian communities. American Journal of Botany 85: 56-63. Stiling, P., D. Moon, G. Hymus, and B. Drake. 2004. Differential effects of elevated CO2 on acorn density, weight, germination, and predation among three oak species in a scrub-oak forest. Global Climate Change 10: 228-232. Taiz, L., and E. Zeiger. 1998. Plant physiology. Sinauer Associates, Inc., Sunderland, Massachusetts. Terashima, I., and T. Seaki. 1983. Light environment within a leaf: I. Optical properties of paradermal sections of Camellia leaves with special reference to differences in the optical properties of palisade and spongy tissues. Plant and Cell Physiology 24: 1493-1501. Valladares, F., and R. W. Pearcy. 1998. The functional ecology of shoot architecture in sun and shade plants of Heteromeles arbutifolia M. Roem., a Californian chaparral shrub. Oecologia 114: 1-10. Van der Werf, A., A. J. Visser, F. Schieving, and H. Lambers. 1993. Evidence for optimal partitioning of biomass and nitrogen at a range of nitrogen availabilities for a fastand slow-growing species. Functional Ecology 7: 63-74. Vogelmann, T. C., and G. Martin. 1993. The functional significance of palisade tissue: penetration of directional versus diffuse light. Plant, Cell and Environment 16: 6572.

PAGE 126

114 Vogelmann, T.C., J. N. Nishio, and W. K. Smith. 1996. Leaves and light capture: light propagation and gradients of carbon fixation within leaves. Trends in Plant Science 1: 65-70. Willis, A. J., and E. W. Yemm. 1961. Braunton Burrows: mineral nutrient status of the dune soils. Journal of Ecology 49: 377-390. Winn, A. A. 1996a. Adaptation to fine-grained environmental variation: an analysis of within-individual leaf variation in an annual plant. Evolution 50: 1111-1118. Winn, A. A. 1996b. The contributions of programmed developmental change and phenotypic plasticity to within-individual variation in leaf traits in Dicerandra linearifolia . Journal of Evolutionary Biology 9: 737-752. Winn, A. A. 1999. Is seasonal variation in leaf traits adaptive for the annual plant Dicerandra linearifolia ? Journal of Evolutionary Biology 12: 306-313. Wright, I. J., P. B. Reich, and M. Westoby. 2001. Strategy shifts in leaf physiology, structure and nutrient content between species of highand low-rainfall and highand low-nutrient habitats. Functional Ecology 15: 423-434. Wright, I. J., M. Westoby, and P. B. Reich. 2002. Convergence towards higher leaf mass per area in dry and nutrient-poor habitats has different consequences for leaf life span. Journal of Ecology 90: 534-543. Zangerl, A. R., and F.A. Bazzaz. 1983. Responses of an early and late successional species of Polygonum to variations in resource availability. Oecologia 56: 397404.

PAGE 127

115 BIOGRAPHICAL SKETCH Amethyst Gail Merchant was born and raised in Macon, Georgia. Influenced by her fatherÂ’s love of the natural world and her grandmotherÂ’s magnificent garden, she appreciated the beauty of all life forms and began asking questions about the plants and animals around her at an early age. While watching her father teach biology to high school students, she realized that she wanted to teach others about the importance of the life around them. She acquired her Bachelor of Science degr ee in biology from Georgia College (now Georgia College and State University) in Milledgeville, Georgia. Her interest in plant ecophysiology became apparent during her senior year. Therefore, she continued her education under the guidance of Dr. Erik T. Nilsen in the Biology Department at Virginia Polytechnic Institute and State University in Blacksburg, Virginia. The thesis she completed for her masterÂ’s degree in biology was concerned with the light and water stress tolerance of two invasive legumes: Cytisus scoparius (Scotch broom) and Spartium junceum (Spanish broom). For her doctorate, she became a graduate student in the Botany Department at the University of Florida under the supervision of Dr. Stephen S. Mulkey. Her continued interest in plant ecophysiology led her to perform the studies discussed in this dissertation that utilized members of the southeastern endemic genus Dicerandra . After completion of this doctorate, Amethyst looks forward to teaching others about the world around them and continuing to satisfy her own curiosity through research.